Vol. XLIV, no.

5, 2017 ISSN: 0390-6663

CLINICAL AND EXPERIMENTAL

OBSTETRICS & GYNECOLOGY
an International Journal

Founding Editor
A. Onnis
Montréal (CND)

Editors-in-Chief
M. Marchetti J.H. Check
Montréal (CND) Camden, NJ (USA)
Assistant Editor
A. Sinopoli
Toronto (CND)

Editorial Board
Andrisani A., Padua (Italy)
Audet-Lapointe P., Montréal (Canada)
Axt-Fliedner R., Lübeck (Germany)
Basta A., Krakow (Poland)
Bender H.J., Dusseldorf (Germany)
Bhattacharya N., Calcutta (India)
Bonilla Musoles F., Valencia (Spain)
Cabero-Roura L., Barcelona (Spain)
Charkviani T., Tbilisi (Georgia)
Chavan N., Mumbai (India)
Dexeus S., Barcelona (Spain)
Eskes T.K.A.B.,
Nijmegen (The Netherlands)
Farghaly S.A., New York (USA)
Friedrich M., Homburg (Germany)
Holub Z., Kladno (Czech Republic)
Kaplan B., Petach Tikva (Israel)
Markowska J., Poznan (Poland)
Marth C., Innsbruck (Austria)
Meden-Vrtovec H., Ljubljana (Slovenia)
Murta E.F.C., Uberaba (Brazil)
Mynbaev O.A. Moscow (Russia)
Papadopoulos N.,
Alexandroupolis (Greece)
Rakar S., Ljubljana (Slovenia)
Rigó J., Budapest (Hungary)
Rouzi A.A., Jeddah (Saudi Arabia)
Stelmachow J., Warsaw (Poland)
Varras M.N., Athens (Greece)
Winter R., Graz (Austria)

7847050 CANADA, Inc.

Administrative Office (M. Beaucage):
4900 Côte St-Luc, Apt #212 – Montréal, Québec, H3W 2H3 (Canada)
Tel. +1-514-4893242 – Fax +1-514-4854513 – e-mail: canlux@mgroup-online.com
Website: www.irog.net
Editorial Office (M. Critelli):
Via Martiri della Libertà, 9, 35137 Padua (Italy)
Tel. +39-049-656521 – Fax +39-049-8752018 – e-mail: irog.canada@gmail.com

CLINICAL AND EXPERIMENTAL OBSTETRICS & GYNECOLOGY – CEOG (ISSN 0390-6663) publishes original work,
preferably brief reports, in the fields of Gynecology, Obstetrics, Fetal Medicine, Gynecological Endocrinology and related subjects,
(Fertility and Sterility, Menopause, Uro-gynecology, Ultrasound in Obstetrics and Gynecology, Sexually Transmitted Diseases, and
Reproductive Biological Section). The Journal is covered by ISI Journal Master List, Index Copernicus International, Science Citation
Index Expanded, Current Contents - Clinical Medicine, Web of Science, Index Medicus/MEDLINE, EMBASE Excerpta Medica,
PubMed, MedSci, Pubget, Genamics JournalSeek, Sciencescape, Unbound Medicine, and PubFacts.com. CEOG is issued bimonthly
in one volume per year by 7847050 CANADA Inc., Montréal (Canada).
Printed in Italy by “Centro Servizi Editoriali S.r.l.” - Grisignano di Zocco - 36040 Vicenza (Italy).
Contents Clinical and Experimental Obstetrics & Gynecology - Vol. XLIV, no. 5, 2017

LETTERS TO THE EDITOR

The circumvallate placenta as a possible culprit of fetomaternal hemorrhage 653
H. Takahashi - Shimotsuke, Tochigi, JAPAN
Circumvallate placenta may be associated with fetomaternal hemorrhage.

Environmental influence on predisposing genes for holoprosencephaly in monochorionic diamniotic twins 655
L. Marin, A. Andrisani - Padua, ITALY
Holoprosencephaly, the most common structural anomaly of the developing forebrain characterized by incomplete separation of
the prosencephalon, is presented.

REVIEW ARTICLES
Renal tumors in pregnancy: a systematic review 657
A. Pontis, F. Congiu, F. Sedda, P. Litta, A. De Lisa, G.B. Melis, S. Angioni - Cagliari, ITALY
Articles on renal tumors during pregnancy published from 1980 to 2015 are reviewed.

Multiple sclerosis management in pregnancy 662

L. Sahin, Y. Ehi - Kars, TURKEY
The different implications of multiple sclerosis during pregnancy are evaluated.

Interventions for treating amniotic fluid embolism: a systematic review with meta-analysis 666
U. Indraccolo, R. Ventrone, G. Scutiero, P. Greco, S.R. Indraccolo - Rome, ITALY
Each intervention for supporting the heart and lung function is perceived as pivotal in the management of amniotic fluid embolism.

ORIGINAL ARTICLES
Indications, limitations and complications of operative hysteroscopy: a retrospective study of an 8-year
experience 678
D. Caserta, S. Picchia, E. Ralli, E. Matteucci, L. Di Benedetto, M. Mallozzi, G. Adducchio, R. Di Iorio,
M. Moscarini† - Rome, ITALY
The hysteroscopic procedures performed over an 8-year experience in a sample of 1,412 women are analyzed.

Office hysteroscopy for removal of retained products of conception: can we predict treatment outcome? 683
A. Cohen, Y. Cohen, S. Sualhi, S. Rayman, F. Azem, G. Rattan - Tel-Aviv, ISRAEL
To evaluate the safety and efficacy of office hysteroscopy in the management of retained product of conception and to identify
those predictors for treatment success.

Protective role of adrenomedullin in heterotopic ovarian transplant 686

E. Erdemoglu, S.G. Gürgen, E. Erdemoglu, K. Kürşat Bozkurt, E. Oz Oyar - Isparta, TURKEY
Adrenomedullin in preventing ischemia and morphological changes in heterotopically transplanted ovary is assessed.

Which type of circumcision is more harmful to female sexual functions? 691

O. Birge, D. Arslan, E.G. Ozbey, M. Adiyeke, I. Kayar, M.M. Erkan, U. Akgör - Nyala-Darfur, SUDAN
Female genital mutilation is common in Sub-Saharan Africa and has been shown to cause sexual dysfunction.

Microwave endometrial ablation at a frequency of 2.45 GHz for menorrhagia: analysis of its efficacy,
recurrence rate, and complications 695
K. Nakayama, K. Nakamura, T. Ishibashi, M. Ishikawa, S. Kyo - Shimane, JAPAN
An investigation of 72 cases evaluated for improvement of menorrhagia after microwave endometrial ablation was conducted.
 Contents 649

Retrospective evaluation of anaesthesia methods in pregnant women with neurological and

neuromuscular syndromes who underwent caesarean section 700
A. Sargin, Z. Pestilci Cağıran, U. Ozdemir Biliç, B. Tanatti Orhanel, S. Karaman - Izmir, TURKEY
The anaesthesia methods used in pregnant women with neurological or neuromuscular disease who underwent caesarean
section was investigated,

Vaginal microbiota in asymptomatic Brazilian women with HIV 704

M.K. Figueiredo Facundo, C.R. de Souza Bezerra Sakano, C.R. Nogueira de Carvalho, A.M. de Oliveira
Machado, N.M. de Góis Speck, J. Chamorro Lascasas Ribalta - São Paulo, BRAZIL
Asymptomatic HIV seropositive women have diversified vaginal flora that may predispose them to local imbalance and acquisi-
tion of genital infections.

Uterus and myoma histomorphology 710

İ. Ceylan, T. Peker, N. Coşkun, S. Ömeroğlu, A. Poyraz - Ankara, TURKEY
The histomorphological differences between myoma uteri and uterus are demonstrated.

The association between cystatin C and metabolic syndrome according to menopausal status in healthy
Korean women 716
Y.J. Lee, K.Y. Yun, S.C. Kim, J.K. Joo, K.S. Lee - Busan, KOREA
The relationship between serum cystatin C level and metabolic syndrome is investigated, in healthy menopausal women.

What is the best initial cycle IVF protocol for patients over 35 years old? 721
P. Telli Celtemen, N. Bozkurt, M. Erdem, M.B. Celtemen, A. Erdem, M. Öktem, R.O. Karabacak - Ankara,
TURKEY
The most effective ovarian hyperstimulation protocol for patients over 35 years of age was assessed.

Effect of aerobic exercises versus foot reflexology on post-menopausal depression 726

A.M. Eman, M.E. Mahmoud - Cairo, EGYPT
Aerobic exercise and foot reflexology were effective adjunct methods in reducing postmenopausal depression but aerobic exer-
cise is more effective than foot reflexology.

Significance of growth differentiation factor 15 in primary ovarian insufficiency: inflammatory,

biochemical, and hormonal correlates 730
S.Y. Tunc, N.Y. Goruk, E. Agacayak, M.S. Icen, F.M. Findik, H. Kusen, M.S. Evsen, H. Yuksel, T. Gul -
Diyarbakir, TURKEY
The results imply that neutrophil-lymphocyte ratio serve as a promising marker for primary ovarian insufficiency patients and role
of inflammatory process in pathogenesis should be investigated in further trials.

Clinicopathological changes of perinatal mortality during the last 20 years in a tertiary hospital of
Greece 734
C. Goudeli, L. Aravantinos, D. Mpotsis, G. Creatsas, A. Kondi-Pafiti - Athens, GREECE
The changes in gestational-age-specific perinatal death causes during a 20-year period and the conditions that led to the pathol-
ogy of the mother, fetus, and membranes are illustrated.

Roles of high-risk human papilloma virus (HR-HPV) E6/E7mRNA in triaging HPV16/18 cases 740
L. Liu, Y.M. Chen, Q.Y. Zhang, C.Z. Li - Jinan, CHINA
The roles of HR-HPV E6/E7mRNA in triaging patients negative for intraepithelial lesion (NILM) accompanied with HPV16/18
infection are investigated.

The relevance of fascial surgical repair in the management of pelvic organ prolapse (POP) 744
F. Nobili, A. Lukic, I. Puccica, M. Vitali, M. Schimberni, F. Manzara, A. Frega, B. Mossa, M. Moscarini†,
D. Caserta - Rome, ITALY
The anatomical, functional, and post-operative outcomes of fascial surgical repair in the management of pelvic organ prolapse
are evaluated.

Inherited thrombophilia and thromboprophylaxis: a retrospective analysis of pregnancy outcomes in

106 patients 749
H. Alptekin, N. Alptekin, R. Selimoğlu, T. Cengiz, S. Barış - Konya, TURKEY
Low-molecular-weight heparin and low-dose aspirin given in combination were evaluated in females with five commonly inherited
thrombophilia polymorphisms to address unexplained recurrent pregnancy loss.
650 Contents

Is there a relationship between maternal blood type and the incidence of gestational diabetes mellitus?
A retrospective review 755
A.M. Oraif, H.A. Jabar, M. Ashi, H. Al Ghanmi, A.R. Al Jarallah - Jeddah, SAUDI ARABIA
The relationship between blood type and incidence of gestational diabetes mellitus are retrospectively reviewed.

Clinical value of transfontanellar ultrasonography for neonatal insular development 758

X.K. Chen, S.H. Chen, G.R. Lv, J.H. You, Z.K. Chen - Quanzhou, CHINA
The morphological characteristics and to establish ultrasonographic standards of normal neonatal insula size using transfontanellar
ultrasonography, and to evaluate the clinical value of this technique are assessed.

Arteriovenous malformations (AVM) of the corpus uteri 764

L. Roncati, T. Pusiol - Rovereto, ITALY
Arteriovenous malformations should be routinely remarked in the histopathological reports, because their presence could be correlated
with an explainable history of dysmenorrhea.

Investigation of the relationship between fear of childbirth and social supports of pregnant women in the
third trimester in Turkey 767
S. Ertekin Pinar, O. Duran Aksoy, B. Cesur, D. Bilgic, G. Daglar, E. Guler - Sivas, TURKEY
The relationship between fear of childbirth and social supports of pregnant women in the third trimester are investigated.

Prevalence of congenital malformations during pregnancy in China: screening by ultrasound examination 772
L.J. Kong, L. Fan, G.H. Li, W.Y. Zhang - Beijing, CHINA
The majority of perinatal deaths are due to complex congenital malformations.

Does increase in body mass index effect primary dysmenorrhea? 777

M. Temur, U. Gök Balci, Y. A. Güçlü, B. Korkmaz, P.Ö. Özbay, N. Soysal, Ö. Yilmaz, T.T. Yilmazer, T. Çift,
K. Öngel - İzmir, TURKEY
The relationship between obesity and dysmenorrhea, and the effects of socio-demographic features on it, were evaluated.

Maternal serum soluble CD40 ligand concentration as a predictor of preeclampsia at first trimester 782
F. Hatiboğlu, S. Kumbasar, B.A. Şık, E. Sever, M. Temur, S. Salman, Ö. Çot, A. Özcan, F. Yazıcıoğlu -
Istanbul, TURKEY
The use of serum soluble CD40 ligand concentration values, measured between 11+0 and 13+6 weeks of pregnancy, in the predic-
tion of preeclampsia development, and to determine the presence of a statistically significant difference was investigated.

CASE REPORTS
Abnormal bending of the umbilical cord due to adhesion of the cord to the placenta 787
Tatsuya Ishiguro, Takao Ishiguro - Sanjo-city, JAPAN
Abnormal adhesion of the cord to the placenta caused fetal distress during delivery.

Umbilical endometriosis: a rare case of spontaneous cutaneous umbilical endometriosis 789

M. Kalinderis, U. Singh - Kent, UNITED KINGDOM
This report describes a case of umbilical endometriosis that developed in the absence of previous abdominal or uterine surgery.

Successful single-port laparoscopic management of abdominal pregnancy in the Douglas pouch 792
X. Yang, K. Ma - Beijing, CHINA
A single-port laparoscopic resection for an abdominal pregnancy, providing a detailed description of the procedures is reported.

A case of disseminated intravascular coagulation developed after surgical management of corpus luteal
hemorrhage in a patient with Klippel-Trenaunay syndrome 795
S.E. Han, Y.H. Kim, S.C. Kim, J.K. Joo, D.S. Suh, K.H. Kim, K.S. Lee - Busan, KOREA
Corpus luteal hemorrhage in patient with Klippel-Trenaunay syndrome is presented.

Breast capillary hemangioma at the tail of Spencer: a rare entity 798

A. Bothou, I. Grammatikakis, N. Evangelinakis, C. Eftichiadis, G. Iatrakis, S. Zervoudis - Athens, GREECE
A palpable breast lump is a frequent clinical finding and preoperative evaluation varies depending on its localization and characteristics.

A very rare case of ectopic intramural pregnancy after IVF-ET 802

B. Bechev, M. Konovalova - Sofia, BULGARIA
The successful use of early medical treatment of ectopic intramural pregnancy with ultrasound-guided laparoscopic methotrexate
injection is reported.
 Contents 651

A rare cause of intractable tachycardia during caesarean section: acute cannabis use 804
B. Tuncali - Izmir, TURKEY
A persistent perioperative tachycardia in a parturient who underwent an emergency caesarean section under combined
spinal epidural anaesthesia is presented.

Confined placental mosaicism of trisomy 16 detected by non-invasive prenatal testing and multiple
abnormalities 806
Ting Wang, Qin Zhang, Haibo Li, Wei Wang - Suzhou, CHINA
A case of confined placental trisomy 16 mosaicism (CPM16) with abnormal amniotic fluid, placental lake, and other abnormal-
ities was investigated.
CEOG Clinical and Experimental
Obstetrics & Gynecology

Letters to the Editor

The circumvallate placenta as a possible culprit

of fetomaternal hemorrhage

H. Takahashi
Department of Obstetrics and Gynecology, Jichi Medical University, Shimotsuke, Tochigi (Japan)

Dear Editor,
We read with interest the article, “Idiopathic massive fe-
tomaternal hemorrhage (FMH) in the third trimester of
pregnancy causing neonatal death” by Peng et al. [1]. They
concluded: “A pregnant woman at late pregnancy with
complaints of unspecific signs such as decreased fetal
movement (DFM) should arouse a high index of clinical
suspicion of idiopathic FMH”. Another recent study
demonstrated that 44% of stillbirth cases due to FMH
showed DFM [2]. It also showed that in two-thirds of still-
birth due to FMH, risk factors of FMH remained unidenti-
fied [2], which Peng et al. referred to as “idiopathic FMH”.
Many FMH are deemed to be unpreventable. Thus, identi-
fying conditions underlying FMH may increase obstetri-
cians’ level of concern for FMH occurrence.
Recently, a new concept of FMH was reported [3]: FMH
was more likely to occur in placentas with “pathological
permeability”, in which fetal blood is more likely to trans-
fer to maternal blood. This is reasonable. An “abnormal tro-
phoblastic invasion” underlies this pathological
permeability: preeclampsia, placental abruption, and pla-
centa previa were listed [3]. Briefly describing a case with
DFM, we suggest that circumvallate placenta, an “abnor-
mal trophoblast invasion”-related condition, may be asso-
ciated with FMH.
At the 32nd week, a 28-year old primiparous woman com-
plained of DFM, showing sinusoidal pattern on car-
diotocogram (Figure 1a) and high middle-cerebral-artery
peak systolic velocity (2.32 MoM). Cesarean section
yielded a 1,486-gram infant (hemoglobin 2.0 g/dL; Apgar
score 4/5 [1/5 minutes]) with maternal hemoglobin F 4.3%,
confirming the diagnosis of FMH. Placental hematoma was
present at the periphery of the circumvallate plate (Figure
1b). With transfusion, the infant was healthy without se-
quelae. An abnormally shallow trophoblastic invasion is
Figure 1. — Cardiotocographic finding (a) and placental gross
finding (b).
(a): Sinusoidal pattern on cardiotocogram at the 32nd week.
(b): Arrowheads indicate circumvallate plate and arrows indicate
peripheral hematoma.
considered to underlie circumvallate placenta [4] similar to
preeclamptic placenta, and may increase the permeability
of the placental barrier, causing FMH. In addition, circum-
vallate placenta frequently accompanies “peripheral
hematoma” [4, 5] as observed here. Pregnant women with
circumvallate placenta sometimes bleed, which was shown
to be of mixed maternal and fetal origin [4]. Hematoma,
destroying the placental barrier there, at least partly, also
may contribute to FMH.
Circumvallate placenta was reported to be associated
with various disorders including preterm delivery, placen-
tal abruption, or fetal growth restriction [4, 5]. To our
knowledge, circumvallate placenta was not associated with
FMH. This may be due to: 1) circumvallate placenta may
be frequently unrecognized, as suggested by the wide range
of its reported incidences (0.62-18.3%) [4], or 2) even if
recognized, significant proportion of obstetricians may con-
sider the circumvallate placenta as non-pathological, and
thus consider concomitant occurrence of circumvallate pla-
centa and FMH as insignificant, and therefore unreported.
If we focus on severe circumvallate placenta, severe to the

Revised manuscript accepted for publication December 1, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog4022.2017
654 H. Takahashi
extent that it causes some perinatal morbidities, FMH may
be present in a significant proportion.
Six decades ago, when FMH was unrecognized, Mitchell
et al. [6] stated that circumvallate placenta may be associ-
ated with “neonatal anemia”. A direct evidence was lacking
whether permeability actually increased in this patient.
However, since abnormal placental trophoblastic invasion
underlies FMH, and circumvallate placenta has this pla-
cental pathology, we assume that circumvallate placenta, at
least partly, may have contributed to FMH, and not a mere
coincidence.
Peng et al. [1] concluded that obstetricians should be sus-
picious of “idiopathic FMH” at DFM. Circumvallate pla-
centa may be hidden among “idiopathic FMH”. Cautious
antenatal ultrasound may reveal circumvallate placenta [5].
Thus, a pregnant woman with antenatally diagnosed cir-
cumvallate placenta may be asked to pay much attention to
fetal movement. Further study is needed to confirm our
suggestion.
References
[1] Peng X., Liu C., Peng B.: “Idiopathic massive fetomaternal hemor-
rhage in the third trimester of pregnancy causing neonatal death”.
Clin. Exp. Obstet. Gynecol., 2016, 43, 284.
[2] O’Leary B.D., Walsh C.A., Fitzgerald J.M., Downey P., McAuliffe
F.M.: “The contribution of massive fetomaternal hemorrhage to an-
tepartum stillbirth: a 25-year cross-sectional study”. Acta Obstet. Gy-
necol. Scand., 2015, 94, 1354.
[3] Umazume T., Yamada T., Morikawa M., Ishikawa S., Kojima T., Cho
K., et al.: “Occult fetomaternal hemorrhage in women with patho-
logical placenta with respect to permeability”. J. Obstet. Gynaecol.
Res. 2016, 42, 632.
[4] Benirschke K., Burton G.J., Baergen R.N.:“Placental shape aberra-
tions”. In: Benirschke K., Burton G.J., Baergen R.N.(ed). Pathology
of the Human Placenta. 6th ed. Berlin: Springer, 2012. 377.
[5] Taniguchi H., Aoki S., Sakamaki K., Kurasawa K., Okuda M., Taka-
hashi T., et al.: “Circumvallate placenta: associated clinical mani-
festations and complications-a retrospective study”. Obstet. Gynecol.
Int., 2014, 2014, 986230.
[6] Mitchell A.P.B., Anderson G.S., Russell J.K.: “Perinatal death from
foetal exsanguination”. Br. Med. J., 1957, 1, 611.
Reply by Peng et al.
Most cases of FMH are idiopathic or “silent” and they
may remain undiagnosed until delivery. The real reason is
unknown but it was speculated that fetal placental blood
vessels have higher blood pressure than the intervillous
space; if there is disruption of the maternal-fetal barrier,
hemorrhage will occur from the fetus to the maternal cir-
culation. It is meaningful to refer that FMH is more likely
to occur in placentas with “pathological permeability”, an
“abnormal trophoblastic invasion” underlies this process,
in which fetal blood is more likely to transfer to maternal
blood. We cannot agree more that it is reasonable. The risk
factors causing FMH include blood type incompatibility,
abdominal trauma, amniocentesis, the external cephalic in-
version, hypertensive disorders such as preeclampsia, pla-
cental abruption or placental, and umbilical cord
abnormalities, such as choriocarcinoma and chorioan-
giomas. The author presented one rare FMH case with cir-
cumvallate placenta, suggesting that circumvallate
placenta, an “abnormal trophoblast invasion”-related con-
dition, may be associated with FMH. It is reasonable and
remarkable. We also speculated that “potential placental ab-
normalities” might be the main cause of idiopathic FMH.
Thus, obstetricians should be suggested to pay much at-
tention to a pregnant woman with any placental pathology.
Corresponding Author:
H. TAKAHASHI, MD, PHD
Department of Obstetrics and Gynecology
Jichi Medical University
3311-1 Shimotsuke
Tochigi 329-0498 (Japan)
e-mail: hironori@jichi.ac.jp
CEOG Clinical and Experimental
Obstetrics & Gynecology
Environmental influence on predisposing genes for
holoprosencephaly in monochorionic diamniotic twins
L. Marin, A. Andrisani
Department of Women and Child Health, Gynecologic and Obstetric Unit, University of Padua, Padua (Italy)
Dear Editor,
Holoprosencephaly (HPE) is the most common structural
anomaly of the developing forebrain characterized by in-
complete separation of the prosencephalon. It is categorized
into three main subtypes in order of decreasing severity: alo-
bar, semilobar, and lobar. HPE encompasses a phenotypic
spectrum that ranges from failure to partition the forebrain
into hemispheres and cyclopia, to mild midfacial anomalies
that occur without forebrain involvement. Clinical manifes-
tations may include facial dysmorphic features, cognitive im-
pairment, seizures, motor impairment, and ophthalmologic
findings. Defects associated with HPE occur very early in em-
bryonic development, at the stage of gastrulation. Incidence
of HPE is 1:250 during embryogenesis [1], but it decreases to
1:16.000 among live deliveries due to the associated high
rates of spontaneous abortion [2].
Case reports on HPE occurring in twins are few, even more
rare in cases of monochorionic twins. Zhang et al. [3] de-
scribed a singular case of discordant alobar HPE in mono-
chorionic diamniotic twins with normal karyotype. The
patient was a 21-year-old woman, gravida 1, para 0. At 24+6
weeks’ gestation, severe brain malformations (alobar HPE,
thalamus fusion, single brain tissue, hydrocephalus),
cheiloschisis, and nose abnormality were detected in one twin
by ultrasonography, while the other one was normal. Amnio-
centesis was performed to obtain amniotic fluid specimen of
normal twin and karyotyping showed 46, XY. Patient denied
any drug abuse, chronic disease, infections, and relevant fa-
milial or obstetric history. At 31 weeks’ gestation polyhy-
dramnios was found in the abnormal twin. Pregnancy was
complicated with severe preeclampsia, intrahepatic cholesta-
sis of pregnancy, and threatened premature labor. At 34+5
weeks’ gestation, a cesarean section was performed. Mal-
formed twin was a male weighing 2,829 grams and died im-
mediately after birth. External examination revealed frontal
bossing, hydrocephaly, hypotelorism of eyes, flat nasal
bridge, macroglossia, and cheilo/palatoschisis. Karyotyping
Revised manuscript accepted for publication December 8, 2016
Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog4323.2017
by G-banding of fetal umbilical blood specimen at birth in
both twins verified 46, XY. Autopsy was not performed be-
cause of the lack of parental consent. The authors did not sug-
gest a possible cause responsible for this singular case report.
Indeed, the etiology of HPE is heterogeneous and still in-
completely understood. It involves a complex interplay be-
tween various environmental factors, syndromic disorders,
chromosomal anomalies, and heterozygous variants in sev-
eral HPE-associated genes [1]. In non-syndromic HPE (30-
40%) heterozygous mutations or small copy number variation
of few genes (SHH, MIM 600725, SIX3, MIM 603714,
ZIC2, MIM 603073, and TGIF, MIM 602630) are found in
approximately 30% of cases, while mutations in any of over
ten additional genes are detected with a much less frequency
[4]. In literature, variable expressivity or incomplete pene-
trance have been reported in multiple cohorts, suggesting a
complex inheritance [5]. Studies on gene-environment inter-
actions in mice have supported this scenario of complex in-
heritance in HPE [6]. In fact, Capobianco et al. [7] highlighted
the teratogenic effect of hyperglycemia, hyperinsulinemia,
and insulin-resistance (IR), describing a case of alobar HPE
and trisomy 13 with maternal gestational diabetes mellitus
(GDM) in dietary treatment.
GDM is defined as any degree of glucose intolerance with
an onset or first recognition during pregnancy. It is a common
complication of pregnancy, associated with a high incidence
of hypertensive disorders, like gestational hypertension, pre-
eclampsia, and eclampsia, and an increase risk of excessive
fetal growth, polyhydramnios, and preterm labor. Physiolog-
ical pregnancy is characterized by elevated levels of hor-
mones and other proteins having insulin-antagonistic effects,
that increase IR in peripheral tissues, inducing compensatory
hyperinsulinemia. Therefore, pregnancy can unmask GDM
in preexisting conditions associated with IR, like obesity, pre-
diabetes, or polycystic ovary syndrome (PCOS).
PCOS is recognized as the most common endocrine-meta-
bolic disorder of reproductive-aged women, characterized by
oligo-anovulation, polycystic ovaries, clinical/ biochemical
656 L. Marin, Andrisani
hyperandrogenism, and often associated with IR even in nor-
moweight women [8, 9]. PCOS patients have also multiple
risk factors that may lead to several complications during
pregnancy, as for example obesity, impaired glucose toler-
ance or diabetes, thyroid disorders, pro-oxidative status, rel-
ative hyperaldosteronism, and hypertension [10]. Indeed, the
prevalence of GDM, preeclampsia, premature delivery, and
caesarean section and other cardiovascular events during
pregnancy is significantly increased in PCOS patients com-
pared with healthy women [11].
In pregnancy liver induction of lower sex hormone binding
proteins reduce bioavailability of androgens; however, this
decrease is impaired in pregnant PCOS women. Controversial
data are available concerning the pathophysiological effects
of hyperandrogenism on the development of complications
during pregnancy. Some studies showed that women, who
were more hyperandrogenic and more insulin resistant during
pregnancy, had a worse pregnancy outcome compared with
those showing lower degree of hyperandrogenism and IR
[12]. Some authors reported increased androgen concentra-
tions in women with high blood pressure and preeclampsia,
suggesting a certain pathological role of androgens in the he-
modynamic changes responsible for the preeclampsia devel-
opment [13, 14].
In the singular case of discordant alobar HPE in mono-
chorionic diamniotic twins with normal karyotype presented
by Zhang et al. [3], we can hypothesize the presence of some
genetic defects partially unmask by maternal condition.
Monozygotic twins originate from a single zygote, and share
the same genetic material and similar intrauterine environ-
ment. It is well known that mutation carriers display highly
variable clinical presentation, because the penetrance and ex-
pressivity of a predisposing mutation is graded by genetic or
environmental modifiers. A forthright GDM could have
caused the development of the syndrome in both twins. In-
stead, PCOS, IR, and an undiagnosed mild hyperglycemia
could have led to a hostile intrauterine environment and only
one fetus could have developed the disease due to different
penetration of HPE.
Cosmi et al. [15] previously evaluated dimensions and vol-
ume of yolk sac in pregnancies complicated by diabetes using
two- and three-dimensional sonography and they observed a
faster growth and a more precocious involution of the yolk
in diabetic pregnancies. The authors speculated that the al-
teration in the speed of growth of the structure could be cor-
related to the diabetic environment. It could be interesting to
evaluate if this finding is present even in women with hyper-
insulinemia and IR, and if it this could be used as a marker of
hostile intrauterine environment. Further studies are needed to
better understand the function of additional genes and gene-
environment interactions. With the development of next-gen-
eration sequencing, more and more potential genetic causes
will be discovered to explore the discordant abnormalities in
monozygotic twins.
References
[1] Hilbrands R., Keymolen K., Michotte A., Marichal M., Cools F.,
Goossens A., et al.: “Pancreas and gallbladder agenesis in a newborn
with semilobar holoprosencephaly, a case report”. BMC Med. Genet.,
2017, 18, 57.
[2] Pallangyo P., Lyimo F., Nicholaus P., Makungu H., Mtolera M.,
Mawenya I.: “Semilobar holoprosencephaly in a 12month-old baby
boy born to a primigravida patient with type 1 diabetes mellitus: a case
report”. J. Med. Case Rep., 2016, 10, 358.
[3] Zhang J., Yang T., Wang X., Yu H.: “Successful management of dis-
cordant alobar holoprosencephaly in monochorionic diamniotic twins
with normal karyotype: a case report”. Clin. Exp. Obstet. Gynecol.,
2015, 42, 114.
[4] Weiss K., Kruszka P., Guillen Sacoto M.J., Addissie Y.A., Hadley D.W.,
Hadsall C.K., et al.: “In-depth investigations of adolescents and adults
with holoprosencephaly identify unique characteristics”. Genet. Med.,
2017, Jun 22. doi: 10.1038/gim.2017.68. [Epub ahead of print]
[5] Solomon B.D., Bear K.A., Wyllie A., Keaton A.A., Dubourg C., David
V., et al.: “Genotypic and phenotypic analysis of 396 individuals with
mutations in Sonic Hedgehog”. J. Med. Gen., 2012, 49, 473.
[6] Kietzman H.W., Everson J.L., Sulik K.K., Lipinski R.J.: “The terato-
genic effects of prenatal ethanol exposure are exacerbated by Sonic
Hedgehog or GLI2 haploinsufficiency in the mouse”. PLoS One, 2014,
19, 9.
[7] Capobianco G., Cherchi P.L., Ambrosini G., Cosmi E., Andrisani A.,
Dessole S.: “Alobar holoprosencephaly, mobile proboscis and trisomy
13 in a fetus with maternal gestational diabetes mellitus: a 2D ultra-
sound diagnosis and review of the literature”. Arch. Gynecol. Obstet.,
2007, 275, 385.
[8] Goodman N.F., Cobin R.H., Futterweit W., Glueck J.S., Legro R.S.,
Carmina E., et al.: “Clinical review: guide to the best practices in the
evaluation and treatment of polycystic ovary syndrome”. Endocr.
Pract., 2015, 21, 1415.
[9] Donà G., Sabbadin C., Fiore C., Bragadin M., Giorgino F.L., Ragazzi
E., et al.: “Inositol administration reduces oxidative stress in erythro-
cytes of patients with polycystic ovary syndrome”. Eur. J. Endocrinol.,
2012, 166, 703.
[10] Armanini D., Sabbadin C., Donà G., Andrisani A., Ambrosini G., Bor-
din L.: “Maternal and fetal outcomes in preeclampsia: interrelations
between insulin resistance, aldosterone, metabolic syndrome, and poly-
cystic ovary syndrome”. J. Clin. Hyertens. (Greenwich), 2015, 17, 783.
[11] Qin J.Z., Pang L.H., Li M.J., Fan X.J., Huang R.D., Chen H.Y.: “Ob-
stetric complications in women with polycystic ovary syndrome: a sys-
tematic review and meta-analysis.” Reprod. Biol. Endocrinol., 2013,
11, 56.
[12] Vejrazkova D., Vcelak J., Vankova M., Lukasova P., Bradnova O.,
Halkova T., et al.: “Steroids and insulin resistance in pregnancy”. J.
Steroid Biochem. Mol. Biol., 2014, 139, 122.
[13] Acromite M.T., Mantzoros C.S., Leach R.E., Hurwitz J., Dorey L.G.:
“Androgens in preeclampsia”. Am. J. Obstet. Gynecol., 1999, 180, 60.
[14] Carlsen S.M., Heimstad R.: “Androgen levels are associated with blood
pressure in pregnant women after term”. Acta Obstet. Gynecol. Scand.,
2012, 91, 232.
[15] Cosmi E., Piazze J.J., Ruozi Y., Anceschi M.M., La Torre R., Andrisani
A., et al.: “Structural-tridimensional study of yolk sac in pregnancies
complicated by diabetes”. J. Perinat. Med., 2005, 33, 132.
Corresponding Author:
A. ANDRISANI, M.D.
Department of Women and Child Health
Gynecologic and Obstetric Unit
University of Padua
Via Giustiniani, 2
35128 Padua (Italy)
e-mail: aleandrisani@yahoo.it
CEOG Clinical and Experimental
Obstetrics & Gynecology

Review Articles

Renal tumors in pregnancy: a systematic review

A. Pontis1, F. Congiu2, F. Sedda2, P. Litta4, A. De Lisa3, G.B. Melis2, S. Angioni2

1 U.O.C Obstetric and Gynecology, Ospedale San Francesco, Nuoro
2 Division of Gynecology and Obstetrics, 3 Division of Urology, Department of Surgical Sciences, University of Cagliari, Cagliari
4 Division of Obstetric and Gynecology, University of Padua, Padua (Italy)

Summary
Renal tumors are rarely observed in pregnancy, and their symptoms may mimic other pregnancy-related conditions, such as renal cal-
culi, cystitis, and pyelectasia. These tumors are generally characterized by magnetic resonance imaging and ultrasonography. The de-
cision to perform surgery depends on the stage of pregnancy. If a patient is diagnosed as having neoplasms in the first trimester, the best
choice is to operate as soon as possible. If the diagnosis is made in the second trimester, a better option would be to wait until the 28th
week of pregnancy to optimize a fetus’ chances of survival in case preterm labor occurs. If the mass is detected in the third trimester,
surgery should be postponed until the end of pregnancy. In this study, the authors reviewed articles on renal tumors during pregnancy
published from 1980 to 2015.

Key words: Renal tumors; Calculi; Cystitis; Pylectasia; Pregnancy.

cases), followed by angiomyolipoma (AML) (82 cases),

Introduction
Renal tumors rarely occur during pregnancy and are usu-
ally accompanied by symptoms that may mimic other preg-
nancy-related conditions (e.g., renal calculi, cystitis,
pyelectasia). Clinical presentation is characterized by a
common triad: palpable masses, flank pain, and hematuria.
The high occurrence of palpable mass during pregnancy is
probably caused by the frequent abdominal examinations
that patients undergo during the course of their pregnancy.
Flank pain is almost always present, but it can mimic renal
colic or pyelonephritis, thus delaying diagnosis. Hematuria
(macroscopic and/or microscopic) presents only in less than
half of pregnant patients. It occurs when a renal mass rup-
tures, and blood flows into the renal ducts [1].
Materials and Methods
The authors systematically examined the scientific literature,
including papers, reviews, and case reports, published by PubMed
from 1980 to 2015. The papers for review were searched using a
combination of the following keywords: renal cancer, pregnancy,
gestation, and renal mass. No language restrictions were imposed
on the selection.
Results
In the last 25 years, 107 diagnoses of renal tumors in
pregnancy have been reported in the literature. The most
common histotype is renal cell carcinoma (RCC) (88
nephroblastoma (Wilms’ tumor) (eight cases), metanephric
adenoma (three cases), fibroma (three cases), sarcoma (two
cases), lymphoma (three cases), juxtaglomerular cell tumor
(three cases), oncocytoma (two cases), reninoma (three
cases), carcinoid, angioma, mesoblastic nephroma, ter-
atoma, and renal pelvis carcinoma (one case each) [2]. The
diagnoses were performed by ultrasonography and mag-
netic resonance imaging (MRI) because pregnant women
are generally prohibited from undergoing CT scanning
given the high exposure of the fetus to radiation during such
tests. Ultrasound techniques are also more sensitive than
intravenous pielography to small renal masses. CT imaging
can be reserved for patients who present symptoms during
puerperium [3].
According to Loughlin [1], the treatment of renal tumors
typically involves radical nephrectomy. Schematically, the
treatment of masses depends on pregnancy period and should
be aimed at ensuring a good prognosis for both the mother
and the fetus [2]. If a patient is diagnosed as having neo-
plasms in the first trimester, the best choice is immediate op-
eration. If the condition is diagnosed in the second trimester,
a better option is to wait until the 28th week of pregnancy to
optimize the fetus’ chances of survival in case preterm labor
occurs. If the mass is detected in the third trimester, surgery
should be scheduled until after the pregnancy.
The type of surgery and other treatment options should be
evaluated in accordance with individual cases and prognoses.
In certain situations wherein benefits outweigh risks, la-

Revised manuscript accepted for publication May 2, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3707.2017
658 A. Pontis, F. Congiu, F. Sedda, P. Litta, A. De Lisa, G.B. Melis, S. Angioni
paroscopy can be performed. The advantages of laparoscopy
are that it involves minimally invasive procedures, reduces
morbidity, decreases pain, entails a short recovery period,
and enables the fast resumption of regular bowel and bladder
function [3]. Despite its benefits, however, it also presents
risks. Difficulties may be encountered in implementing tech-
niques because of a large pregnant uterus; the pregnant uterus
may be damaged, and abdominal pressure may occur,
thereby reducing the amount of blood pumped back to the
heart and decreasing placental perfusion. Given the different
histotypes, clinical presentations, and prognoses identified
in the reviewed papers, the present authors organized the
discussion according to the types of tumors diagnosed.
Renal cell carcinoma (RCC)
RCC is the most frequently diagnosed renal tumor in
pregnancy. Numerous findings support the idea that preg-
nancy-related hormonal changes (e.g., high estrogen and
progestin levels) stimulate renal cell proliferation. Renal
cells have both estrogen and progestin receptors. At the
same time, the hyperfiltration of kidneys during pregnancy
causes nephrons to develop glomerulosclerosis, thus in-
creasing their susceptibility to carcinogenesis [2]. RCCs
double in volume in 500 days, with the tumors demon-
strating malignancy but a low replication rate. On the basis
of histological characteristics and genetic alterations, the
Heidelberg classification (1997) identifies four groups of
RCCs: common, papillary, chromophobe, and collecting
duct carcinomas [4]. Clear cell carcinoma is the most com-
mon of the four types of RCCs. It can also be one of the tu-
mors that grow as a symptom of von Hippel-Lindau (VHL)
syndrome, which is an autosomal disorder that induces the
development of several benign and malignant tumors in
various body districts. These tumors include hemangio-
blastomas of the retina, cerebellum, brain stem, and spinal
cord, adenocarcinomas of the lymphatic sac, cysts or tu-
mors of the pancreas, and renal lesions. In pregnant women
with a familial history of VHL disease, ultrasound screen-
ing is a suitable diagnostic procedure given that such
screening avoids the risk of puncturing the hypervascular
tumor of the spinal cord during the administration of
epidural anesthesia . This tumor can be very aggressive and
spreads early with diffuse metastasis [5].
Chromophobe RCC is a rare condition, with only ap-
proximately 50 cases described in the literature. Its main
symptoms are microscopic hematuria (47% of cases),
which is present in less than half of diagnosed patients, hy-
pertension (18%) that is often be poorly controlled by med-
ical therapy and mimics preeclampsia, and renal masses
(88% of patients) that are difficult to identify because of
the large dimensions of the uterus [9, 10]. The prognosis
for this disease is better than that for clear cell carcinoma,
with a five-year survival of 92% (chromophobe RCC)
against 50% (clear cell carcinoma) [6].
Collecting duct renal cancer is considerably more malig-
nant than other renal cancers. It often occurs in young pa-
tients, and at diagnosis, metastasis has often already spread.
The first symptoms are usually bone metastasis and weight
loss [7].
Mucinous tubular and spindle cell carcinoma is a low-
grade variant of RCC. It is the rarest type and presents non-
specific characteristics during MRI and ultrasound
screening, thus leading to difficult diagnoses [8]. During
the studied period, 88 pregnant patients were diagnosed
with renal cancer, 29 of whom were diagnosed from 2000
to 2015. The gestational age at diagnosis was identified in
24 of the 29 cases. Specifically, five, 13, and three patients
were diagnosed in the first, second, and third trimesters, re-
spectively. Tumor localization was specified in 12 of 29
cases: in eight and four of the subjects, the tumors were on
the right and left kidneys, respectively. Applied therapy was
specified in 27 cases: 18 patients underwent laparotomic
nephrectomy, and nine were treated laparoscopically. Mode
of delivery was indicated in 22 cases: ten patients delivered
vaginally; ten underwent a cesarean section (CS) (one of
the patients also underwent hysterectomy directly after CS),
and two underwent medical abortion. From 1980 to 2015,
at least 18 patients gave birth to healthy infants by vaginal
delivery [1, 8-16].
Angiomyolipoma (AML)
AML is a benign amartomatous tumor composed of
blood vessels, smooth muscle cells, and fat tissue. In 30%
of cases, AML is associated with tuberous sclerosis (TS),
whereas the remaining 70% occurs in a sporadic manner
[17]. It is a rare condition, accounting for only 3% of renal
solid masses and presenting in only 0.3% of the general
population. AML is usually found in kidneys but may also
occur in the spleen, liver, uterus, or tubes [18].
TS is a neurocutaneous hereditary disease that can af-
fect almost every tissue in the body. The most frequent lo-
calizations are the skin, brain, kidneys, and heart. TS
incidence varies from cosmetic skin alterations to severe
organ damage. About 80% of patients with TS present
AML [19]. In many cases of TS and AML association,
AML and lymphangio-leiomyomatosis coexist; the latter is
a lung disease characterized by shortness of breath, pneu-
mothorax, and fatigue and causes severe respiratory im-
pairment [20]. When AMLs are associated with TS, they
are often large (generally > 4 cm), multiple, and suscepti-
ble to ruptures and massive bleeding [3]. The high risk of
bleeding is attributed to increased plasma volume during
pregnancy [21].
AML is also strongly correlated with estrogen exposure,
both in pregnancy and in combined contraceptive therapy.
It is typical of women of fertile age. AML causes morbid-
ity in two ways: bleeding and renal failure [17]. It is fre-
quently asymptomatic. In symptomatic patients, for whom
the possibility of AML rupture is a necessary consideration,
symptoms are usually flank pain, palpable masses (often
 Renal tumors in pregnancy: a systematic review
detected in pregnancy because of the high number of ab-
dominal palpations), and hematuria (due to intracapsular or
retroperitoneal rupture of aneurysms generated by the ves-
sels of the tumor) [18].
The risk of ruptures increases with tumor dimensions (a
tumor > 4 cm can be regarded as presenting high risk), co-
existence with TS, and other symptoms. The presence of fat
enables ultrasound, MRI, and CT imaging of AML [17]. The
disease is effectively diagnosed by ultrasonography because
the findings of this technique are strongly suggestive (hy-
perechoic mass). MRI is then used as a confirmatory screen-
ing approach, and a biopsy is rarely performed [21]. In
asymptomatic patients with AMLs less than 4 cm, treatment
is unnecessary, and ultrasound control every 6 to 12 months
is sufficient [22]. Interventions for patients requiring treat-
ment can involve surgery, with partial or radical nephrec-
tomy, and endovascular treatment that entails transcatheter
embolization with liquid agents (e.g., NBCA), polyvinyl al-
cohol particles, gel foam, and metallic coils; these treat-
ments constitute a conservative approach to renal
parenchyma [3]. Therapeutic embolization can be used in
different cases: (a) in patients whose AMLs are larger than
4 cm, for which preventive intervention is necessary; b) in
acute bleeding of AML and in the stabilization of renal func-
tion when TS is present. The most common approach to
transarterial embolization is the femoral technique because
it enables the easiest access to the femoral artery. When this
treatment is contraindicated, however, transradial em-
bolization is a suitable alternative. Instances of contraindi-
cation are the presence of acute aorto-renal angles, renal
artery stenosis, or severe peripheral arterial stenosis, and
when avoiding high exposure of the pelvic region to radia-
tions is advised (as is the case in pregnancy). Another alter-
native treatment is radical nephrectomy, which is considered
for patients who are hemodynamically unstable [23-30].
Reninoma
Reninoma differs from other tumors. In fact it is induced
by the hormonal secretion of renin, which then activates
the renin-angiotensin-aldosterone system. This activation
causes hypokalemia and severe hypertension that is unre-
sponsive to most medical treatments. Differential diag-
noses of pheochromocytoma include urine catecholamine
and metanephrine testing, which generate normal results
in the presence of reninoma [31]. Severe hypertension can
simulate preeclampsia and lead to abruptio placentae,
preterm delivery, and HELLP syndrome, thus threatening
the patient’s life.
Adult Wilms’ tumor
The present authors found six cases of adult Wilms’
tumor, which is one of the most common tumors in chil-
dren and rarely occur in adults. The criteria typically used
to diagnose a nephroblastoma are (1) primary renal neo-
plasms, (2) primitive blastomatous spindle or round cell
659
components, (3) the formation of abortive or embryonal tu-
bular or glomeruloid structures, (4) the absence of hyper-
nephromatous areas, (5) pictorial confirmation of
histological findings, and (6) > 15 years of age.
Standard treatment for adult Wilms’ tumor should consist
of radical nephrectomy accompanied by chemotherapy with
or without radiotherapy. During pregnancy, chemotherapy
and radiotherapy are contraindicated (primarily in the first
trimester because of teratogenic effects, but also in the sec-
ond and third trimesters because of effects on placental func-
tioning and fetal growth) [32]. Surgical timing should
adhere to Loughlin’s recommendations [1].
Metanephric adenoma
Metanephric adenoma derives from the same structure as
Wilms’ tumor and metanephric blastema, one of the em-
bryologic precursors to the development of the kidney.
Metanephric adenoma is thus considered the counterpart of
Wilms’ tumor. It can present not only with the classical
symptoms of renal tumors (hematuria, palpable masses, or
flank pain), but also with paraneoplastic syndromes, such as
polycythemia or hypercalcemia. With imaging findings
alone, differential diagnoses of RCC and Wilms’ tumor can
be very difficult, thus motivating the adoption of an ag-
gressive approach even with a benign tumor. Only two
cases of metanephric adenoma are described in the litera-
ture [33].
Sarcoma
Renal sarcomas are rare in pregnancy, as evidenced by
the only two cases reported in the literature. The first case
was a rhabdomyosarcoma, a malignant tumor of mes-
enchymal origin that presents with non-specific symptoms.
Only ten cases of rhabdomyosarcoma are found in the lit-
erature, both for pregnant and non-pregnant patients and
with no differences in incidence between men and women.
Typical presentations are weight loss, fatigue, hematuria,
and abdominal pain. It is generally aggressive and progno-
sis is worse in adults than in children [34]. No standardized
treatment is employed and the therapeutic approach should
be similar to the classical intervention suggested by Lough-
lin for solid renal masses, independent of histotype [1]. The
second sarcoma case during pregnancy involved a patient
who presented gross hematuria without signs of urinary in-
fection [35]. Both of these cases occurred during the third
trimester of pregnancy with a unilateral tumor; the patients
underwent simultaneous CS and total nephrectomy. Both
patients presented free margins of excisions and were dis-
ease free after several years of follow-up.
Oncocytoma
Oncocytoma is a benign tumor without invasion or
metastasis. It is usually accidentally diagnosed in asymp-
tomatic patients. The literature discusses two cases of on-
cocytoma in pregnant patients. In one of the patients, the
660 A. Pontis, F. Congiu, F. Sedda, P. Litta, A. De Lisa, G.B. Melis, S. Angioni
tumor presented with severe hypertension with superim-
posed preeclampsia, uncontrollable by therapy, thereby
leading to diffuse edema and acute pulmonary edema. This
tumor’s behavior was uncommon, driving the need for in-
tensive care and radical nephrectomy for the mother and
causing fetal death. Blood pressure normalized after the in-
tervention [36].
Teratoma
Only one case of mature cystic teratoma is reported in
the literature. The patient, at 25 weeks of gestation, was ini-
tially diagnosed with renal abscess. Ultrasound and clinical
examination indicated spontaneous abortion with pyometra.
The patient underwent nephrectomy and evacuation of re-
tained products of conception and developed sepsis. Diag-
nosing renal teratoma necessitates the presence of two
criteria: the tumor should unequivocally be determined as
being of renal origin and must show heterotopic organo-
genesis [37].
Choriocarcinoma
Choriocarcinoma is a highly aggressive cancer composed
of atypical cytotrophoblast and syncytiotrophoblast. In
women, it derives frequently from a previous pregnancy,
mainly molar. The main site of occurrence is inside the
uterus, deriving from eutopic pregnancies, followed by ovar-
ian and fallopian localization (derived from ectopic preg-
nancy). In rare cases, it has been described as a primary
tumor that occurs in other districts of the body. Metastasis is
more common in other organs than in genital ones. The lit-
erature describes only one case of renal choriocarcinoma (of
gestational origin), detected in a patient with three previous
full-term deliveries. At diagnosis, the tumor had already
spread by pulmonary metastasis. The uterus did not present
signs of the tumor. The patient underwent chemotherapy and
was disease free at two years of follow-up [38].
Conclusion
Although rare, renal tumors can threaten the lives of preg-
nant women and infants [18]. These conditions require close
examination upon presentation of flank pain, palpable
masses, hematuria, and severe hypertension. Accurate dif-
ferential diagnosis is also essential. Common diseases that
can be confused with renal tumors include urinary tract in-
fections (abscesses and pyelonephritis), renal calculi,
preeclampsia, and renal endometriosis [39-41] .Given their
generally poor progression, the early diagnosis of renal tu-
mors permits a less invasive treatment, thus preserving renal
function and securing positive pregnancy outcomes. With the
evolution of laparoscopic and endovascular techniques in re-
cent years, minimally invasive therapies are now available
and minimize maternal and fetal risks [19, 42, 43].
References
[1] Loughlin K.R.: “The management of urological malignancies during
pregnancy”. Br. J. Urol., 1995, 76, 639.
[2] Boussios S., Pavlidis N.: “Renal cell carcinoma in pregnancy: a rare
coexistence”. Clin. Transl. Oncol., 2014, 16, 122.
[3] Idilman I.S., Vesnic S., Cil B., Peynircioglu B.: “Giant renal artery
pseudoaneurysm caused by rupture of renal angiomyolipoma fol-
lowing pregnancy: endovascular treatment and review of the litera-
ture.” Saudi J. Kidney Dis. Transpl., 2014, 25, 385.
[4] Szendrôi A., Rusz A., Székely E., Riesz P., Kelemen Z.: “Renal
tumor causing haematuria and sepsis”. Pathol. Oncol. Res., 2003, 9,
246.
[5] Simon I., Rorive S., Kirkpatrick C., Roumeguere T., Nortier J.L.:
“Clear cell renal carcinoma presenting as a bleeding cyst in preg-
nancy: inaugural manifestation of a von Hippel-Lindau disease”.
Clin. Nephrol., 2008, 69, 224.
[6] Guven S., Guvendag Guven ES., Islamoglu E., Gunalp GS., Ozen
H.: “Successful management of chromophobe type renal cell carci-
noma in pregnancy”. Aust. N. Z. J. Obstet. Gynaecol., 2004, 44, 362.
[7] Mancuso A., Macrì A., Palmara V, Scuderi G., Grosso M., Famu-
lari C.: “Chromophobe renal cell carcinoma in pregnancy: case re-
port and review of the literature”. Acta Obstet. Gynecol. Scand.,
2001, 80, 967.
[8] Noon AP., Smith D.J., McAndrew P.: “Magnetic resonance imaging
characterization of a mucinous tubular and spindle cell carcinoma
of the kidney detected incidentally during an ectopic pregnancy”.
Urology, 2010, 75, 247.
[9] Gnessin E., Dekel Y., Baniel J.: “Renal cell carcinoma in preg-
nancy”. Urology, 2002, 60, 1111.
[10] Gladman M.A., MacDonald D., Webster J.J., Cook T., Williams G.:
“Renal cell carcinoma in pregnancy”. J. R. Soc. Med., 2002, 95, 199.
[11] Pearson G.A., Eckford S.D.: “Renal cell carcinoma in pregnancy”.
J. Obstet. Gynaecol., 2009, 29, 53.
[12] Lee D., Abraham N.: “Laparoscopic radical nephrectomy during
pregnancy: case report and review of the literature”. J. Endourol.,
2008, 22, 517.
[13] Fynn J., Venyo A.K.: “Renal cell carcinoma presenting as hyperten-
sion in pregnancy”. J. Obstet. Gynaecol., 2004, 24, 821.
[14] O’Connor JP., Biyani C.S., Taylor J., Agarwal V., Curley P.J.,
Browning A.J.: “Laparoscopic nephrectomy for renal-cell carcinoma
during pregnancy”. J. Endourol., 2004, 18, 871.
[15] Casella R., Ferrier C., Giudici G., Dickenmann M., Giannini O.,
Hosli I., et al.: “Surgical management of renal cell carcinoma during
the second trimester of pregnancy.” Urol Int. 2006, 76, 180.
[16] Ceglowska A., Michalski A.: “Renal cell carcinoma during preg-
nancy”. Ginekol. Pol., 2004, 75, 145.
[17] Abrams J., Ye DC, Clark T.W.: “Transradial embolization of a bleed-
ing renal angiomyolipoma”. Vasc. Endovascular Surg., 2011, 45,
470.
[18] Pontis A., Piras B., Meloni A., De Lisa A., Melis G.B., Angioni S.:
“Rupture of renal angiomyolipoma in pregnancy”. J. Obstet. Gy-
naecol., 2013, 33, 628.
[19] Ferianec V., Gábor M., Caňo M., Papcun P., Holoman K.: “Severe
retroperitoneal haemorrhage in the first trimester of a multiple preg-
nancy after spontaneous rupture of renal angiomyolipoma”. Arch.
Gynecol. Obstet., 2013, 288, 1193.
[20] Meredith W.T., Levine E., Ahlstrom N.G., Grantham J.J.: “Exacer-
bation of familial renal lymphangiomatosis during pregnancy”. AJR
Am. J. Roentgenol., 1988, 151, 965.
[21] Gyimadu A.O., Kara O., Basaran D., Esinler I.: “Conservative man-
agement of a retroperitoneal hemorrhage following a ruptured renal
angiomyolipoma in pregnancy”. J. Obstet. Gynaecol. Res., 2011, 37,
156.
[22] Kontos S., Politis V., Fokitis I., Lefakis G., Koritsiadis G., Simaio-
foridis V., et al.: “Rupture of renal angiomyolipoma during preg-
nancy: a case report.” Cases J., 2008, 1, 245.
[23] Zapardiel I., Delafuente-Valero J., Bajo-Arenas J.M.: “Renal an-
 Renal tumors in pregnancy: a systematic review
giomyolipoma during pregnancy: review of the literature”. Gynecol.
Obstet. Invest., 2011, 72, 217.
[24] Raft J., Lalot J.M., Meistelman C., Longrois D.: “Renal angiomy-
olipoma rupture during pregnancy”. Gynecol. Obstet. Fertil., 2006,
34, 917.
[25] Morales J.P., Georganas M., Khan M.S., Dasgupta P., Reidy J.F.:
“Embolization of a bleeding renal angiomyolipoma in pregnancy:
case report and review”. Cardiovasc. Intervent. Radiol., 2005, 28,
265.
[26] Dumas J.P., Colombeau P., Steiner E., Jouvie J. “Tumors of the kid-
ney and pregnancy.” Ann. Urol. (Paris), 1984, 18, 339.
[27] Capobianco G., Angioni S., Dessole M., Cherchi P.L.: “Cesarean
section: to be or not to be, is this the question?” Arch. Gynecol. Ob-
stet., 2013, 288, 461.
[28] Ponsot Y., Blouin D., Carmel M.: “Hemorrhagic rupture of an an-
giomyolipoma during pregnancy. Review of the literature apropos
of a case”. Prog. Urol., 1994, 4, 578.
[29] Fernández Arjona M., Mínguez R., Serrano P., Sanz J., Teba F.,
Peinado F., et al.: “Rapidly-growing renal angiomyolipoma associ-
ated with pregnancy.” Actas Urol. Esp., 1994, 18, 755.
[30] Russo O., Beneventano G., Trovato R., Pappalardo F., Gangemi P.,
Ciuni S.: “Angiomyolipoma of the kidney”. Ann. Ital. Chir., 1995,
66, 109.
[31] Diker-Cohen T., Abraham S.B., Rauschecker M., Papadakis G.Z.,
Munir K.M., Brown E., et al.: “Reninoma presenting in pregnancy”.
J. Clin. Endocrinol. Metab., 2014, 99, 2625.
[32] Rodrigues F.A., Ribeiro E.C., Maroccolo Filho R., Silva E.A., Diaz
F.A.: “Adult Wilms tumor during gestational period”. Urology, 2009,
73, 929.
[33] Lerut E., Roskams T., Joniau S., Oyen R., Achten R., Van Poppel
H., et al.: “Metanephric adenoma during pregnancy: clinical pres-
entation, histology, and cytogenetics”. Hum. Pathol., 2006, 37,
1227.
[34] Meir K., Wygoda M., Reichman O., Gofrit ON., Pizov G.: “Puer-
peral renal rhabdomyosarcoma: case report and review of the litera-
ture”. Urol. Oncol., 2006, 24, 40.
[35] Bettendorf O., Bierer S., Köhler G., Piechota H.J., Mesters R.M.,
Klockenbusch W.: “Neoplasia of the kidney—a rare event during
pregnancy”. Eur. Urol., 2006, 50, 148.
661
[36] Torres R., Borges A., Campos A.: “Renal oncocytoma in preg-
nancy—an unusual presentation of secondary hypertension”. Rev.
Port. Cardiol., 2012, 31, 385.
[37] Nzegwu M.A., Aligbe J.U., Akintomide G.S., Akhigbe A.O.: “Ma-
ture cystic renal teratoma in a 25-year-old woman with ipsilateral
hydronephrosis, urinary tract infection and spontaneous abortion”.
Eur. J. Cancer Care (Engl.), 2007, 16, 300.
[38] Vereczkey I., Csernák E., Olasz J., Kuronya Z., Szentirmay Z., Toth
E.: “Renal choriocarcinoma: gestational or germ cell origin?” Int. J.
Surg. Pathol., 2012, 20, 623.
[39] Angioni S., Nappi L., Pontis A., Sedda F., Luisi S., Mais V., et al.:
“Dienogest. A possible conservative approach in bladder endometrio-
sis. Results of a pilot study”. Gynecol. Endocrinol., 2015, 31, 406.
[40] Pontis A., Arena I., Angioni S.: “Umbilical endometriosis primary
site without pelvic endometriosis and previous surgery: a case re-
port”. Giornale Italiano di Ostetricia e Ginecologia, 2014, 36, 336.
[41] Locci R., Nisolle M., Angioni S., Foidart J.M., Munaut C.: “Ex-
pression of the gamma 2 chain of laminin-332 in eutopic and ectopic
endometrium of patients with endometriosis”. Reprod. Biol. En-
docrinol., 2013, 11, 94.
[42] Liliana M., Alessandro P., Giada C., Luca M.: “Single-port access
laparoscopic hysterectomy: a new dimension of minimally invasive
surgery”. J. Gynecol. Endosc. Surg., 2011, 2, 11.
[43] Mencaglia L., Mereu L., Carri G., Arena I., Khalifa H., Tateo S., et
al.: “Single port entry - are there any advantages?” Best. Pract. Res.
Clin. Obstet. Gynaecol., 2013, 27, 441.
Corresponding Author:
S. ANGIONI, M.D., PhD
Department of Surgical Sciences
University of Cagliari, Italy
Azienda Ospedaliero Universitaria Blocco Q
SS554, 09124 Monserrato (Italy)
e-mail: sangioni@yahoo.it
CEOG Clinical and Experimental
Obstetrics & Gynecology

Multiple sclerosis management in pregnancy

L. Sahin, Y. Ehi
Department of Obstetrics and Gynecology, Kafkas University Medical School, Kars (Turkey)

Summary
Multiple sclerosis (MS) is the most common chronic neurologic disability in young adults in their childbearing ages of 20 to 45 years.
The disease affects especially women that is worthy of discussion among pregnancy-related conditions in a woman with MS. Prenatal
counseling to discuss the safety of medications in pregnancy, the antepartum period, along with what the patient can expect during
birth, and the postpartum period will be discussed.

Key words: Multiple sclerosis; Pregnancy; Antepartum; Postpartum.

Immunological background of MS
Neuroinflammation is involved in several neurodegenera-
tive disorders including multiple sclerosis (MS) and emerg-
ing evidence indicates that it constitutes a critical process
that is required for the progression of neurodegeneration. Mi-
croglial activation constitutes a central event in neuroin-
flammation with microglial cells being the main source of
reactive oxygen species (ROS) and nitrogen species, gluta-
mate, and TNF-α [1, 2]. MS is a neurodegenerative autoim-
mune disorder in which axon demyelination lesions develop
in the central nervous system and T-cell-mediated response
has been known to be involved for more than a decade [3].
MS is characterised by the progressive loss of neurologi-
cal function caused by the destruction of the axonal myelin
sheath in several areas of the brain and the spinal cord, which
is mediated, mainly, by self-reactive CD4+ T-cells [4].
The loss of myelin is manifested in clinical symptoms
such as: paralysis, muscle spasms, optic neuritis, and neuro-
pathic pain [5]. The pathological features of MS lesions in-
volve: blood-brain barrier (BBB) permeability, myelin sheath
destruction, axonal damage, glial scar formation, and pres-
ence of inflammatory cells infiltrated into the CNS [6].
Experimental model of MS
The most used and accepted animal model equivalent of
MS is experimental autoimmune encephalomyelitis (EAE),
which corresponds to an induced autoimmunity in mice [7].
The administration of myelin-derived antigens in an im-
munogenic context induces the activation of self-reactive T-
cells that are specific for myelin antigens, mediating myelin
destruction. This induced autoimmunity is characterised by
focal areas of demyelination along the brain and spinal
cord, with axonal loss that results in ascending paralysis,
affecting first the tail and then the hind limbs [8]. This ini-
tial neuroinflammatory process results in BBB disruption
and the entrance of leukocytes into the CNS parenchyma.
Once T-cells enter the CNS, they are re-stimulated by res-
ident antigen-presenting cells (APCs), such as astrocytes,
microglia or infiltrated APCs such as dendritic cells and
macrophages [9].
Th17: main culprit in MS
The experimental autoimmune EAE model of MS provided
the first clues to the possibility that other T cell effector func-
tions, beyond those attributed to the Th1 and Th2 subsets,
could be contributing to the onset and progression of MS [10].
Researchers have further explored the link between autoim-
munity and environmental factors by looking at the effect of
a high salt diet, such as is seen in a typical “Western diet”, in
the pathology of autoimmunity, specifically Th17 cells [11].
Vitamin D can act as another source of Th17 regulation, as the
vitamin D receptor is induced in Th17 cells [12]. A metabo-
lite of vitamin A, retinoic acid, has also come into the spot-
light as a potent attenuator of immune function and has been
shown to have effects on T cell differentiation and function
[13]. Human blood-brain barrier endothelial cells were found
to express receptors for IL-17 and IL-22, thus making it pos-
sible for IL-17 and IL-22 to disrupt blood brain barrier junc-
tions [14]. It was also found that human Th17 lymphocytes
were able to migrate past blood-brain barrier-associated cells,
where they continued to promote inflammation through
CD4+ T cell recruitment and inflammatory cytokine produc-
tion [14]. With regards to the role of microbiota in the pre-
vention or progression of EAE and MS, reports have thus far
been varied. Researchers have shown that the use of specific
probiotic mixtures is able to suppress EAE development
though Th1/Th17 polarization inhibition, and increases

Revised manuscript accepted for publication November 9, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3420.2017
 L. Sahin, Y. Ehi
Foxp3+ Treg numbers and IL10 production [15]. Yokote et
al. also demonstrated that the administration of antibiotics al-
tered gut flora composition and ameliorated EAE develop-
ment through a possible invariant natural killer cell-Th17
interaction-dependent mechanism [16].
MS symptoms
MS is caused by an autoimmune response involving
macrophages and cytotoxic T cells that cause a chronic in-
flammatory state resulting in damage to the myelin sheath
around nerve cells. The demyelination of axons causes both
acute and chronic lesions to develop in the central nervous
system altering the conduction of electrical impulses to mus-
cle groups. Varying MS symptoms occur based on the loca-
tion of lesions such as pain, bowel and bladder dysfunction,
seizures, sexual dysfunction, discoordination-tremor, vertigo,
and sensory changes. Moreover vision, speech, muscle
strength, and sensation are commonly affected by MS. Phys-
ical symptoms can be accompanied by memory loss, cogni-
tive impairment, depression, mood swings, and fatigue [17,
18]. Infection, stress, pregnancy, postpartum period, fatigue,
heat, and heavy metal exposure may cause exacerbation of
MS symptoms [19].
Diagnosis
Although onset of MS is uncommon during pregnancy,
women’s healthcare providers require an understanding of di-
agnostic criteria to better discuss prognosis and pregnancy-
related concerns. In addition to clinical symptoms and physical
examination finding (Lhermitte’s sign, dysesthesias, Tic
douloureux), the diagnosis includes evidence of MRI lesions
in at least two places in the central nervous system, at least
one month apart, and differential diagnoses are ruled out. Over
half of MS women had MRI changes in the last trimester of
pregnancy and within four to 12 weeks of postpartum. An in-
crease in new or enlarging MRI lesions postpartum was re-
ported in several studies [20]. Differential diagnoses include
migraine, cerebral neoplasms, nutritional deficiencies of vita-
min B12 or copper, compressive lesions of the spinal cord,
human immunodeficiency virus, syphilis, lupus, or psychiatric
disease.
MS in pregnancy
Prenatal counseling
MS is believed to be caused by a combination of several
factors including immunologic, genetic, and environmental
factors. Previously, pregnancy was discouraged in women
with MS; however, recent studies have shown pregnancy as
potentially having a beneficial role on MS relapse rates with
no effects to long-term progression of the disease [21, 22].
Babies of mothers with MS have a higher risk (4%) of de-
veloping the disease compared with the general population
663
(0.1%) [23]. If both parents are affected by the disease, the
risk for offspring increases to 30.5% [24]. Currently, there
are no prenatal diagnostic tests that predict MS.
Antenatal care in MS
Preterm labor rates are not higher in women with MS; if
MS woman have impaired sensation she may not perceive
preterm contractions or pelvic pressure [25, 26]. If preterm
labor requires tocolysis, calcium channel blockers should be
used. Caution is used with administration of magnesium sul-
fate, which causes fatigue and weakness. When strict bed
rest is prescribed, venous thromboembolism prophylaxis is
recommended until the woman is able to ambulate.Women
with MS have increased risk of urinary tract infection during
pregnancy that may cause exacerbation of the woman’s MS
symptoms. Recommended routine monitoring for urinary
tract infection and chronic antibiotic prophylaxis may be ap-
propriate [26].
Medical therapy of pregnant women with MS
Pregnancy appears to have no influence on the progres-
sion of disability in MS. Confavreux et al. reported that the
frequency of relapses, decreased by 80% up to the third
trimester, increased in the immediate postpartum period, and
returned to pre-pregnancy levels at six months [27]. Eight
therapies are currently approved for treatment of MS: glati-
ramer acetate (GA), subcutaneous interferon b (IFNb)-1a,
intramuscular IFNb-1a, subcutaneous IFNb-1b, fingolimod,
natalizumab, teriflunomide, and mitoxantrone. Based on cur-
rent literature data, IFN-β and glatiramer acetate (GA) ap-
pear to be most suitable for use up until the time of confirmed
pregnancy. They are not associated with teratogenic risk or a
higher risk of miscarriage. Relapses during pregnancy can be
treated with corticosteroids but caution is advised prior to ges-
tational week 12 because of the risk of cleft palate. In the case
of severe relapse in the first trimester, the preferred treatment
is prednisolone as it is inactivated in the placenta, because
only about 10% reaches the fetus versus 100% with dexam-
ethasone. In women receiving continuous corticosteroid ther-
apy, premature rupture of the membranes can occur.
Natalizumab is unlikely to cross the placental barrier to any
great extent during early pregnancy. Therefore natalizumab
should be discontinued at a maximum of three months prior
to pregnancy. If natalizumab is administered in the third
trimester of pregnancy, some haematological abnormalities
such as thrombocytopenia and anaemia may occur. Effective
contraception is advised for at least two months after cessa-
tion of fingolimod treatment and breastfeeding is contraindi-
cated. Teriflunomide has teratogenic effect in animal studies.
Therefore pregnant women have teriflunomide exposure dur-
ing pregnancy should be using cholestyramine or activated
charcoal. Alemtuzumab use in pregnancy has no harmful ef-
fect with respect to obstetrical and fetal outcomes. For MS
patients with aggressive disease, treatments include im-
munosuppression or chemotherapy, followed by autologous
664 Multiple sclerosis management in pregnancy
stem cell transplant [20]. Mitoxantroneis, a cytotoxic
chemotherapeutic agent is used for secondary progressive
MS. Use of mitoxantrone during pregnancy has shown ter-
atogenic effects in the developing fetus and intrauterine
growth restriction at low doses in rats.
Breastfeeding in MS
Recent meta-analysis reported that association between
breastfeeding and MS relapse showed a tendency for fewer
relapses in women who breastfed, suggesting that exclusive
breastfeeding may reduce early postpartum relapses [28].
Furthermore, breastfeeding appears to be safe while patients
receive IFN-β and GA, natalizumab, and other non-depleting
monoclonal antibodies, but not small molecules such as fin-
golimod and dimethyl fumarate.
Obstetrical outcome in MS
Pregnant women have high levels of circulating lympho-
cytes and macrophages and decreased production of proin-
flammatory cytokines. These immune system changes
appear to have a cumulative, beneficial effect on MS by de-
creasing the incidence of a first clinical demyelinating event.
In addition, there is evidence that with each additional preg-
nancy that the incidence continues to decrease. The fre-
quency of relapse during pregnancy drops, with the third
trimester being the lowest. However, for three to six months
after delivery, the relapse rate significantly increases [29, 30].
With regards to obstetrical outcomes, the course of preg-
nancy is similar to that of women without MS but with a ten-
dency towards assisted delivery/cesarean section and
possibly lower neonatal birth weights. One study reported
that women who received epidural anesthesia of bupivacaine
had a higher incidence of postpartum relapses [31]. How-
ever, other studies failed to demonstrate any evidence of an
increase in postpartum relapse rate, based on the mode of ob-
stetrical anesthesia [32]. But another study conducted in
women with MS were found to have a higher incidence of
labor induction, longer second stage of labor, more opera-
tive vaginal births, and cesarean deliveries, and their infants
had lower birth weights [33]. However, neonatal mortality
rates were not increased. Reported most recently was a 30%
higher risk for cesarean birth and a 70% increase of in-
trauterine growth restriction, as compared to healthy women
[34]. MS has no any effect for miscarriage or congenital mal-
formation [32].
Oral contraceptives (OCs) use and other hormonal treat-
ment in MS
The estrogenic and progestagenic influence of oral con-
traceptive use increase the prevalence of MS. Another study
suggests that estrogen may have protective effects against
disease progression and have no adverse effects of incidence,
overall prognosis, or disability severity in women with MS
[35, 36]. The results of a well-matched case-control study
adjusted for MS risk factors showed an increased risk of MS
with use of combined OCs [35]. Short-term methylpred-
nisolone therapy does not appear to have an adverse effect on
fertility, whereas menstrual disturbances have been reported
with interferon beta and permanent amenorrhea with mitox-
antrone, especially in women older than 35 years [37].
Infertility treatment in MS
In general, fertility does not seem to be reduced in women
with MS. However, infertility and MS might just occur co-
incidentally due to a higher incidence of hyperprolactinemia,
thyroid disorders and endometriosis [37], higher levels of
prolactin, follicle-stimulating hormone, luteinizing hormone,
total and free testosterone, 5-α dihydrotestosterone, δ-4 an-
drostenedione levels, and decreased levels of estrone sulfate
[38]. Men with MS may have impaired fertility due to de-
creased sperm count, mobility, and normal sperm develop-
ment [39]. Therefore, MS patients might undergo assisted
reproductive treatment (ART).
Studies reported an increase in annualized relapse rate
after ART, particularly in the first three months after unsuc-
cessful cycles [37, 40]. Recent study reported that infertile
women with MS who underwent IVF showed a seven-fold
increase in the risk of MS exacerbation and a nine-fold in-
crease in the risk of disease progression with new brain le-
sions [41]. The risk appears to be greater with use of GnRH
agonists cycles. Putative mechanisms involved in MS wors-
ening after ART include: temporary interruption of disease
modified therapies, stressful events associated with IVF
treatment, and immunological changes induced by cytokines
and hormones, such as increase in pro-inflammatory cy-
tokines, estrogens, and progesterones, as well as an increase
in immune cell migration across the blood-brain-barrier.
Overall, obstetricians and neurologists should be aware of
this risk and discuss the pros and cons of the procedures with
MS patients [42]. MS patients requiring ART should be sta-
bilized before patients undergo the IFV procedure.
Conclusion
Although onset of MS in pregnancy is uncommon, large
numbers of female patients express desire to become preg-
nant after receiving MS diagnosis. Therefore, issues of con-
traception, conception, pregnancy, childbirth, and child
rearing become critically important in the overall manage-
ment strategies of MS patients and have been of great inter-
est to neurologists, obstetricians, and reproductive
specialists. It is known that the risk of MS relapse declines
during pregnancy but increases in the first three to six months
postpartum. It is also known that this risk is not affected by
delivery method, anesthesia type, or parity. IFN-β and glati-
ramer acetate appear to be most suitable for use up until the
time of confirmed pregnancy. Decision regarding the mode
of delivery is usually based on obstetrical rather than neuro-
logical factors. Infertile women with MS who underwent
IVF showed a seven-fold increase in the risk of MS exacer-
 L. Sahin, Y. Ehi 665

bation and a nine-fold increase in the risk of disease pro- [22] Koch M., Uyttenboogaart M., Heersema D., Steen C., de Keyser J.:
gression with new brain lesions. The hormonal influence of “Parity and secondary progression in multiple sclerosis”. J. Neurol.
Neurosurg. Psychiatry, 2009, 80, 676.
oral contraceptive use increase the prevalence of MS. [23] Dyment D.A., Ebers G.C., Schubert A.: “Genetics of multiple sclero-
sis”. Lancet Neurol., 2004, 3, 104.
[24] Crabtree-Hartman E.: “Sex differences in multiple sclerosis”. Contin-
References uum (Minneap Minn.), 2010, 16, 193.
[25] van der Kop M.L., Pearce M.S., Dahlgren L., Synnes A., Sadovnick
[1] Lucin K.M., Wyss-Coray T.: “Immune activation in brain aging and
D., Sayao A.L., Tremlett H.: “Neonatal and delivery outcomes in
neurodegeneration: too much or too little?” Neuron., 2009, 64, 110.
women with multiple sclerosis”. Ann. Neurol., 2011, 70, 41.
[2] Gonzalez H., Elgueta D., Montoya A., Pacheco R.: “Neuroimmune reg-
[26] Signore C., Spong C.Y., Krotoski D., Shinowara N.L., Blackwell S.C.:
ulation of microglial activity involved in neuroinflammation and neu-
“Pregnancy in women with physical disabilities”. Obstet. Gynecol.,
rodegenerative diseases”. J. Neuroimmunol., 2014, 274, 1.
2011, 117, 935.
[3] Egwuagu C.E., Larkin Iii J.: “Therapeutic targeting of STAT pathways
[27] Confavreux C., Hutchinson M., Hours M.M., Cortinovis-Tourniaire P.,
in CNS autoimmune diseases”. JAKSTAT, 2013, 2, e24134..
Moreau T.: “Rate of pregnancy-related relapse in multiple sclerosis.
[4] Goverman J.: “Autoimmune T cell responses in the central nervous
Pregnancy in Multiple Sclerosis Group”. N. Engl. J. Med., 1998, 339,
system”. Nat. Rev. Immunol., 2009, 9, 393.
285.
[5] Chastain E.M., Duncan D.S., Rodgers J.M., Miller S.D.: “The role of
[28] Pakpoor J., Disanto G., Lacey M.V., Hellwig K., Giovannoni G., Ra-
antigen presenting cells in multiple sclerosis”. Biochim. Biophys. Acta,
magopalan S.V.: “Breastfeeding and multiple sclerosis relapses: a meta-
2011, 1812, 265.
analysis”. J. Neurol., 2012, 259, 2246.
[6] Jadidi-Niaragh F., Mirshafiey A.: “Histamine and histamine receptors
[29] Drew P.D., Chavis J.A.: “Female sex steroids: effects upon microglial
in pathogenesis and treatment of multiple sclerosis”. Neuropharma-
cell activation”. J. Neuroimmunol., 2000, 111, 77.
cology, 2010, 59, 180.
[30] PPonsonby A.L., Lucas R.M., van der Mei I.A., Dear K., Valery P.C.,
[7] Prado C.,Contreras F., Gonzales H., Diaz P., Elqueta D., Barrientos M.,
Pender M.P., et al.: “Offspring number, pregnancy, and risk of a first
et al.: “Stimulation of dopamine receptor D5expressed on dendritic
clinical demyelinating event: the AusImmune Study”. Neurology, 2012,
cells potentiates Th17-mediated immunity”. J. Immunol., 2012, 188,
78, 867.
3062.
[31] Bader A.M., Hunt C.O., Datta S., Naulty J.S., Ostheimer G.W.: “Anes-
[8] Reboldi A., Coisne C., Baumjohann D., Benvenuto F., Bottinelli D.,
thesia for the obstetric patient with multiple sclerosis”. J. Clin. Anesth.,
Lira S., et al.: “C-C chemokine receptor 6-regulated entry of TH-17
1988, 1, 21.
cells into the CNS through the choroid plexus is required for the initi-
[32] Vukusic S., Hutchinson M., Hours M., Moreau T., Cortinovis-Tourni-
ation of EAE”. Nat. Immunol., 2009, 10, 514.
aire P., Adeleine P., et al.: “Pregnancy and multiple sclerosis (The
[9] Carson M.J., Doose J.M., Melchior B., Schmid C.D., Ploix C.C.: “CNS
PRIMS Study): clinical predictors of post-partum relapse”. Brain, 2004,
immune privilege: hiding in plain sight”. Immunol. Rev., 2006, 213, 48.
127, 1353.
[10] Chu C.Q., Wittmer S., Dalton D.K.: “Failure to suppress the expansion
[33] Dahl J., Myhr K.M., Daltveit A.K., Hoff J.M., Gilhus N.E.: “Pregnancy,
of the activated CD4 T cell population in interferon -deficient mice
delivery, and birth outcome in women with MS”. Neurology, 2005, 65,
leads to exacerbation of experimental autoimnaune encephalomyelitis”.
1961.
J. Exp. Med., 2000, 192, 123.
[34] Kelly V.M., Nelson L.M., Chakravarty E.F.: “Obstetric outcomes in
[11] Kleinewietfeld M., Manzel A., Titze J., Kvakan H., Yosef N., Linker
women with MS and epilepsy”. Neurology, 2009, 73, 1831.
R.A., et al.: “Sodium chloride drives autoimmune disease by the in-
[35] Hellwig K., Haghikia A., Rockhoff M., Gold R.: “Multiple sclerosis
duction of pathogenic TH17 cells”. Nature, 2013, 7446, 518.
and pregnancy: experience from a nationwide database in Germany”.
[12] Staeva-Vieira T.P., Freedman L.P.: “1,25-Dihydroxyvitamin D3 inhibits
Ther. Adv. Neurol. Disord., 2012, 5, 247.
IFN- and IL-4 levels during in vitro polarization of primarymurine
[36] Ferrero S., Esposito F., Pretta S., Ragni N.: “Fetal risks related to the
CD4+ T cells”. J. Immunol., 2002, 168, 1181.
treatment of MS during pregnancy and breast-feeding”. Expert Rev.
[13] Hall J.A., Grainger J.R., Spencer S.P., Belkaid Y.: “The role of retinoic
Neurother., 2006, 6, 1823.
acid in tolerance and immunity”. Immunity, 2011, 35, 13.
[37] Cavalla P.: “Fertility in patients with multiple sclerosis: current knowl-
[14] Kebir H., Kreymborg K., Ifergan I., Dodelet-Devillers A., Cayrol R.,
edge and future perspectives”. Neurol. Sci., 2006, 27, 231.
Bernard M., et al.: “Human TH17 lymphocytes promote blood-brain
[38] Grinsted L., Heltberg A., Hagen C., Djursing H.: “Serum sex hormone
barrier disruption and central nervous system inflammation”. Nat. Med.,
and gonadotropin concentrations in premenopausal women with mul-
2007, 13, 1173.
tiple sclerosis”. J. Intern. Med., 1989, 226, 241.
[15] Kwon H.K., Kim G.C., Kim Y., Hwang W., Jash A., Sahoo A., et al.:
[39] Demirkiran M., Sarica Y., Uguz S., Yerdelen D., Aslan K.: “Multiple
“Amelioration of experimental autoimmune encephalomyelitis by pro-
sclerosis patients with and without sexual dysfunction: are there any
biotic mixture is mediated by a shift in T helper cell immune response”.
differences?” Mult. Scler., 2006, 12, 209.
Clin. Immunol., 2013, 146, 217.
[40] Hellwig K., Chen L., Langer-Gould A.: “Hormonal contraceptives and
[16] Yokote H., Miyake S., Croxford J.L., Oki S., Mizusawa H., Yamamura
multiple sclerosis susceptibility”. Neurology, 2014, 82, S34.
T.: “NKT cell-dependent amelioration of a mouse model of multiple
[41] Voskuhi R.R.: “Assisted reproduction technology in multiple sclerosis.
sclerosis by altering gut flora”. Am. J. Pathol., 2008, 173, 1714.
Giving birth to a new avenue of research in hormones and autoimmu-
[17] Marchiondo K.: “Multiple sclerosis”. Medsurg Nurs., 2010, 19, 303.
nity”. Ann. Neurol., 2012, 72, 631.
[18] Hauser S.L., Goodin D.S.: “Multiple sclerosis and other demyelinat-
[42] Hellwig K., Correale J.: “Artificial reproductive techniques in multiple
ingdiseases”. In: Kasper D.L., Harrison T.R. (eds). Harrison’s Princi-
sclerosis”. Clin Immunol., 2013, 149, 219.
ples of Internal Medicine. 17th ed. New York, NY: McGraw-Hill
Medical Pub Division, 2008, 2611.
[19] Houtchens M.K.: “Bradley’s neurology in clinical practice”. 6th ed. Corresponding Author:
Philadelphia, PA: Elsevier; 2012, 1283. L. SAHIN, M.D.
[20] Tsang B.: “MS: diagnosis, management and prognosis”. Aust. Fam. Kafkas University Medical School
Phys., 2011, 40, 948. Department of Obstetrics and Gynecology
[21] Ramagopalan S., Yee I., Byrnes J., Guimond C., Ebers G., Sadovnick Sehitler Mahallesi
D.: “Term pregnancies and the clinical characteristics of multiple scle- 36000 Merkez-Kars (Turkey)
rosis: a population-based study”. J. Neurosurg. Psychiatry, 2012, 8,
e-mail: leventsahinmd@yahoo.com
793.
CEOG Clinical and Experimental
Obstetrics & Gynecology

Interventions for treating amniotic fluid embolism:

a systematic review with meta-analysis

U. Indraccolo1, R. Ventrone2, G. Scutiero3, P. Greco3, S.R. Indraccolo2

1Complex Operative Unit of Obstetrics and Gynecology, “Alto Tevere” Hospital of Città di Castello, ASL 1 Umbria, Città di Castello
2Department of Gynecological, Obstetrical, and Urological Sciences; “Sapienza” University of Rome, Rome
3Department of Morphology, Surgery and Experimental Medicine, University of Ferrara, Ferrara (Italy)

Summary
Purpose: Assessing to what extent the interventions for treating amniotic fluid embolism (AFE) are effective. Materials and Meth-
ods: A systematic review of cases of AFE available on PubMed, Scielo, Scopus, and AJOL databases from 1990 to 2015 was carried
out. The perception of effectiveness of each kind of intervention on heart, lungs, coagulopathy, and the importance of immediate par-
turition was quantified semi-quantitatively, by scoring textual information (from 0 to 3). Scores 2 and 3 were considered positive scores
(effective intervention), while 0 and 1 were considered negative scores (ineffective intervention). Rates of such scores were compared
with a random distribution of scores from 0 to 3. Sub-groups analyses were carried out. Results: One hundred twenty-one typical AFE
cases were assessed. Each intervention for supporting the heart and, predominantly, lung function is perceived as pivotal. Conclusion:
The management of AFE should be focused on supporting the lung and the heart function.

Key words: Amniotic fluid embolism; Outcome; Treatments; Meta-analysis.

Introduction as effective by authors of the case reports. This meta-analy-

Amniotic fluid embolism (AFE) is a rare and potentially
catastrophic syndrome of pregnancy. It has been reported
[1] that an estimated AFE rate of 1/15,200 deliveries is
present in North America, and occurs at 1/53,800 deliver-
ies in Europe. Additionally, the maternal mortality ratio for
AFE ranges between 0.5 - 1.7 deaths per 100,000 live
births. Due to its rarity and lack of unequivocal diagnostic
criteria, AFE may have unknown true rates of occurrence,
and data is lacking regarding clinical onset, outcomes, and
the effectiveness of treatments provided [2]. As it has been
reported that the survival rate is improving [1, 3], one could
hypothesize that some interventions provided for manag-
ing AFE are effective for improving AFE outcome. To
check that hypothesis, observational or randomized stud-
ies to prove the efficacy of treatments in AFE cases could
be carried out. However, AFE is too rare to allow studies of
large series. Therefore, all the evidence regarding AFE has
been extracted from case reports or small series, with the
best evidence available from national registries or popula-
tion-based reviews. Therefore, the present authors assessed
the effectiveness of intervention in cases of AFE by meta-
analyzing the AFE case reports. They hypothesize that in-
terventions provided in amniotic fluid embolism are
effective, and to prove it, they quantified from textual in-
formation how much an intervention for AFE is perceived
sis has been registered on the International Prospective
Register of Systematic Review (PROSPERO) site (number
34104).
Materials and Methods
On January 20th, 2016, the present authors performed a sys-
tematic review of the literature by introducing “amniotic” AND
“fluid” AND “embolism” as key words in the PubMed, Scielo,
AJOL (African Journal Online), and Scopus search engines. The
time frame of the search was limited to run from 1990 to the pres-
ent. This range was given because the care for AFE could have
changed over the years. No languages limitations were set.
The PubMed search returned 607 references, Scopus returned
1,750 references, Scielo returned 16 references, and AJOL re-
turned one reference. The whole body of references was checked
for duplicates by both a manual check and the use of EndNote.
After that, the authors read the titles and abstracts, checking for
case reports, letters to the editors, and small series. They searched
case descriptions in which the clinical picture of each case of AFE
was reported. If titles and abstracts were not exhaustive, they read
the full text article (if available) to understand if cases and small
series of cases reported useful clinical information. They were
able to read articles in Italian, English, French, Spanish, and Por-
tuguese in their original languages, while other languages were
translated into English or Italian.
Full texts were found on the “Sapienza” University of Rome
electronic database, and by using the Italian interlibrary free ex-
change of full- text articles (NILDE tool) for obtaining full texts

Revised manuscript accepted for publication November 24, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog4013.2017
 U. Indraccolo, R. Ventrone, G. Scutiero, P. Greco, S.R. Indraccolo
Table 1. — Discarded references.
Discarded because unavailability of the full text:
1. Alamo F et al. [On amniotic fluid embolism]. Rev Esp Anestesiol
Reanim. 2001;48:251-2.
2. Burrows A. The amniotic fluid embolism syndrome: 10 years’ expe-
rience at a major teaching hospital. Aust N Z J Obstet Gynaecol.
1995;35:245-50.
3. Bussen S et al. [Clinical aspects and therapy of amniotic fluid em-
bolism. Illustration based on a case report]. Z Geburtshilfe Neonatol.
1997;201:95-8.
4. Davies MG et al. Amniotic fluid embolism: maternal mortality re-
visited. Br J Hosp Med. 1992;47:775-6.
5. Delvenne P, et al. [Anatomo-clinical correlation—pulmonary em-
bolism not originating from the leg]. Rev Med Liege. 1990;45:649-
59.
6. Demianczuk CE et al. Succesful pregancy after amniotic fluid em-
bolism: a case report. J Obstet Gynaecol Can. 2005;27:699-701.
7. Hasenburg A et al. [Fatal amniotic fluid embolism 20 hours after
secondary cesarean section]. Gynakologe. 1994;27:352-4.
8. Heidenreich W. [Clinical aspects of amniotic fluid embolism]. Zen-
tralbl Gynakol. 1993;115:233-7.
9. Katsulov A. [Amniotic embolism—a case in forensic medicine
practice and literature review]. Akush Ginekol (Sofia).
2005;44Suppl1:15-8.
10. Hyuga S, et al. [Prompt resuscitation by obstetric anesthesiologists
saved a parturient with amniotic fluid embolism: a case report].
Masui. 2013;62:1435-9.
11. Masson RG. Amniotic fluid embolism. Clin Chest Med.
1992;13:657-65.
12. Nadesan K et al. Sudden maternal deaths from amniotic fluid em-
bolism. Ceylon Med J. 1997;42:185-9.
13. Nadjem H, et al. [A case of fatal amniotic fluid embolism]. Arch
Kriminol. 2001;207:89-96.
14. Observer AF. The ties that bind medical intervention and maternal
mortality: Mary’s story. Birth Gaz. 1999;15:23-5.
15. Oi C, et al. [A Successful Case of Recovery from the Amniotic Fluid
Embolism with Cardiac Arrest during the Operation]. Masui.
2015;64:1186-9.
16. Petroff S, et al. [Amniotic embolism with favourable outcome]. Ann
Fr Anest Reanim. 1994;13:135-7.
17. Roungsipragarn et al. Amniotic fluid embolism: a case report. J
Med Assoc Thai. 1993;76Suppl1:105-7.
18. Sugasawa Y, et al. [Successful anesthetic management of a post-
partum patient with amniotic fluid embolism]. Masui. 2011;60:91-
5.
19. Tammelleo AD. Legal cases brief for nurses. S.D.: agency’s liabil-
ity for unemployment contributions; MI.: ”lost opportunity to sur-
vive”: CRNA involved. Regan Rep Nurs Law. 1991;31:3.
20. Yorifuji H et al. [A survival case of amniotic fluid embolism]. Nihon
Sanka Fujinka Gakkai Zasshi. 1990;42:491-4.
Discarded because other reasons:
1. Abecassis P, et al. Is amniotic fluid embolism likely to recur in a
subsequent pregnancy? Int J Obstet Anesth. 2006;15:90
2. Almaguer Almaguer JA, et al. [Amniotic fluid embolism vs primary
pulmonary hypertension. A case Report]. Rev Cubana Obstet Gy-
necol. 2007;33(3).
3. Balazic J, et al. Amniotic fluid embolism with involvement of the
brain, lungs, adrenal glands, and heart. Int J Legal Med.
2003;117:165-9.
4. Bauer P, et al. Thrombo-embolism during amniotic fluid embolism.
Intensive Care Med. 1995;21:384.
5. Benson MD, et al. A mild case of amniotic fluid embolism? J
Matern Fetal Neonatal Med. 2007;20:261-2.
6. Burtelow M, et al. How we treat: management of life-threatening
primary postpartum hemorrhage with a standardized massive trans-
667
fusion protocol. Transfusion. 2007;47:1564-72.
7. Cabrera-Samithi I, et al. [Amniotic fluid embolism in preterm labor:
a case report]. CIMEL. 2008;13:78-81.
8. Caputo M, et al. Severe fetal ischemic complications caused by sec-
ond trimester amniotic fluid embolism. Int J Gynaecol Obstet.
2012:116:175-6.
9. Carrillo-Galindo A, et al. [Amniotic fluid embolism direct cause of
maternal death. Clinical case report]. Gac Med Mex. 2003;139:607-
9.
10. Cheung AN, et al. The importance of extensive sampling and ex-
amination of cervix in suspected cases of amniotic fluid embolism.
Arch Gynecol Obstet. 1994;255:101-5.
11. D’Addato F, et al. [Amniotic fluid embolism in trial of labor. A case
report]. Minerva Ginecol. 1997;49:217-20.
12. Dorairajan G, et al. Maternal death after intrapartum saline am-
nioinfusion—report of two cases. BJOG. 2005;112:1331-3.
13. Farrar SC, et al. Serum tryptase analysis in a woman with amniotic
fluid embolism. A case report. J Reprod Med. 2001;46:926-8.
14. Feng ZY, et al. Successful pregnancy after amniotic fluid embolism.
Chin Med J (Engl). 2013;126:2799.
15. Franchitto N, et al. Medical responsibility in the operating room:
the example of an amniotic fluid embolism. J Forensic Sci.
2012;57:1120-3.
16. Gogola J, et al. Amniotic fluid embolism in progress: a manage-
ment dilemma! Am J Perinatol. 1998;15:491-3.
17. Guerry MJ, et al. [Septic shock and amniotic fluid embolism]. Med
Mal Infect. 2012;42:86-8.
18. Haines J, et al. Non-fatal amniotic fluid embolism after cervical su-
ture removal. Br J Anaesth. 2003;90:244-7.
19. Harboe T, et al. Cardiopulmonary distress during obstetrical anaes-
thesia: attempts to diagnose amniotic fluid embolism in a case se-
ries of suspected allergic anaphylaxis. Acta Anaesthesiol Scand.
2006;50:324-30.
20. Hosoya Y, et al. Cardiac magnetic resonance imaging in a patient
with amniotic fluid embolism associated with severe cardiopul-
monary complications. Int Heart J. 2013;54:119-22.
21. Howes LJ. Anaphylactoid reaction possibly caused by amniotic
fluid embolism. Int J Obstet Anesth. 1995;4:51-4.
22. Humikowski CA. A piece of my mind. For what it’s worth. JAMA.
2015;313:799-800.
23. Hussain SA, et al. Amniotic fluid embolism with late respiratory
failure. Hosp Phys. 2001;Dec:40-3.
24. Jeanneteau P, et al. Amniotic fluid embolism after uterine artery em-
bolization for uterine fibroids. Eur J Obstet Gynecol Reprod Biol.
2015;191:144-5.
25. Kim IT, et al. Occlusions of branch retinal arterioles following am-
niotic fluid embolism. Ophthalmologica. 2000;214:305-8.
26. Kobayashi et al. A simple, noninvasive, sensitive, method for the
diagnosis of amniotic fluid embolism by monoclonal antibody
TKH-2 that recognizes NeuAc alpha 2-6GalNAc. Am J Obstet Gy-
necol. 1993;168(3 Pt 1):848-53.
27. Kobayashi H, et al. Histological diagnosis of amniotic fluid em-
bolism by monoclonal antibody TKH-2 that recognizes NeuAc
alpha 2-6GalNAc epitope. Hum Pathol. 1997;28:428-33.
28. Laforga JB. Amniotic fluid embolism. Report of two cases with co-
agulation disorder. Acta Obstet Gynecol Scand. 1997;76:805-6.
29. Lee PH, et al. Surgical treatment of an amniotic fluid embolism with
cardiopulmonary collapse. Ann Thorac Surg. 2010;90:1694-6.
30. Less A, et al. Vaginal prostaglandin E2 and fatal amniotic fluid em-
bolus. JAMA. 1990;263:3259-60.
31. Liao WC, et al. A noninvasive evaluation analysis of amniotic fluid
embolism and disseminated intravascular coagulopathy. J Matern
Fetal Neonatal Med. 2011;24:1411-5.
32. Lopez Matta J, et al. Embolism by the water of life. Neth J Crit
Care. 2014;18:26-9.
33. Lunetta P, et al. Immunohistochemical identification of syncy-
tiotrophoblastic cells and megakaryocytes in pulmonary vessels in
668 Interventions for treating amniotic fluid embolism: a systematic review with meta-analysis
a fatal case of amniotic fluid embolism. Int J Legal Med.
1996;108:210-4.
34. Maher JE, et al. Amniotic fluid embolism after saline amnioinfu-
sion: two cases and review of the literature. Obstet Gynecol.
1994;83(5 Pt 2):851-4.
35. Mahshid N, et al. Sudden cardiac arrest during cesarean section —
a possible case of amniotic fluid embolism. Middle East J Anes-
thesiol. 2009;20:315-7.
36. Malvino E, et al. [Amniotic fluid embolism. Diagnostic criteria in
two fatal cases]. Medicina (B Aires). 2008;68:59-61.
37. Marcus BJ, et al. Ancillary studies in amniotic fluid embolism: a
case report and review of the literature. Am J Forensic Med Pathol.
2005;26:92-5.
38. Margarson MP. Delayed amniotic fluid embolism following caesarean
section under spinal anaesthesia. Anaesthesia. 1995;50:804-6.
39. Martindale E, et al. Amniotic fluid embolism and Pgs. Br J Hosp
Med. 1992;48:135.
40. Michel C, et al. [Prediction by means of endogenous and exoge-
nous evoked potentials of the favorable evolution of a prolonged
coma]. Neurophysiol Clin. 1998;28:343-59.
41. Nakagami H, et al. Amniotic components in the uterine vasculature
and their role in amniotic fluid embolism. J Obstet Gynaecol Res.
2015;41:870-5.
42. Nishio H, et al. A fatal case of amniotic fluid embolism with eleva-
tion of serum mast cell tryptase. Forensic Sci Int. 2002;126:53-6.
43. Pang AL, et al. Amniotic fluid embolism during Caesarean section
under spinal anaesthesia. Is sympathetic blockade a risk factor? Aust
N Z J Obstet Gynaecol. 2001;41:342-3.
44. Papaioannou VE, et al. A step-by-step diagnosis of exclusion in a
twin pregnancy with acute respiratory failure due to non-fatal am-
niotic fluid embolism: a case report. J Med Case Reports.
2008;2:177.
45. Richa F et al. Obstetric admissions to the intensive care unit: an
eight-year review. J Med Liban. 2008;56:215-9.
46. Van Cortenbosch B et al. [Biologic tests for the diagnosis of amni-
otic fluid embolism]. Ann Biol Clin (Paris). 2007;65:153-60.
47. Rodgers L, et al.. Acute circulatory and respiratory collapse in ob-
stetrical patients: a case report and review of the literature. AANA
J. 2000; 68:444-50.
48. Romero C, et al. Embolía por Líquido Amniótico: una puesta al día
a propósito de un caso. Rev Chil Med Inten. 2005; 20:38-41.
49. Romero R, et al. Maternal death following cardiopulmonary col-
lapse after delivery: amniotic fluid embolism or septic shock due
from other Italian libraries. Additional articles were freely down-
loaded from Google Scholar, if available, or were purchased. Fi-
nally, some articles were provided from some authors of case
reports by e-mail.
To the remaining 233 references, seven additional references
incidentally found on Google Scholar during the full texts search
process were added. The authors were unable to collect 20 full
texts. Therefore, they were forced to discard 20 references.
To best assess the perception of effectiveness for interventions
in AFE cases, the authors focused on the cases description and
discussion. They chose to discard references in which the effec-
tiveness of interventions for AFE was not reported or not dis-
cussed (51 references). Moreover, they discarded six references on
small series not assessing the treatment of AFE, seven cases in
which AFE was unlikely, one case in which an amniotic embolus
was found and treated before it would have reached the heart, and
one case reported in a retracted article.
Because some references reported more than one case, the au-
thors thus collect a database of 181 cases of suspected AFE cases.
However, they were forced to discard three cases due to insuffi-
cient information and one case where AFE seemed unlikely.
Figure 1 shows a flow-chart of phases for building the database
to intrauterine infection? Am J Reprod Immunol. 2010;64:113-25.
50. Rosain J, et al. [Last updates about amniotic fluid embolism and a
case report]. Ann Biol Clin (Paris). 2013;71:633-7.
51. Rufforny-Doudenko I, et al. Pathologic quiz case. A 30-year-old
woman with severe disseminated intravascular coagulation during
delivery. Arch Pathol Lab Med. 2002;126:869,870.
52. Saad A, et al. Echocardiographically detected mass “in transit” in
early amniotic fluid embolism. Eur J Echocardiogr. 2006;7:332-5.
53. Sené T, et al. [Probable spontaneous amniotic fluid embolism at 28
weeks of term]. Ann Fr Anesth Reanim. 2010;29:736-7.
54. St-Amand J. Medicolegal nightmare: a tragic case, a needless trial.
CMAJ. 1993;148:806-9.
55. Styron AG, et al. Multidisciplinary management of placenta perc-
reta complicated by embolic phenomena. Int J Obstet Anesth.
2008;17:262-6.
56. Tamura N, et al. Amniotic fluid embolism induces uterine anaphy-
laxis and atony following cervical laceration. Gynecol Obstet In-
vest. 2014;78:65-8.
57. Truong HT, et al. Anaphylaxis-induced hyperfibrinolysis in preg-
nancy. Int J Obstet Anesth. 2015;24:180-4.
58. Umazume T, et al. Protein S deficiency present in a pregnant woman
with dyspnea, abdominal pains, restlessness, agitation and hy-
pofibrinogenemia. Clin Case Rep. 2015;3:213-6.
59. Van Cortenbosch B, et al. [Biologic tests for the diagnosis of amni-
otic fluid embolism]. Ann Biol Clin (Paris). 2007;65:153-60.
60. van Liempt SW, et al. Essentials in cardiac arrest during cesarean
section. Clin Pract. 2015;5:668.
61. Wernet A, et al. Use of insulin-like growth factor binding protein-
1 for retrospective diagnosis of amniotic fluid embolism in first
trimester. Arch Gynecol Obstet. 2014;289:461-2.
62. Willging KJ. Falcon v. Memorial Hospital: a rational approach to
loss-of-chance tort actions. J Contemp Health Law Policy.
1993;9:545-62.
63. Yang JI, et al. Amniotic fluid embolism with isolated coagulopa-
thy: a case report. J Reprod Med. 2006;51:64-6.
64. Woo YS, et al. Ischemic stroke related to an amniotic fluid em-
bolism during labor. J Clin Neurosci. 2015;22(4):767-8.
65. Žáková D, et al. [Different outcome in two case reports of births
showing symptoms of amniotic fluid embolism in the years 2002
and 2012 in the district hospital of Decin]. Ceska Gynekol.
2015;80:302-8.
66. Zamurović M, et al. Pulmonary embolism during delivery—treat-
ment and outcome. Clin Exp Obstet Gynecol. 2015;42:833-4.
of 177 presumed cases of AFE.
The interventions considered to be effective for managing AFE
cases were: 1) pulmonary interventions (cardio-pulmonary resus-
citation, extracorporeal membrane oxygenation, cardio-pul-
monary by-pass, embolectomy), 2) cardiac interventions
(cardio-pulmonary resuscitation, defibrillation, inotropic agents,
fluid infusions, cardio-pulmonary by-pass, aortic balloon counter-
pulsation, open heart massage, aortic compression), 3) interven-
tions on the coagulopathy (fresh frozen plasma infusion,
cryoprecipitate, recombinant factor VIIa infusion, fibrinogen con-
centrate infusion, tranexamic acid, platelets infusion, antithrom-
bin III infusion, desmopressin acetate therapy, aprotinin, C1
esterase inhibitor), and 4) immediate delivery (each operative in-
tervention for delivery the baby or for interrupting the pregnancy
and emptying the uterus).
The effect size to be meta-analysed is the perception of effec-
tiveness of the interventions in AFE cases, as reported authors in
their cases description and discussion. To provide an unequivo-
cal measure of such effectiveness, the authors built a semi-quan-
titative score based on textual information. They assigned 0 if the
interventions are not reported or if they were considered totally in-
effective. They assigned 1 if the interventions were perceived to
 U. Indraccolo, R. Ventrone, G. Scutiero, P. Greco, S.R. Indraccolo
Table 2. — References used for meta-analysis.
1. [No authors listed]. Case records of the Massachusetts General Hos-
pital. Weekly clinicopathological exercises. Case 9-1998. Cardio-
vascular collapse after vaginal delivery in a patient with a history of
cesarean section. N Engl J Med. 1998;338:821-6..
2. Adibe EC, et al. Amniotic fluid embolism: case report in ABSUTH.
ABSUMSAJ. 2007;4:41-2.
3. Aguilera LG, et al. Fatal amniotic fluid embolism diagnosed histo-
logically. Acta Anaesthesiol Scand. 2002;46:334-7.
4. Almeida PE, et al. [Pulmonary amniotic fluid embolism syndrome.
Case report and literature review]. Rev Bras Ter Intensiva.
2007;19:238-41.
5. Alon E, et al. Successful cardiopulmonary resuscitation of a par-
turient with amniotic fluid embolism. Int J Obstet Anesth.
1992;1:205-7.
6. Annecke T, et al. Algorithm-based coagulation management of cat-
astrophic amniotic fluid embolism. Blood Coagul Fibrinolysis.
2010;21:95-100.
7. António C, et al. [Amniotic fluid embolism]. Acta Med Port.
2011;24:1087-90.
8. Arafa A, et al. A novel case of amniotic fluid embolism in multiple
pregnancy. J Obstet Gynaecol. 2009;29:64-5.
9. Awad IT, et al. Amniotic fluid embolism and isolated coagulopa-
thy: atypical presentation of amniotic fluid embolism. Eur J Anaes-
thesiol. 2001;18:410-3.
10. Aziz ES. A successfully resuscitated case of amniotic fluid em-
bolism. Eur J Anaesthesiol. 2000;17:712-3.
11. Baghirzada L, et al. Maternal cardiac arrest in a tertiary care centre
during 1989-2011: A case series. Can J Anaesth. 2013;60:1077-84.
12. Bastien JL, et al. Atypical presentation of amniotic fluid embolism.
Anesth Analg. 1998;87:124-6.
13. Belfort MA, et al. Aortic compression and cross clamping in a case
of placenta percreta and amniotic fluid embolism: a case report. AJP
Rep. 2011;1:33-6.
14. Ben Ismail M, et al. [Successful outcome of amniotic fluid em-
bolism with cardiac arrest during labour]. Rev Med Perinatal.
2014;6:265-7.
15. Benson MD. Nonfatal amniotic fluid embolism. Three possible
cases and a new clinical definition. Arch Fam Med. 1993;2:989-4.
16. Bermejo-Alvarez MA, et al. [Fulminant course of amniotic fluid
embolism]. Rev Esp Anestesiol Reanim. 2006;53:114-8.
17. Biron-Andreani C, et al. Amniotic fluid embolism with haemosta-
sis complications: primary fibrinogenolysis or disseminated in-
travascular coagulation? Pathophysiol Haemost Thromb.
2003;33:170-1.
18. Bøgeskov RA, et al. [Thromboelastography-guided transfusion in a
patient with amniotic fluid embolism and massive coagulopathic
bleeding]. Ugeskr Laeger. 2015;177:pii:V05150453.
19. Bolden N,et al. Making the case for obstetric “response teams” and
simulation in labor and delivery: management of catastrophic am-
niotic fluid embolism during labor. J Clin Anesth. 2012;24:517-8.
20. Bouman EA, et al. [Complicated but successful resuscitation after
amniotic fluid embolism]. Ned Tijdschr Geneeskd. 2001;145:747-9.
21. Buechel J, et al. Unresponsive primipara after rupture of mem-
branes. BMJ Case Rep. 2015;pii: bcr2015209765.
22. Calvo N, et al. [Late amniotic fluid embolism among the differen-
tial diagnosis of acute respiratory failure in the postpartum: case re-
port and review]. Med Intensiva. 2008;32:404-5.
23. Campanharo FF, et al. Amniotic embolism with complement activa-
tion in a lupic pregnant woman. J Obstet Gynaecol. 2015;35:416-7.
24. Chanimov M, et al. Amniotic fluid embolism: a plea for better brain
protection. Isr Med Assoc J. 2008;10:154-5.
25. Choi DM, et al. Amniotic fluid embolism. Anaesth Intensive Care.
1995;23:741-3.
26. Collier C. Recurring amniotic fluid embolism. Anaesth Intensive
Care. 1998;26:599-600.
669
27. Collins NF, et al. Hyperfibrinolysis diagnosed by rotational throm-
boelastometry in a case of suspected amniotic fluid embolism. Int
J Obstet Anesth. 2013;22:71-6.
28. Cragun JM, et al. Amniotic fluid embolism presenting after uterine
manipulation under general anesthesia: a case with a positive out-
come. Resuscitation. 2008;77:145-6.
29. Davies S. Amniotic fluid embolism and isolated disseminated in-
travascular coagulation. Can J Anaesth. 1999;46(5 Pt 1):456-9.
30. Davis D. Amniotic fluid embolism. Exploring this rare but typically
fatal condition. AWHONN Lifelines. 2003;7:126-31.
31. De Jong MJ, et al. Anaphylactoid syndrome of pregnancy. A dev-
astating complication requiring intensive care. Crit Care Nurse.
2003;23:42-8.
32. de Rooij GM, et al. Severe maternal respiratory distress due to the
amniotic fluid embolism syndrome in a twin pregnancy. Neth J
Med. 2003;61:337-40.
33. Dean LS, et al. Case scenario: amniotic fluid embolism. Anesthesi-
ology. 2012;116:186-92.
34. Devriendt J, et al. Amniotic fluid embolism: another case with non-
cardiogenic pulmonary edema. Intensive Care Med. 1995;21:698-9.
35. Dib N, et al. Amniotic fluid embolism causing severe left ventricu-
lar dysfunction and death: case report and review of the literature.
Cathet Cardiovasc Diagn. 1996;39:177-80.
36. Dorne R, et al. [Amniotic fluid embolism: successful evolution
course after uterine arteries embolization]. Ann Fr Anesth Reanim.
2002;21:431-5.
37. Ducloy-Bouthors AS, et al. [Amniotic fluid embolism suspected in a
case of seizure and mild uterine haemorrhage with activation of co-
agulation and fibrinolysis]. Ann Fr Anesth Reanim. 2004;23:149-52.
38. Ecker JL, et al. Case records of the Massachusetts General Hospi-
tal. Case 40-2012. A 43-year-old woman with cardiorespiratory ar-
rest after a cesarean section. NEJM. 2012:367:2528-36.
39. Elbahraoui H, et al. [Amniotic fluid embolism: report of two cases].
Pan Afr Med J. 2012;11:74.
40. Ellingsen CL, et al. Amniotic fluid embolism after blunt abdominal
trauma. Resuscitation. 2007;75:180-3.
41. Esposito RA, et al. Successful treatment of postpartum shock
caused by amniotic fluid embolism with cardiopulmonary bypass
and pulmonary artery thromboembolectomy. Am J Obstet Gynecol.
1990;163:572-4.
42. Evans S, et al. Survival after an amniotic fluid embolism following the
use of sodium bicarbonate. BMJ Case Rep. 2014;pii:bcr2014204672.
43. Falzone E, et al. [Amniotic fluid embolism: report of the success-
ful outcome of a case with inaugural cardiac arrest and early DIVC
complicated by hemoperitoneum of iatrogen origin and bleeding of
an hepatic adenoma] Ann Fr Anesth Reanim. 2012;31:802-5.
44. Fava S, et al. Amniotic fluid embolism. Br J Obstet Gynaecol.
1993;100:1049-50.
45. Fekhkhar K, et al. [Amniotic fluid embolism during curettage for a
pregnancy arrest. Case report]. Ann Fr Anesth Reanim.
2009;28:795-8.
46. Firstenberg MS, et al. Temporary extracorporeal circulatory sup-
port and pulmonary embolectomy for catastrophic amniotic fluid
embolism. Heart Surg Forum. 2011;14:e157-9.
47. Fletcher SJ, et al. Amniotic fluid embolism: a case report and re-
view. Resuscitation. 2000;43:141-6.
48. Furukawa S, et al. A rare case of amniotic fluid embolism with mas-
sive platelet aggregations in pulmonary capillaries. J Obstet Gy-
naecol Res. 2010;36:397-400.
49. Gad N. Amniotic fluid embolism (it can present without respiratory
manifestations). Case report. Clin Exp Obstet Gynecol.
1996;23:248-51.
50. Gamerre L, et al.[Amniotic fluid embolism: successful evolution
course of isolated disseminated intravascular coagulation and early
biological diagnosis]. Ann Fr Anesth Reanim. 2006;25:633-7.
51. Goldszmidt E, et al. Two cases of hemorrhage secondary to amni-
otic fluid embolus managed with uterine artery embolization. Can
670 Interventions for treating amniotic fluid embolism: a systematic review with meta-analysis
J Anaesth. 2003;50:917-21.
52. Green BT, et al. Amniotic fluid embolism. South Med J.
2000;93:721-3.
53. Guenther SPW, et al. [Fulminant amniotic fluid embolism: mobile
ECLS team in the delivery ward]. Z Herz Thorax Gefäßchir.
2015;29:312-6.
54. Gurke B, et al. [Amniotic fluid embolism as an obstetric emer-
gency—description of the syndrome and a case with fatal outcome].
Anasthesiol Intensivmed Notfallmed Schmerzther. 2001;36:247-9.
55. Hardin L, et al. Amniotic fluid embolism. South Med J.
1991;84:1046-8.
56. Healy BR, et al. Surviving anaphylactoid syndrome of pregnancy:
a case study. Clin Lab Sci. 2013Spring;26:72-5.
57. Hernández Iglesias R, et al. [Amniotic fluid embolism. A report of
a case]. Rev Cuba Anestesiol Reanim. 2010;9:223-31.
58. Hikiji W, et al. Fatal amniotic fluid embolism with typical patho-
histological, histochemical and clinical features. Forensic Sci Int.
2013;226:e16-9.
59. Ho CH, et al. Early application of extracorporeal membrane oxy-
genation in a patient with amniotic fluid embolism. Acta Anaesthe-
siol Taiwan. 2009;47:99-102.
60. Hosokawa S, et al.[A survival case of amniotic fluid embolism
treated by percutaneous cardiopulmonary support and thrombolysis
with tissue-plasminogen activator]. Nippon Naika Gakkai Zasshi.
2001;90:2074-6.
61. Hosono K, et al. Successful recovery from delayed amniotic fluid
embolism with prolonged cardiac resuscitation. J Obstet Gynaecol
Res. 2011;37:1122-5.
62. Hrgović Z, et al. [Fatal amnioinfusion with previous choriocarci-
noma in a parturient woman]. Zentralbl Gynakol. 2004;126:93-5.
63. Hsieh YY, et al. Successful application of extracorporeal membrane
oxygenation and intra-aortic balloon counterpulsation as lifesaving
therapy for a patient with amniotic fluid embolism. Am J Obstet
Gynecol. 2000;183:496-7.
64. Hunka I, et al. [Amniotic fluid embolism in a 24-year-old female
after caesarean section - a case report]. Kardiol Pol. 2003;59:325-7.
65. Huybrechts W, et al. Amniotic fluid embolism: a rare cause of acute
left-sided heart failure. Acta Cardiol. 2006;61:643-9.
66. Hwang JJ, et al. Successful resuscitation of amniotic fluid embolism
during cesarean section: a case report. Ma Zui Xue Za Zhi.
1993;31:191-4.
67. Ihara K, et al. Succesfull recovery from an acute kidney injury due
to amniotic fluid embolism. Intern Med. 2015;54:49-54.
68. Imanaka H, et al. Chest computed tomography of a patient reveal-
ing severe hypoxia due to amniotic fluid embolism: a case report. J
Med Case Rep. 2010;4:55.
69. James CF, et al. Massive amniotic fluid embolism: diagnosis aided
by emergency transesophageal echocardiography. Int J Obstet
Anesth. 2004;13:279-83.
70. Ječmenica D, et al. Amniotic fluid embolism-apropos two consec-
utive cases. J Forensic Sci. 2011;56(Suppl 1):S247-51.
71. Jelks A, et al. Nonpneumatic antishock garment combined with
bakri balloon as a nonoperative “uterine sandwich” for temporiza-
tion of massive postpartum hemorrhage from disseminated in-
travascular coagulation. Case Report Obstet Gynecol. 2015;
2015:124157.
72. Judich A, et al. Amniotic fluid embolism following blunt abdomi-
nal trauma in pregnancy. Injury. 1998;29:475-7.
73. Jung JY, et al. Usefulness of the bispectral index during cardiopul-
monary resuscitation -A case report-. Korean J Anesthesiol.
2013;64:69-72.
74. Kadiroğlu AK, et al. [A rare cause of disseminated intravascular co-
agulation and pulmonary thromboembolism and review of litera-
ture: amniotic fluid embolism]. Yoğun Bakım Derg. 2013;4:32-6.
75. Kahyaoglu I, et al. Factor VIIa treatment of DIC as a clinical man-
ifestation of amniotic fluid embolism in a patient with fetal demise.
Arch Gynecol Obstet. 2009;280:127-9.
76. Kamper CH, et al. [Double cardiac arrest caused by amniotic fluid em-
bolism during caesarean section]. Ugeskr Laeger. 2009;171:321-2.
77. Kane SK. Historical perspective of amniotic fluid embolism. Int
Anesthesiol Clin. 2005;43:99-108.
78. Kaneko Y, et al. Continuous hemodiafiltration for disseminated in-
travascular coagulation and shock due to amniotic fluid embolism:
report of a dramatic response. Intern Med. 2001;40:945-7.
79. Karsten J, et al. Electrical impedance tomography may optimize
ventilation in a postpartum woman with respiratory failure. Int J
Obstet Anesth. 2013;22:67-71.
80. Kelly MC, et al. A case of amniotic fluid embolism in a twin preg-
nancy in the second trimester. Int J Obstet Anesth. 1995;4:175-7.
81. Kendrick JE 4th, et al. Use of a fascial prosthesis for management
of abdominal compartment syndrome secondary to obstetric hem-
orrhage. Obstet Gynecol. 2006;107(2 Pt 2):493-6.
82. Kent KJ, et al. Presumed antepartum amniotic fluid embolism. Ob-
stet Gynecol. 2003;102:493-5.
83. Kim JY, et al. Decrease in bispectral index prior to cardiovascular
collapse during Caesarean sections. Acta Anaesthesiol Scand.
2014;58:123-6.
84. Koegler A, et al. Amniotic fluid embolism: a case with non-cardio-
genic pulmonary edema. Intensive Care Med. 1994;20:45-6.
85. Kostamovaara PA, et al. Severe maternal hypoxemia during a twin
pregnancy. Acta Obstet Gynecol Scand. 2000;79:82-3.
86. Kretzschmar M, et al. [Pathophysiological and therapeutic aspects
of amniotic fluid embolism (anaphylactoid syndrome of preg-
nancy): case report with lethal outcome and overview]. Anaesthe-
sist. 2003;52:419-26.
87. Kumar S, et al. Amniotic fluid embolism complicated by paradox-
ical embolism and disseminated intravascular coagulation. Am J
Crit Care. 2010;19:379-82.
88. Kumar V, et al. Paradoxical amniotic fluid embolism presenting be-
fore caesarean section in a woman with an atrial septal defect. Int J
Obstet Anesth. 2010;19:94-8.
89. Lee JH, et al. Amniotic fluid embolism that took place during an
emergent Cesarean section -A case report-. Korean J Anesthesiol.
2010;59Suppl:S158-62.
90. Legrand M, et al. [Amniotic fluid embolism: an update]. Ann Fr
Anesth Reanim. 2013;32:189-97.
91. Leitner L, et al. [Amniotic fluid embolism—a case report with pos-
itive outcome]. Anasthesiol Intensivmed Notfallmed Schmerzther.
1995;30:113-5.
92. Levy R, et al. Fetal bradycardia and disseminated coagulopathy:
atypical presentation of amniotic fluid emboli. Acta Anaesthesiol
Scand. 2004;48:1214-5.
93. Li N, et al. Successful resuscitation of an obstetric patient with sus-
pected amniotic fluid embolism—case report. Middle East J Anes-
thesiol. 2010;20:755-8.
94. Lim Y, et al. Recombinant factor VIIa after amniotic fluid embolism
and disseminated intravascular coagulopathy. Int J Gynaecol Ob-
stet. 2004;87:178-9.
95. Lionel J. Amniotic fluid embolism. Trop Doct. 2005;35:44-5.
96. Malhotra P, et al. Delayed presentation of amniotic fluid embolism:
lessons from a case diagnosed at autopsy. Respirology.
2007;12:148-50.
97. Manchanda R, et al. Anaphylactoid syndrome caused by amniotic
fluid embolism following manual removal of placenta. J Obstet Gy-
naecol. 2005;25:201-2.
98. Martin PS, et al. Emergency. Amniotic fluid embolism. Am J Nurs.
2001;101:43-4.
99. Mathelier AC, et al. Placenta previa and accreta complicated by am-
niotic fluid embolism. Int J Fertil Womens Med. 2006;51:28-32.
100. Mato J. Suspected amniotic fluid embolism following amniotomy:
a case report. AANA J. 2008;76:53-9.
101. Matsuda Y, et al. Amniotic fluid embolism: a comparison between
patients who survived and those who died. J Int Med Res.
2009;37:1515-21.
 U. Indraccolo, R. Ventrone, G. Scutiero, P. Greco, S.R. Indraccolo
102. McDonnell NJ, et al. Rapid reversal of critical haemodynamic com-
promise with nitric oxide in a parturient with amniotic fluid em-
bolism. Int J Obstet Anesth. 2007;16:269-73.
103. McDougall RJ, et al. Amniotic fluid embolism syndrome: case re-
port and review. Anaesth Intensive Care. 1995;23:735-40.
104. Meletti JF, et al. Amniotic fluid embolism during vaginal delivery
under analgesia. Case report. Rev Bras Anestesiol. 2008;58:399-
402,397-9.
105. Moon JM, et al. Reversible posterior leukoencephalopathy syn-
drome. J Emerg Med. 2010;38:e1-4.
106. Morau E, et al. [Amniotic fluid embolism during labor]. Ann Fr
Anesth Reanim. 2002;21:744-7.
107. Nagar MP, et al. Survival following amniotic fluid embolism and
cardiac arrest complicated by sub-capsular liver haematoma. Int J
Obstet Anesth. 2005;14:62-5.
108. Nagarsheth NP, et al. Successful placement of a right ventricular
assist device for treatment of a presumed amniotic fluid embolism.
Anesth Analg. 2008;107:962-4.
109. Noble WH, et al. Amniotic fluid embolus. Can J Anaesth.
1993;40:971-80.
110. Ocegueda-Pacheco C, et al. Therapeutic hypothermia for cardio-
vascular collapse and severe respiratory distress after amniotic fluid
embolism. Ther Hypothermia Temp Manag. 2014;4:96-8.
111. Ogihara T, et al. Continuous hemodiafiltration for potential amni-
otic fluid embolism: dramatic responses observed during a 10-year
period report of three cases. Ther Apher Dial. 2012;16:195-7.
112. Patel D, et al. An example of prompt and appropriate multidiscipli-
nary management leading to an exceptionally good outcome: a case
complicated by amniotic fluid embolism. BMJ Case Rep.
2015;2015.pii:bcr2015211462.
113. Peitsidou A, et al. Amniotic fluid embolism managed with success
during labour: report of a severe clinical case and review of litera-
ture. Arch Gynecol Obstet. 2008;277:271-5.
114. Penfold PR, et al. Amniotic fluid embolism in an HIV-positive par-
turient. S Afr J Anaesthesiol Analg. 2010;16:43-7.
115. Perdrix C, et al. [Fatal amniotic fluid embolism. Extraction by Cae-
sarean section of a living child without neurological after-effects].
Ann Fr Anesth Reanim. 2004;23:912-6.
116. Pevey WJ. An unusual occurrence of amniotic fluid embolism. J La
State Med Soc. 1990;142:31-2.
117. Pluymakers C, et al. Amniotic fluid embolism after surgical trauma:
two case reports and review of the literature. Resuscitation.
2007;72:324-32.
118. Porat S, et al. Transient intracardiac thrombi in amniotic fluid em-
bolism. BJOG. 2004;111:506-10.
119. Prosper SC, et al Recombinant factor VIIa to successfully manage
disseminated intravascular coagulation from amniotic fluid em-
bolism. Obstet Gynecol. 2007;109(2 Pt 2):524-5.
120. Quinn A, et al. Delayed onset of coagulopathy following amniotic
fluid embolism: two case reports. Int J Obstet Anesth. 1993;2:177-80.
121. Rainio J, et al. Amniotic fluid embolism as cause of death in a car
accident—a case report. Forensic Sci Int. 2003;137:231-4.
122. Rault S, et al. [Amniotic fluid embolism: two case reports]. Gy-
necol Obstet Fertil. 2006;34:127-30.
123. Ray BK, et al. Amniotic fluid embolism with second trimester preg-
nancy termination: a case report. Can J Anaesth. 2004;51:139-44.
124. Revello-Álvarez R, et al [Amniotic fluid embolism. Report of four
cases at Hospital Universitario La Paz, in Madrid, Spain, and re-
view of literature]. Rev Colomb Obstet Gynecol. 2013;64:60-6.
125. Robillard J, et al. The syndrome of amniotic fluid embolism: a po-
tential contribution of bradykinin. Am J Obstet Gynecol.
2005;193:1508-12.
126. Rodrigues R, et al. [Amniotic fluid embolism]. Acta Med Port.
2007;20:93-5.
127. Rogers WK, et al. Acute hypotension associated with intraopera-
tive cell salvage using a leukocyte depletion filter during manage-
ment of obstetric hemorrhage due to amniotic fluid embolism.
671
Anesth Analg. 2013;117:449-52.
128. Sanders GM. Amniotic fluid embolism in a patient with SC sickle
cell disease. Anaesthesia. 1999;54:614-6.
129. Shechtman M, et al. Amniotic fluid embolism: early findings of
transesophageal echocardiography. Anesth Analg. 1999;89:1456-8.
130. Shen HP, et al. Amniotic fluid embolism treated with emergency
extracorporeal membrane oxygenation: a case report. J Reprod
Med. 2009;54:706-8.
131. Shojai R, et al. Amniotic fluid embolism during late term termina-
tion of pregnancy. Prenat Diagn. 2003;23:950-1.
132. Sisodia SM, et al. Amniotic fluid embolism: a cause of sudden ma-
ternal death and police inquest. Am J Forensic Med Pathol.
2013;33:330-4.
133. Sistac Ballarín JM, et al. [Amniotic fluid embolism during the birth
process. Apropos of an atypical presentation]. Rev Esp Anestesiol
Reanim. 2000;47:424-5.
134. Sprung J, et al. Amniotic fluid embolism during epidural anesthe-
sia for cesarean section. Acta Anaesthesiol Belg. 1991;42:225-31.
135. Stanten RD, et al. Amniotic fluid embolism causing catastrophic
pulmonary vasoconstriction: diagnosis by transesophageal echocar-
diogram and treatment by cardiopulmonary bypass. Obstet Gynecol.
2003;102:496-8.
136. Stehr SN, et al. Closing the gap between decision and delivery—am-
niotic fluid embolism with severe cardiopulmonary and haemostatic
complications with a good outcome. Resuscitation. 2007;74:377-81.
137. Stiller RJ, et al. Successful pregnancy after suspected anaphylac-
toid syndrome of pregnancy (amniotic fluid embolus). A case re-
port. J Reprod Med. 2000;45:1007-9.
138. Stroup J, et al. Aprotinin in the management of coagulopathy associ-
ated with amniotic fluid embolus. Pharmacotherapy. 2006;26:689-93.
139. Syed SA, Dearden CH: Amniotic fluid embolism: emergency man-
agement. J Accid Emerg Med. 1996;13:285-6.
140. Tan K, et al. Maternal-fetal survival following amniotic fluid em-
bolism: 2 case reports. Internet J Anesthesiol. 2001;5.
141. Terasako K. Amniotic fluid embolism. J Anesth. 2001;15:238.
142. Todo Y, et al. Therapeutic application of C1 esterase inhibitor con-
centrate for clinical amniotic fluid embolism: a case report. Clin
Case Rep. 2015;3:673-5.
143. Tomasa TM, et al. Amniotic fluid embolism diagnosed by fibre-
optic bronchoscopy. Anaesthesia. 2010;65:1230-1.
144. Tramoni G, et al. Amniotic fluid embolism during caesarean sec-
tion. Int J Obstet Anesth. 2004;13:271-4.
145. van Dorp W, et al. [Resuscitation of a pregnant patient—don’t hes-
itate to perform a perimortem caesarean section]. Ned Tijdschr Ge-
neeskd. 2010;154:A2370.
146. Van Heerden PV, et al. Inhaled aerosolized prostacyclin as a selec-
tive pulmonary vasodilator for the treatment of severe hypoxaemia.
Anaesth Intensive Care. 1996;24:87-90.
147. Vanmaele L, et al. Transient left heart failure in amniotic fluid em-
bolism. Intensive Care Med. 1990;16:269-71.
148. Vellayappan U, et al. Paradoxical embolization by amniotic fluid
seen on the transesophageal echocardiography. Anesth Analg.
2009;108:1110-2.
149. Verroust N, et al. Ventricular fibrillation during termination of preg-
nancy. Lancet. 2007;369:1900.
150. Wallace F, et al. Successful resuscitation following amniotic fluid
embolism in a patient undergoing induction of labour for late mis-
carriage. Int J Obstet Anesth. 2012;21:202-3.
151. Weksler N, et al. Continuous arteriovenous hemofiltration in the treat-
ment of amniotic fluid embolism. Int J Obstet Anesth: 1994;3:92-6.
152. Wilhite L, et al. Amniotic fluid embolism: a case study. Am J Nurs.
2008;108:83-5.
153. Yao J, et al. Surgical treatment of late-onset amniotic fluid pul-
monary embolism. J Obstet Gynaecol. 2014;34:209-11.
154. Yufune S, et al. Successful resuscitation of amniotic fluid embolism
applying a new classification and management strategy. JA Clin
Rep. 2015;1:1-5.
672 Interventions for treating amniotic fluid embolism: a systematic review with meta-analysis

Table 3. — Descriptive statistics with negative and positive scores.

Atypical AFE Atypical AFE + Typical AFE Typical AFE + Death Onset before
16.4% confirmations* 42.4% confirmations* 29.9% delivery
15.3% 26% 59.9%
Pulmonary Negative 0 31% 22.2% 10.7% 15.2% 18.9% 14.2%
interventions scores 1 41.4% 55.6% 30.7% 50% 50.9% 42.5%
Positive 2 24.1% 14.8% 49.3% 21.7% 26.4% 36.8%
scores 3 3.4% 7.4% 9.3% 13% 3.8% 6.6%
Cardiac Negative 0 27.6% 29.6% 9.3% 17.4% 22.6% 14.2%
interventions scores 1 41.4% 48.1% 33.3% 47.8% 50.9% 38.7%
Positive 2 31% 7.4% 50.7% 13% 22.6% 39.6%
scores 3 0% 14.8% 6.7% 21.7% 3.8% 7.5%
Interventions on Negative 0 31% 51.9% 10.7% 13% 24.5% 20.8%
the coagulopathy scores 1 31% 33.3% 26.7% 47.2% 47.2% 32.1%
Positive 2 24.1% 7.4% 48% 19.6% 18.9% 34.9%
scores 3 13.8% 7.4% 14.7% 21.7% 9.4% 12.3%
Immediate Negative 0 69% 70.4% 50.7% 60.9% 60.4% 49.1%
delivery scores 1 24.1% 14.8% 17.3% 23.9% 20.8% 25.5%
Positive 2 3.4% 7.4% 24% 10.9% 15.1% 17%
scores 3 3.4% 7.4% 8% 4.3% 3.8% 8.5%
*Pathological or blood/serum confirmations (autopsy, pathological examination of specimens, bronco-alveolar lavage, blood keratinocytes, blood amniocytes,
rise serum in IGF-BP1, and/or Syalil Tn, and/or Zn CP1, and/or tryptase).

Table 4. — Multivariate logistic regression analyses, cheeking for heterogeneity.

Pulmonary interventions Cardiac interventions Interventions on Immediate delivery
OR for positive scores OR for positive scores the coagulopathy OR for positive scores
OR for positive scores
Adjusted OR Adjusted OR Adjusted OR Adjusted OR
95% CI 95% CI 95% CI 95% CI
p p p p
AFE diagnosis + + + +
Atypical AFE 1 1 1 1

Atypical AFE + confirmations* 0.750 0.771 0.357 2.294

0.222-2.538 0.223-2.667 0.095-1.338 0.381-13.802
0.644 0.681 0.126 0.364

Typical AFE 3.726 3.240 2.874 6.582

1.463-9.491 1.273-8.244 1.174-7.035 1.433-30.227
0.006 0.014 0.021 0.015

Typical AFE + confirmations* 1.400 1.809 1.669 2.629

0.507-3.865 0.619-5.283 0.604-4.618 0.501-13.779
0.516 0.279 0.323 0.253
Death 0.598 0.428 0.399 1.037
0.281-1.275 0.195-0.939 0.184-0.866 0.416-2.584
0.183 0.034 0.020 0.938
Onset before delivery 1.215 1.913 1.244 2.149
0.635-2.327 0.984-3.720 0.645-2.402 0.944-4.894
0.556 0.056 0.515 0.069
*Pathological or blood/serum confirmations (autopsy, pathological examination of specimens, bronco-alveolar lavage, blood keratinocytes, blood amniocytes, rise
serum in IGF-BP1, and/or Syalil Tn, and/or Zn CP1, and/or tryptase).

be of some effectiveness, 2 if the interventions were perceived as
effective, and 3 if the interventions were perceived to be very ef-
fective. Consensus among meta-analysts was used to carefully at-
tribute those scores. Thus 0 and 1 were considered negative
scores, while 2 and 3 were considered positive scores.
The AFE diagnosis is entirely clinical, basing on a typical triad
of signs [4]. These are: cardiac arrest/ hypotension, respiratory
failure, coagulopathy. Some additional information supporting a
presumptive amniotic embolization can be provided pathologi-
cally or by checking for some component of the amniotic fluid [2,
5-8] in maternal blood. On the other hand, it has been reported
that some cases of AFE are atypical [9], usually presenting with
 U. Indraccolo, R. Ventrone, G. Scutiero, P. Greco, S.R. Indraccolo 673

Table 5. — Analyses on the whole database.

WHOLE DATABASE
(177 cases; 53 died, 124 survived)
Pulmonary interventions
Scores Given Random
-0 30 (16.9%) 34 (19.2%)
-1 73 (41.2%) 53 (29.9%)
-2 58 (32.8%) 56 (31.6%)
-3 16 (9%) 34 (19.2%)
Chi square 9.94, p = 0.01909, 3 degrees of freedom.
Cardiac interventions
Scores Given Random
-0 31 (17.5%) 32 (18.1%)
-1 72 (40.7%) 56 (31.6%)
-2 61 (34.5%) 57 (32.2%)
-3 13 (7.3%) 32 (18.1%)
Chi square 10.17, p = 0.01715, 3 degrees of freedom.
Interventions on the coagulopathy
Scores Given Random
-0 37(20.9%) 32 (18.1%)
-1 59 (33.3%) 48 (27.1%)
-2 54 (30.5%) 70 (39.5%)
-3 27 (15.3%) 27 (15.3%)
Chi square 3.558, p = 0.3134, 3 degrees of freedom.
Immediate birth
Scores Given Random
-0 105 (59.3%) 25 (14.1%)
-1 35 (19.8%) 61 (34.5%)
-2 26 (14.7%) 64 (36.2%)
-3 11 (6.2%) 27 (15.3%)
Chi square 79.05, p < 0.0001, 3 degrees of freedom.

signs and symptoms of a coagulopathy without other clinical man-
ifestations. Therefore, the AFE diagnosis is elusive, and should
be considered after the exclusion of other diseases mimicking an
AFE.
By reviewing a database of cases reported as AFEs, the authors
acknowledge that some cases are true and typical AFE cases and
that other cases are atypical AFE cases or may not be AFE cases.
They therefore were aware that they were analysing a heteroge-
neous database. The clinical picture of AFE along with the AFE
outcome (death or survival) and its onset (before and after birth)
could have conditioned the authors’ perception of the effective-
ness of their interventions. Therefore, they distinguished AFE
cases as typical (cardiovascular collapse/hypotension, pulmonary
failure, coagulopathy), or atypical (lack of one or more of the typ-
ical signs and symptoms), with or without pathological or blood
confirmations. The pathological or laboratory confirmations were
considered to be: autopsy findings of amniotic debris in the pul-
monary bed, pathological examination of uterine specimens
demonstrating amniotic debris in uterine vessels, bronco-alveo-
lar lavage with amniocytes or keratinocytes, keratinocytes in ma-
ternal blood, amniocytes in maternal blood, rise in IGF-BP1
and/or Syalil Tn, and/or Zn CP1, and/or tryptase in maternal
blood).
Logistic regression analyses were built to check the hetero-
geneity of the sample, assuming that the clinical picture of AFE
(typical with pathological or laboratory confirmations, typical
without any confirmation, atypical with pathological or labora-
tory confirmations, atypical without any confirmation), AFE out-
come (death or alive), AFE onset (before and after birth) can have
conditioned the positive scores.
To compare scores assigned to the interventions in AFE cases,
the authors generated a random sequence of 177 numbers, from 0
to 3, by using Open.epi 3.03a, and they used the frequencies of
such random distribution as a contrast. They checked the null hy-
pothesis that the scores given from textual information were ran-
domly distributed. Based on results of logistic regression analyses,
they performed sub-groups analyses checking that, in the sub-
groups, the scores given from textual information were also ran-
domly distributed. The Chi square test was used for inference and
p < 0.05 was set for the level of significance.
Results
Table 1 lists the discarded references. Table 2 reports the
list of 154 references used for meta-analysis. Table 3 re-
ports the descriptive statistics, focusing on the rates of pos-
itive and negative scores.
Logistic regression analyses demonstrated that more pos-
itive scores were found in typical AFE cases, whereas ob-
viously more negative scores were found if the patient died
(Table 4). These results confirm heterogeneity. Therefore,
the authors performed sub-group analyses on patients with
674 Interventions for treating amniotic fluid embolism: a systematic review with meta-analysis

Table 6. — Sub-group analyses: typical AFEs with or without pathological and/or lab confirmations of AFE.
TYPICAL AFEs WITH OR WITHOUT AFE CONFIRMATIONS*
(121 cases; died 37, survived 84)
Pulmonary interventions
Scores Given Random
-0 15 (12.4%) 26 (21.4%)
-1 46 (38%) 38 (31.4%)
-2 47 (38.8%) 34 (28.1%)
-3 13 (10.7%) 23 (19%)
Chi square 8.577, p = 0.03547, 3 degrees of freedom.
Cardiac interventions
Scores Given Random
0 15 (12.4%) 20 (16.5%)
-1 47 (38.8%) 38 (31.4%)
-2 50 (41.3%) 41 (33.9%)
-3 9 (7.4%) 22 (18.2%)
Chi square 8.009, p = 0.04583, 3 degrees of freedom.
Interventions on the coagulopathy
Scores Given Random
-0 14 (11.6%) 22 (18.2%)
-1 41 (33.9%) 35 (28.9%)
-2 45 (37.2%) 50 (41.3%)
-3 21 (17.4%) 14 (11.6%)
Chi square 3.915, p = 0.2708, 3 degrees of freedom.
Immediate birth
Score Given Random
-0 66 (54.5%) 16 (13.2%)
-1 24 (19.8%) 41 (33.9%)
-2 23 (19%) 49 (40.5%)
-3 8 (6.6%) 15 (12.4%)
Chi square 46.45, p < 0.00001, 3 degrees of freedom.
*Pathological or blood/serum confirmations (autopsy, pathological examination of specimens, bronco-alveolar lavage, blood keratinocytes, blood amniocytes, rise
serum in IGF-BP1, and/or Syalil Tn, and/or Zn CP1, and/or tryptase).

typical AFE (with and without pathological or laboratory
confirmation), in patients with typical AFE (without any
pathological or laboratory confirmation) and in patients
who died.
Table 5 reports results for the whole database. Pulmonary
interventions (Chi square 9.94, p = 0.01909) and cardiac
interventions (Chi square 10.17, p = 0.01715) are perceived
to be of some relevance in treating AFE cases (higher rates
for score 1) while it is not perceived as pivotal (Chi square
79.05, p < 0.0001) to immediately giving birth (higher rate
for score 0).
Table 6 reports results on typical AFE cases (with and
without pathological and laboratory confirmation). Pul-
monary interventions (Chi square 8.577, p = 0.03547) and
cardiac interventions (Chi square 8.009, p = 0.04583) are
perceived as effective in treating AFE cases (higher rates
for score 2), while it is not perceived as pivotal (Chi square
46.45, p < 0.00001) to immediately giving birth (higher rate
for score 0).
Table 7 reports results on typical AFE cases (without
pathological and laboratory confirmation). Pulmonary in-
terventions (Chi square 12.1, p = 0.007035) are perceived
as effective in treating AFE cases (higher rate for score 2).
Cardiac interventions fail to reach the significance level
(Chi square 7.409, p = 0.05995), despite the higher rate for
score 2). It is not perceived as pivotal (Chi square 30.89, p
< 0.00001) to immediately giving birth (higher rate for
score 0).
Table 8 reports results on cases of death for AFE. Pul-
monary interventions (Chi square 14.59, p = 0.002198) and
cardiac interventions (Chi square 9.667, p = 0.02162) are
perceived of some relevance in treating AFE cases (higher
rates for score 1) while it is not perceived as pivotal (Chi
square 22.43, p = 0.00005) to immediately giving birth
(higher rate for score 0). Interventions on the coagulopa-
thy fail to reach significance (p = 0.0522) and do not seem
pivotal.
Many authors of AFE case reports agree that immediate
interventions should be provided and some reports show
positive findings if extracorporeal membrane oxygenation
is provided.
Discussion
The present meta-analysis is adapted to case reports, and
aimed to synthesize the “a priori” perception of the effec-
 U. Indraccolo, R. Ventrone, G. Scutiero, P. Greco, S.R. Indraccolo 675

Table 7. — Sub-group analyses. Typical AFE cases without any pathological and/or lab confirmation of AFE.
TYPICAL AFEs WITHOUT AFE CONFIRMATIONS*
(75 cases: 12 died, survived 63)
Pulmonary interventions
Scores Given Random
-0 8 (10.7%) 15 (20%)
-1 23 (30.7%) 24 (32%)
-2 37 (49.3%) 19 (25.3%)
-3 7 (9.3%) 17 (22.7%)
Chi square 12.1, p = 0.007035, 3 degrees of freedom.
Cardiac interventions
Scores Given Random
-0 7 (9.3%) 13 (17.3%)
-1 25 (33.3%) 22 (29.3%)
-2 38 (50.7%) 27 (36%)
-3 5 (6.7%) 13 (17.3%)
Chi square 7.409, p = 0.05995, 3 degrees of freedom.
Interventions on the coagulopathy
Scores Given Random
-0 8 (10.7%) 14 (18.7%)
-1 20 (26.7%) 22 (29.3%)
-2 36 (48%) 29 (38.7%)
-3 11 (14.7%) 10 (13.3%)
Chi square 2.533, p = 0.4693, 3 degrees of freedom.
Immediate birth
Score Given Random
-0 38 (50.7%) 7 (9.3%)
-1 13 (17.3%) 28 (37.3%)
-2 18 (24%) 31 (41.3%)
-3 6 (8%) 9 (12%)
Chi square 30.89, p < 0.00001, 3 degrees of freedom.
*Pathological or blood/serum confirmations (autopsy, pathological examination of specimens, bronco-alveolar lavage, blood keratinocytes, blood amniocytes, rise
serum in IGF-BP1, and/or Syalil Tn, and/or Zn CP1, and/or tryptase).

tiveness of interventions provided by authors who man-
aged AFE cases and who wrote their case reports. Usu-
ally, in contrast, the traditional meta-analysis summarizes
“a posteriori” results, by averaging an effect size,
weighted for the inverse of the variance [10]. In very rare
diseases, however, we cannot build large series to obtain
data by observational or randomized studies, and, there-
fore, we cannot meta-analyse the data by averaging an
effect size weighted for the inverse of the variance. The
knowledge from the rarer diseases must be drawn from
case reports or small series, which are written by authors
focusing on their personal feelings, knowledge, and ex-
pertise. This subjective way to communicate information
may be biased by personal opinions and produce evi-
dence of limited quality. Therefore, readers may feel it is
inappropriate to perform an analysis starting from sub-
jective scores given from meta-analysts to the texts. It
has been reported by some authors [11, 12] that semi-
quantitative scores from textual information can summa-
rize each personal feelings in a more objective way,
allowing statistical inference and improving the level of
evidence. This is what already happens when practice
guidelines are issued by opinion leaders. Interestingly,
findings from the current meta-analysis are reported in
the same way in recent practice guidelines for the man-
agement of AFE [13].
The elusive diagnosis of AFE is always a concern. In na-
tional data sources or registries, stringent clinical criteria
were adopted for indexing cases as AFE. This policy leads
to the loss of atypical AFE cases, which could have a more
favourable prognosis [9]. The present authors decided to
perform sub-group analysis for assessing if some interven-
tions are more effective in more likely cases of AFE (typi-
cal ones, with and without laboratory confirmation),
thereby remedying the heterogeneity of the data. They
found that each intervention aiming to support the heart and
the lung was perceived as effective in treating typical AFEs.
More pivotal seems to be support for the lung function.
Typical AFE leads to cardiovascular and pulmonary fail-
ure: it is therefore logical to believe that supporting the
heart and the lungs does improve AFE outcome. The body
of evidence from case reports agrees that AFE requires im-
mediate interventions. This is also in agreement with what
was reported by Fitzpatrick et al. [14], who stated that the
676 Interventions for treating amniotic fluid embolism: a systematic review with meta-analysis

Table 8. — Sub-group analyses. Analyses of patients who died.

DEATHS
53 cases
Pulmonary interventions
Scores Given Random
-0 10 (18.9%) 13 (24.5%)
-1 27 (50.9%) 10 (18.9%)
-2 14 (26.4%) 20 (37.7%)
-3 2 (3.8%) 10 (18.9%)
Chi square 14.59, p = 0.002198, 3 degrees of freedom.
Cardiac interventions
Scores Given Random
-0 12 (22.6%) 11 (20.8%)
-1 27 (50.9%) 15 (28.3%)
-2 12 (22.6%) 17 (32.1%)
-3 2 (3.8) 10 (18.9%)
Chi square 9.667, p = 0.02162, 3 degrees of freedom.
Interventions on the coagulopathy
Scores Given Random
-0 13 (24.5%) 8 (15.1%)
-1 25 (47.2%) 16 (30.2%)
-2 10 (18.9%) 21 (39.6%)
-3 5 (9.4%) 8 (15.1%)
Chi square 7.762, p = 0.05120, 3 degrees of freedom.
Immediate birth
Score Given Random
-0 32 (60.4%) 9 (17%)
-1 11 (20.1%) 16 (30.2%)
-2 8 (15.1%) 21 (39.6%)
-3 2 (3.8%) 7 (13.2%)
Chi square 22.43, p = 0.00005, 3 degrees of freedom.

AFE outcome can be improved if immediate interventions

are provided. Each intervention for sustaining the heart and
lungs is effective. From a practical point of view, after im-
mediate resuscitation, it should be considered appropriate
to perform a cardiopulmonary by-pass and extracorporeal
membrane oxygenation, because this procedure also allows
the purification of blood from amniotic debris, thereby
avoiding worsening of the AFE [15]. The same concept was
expressed by Ihara et al. [16] in a case of renal replacement
therapy during AFE. However, it must be pointed out that
the present authors were unable, from the available data, to
establish if cardiopulmonary by-pass and extracorporeal
membrane oxygenation can change the AFE outcome.
Interestingly, the efforts of treating the coagulopathy are
not perceived as pivotal for the managing of AFE. Atypical
cases of AFE are usually the ones presenting with the co-
agulopathy only. In such cases, authors have not high-
lighted the importance of the supportive treatment of the
disseminated intravascular coagulation (DIC). This per-
ception of authors could be explained because DIC has
overall a standard treatment [17, 18], irrespective from the
cause of DIC.
Figure 1. — Flow-chart of the systematic review.
Finally, the immediate birth approach is rarely reported
by authors as pivotal for managing AFE, despite it being
acknowledged that it allows more effective resuscitation,
improving both maternal and fetal survival chances [19].
This is due to the omission of authors of discussing and re-
porting the topic. As a result, the present authors found
higher rates of the 0 score for the item “immediate birth”.
Maybe those authors have implied that their readers are
well aware that successful resuscitation is more likely after
giving immediate birth, though this is certain.
Further study and national data sources should assess
which impact would have on the outcome of AFE each in-
tervention able to filtrate and purify the blood from amni-
otic debris, and to what extent pulmonary and heart support
along with blood purification are effective for improving
the AFE outcome.
 U. Indraccolo, R. Ventrone, G. Scutiero, P. Greco, S.R. Indraccolo
Conclusion
This meta-analysis provides “a priori” evidence that each
immediate intervention for supporting the heart and,
mostly, the lungs in AFE cases are perceived as effective for
the management of AFE.
Acknowledgements
The authors are grateful to Dr. L.G. Aguilera (Servicio
de Anestesiología y Reanimación, Hospital del Mar-Es-
perança. IMAS, Hospital Clínic, Barcelona, Spain), Dr.
M. Dabrowski (Klinika Kardiologii, Instytutu Kardi-
ologii, Szpital Bielanski, Zespol Badawczo-Leczniczy
Chorob Ukladu Krazenia ICMDiK PAN, Warsaw,
Poland), Dr. E.A. Bouman (Maastricht Universitair
Medisch Centrum, Anesthesiology and Pain Medicine,
Maastricht, Netherlands), Dr. W. van Dorp (Erasmus MC,
Department of Obstetrics and Gynaecology, Rotterdam,
Netherlands), Dr. L. Baghirzada (Department of Anaes-
thesia, University of Calgary, Calgary, Canada), Prof. S.
Gerli and Prof. G.C. Di Renzo (Dipartimento di Ostetricia
e Ginecologia – Azienda Ospedaliera di Perugia, Univer-
sità di Perugia, Perugia, Italy), Dr. R. Bøgeskov (Anæste-
siologisk Afdeling, Herlev Hospital, Denmark), Dr. M.
Ben Ismail (Service de gynécologie obstétrique, centre
hospitalier François Quesnay, Mantes-la-Jolie cedex,
France) who assisted in finding some articles. The authors
also thank Prof. E. Indraccolo (Dipartimento di Scienze
Economiche e Statistiche, Università di Salerno, Salerno,
Italy) who freely translated from German, and Mrs. A.
Dobrowolska (Unità Operativa Complessa di Ematologia
e Medicina Interna, Ospedale di Civitanova Marche, Area
Vasta 3 - Marche, Civitanova Marche (MC), Italy) who
freely translated from Polish.
References
[1] Conde-Agudelo A., Romero R.: “Amniotic fluid embolism: an evi-
dence-based review”. Am. J. Obstet. Gynecol., 2009, 201, 445.e1.
[2] Clark SL.: “Amniotic fluid embolism”. Obstet. Gynecol., 2014,
123(2 Pt 1), 337.
[3] Gilbert W.M., Danielsen B.: “Amniotic fluid embolism: decreased
mortality in a population-based study”. Obstet. Gynecol., 1999, 93,
973.
[4] Clark S.L., Hankins G.D.V., Dudley D.A., Dildy G.A., Porter T.F.:
“Amniotic fluid embolism: analysis of the national registry”. Am. J.
Obstet. Gynecol., 1995, 172, 1158.
[5] Legrand M., Rossignol M., Dreux S., Luton D., Ventré C., Barranger
E., et al.: “Diagnostic accuracy of insulin-like growth factor binding
protein-1 for amniotic fluid embolism”. Crit. Care Med., 2012, 40,
677
2059.
[6] Wernet A., Luton D., Muller F., Ducarme G.: “Use of insulin-like
growth factor binding protein-1 for retrospective diagnosis of amni-
otic fluid embolism in first trimester”. Arch. Gynecol. Obstet., 2014,
289, 461.
[7] Kanayama N., Tamura N.: “Amniotic fluid embolism: pathophysi-
ology and new strategies for management”. J. Obstet. Gynaecol.
Res., 2014, 40, 1507.
[8] Oi H., Naruse K., Noguchi T., Sado T., Kimura S., Kanayama N., et
al.: “Fatal factors of clinical manifestations and laboratory testing in
patients with amniotic fluid embolism”. Gynecol. Obstet. Invest.,
2010, 70, 138.
[9] Shen F., Wang L., Yang W., Chen Y.: “From appearance to essence:
10 years review of atypical amniotic fluid embolism”. Arch. Gynecol.
Obstet., 2016, 293, 329.
[10] Lipsey M.W., Wilson D.B.: “Analysis issues and strategies”. In:
Bickman L., Rog D.J., (eds). Practical meta-analysis. Thousand
Oaks, London, New Delhi: SAGE publications, Inc., 2001, 105.
[11] Indraccolo U., Graziani C., Di Iorio R., Corona G., Bonito M., In-
draccolo SR.: “External cephalic version for singleton breech pres-
entation: proposal of a practical check-list for obstetricians”. Eur.
Rev. Med. Pharmacol. Sci., 2015, 19, 2340.
[12] Indraccolo U., Di Iorio R., Matteo M., Corona G., Greco P., Indrac-
colo SR.: “The pathogenesis of endometrial polyps: a systematic,
semi-quantitative review”. Eur. J. Gynaecol. Oncol., 2013, 34, 5.
[13] Society for Maternal-Fetal Medicine (SMFM) with the assistance of.
Electronic address: pubs@smfm.org, Pacheco L.D., Saade G., Han-
kins G.D., Clark S.L.: “Amniotic fluid embolism: diagnosis and
management”. Am. J. Obstet. Gynecol., 2016, 215, B16.
[14] Fitzpatrick K., Tuffnell D., Kurinczuk J., Knight M.: “Incidence, risk
factors, management and outcomes of amniotic-fluid embolism: a
population-based cohort and nested case-control study”. BJOG,
2015, 123, 100.
[15] Oghiara T., Morimoto K., Kaneko Y.: “Continuous hemodiafiltration
for potential amniotic fluid embolism: dramatic responses observ-
ing during 10-year period report of three cases”. Ther. Apher. Dial.,
2012, 16, 195.
[16] Ihara K., Naito S., Okado T., Rai T., Mori Y., Toda T., et al.: “Suc-
cessful recovery from an acute kidney injury due to amniotic fluid
embolism”. Intern. Med., 2015, 54, 49.
[17] Levi M., Toh C.H., Thachil J., Watson H.G.: “Guidelines for the di-
agnosis and management of disseminated intravascular coagulation.
British Committee for Standards in Haematology”. Br. J. Haematol.,
2009, 145, 24.
[18] Di Nisio M., Baudo F., Cosmi B., D’Angelo A., De Gasperi A.,
Malato A., et al., Italian Society for Thrombosis and Haemostasis.:
“Diagnosis and treatment of disseminated intravascular coagulation:
guidelines of the Italian Society for Haemostasis and Thrombosis
(SISET)”. Thromb. Res., 2012, 129, e177.
[19] Benson M.D., Padovano A., Bourjeily G., Zhou Y.: “Maternal col-
lapse: challenging the four-minute rule”. EbioMedicine, 2016, 6,
253.
Corresponding Author:
U. INDRACCOLO, M.D., PHD.
Via P. Veronese 2/c
06024 Gubbio (PG) (Italy)
e-mail: ugo.indraccolo@libero.it
CEOG Clinical and Experimental
Obstetrics & Gynecology

Original Articles

Indications, limitations and complications of operative

hysteroscopy: a retrospective study of an 8-year experience

D. Caserta, S. Picchia, E. Ralli, E. Matteucci, L. Di Benedetto, M. Mallozzi,

G. Adducchio, R. Di Iorio, M. Moscarini†
Department of Obstetrics, Gynaecological and Urological Sciences, Sant’Andrea Hospital, “Sapienza” University of Rome, Rome (Italy)

Summary
Operative hysterectomy (HSC) is now considered the gold standard treatment of most benign intrauterine pathologies [2]. The exam
is performed using general anesthesia in day surgery procedure. Operative HSC enables the gynecologist to make diagnoses, obtain tar-
geted endometrial specimens for histological examination, apply therapies (e.g. endometrial ablation), and perform a variety of surgi-
cal procedures (e.g. adhesiolysis, myomectomy, polypectomy). Operative HSC is also indicated for Müllerian anomalies (e.g. uterine
septa), retained intrauterine contraceptives, endocervical lesions, and abnormal uterine bleeding unresponsive to medical treatment.
The aim of the study was to analyze hysteroscopic procedures performed over an 8-year experience, highlighting indications, limita-
tions, and complications of this technique in a sample of 1,412 women.

Key words: Operative hysterectomy; Benign intrauterine pathologies; Endometrial specimens.

Introduction hysteroscopy can be also classified into those caused by

The first successful operative hysteroscopy (HSC) was
reported by Pantaleoni in 1869 [1]. Its use has increased
with time and within few years it has become very popular
among gynecologists. Operative HSC is now considered
the gold standard treatment of most benign intrauterine
pathologies [2]. The exam is performed using general anes-
thesia in day surgery procedure. Operative HSC enables the
gynecologist to make diagnoses, obtain targeted endome-
trial specimens for histological examination, apply thera-
pies (e.g. endometrial ablation), and perform a variety of
surgical procedures (e.g. adhesiolysis, myomectomy, and
polypectomy). Operative HSC is also indicated for Mül-
lerian anomalies (e.g. uterine septa), retained intrauterine
contraceptives, endocervical lesions, and abnormal uterine
bleeding unresponsive to medical treatment [3]. As regards
hysteroscopy-related complications, they are limited and
can be categorized in intraoperative (e.g. cervical lacera-
tions, uterine perforations, hemorrhages, bowel or bladder
injuries, gas embolization, fluid overload, hyponatremia,)
and postoperative (e.g. endometritis, postoperative
synechiae, haematometra, procedure failure, myometrial
damage, and obstetrical morbidity) [3-5]. Complications of
hysteroscopic approach (e.g. perforation) and those caused
by hysteroscopic technique (e.g. electrosurgery, inflow
pressure) [6]. However, with strict preoperative evaluation,
rigorous procedure and monitoring, complications are
largely preventable. The short operating times and the
avoidance of cutting too deeply into the myometrium are
some of the parameters to be considered when hys-
teroscopy is being performed [7]. Technologic advances,
ongoing research, and postgraduate training in hystero-
scopic technique continue to expand the safe and benefi-
cial applications of hysteroscopy into the next century [8].
The aim of the study was to analyze hysteroscopic proce-
dures performed over 8-year experience, highlighting indi-
cations, limitations, and complications of this technique in
a sample of 1,412 women.
Materials and Methods
This retrospective study was performed on data stored in the
database of the Gynecological Unit of Sant’Andrea Hospital of
Rome (“Sapienza” University of Rome), pertaining to all patients
who underwent operative hysteroscopy between January 2005 to
May 2013 (n=1,412). The authors examined: the age of the pa-

Revised manuscript accepted for publication June 30, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3031.2017
 D. Caserta, S. Picchia, E. Ralli, E. Matteucci, L. Di Benedetto, M. Mallozzi, G. Adducchio, R. Di Iorio, M. Moscarini
tients, the medical histories, the indications for surgery, the sur-
gical procedure, the duration of the procedures (operating time),
the causes of the eventual transfer to the ordinary hospitalization,
the intraoperative and postoperative complications, the results of
histological examinations, and the results of the gynecological
exam performed 15 days after the surgery. Absolute contraindi-
cations to the procedure were pregnancy, active pelvic infections,
and known cervical or uterine cancer. All patients underwent vagi-
nal examination, transvaginal ultrasound, blood tests (glycemia,
electrolyte, blood count, creatininemia, azotemia, coagulation
function, β-hCG), sonohysterography which gave indication for
operative hysteroscopy, and anesthesiological examination. Be-
fore undergoing the procedure, it was necessary to perform an
electrocardiogram and to view a recent (≤ 1 year) vaginal cytol-
ogy not showing inflammation, precancerous, and/or cancerous
lesions. An antibiotic prophylaxis was performed with Ceftriax-
one (1 g; iv). Patients were counselled about all potential risks
and benefits of the operative HSC and were asked to sign an in-
formed consent form. The operation was performed under gen-
679
Figure 1. — Histological diagnosis at op-
erative hysteroscopy in the present med-
ical Centre.
Figure 2. — Mean duration of each oper-
ative hysteroscopic procedure (operating
time).
eral anesthesia by inserting a 10-mm rigid hysteroscope after di-
latation of the cervix with a Hegar n. 10, and the introduction of
the hysterometer. The uterus was distended with a mixture of 3%
sorbitol and 0.5% mannitol to a mean pressure of 120 mmHg and
at least a volume of 500 mL; during the procedure blood pressure,
temperature, pulse rate, and inflow and outflow were constantly
supervised. All procedures were video monitored and all resected
specimens were collected for histologic examination.
Results
The mean age of the patients was 45 (range 18 to 89 ).
years. The most common indication for operative HSC was
endometrial neoformation or thickening revealed by trans-
vaginal ultrasound; other indications were cervical neo-
formation, menorrhagia and/or metrorrhagia, sterility or
infertility, abnormal uterine bleeding in the post-
menopausal period, intrauterine fluid collection, septum
680 Indications, limitations and complications of operative hysteroscopy: a retrospective study of an 8-year experience

Table 1. — The most common indications for operative Table 2. — Causes of transfer to ordinary hospitalization
HSC in the present medical Centre. due to the inadequate ASA-criteria in the present medical
Indications Number of Rate Centre.
patients (per 100) Inadequate ASA-criteria Number of Rate
Endometrial neoformation or thickening 1246 88 patients (per 100)
Cervical neoformation 121 8.6 Heart disease (cardiomyopathy, valvular
Menorrhagia/metrorrhagia 112 8 7 8
insufficiency, drug-refractory arrythmia)
Sterility/infertility 34 2.4 Body mass index >30 35 38
Abnormal uterine bleeding 27 1.9 Drug-resistant hypertension 16 18
Intrauterine fluid collection 8 0.6 Unstable diabetes 6 7
Septum uterus 8 0.6 Chemotherapy for breast cancer 3 3
Lost IUD 6 0.4 Pre-operative haemorrhage 13 14
Synechiae 4 0.3 Non-stabilised hypothyroidism 2 2
Ashermann’s syndrome 3 0.2 Neurological disease 2 2
Retained placenta 2 0.1 Anorexia 1 1

Table 3. — Intraoperative complications of operative HSC

uterus, lost intrauterine devices (IUD), synechiae, Asher-
mann’s syndrome, and retained placenta (Table 1). The
most common histological diagnosis was endometrial
polyp (n=1002; 71%) followed by simple typical endome-
trial hyperplasia (n=160; 11%), myomas (n=144; 10%),
cervical polyps (n=131; 9%), complex typical endometrial
hyperplasia (n=30; 2.1%), simple atypical endometrial hy-
perplasia (n= 27; 1.9%), endometrial atrophy (n=20;
1.4%), complex atypical endometrial hyperplasia (n=11;
0.8%) and endometrial adenocarcinomas (n=3; 0.2%) (Fig-
ure 1). The mean duration of the surgical procedure was 19
minutes, with the shortest lasting 13 minutes (endometrial
biopsy) and the longest lasting 30 minutes (endometrial
ablation ) (Figure 2). In the present study the main limita-
tion was the need to perform operative HSC in ordinary
hospitalization in 85 patients (6%) because the anesthetic
criteria ASA I and II (Classification from the American So-
ciety of Anesthesiologists) were not insured, thus they
were not suitable for day surgery procedure. The causes of
transfer to ordinary hospitalization because of the inade-
quate ASA criteria are shown in Table 2. Another limita-
tion was the need to perform the myomectomy of large
submucous myomas with two subsequent operative HSCs
(e.g. two-step procedure) in three patients (0.2%). Intra-
operative complications occurred in 24 patients (1.7%) and
experienced panic attack (n=3; 0.2%), bronchospasm (n=2;
0.1%), hypertensive crisis (n=1; 0.07%), stenotic uterine
ostium (n=8; 0.5%), laceration of the anterior cervical lip
during dilatation due to excessive traction using the clamp
(n=8; 0.5%), and uterine perforation (n=2; 0.1%) (Table
3). One case of perforation occurred in a patient operated
for a lost IUD. The discharge from the uterus was not de-
tected by the preoperative ultrasound, but just after the in-
troduction of the hysteroscopic optical. Thus, the
perforation was caused by the IUD itself. On the other
hand, the second case of perforation was caused by the in-
troduction of the hysterometer. Except for the eight pa-
tients with the laceration of the anterior cervical lip (in
procedures in the present medical Centre.
Intraoperative complications Number of Rate
patients (per 100)
Panic attack 3 0.2
Bronchospasm 2 0.1
Hypertensive crisis 1 0.07
Stenotic uterine ostium 8 0.5
Laceration of the anterior cervical lip 8 0.5
Uterine perforation 2 0.1
which the lesion was repaired with absorbable sutures, not
requiring a longer hospital stay), the rest of the patients
with intraoperative complications needed to interrupt the
procedure and were transferred to ordinary hospitalization
for a watchful observation, some of them turning into other
surgical techniques, and some others postponing the oper-
ative HSC. The authors did not record any postoperative
complication, both during the hospital stay and at the gy-
necological examination performed 15 days after surgery.
Moreover, none of the patients returned to report any late
complication.
Discussion
Over the years operative HSC has increased as a surgical
option for various gynecological disorders because it has a
great accuracy in diagnosis and treatment and it reduces pa-
tients’ hospital stay, convalescence period, and healthcare
costs compared to major surgery [9-11]. In the present se-
ries, the most common indication of operative HSC was
endometrial neoformation or thickening (88%). In litera-
ture, the authors found it as the second most common indi-
cation in a study conducted by Mettler et al. in 2002 [12],
demonstrating the high prevalence of this pathological con-
dition. It is therefore important to perform a transvaginal
ultrasound screening in adult women to diagnose endome-
trial pathologies before clear symptoms appear and to treat
them at an early stage with better prognosis.
 D. Caserta, S. Picchia, E. Ralli, E. Matteucci, L. Di Benedetto, M. Mallozzi, G. Adducchio, R. Di Iorio, M. Moscarini
Table 4. — Reported complications of operative HSC procedures.
Cervical Uterine
laceration perforation
Caserta D. et al. (2013) 0.5% 0.1%
Propst et al. (2000) - 0.4%
Istre O. (2009) - 1%
Jansen F.W. et al. (2000) - 0.8%
Agostini A. et al. (2002) - -
Wortman M. et al. (2013) - -
Agostini A. et al. (2002) - -
Mencaglia L. et al. (2013) 0.7% 0.7%
Izetbegovic S. (2002) - 0.3%
a
litres.
In the present study, the authors found a low rate of sur-
gery-related complications compared to literature (Table
4) [6, 13-19]. Among the 24 (1.7%) intraoperative com-
plications, 15 were patient-related (panic attack, bron-
chospasm, hypertensive crisis, stenotic uterine ostium, and
uterine perforation caused by the lost IUD), while only
nine were surgery-related: eight (0.5%) cervical lacera-
tions and one (0.07%) uterine perforation caused by the
hysterometer. The laceration of the anterior cervical lip,
caused by excessive traction during the dilatation when the
top portion of the cervix was grasped with a clamp, was the
main surgical-related complication, representing just the
0.5% in this series. Furthermore, it should be noted that
these lacerations had been repaired with just absorbable
sutures, not requiring a longer hospital stay. The compli-
cation rate of cervical lacerations (0.5%) was similar com-
pared to literature findings, for instance a study conducted
by Mencaglia et al. in 2013 reported a rate of 0.7% [13].
On the other hand, the rate of uterine perforation is not
completely in agreement with literature findings: a study
conducted by Istre in 2009 reported a rate of 1% [14],
while Janszen et al. in 2000 reported a rate of 0.8% [6]. In
the present series, the authors had not experienced fluid
overload (defined as the absorption of more than 1,500 mL
of distension medium). The present results were com-
pletely in disagreement with that of Propst et al. [15],
which reported the phenomenon as the most common com-
plication (0.7%), and with the study of Istre, which re-
ported that fluid overload of 1-2 litres and > 2 litres,
respectively, occurred in 5.2% and in 1% of cases [14].
The lack of fluid overload may be partly explained with
the particular attention avoiding long operating time and
with the meticulous monitoring of the inflow pressure.
Moreover, the authors did not experience any early or late
postoperative complications. Regarding postoperative in-
fections, (i.e. endometritis and urinary tract infections)
Agostini et al. reported a rate of 1.42% [16] and Wortman
et al. a rate of 1.9% [17]. In the present study the authors
had not recorded any postoperative infection, which was
probably due to the accurate antibiotic prophylaxis per-
681
Overflow Postoperative Postoperative Intraoperative
infections haemorrhage haemorrhage
0% 0% 0% 0%
0.7% 0.2% - 0%
5.2%< 2 lta; 1%>2 lta - - 3%
0.2% - - -
- 1.42% - -
- 1.9% - -
- - - 0.61%
0.7% - - -
0.3% - 0.6% 0.9%
formed in all patients. Moreover, they had not experienced
intraoperative or postoperative hemorrhages; with regards
to intraoperative ones, Agostini et al. reported a rate of
0.61% [18] and Istre, a rate of 3% [14]. Regarding post-
operative ones, Izetbegovic reported a rate of 0.6% [19].
Haemorrhages were likely prevented by minimizing tis-
sue’s trauma, both limiting the handling of tissues for a
safe completion of the procedure and reducing operating
times. In the present study, the main limitation was the
need to deny the day surgery procedure to 84 patients with
ASA ≥ III and the subsequent performance of HSC in or-
dinary hospitalization. The most common ASA-criterion
responsible for the impossibility to perform operative HSC
in day surgery was a body mass index of >30. It is impor-
tant to note that this limitation was not related to the oper-
ative HSC itself, but to the health condition of the patient
which could be overcome with an ordinary admission to
the hospital, just requiring longer preoperative and post-
operative observation. Another limitation of this study was
the need to perform the myomectomy of large submucous
myomas with the two-step procedure, although undergoing
two surgical operations may seem more dangerous for the
patient, rather this method is in agreement with studies re-
garding the resection of submucous myomas, in which a
two-session approach was recommended, i.e. re-hys-
teroscopy after several weeks, because after this time the
intramural portion of the myoma will have shifted into the
uterine cavity due to a decrease in internal pressure [20].
The present study confirms that operative HSC is a safe,
effective, and minimally invasive procedure, with few lim-
itations and complications, especially giving particular at-
tention to the prevention of risks factors. Prevention begins
with an accurate patient selection, analyzing the medical
history and the anaesthesiological examination, and even-
tually transferring patients (preoperatively and/or postop-
eratively) to ordinary hospitalization for a watchful
observation. Prevention is also ensured avoiding long op-
erating time, the depth of resection, the manipulating of
tissues, by performing an antibiotic prophylaxis in all pa-
tients, and a meticulous monitoring of the inflow pressure.
682 Indications, limitations and complications of operative hysteroscopy: a retrospective study of an 8-year experience

[13] Mencaglia L., Carri G., Prasciolu C., Giunta G., Albis Florez E.D.,
Cofelice V., Mereu L.: “Feasibility and complications in bipolar re-
References sectoscopy: preliminary experience”. Minim. Invasive Ther. Allied
Technol., 2013, 22, 50.
[1] Pantaleoni D.C.: “On endoscopic examination of the cavity of the
[14] Istre O.: “Managing bleeding, fluid absorption and uterine perfora-
womb”. Medical Press. Circular., 1869, 8, 26.
tion at hysteroscopy”. Best. Pract. Res. Clin. Obstet. Gynaecol.,
[2] Parker W.H.: “Uterine myomas: management”. Fertil. Steril., 2007,
2009, 23, 619.
88, 255.
[15] Propst A.M., Liberman R.F., Harlow B.L., Ginsburg E.S.: “Compli-
[3] DaCosta V., Wynter S., Harriott J., Christie L., Berry E., Frederick-
cations of hysteroscopic surgery: predicting patients at risk”. Obstet.
Johnston S., Frederick J.: “Operative hysteroscopy in a Jamaican co-
Gynecol., 2000, 96, 517.
hort”. West Indian Med. J., 2011, 60, 641.
[16] Agostini A., Cravello L., Shojai R., Ronda I., Roger V., Blanc B.:
[4] Cooper J.M., Brady R.M.: “Late complications of operative hys-
“Postoperative infection and surgical hysteroscopy”. Fertil. Steril.,
teroscopy”. Obstet. Gynecol. Clin. North Am., 2000, 27, 367.
2002, 77, 766.
[5] Sentilhes L., Sergent F., Roman H., Verspyck E., Marpeau L.: “Late
[17] Wortman M., Daggett A., Ball C.: “Operative hysteroscopy in an of-
complications of operative hysteroscopy: predicting patients at risk
fice-based surgical setting: review of patient safety and satisfaction
of uterine rupture during subsequent pregnancy”. Eur. J. Obstet. Gy-
in 414 cases”. J. Minim. Invasive Gynecol., 2013, 20, 56.
necol. Reprod. Biol., 2005, 120, 134.
[18] Agostini A., Cravello L., Desbrière R., Maisonneuve A.S., Roger V.,
[6] Jansen F.W., Vredevoogd C.B., Van Ulzen K., Hermans J., Trimbos
Blanc B.: “Hemorrhage risk during operative hysteroscopy”. Acta
J.B., Trimbos-Kemper T.C.: “Complications of hysteroscopy: a
Obste.t Gynecol. Scand., 2002, 81, 878.
prospective, multicenter study”. Obstet. Gynecol., 2000, 96, 266.
[19] Izetbegović S.: “Early and late complications in patients treated with
[7] Paschopoulos M., Polyzos N.P., Lavasidis L.G., Vrekoussis T., Dal-
hysteroscopic surgery”. Med. Arh., 2002, 56, 217.
kalitsis N., Paraskevaidis E.: “Safety issues of hysteroscopic sur-
[20] Brandner P., Neis K.J., Diebold P.: “Hysteroscopic resection of sub-
gery”. Ann. N.Y. Acad. Sci., 2006, 1092, 229.
mucous myoma”. Contrib. Gynecol. Obstet., 2000, 20, 81.
[8] Cooper J.M., Brady R.M.: “Intraoperative and early postoperative
complications of operative hysteroscopy”. Obstet. Gynecol. Clin.
North Am., 2000, 27, 347.
[9] Taylor P.J.: “Hysteroscopy:where have we been, where are we
going?” J. Reprod. Med., 1993, 38, 757. Corresponding Author:
[10] Lindemann H.J., Gallinat A.: “Physical and physiological principles D. CASERTA, M.D., Ph.D.
of CO2 hysteroscopy”. Geburtshilfe Frauenheilkunde, 1976, 36, Department of Obstetrics, Gynaecological and
729. Urological Sciences
[11] Wood C., Maher P.: “Minimally invasive gynaecological surgery”.
“Sapienza” University of Rome, Sant’Andrea Hospital
Aust. Fam. Physician, 1992, 21, 772.
[12] Mettler L., Wendland E.M., Patel P., Caballero R., Schollmeyer T.: Via di Grottarossa 1035-1039
“Hysteroscopy: an analysis of 2-years’ experience”. JSLS, 2002, 6, 00189 Rome (Italy)
195. e-mail: donatella.caserta@uniroma1.it
CEOG Clinical and Experimental
Obstetrics & Gynecology

Office hysteroscopy for removal of retained products

of conception: can we predict treatment outcome?

A. Cohen*, Y. Cohen*, S. Sualhi, S. Rayman, F. Azem, G. Rattan

1 Department of Obstetrics and Gynecology, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, Tel Aviv
Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv (Israel)

Summary
Purpose of investigation: To evaluate the safety and efficacy of office hysteroscopy in the management of retained product of con-
ception (RPOC) and to identify those predictors for treatment success. Study Design: A retrospective cohort study that was conducted
in tertiary university-affiliated medical center. One hundred and eight women with sonographic findings of RPOC, who underwent see-
and-treat hysteroscopy, were included in this study. Demographic data, indication for treatment, and preoperative patient characteris-
tics and ultrasound findings were evaluated as predictors for treatment outcome. Results: Office-hysteroscopy was well tolerated by most
of the patients (96%), with an overall success rate of 65%. Causes of treatment failure were: actual RPOC size (assessed during see-
and-treat hysteroscopy), bleeding, and pain. In univariate analysis, none of the examined factors was shown to predict complete removal
of RPOC. Furthermore, RPOC size assessed by ultrasound was not shown to be valuable predictors for treatment outcome. Conclusions:
The efficacy of office hysteroscopy for removal of RPOC is limited. Ultrasound measurement of RPOC size should not be used as a
predictor for treatment outcome.

Key words: Hysteroscopy; Office hysteroscopy; Retained products of conception; See-and-treat hysteroscopy.

Introduction ting [10]. This technique allows evaluation, definitive di-

Retained products of conception (RPOC) are a well-
known complication after delivery (vaginal or cesarean),
termination of pregnancy (medical or surgical) and mis-
carriage [1, 2]. Although the precise incidence of RPOC is
unknown, evidence of suspected RPOC using color
Doppler was identified in 6.3% of women following deliv-
ery or termination of pregnancy [3]. Clinical signs at pres-
entation include abdominal pain, vaginal bleeding, fever,
and intrauterine finding on ultrasound examination. How-
ever, the reliability of ultrasonographic imaging as a diag-
nostic tool of RPOC showed variable accuracy in different
studies [1, 4-6].
Intrauterine adhesions formation is considered to be a se-
rious complication of RPOC, which may result in infertil-
ity, recurrent pregnancy loss, and menstrual abnormalities
[7]. Until recently, the management of RPOC has been di-
latation and curettage (D&C). Hysteroscopic removal of
RPOC was shown to be an alternative for D&C, allowing
a more selective procedure with the advantage of reduced
intrauterine adhesions and increase pregnancy rate [8, 9].
The development of small diameter operative hystero-
scopes enabled surgeons to perform small operative proce-
dures in an office-based setting. Moreover, the introduction
of the vaginoscopic approach by Bettocchi et al. allowed
abandoning the use of tenaculum and speculum and there-
fore avoiding the use of anesthesia and analgesia in this set-
*
Major and equal contribution.
agnosis and treatment at a single office procedure (See-and-
treat hysteroscopy). Furthermore, it was to shown to be
feasible in a variety of medical conditions such as polypec-
tomy, myomectomy, and adhesiolysis, with the advantage
of cost saving, reduced operation time and a high degree of
patient satisfaction [11, 12].
The aim of our study was to examine the safety and effi-
cacy of office hysteroscopy in the management of RPOC
and to evaluate clinical parameters that allow optimal pa-
tient selection and predict successful treatment.
Materials and Methods
The present authors conducted this retrospective cohort study at
a tertiary, university-affiliated medical center. IRB approval was
obtained from the local ethics committee. Medical records of all
patients who underwent see-and-treat hysteroscopy for RPOC in
the present department between 2011 and 2014 were reviewed.
Women after vaginal delivery, cesarean section or abortion
(medical and surgical) with clinical and sonographic suspicion of
RPOC were included in the study. All patients underwent outpa-
tient hysteroscopy with a semi-rigid hysteroscope. Distension of
the uterine cavity was achieved by a continuous infusion of saline
solution. The present authors used the vaginoscopic approach in
all the procedures, without using either tenaculum or speculum.
Neither did they use analgesia nor anesthesia during the proce-
dures. During the procedure, residual tissue was removed with
hysteroscopic forceps. They defined treatment success in those
cases where residual tissue was completely removed (either in one

Revised manuscript accepted for publication July 5, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3827.2017
684 A. Cohen, Y. Cohen, S. Sualhi, S. Rayman, F. Azem, G. Rattan

or more attempts), whereas procedures failure was defined as

those cases where residual tissue could not be removed and there- Table 1. — Analysis of demographic characteristics.
fore the patients were sent for hysteroscopy procedure under anes- Characteristics Successful group Failed group p
thesia in the authors’ day-hospitalization unit. Age (years),
For the purpose of the study, women who were successfully 32.62 (5.42) 31.95 (5.54) 0.54
mean (SD)
treated for RPOC by office hysteroscopy were compared with
those women with treatment failure. The retrieved data included Gravidity,
2.21 (1.56) 2.18 (1.64) 0.93
demographic data (age, parity, and gravidity), indication for treat- mean (SD)
ment (index pregnancy), preoperative patient’s complaint of vagi- Nulliparity, n (%) 19 (26.76%) 14 (37.84%) 0.23
nal bleeding, ultrasound finding (RPOC size in its greatest Obstetric event
dimension and Doppler studies), and intra- and postoperative Delivery, n (%) 37/71 (52.11%) 13/37 (35.14%) 0.61
complications. First trimester
34 (47.89%) 24 (64.86%) 0.11
The statistical analysis was carried out using SAS version 9.2. TOP , n (%)
The authors used univariate analysis to characterize the different Data are presented as mean ± standard deviation or absolute numbers (percentage).
variables with respect to both groups. Pearson chi-square and TOP: termination of pregnancy
Fisher exact tests were used to compare categorical variables,
while Two Sample T-test and Two Sample Wilcoxon test were
used to compare continuous variables. Continuous variables were
reported by means and standard deviations, while categorical vari-
Table 2. — Analysis of clinical and sonographic predictors
ables were reported by their relative frequencies. for successful office hysteroscopy.
Characteristics Successful group Failed group p

Results Time elapsed after

index pregnancy, 10.39 (6.42) 8.31 (4.79) 0.12
During the study period, 870 office hysteroscopies were weeks (SD)
performed in the present unit. One hundred eight women Bleeding, n (%) 13 (18.30%) 6 (16.22%) 0.76
after first trimester termination of pregnancy, spontaneous RPOC size (mm)
15 (9) 16(8.3) 0.55
abortion or delivery, were referred with ultrasonographic by US, mean (SD)
finding of RPOC. In 71 cases (65.74%), complete removal US Doppler flow,
16/30 (53.33%) 16/21 (76.19%) 0.09
n (%)
of RPOC was feasible by office hysteroscopy. In six
women, it was accomplished by a second office hys- Data are presented as mean ± standard deviation or absolute numbers (percentage).
teroscopy. The mean RPOC size measured by ultrasound RPOC: Retained products of conception, US- ultrasound
was 17.48 ± 8.75 mm (mean ± SD). Since all cases were re-
ferred to the present tertiary medical center for see-and-
treat hysteroscopy after initial evaluation in the community,
sis to search for predictors of successful removal of RPOC
the prevalence of RPOC after delivery or TOP in this study
by see-and-treat hysteroscopy: time elapsed after preg-
does not represent the true prevalence in the general popu-
nancy, patient age, previous deliveries, vaginal bleeding,
lation. The demographic characteristics of women enrolled
third trimester pregnancy vs. first trimester abortion, med-
in this study are summarized in Table 1.
ical vs. surgical TOP. None of these factors can be used to
The procedure was well tolerated by most patients and
predict complete evacuation of RPOC. Furthermore, RPOC
only four see-and-treat procedures (3.70%) were discon-
size and blood flow assessed by ultrasound were not found
tinued due to patient discomfort. The main reason for fail-
be valuable predictors (Table 2).
ure was the actual size of the RPOC (as evaluated during
OH). In 19 cases (17.59%), full evacuation of the uterus
was not feasible due to the size of RPOC. The second most Discussion
common reason of failure was preoperative bleeding. Nine
Office hysteroscopy allows minimally invasive diagnos-
women (8.33%), bled before the procedure, resulting in vi-
tic and therapeutic procedure for removal of RPOC. It can
sual impairment, therefore preventing the completion of the
serve as an alternative for operative hysteroscopy, obviat-
procedure. In three cases, removal of RPOC by see-and-
ing general anesthesia, and saving operating room time and
treat hysteroscopy was not attempted because of suspected
costs. As with every minimal invasive office based proce-
arterial venous malformation (AVM), large fibroid in the
dure, patient selection is of utmost importance to ensure
cavity, and cervical stenosis. There were no procedure re-
both patient satisfaction and safety. The results of this study
lated complications such as accidental uterine perforation,
show that complete removal of small size RPOC can be ac-
excessive bleeding, and fluid overload during the study.
complished by office-hysteroscopy with minimal patient
Uterine abnormalities were diagnosed in five women (bi-
discomfort and without complications.
cornuate uterus: two, uterus didelphys: one, intrauterine ad-
Patient age and gravidity did not predict successful
hesions: two). Nevertheless, it was possible to completely
RPOC removal by see-and-treat hysteroscopy. Further-
remove RPOC despite these findings.
more, the present authors hypothesized that preoperative
The following factors were analyzed in univariate analy-
 Office hysteroscopy for removal of retained products of conception: can we predict treatment outcome?
parameters like parity, vaginal bleeding, and time elapsed
from index pregnancy can effect cervical dilatation and tis-
sue organization within the uterine cavity. Thus, they can
potentially increase the technical difficulty and cause pa-
tient discomfort during office-hysteroscopy. However, the
authors found that patient selection for RPOC removal by
office hysteroscopy cannot be based on preoperative clini-
cal signs and patient complaints.
Surprisingly, RPOC size by ultrasound examination was
not shown to be a predictor for treatment outcome. In con-
trast, it was shown that actual RPOC size (as documented
during hysteroscopy) was a major reason for procedure fail-
ure. Ultrasonography is considered to be an important di-
agnostic tool regarding RPOC, however, its reliability was
shown to vary in different studies (1, 4-6). In his study,
Sawyer et al. (13) examined the significance of endome-
trial thickness and volume as predictors for the presence of
RPOC. In their study, the authors did not identify a cut-off
value for endometrial thickness nor volume that could be
used to diagnose RPOC. These results are further supported
by the study of Levin et al. (14), who showed in his study
that surgeon opinion based on hysteroscopic findings is a
predictor for RPOC, as opposed to other clinical parameters
and sonographic finding. There is no doubt that actual
RPOC size is a limiting factor in successful outcome,
mainly due to extended procedure time, patient discomfort,
and tissue adherence. However, it seems that RPOC size
by ultrasound examination does not reflect the actual size
as was perceived by the surgeon during office hysteroscopy,
and therefore was not shown to be a predictor for treatment
outcome.
Based on the results of this study and the limitation of ul-
trasound in accurately diagnosing RPOC, the authors sug-
gest the following office hysteroscopy for the diagnosis and
treatment of RPOC: women with suspected RPOC will ini-
tially undergo office hysteroscopy for diagnosis, followed
by a trial of removal in cases with small size RPOC. In
women with large size RPOC or those where complete
evacuation of the tissue has failed, referral for operative
hysteroscopy under general anesthesia should be advised.
Conclusion
The results of this study show that office hysteroscopy
can be used as an accessible diagnostic tool, overcoming
the limitations of ultrasonographic diagnosis of RPOC.
However, its use as a treatment tool for removal of RPOC
should be limited to small size residual tissue. Future ran-
domized controlled studies comparing office hysteroscopy
and operative hysteroscopy are required to define evidence-
based clinical guidelines for removal of RPOC.
685
References
[1] Achiron R., Goldenberg M., Lipitz S., Mashiach S.: “Transvaginal
duplex Doppler ultrasonography in bleeding patients suspected of
having residual trophoblastic tissue”. Obstet. Gynecol., 1993, 81,
507.
[2] Zalel Y., Cohen S.B., Oren M., Seidman D.S., Zolti M., Achiron R.,
Goldenberg M.: “Sonohysterography for the diagnosis of residual
trophoblastic tissue”. J. Ultrasound Med., 2001, 20, 877.
[3] van den Bosch T., Daemen A., Van Schoubroeck D., Pochet N., De
Moor B., Timmerman D.: “Occurrence and outcome of residual tro-
phoblastic tissue: a prospective study”. J. Ultrasound Med., 2008,
27, 357.
[4] Alcazar J.L.: “Transvaginal ultrasonography combined with color
velocity imaging and pulsed Doppler to detect residual trophoblas-
tic tissue”. Ultrasound Obstet. Gynecol., 1998, 11, 54.
[5] Ben-Ami I., Schneider D., Maymon R., Vaknin Z., Herman A.,
Halperin R.: “Sonographic versus clinical evaluation as predictors
of residual trophoblastic tissue”. Hum. Reprod., 2005, 20, 1107.
[6] Wolman I., Hartoov J., Pauzner D., Grutz A., Amit A., David M.P.,
Jaffa A.J.: “Transvaginal sonohysterography for the early diagnosis
of residual trophoblastic tissue”. J. Ultrasound Med., 1997, 16, 257.
[7] Yu D., Wong Y.M., Cheong Y., Xia E., Li T.C.: “Asherman syn-
drome—one century later”. Fertil Steril., 2008, 89, 759.
[8] Ben-Ami I., Melcer Y., Smorgick N., Schneider D., Pansky M.,
Halperin R.: “A comparison of reproductive outcomes following
hysteroscopic management versus dilatation and curettage of re-
tained products of conception”. Int. J. Gynaecol. Obstet., 2014, 127,
86.
[9] Rein D.T., Schmidt T., Hess A.P., Volkmer A., Schondorf T., Brei-
denbach M.: “Hysteroscopic management of residual trophoblastic
tissue is superior to ultrasound-guided curettage”. J. Minim. Inva-
sive Gynecol., 2011, 18, 774.
[10] Bettocchi S., Ceci O., Nappi L., Di Venere R., Masciopinto V.,
Pansini V., et al.: “Operative office hysteroscopy without anesthesia:
analysis of 4863 cases performed with mechanical instruments”. J.
Am. Assoc. Gynecol. Laparosc., 2004, 11, 59.
[11] Penketh R.J., Bruen E.M., White J., Griffiths A.N., Patwardhan A.,
Lindsay P., et al.: “Feasibility of resectoscopic operative hys-
teroscopy in a UK outpatient clinic using local anesthetic and tradi-
tional reusable equipment, with patient experiences and comparative
cost analysis”. J. Minim. Invasive Gynecol., 2014, 21, 830.
[12] Wortman M., Daggett A., Ball C.: “Operative hysteroscopy in an of-
fice-based surgical setting: review of patient safety and satisfaction
in 414 cases”. J. Minim. Invasive Gynecol., 2013, 20, 56.
[13] Sawyer E., Ofuasia E., Ofili-Yebovi D., Helmy S., Gonzalez J., Ju-
rkovic D.: “The value of measuring endometrial thickness and vol-
ume on transvaginal ultrasound scan for the diagnosis of incomplete
miscarriage”. Ultrasound Obstet. Gynecol., 2007, 29, 205.
[14] Levin I., Almog B., Ata B., Ratan G., Many A.: “Clinical and sono-
graphic findings in suspected retained trophoblast after pregnancy
do not predict the disorder”. J. Minim. Invasive Gynecol., 2010, 17,
66.
Corresponding Author:
A. COHEN, M.D.
Lis Maternity Hospital
6 Weizman St
Tel Aviv 64239 (Israel)
e-mail: co.aviad@gmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Protective role of adrenomedullin

in heterotopic ovarian transplant

E. Erdemoglu1, S.G. Gürgen2, E. Erdemoglu3, K. Kürşat Bozkurt4, E. Oz Oyar5

1 Department of Gynecology and Obstetrics, Womens and Children Hospital, Isparta
2 Department of Histology and Embryology, Celal Bayar University Medical Faculty, Manisa
3 Department of Gynecologic Oncology, Suleyman Demirel University Medical Faculty, Isparta
4 Department of Pathology, SuleymanDemirel University Medical Faculty, Isparta
5 Department of Physiology, Katip Celebi University Medical Faculty, Izmir (Turkey)

Summary
Purpose of investigation: To study adrenomedullin (ADM) in preventing ischemia and morphological changes in heterotopically
transplanted ovary. Materials and Methods: Forty female Sprague-Dawley rats were divided into four groups. In groups 1 and 2 each
ovary was transplanted to the corresponding inguinal region by heterotopic transplantation. In groups 3 and 4, ovaries were left intact
without transplantation. Treatment was injected to left inguinal region. ADM was given in groups 1 and 3 and placebo was given in
groups 2 and 4. Left ovaries showing local treatment effect in heterotopic transplantation was designated as A (1A, 2A, 3A, and 4A),
right ovaries showing systemic treatment effect were designated as (1B, 2B,3B, and 4B). Main outcome measures were ischemia, fol-
licle count, and CD 31 expression. Results: Ovaries treated with local ADM (group 1A) were in consonance with normal rat ovaries.
The ovaries of rats in group 1B and placebo treated transplant group (groups 2A, 2B) exhibited varying effects of ischemia. The mean
follicle numbers in groups 1A, 1B, 2A, 2B were 28 ± 3.3, 16.8 ± 2.5, 17.7 ± 2.1, 16.4 ± 2.9, respectively. The mean follicle number in
groups 3A, 3B, 4A, and 4B were 27.7 ± 2.0, 28.3 ± 2.2, 28.3 ± 2.2, 27.8 ± 1.9, respectively. Corpus luteum number and CD31 expres-
sion was found to be significantly higher in group 1A. Conclusion: Subcutaneous injection of ADM to heterotopic ovarian graft site
causes vasodilatation and increases angiogenesis and may protect ovarian graft against hypoxic damage.

Key words: Heterotopic transplantation; Adrenomedullin; Ovary; Ischemia; CD 31.

Introduction Fresh transplantation of ovaries far away from the field

Population of oocytes is fixed before birth and oocytes
cannot be renewed in mammalians after birth [1]. Deple-
tion of oocytes and cessation of ovarian function result in
menopause. Pelvic radiotherapy in young patients with cer-
vical cancer can lead to premature ovarian failure. Ap-
proximately 30,000 cases of cancer are diagnosed annually
in women aged between 25–49 years old, and cervical can-
cer comprises 2% of female cancers in reproductive period
[2]. Forty-one percent of cervical cancer patients are less
than 45 years of age [3]. Young survivors of cancer experi-
ence somatic symptoms of menopause and are at increased
risk of osteoporosis, bone fracture, cardiovascular diseases,
hot flushes, stroke, anxiety, and sexual dysfunction [4, 5].
Therefore, preservation of gonadal function is of great con-
cern in young patients with cancer. Risk of ovarian metas-
tasis in cervical cancer is very low if the ovaries are
macroscopically normal [6]. Ovarian transposition is the
choice of surgical procedure to preserve ovarian function
by transposing the ovaries out of the field of radiotherapy,
however the effectiveness of ovarian transposition is de-
batable in the literature [7, 8].
of pelvic irradiation (heterotopic transplantation) may be
an alternative to young patients with cancer to preserve
ovarian function [9]. Transplantation of ovary under the
skin is the preferred grafting site, because monitoring of
ovary is simple [10]. Vascularization of the ovarian graft is
the major factor limiting transplantation, particularly in het-
erotopic transplantation. Ischemic damage results in apop-
tosis of ovarian follicles. Majority of the follicles die due to
ischemia during transplantation, and finding ways to ac-
celerate graft vascularization is essential for development
of reproducible and reliable procedures for ovarian trans-
plantation [11].
Adrenomedullin (ADM) is a 52-amino acid peptide,
structurally and functionally related to calcitonin, calci-
tonin gene-related peptide [12]. ADM has been shown to
mediate multifunctional responses in cell culture and ani-
mal systems, particularly regulation of growth and apop-
tosis [13]. ADM is a potent vasodilator [14]. ADM
regulates vascular permeability and plays a major role in
forming blood vessels in physiological and pathological
conditions [14-16]. Immunostaining by CD31 revealed

Revised manuscript accepted for publication January 25, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3536.2017
 E. Erdemoglu, S.G. Gürgen, E. Erdemoglu, K. Kürşat Bozkurt, E. Oz Oyar
that ADM increased capillary formation in ischemia, hy-
poxia, neoplasia, and tumor angiogenesis [16, 17]. In-
flammation and hypoxia increases ADM expression in
tumors [17]. Elevated levels of ADM is shown to be asso-
ciated with tumor neovascularization in xenografted en-
dometrial tumors and renal cell carcinoma [17, 18]. ADM
also acts as a cell survivor factor; it plays a potent protec-
tive role against apoptosis and maintains cellular integrity
[19]. Protective effect of ADM by preventing apoptosis
and inducing angiogenesis is reported in ischemia and hy-
poxic conditions such as stroke, age-related macular dys-
function, intrauterine growth restriction, myocardial
infarct, and carcinomas of various organs [19-22]. How-
ever, protective role of ADM in transplantation has not
been studied before in the literature. Aim of the present
study was to investigate whether ADM can prevent is-
chemia and morphological changes in heterotopically
transplanted whole ovary.
Materials and Methods
An animal research was designed to study the effect of ADM in
heterotopic whole ovary transplantation. The study was approved
by institutional review board and all procedures were approved
by the Animal Care and Use Committee of the University.
Forty female Sprague-Dawley rats, weighing 160-210 grams,
were allowed to acclimatize for seven days. Rats were housed in
a controlled environment at 23 ± 2°C on an illumination schedule
of 12 hours of light and 12 hours of darkness each day. Rats were
fed standard food and tap water ad libitum. The animals were
maintained in accordance with Animal Care and Use Committee
regulations. After acclimatization, rats were divided into four
groups and the surgical procedures were undertaken. The proto-
cols in this study followed guiding ethics for research involving
animals as recommended by the Declaration of Helsinki and the
Guiding Principles in the Care and use of Animals [23].
All rats were operated with ketamine (80 mg/kg body weight)
and xylazine (ten mg/kg body weight) for anesthesia. Abdominal
and inguinal skin was shaved and sterilized with iodine. A two-cm
skin incision was made and the subcutaneous transplantation site
was prepared. Ovaries were harvested by midline laparotomy.
After being freed from any foreign tissue, the ovary which was to
be transplanted was washed in physiological solution (0.9%
NaCl), just before transplantation. After removal, cleaning and
rinsing, the entire ovary was immediately placed into previously
prepared subcutaneous transplantation site without vascular anas-
tomosis. The incision was sutured using 4–0 sutures.
Rats were divided into four groups. In groups 1 and 2, hetero-
topic transplantation was performed; in groups 3 and 4 ovaries
were left intact without transplantation. In groups 1 and 2, both
ovaries were harvested and each ovary was transplanted to right
and left inguinal region. ADM treatment was given to group 1 rats
and placebo was given in group 2 rats for control. Groups 3 and
4 were control groups without transplantation. In groups 3 and 4,
only midline laparotomy was made. Treatment by ADM was ad-
ministered to rats in group 3 and placebo was administered to rats
in group 4.
ADM was administered ten µg/kg, daily for 21 days. ADM was
administered in group 1 (transplantation and ADM treated group)
and group 3 (no transplantation, ADM treated group). 0.9% NaCl
687
was used for placebo in group 2 (transplantation, no treatment
group) and group 4 (no transplantation, no treatment). Injections
were made to left inguinal operation site.
Left heterotopic ovaries of the rats from groups 1 and 2 showed
the local treatment effect, while right heterotopic ovaries showed
the systemic treatment effect after subcutaneous injection. Left
ovaries were showing local treatment effect in heterotopic trans-
plantation was designated as A (1A, 2A, 3A, and 4A), right
ovaries showing systemic treatment effect were designated as (1B,
2B, 3B, and 4B). Rats were sacrificed at the end of 21-day treat-
ment, ovaries were removed for macroscopic gross examination
and microscopic examination.
Macroscopic examination for viability was assessed and then tis-
sue samples were fixed in a 10% formalin solution, dehydrated
through a graded ethanol series, cleared in xylene, and processed for
embedding in paraffin wax, according to routine protocols. Five-
μm-thick sections were cut by microtome and stained with hema-
toxylin and eosin (H&E) according to the standard method. Sec-
tions were evaluated to detect the follicle and corpus luteum number
by using a CX41 bright-field microscope.
Immunohistochemical analysis for CD31 (rabbit polyclonal,
1:200) was performed on formalin fixed, paraffin-embedded tis-
sue blocks using the streptavidin-biotin-peroxidase tech-
nique.Following antigen retrieval, four-μm-thick sections were
washed gently in deionized water, then treated with 2% trypsin in
Tris buffer (50 mMTris base and 150 mMNaCl dissolved in deion-
ized H2O) at 37oC for 15 minutes. Endogenous peroxidase was
blocked with 3% hydrogen peroxide for ten minutes. Slides were
incubated with avidin and biotin blocking solutions for 15 minutes
each, and 3% normal goat serum for 20 minutes in order to pre-
vent nonspecific staining in the background. All slides underwent
overnight incubation at 4°C. Negative controls for immunostain-
ing were provided by omitting the primary antibody step. After
washing with TBST, biotinylated goat anti-rabbit IgG (1:1000)
were applied to the sections for 30 minutes at room temperature.
Then all of the sections were incubated with strepavidin-HRP for
30 minutes at room temperature. Finally, 3-amino-9-
ethylcarbazole was used as the chromagen and hematoxylin as the
counterstain.
Depending on the size of the H&E section, three to five high
power areas within the slide were selected randomly for evalua-
tion. Image-analyzing software was used to lock on these prese-
lected areas for each histological section of the same paraffin
block. The microvessel density (MVD) measurements for CD31
were performed within each area at a ×100 magnification. The
MVD was measured based on Weidner’s method [24]. Each pos-
itive endothelial cell cluster of immunoreactivity in contact with
the selected field was counted as an individual vessel in addition
to the morphologically identifiable vessels with a lumen.
Results
Macroscopic and microscopic examination of ovaries
revealed obvious tissue modifications between hetero-
topic transplant group rats treated with ADM and placebo.
Ovaries from control group rats (groups 3 and 4) were
normal and they had follicles in different stages of devel-
opment and a cellular stroma. Ovaries treated with local
ADM (group 1A) were in consonance with normal rat
ovaries as described before [20, 21] and control groups.
The ovaries of rats in group 1B (systemic effect of ADM)
and placebo treated transplant group (groups 2A and 2B)
were exhibiting varying effects. There were changes in
688 Protective role of adrenomedullin in heterotopic ovarian transplant

Figure 1. — There is marked va-

sodilation by ADM local injec-
tion in heterotopically trans-
planted ovary (CD31, ×100). B)
CD31 expression (arrow) in the
vascular endothelium in group 1A
(CD31, ×400). C) Corpus luteum
(C) number was significantly
higher along with the vascular-
ization (*) in group 1A (CD31,
×100). D) Remarkable lipid accu-
mulation in the CL of ADM
treated rats and increased vascu-
larization (arrows). This might be
due to the increased steroid hor-
mone synthesis by ADM (H&E,
×400).

Table 1. — A: follicule counts of transplantation and con- Table 2. — CD31 expressions of transplantation and con-
trol groups. B: corpus luteum counts of transplantation and trol groups.
control groups. Transplantation Group 1A Group 1B Group 2A Group 2B
Group Follicle count CL count 283±8.9 180±7.2 180±7.3 178±9.6
1A 28.0±3.3 26.4±2.4
2A 17.7±2.1 15.4±2.0 Controls Group 3A Group 3B Group 4A Group 4B
3A 27.7±2.0 23.8±2.2 110±7.5 104±3.4 100.6±3.7 101±4.5
4A 28.3±2.2 20.6±2.5
1B 16.8±2.5 15.7±1.8
2B 16.42.9 15.4±2.0
3B 28.3±2.2 22.0±2.1
4B 27.8±1.9 21.2±2.0 whereas it was significantly higher than the mean follicle
number in groups 1B and group 2. This finding showed
that ADM was useful locally to preserve viability of
ovary. The corpus luteum number was also found to be
the color, size, and macroscopic viability of ovaries. significantly higher in group 1A. CD31 expression was
Microscopic examination was in concordance with increased in heterotopic transplant groups (groups 1 and
macroscopic evaluation. Morphology of follicles was pro- 2). CD31 expression was highest in group 1A. CD31 ex-
tected in locally ADM treated group 1A rats. There were pression was similar in groups 1B, 2A, and 2B (Table 2).
significantly more corpus luteum structures and lipid ac-
cumulation in the corpus luteum was remarkable. There
was aberrant vascular dilatation in this group (Figure 1); Discussion
however, the follicles were remarkably scant in groups Hypoxic-ischemic damage is the major challenge for
1B, 2A, and 2B. There was no significant vascular dilata- large grafts in heterotopic transplantation. Ischemia causes
tion in groups 1B, 2A, and 2B. depletion of follicles during the first days after transplan-
The mean follicle number in groups 1A, 1B, 2A, 2B tation and this continues for one week [22, 25]. The pres-
were 28 ± 3.3, 16.8 ± 2.5, 17.7 ± 2.1, 16.4 ± 2.9, respec- ent authors have studied local and systemic effect of ADM
tively. The mean follicle number in groups 3A, 3B, 4A, to salvage transplanted ovaries from ischemia. This is the
and 4B were 27.7 ± 2.0, 28.3 ± 2.2, 28.3 ± 2.2, 27.8 ± first study investigating the novel, angiogenetic, anti-apop-
1.9,respectively (Table 1). The mean number of follicles totic factor (ADM) in heterotopic transplantation. Subcu-
in group 1A was similar to control groups 3 and 4, taneous injection of ADM to transplantation field was
 E. Erdemoglu, S.G. Gürgen, E. Erdemoglu, K. Kürşat Bozkurt, E. Oz Oyar
found to protect the ovarian follicles from hypoxia and in-
creased the vascularization of the graft after seven days.
Subcutaneous injection of ADM to another site did not pro-
tect the transplanted ovary.
One of the limitations of the present study was that the
authors were not able to detect hormones such as FSH, LH,
estradiol, inhibin, or anti-Müllerian hormone (AMH) with
a sensitive analysis. It is reported that these markers may
not be useful as a marker of ovarian reserve after trans-
plantation, although they reflect ovarian function in other
clinical situations [26, 27]. Ovarian function may also re-
turn late after transplantation [27]. Longer follow-up after
transplantation is needed to evaluate hormonal function of
the ovary. The present aim was to study the viability, loss
of follicles, and vasculogenesis of the ovary by ADM in-
jection after heterotopic transplantation. ADM may also ef-
fect the secretion of hormones, which may cause a bias for
hormonal status [28, 29]. Hormonal function, development
of follicles, and fertility may further be investigated in fu-
ture studies with longer duration. Another limitation of the
present study may be the follow-up after transplantation.
The decision to sacrifice animals on the eighth day of
transplantation is based on previous studies. Experimen-
tal studies showed that the fall in number of follicles in
grafted tissue due to hypoxia and delay re-vascularization
occurs in two to four days [26]. The critical time for the
full recovery of ovarian function after transplant happens
during the first 24 hours [30]. Dissen et al. showed pro-
fuse re-vascularization 48 hours after transplantation [31].
Neovascularization is completed by seven to ten days after
transplantation [32,33].
CD31 is a reliable marker to evaluate angiogenesis in
tranplants [16, 34]. The present authors have found the mi-
crovessel density (expression of CD31) to be significantly
higher in locally ADM treated transplanted ovaries. ADM
is released under hypoxic conditions and ADM is a potent
vasodilator and induces angiogenesis [15-17]. ADM also
regulates vascular permeability [15], which may contribute
to the graft nourishing and survival [33]. ADM was pro-
tective against hypoxia; number of follicles and corpus lu-
teum was significantly higher in locally treated ovarian
grafts. Chaung et al. reported that peripheral administra-
tion of ADM may reduce stroke-induced brain ischemic in-
jury [35]. In the present study, the authors found that ADM
was protective only in local injection to the site of trans-
plantation.
Damous et al. reported that corpus luteum appeared
seven days after transplantation [32]. Presence of corpus
luteum indicates immature follicles can grow and mature
to the preovulatory stage, eventually ovulating and forming
corpora lutea. These findings suggested that heterotopic
ovarian transplantation into subcutaneous tissue can resume
normal function and estrous cycle. The number of corpus
luteum was higher in group 1A. Corpus Luteum is a well
vascularized structure and formation of CL may be associ-
689
ated with ADM. ADM and its mRNA were reported in the
follicles and the corpora lutea (CL) of rat and human
ovaries [28,36]. In human ovary, ADM levels were low in
the mature follicle but increased in the CL of the mid-luteal
phase and remained high in CL of early pregnancy [36].
The present authors have observed remarkable lipid accu-
mulation in the CL of ADM treated rats. This might be due
to the increased steroid hormone synthesis by ADM.
Conclusion
Subcutaneous injection of ADM to heterotopic ovarian
graft site causes vasodilatation and increases angiogenesis.
These benefical effects of ADM may protect the ovarian
graft against hypoxic damage, and depletion of follicles.
The heterotopic graft treated by ADM is viable, can resume
normal function, and the follicles may develop to CL. Fur-
ther studies on ADM and ovarian transplantation is neces-
sary for a better understanding of interaction of ADM,
ischemia, and grafts.
References
[1] Begum S., Papaioannou V.E., Gosden R.G.: “The oocyte population
is not renewed in transplanted or irradiated adult ovaries”. Hum. Re-
prod., 2008, 23, 2326.
[2] King J., Wynne C.H., Assersohn L., Jones A.: “Hormone replace-
ment therapy and women with premature menopause a cancer sur-
vivorship issue”. Eur. J. Cancer, 2011, 47, 1623.
[3] Rasool N., Rose P.G.: “Fertility-preserving surgical procedures for
patients with gynecologic malignancies”. Clin. Obstet. Gynecol.,
2010, 53, 804.
[4] Sayakhot P., Vincent A., Deeks A., Teede H.: “Potential adverse im-
pact of ovariectomy on physical and psychological function of
younger women with breast cancer”. Menopause, 2011, 18, 786.
[5] Clemons M., Clamp A., Anderson B. “Management of the
menopause in cancer survivors”. Cancer Treat. Rev., 2002, 28, 321.
[6] Landoni F., Zanagnolo V., Lovato-Diaz L., Maneo A., Rossi R.,
Gadducci A., et al.: “Ovarian metastases in early-stage cervical
cancer (IA2-IIA): a multicenter retrospective study of 1965 patients
(a Cooperative Task Force study)”. Int. J. Gynecol. Cancer, 2007,
17, 623.
[7] Dursun P., Ayhan A., Yanik F.B., Kuşçu E.: “Ovarian transposition
for the preservation of ovarian function in young patients with cer-
vical carcinoma”. Eur. J. Gynaecol. Oncol., 2009, 30, 13.
[8] Feeney D.D., Moore D.H., Look K.Y., Stehman F.B., Sutton G.P.:
“The fate of the ovaries after radical hysterectomy and ovarian trans-
position”. Gynecol. Oncol., 1995, 56, 3.
[9] Chiu D.T.W., Hu G.: “Evaluation of the hormonal function and his-
tological features of heterotopic isogenic ovarian transplantation in
rats”. Plast and Reconstr Surg., 2003,111,1646.
[10] Gao J.M., Yan J., Li R., Li M., Yan L.Y., Wang T.R., et al.: “Im-
provement in the quality of heterotopicallotransplantedmouseovari-
antissues with basic fibroblast growth factor and fibrin hydrogel”.
Hum. Reprod., 2013, 28, 2784
[11] Israely T., Nevo N., Harmelin A., Neeman M., Tsafriri A.: “Reduc-
ing ischaemic damage in rodent ovarian xenografts transplanted into
granulation tissue”. Hum. Reprod., 2006, 21, 1368.
[12] Kitamura K., Kangawa K., Kawamoto M., Ichiki Y., Nakamura S.,
Matsuo H., et al.: “Adrenomedullin, A novel hypotensive peptide
isolated from human pheochromocytoma”. Biochem. Biophys. Res.,
Commun., 1993, 192, 553.
[13] Uzan B., Villemin A., Garel J.M., Cressent M.: “Adrenomedullin is
690 Protective role of adrenomedullin in heterotopic ovarian transplant
anti-apoptotic in osteoblasts through CGRP1 receptors and MEK-
ERK pathway”. J. Cell. Physiol., 2008, 215, 122.
[14] Hasbak P., Sheykhzade M., Schifter S., Edvinsson L.: “Potentiated
adrenomedullin-induced vasorelaxation during hypoxia in organ cul-
tured porcine coronary arteries”. J. Cardiovasc. Pharmacol., 2014,
63, 58.
[15] Sakimoto S., Kidoya H., Kamei M., Naito H., Yamakawa D., Sak-
aguchi H., et al.: “An angiogenic role for adrenomedullin in
choroidal neovascularization”. PLoS One, 2013, 8, e58096.
[16] Limuro S., Shindo T., Moriyama N., Amaki T., Niu P., Takeda N.,
et al.: “Angiogenic effects of adrenomedullin in ischemia and tumor
growth”. Circ Res., 2004, 95, 415.
[17] Kim W., Moon S.O., Sung M.J., Kim S.H., Lee S., So J.N., et al.:
“Angiogenic role of adrenomedullin through activation of Akt, mi-
togen-activated protein kinase, and focal adhesion kinase in en-
dothelial cells”. FASEB J., 2003, 17, 1937.
[18] Oehler M.K., Hague S., Rees M.C., Bicknell R.: “Adrenomedullin
promotes formation of xenografted endometrial tumors by stimula-
tion of autocrine growth and angiogenesis”. Oncogene, 2002, 21,
2815.
[19] Penchalaneni J., Wimalawansa S.J., Yallampalli C.:
“Adrenomedullin antagonist treatment during early gestation in rats
causes fetoplacental growth restriction through apoptosis”. Biol. Re-
prod., 2004, 71, 1475.
[20] Xia C.F., Yin H., Borlongan C.V., Chao J., Chao L.: “Post ischemic
infusion of adrenomedullin protects against ischemic stroke by in-
hibiting apoptosis and promoting angiogenesis”. Exp. Neurol., 2006,
197, 521.
[21] Yin H., Chao L., Chao J.: “Adrenomedullin protects against my-
ocardial apoptosis after ischemia/reperfusion through activation of
Akt-GSKsignaling”. Hypertension, 2004, 43, 109.
[22] Abasolo I., Montuenga L.M., Calvo A.: “Adrenomedullin prevents
apoptosis in prostate cancer cells“. Regul. Pept., 2006, 133, 115.
[23] American Physiological Society, World Medical Association Gen-
eral Assembly: “Guiding principles for research involving animals
and human beings”. Am. J. Physiol. Cell. Physiol., 2002, 282, 3.
[24] Weidner N. “Current pathologic methods for measuring intratumoral
microvessel density within breast carcinoma and other solid tumors”.
Breast Cancer Res. Treat, 1995, 36, 169.
[25] Soleimani R., Heytens E., Oktay K.: “Enhancement of neoangio-
genesis and follicle survival by sphingosine-1-phosphate in human
ovarian tissue xenotransplants”. PLoS One, 2011, 29, e19475.
[26] Janse F., Donnez J., Anckaert E., de Jong F.H., Fauser B.C., Dol-
mans M.M.: “Limited value of ovarian function markers following
orthotopic transplantation of ovarian tissue after gonadotoxic treat-
ment”. J. Clin. Endocrinol. Metab., 2011, 96, 1136.
[27] Oktay K., Buyuk E., Veeck L., Zaninovic N., Xu K., Takeuchi T.:
“Embryo development after heterotopic transplantation of cryopre-
served ovarian tissue”. Lancet, 2004, 363, 837.
[28] Li L., O W.S., Tang F.: “Adrenomedullin in rat follicles and corpora
lutea: expression, functions and interaction with endothelin-1”. Re-
prod. Biol. Endocrinol., 2011, 9, 111.
[29] Moriyama T., Otani T., Maruo T.: “Expression of adrenomedullin
by human granulose lutein cells and its effect on progesterone pro-
duction”. Eur. J. Endocrinol., 2000, 142, 671.
[30] Damous L.L., da Silva S.M., Simões Rdos S., Sakano C.R., Simões
Mde J., Montero E.F.: “Indirect evaluation of estrogenic activity post
heterotopic ovarian autograft in rats”. Acta Cir. Bras., 2008, 23, 372.
[31] Dissen G.A., Lara H.E., Fahrenbach W.H., Costa M.E., Ojeda S.R.:
“Immature rat ovaries become revascularized rapidly after auto-
transplantation and show a gonadotropin-dependent increase in an-
giogenic factor gene expression”. Endocrinology, 1994, 134, 1146.
[32] Damous L.L., Silva S.M., Carbonel A.P., Simões R.S., Simões M.J.,
Montero E.F.: “Effect of remote ischemic preconditioning on rates
tradiol serum levels and follicular development after ovarian trans-
plantation”. Transplant Proc., 2009, 41, 830.
[33] Wang Y., Chang Q., Sun J., Dang L., Ma W., Hei C., et al.: “Effects
of HMG on revascularization and follicular survival in heterotopic
autotransplants of mouse ovarian tissue”. Reprod. Biomed., 2012,
24, 646.
[34] Wang D., Stockard C.R., Harkins L., Lott P., Salih C., Yuan K., et al.:
“Immunohistochemistry in the evaluation of neovascularization in
tumorxenografts”. Biotech. Histochem., 2008, 83, 179.
[35] Chaung W.W., Wu R., Ji Y., Wang Z., Dong W., Cheyuo C., et al.:
“Peripheral administration of human adrenomedullin and its bind-
ing protein attenuates stroke-induced apoptosis and brain injury in
rats”. Mol. Med., 2011, 17, 1075.
[36] Abe K., Minegishi T., Ibuki Y., Kojima M., Kangawa K.: “Expres-
sion of adrenomedullin in the human corpus luteum”. Fertil.
Steril., 2000, 74, 141.
Corresponding Author:
E. ERDEMOGLU, M.D.
Department of Obstetrics and Gynecology
Isparta Women and Children Hospital
Isparta 32260 (Turkey)
e-mail: ebru.md@hotmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Which type of circumcision is more harmful

to female sexual functions?

O. Birge1, D. Arslan2, E.G. Ozbey2, M. Adiyeke3, I. Kayar4, M.M. Erkan5, U. Akgör1

1
Department of Gynecology and Obstetrics, Nyala Sudan Turkey Training and Research Hospital, Nyala-Darfur
2
Department of Urology, Nyala Sudan Turkey Training and Research Hospital, Nyala-Darfur (Sudan)
3
Department of Gynecology and Obstetrics, Bergama State Hospital, Izmir
4
Department of Gyneology and Obstetrics, Osmaniye State Hospital, Osmaniye
5
Department of Gynaecology and Obstetrics, Seferihisar State Hospital, Izmir (Turkey)

Summary
Background: Female genital mutilation (FGM) is common in Sub-Saharan Africa. It has been shown that it can cause sexual dys-
function. Materials and Methods: A total of 239 volunteer women were included in the study, which was conducted between April
2014 and January 2015; 210 of these women were circumcised and 29 were uncircumcised. Sexual functions of the women were eval-
uated by using Female Sexual Functioning Index (FSFI). Statistical analyses were performed by using the Mann-Whitney U-test,
Kruskal-Wallis test, and chi-square test. Results: The ages of women examined ranged between 17 and 65 years (mean: 33.54 ± 10.25).
The ratio of women circumcised was determined as 87.9%. The most commonly performed circumcision type was Type 2 (51%), fol-
lowed by Type 3 (25.7%); the remaining cases were determined to be Type 1 circumcision (23.3%). Both the total FSFI scores and each
individual score of sexual desire, arousal, lubrication, orgasm, satisfaction, and pain differed between the uncircumcised and circum-
cised women; these differences were statistically significant. When the circumcision types were compared to each other, the difference
between Type 1 and 2 was not statistically significant, whereas the differences between Type 1 and 3, and between Type 2 and 3 were
statistically significant. Discussion: Type 3 FGM is the most severe form of FGM, in which almost all of the female external genitalia
is excised, and the remaining parts are sewn together; this procedure narrows the opening of the genital organ. In the current study, the
lowest FSFI scores were determined in women with Type 3 FGM. Circumcision, and especially the Type 3, is still an important health
problem causing female sexual function disorders in the women living in Darfur, Sudan.

Key words: Female genital mutilation; Sexual function; Type 3 female circumcision.

Infibulation is the most severe form of FGM, and it is

Introduction
Female genital mutilation (FGM) is defined by the World
Health Organization (WHO) as all procedures involving
partial or total removal of the female external genitalia for
non-medical reasons. Female genital mutilation is concen-
trated in Africa and Middle Eastern countries [1]. It is gen-
erally performed on girls between the ages of five and 12,
and sometimes on adult girls after puberty [2]. This proce-
dure is performed with or without local anesthesia, by hold-
ing them forcibly, and by using knives, razor blades, or
pieces of broken glass. FGM is classified by the WHO in
four types: Type 1 – excision of the clitoris prepuce and/or
total or partial clitorectomy; Type 2 – total or partial exci-
sion of the labium minus and/or clitorectomy; Type 3 – ex-
cision of the labium majus and sewing the remaining parts
of the outer lips together (infibulation); Type 4 – punctur-
ing, piercing, cutting, or cauterization without extracting
any part.
mainly performed in Djibouti, Eritrea, Ethiopia, Somalia,
and Sudan [3]. After infibulation, i.e. after sewing the re-
maining parts of the outer lips together, the feet of girls are
tied together and they are kept in that position for many
days in order to support the excised remaining parts to join
together [4]. Complications depend on the environmental
and procedural hygiene, the tools used, experience of the
person who carries out the procedure, or the type of the
FGM procedure [1, 5]. In patients who underwent Type 3
FGM, some problems experienced included: inability to
have sexual intercourse, infertility, dysmenorrhea, en-
dometriosis, difficulty in urination, and prolonged and dif-
ficult labor and delivery because of the narrow openings
for urine and menstrual flow [6]. Many previous studies
have determined that FGM causes various degrees of sex-
ual function disorders, decrease in sexual pleasure, and
feelings of humiliation and inadequacy in women when
compared to their congeneric [7, 8].

Revised manuscript accepted for publication November 24, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3464.2017
692 O. Birge, D. Arslan, E.G. Ozbey, M. Adiyeke, I. Kayar, M.M. Erkan, U. Akgör

Table 1. — Ratios of circumcised women. Table 3. — Total FSFI scores of the women.
Frequency Percentage Valid percentage FSFI score Number Minimum Maximum Mean Standard
Circumcised 210 87.1 87.9 of women score score score Deviation
Uncircumcised 29 12 12.1 Uncircumcised 29 7.20 32.60 24.4138 5.69177
Total 239 99.2 100 Type 1 49 8.00 27.90 20.3204 4.15421
Type 2 107 7.80 33.50 20.2421 4.08735
Type 3 54 2.30 27.60 16.6741 5.21891
Table 2. — Distribution of circumcised women according
to the types of circumcision.
FGM Type Frequency Percentage Valid Cumulative
Table 4. — Total FSFI scores and the mean scores accord-
percentage percentage ing to the types of circumcision.
Type 1 49 20.3 23.3 23.3 Type of circumcision Number Minimum Maximum Mean Std.
Type 2 107 44.4 51 74.3 of women Deviation
Type 3 54 22.4 25.7 100 Unc. FSFI 29 7.20 32.60 24.4138 5.69177
Total 210 87.1 100 100 Desire 29 2.00 10.00 6.5862 1.91828
Arousal 29 4.00 19.00 13.7241 3.90875
Lubrication 29 4.00 18.00 13.9310 3.11598
Orgasm 29 3.00 14.00 10.4828 2.50172
Satisfaction 29 3.00 15.00 10.0690 2.86520
Female circumcision, which is a common procedure in Pain 29 3.00 13.00 9.8621 2.24760
Sudan, is performed in different areas of the country, in dif- Type 1 FSFI 49 8.00 27.90 20.3204 4.15421
fering ratios. FGM is also performed in high ratios in the Desire 49 2.00 8.00 5.1429 1.60728
Darfur district of Sudan. The Female Sexual Functioning Arousal 49 4.00 17.00 10.6327 3.16026
Index (FSFI) is used to evaluate sexual functions in women. Lubrication 49 4.00 17.00 11.8776 2.98351
The present study aimed to evaluate sexual functions in cir- Orgasm 49 3.00 12.00 8.7347 2.01841
cumcised and uncircumcised women in the Darfur region Satisfaction 49 3.00 12.00 7.9184 2.42244
of Sudan, and to determine if sexual functions differ de- Pain 49 3.00 15.00 9.5510 2.28274
Type 2 FSFI 107 7.80 33.50 20.2421 4.08735
pending on the type of female circumcision.
Desire 107 2.00 10.00 5.1028 1.47261
Arousal 107 4.00 20.00 10.7850 3.04069
Materials and Methods Lubrication 107 4.00 18.00 11.9533 2.83969
Orgasm 107 3.00 14.00 8.7664 1.88129
A total of 239 volunteer women were included in the study, Satisfaction 107 3.00 13.00 7.9626 2.06920
which was conducted between April 2014 and January 2015; 210
Pain 107 3.00 15.00 9.1682 2.22544
of these women were circumcised and 29 were uncircumcised.
Type 3 FSFI 54 2.30 27.60 16.6741 5.21891
The women were from Sudan Nyala Training and Research Hos-
pital, either working or accompanying the patients there, or who Desire 54 2.00 8.00 4.3333 1.37361
attended a department other than Obstetrics and Gynecology. Arousal 54 1.00 16.00 8.7778 3.26020
They were questioned about their ages and if they were circum- Lubrication 54 0.00 16.00 9.8148 3.63979
cised. They were examined to determine the type of circumcision, Orgasm 54 0.00 12.00 7.0556 2.49843
and the results were noted. Their sexual functions were evaluated Satisfaction 54 2.00 12.00 6.8333 2.32906
by using FSFI translated into Arabic. FSFI is an approved, brief, Pain 54 0.00 12.00 7.3519 2.70679
composite, and multidimensional questionnaire measure that eval-
uates female sexual functions. It consists of a total of 19 ques-
tions, and their distributions and items for being evaluated are as
follows: 2 for sexual desire (libido), 4 for subjective arousal, 4 for
lubrication, 3 for orgasm, 3 for satisfaction, and 3 for pain. Each
termined to be Type 1 circumcision (23.3%) (Table 2).
question is scored between 0 and 5. The scores of each question-
naire item are added individually within itself, and the total score Both the total FSFI scores and each individual score of
is calculated as previously stated [9]. sexual desire, arousal, lubrication, orgasm, satisfaction, and
Statistical analyses were performed by using the Mann-Whit- pain differed between the uncircumcised and circumcised
ney U-test, Kruskal-Wallis test, and chi-square test. A p value < women; these differences were statistically significant. In
0.05 was accepted as significant. particular, the difference determined between uncircum-
cised women and Type 3 circumcised cases was extremely
Results prominent (Table 3).
When the circumcision types were compared to each
The ages of women examined ranged between 17 and 65
other, the difference between Type 1 and Type 2 was not
years (mean: 33.54 ± 10.25). The ratio of women circum-
statistically significant, whereas the differences between
cised was determined as 87.9% (Table 1). The most com-
Type 1 and Type 3, and between Type 2 and Type 3 were
monly performed circumcision type was Type 2 (51%),
statistically significant.
followed by Type 3 (25.7%); the remaining cases were de-
 Which type of circumcision is more harmful to female sexual functions?
Discussion
The definition of sexual health of the WHO refers to a
state of physical, emotional, mental, and social well-being in
relation to sexuality [10]. Sexuality for women is a concept
that includes desirability, ability to give birth to a baby, and
body image, in addition to emotional, intellectual, and soci-
ocultural components [11]. Therefore, the problems experi-
enced related to sexual functions are extremely private,
irritable, and physically and socially destructive for women,
and may lead to emotional stress, partner disagreements,
and divorces. These problems decrease self-confidence and
quality of life in women, and affect their mental states [12].
It is essential to have a functioning body for a healthy
and happy sex life. FGM involves partial or total removal
of the female external genitalia. According to current esti-
mations, approximately 100-140 million women have al-
ready been circumcised, and additionally, approximately
two million women will be circumcised each year [1]. Type
1 and 2 are commonly performed in West African coun-
tries, and Type 3 in Somalia, Djibouti, Ethiopia, Egypt, and
Sudan. In the current study, FGM Type 2 was determined
to be the most concentrated in Darfur (51%), which was
followed by Type 3 (25.7%), and Type 1 (23.3%), respec-
tively (Table 2).
Circumcision is an illegal procedure in Sudan. It is
known that the performers and those allowing circumcision
will be sentenced or fined if they are caught; therefore, cir-
cumcision currently continues to be performed under un-
suitable and secret conditions. In central areas where people
with high levels of education live, the ratio of circumcision
is low (32-45% in Khartoum), whereas this ratio is high in
rural areas where inhabitants have low levels of education
(87% in Haj Yousif, 99.6% in Shendi) [13, 14]. According
to the Sudan Demographic and Health Survey (SDHS) per-
formed in 1989-1990, the circumcision ratio in the Darfur
area was reported to be 65%; in the current study, this ratio
was 87% [15]. This higher ratio was explained by the in-
tensive migration of people and refugees from the rural
areas to the center of Nyala, in which the present study was
performed, due to the fire fights and battles that began in
Darfur in 2003.
Many studies have determined that damage to the exter-
nal genitalia negatively affects women’s sexual life. In a
meta-analysis of the results of 15 studies examining the
sexual consequences of FGM in 12,671 cases from seven
different countries, performed by Berg and Denison, they
determined dyspareunia in 52% of women with FGM, ab-
sence of sexual desire in almost half of them, and decreased
sexual satisfaction in one-third of them [16]. Additionally,
in some studies, anxiety, depression, and post-traumatic
stress disorder were reported in women with FGM [17].
When sexual desire, arousal, lubrication, orgasm, sexual
satisfaction, pain, and total FSFI scores were compared in
the present study between 210 circumcised cases and 29
693
uncircumcised women, uncircumcised women were deter-
mined to possess higher scores (Table 4).
Type 1 is generally called ‘sunna’ in Sudan, and it refers
to removal of the clitoral prepuce only [18]. However, the
current study determined total or partial clitorectomy in all
cases with Type 1 FGM. The clitoris is important in sexual
arousal, sexual satisfaction, and orgasm. Partial or total re-
moval of this organ decreases sexual pleasure in women.
This fact was proven by the significant correction of sexual
functions in women with FGM after clitoral reconstruction
[19]. When sexual function scores were compared between
the types of circumcision in the current study, Type 1 and
2 did not differ significantly (Type 1 20.32 ± 4.15, Type 2
20.24 ± 4.087). This result was explained by the removal of
the clitoris in both types.
In Type 3,which is the most severe form of FGM, almost
all of the female external genitalia is excised, and the re-
maining parts are sewn together; this procedure narrows
the opening of the genital organ. Girls’ feet are then tied to-
gether and they are forced to stay in that position for many
days, in order to support the remaining excised parts to
join together. The resultant scars and adhesions in most of
them cause pain during sexual intercourse and difficulty
in intercourse, and therefore lead to sexual function disor-
ders [20, 21]. In the current study, the lowest FSFI scores
were determined in women with Type 3 FGM (FSFI score
16.67 ± 5.218). When the types of circumcision were com-
pared, sexual desire, arousal, lubrication, orgasm, sexual
satisfaction, and pain scores, as well as total FSFI scores
differed significantly between Type 1 and 3, and also be-
tween the Type 2 and 3 (Table 3).
Conclusion
In spite of laws against circumcision, and the intensive
studies of the WHO, UNICEF, and various local non-gov-
ernmental organizations, FGM continues to be performed
in high ratios in Sudan. Circumcision, and especially Type
3, is still an important health problem causing female sex-
ual function disorders in the women living in Darfur,
Sudan.
References
[1] World Health Organization, Department of Reproductive Health and
Research: “Eliminating female genital mutilation: an interagency
statement”. Geneva: WHO, 2008. Available at: http://apps.who.int/
iris/bitstream/10665/43839/1/9789241596442_eng.pdf
[2] Shah G., Susan L., Furcroy J.: “Female circumcision: history, med-
ical and psychological complications, and initiatives to eradicate this
practice”. Can. J. Urol., 2009, 16, 4576.
[3] Yoder P.S., Khan S., USAID: “Numbers of women circumcised in
Africa: the production of a total. DHS Working Paper 39”. Claver-
ton MD: Macro International, 2004.
[4] Perron L., Senikas V., Burnett M., Davis V., Society of Obstetricians
and Gynaecologists of Canada: “Female genital cutting”. J. Obstet.
Gynaecol. Can., 2013, 35, 1028.
694 O. Birge, D. Arslan, E.G. Ozbey, M. Adiyeke, I. Kayar, M.M. Erkan, U. Akgör
[5] MacLeod T.L.: “Female genital mutilation”. J. SOGC, 1995, 17, 333.
[6] Nour N.M.: “Female genital cutting: clinical and cultural guide-
lines”. Obstet. Gynecol. Surv., 2004, 59, 272.
[7] Catania L., Abdulcadir O., Puppo V., Verde J.B., Abdulcadir J., Ab-
dulcadir D.: “Pleasure and orgasm in women with Female Genital
Mutilation/Cutting (FGM/C)”. J. Sex. Med., 2007, 4, 1666.
[8] Utz-Billing I., Kentenich H.: “Female genital mutilation: an injury,
physical and mental harm”. J. Psychosom. Obstet. Gynaecol., 2008,
29, 225.
[9] Rosen R., Brown C., Heiman J., Leiblum S., Meston C., Shabsigh R.,
et al.: “The Female Sexual Function Index (FSFI): a multi-dimen-
sional self report instrument for the assessment of female sexual
function”. Sex. Marital. Ther., 2000, 26, 191.
[10] World Health Organization: “Developing sexual health programmes.
A framework for action”. WHO/RHR/HRP/10.22, 2010. Available
at: http://apps.who.int/iris/bitstream/10665/70501/1/WHO_RHR_
HRP_10.22_eng.pdf
[11] Shifren J.L., Monz B.U., Russo P.A., Segreti A., Johannes C.B.:
“Sexual problems and distress in United States women: prevalence
and correlates”. Obstet. Gynecol., 2008, 2, 970.
[12] Laumann E.O., Glasser D.B., Neves R.C., Moreira E.D. Jr.,
GSSAB Investigators’ Group: “A population-based survey of sex-
ual activity, sexual problems and associated help-seeking behavior
patterns in mature adults in the United States of America”. Int. J.
Impot. Res., 2009, 21, 171.
[13] Islam M.M,. Uddin M.M.: “Female circumcision in Sudan: future
prospects and strategies for erradication. International Family Plan-
ning Perspectives, 2001, 27, 71.
[14] Al Nagar S., Tønnessen L.: “Sudan Country Case Study: Child
Rights”. Commissioned by Norad and Sida, 2011:3 UTV Working
Paper, 2011: 3. Available at: http://www.sida.se/contentassets/
e3a7b0cb84274558afc2720451885d62/20113-sudan-country-case-
study-child-rights_3127.pdf
[15] Department of Statistics: “Sudan Demographic and Health Survey
1989/1990” Khartoum, Sudan: Department of Statistics; Columbia,
MD, USA: Institute of Resource Development/Macro International,
1991. Available at: http://dhsprogram.com/pubs/pdf/fr36/fr36.pdf
[16] Berg R.C., Denison E.: “Does female genital mutilation/cutting
(FGM/C) affect women’s sexual functioning? A systematic review of
the sexual consequences of FGM/C”. Sexual Res. Social Policy, 2012,
9, 41.
[17] Reisel D., Creightonb S.M.: “Long term health consequences of Fe-
male Genital Mutilation (FGM)”. Maturitas, 2015, 80, 48.
[18] Dareer A.E., Kheir H.M., Kumar S., Cross A.R., Islam M.: “Baseline
Survey on Reproductive Health and Family Planning”. Khartoum,
Sudan: Central Bureau of Statistics; New York: UNFPA, 1999. Avail-
able at: https://www.guttmacher.org
[19] Abdulcadir J., Rodriguez M.I., Petignat P., Say L.: “Clitoral recon-
struction after female genital mutilation/cutting: case studies”. J. Sex.
Med., 2015, 12, 274.
[20] Althaus F.A.: “Female circumcision: rite of passage or violation of
rights”. International Family Planning Perspectives, 1997, 23, 130.
[21] Shandall A.A.: “Circumcision and infibulation of females” Sudan
Medical Journal, 1967, 5, 178.
Corresponding Author:
O. BIRGE, M.D.
Department of Gynecology and Obstetrics
Nyala Sudan Turkey Training and Research Hospital
Pirsultan Abdal Mah. 809, District no. 42
Iğdır (Turkey)
e-mail: ozbirge@gmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Microwave endometrial ablation at a frequency of 2.45 GHz

for menorrhagia: analysis of its efficacy,
recurrence rate, and complications

K. Nakayama, K. Nakamura, T. Ishibashi, M. Ishikawa, S. Kyo

Department of Obstetrics and Gynecology, Shimane University School of Medicine, Shimane (Japan)

Summary
In late years, microwave endometrial ablation (MEA) has been attracting attention as an effective and minimally invasive treatment
alternative to hysterectomy. Microwave irradiation removes whole endometrium including its basal layer and reduces the amount of men-
strual bleeding. The authors performed MEA in 103 patients with hypermenorrhea from August 2007 to October 2012. As a note, all
patients had no hope of delivering. Among those patients, 72 cases were able to be enrolled for the evaluation. Then, the effectiveness
of MEA for the excessive menstruation was evaluated. As a result, the authors have reached the conclusion that MEA is a new effec-
tive treatment with safety and good cost performance for excessive menstruation. MEA should be considered as a standard treatment
for the conservative therapy-resistant excessive menstruation.

Key words: Hysterectomy; Microwave endometrial ablation; Menorrhagia.

Introduction cases that could be evaluated for improvement of menorrha-

Approximately six million women in Japan are said to be
suffering from menorrhagia. Currently, we are facing an
enormous revolution in how it can be treated. As of April
2014, microwave endometrial ablation (MEA) has been
listed for medical insurance coverage under “K863-C 3 Hys-
teroscopic Endometrial Ablation: 17,810 points”. Endome-
trial ablation was developed as a treatment to replace total
hysterectomy. Using microwaves or radiowaves, it induces
necrosis of the endometrial tissue and reports on this treat-
ment have been published since the 1980s. MEA using a
2.45-GHz microwave is a novel treatment for menorrhagia
that was developed by Kanaoka et al. [1]. It was first intro-
duced as minimally invasive treatment in August 2007 at the
Department of Obstetrics and Gynecology, School of Med-
icine, Shimane University. In June 2009, the present authors
became the fourth certified institution for advanced medical
care in Japan, and have progressively employed this treat-
ment in their practice. In Japan, an estimated 40,000 cases of
hysterectomy are conducted every year, and it is speculated
that 10,000 hysterectomies are avoided thanks to MEA. In
the present department alone, the authors performed 116
cases of MEA surgeries over the past five years and four
months, and their institution boasts experience with the
largest number of these surgeries in Japan. Until now, they
have published reports on the efficacy, safety, and minimal
invasiveness of MEA compared to traditional surgery [2-5].
In this study, the authors conducted an investigation of 72
gia after MEA, looking primarily at recurrences or compli-
cations, and would like to report their results.
Materials and Methods
MEA was performed in 72 patients with a chief complaint of men-
orrhagia who had no wish to bear children and who presented to the
present gynecology department between August 2007 and April
2012. The 72 cases who were six months or longer post-MEA and
in whom the authors could evaluate menorrhagia improvement, were
assessed for menstrual interstitial myoma volume, painful periods,
and satisfaction with treatment using visual analog scale (VAS)
scores. Before MEA was listed for insurance coverage, the Shimane
University Institutional Review Board had approved MEA treat-
ments. In addition, written informed consent was obtained after pa-
tients were provided with both written and oral explanations of the
procedure. After epidural anesthesia and placement in a lithotomy
position, the women underwent isodine disinfection of the lower ab-
domen, genitalia, thighs, and intravaginal area. Using a transab-
dominal or transrectal ultrasound guide, the entire surface of the
endometrium was ablated by MEA, while confirming coagulation of
the endometrium throughout the operation. A color Doppler was
used with the ultrasound guide, making it easy to confirm which
parts of the endometrium had been ablated (Figure 1). Using Mi-
crotas output 70W, the energizing time per coagulation was 50 sec-
onds in accordance with the MEA Treatment Guidelines [6].
Results
The 72 women treated with MEA ranged in age from 37
to 53 years with a median age of 46 years. All patients had

Revised manuscript accepted for publication November 19, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3467.2017
696 K. Nakayama, K. Nakamura, T. Ishibashi, M. Ishikawa, S. Kyo

a chief complaint of menorrhagia. Clinical diagnoses of the

Figure 1. — Ultrasound imaging of endometrium during MEA.
Colored area (arrow) indicates an irradiated site during ablation of
endometrium.
72 cases comprised 47 cases of myoma, 18 cases of uterine
adenomyoma (includes four cases of uterine myoma com-
plicated by uterine adenomyoma), nine cases of functional
menorrhagia, two cases of intrauterine polyps, and one case
of uterine cancer. None of the patients had plans to undergo
surgery and presented with massive genital bleeding when
they consulted the outpatient clinic, so that emergency
MEA was required in 23.6% of these patients. Among these
patients, there was one case of uterine cancer, but when the
patient presented to the outpatient clinic, she was suffering
from uncontrolled genital hemorrhage and so MEA was
performed to provide as much hemostasis as possible.
Presurgical Hb values ranged from 6.3-13.9 g/dL, with a
mean level of 10.0 g/dL. Surgical times ranges from 14 to
74 minutes with the mean at 37.4 minutes. Hemorrhage vol-
umes ranged from 0 to 300 mL with a mean of 17.0 mL. Hos-
pitalizations ranged from one to 12 days with a mean of 1.5
days. At six months or longer after MEA, menstrual volume,
painful menstruation, and satisfaction with treatment were

A N=72 B N=72

P 0.001 P 0.001
Dysmenorrhea VAS
VAS
Menorrhagia

Before MEA After MEA Before MEA After MEA

C N=72 D N=72
Number of patients
Hb g/dl

P 0.001

Before MEA After MEA Satisfaction (VAS)

Figure 2. — A: Change in the visual analog scale (VAS) score for menorrhagia prior to and following microwave endometrial ablation
(MEA). B: Change in VAS score for dysmenorrhea prior to and following MEA. C: Summary of patient satisfaction for MEA based on
VAS score. D: Change in the hemoglobin prior to and following MEA. Revised and cited from references 2, 3, 4, and 5.
Microwave endometrial ablation at a frequency of 2.45 GHz for menorrhagia: analysis of its efficacy, recurrence rate, and complications 697

Table 1. — Clinical factors and amenorrhea after MEA. were added to the present MEA treatment result data, and
Factors Patients Amenorrhea p-value some of these cases were added to factor analyses.
(number) Negative Positive
Age (years)
< 45 21 15 6 Discussion
≥ 45 51 33 18 0.58
Adenomyosis Although the present data on cases that illustrate the ther-
Positive 18 9 9 apeutic efficacy and effectiveness of MEA are limited, the
Negative 54 39 15 0.08 authors have reported their findings in these cases in the
Myoma: submucosal past [2-5]. This time, they conducted an investigation in a
Positive 28 21 7 larger number of cases, and found that the results remained
Negative 44 27 17 0.23
the same. They were again able to confirm MEA’s useful-
Myoma: intramural
Positive 27 25 2 ness. Improved VAS scores for menorrhagia and menstrual
Negative 45 23 22 0.0003 pain suggest major improvement in the patients’ QOL in
Myoma diameter > 5 cm terms of their menorrhagia. Objective evaluations are often
Positive 19 18 1 based on Hb values before and after treatment, and the Hb
Negative 53 30 23 0.003 value had improved by 2.3 g/dL at six months after MEA,
Multiple myomata showing that it had been effective in improving symptoms
Positive 18 15 3 of anemia. The incidence of menorrhagia recurrence after
Negative 54 33 21 0.08
MEA was 5.5% (4/72), which almost matches the results
Uterine sounding > 9 cm
Positive 42 31 11 of Kanaoka et al. [7]. In addition, the present authors in-
Negative 30 17 13 0.13 vestigated clinically-related factors that might be associ-
ated with the recurrence of menorrhagia, but were
MEA: microwave endometrial ablation.
unfortunately unable to identify any statistically significant
clinical factors that could be linked to this finding. Since
they may have failed to find statistically significant clinical
factors because there were too few cases, they are currently
evaluated through VAS scoring. VAS scores for menstrual
in the process of collecting cases with Dr. Kanaoka et al. for
volume improved from an average of 10 before surgery to a
use in a collaborative, large scale, multicenter study.
mean of 1.8 after surgery (p < 0.0001) (Figure 2A). Presur-
When the present authors looked at postoperative com-
gical VAS scores for painful menstruation improved from a
plications, not even one case of serious complications re-
mean of 5.0 to a postoperative mean of 1.4 (p < 0.0001) (Fig-
quiring emergency surgery for situations, such as
ure 2B). At six months post-MEA, Hb values increased from
postsurgical intestinal heat damage was noted. They sus-
10.2 before surgery to 12.5 g/dL post-surgery (p < 0.0001)
pect this was because they adhered to the indications call-
(Figure 2C). VAS scores for mean treatment satisfaction
ing for a normal myometrial thickness of one cm or
were 8.7 (Figure 2D). At six months after surgery, 25 out of
thicker as specified in the MEA guidelines. At the present
the 72 cases became amenorrheic, with a rate of 34.7%. In
institution, in order to further ensure patient safety, the
the 72 patients who could be evaluated for six or more
authors include color Doppler imaging during transab-
months after surgery, there were five patients in whom treat-
dominal and transrectal ultrasonography guidance (Fig-
ment was ineffective or menorrhagia recurred (5.3%). The
ure 1). As a result, it was easy to confirm the ablation
mean duration until recurrence was 9.7 months. Clinical fac-
points of the sounding applicator, allowing them to per-
tors that could have been associated with ineffectiveness or
form MEA more safely. The US FDA has also added a
recurrence were studied, but the present authors were unable
condition that the myometrium must be at least 1.0 cm in
to identify any statistically significant factors (data not
thickness to approve use of the MEA system. As of that
shown). In addition, ten out of the 72 cases (13.1%) devel-
time, the incidence of extrauterine organ damage in the
oped complications such as myometritis/endometritis. In
subsequent 5,000 cases performed has been 0 cases [8].
these patients, symptoms were alleviated with oral antibiotic
The present authors believe it is important to adhere to
treatment in six cases (60%), while four cases (40%) required
these guidelines strictly in order to prevent serious com-
i.v. antibiotic infusions. The present authors attempted to
plications, especially since MEA is predicted to spread
identify clinical factors that could have been related to the
rapidly throughout Japan hereafter [6]. In addition, as a
development of myometritis and endometritis, but nothing
mild complication of MEA, endometritis and myometri-
proved to be statistically significant (data not shown). In ad-
tis was observed in 13.8% (10/72). Interestingly, at one
dition, cases with myometrial gliomas and cases free of large
week post-MEA, not a single case presented with my-
interstitial myomas with diameters of five cm or more, were
ometritis/endometritis at the present outpatient clinic, and
significantly more likely to become amenorrheic after sur-
all cases of myometritis/endometritis developed their dis-
gery (Table 1). Furthermore, cases from the literature [2-5]
698 K. Nakayama, K. Nakamura, T. Ishibashi, M. Ishikawa, S. Kyo

Insurance coverage
20
19
18
18

16
Advanced medical treatment
14

12
No. of MEA

12

10
10
9
Non insurance coverage 8 8 8
8
7
6
6

4
4
3
2 2
2
1 1

0
20 ~4

10 8

20 ~4

20 5~8

20 ~4

20 ~9
20 5~8

20 ~4

20 5~8
20 ~4

20 12

20 12
20 12

20 12

20 12

Figure 3. — Number of MEA cases

0-
~
/1

/5

/1

/1

/5
/1
~

~
/1

/1
/
/

/9

/9
/8

/9

/9

10

10

11

11

12

12
08

09

09
08

12
increased from period of non-insur-
11
07

08

09

20
20

ance to period of advanced medical

Months treatment or insurance coverage.

ease two weeks after MEA. Up to April 2012 patients
were allowed to take baths one week after MEA. In other
words, there is a possibility that transvaginal bacterial in-
fection during bathing after the first week post-MEA
could have been the cause of these infections. Therefore,
as of June, 2012, bathing was prohibited for two weeks
after an MEA procedure. Thereafter, until now (end of De-
cember 2012), not a single case of myometritis/en-
dometritis has occurred. We must take into consideration
that there is a possibility that for about two weeks after
MEA, the area around the external opening of the uterus
has compromised contractility. In addition, cases that do
not have interstitial myomas, or those without myomas
with a larger diameter of five cm or more, were signifi-
cantly more likely to develop post-surgical amenorrhea
(Table 1). Cases that satisfy these conditions suggest that
the treatment efficacy of MEA and complete hysterec-
tomy are roughly the same. In other words, patients with
menorrhagia that satisfy these conditions should actively
avoid radical hysterectomies.
Recently, there have been reports from multiple institu-
tions in Japan on treatment results with MEA [9-12]. All
reports show similar data, suggesting the safety and effi-
cacy of MEA has been confirmed. Among these reports,
there is one paper that includes the results from an office
gynecology [12], and now that insurance coverage will pay
for the procedure, it is predicted that MEA will become
even more popular in Japan. In the UK where endometrial
ablation is already very popular, a meta-analysis reported
that bipolar radiofrequency and microwave ablative device
use was more effective than thermal balloon or fluid abla-
tion methods [13]. Moreover, patients were equally satis-
fied with bipolar radio frequency and microwave ablative
device use [14]. These reports should help MEA to become
more widespread within Japan.
MEA was certified as advanced medical care in Decem-
ber 2008 by the MHLW. Since that time, mixed healthcare
(mix of insurance-covered treatment with medical treatment
at one’s own expense) has become an option, and compared
to the previous situation when patients were required to pay
all costs themselves, the economic burden on patients has
been greatly alleviated. The number of cases undergoing
MEA at the present department has increased since June
2009 when it was certified as advanced medical care. Fur-
thermore, since April 2012, it has been listed as a treatment
covered by medical insurance under “K863-3 hysteroscopic
endometrial ablation surgery: 17,810 points”. Alfresa
Pharma Corporation and Kanaoka, Asakawa et al. held a
press seminar in Tokyo, in June 2012, and explained insur-
ance coverage for MEA [15]. Thereafter, many mass media
outlets began to cover MEA and this became the trigger that
led to laypersons learning about MEA. At the Department of
Obstetrics and Gynecology, School of Medicine, Shimane
University, the present authors began to see patients come to
their clinic for MEA treatment from other prefectures in the
Chugoku region (Hiroshima, Okayama, Yamaguchi), and
the number of cases with indications for MEA is rapidly in-
creasing. In Figure 3 the authors show the number of pa-
Microwave endometrial ablation at a frequency of 2.45 GHz for menorrhagia: analysis of its efficacy, recurrence rate, and complications 699

tients treated by MEA at their department, but after MEA [8] U.S. Food and Drug Administration (FDA): “Summary of safety and
was listed for medical insurance coverage, a sudden and dra- effectiveness data. Microwave Endometrial Ablation System (MEA)
- P20031”, 2003. Available at: http://www.fda.gov/ohrms/dock-
matic increase in the number of patients has been seen. ets/dailys/04/jan04/012704/04m-0031-aav0001-03-summary-of-
However, MEA recognition and knowledge has yet to safety-vol1.pdf
spread to laypersons and gynecologists at private clinics, [9] Tsuda A., Kanaoka Y.: “Outpatient transcervical microwave myol-
hence further activities will be necessary to promote dis- ysis assisted by transabdominal ultrasonic guidance for menorrha-
gia caused by submucosal myomas”. Int. J. Hyperthermia, 2015,
semination. 31, 588.
[10] Kanaoka Y., Imoto H.: “Transcervical interstitial microwave abla-
tion therapy for the treatment of adenomyosis: a novel alternative to
References hysterectomy”. Open J. Obstet. Gynecol., 2014, 4, 840.
[11] Ishikawa M., Katayama K., Yoshida H., Hirahara F.: “Therapeutic
[1] Kanaoka Y., Hirai K., Ishiko O., Ogita S.: “Microwave endometrial
outcomes and postoperative courses in microwave endometrial ab-
ablation at a frequency of 2.45 GHz. A pilot study”. J. Reprod. Med.,
lation for menorrhagia”. Journal of Microwave Surgery, 2012, 30,
2001, 46, 559.
253.
[2] Nakayama K., Yeasmin S., Katagiri A., Rahman M.T., Rahman M.,
[12] Tsuda A: “Microwave endometrial ablation for menorrhagia in office
Ishikawa M., et al.: “A comparative study between microwave en-
gynecology”. Journal of Microwave Surgery, 2012, 30, 71.
dometrial ablation and conventional surgical procedures for treat-
[13] Daniels J.P., Middleton L.J., Champaneria R., Khan K.S., Cooper
ment of menorrhagia”. Clin. Exp. Obstet. Gynecol., 2011, 38, 33.
K., Mol B.W., et al.: “Second generation endometrial ablation tech-
[3] Nakayama K., Ishibashi T., Ishikawa M., Katagiri A., Katagiri H.,
niques for heavy menstrual bleeding: network meta-analysis”. BMJ,
Iida K., et al.: “Microwave endometrial ablation at a frequency of
2012, 344, e2564.
2.45 GHz for menorrhagia: analysis of treatment results at a single
[14] Singh N., Hassanaein M.: “Comparing satisfaction rates of mi-
facility”. J. Obstet. Gynaecol. Res., 2014, 40, 224.
crowave and bipolar impedance controlled endometrial ablation”.
[4] Nakamura K., Nakayama K., Ishikawa M., Katagiri H., Katagiri A.,
Arch. Gynecol. Obstet., 2012, 285, 1301.
Ishibashi T., et al.: “Efficacy of multiple microwave endometrial ab-
[15] “Microwave endometrial ablation for menorrhagia”. Available at:
lation technique for menorrhagia resulting from adenomyosis”. J.
http://www.alfresa-pharma.co.jp/event/pdf/20120629MEA_
Obstet. Gynaecol. Res., 2015, 41, 1769
[5] Nakamura K., Nakayama K., Ishikawa M., Katagiri H., Ishibashi T., press_seminar.pdf
Sato E., et al.: “Efficacy of microwave ablation for endometrial car-
cinoma: a single center experience of 3 patients”. Oncol. Lett., 2016,
11, 3025. Epub 2016 Mar 23. Corresponding Author:
[6] Kanaoka Y., Ishikawa N., Asakawa Y., Nakayama K.: “Practice K. NAKAYAMA, M.D., Ph.D.
Guideline of MEA 2012”. Available at: http://www.alfresa- Shimane University School of Medicine
pharma.co.jp/microtaze/MEAguideline2012.pdf Enyacho 89-1, Izumo
[7] Kitaoka M., Kanaoka Y., Hirai K., Yasui T., Hattori K., Tokuyama O.,
Shimane, 6938501 (Japan)
Ishiko O.: “Microwave endometrial ablation for menorrhagia caused
by submucosal myomas”. Journal of Microwave Surgery, 2003, 21, e-mail: kn88@med.shimane-u.ac.jp
39.
CEOG Clinical and Experimental
Obstetrics & Gynecology

Retrospective evaluation of anaesthesia methods

in pregnant women with neurological and neuromuscular
syndromes who underwent caesarean section

A. Sargin, Z. Pestilci Cağıran, U. Ozdemir Biliç, B. Tanatti Orhanel, S. Karaman

Department of Anaesthesiology and Reanimation, Ege University School of Medicine, Izmir (Turkey)

Summary
Purpose: The purpose of this study was to investigate the anaesthesia methods used in pregnant women with neurological or neuro-
muscular disease who underwent caesarean section. Materials and Methods: Demographics; pregnancy weeks, urgent or elective cae-
sarean section, accompanying neurological or neuromuscular diseases, and anaesthesia type. Results: Of the pregnant women operated
on, 72% (16),14% (three) and 14% (three) were diagnosed with epilepsy, multiple sclerosis (MS), and myasthenia gravis (MG), re-
spectively. General anaesthesia was administered in 45%, 40%, and 25% of epileptic pregnant women, patients with MS, and those di-
agnosed with MG, respectively. Spinal anaesthesia was administered in 55%, 20%, and 75% of epileptic pregnant women, those with
MS, and those diagnosed with MG, respectively. Conclusion: Regional anaesthesia may be an appropriate option in pregnant women
with neurological or neuromuscular diseases. Epidural anaesthesia may be a safer method in terms of ensuring the control of block
level.

Key words: Epilepsy; Myasthenia gravis; Multiple sclerosis; Obstetrical anaesthesia.

Introduction
Pregnant women with neurological or neuromuscular dis-
ease have become more frequently observed due to an in-
crease in treatment options. Patients in this pregnancy
group are at a high risk of maternal mortality and morbid-
ity [1, 2]. Generally, neurological diseases are observed
more commonly in women and during childbearing years.
In addition, teratogenic complications associated with treat-
ment, along with the maternal risks, complicate cases.
Therefore, a multidisciplinary management approach is re-
quired.
Neurological and neuromuscular diseases cause some
structural and functional changes. Therefore, in order to
make an accurate decision and manage the anaesthesia
method competently, obstetric anaesthesiologists should be
aware of the pathophysiology of these diseases. The pres-
ent authors aimed to evaluate the methods of anaesthesia
performed in pregnant women diagnosed with epilepsy,
multiple sclerosis (MS), and myasthenia gravis (MG) over
a ten-year period, and their effect on the mother and new-
born in the early postoperative period.
Materials and Methods
Following the approval of the present hospital’s ethics commit-
tee, the records of pregnant women with neurological or neuro-
muscular disease (epilepsy, MS, and MG), who had undergone
caesarean section from 2004–2014, were examined retrospectively.
Out of 53 cases, 13 were excluded from the evaluation because of
vaginal delivery. For the included 40 patients, the following infor-
mation was extracted: demographics (age and weight), pregnancy
number, week, and parity, urgent or elective caesarean section, ac-
companying neurological or neuromuscular diseases, administered
treatment, disease history, anaesthesia type, new-born Apgar scores,
and the intensive care requirements of the mother and new-born.
Results
Of the 53 pregnant women with neurological or neuro-
muscular disease, 13 (25%) had undergone vaginal delivery
and 40 (75%) caesarean section. Of these pregnant women,
31, five, and four were diagnosed with epilepsy, MS, and
MG, respectively (Figure 1, Table 1). Caesarean section
was performed urgently in 55% (n=22) of cases and elec-
tively in 45% (n=18). Of the urgent cases, 16 (72%), three
(14%), and three (14%) patients were diagnosed with
epilepsy, MS, and MG, respectively.
When the time of diagnosis in pregnant women with
epilepsy was evaluated, in 14 patients, diagnosis was made
in ≤ two years, 12 were diagnosed between ten to 20 years,
and five had a history of epilepsy for > 20 years. Six of the
31 pregnant women diagnosed with epilepsy had a history
of seizure in the previous month. Seventeen (55%) patients
had spinal anaesthesia while 14 (45%) received general
anaesthesia (Figure 2).
When the pregnancy stage was considered, three patients

Revised manuscript accepted for publication March 16, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3638.2017
 A. Sargin, Z. Pestilci Cağıran, U. Ozdemir Biliç, B. Tanatti Orhanel, S. Karaman 701

Table 1. — History of pregnant women with neurological

disease.
Diagnosis (n) Time of diagnosis Treatments Clinical
(years) during pregnancy symptoms
E (31) 1–10 (14) No (10) No (25)
11–20 (12) AED (21) Yes (6)
≥ 21 (5)
MS (5) 1–5 (5) No (4) No (3)
CS(1) Yes (2)
MG (4) 1–5 (1) No (3) No (1)
5–10 (3) Mestinon (1) Yes (3)
AED: antiepileptic drug; CS: corticosteroid; E: epilepsy;
MS: multiple sclerosis; MG: myasthenia gravis.

Figure 1. — Proportion (%) of patient diagnoses. E: epilepsy, MS:

multiple sclerosis, MG: myasthenia gravis. Table 2. — Pregnancy period in patients diagnosed with
neurological disease.
Diagnosis (n) Pregnancy Neurological finding Caesarean
weeks (n) in pregnancy (n) type (n)
were ≤ 24 weeks pregnant. A total of seven patients were at E (31) ≤ 24 (3) No (25) Urgent (16)
25–36 weeks of pregnancy (Table 2). Only one woman was 25–35 (7) Yes (6) Elective (15)
≥ 36 (21)
25 weeks pregnant and was transferred to the intensive care
MS (5) ≤ 24 (0) No (1) Urgent (3)
unit due to neonatal respiratory distress. The Apgar scores 25–35 (1) Yes (4) Elective (2)
of new-borns of the remaining pregnant women were 9–10 ≥ 36 (4)
at one and five minutes. Intensive care was not required for MG (4) ≤ 24 (0) No (1) Urgent (3)
any of the patients in the post-partum period. 25–35 (4) Yes (3) Elective (1)
MS was diagnosed in four patients in the previous five- ≥ 36 (0)
year period. One of the pregnant women with a clinical E: epilepsy; MS: multiple sclerosis; MG: myasthenia gravis.
symptom had visual impairment, while another had numb-
ness in the hands (Table 2). Only one patient received cor-
ticosteroid treatment. The anaesthesia types of the pregnant
women are shown in Figure 2. Only one patient was less nant women had visual impairment, myopathy, and numb-
than 36 weeks pregnant. The Apgar scores of the new-borns ness in the hands (Table 2). While general anaesthesia was
of the pregnant women were 9–10 at one and five minutes. administered in one pregnant woman, spinal anaesthesia was
Intensive care was not required in any of the patients in the performed in the remaining three (Figure 2). The only pa-
post-operative period. tient who received drug treatment was the patient with my-
Four patients with MG were recorded. One patient was di- opathy. When the pregnancy stage was considered, all
agnosed two years previously. One patient had no clinical patients were between 29–33 weeks; however, new-born
symptoms and one patient was paraplegic. Two of the preg- Apgar scores were 9–10 at one and five minutes. Intensive

Figure 2. — Pregnant women with neurological disease and the type of anaesthesia selected. E: epilepsy, MS: multiple sclerosis, MG:
myasthenia gravis, Ep: epidural anaesthesia, S: spinal anaesthesia, GA: general anaesthesia.
702 Retrospective evaluation of anaesthesia methods in pregnant women with neurological and neuromuscular syndromes who underwent etc.
care was not required in any of the mothers or new-borns.
Discussion
Epilepsy is the most commonly observed neurological
disorder; it is associated with recurring seizures [1]. Preg-
nant women diagnosed with epilepsy are at a high risk of
sudden death. In addition, in order to be protected from the
teratogenic effect of anticonvulsant drugs, discontinuation
of treatment or reducing the dose and switching to other
drugs, can aggravate the disease. Hormonal changes can
also affect outcomes. Elevation of estrogen levels in the
early period of pregnancy triggers seizures and an increase
in progesterone levels causes an antiepileptic effect [2].
Previous studies on epileptic pregnant women have found
no difference in seizure frequency in 25–50% of the pa-
tients, whereas a reduction was observed in 13–14% [3, 4].
In approximately 17–32% of patients, an increase in the
frequency of seizures has been observed [3, 4]. In the pres-
ent study, 19% of patients had a history of seizure in the
previous month.
When selecting general or regional anaesthesia in epilep-
tic pregnant women, both adverse-effects of antiepileptic
drugs and their interaction with anaesthetic agents should
be known. For instance, haematological (including agran-
ulocytosis and aplastic anaemia) and neurological adverse-
effects (peripheral neuropathy) of anticonvulsant drugs,
such as phenytoin, barbiturates, and carbamazepine, are as-
sociated with regional anaesthesia. Due to the elevation of
anticonvulsant drugs in the liver, microsomal enzyme sen-
sitivity to opioids, neuromuscular blockers, and inhalation
agents increase. In addition, agents such as etomidate and
ketamine are known to be epileptogenic. It should be con-
sidered that, although the amide group of local anaesthet-
ics are also anticonvulsants during regional anaesthesia, the
high serum concentrations also cause convulsions [5]. All
maternal, fetal, and obstetric factors should be well-evalu-
ated and appropriate actions taken.
MS is a chronic, immune-mediated, inflammatory disease
characterised by neuroinflammation and neurodegeneration
in the central nervous system. The disease has an episodic
course. Some studies have demonstrated that MS remits in
the third trimester, in particular, during the pregnancy pe-
riod; however, in the first three months post-partum, it
shows a 70% increase compared with the pre-pregnancy pe-
riod [6-9]. This can be attributed to the suppression and
stimulation of cellular and humoral immunity that develop
during pregnancy. It has been suggested that cytokines se-
creted from fetoplacental structures cause an increase in cel-
lular immunity and sex steroids in humoral immunity [4]. In
a study conducted by the National Multiple Sclerosis Soci-
ety, no statistically significant difference was found between
epidural and general anaesthesia in groups of pregnant
women. In the same study, spinal anaesthesia caused an in-
crease in episode frequency due to an increase in the neu-
rotoxic effects of local anaesthetics [10]. Currently, this sub-
ject is still controversial, as some authors have failed to re-
port a difference with spinal anaesthesia [11].
It is essential to sufficiently evaluate the motor functions
of some patients, while determining the anaesthesia
method. The localisation of the demyelinating region in the
central nervous system and its effects on respiratory func-
tion should be carefully controlled. It should be known that
autonomic dysfunction in the lesions of the upper thoracic
region, in particular, may cause haemodynamic instability,
and hypotension that develops after regional anaesthesia
may have very serious consequences. In patients with poor
respiratory function, pre-oxygenation is vital during gen-
eral anaesthesia [5]. The choice between intravenous or in-
halation anaesthesia has not yet been clarified. Drugs
administered in the treatment MS are important for the
choice of anaesthesia method since baclofen causes sensi-
tivity to muscle relaxants and steroid use leads to hypoten-
sion as a result of adrenal insufficiency [12, 13].
MG is a neuromuscular disease that manifests as weak-
ness and loss of tonus in muscles. Autoantibodies develop-
ing against postsynaptic acetylcholine receptors are
responsible for its pathology. It is two-fold more common
in women and tends to emerge during childbearing years.
Hopkins et al. reported an improvement in symptoms in
30–40% of pregnant women with MS, no change in 30–
40%, and a clinical deterioration in approximately 20–30%
of patients. This study stated that symptoms worsened par-
ticularly in the first trimester and post-partum period [5].
Some studies have demonstrated that the risk of preterm
labour increases [14], as similarly observed in all patients
in the present study.
Currently, anaesthesia evaluation in pregnant women is
recommended in the antenatal period. Pulmonary and my-
ocardial functions of patients and the treatment adminis-
tered are of major importance in the choice of anaesthesia.
In pregnant women with MG, epidural anaesthesia or com-
bined epidural-spinal anaesthesia is more widely accepted
[5]. In order to prevent episodes in patients with MG, it is
beneficial to ensure the control of postoperative pain. In the
choice of local anaesthetic agents, the amide group of
anaesthetics should be selected due to the degradation of
the ester group of agents by anticholinesterases. If general
anaesthesia is used, the sensitivity to neuromuscular agents
is high. Therefore, many studies and case presentations
have reported that an adequate muscle relaxation can only
be achieved by inhalation agents. However, upon the avail-
ability of sugammadex, results indicate safe use of rocuro-
nium in this group of patients [15, 16].
Conclusion
The authors conclude that general anaesthesia can be ad-
ministered in the presence of neurological deficit, and re-
gional anaesthesia may be an appropriate option in pregnant
 A. Sargin, Z. Pestilci Cağıran, U. Ozdemir Biliç, B. Tanatti Orhanel, S. Karaman
women with neurological or neuromuscular diseases. They
propose that epidural or combined epidural-spinal anaesthe-
sia may be a safer method with regards to allowing block
level control.
References
[1] Pschirrer E.R.: “Seizure disorders in pregnancy”. Obstet. Gynecol.
Clin. North Am., 2004, 31, 73.
[2] Mattson R.H., Cramer J.A.: “Epilepsy, sex hormones and antiepilep-
tic drugs”. Epilepsia. 1987, 28, 29.
[3] The EURAP Study Group: “Seizure control and treatment in preg-
nancy: observations from the EURAP epilepsy pregnancy registry”.
Neurology, 2006, 66, 354.
[4] Gjerde I.O., Strandjord R.E., Ulstein M.: “The course of epilepsy
during pregnancy: a study of 78 cases“. Acta Neurol. Scand., 1988,
78, 198.
[5] Hopkins A.N., Alshaeri T., Akst S.A., Berger J.S.: “Neurologic dis-
ease with pregnancy and considerations for the obstetric anesthesi-
ologist“. Semin Perinatol., 2014, 38, 359.
[6] Confavreux C., Hutchinson M., Hours M.M., Cortinovis-Tourniaire
P., Moreau T., PRIMS Group: “Rate of pregnancy-related relapse in
multiple sclerosis”. N. Eng. J. Med., 1998, 339, 285.
[7] Korn-Lubetzki I., Kahana E., Cooper G., Abramsky O.: “Activity of
multiple sclerosis during pregnancy and puerperium”. Ann Neurol.,
1984, 16, 229.
[8] Douglass L., Jorgensen C.: “Pregnancy and multiple sclerosis”. Ann.
Obstet. Gynecol., 1948, 55, 332.
[9] Vukusic S., Hutchinson M., Hours M., Moreau T., Cortinovis-Tour-
niaire P., Adeleine P., et al.: “Pregnancy and multiple sclerosis (the
PRIMS study): clinical predictors of post-partum relapse“. Brain,
2004, 127, 15.
703
[10] Bennet K.A.: “Pregnancy and multiple sclerosis”. Clin. Obstet. Gy-
necol., 2005, 48, 38.
[11] Martucci G., Di Lorenzo A., Polito F., Acampa L.: “A 12-month fol-
low-up for neurological complication after subarachnoid anesthesia
in a parturient affected by multiple sclerosis”. Eur. Rev. Med. Phar-
macol. Sci., 2011, 15, 458.
[12] Dorottta I.R., Schubert A.: “Multiple sclerosis and anesthetic impli-
cations”. Curr. Opin. Anaesthesiol., 2002, 15, 365.
[13] Lee K.H., Park J.S., Lee S.I., Kim J.Y., Jim K.T., Choh W.J.: “Anaes-
thetic management of the emergency laparotomy for a patient with
multiple sclerosis- A case report”. Korean J. Anesthesiol., 2010, 59,
359.
[14] Stafford I.P., Dildy G.A.: “Myasthenia gravis and pregnancy”. Clin.
Obstet. Gynecol., 2005, 48, 48.
[15] Soyoral L., Goktas U., Cegin M.B., Baydi V.: “Successful use of
sugammadex for caesarean section in a patient with myasthenia
gravis”. Rev. Bras. Anestesiol., 2014. doi:10.1016/j.bjane.2014.
08.008
[16] Garcia V., Diemunsch P., Boet S.: “Use of rocuronium and sugam-
madex for caesarean delivery in a patient with myasthenia gravis”.
Int. J. Obstet. Anaesth., 2012, 21, 286.
Corresponding Author:
A. SARGIN, M.D.
Department of Anaesthesiology and Reanimation
Ege University School of Medicine
Kazim Dirik Mah./Bornova
35040 Izmir (Turkey)
e-mail: asuozdemir@hotmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Vaginal microbiota in asymptomatic Brazilian women with HIV

M.K. Figueiredo Facundo1, C.R. de Souza Bezerra Sakano2, C.R. Nogueira de Carvalho3,
A.M. de Oliveira Machado4, N.M. de Góis Speck1, J. Chamorro Lascasas Ribalta1
1 Gynecological Disease Prevention Nucleus (NUPREV) of the Gynecology Department of the Federal University of São Paulo, São Paulo
2 Pathology Department, of the Federal University of São Paulo, São Paulo
3 Gynecology Department of the Federal University of São Paulo, São Paulo
4 São Paulo Hospital’s Central Laboratory, Federal University of São Paulo, São Paulo (Brazil)

Summary
The purpose of this study was to evaluate the prevalence of different microorganisms, and the influence of menstrual cycle, CD4+
cells and viral load in vaginal flora, and compare different diagnosis methods in asymptomatic Human immunodeficiency virus HIV−
and HIV+ women. Variables like contraception methods, type of sexual intercourse, and menstrual cycle phase were significant between
groups. The clinical evaluation of vaginal pH and type of discharge, besides intraepithelial lesions, do not seem to have influence in mi-
croflora. Fresh wet-mount microscopy and bacterioscopy demonstrated no difference. HIV+ presented predominance of Gardnerella,
Candida, Trichomonas, and Mobiluncus in cervicovaginal cytology, and vaginal culture exhibited higher prevalence of Gram+ and co-
agulase-negative staphylococci. Fresh wet mount microscopy showed a sensitivity of 88.9%, and the bacterioscopy sensitivity was
75%. Clinical exam specificities were 76.3% and 94.9%, respectively. Asymptomatic HIV+ women may present diversified vaginal mi-
croenvironment, possibly making them more prone to pelvic inflammatory disease, sexually transmitted infection (STI), and infertil-
ity.

Key words: HIV; HIV-seropositive; Vaginal microenvironment; Bacterial vaginosis; LGT infection; Asymptomatic women.

Introduction more, this disorder can be asymptomatic in approximately

Human immunodeficiency virus (HIV) has been pre-
senting new cases of infections in Brazil. In 2014, the rates
recorded 47% of all new cases counted in Latin America,
and nearly 734,000 people are living with HIV. The preva-
lence in women aged 15-49 years old was 0.4% [1], and
appear to be more easily infected with HIV than men [2].
The predominant mode of transmission is sexual inter-
course, due to diversity in sexual behavior patterns, and
variations in biological and behavioral co-factors [3]. Con-
dom use, anal intercourse, male circumcision, and hor-
monal contraception can implicate in HIV transmission [2];
besides the stage of disease in the HIV infected partner,
treatment with antiretroviral drugs, the presence of another
sexually transmitted infection (STI), and bacterial vaginosis
(BV) may be the most important co-factors [3].
BV is a disorder where some microbiological alterations
of vaginal microflora takes place, characterized by de-
creased Lactobacillus sp. and overgrowth of Gardnerella
vaginalis, together with anaerobes and potentially patho-
genic bacteria [4, 5]. BV can be associated with HSV-2,
gonorrhea, syphilis, Trichomonas vaginalis, and HIV in-
fections [6, 7]. The presence of BV and absence of lacto-
bacilli decrease the H2O2 concentration produced by
Lactobacillus sp., which showed to be protective against
HIV and other inflammatory conditions [1, 7]. Further-
half of the women who develops it [4].
Supposedly, there are three different mechanisms for in-
creasing susceptibility to HIV infection in women with BV:
disruption of the vaginal epithelium and subsequent trans-
mission of HIV to subepithelium; number reduction of Lac-
tobacillus species leading to increased pH and reduced
H2O2 concentration [1, 8]; significant and reversible alter-
ations in cervical immune cells and local inflammatory cy-
tokines, influencing local HIV replication [9].
In this study, the appraisal of the vaginal microenviron-
ment (VM) in asymptomatic HIV+ women might con-
tribute to a better prognosis and knowledge about BV. The
aims of the present study was to determine and evaluate the
prevalence of different microorganisms in VM with non-
molecular affordable tests, the phase of the menstrual cycle
in relation to vaginal flora, the counting of CD4+ cells and
viral load, and the comparison of different diagnostic meth-
ods effectiveness in asymptomatic HIV+ women.
Materials and Methods
Study population
The transversal case-control observational study was conducted
at Gynecological Disease Prevention Nucleus (NUPREV) of the
Gynecology Department of the Universidade Federal de São
Paulo - UNIFESP/EPM, within the period from October 2008

Revised manuscript accepted for publication December 22, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3509.2017
Figueiredo Facundo, de Souza Bezerra Sakano, Nogueira de Carvalho, de Oliveira Machado, de Góis Speck, Chamorro Lascasas Ribalta
through April 2010. Ethics and Research Committee of the Uni-
versidade Federal de São Paulo UNIFESP/EPM under protocol
number 0510/08 approved the study. Written informed consent
was obtained from all participants prior to enrollment.
The authors selected 98 women and the age of patients ranged
from 17-40 years, that were divided into two groups identified as
control or HIV− and case or HIV+. All patients presented no gen-
ital symptoms. The non-inclusion criteria consisted in women dur-
ing the menstrual period, pregnancy and puerperal cycle, non-HIV
immunosuppressing conditions as diabetes mellitus or corticoid
and antibiotic therapy users, sexual intercourse history, and douch-
ing in the last 48 hours. Each group was submitted to a clinical
exam, vaginal fresh wet mount microscopy, bacterioscopy, cul-
ture, and cytology.
Clinical exam
Briefly, the patients were submitted to an anamnesis and they
were conducted to general physical exams, which included gyne-
cological examination using a non-lubricated speculum, expos-
ing the vaginal walls and characterizing the vaginal content related
to color, odor, and appearance, following the Amsel standard
methods. The Whiff test was performed using KOH 10%, fol-
lowing the standard method when necessary. The pH of vaginal
content was measured by color-fixed indicator strips, which was
evaluated and compared through the staining on the pH indicator
with a measurement range from 0 to 14, previously established,
according to manufacturer instructions.
The cervicovaginal swab specimens were collected according
to the standard technique for fresh wet mount microscopy, bacte-
rioscopic exam using Gram staining, aerobic bacteria culture, and
cytology.
All women underwent colposcopy after initial exams. Patients
with colposcopic alterations were forwarded to cervical and/or
vaginal biopsy.
Fresh wet mount microscopy
Collected cells were initially prepared for fresh wet mount mi-
croscopy by directly spreading cells onto a glass slide, immedi-
ately applying a drop of sodium chloride 0.9%, and covering with
a coverslip. The samples were observed through an optical mi-
croscope and the microorganisms identified.
Bacterioscopy
The samples collected by sterile swab were spread onto a glass
slide and dried without fixer, and then were forwarded to the Cen-
tral Laboratory of Hospital São Paulo. The slide glasses were
stained by the Gram method, where initially crystal violet dye was
used for one minute, followed by wash. The incubation with lugol
2% lasted one minute and the samples were washed with water.
The discoloration occurred by applying acetone-alcohol 30% and
washed with water, the samples were incubated in Fuchsin Ziehl,
diluted in water 1:10 for 30 seconds, and finally were washed,
dried, and examined in optical microscope. The assessment was
conducted according to the previously established method in the
Central Laboratory of Hospital São Paulo.
Culture
The samples for aerobic culture, equally collected by sterile
swab and stored in tubes, were inoculated into chocolate agar
PolyViteX (PVX) plate, blood agar - Columbia CNA agar + 5%
sheep blood plate, and eosin methylene blue (EMB) agar plate.
The samples were incubated in bacteriological incubator for five
to seven days. The conditions of incubation for chocolate agar and
blood agar was 35 ± 2ºC with 5-10% CO2, and EMB agar was 35
± 2ºC. When more than one type of bacterial growth was ob-
705
served, the specimens were isolated for 24 hours, and were then
identified. In aerobic culture some microaerophilic organisms
were highlighted.
Some plates showed fungal growth, which were separated and
forward to mycological laboratory. The sample for fungal culture
was inoculated into Sabouraud’s glucose agar Difco and Mycosel
agar, and were incubated at 36ºC. The yeast colonies were sepa-
rated from colonies of filamentous fungi. For bacteria, depending
on the morphology, the colonies were selected and the species
identified. As for fungi, the colonies were selected and identified
by phenotype depending on the morphology.
Cytology
Cervicovaginal swab specimens were collected from vaginal
fornix, ectocervix, and endocervix for cytology. The cytological
exams were performed according to the standard technique. Col-
lected cells were initially prepared for conventional cytology by
directly spreading cells onto a glass slide and immediately im-
mersed in ethyl alcohol 95% for fixation. The samples were im-
mersed in ethanol absolute for 30 minutes, followed by alcohol
70% and alcohol 50% for one minute each. After that, the glass
slides were washed with water for one minute, stained with Har-
ris’ hematoxylin solution for one minute, and washed again. The
samples were immersed in alcohol solution 50% /one minute, al-
cohol solution 70%/one minute, ethanol absolute /one minute, and
stained Orange G 6 solution. Three baths were performed with
ethanol absolute for one minute each and followed by Polychrome
solution EA 31 for one minute. Subsequently, the samples received
three baths of ethanol absolute for one minute each one, xylene ,
and ethanol absolute solution (half of each) for one minute and
only xylene for ten minutes. Finally, two drops of synthetic
Canada balsam were placed on the glass slides and the samples
were observed with microscopy.
Statistical analysis
Analyses were performed using SPSS version 16.0. Continu-
ous values are expressed as mean ± standard deviation and ana-
lyzed by the Student t-test or Mann-Whitney U- test, when the
normality was not observed. Categorical variables are presented
as absolute numbers and analyzed by the chi-square test or exact
test of Fisher. Two-sided p-values ≤ 0.05 indicate statistical sig-
nificance.
Results
The analysis included 51 women in the control group
ranging from 17 to 40 years old with a mean age of 29.69
± 7.20 years, and 47 women in the case group ranging from
25 to 40 years old, with a mean of 33.96 ± 3.96 years. The
mean of menarche age was 12.90 ± 1.72 years in the con-
trol group and 12.87 ± 1.83 years in the case group. Mean-
while, the academic level (p = 0.04), contraception (p <
0.001), and type of sexual intercourse (vaginal, oral, oral
and/or vaginal, vaginal and anal) were statistically signifi-
cant (p = 0.04).
Menstrual cycle was valued in both studied groups; in
HIV− women 31.4% were in the first phase, 21.6% were in
the second phase, and 47.1% were in the single phase due
to hormonal contraception use. In HIV+ women, 27.7%
were in the first phase, 61.7% in the second phase, and
10.6% in the single phase. The results were highly signifi-
cant (p < 0.001).
706 Vaginal microbiota in asymptomatic Brazilian women with HIV

Table 1. — Frequency of performed essays in 98 asympto- Table 2. — Vaginal flora of HIV− women and distribu-
matic women with HIV− and HIV+ and microorganisms tion according to the counting of CD4+ T cells.
detected. Culture CD4+ T - lymphocyte title (cells/mm3) p*
Groups Control HIV+ Total p* < 200 200-500 > 500 Total
N (%) N (%) N (%) N (%) N (%) N (%) N (%)
FRESH EXAM 0.39 Staphylococcus
1 (4.0) 12 (48.0) 12 (48.0) 25 (100)
Clue Cells 11 (21.6) 12 (25.5) 23 (23.5) coagulase -
Doderlein 29 (56.9) 24 (51.1) 53 (54.1) Staphylococcus
1 (50.0) - 1 (50.0) 2 (100)
Trichomonas -- 3 (6.4) 3 (3.1) aureus
Hyphae 6 (11.8) 6 (12.8) 12 (12.2) Gram+ bacteria 1 (10.0) 4 (40.0) 5 (50.0) 10 (100)
Intermediate 5 (9.8) 2 (4.3) 7 (7.1) Escherichia coli - - 1 (100) 1 (100)
Total 51 (100) 47 (100) 98 (100) Enterococcus sp. - 1 (100) - 1 (100)
0.496
BACTERIOSCOPY 0.90 Streptococcus B - - 2 (100) 2 (100)
Gram+ cocci 13 (25.5) 11 (23.4) 24 (24.5) Candida glabrata - - 1 (100) 1 (100)
Gram+ bacilli 29 (56.9) 24 (51.1) 53 (54.1) Candida albicans - 1 (100) - 1 (100)
Yeasts 1 (2.0) 2 (4.3) 3 (3.1) Klebsiella
- - 1 (100) 1 (100)
Gram+ and - bacilli 7 (13.7) 9 (19.1) 16 (16.3) pneumoniae
Gram+ bacilli 1 (2.0) 1 (2.1) 2 (2.0) No growth 1 (33.3) 1 (33.3) 1 (33.3) 3 (100)
Total 51 (100) 47 (100) 98 (100) Total 4 (8.5) 19 (40.4) 24 (51.1) 47 (100)
CYTOLOGY 0.05 *p value determined by chi-square.
Lactobacillus sp. 38 (74.5) 27 (57.4) 65 (66.3)
Trichomonas -- 2 (4.3) 2 (2.0)
Gardnerella 3 (5.9) 8 (17.0) 11 (11.2)
Cocci 8 (15.7) 4 (8.5) 12 (12.2)
Candida 1 (2.0) 5 (10.6) 6 (6.1) outcome was not significant.
Other 1 (2.0) 1 (2.1) 2 (2.0) In the cytology of cervicovaginal samples, a high number
Total 51 (100) 47 (100) 98 (100)
of women exhibited Lactobacillus sp., represented by
CULTURE 0.001
Staphylococcus coagulase 26 (51.0) 25 (53.2) 51 (52.0)
74.5% of HIV− and 57.4% of HIV+ women. Some mi-
Staphylococcus aureus 3 (5.9) 2 (4.3) 5 (5.1) croorganisms, such as T. vaginalis, Gardnerella vaginalis,
Gram+ bacteria -- 10 (21.3) 10 (10.2) cocci, Candida sp., Mobiluncus, and Actinomyces were
Escherichia coli 5 (9.8) 1 (2.1) 6 (6.1) found in some samples. Due to the low number found, the
Doderlein 12 (23.5) -- 12 (12.2) Mobiluncus and Actinomyces were grouped and named
Enterococcus sp. 2 (3.9) 1 (2.1) 3 (3.1) “other”; there was statistical significance (p = 0.05).
Streptococcus B -- 2 (4.3) 2 (2.0) Culture of aerobic microorganisms from vaginal specimen
Citrobacter sp. 1 (2.0) -- 1 (1.0) is described in Table 1. When comparing the menstrual cycle
Candida glabrata 1 (2.0) 1 (2.1) 2 (2.0)
phases and culture of microorganisms in HIV− and HIV+
Candida albicans -- 1 (2.1) 1 (1.0)
Klebsiella pneumoniae -- 1 (2.1) 1 (1.0)
patients, the coagulase-negative staphylococci highlighted
No growth 1 (2.0) 3 (6.4) 4 (4.1) the high incidence in the single phase in 58.6% of women.
Total 51 (100) 47 (100) 98 (100) The first phase showed that 51.7% of women had coagulase-
negative staphylococci, and 47.5% in the second phase, fol-
* p value determined by chi-square.
lowed by 20% of Gram-positive (Gram+) bacteria in the
second phase of menstrual cycle, 17% of Doderlein in the
first, and 13.8% in the single phase (p = 0.60).
The analysis of clinical exam presented 86.3% of HIV− The expression of Escherichia coli, Enterococcus sp.,
and 72.3% of HIV+ women with apparently physiological Candida glabrata, Candida albicans, and Klebsiella pneu-
vaginal discharge. Bacterial vaginosis were detected in moniae were low, but 52% of all women presented coagu-
13.7% and 27.7%, respectively. The vaginal pH showed that lase-negative staphylococci, 12.2% Doderlein, and 10.2%
45.1% and 34% of control and case patients were < 4.5 and Gram+ bacteria (p < 0.001).
54.9%, and 66% were > 4.5, respectively. Immune status of 47 HIV+ women was evaluated by
Fresh wet mount microscopic test presented clue cells means of CD4+ T cells and viral load quantification; 8.5%
and Doderlein bacilli in more than half the women. The T. showed values lower than 200 cells/mm3 featuring acquired
vaginalis, fungus shaped hyphae, intermediate flora that immunodeficiency syndrome (AIDS) (Table 2). There were
consists in bacilli and other bacterias (Table 1) were found no significant differences in the vaginal flora between
in the samples; there was no statistical significance (p = groups according to standard method (p = 0.496).
0.39). The analysis of bacterioscopy by Gram stain showed HIV+ women were divided into three subgroups classi-
the presence of Gram+ cocci, Gram+ bacilli, yeasts, Gram+ fied by viral load: viral load lower than 10,000 copies/ml,
and Gram-negative (Gram-) bacilli, and Gram- bacilli; the viral load between 10,000-50,000 copies/ml, and viral load
Figueiredo Facundo, de Souza Bezerra Sakano, Nogueira de Carvalho, de Oliveira Machado, de Góis Speck, Chamorro Lascasas Ribalta
more than 50,000 copies/ml. The majority exhibited viral
load lower than 10,000 copies/ml, i.e. 87.3% of women;
2.1% showed more than 50,000 viral copies/ml with 515
cells/mm3 of CD4+ T cells (p = 0.242).
In the HIV+ group, 12.8% were not using antiretroviral
drugs, but 87.2% of women were using the therapy and the
vaginal flora showed more diversity, included in these sam-
ples E. coli, Enterococcus sp., S. agalactiae, C. glabrata, C.
albicans, S. aureus, and K. pneumoniae with low incidence.
The higher prevalence was observed for coagulase-nega-
tive staphylococci and Gram+ bacteria. The use of anti-
retroviral drugs did not influence the type of vaginal flora
of HIV+ women (p = 0.619).
The evaluation of sensibility and specificity of diagnos-
tic methods used, such as clinical exam, fresh wet mount
microscopy test, bacterioscopy, and cytology for identifi-
cation of vaginal flora of all asymptomatic women were
compared to culture, which is the gold standard method.
The comparison of sensibility among methods in the con-
trol group was 41.7% to clinical exam, 50% to cytology,
66.7% to fresh wet mount microscopy test, and 75% to bac-
terioscopy. In the case group, 44.4% to clinical exam,
55.6% to bacterioscopy, 77.8% to cytology, and 88.9%
fresh wet mount microscopy test. This proportion of subject
with positive outcome were properly identified by the test.
Moreover, the analysis displayed that the fresh wet mount
microscopy test had 64.1% specificity; the bacterioscopy
had 66.7%. The specificity value was 82.1% to cytology
and 94.9% to clinical exam in HIV− group. The HIV+
group showed higher specificity to clinical exam with
76.3%, followed by cytology with 65.8%.
After colposcopic examination, 54.9% of all women did
not undergo biopsy. The intraepithelial lesion in cervix
and/or vagina were confirmed by histopathological exam-
ination, and the HIV− and HIV+ groups showed that 11.8%
and 23.4% presented low-grade intraepithelial lesions, re-
spectively. In the high-grade intraepithelial lesions, 7.8%
of HIV− and 6.4% of HIV+ groups were confirmed by
biopsy. Chronic cervicitis was observed in 25.5% of HIV(-
) and 10.6% of HIV(+) women (p = 0.17).
Discussion
The vaginal microenvironment is highly complex, due to
the hormonal cycles, that results in mucosal changes, and to
the multiple sexually transmitted pathogens [10, 11].
The absence of symptoms does not characterize the vagi-
nal mucosa as a healthy microenvironment [12, 13]. Dif-
ferent microorganisms might be found in the lower genital
tract, such as intermediate flora and BV [12]. The BV as-
sociated pathogens may activate the metabolic pathways
that influence certain innate and adaptive immune re-
sponses [1, 14]. The present results revealed the presence of
11.2% of G. vaginalis in all samples by cytology. Gopinath
and Iwasaki explained that Gardnerella dominance is not
707
significantly associated with inflammatory cytokines, indi-
cating that a single genus, even Gardnerella, is not a con-
sistent marker of vaginal inflammation [15].
Levels of estrogen vary depending on the menstrual cycle
phase and contraception use. Among the HIV+ women,
6.4% use hormonal contraception and almost all 61.7%
were in the second phase of menstrual cycle. Estrogen in-
creases the levels of available glycogen in epithelial cells,
which facilitates lactobacilli growth and lactic acid lower-
ing the vaginal pH [13, 15]. Moreover, the comparison be-
tween HIV+ and HIV− women and menstrual cycle seems
to be highly correlated. Wijgert et al. affirmed that menses
are the largest disturbing factor during the menstrual cycle
and might contribute to lactobacilli reduction, therefore
some shifts may occur that favor the appearance of BV as-
sociated bacteria, streptococci or other Gram-positive cocci
[13]. The menstrual cycle phase did not influence in the
composition of vaginal flora in both groups and the preva-
lence was the same in the first and the second phase.
The pH measured in both groups exhibited that 60.2%
(59) of women were with pH more than 4.5 and 66% (31)
of them were HIV+. Several studies highlighted that pH is
an important characteristic and the pH less than 4.5 may
prevent the transmission of pathogenic bacteria and viruses
including HIV [1, 3, 7, 16].
Antiretroviral drugs use, the number of CD4+ T cells, and
viral load did not involve in the statistically significant dif-
ference of vaginal flora, which was emphasized by the cul-
ture or by other diagnostic methods, including the women
with CD4+ T cells lower than 200 cells/mm3 and featured in
stage 3 of infection by HIV.
The clinical exam of vaginal content was considered
physiological in 79.6% of all women, the BV was observed
in 20.4% of them, and the prevalence was in the HIV+
women. Regardless of HIV+ or HIV− women, they pre-
sented similar quantity of clue cells, Dordelein bacilli, hy-
phae, and intermediate flora. The features in the
composition of vaginal flora in the fresh wet mount mi-
croscopy did not show large differences, but HIV+ women
had T. vaginalis. The bacterioscopy did not exhibited ex-
pressive differences: the control group presented a little
more Gram+ cocci and Gram+ bacilli than in the case
group.
In the cytology and aerobic cultures, there were discrep-
ancies statistically significant between the groups. Cervi-
covaginal cytology is used worldwide to observe cervical
and/or vaginal precursor lesions, but the vaginal microflora
were detected and showed statistical significance. A large
diversity of the vaginal microorganisms was observed in
the HIV+ women, besides G. vaginalis, Candida sp. and T.
vaginalis were present in the samples. Viral infections in
women with BV might be asymptomatic; once the women
in both groups resulted with high-grade intraepithelial le-
sion in 17.3% and 7.1% with low-grade, meanwhile more
studies are necessary.
708 Vaginal microbiota in asymptomatic Brazilian women with HIV
The culture is considered the main method used and it
was possible to observe that 18.37% of all asymptomatic
women presented different pathogenic microorganisms
such as S. aureus, E. coli, Enterococcus sp., Beta-hemolytic
streptococci, K. pneumoniae, and Citrobacter sp. The
Dorderlein bacilli were classified in the HIV+ group like
Gram+ bacteria, due to other types of Gram+ bacteria
found in the samples.
In the comparison between assays, the sensibility was
higher in bacterioscopy in the HIV− women, while the
fresh wet mount microscopy was higher in the HIV+
women. Nevertheless, the specificity was much higher to
clinical exam in control HIV− and case (HIV+) groups, as
well as the positive predictive value to control group
(HIV−) with 71.4%. The positive predictive value to case
(HIV+) was 35% in the cytology. This outcome emphasizes
the idea that the clinical exam is important in the diagnosis
of asymptomatic BV infections [17, 18]. G. vaginalis and
clue cells were confirmed in more than half of the women
with clinical exam.
Bacterioscopy, fresh wet mount microscopy, and cytol-
ogy did not differ with the cocci classification by subtypes
and an accurate evaluation by the means of culture is nec-
essary to identify the pathogenic one. The implementations
of inexpensive and simple execution exams such as vaginal
pH measure, fresh wet mount microscopy, and perhaps bac-
terioscopy by Gram stain and culture of vaginal content
may select women with susceptibility to acquire infectious
diseases.
Mirmonsef and Spear, in a review analysis, asserted that
microbiota of the genital tract appear to play an important
role in the HIV epidemic [16], and BV might be an inde-
pendent risk factor for the acquisition of STIs [19-21].
In conclusion, asymptomatic HIV-seronegative and
HIV+ seropositive women resulted with a higher preva-
lence of Doderlein bacilli and coagulase-negative staphy-
lococci in the vaginal content. The G. vaginalis,
Trichomonas sp. and Candida sp. come across as being
more common in HIV+ than HIV− women. The composi-
tion of vaginal microenvironment in both groups did not
seem to suffer the influence of menstrual cycle; CD4+ T
cells and viral load also did not appear to be influenced in
vaginal flora in this restricted group of HIV(+) women.
Clinical exam exhibited less sensibility and higher speci-
ficity in HIV+ woman and the higher sensibility was in the
fresh wet mount microscopic test, while the bacterioscopy
presented lower specificity. It is important to take into ac-
count that the evidences may improve the screening BV in
asymptomatic women with HIV, but more studies are nec-
essary to outline a profile of these women.
HIV+ women with BV confirmed by exams, even with-
out gynecological symptoms, may present more diversified
vaginal flora when compared to the HIV− women, and pos-
sibly this makes them more prone to pelvic inflammatory
disease, STI, and infertility.
Acknowledgements
Research supported by CAPES grant.
References
[1] Ministério da Saúde: “Boletim epidemiológico HIV/AIDS”. Min-
istério da Saúde Secretaria de Vigilância em Saúde. Brasília 2014, 1,
1.
[2] Petrova M. I., van den Broek M., Balzarini J., Vanderleyden J.,
Lebeer S.: “Vaginal microbiota and its role in HIV transmission and
infection”. FEMS Microbiol. Rev., 2013, 37, 762.
[3] Buvé A., Jespers V., Crucitti T., Fichorova R. N.: “The vaginal mi-
crobiota and susceptibility to HIV”. Aids, 2014, 28, 2333.
[4] Biagi E., Vitali B., Pugliese C., Candela M., Donders G. G. G.,
Brigidi, P.: “Quantitative variations in the vaginal bacterial popula-
tion associated with asymptomatic infections: A real-time poly-
merase chain reaction study”. Eur. J. Clin. Microbiol. Infect. Dis.,
2009, 28, 281.
[5] Baka S., Tsirmpa I., Chasiakou A., Tsouma I., Politi E., Gennimata
V., Kouskouni E.: “Inflammation on the cervical papanicolaou
smear: evidence for infection in asymptomatic women?”. Infect. Dis.
Obstet. Gynecol., 2013, 2013, 184302.
[6] Gallegos-Arreola M. P., Figuera-Villanueva L. E., Puebla-Pérez A.
M., Montoya-Fuentes H., Suarez-Rincon A E., Zúñiga-González G.
M.: “Association of TP53 gene codon 72 polymorphism with en-
dometriosis in Mexican women”. GMR, 2012 11,1401.
[7] Cone R. A.: “Vaginal Microbiota and Sexually Transmitted Infec-
tions That May Influence Transmission of Cell-Associated HIV”. J.
Infect. Dis., 2014, 210, 616.
[8] Witkin S. S., Linhares I. M., Giraldo P.: “Bacterial flora of the female
genital tract: function and immune regulation”. Clin. Obstet. Gy-
naecol., 2007, 21, 347.
[9] Rebbapragada A., Howe K., Wachihi C., Pettengell C., Sunderji S.,
Huibner S., et al.: “Bacterial vaginosis in HIV-infected women in-
duces reversible alterations in the cervical immune environment”. J.
Acquir. Immune. Defic. Syndr., 2008, 49, 520.
[10] Spear G. T., Sikaroodi M., Zariffard M. R., Landay A. L., French L.,
Gillevet P. M.: “Comparison of the diversity of the vaginal micro-
biota in HIV infected and -uninfected women with and without bac-
terial vaginosis”. J. Infect. Dis. 2008, 198, 1131.
[11] Ameur A., Meiring T. L., Bunikis I., Häggqvist S., Lindau C., Lind-
berg J. H., et al.: “Comprehensive profiling of the vaginal micro-
biome in HIV positive women using massive parallel semiconductor
sequencing”. Sci. Rep., 2014, 4, 4398.
[12] Zozaya-Hinchliffe M., Lillis R., Martin D. H., Ferris M. J.: “Quan-
titative PCR assessments of bacterial species in women with and
without bacterial vaginosis”. J. Clin. Microbiol., 2010 48, 1812.
[13] Van de Wijgert J. H. H. M., Borgdorff H., Verhelst, R., Crucitti T.,
Francis S., Verstraelen H., Jespers, V.: “The vaginal microbiota: what
have we learned after a decade of molecular characterization?”. PLoS
One, 2014, 9, 105998.
[14] Kyongo J. K., Crucitti T., Menten J., Hardy L., Cools P., Michiels J.,
et al.: “A cross-sectional analysis of selected genital tract immuno-
logical markers and molecular vaginal microbiota in Sub-Saharan
African women with relevance to HIV risk and prevention”. Clin.
Vaccine Immunol., 2015, 22, 526.
[15] Gopinath S., Iwasaki A.: “Cervicovaginal Microbiota: Simple Is Bet-
ter”. Immunity, 2015, 42, 790.
[16] Mirmonsef P., Spear G. T.: “The barrier to HIV transmission pro-
vided by genital tract Lactobacillus colonization”. Am. J. Reprod.
Immunol., 2014, 71, 531.
[17] Anderson M. R., Klink K., Cohrssen A.: “Evaluation of vaginal com-
plaints”. JAMA, 2004, 291, 1368.
[18] Yin Y., Wu Z., Lin C., Guan J., Wen, Y., Li L., et al.: “Syndromic and
laboratory diagnosis of sexually transmitted infection: a compara-
tive study in China”. Int. J. STD AIDS, 2008, 19, 381.
Figueiredo Facundo, de Souza Bezerra Sakano, Nogueira de Carvalho, de Oliveira Machado, de Góis Speck, Chamorro Lascasas Ribalta 709

[19] Saxena D., Li Y., Yang L., Pei Z., Poles M., Abrams W. R., Malamud Corresponding Author:
D.: “Human microbiome and HIV/AIDS”. Curr. HIV/AIDS Rep., M.K. FIGUEIREDO FACUNDO, M.D.
2012, 9, 44. Núcleo de Prevenção de Doenças Ginecológicas
[20] Nardis C., Mosca L., Mastromarino P.: ”Vaginal microbiota and viral
Departamento de Ginecologia
sexually transmitted diseases”. Ann. Ig., 2013, 25, 443
[21] Sharma H., Tal R., Clark N. A., Segars, J. H.: ”Microbiota and pelvic
Universidade Federal de São Paulo
inflammatory disease”. Semin. Reprod. Med., 2014, 32, 43. Rua Borges Lagoa, 380 – Vila Clementino
04038-000 São Paulo – SP (Brazil)
e-mail: mayara.kff@gmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Uterus and myoma histomorphology

İ. Ceylan1, T. Peker2, N. Coşkun3, S. Ömeroğlu3, A. Poyraz5

Ankara University, Institute of Health, Neuroscience Department, Sihhiye, Ankara
1

Department of Anatomy, Faculty of Medicine, Gazi University, Beşevler, Ankara

2
3
Department of Histology and Embryology, Faculty of Medicine, Gazi University, Beşevler, Ankara
4
Department of Pathology, Faculty of Medicine, Gazi University, Beşevler, Ankara (Turkey)

Summary
Objective: Uterine fibroids, or leiomyomas are common benign neoplasms of the myometrium. These neoplasms are composed of
large amounts of extracellular matrix and disarrayed smooth muscle tissue. The aim of this study was to examine the histomorpholog-
ical differences between myoma uteri and uterus. Materials and Methods: Thickness of muscle fascicles and collagen fibers, and vas-
cular and histological structures were evaluated by using morphometric, histomorphologic and immunohistochemical analyses, and
findings represented by using three-dimensional (3D) modeling. The authors used a light microscope and photos were captured using
a specific program for analysis. Light micrographs were assembled into 3D images. Results and Conclusion: Histological and 3D find-
ings demonstrated that the muscle fiber is a vital part of the myometrium and loss of its contractility indicates a significant deviation
from the uterine structure.

Key words: Histomorphology; Immunohistochemistry; Three-dimensional modeling; Uterine myoma; Uterus; Leiomyomas.

Introduction The authors investigated the histomorphology of normal

Uterine leiomyomas (fibroids or myomas) are benign
clonal tumors that arise from the smooth muscle cells of
the uterus. They appear clinically in approximately 25%
of women, although with the application of new imaging
techniques, the actual clinical prevalence may be higher
[1, 2].
At the microscopic level, myomas consist of whorled,
anastomosing fascicles of uniform, spindle-shaped smooth
muscle cells. The cells have indistinct borders and
eosinophilic cytoplasm; the nuclei are elongated and ex-
hibit finely dispersed chromatin. Myomas may show areas
of hemorrhage as well as cystic degeneration and micro-
calcification in some cases [3].
Deviations of uterine myomas from the normal anatomic
structure have been described elsewhere [4, 5]. It is diffi-
cult, however, to demonstrate these deviations in three-di-
mensional (3D) form. Computers can generate true 3D
models and several 3D visualization techniques have been
developed to enable one to visualize not only a flat (hori-
zontal) representation of a 3D object, but also an approx-
imation of its Z-axis [6].
It is useful to demonstrate microscopic structures and re-
lated pathologies using 3D images or animations, espe-
cially for medical education. Use of web-based 3D models
for anatomy training enhances education, but such 3D
models must be used as part of an integrated training pack-
age that includes other material including video clips, text
book descriptions, and self-assessment tools [7].
myometrium and uterine myoma using histochemical and
immunohistochemical techniques. Light microscopy was
used for image analysis. In addition, two-dimensional light
micrographs were assembled into 3D images that may be
useful for medical education, medical research, and clini-
cal studies.
Materials and Methods
Leiomyoma (n=8) and normal myometrium (n=8) tissue were
collected by consent from women undergoing laparoscopic my-
omectomy from the Department of Gynecology and Obstetrics
(Gazi University Faculty of Medicine). The present study was
approved by the Ethics Committee of Gazi University Faculty of
Medicine.
Uterine tissues were fixed in 10% neutral buffered formalin
and embedded in paraffin after routine histological procedures
were performed. Then, four-µm sections were obtained from
each paraffin block and stained with haematoxylin and eosin
(H&E), Masson trichrome, and Van Gieson [8].
The avidin-biotin peroxidase method was used for the im-
munohistochemical studies to investigate α-smooth muscle actin,
desmin, S-100, PGP 9.5, and CD56 activities [9]. The activity of
these antibodies was assessed semiquantitatively. Histological
and immunohistochemical analysis was performed using a light
microscope and photos were captured by using the Leica Q Win
3 program.
Light micrographs of myometrium and fibroid tissues were re-
constructed into 3D images using Cinema 4D Release 15 3D
modeling and animation software. The 3D modelling process in-
cluded sculpting, texturing, lightning, and image editing, re-
spectively. Sculpting of 3D histological structures was created

Revised manuscript accepted for publication May 12, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3744.2017
 İ. Ceylan, T. Peker, N. Coşkun, S. Ömeroğlu, A. Poyraz
Figure 1. Normal myometrium. ifa: interfascicular area; bv: blood
vessel; mf: muscle fascicle (original magnification H&E ×10).
Figure 3. — Myoma tissue. Dense collagen fiber bundles (co) are
visible (original magnification H&E ×40).
Figure 5. — Myoma tissue. Note hyalinized matrix (hm) (original
magnification Van Gieson ×10).
711
Figure 2. — Normal myometrium. Myocytes (m) are dark red and
collagen fibers (co) are blue. Endomysium (e) surrounds a single
muscle fiber and perimysium (p) surrounds muscle bundles (ori-
ginal magnification Masson trichrome ×10).
Figure 4. — Myoma tissue. Vacuolization (v) owing to atrophy
of myocytes can be seen (original magnification Van Gieson ×40).
according to histological sections. Three-dimensional images
were rendered using HDRI in tiff format at 800×600 dpi. This
allows preservation of details that may be lost due to limiting
contrast ratios. Histological 3D models benefit from this as it cre-
ates more realistic scenes than with the more simplistic lighting
models used.
Masson trichrome stained slides were measured for muscle fas-
cicle and the collagen fiber quantity in the connective tissue. Like-
wise Van Gieson stained sections were counted blood vessels for
analysis of vascular structures. All these analyzes were performed
using a specific program and ten areas were randomly selected in
six cross-sections from myoma uteri and uterus tissue. Statistical
analyses were performed using SPSS statistical software. All data
are expressed as means ± SD. Data obtained from the counts in bi-
nary groups were evaluated using the Mann Whitney U test; p val-
ues less than 0.05 were accepted as statistically significant [9].
712 Uterus and myoma histomorphology

Figure 6. — Actin immunoreactivity. a) Myometrial tissue (original magnification actin

antibody ×10). b) Myoma tissue (original magnification actin antibody ×40). Desmin
immunoreactivity. c) Myometrial tissue (original magnification desmin antibody ×10).
d) Myoma tissue, immunoreactivity in the walls of blood vessels (original magnifica-
tion desmin antibody ×40). S-100 immunoreactivity. e) Myometrial tissue (original
magnification S-100 antibody ×10). f) Myoma tissue (original magnification S-100 an-
tibody ×40). PGP 9.5 immunoreactivity. g) Myometrial tissue (original magnification
PGP 9.5 antibody ×10). h) Myoma tissue (original magnification PGP 9.5 antibody
×10). CD-56 immunoreactivity. i) Myometrial tissue (original magnification CD-56 an-
tibody ×10). j) Myoma tissue (original magnification CD-56 antibody ×10).

Results
Histological sections of normal uterine myometrium
layer have fusiform, smooth, and tightly packed muscle
fibers in to the fascicles. The authors found collagen fibers
in interfascicular areas together with a few blood vessels
of various diameters (Figure 1). Also, after Masson’s
trichrome and H&E staining, muscle fibers seen red and
collagen fibers seen blue. Most of the collagen fibers lo-
cated between the muscle fascicles. Interfascicular col-
lagenous stroma was not abundant and smooth muscle
fibers were in parallel to each other in the linear orienta-
tion. Also myocytes were fusiform, and nuclei were located
in the center and oval-shaped. (Figure 2). In the sections
that stained with van Gieson, blood vessels were clearly
seen in the interfascicular area.
 İ. Ceylan, T. Peker, N. Coşkun, S. Ömeroğlu, A. Poyraz 713

Figure 7. — Three dimensional models. a) Myometrium. b) Myoma. c) Hyalinized matrix. co: collagen; m: myocyte; bv: blood vessel;
n: nucleus; h: hyalinization; v: vacuolization. 3D models designed by Tuncay Peker.

Figure 8. — Measurements of diameter in collagen fibers and Figure 9. — Number of blood vessels.
muscle fascicles.

Table 1. — Semiquantitation of immunohistochemical Table 2. — Muscle fascicle, collagen of connective tissue

staining results. and blood vessel counts.
Myometrium Myoma Mean±SD Median (min/max)
Actin + +++++ Myometrium
Desmin + ++++ Muscle fascicle 189.10±409.23 175.81 (146.68/239.89)
S-100 ++ + Collagen 131.75±450.04 119.60 (84.99/211.80)
PGP 9,5 + ++ Blood vessel 10.50±2.41 10.25 (8.00/13.50)
CD56 + +++ Myoma
Muscle fascicle 49.91±119.93 51.21 (31.86/67.70)
Collagen 341.90±587.78 332.42 (271.42/448.94)
Blood vessel 4.33±1.63 4.00 (3.00/7.00)
Histological sections of myoma uteri showed a fibroid
structure due to increase of collagen fibers. Myocytes in
the fibroid structure lost their characteristic structure, with
mitotic and cytoplasmic atrophy, vacuolization, and had a and may have reflected loss of myofilaments or atrophy in
pale cytoplasm. In Masson’s trichrome stained sections of myocytes (Figure 4). Figure 5 shows an amorphous (hya-
myomas and muscle fibers with intense dark blue. With line) matrix between elongated and atrophic myocytes and
H&E stained sections, more collagenous activity was ob- decreased vascularity due to the increase fibroid areas.
served and disorganized arrangement of smooth muscle Actin (Figures 6a, b) and desmin (Figures 6c, d) antibody
fibers and collagen bundles also were thicker in fibroids are responsible for contraction in smooth muscle cells and
than in normal tissue (Figure 3). In the sections that stained CD56 (Figures 6i, j) antibody indicates an increase in pos-
with Van Gieson, cytoplasmic vacuolization was common itive carcinomas showing significantly strong immuno-
714 Uterus and myoma histomorphology
reactivity in the myometrial layers of myoma uteri with
semiquantitative evaluation of immunohistochemical stain-
ing. Similarly, semiquantitative evaluation of immunohis-
tochemical staining for S-100 (Figures 6e, f) and PGP 9.5
(Figures 6g, h) showed significantly weak immunoreactiv-
ity due to the decrease vascularity in the myometrial layers
of myoma uteri (Table 1).
When muscle fascicle diameter (p = 0.004) and counted
blood vessels (p = 0.004) were measured of the myoma
uteri, they were significantly decreased compared to normal
myometrium. On the other hand, diameter of collagen
fibers (p = 0.004), of the normal myometrium was signifi-
cantly decreased than myoma uteri (Table 2) (Figures 8, 9).
Traditional learning methods in anatomy and histology
focus on the use of textbooks, 2D illustrations or diagrams.
Spatial relations are difficult to appreciate, and this cer-
tainly applies to the histomorphological differences be-
tween normal uterus and myoma uteri. Visualising these
types of structures in 3D with interactive educational ma-
terial can greatly improve the understanding of spatial re-
lationships and retention of that knowledge. Therefore,
light microscopic images were reconstructed into 3D im-
ages, which made normal and pathologic findings more ap-
parent (Figures 7a–c).
Discussion
Because smooth muscle is important for coordinated con-
traction of the myometrium, disorganized arrangement of
smooth muscle fibers signals an important departure from
the normal contractile structure [4].
The present authors focused their investigation on
smooth muscle fascicles, collagen organization, and blood
vessel content of normal and uterine fibroid tissues. The fi-
broids exhibited more collagen than normal uterus; normal
uterus had more smooth muscle fibers and was more vas-
cular than fibroids. The present observations are consistent
with previous reports [4, 5, 10]. They are also consistent
with a structural transformation from smooth muscle cells
in normal uterus to fibroblasts in fibroids. Flake et al. re-
ported that after structural transformation from smooth
muscle cells in normal uterus to fibroblast-like cells in fi-
broids, collagen deposition within the tumors was variably
interfascicular, intrafascicular or a combination of these and
that the collagenous stroma was produced by the myocytes
[4]. These investigators also reported a loosely parallel ori-
entation of fibers and a disorganized arrangement of the fi-
broids. They found that Masson trichrome, H&E, and Van
Gieson stained fibroids exhibited more collagen than nor-
mal tissues.
Flake et al. found pale cytoplasm in myomas and my-
ocytes that were transformed to slender and elongated
structures in myoma tissues [4]. Atrophy of the cytoplasm
and nuclei of myocytes was common. These investigators
also reported cytoplasmic vacuolization, which may reflect
loss of myofilaments and other changes such as hyaliniza-
tion related to severe atrophy. [4].
Weiss et al. reported a 3D study of the muscle and colla-
gen fiber architecture of the human uterus [10]. Immuno-
histochemical staining of alpha-smooth muscle actin
demonstrated that fibrotic leiomyomas consisted of abun-
dant collagen fibrils arranged in a non-parallel manner,
whereas in healthy myometrium collagen bundles adjacent
to smooth muscle cells, they were sparse and well-aligned;
the present findings were comparable.
Interstitial ischemia results from excessive production of
collagen, which causes decreased microvascular density,
increased distance between myocytes and capillaries, nu-
tritional deprivation, and myocyte atrophy. The end stage of
this process is death of myocytes. Further studies are re-
quired to determine whether necrosis may be used to dis-
tinguish ischemic necrosis from malignant uterine smooth
muscle tumors.
Anatomy teaching is undergoing significant changes
owing to time constraints, limited availability of cadavers,
and advances in computer-assisted learning. Web3D offers
the ability to simulate the spatial relationships among
anatomical structures. More research is required, however,
to evaluate these resources, before they are introduced rou-
tinely into the undergraduate medical curriculum [11].
The present authors investigated histomorphological
differences between uterine myoma and normal uterus
using histochemical and immunohistochemical methods.
In addition, light micrographs were reconstructed into 3D
images. In this way, normal and pathologic findings were
rendered more understandable. Use of the 3D method pre-
sented here can give further insight into the scientific re-
search of the uterus.
References
[1] Buttram V.C. Jr., Reiter R.C.: “Uterine leiomyomata: etiology, symp-
tomatology, and management”. Fertil. Steril., 1981, 36, 433.
[2] Cramer S.F., Patel A.: “The frequency of uterine leiomyomas”. Am.
J. Clin. Pathol., 1990, 94, 435.
[3] Lynch A.M., Morton C.C.: “Uterus: leiomyoma”. Atl. Genet. Cyto-
genet. Oncol. Haematol., 2008, 12, 68.
[4] Flake G.P., Moore A.B., Sutton D., Kissling G.E., Horton J., Wicker
B., et al.: “The natural history of uterine leiomyomas: light and elec-
tron microscopic studies of fibroid phases, interstitial ischemia,
inanosis and reclamation”. Obstet. Gynecol. Int., 2013, 2013,
528376.
[5] Berto A.G.A., Sampaio L.O., Franco C.R.C., Cesar R.M., Michelacci
Y.M.: “A comparative analysis of structure and spatial distribution of
decorinin human leiomyoma and normal myometrium”. Biochim.
Biophys. Acta., 2003, 1619, 98.
[6] Rydmark M., Kling-Petersen T., Pascher R., Philip F.: “3D visuali-
zation and stereographic techniques for medical research and edu-
cation”. Stud. Health Technol. Inform., 2001, 8, 434.
[7] John N.W.: “The impact of Web3D technologies on medical educa-
tion and training”. Comp. Ed., 2007, 49, 19.
[8] Sheehan D., Hrapchak B.: “Theory and practice of histotechnology”.
Ohio: Battelle Press, 1980, 189.
[9] Coşkun N., Hatipoğlu M.T., Özoğul C., Korkmaz C., Akyol S.N.,
Micili S.C., et al.: “The protective effects of acetyl L-carnitine on
 İ. Ceylan, T. Peker, N. Coşkun, S. Ömeroğlu, A. Poyraz 715

testis gonadotoxicity induced by cisplatin in rats”. Balk. Med. J., Corresponding Author:
2013, 30, 235. T. PEKER, M.D.
[10] Weiss S., Jaermann T., Schmid P., Steampfli P., Boesiger P., Niederer Gazi University Faculty of Medicine
P., et al.: “Three dimensional fiber architecture of the non-pregnant
Department of Anatomy
human uterus determined ex vivo using magnetic resonance diffu-
sion tensor imaging”. Anat. Rec. Part A., 2006, 288A, 84.
Beşevler 06510 Ankara (Turkey)
[11] Brenton H., Hernandez J., Bello F., Strutton P., Purkayastha S., Firth e-mail: tpeker@gazi.edu.tr
T., Darzi A.: “Using multimedia and Web3D to enhance anatomy
teaching”. Comp. Ed., 2007, 49, 32.
CEOG Clinical and Experimental
Obstetrics & Gynecology

The association between cystatin C and metabolic syndrome

according to menopausal status in healthy Korean women

Y.J. Lee, K.Y. Yun, S.C. Kim, J.K. Joo, K.S. Lee
Department of Obstetrics and Gynecology, Medical Research Institute, Pusan National University School of Medicine, Busan (Korea)

Summary
Objective: Cystatin C (Cys-C) is used as a marker for the measurement of the glomerular filtration rate (GFR) in chronic kidney dis-
ease and as a predictive marker of cardiovascular disease. The authors investigated the relationship between serum Cys-C level and meta-
bolic syndrome (MetS). Materials and Methods: A total of 3,670 women who visited the health promotion center were included in the
cross-sectional study. Single logistic regression analysis was used to analyze the relationship between Cys-C and MetS in premenopausal
and postmenopausal women. One-way analysis with linear trends was performed to determine the association between serum Cys-C
level and MetS components. Results: The authors divided the subjects into four groups (premenopausal women with or without MetS,
postmenopausal women with or without MetS) and compared the interquartile range (IQR) of the basic characteristics in each group.
The level of Cys-C was increased only in postmenopausal women with MetS. The mean value of Cys-C increased progressively as the
number of MetS components increased and the p value for the trend (p < 0.0001) was lower in postmenopausal women with MetS. Lo-
gistic regression analysis showed that the mean value of Cys-C was higher in MetS women and that the odds ratio was higher in post-
menopausal women with MetS than in premenopausal women with MetS. These results indicated that the interaction between Cys-C
and MetS was higher in postmenopausal women with MetS. Conclusions: Higher Cys-C level was found to have a positive correlation
with MetS in Korean premenopausal and postmenopausal women. These interactions were more significant in postmenopausal women.

Key words: Cystatin C; Menopause; Metabolic syndrome.

Introduction
Cystatin C (Cys-C) is a member of the human cysteine
superfamily and is an extracellular inhibitor of cysteine
proteases. This low molecular protein (13.4kDa) can be
freely filtered by glomerulus but is not secreted and re-
absorbed at the renal tubule. It can be used to represent
the changes in the glomerular filtration rate (GFR), sim-
ilar to serum creatinine. Several studies were performed
to evaluate the usefulness of serum Cys-C level as a
marker of GFR. Serum Cys-C level was found to be a
better marker than serum creatinine level and also a bet-
ter representative of GFR than the serum level of beta 2-
microglobulin [1-3].
Renal function is an important prognostic factor for
cardiovascular disease (CVD). Hence, Cys-C has also
been studied as a biomarker for risk prediction of CVD.
Shlipak et al. reported that elderly patients with higher
Cys-C levels had a higher risk of mortality and cardio-
vascular events [4]. Taglieri et al. reported that increased
Cys-C was related to a higher risk of developing both
CVD and chronic kidney disease (CKD) and it was also
strongly associated with CVD [5]. However, the patho-
physiologic mechanisms of Cys-C in cardiorenal meta-
bolic syndrome are not totally understood [6].
Metabolic syndrome (MetS) has at least three of the
following clinical characteristics: abdominal obesity, in-
creased blood pressure (BP), impaired glucose tolerance
or diabetes, dyslipidemia (elevated levels of triglycerides
and low concentration of high-density proteins) [7].
These characteristics are relevant for the development of
CKD and CVD. Additionally, MetS is known as an im-
portant risk factor for cardiovascular disease incidence
and mortality [8-10]. Due to this intimate relationship be-
tween MetS and renal disease, as well as CVD, several
studies have been conducted to explore the relationship
between MetS and Cys-C. Magnusson et al. demon-
strated that Cys-C affected metabolic factors, particularly
abdominal obesity, thus contributing to the development
of MetS [11]. As described above, there were several re-
ports on the relationship between Cys-C and MetS in pa-
tients with underlying diseases but studies involving
healthy people are scarce.
Many clinical findings on MetS have emerged in post-
menopausal women. Menopause may be a predictor and
an independent risk factor for MetS [12-14]. Therefore,
the present authors studied the correlation between Cys-
C and MetS, as well as the differences in the interaction
between Cys-C and MetS in healthy women according to
menopausal status.

Revised manuscript accepted for publication January 11, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3529.2017
 Y.J. Lee, K.Y. Yun, S.C. Kim, J.K. Joo, K.S. Lee 717

Table 1. — The basic characteristics of premenopausal women with or without MetS and postmenopausal women with or
without MetS.
Variables Premenopausal women Premenopausal women Postmenopausal women Postmenopausal women
without MetS (n=1290) with MetS (n=116) without MetS (n=1777) with MetS (n=487)
Age, median (IQR), years 44.0 (38.0 - 48.0) 47.0 (42.5 - 50.5) 57.0 (54.0 - 62.0) 61.0 (56.0 - 66.0)
Body weight, median (IQR), kg 55.1 (50.8 - 60.1) 63.2 (59.0 - 71.2) 55.9 (51.5 - 59.9) 60.9 (56.1 - 66.4)
Waist circumference, median (IQR), cm 75.5 (71.0 - 80.0) 86.0 (82.5 - 90.0) 79.0 (74.0 - 84.0) 86.0 (82.0 - 91.0)
BMI, median (IQR), kg/m2 21.7 (20.2 - 23.6) 25.6 (23.9 - 27.9) 22.7 (21.1 - 24.5) 25.3 (23.6 - 27.4)
SBP, median (IQR), mmHg 111.0 (102.0 - 120.0) 131.0 (121.5 - 140.5) 117.0 (107.0 - 128.0) 132.0 (119.0 - 143.5)
DBP, median (IQR), mmHg 68.0 (63.0 - 74.0) 79.0 (73.5 - 86.0) 72.0 (65.0 - 78.0) 79.0 (71.5 - 86.0)
Total bilirubin, median (IQR), mg/dl 0.9 (0.7 - 1.1) 0.9 (0.7 - 1.0) 0.9 (0.7 - 1.1) 0.8 (0.7 - 1.1)
Direct bilirubin, median (IQR), mg/dl 0.2 (0.2 - 0.2) 0.2 (0.1 - 0.2) 0.2 (0.2 - 0.2) 0.2 (0.1 - 0.2)
Total protein, median (IQR), g/dl 7.2 (6.9 - 7.5) 7.3 (7.0 - 7.5) 7.2 (6.9 - 7.4) 7.3 (7.0 - 7.6)
Albumin, median (IQR), g/dl 4.3 (4.2 - 4.5) 4.4 (4.2 - 4.5) 4.4 (4.2 - 4.5) 4.4 (4.2 - 4.5)
BUN, median (IQR), mg/dl 12.6 (10.7 - 15.1) 13.0 (10.2 - 14.4) 14.6 (12.5 - 17.3) 14.6 (12.2 - 17.4)
Creatinine, median (IQR), mg/dl 0.7 (0.7 - 0.8) 0.7 (0.7 - 0.8) 0.7 (0.7 - 0.8) 0.7 (0.6 - 0.8)
Estimated GFR, median (IQR), ml/min/1.73m2 94.9 (85.6 - 105.9) 93.6 (81.5 - 101.8) 89.3 (78.8 - 100.1) 89.1 (78.3 - 100.9)
Phosphate, median (IQR), mg/dl 3.7 (3.4 - 4.0) 3.7 (3.4 - 4.1) 3.9 (3.5 - 4.2) 3.9 (3.5 - 4.3)
Calcium, median (IQR), mg/dl 9.3 (9.1 - 9.6) 9.4 (9.3 - 9.6) 9.5 (9.2 - 9.7) 9.6 (9.3 - 9.8)
Cystatin C, median (IQR), mg/l 0.7 (0.7 - 0.8) 0.7 (0.7 - 0.8) 0.8 (0.7 - 0.9) 0.9 (0.8 - 1.0)
Total cholesterol, median (IQR), mg/dl 186.0 (166.0 - 209.0) 202.0 (174.5 - 235.0) 209.0 (183.0 - 234.0) 209.0 (182.0 - 238.0)
Triglyceride, median (IQR), mg/dl 69.5 (53.0 - 94.0) 156.0 (116.0 - 195.0) 79.0 (58.0 - 107.0) 151.0 (103.0 - 190.0)
HDL-cholesterol, median (IQR), mg/dl 60.0 (52.0 - 70.0) 43.0 (38.0 - 48.0) 60.0 (52.0 - 71.0) 44.0 (39.0 - 49.0)
LDL-cholesterol, median (IQR), mg/dl 111.0 (92.0 - 133.0) 127.0 (103.5 - 155.0) 132.0 (108.0 - 155.0) 138.0 (112.0 - 163.0)
Glucose, median (IQR), mg/dl 85.0 (79.0 - 90.0) 93.5 (87.0 - 101.0) 87.0 (82.0 - 94.0) 98.0 (89.0 - 117.0)
Insulin, median (IQR), uIU/ml 3.6 (2.6 - 4.6) 5.3 (4.1 - 7.1) 4.0 (3.1 - 5.2) 5.4 (4.1 - 7.8)
Free T4, median (IQR), ng/dl 1.3 (1.2 - 1.4) 1.3 (1.2 - 1.5) 1.3 (1.2 - 1.4) 1.3 (1.2 - 1.4)
TSH, median (IQR), uIU/ml 1.6 (1.1 - 2.5) 1.7 (1.2 - 2.7) 1.7 (1.1 - 2.6) 1.7 (1.2 - 2.7)
All data are presented as the interquartile range (IQR). P-value of all data < 0.05. MetS: metabolic syndrome; IQR: interquartile range; BMI: body mass index;
SBP: systolic blood pressure; DBP: diastolic blood pressure; BUN: blood urea nitrogen; eGFR: estimated glomerular filtration rate; HDL: high-density lipoprotein;
LDL: low-density lipoprotein; TSH: thyroid stimulating hormone.

Materials and Methods
A total of 3,670 women who visited the health promotion cen-
ter at Pusan National University hospital from January 2011 to
December 2014 were included in the cross-sectional study. Infor-
mation on gynecological history including menstrual history, op-
erative history, and gynecological disease history, in addition to
medical history, medication use, and lifestyle were obtained using
self-report questionnaires and interviews with healthcare
providers. Using a standing stadiometer, body weight and height
were measured with light clothing while the subjects were bare-
foot. The values were rounded to the nearest 0.1 kg and 0.1 cm.
Body mass index (BMI) was calculated as the weight in kilograms
divided by the height in meters squared. BP was measured with an
automatic machine in a sitting position after a ten-minute rest.
The authors included all patients who completed their self-re-
port questionnaires without exception. In order to apply this study
to the normal population group, the authors set the exclusion cri-
teria as follows: (1) to minimize the effect of renal or liver dys-
function, patients with renal disease or liver disease were
excluded. Patients whose blood test results raised the suspicion
of renal disease and liver disease were also excluded (alanine
aminotransferase level higher than 60 U/L, a total bilirubin level
higher than 1.5 mg/L, an eGFR less than 60 ml/minute/1.73 m2);
(2) patients that received hormone therapy within the last 12
months; (3) those currently undergoing chemotherapy or radia-
tion therapy; (4) patients who had amenorrhea within the last year.
Blood was drawn from the antecubital vein for all subjects be-
tween 8:30 and 10:00 am following at least eight hours of fast-
ing. Cys-C was analyzed using the turbidimetric immunoassay
method. Liver and renal function tests included alanine amino-
transferase, aspartate aminotransferase, lipid profiles, blood urea
nitrogen (BUN), serum creatinine, and total bilirubin were meas-
ured.
The authors defined menopause according to the International
Menopause Society terminology. Menopause was recognized to
have occurred after 12 consecutive months of amenorrhea with
no other obvious pathological or physiological causes [15]. If the
patient had a hysterectomy, menopause was diagnosed as serum
FSH above 40 IU/ml.
As the National Cholesterol Education Program reported in the
Adult Treatment Panel III, metabolic syndrome can be defined by
applying one of three or more out of five diagnostic criteria. The
diagnostic criteria are as follows: 1) abdominal circumference
over 80 cm (Asian); 2) triglyceride level over 150 mg/dl; 3) HDL
cholesterol less than 50 mg/dl; 4) FBG over 110 mg/dl or DM; 5)
BP over 130/85 or hypertension medication [7].
The Statistical Analysis System (SAS) 9.3 program was used
for statistical analysis. All data were entered into a database and
were verified by a second independent person. The authors di-
vided the study population into four groups: premenopause with-
out MetS, premenopause with MetS, postmenopause without
MetS, and postmenopause with MetS. The basic characteristics
of the groups were investigated by single logistic regression analy-
sis with a significance level of 5% or less and each median level
was determined by interquartile range (IQR). The authors com-
pared the mean value of Cys-C and the number of MetS compo-
718 The association between cystatin C and metabolic syndrome according to menopausal status in healthy Korean women

Table 2. — Relationship between mean value of Cys-C and Results

number of MetS components. The basic characteristics of the study groups are presented
Total (n=3670) N (%) Cystatin C level,
mean (SD)
in Table 1. Compared to women without MetS, women with
0 components 1143 (31.1%) 0.75 (0.11) MetS had higher basic characteristics as follows: age, body
1 component 1142 (31.1%) 0.77 (0.13) weight, waist circumference, BMI, systolic BP, diastolic BP,
2 components 782 (21.3%) 0.81 (0.14) calcium, triglycerides, LDL-cholesterol, glucose, and in-
3 components 404 (11.0%) 0.84 (0.17) sulin. The estimated GFR and HDL-cholesterol were lower
4 components 163 (4.44%) 0.89 (0.22) in both premenopausal and postmenopausal women with
5 components 36 (0.98%) 0.89 (0.17) MetS (p < 0.05) The level of Cys-C was increased only in
p value for trend < 0.0001 postmenopausal women with MetS.
The relationship between the mean value of Cys-C and
Premenopausal women (n=1406) 1406 (38.4%) Cystatin C level,
mean (SD) the number of MetS components are shown in Table 2. The
0 components 626 (44.5%) 0.72 (0.12) mean level of Cys-C progressively increased as the number
1 component 455 (32.4%) 0.73 (0.12) of MetS components increased in both premenopausal and
2 components 209 (14.9%) 0.73 (0.10) postmenopausal women. However, this linear p value trend
3 components 91 (6.47%) 0.76 (0.10) line was lower in postmenopausal women (p < 0.0001)
4 components 22 (1.56%) 0.73 (0.13) than in premenopausal women (p < 0.0017) and both p-
5 components 3 (0.21%) 0.92 (0.16) value trend lines were significant in all groups.
p value for trend 0.0017 Table 3 shows the interactions between Cys-C and MetS
with the odds ratio. The predicted probabilities of MetS ac-
Postmenopausal women (n=2264) 2264 (61.6%) Cystatin C level,
mean (SD) cording to the mean value of Cys-C are presented in Figure
0 components 517 (22.8%) 0.78 (0.11) 1. Depending on the method used to calculate the mean
1 component 687 (30.3%) 0.80 (0.12) value of Cys-C, the increase in the mean Cys-C level dif-
2 components 573 (25.3%) 0.83 (0.14) fered in all groups and in the premenopausal and post-
3 components 313 (13.8%) 0.86 (0.17) menopausal groups according to the presence of MetS. The
4 components 141 (6.23%) 0.92 (0.22) median IQR of Cys-C was only higher in postmenopausal
5 components 33 (1.46%) 0.89 (0.17) women with MetS, but the mean standard deviation of Cys-
p value for trend < 0.0001
C level was higher in all three groups with MetS. P value
MetS: metabolic syndrome. was statistically significant in all three groups, but that of
premenopausal women groups was slightly higher. The
odds ratio of the Cys-C level was higher in postmenopausal
women than in premenopausal women. The predicted prob-
nents by single logistic regression analysis with a trend test. The abilities of MetS according to the mean value of Cys-C
interactions between Cys-C and MetS were evaluated to deter-
mine the predicted probabilities of MetS according to the mean were also higher and had steeper slopes in postmenopausal
level of Cys-C with the odds ratio. women (Figure 1).

Table 3. — The interactions between Cys-C and MetS with odds ratio.
Total Patients without MetS Patients with MetS p-value Odds 95% CI
3670 (100.0%) 3067 (83.6%) 603 (16.4%) ratio
Cystatin C (all)
Mean ± Std 0.79 ± 0.14 0.77 ± 0.13 0.86 ± 0.18 < 0.001 45.078 (24.457 - 83.087)
Median (IQR) 0.8 (0.7 - 0.9) 0.8 (0.7 - 0.8) 0.8 (0.7 - 0.9)
Range 0.4 - 2.7 0.4 - 1.9 0.4 - 2.7
Cystatin C (premenopausal)
Mean ± Std 0.73 ± 0.12 0.73 ± 0.12 0.76 ± 0.11 0.008 5.626 (1.574 - 20.106)
Median (IQR) 0.7 (0.7 - 0.8) 0.7 (0.7 - 0.8) 0.7 (0.7 - 0.8)
Range 0.4 - 1.9 0.4 - 1.9 0.5 - 1.1
Cystatin C (postmenopausal)
Mean ± Std 0.82 ± 0.15 0.80 ± 0.13 0.88 ± 0.19 < 0.001 33.240 (16.293 - 67.812)
Median (IQR) 0.8 (0.7 - 0.9) 0.8 (0.7 - 0.9) 0.9 (0.8 - 1.0)
Range 0.4 - 2.7 0.5 - 1.7 0.4 - 2.7
MetS: metabolic syndrome.
 Y.J. Lee, K.Y. Yun, S.C. Kim, J.K. Joo, K.S. Lee
Discussion
Cys-C can be used to estimate GFR and is less affected
by renal factors such as inflammatory, infectious and liver
diseases, and by extrarenal factors like age, gender, diet,
and body composition. Therefore, it was reported to be a
better marker of GFR than serum creatinine level [16, 17].
Liu et al. reported that serum Cys-C was closely related to
MetS components and indicated that monitoring Cys-C
might help to predict the development and prognosis of
MetS in the elderly [18]. However, the relationship is un-
clear in relatively young people. In the present study, the
average age was relatively young and the results showed
positive relationships between serum Cys-C level and MetS
component, especially in postmenopausal women. It is un-
clear whether these results are due to menopause or aging.
Also, it is still unclear whether menopause is one of the
causes of MetS and weight gain or not. However, from re-
cently published reports, the present authors can postulate
that the change in body fat distribution after menopause and
the effect on insulin are responsible for these results rather
than menopause itself. Postmenopausal women exhibit
metabolic changes, such as central fat redistribution and el-
evated fasting plasma glucose levels [19]. Estrogen defi-
ciency in menopausal women leads to decreased insulin
secretion and elimination, as well as increased insulin re-
sistance. Insulin resistance is also well known to be an im-
portant factor affecting renal dysfunction and disease
[20-22]. Insulin resistance is defined as decreased cellular
sensitivity to insulin and is connected to atherogenic dys-
lipidemia, hypertension, and prothrombotic state. Insulin re-
719
Figure 1. — The predicted probabilities
for metabolic syndrome according to the
mean level of cystatin-C. Menopause 0
(blue line); premenopause, menopause 1
(red line); postmenopause.
sistance may be mediated through multiple metabolic path-
ways [23, 24]. These relationships suggest the potential of
Cys-C as an indicator of MetS in postmenopausal women.
In the present study, the odds ratio of Cys-C level and the
probability of MetS were higher in postmenopausal women
at the same Cys-C level. Therefore, Cys-C can be used to
predict the likelihood of MetS and the severity of the MetS.
There are some limitations in this study. First, the basic
characteristics of the groups and the mean value of Cys-C,
as well as the number of MetS components were investi-
gated by single logistic regression analysis. Since metabolic
syndrome can be affected by various factors, eliminating
the effects of these variables using multiple logistic re-
gression analysis would have helped us determine a more
precise correlation between Cys-C and MetS. Second, the
present data showed the predicted probabilities of MetS ac-
cording to the mean level of Cys-C, but the authors could
not determine the exact cut-off value for Cys-C. When they
performed multiple logistic regression analysis to reduce
the effect between factors using the Chi-squared Automatic
Interaction Detector (CHAID) algorithm, the calculated p-
value for the cut-off level exceeded 0.05 and was not sta-
tistically significant. However, when the mean Cys-C level
was 1 to 1.5, the predicted probabilities of MetS was around
50 percent. This linear relationship can be used to predict
the likelihood of MetS according to the Cys-C level.
Despite these limitations, this study has value in that the
authors found that the serum Cys-C level has a positive cor-
relation with MetS in Korean premenopausal and post-
menopausal women, and the probability of MetS was
predicted according to the mean value of Cys-C. The au-
720 The association between cystatin C and metabolic syndrome according to menopausal status in healthy Korean women
thors also verified the relationship between serum Cys-C
level and MetS in middle aged women in this study. Further
investigations with larger patient groups and well-con-
trolled variables will be needed to establish the cut-off
value for Cys-C in predicting MetS.
Acknowledgement
The authors acknowledge assistance with statistical
analysis from the Pusan National University Hospital Clin-
ical Trial Center Biostatistics Office.
References
[1] Coll E., Botey A., Alvarez L., Poch E., Quinto L., Saurina A., et al.:
“Serum cystatin C as a new marker for noninvasive estimation of
glomerular filtration rate and as a marker for early renal impair-
ment”. Am. J. Kidney Dis., 2000, 36, 29.
[2] Salgado J.V., Neves F.A., Bastos M.G., Franca A.K., Brito D.J., San-
tos E.M., et al.: “Monitoring renal function: measured and estimated
glomerular filtration rates - a review”. Braz. J. Med. Biol. Res., 2010,
43, 528.
[3] Simonsen O., Grubb A., Thysell H.: “The blood serum concentra-
tion of cystatin C (gamma-trace) as a measure of the glomerular fil-
tration rate”. Scand. J. Clin. Lab. Invest., 1985, 45, 97.
[4] Shlipak M.G., Sarnak M.J., Katz R., Fried L.F., Seliger S.L., New-
man A.B., et al.: “Cystatin C and the risk of death and cardiovascu-
lar events among elderly persons”. N. Engl. J. Med., 2005, 352, 2049.
[5] Taglieri N., Koenig W., Kaski J.C.: “Cystatin C and cardiovascular
risk”. Clin. Chem., 2009, 55, 1932.
[6] Salgado J.V., Souza F.L., Salgado B.J.: “How to understand the as-
sociation between cystatin C levels and cardiovascular disease: Im-
balance, counterbalance, or consequence?” J. Cardiol., 2013, 62,
331.
[7] National Cholesterol Education Program (NCEP) Expert Panel on
Detection, Evaluation, and Treatment of High Blood Cholesterol in
Adults (Adult Treatment Panel III): “Third Report of the National
Cholesterol Education Program (NCEP) Expert Panel on Detection,
Evaluation, and Treatment of High Blood Cholesterol in Adults
(Adult Treatment Panel III) final report”. Circulation, 2002, 106,
3143.
[8] Galassi A., Reynolds K., He J.: “Metabolic syndrome and risk of car-
diovascular disease: a meta-analysis”. Am. J. Med., 2006, 119, 812.
[9] Gluba A., Mikhailidis D.P., Lip G.Y., Hannam S., Rysz J., Banach M.:
“Metabolic syndrome and renal disease”. Int. J. Cardiol., 2013, 164,
141.
[10] Nashar K., Egan B.M.: “Relationship between chronic kidney dis-
ease and metabolic syndrome: current perspectives”. Diabetes
Metab. Syndr. Obes., 2014, 7, 421.
[11] Magnusson M., Hedblad B., Engstrom G., Persson M., Nilsson P.,
Melander O.: “High levels of cystatin C predict the metabolic syn-
drome: the prospective Malmo Diet and Cancer Study”. J. Intern.
Med., 2013, 274, 192.
[12] Carr M.C.: “The emergence of the metabolic syndrome with
menopause”. J. Clin. Endocrinol. Metab., 2003, 88, 2404.
[13] Cho G.J., Lee J.H., Park H.T., Shin J.H., Hong S.C., Kim T., et al.:
“Postmenopausal status according to years since menopause as an
independent risk factor for the metabolic syndrome”. Menopause,
2008, 15, 524.
[14] Eshtiaghi R., Esteghamati A., Nakhjavani M.: “Menopause is an in-
dependent predictor of metabolic syndrome in Iranian women”. Ma-
turitas, 2010, 65, 262.
[15] Utian W.H.: “The International Menopause Society menopause-re-
lated terminology definitions”. Climacteric, 1999, 2, 284.
[16] Dharnidharka V.R., Kwon C., Stevens G.: “Serum cystatin C is su-
perior to serum creatinine as a marker of kidney function: a meta-
analysis”. Am. J. Kidney Dis., 2002, 40, 221.
[17] Seronie-Vivien S., Delanaye P., Pieroni L., Mariat C., Froissart M.,
Cristol J.P.: “Cystatin C: current position and future prospects”. Clin.
Chem. Lab. Med., 2008, 46, 1664.
[18] Liu P., Sui S., Xu D., Xing X., Liu C.: “Clinical analysis of the rela-
tionship between cystatin C and metabolic syndrome in the elderly”.
Rev. Port. Cardiol., 2014, 33, 411.
[19] Otsuki M., Kasayama S., Morita S., Asanuma N., Saito H., Mukai
M., et al.: “Menopause, but not age, is an independent risk factor for
fasting plasma glucose levels in nondiabetic women”. Menopause,
2007, 14, 404.
[20] Charlesworth J.A., Kriketos A.D., Jones J.E., Erlich J.H., Campbell
L.V., Peake P.W.: “Insulin resistance and postprandial triglyceride
levels in primary renal disease”. Metabolism, 2005, 54, 821.
[21] Nistala R., Whaley-Connell A.: “Resistance to insulin and kidney
disease in the cardiorenal metabolic syndrome; role for angiotensin
II”. Mol. Cell. Endocrinol., 2013, 378, 53.
[22] Stevenson J.C.: “Metabolic effects of the menopause and oestrogen
replacement”. Baillieres Clin. Obstet. Gynaecol., 1996, 10, 449.
[23] Grundy S.M.: “Hypertriglyceridemia, insulin resistance, and the
metabolic syndrome”. Am. J. Cardiol., 1999, 83, 25f.
[24] Miranda P.J., DeFronzo R.A., Califf R.M., Guyton J.R.: “Metabolic
syndrome: definition, pathophysiology, and mechanisms”. Am. Heart
J., 2005, 149, 33.
Corresponding Author:
J.K. JOO, M.D.
Department of Obstetrics and Gynecology
Pusan National University, School of Medicine
Gu-Deok-Ro 305, Seo-Gu
Busan (Republic of Korea)
e-mail: jkjoo@pusan.ac.kr
CEOG Clinical and Experimental
Obstetrics & Gynecology

What is the best initial cycle IVF protocol for patients

over 35 years old?

P. Telli Celtemen1, N. Bozkurt1, M. Erdem1, M.B. Celtemen2, A. Erdem1, M. Öktem1, R.O. Karabacak1
1 Gazi University Department of Obstetrics and Gynecology, Ankara
2 Cankiri Government Hospital, Department of Obstetrics and Gynecology Cankiri (Turkey)

Summary
Purpose: The aim of this study was to find the most effective ovarian hyperstimulation protocol for patients over 35 years old. Mate-
rials and Methods: This is a retrospective study of 390 first IVF cycles of patients older than 35 years, that had serum follicle stimulat-
ing hormone < 10 IU/L, and that had no co-existing endocrine disorders. Long (n=181), antagonist (n=71), and micro-dose flare-up
(n=138) protocols were evaluated. Results: Clinical pregnancy and live birth rates were highest in long protocol group and lowest in
micro-dose protocol group. The difference between long and micro-dose protocol groups was statistically significant (p < 0.05). In mul-
tivariate logistic regression analysis, picked-up oocyte count (p = 0.005), endometrium thickness at hCG day (p = 0.006), age (p = 0.006),
and antral follicle numbers (p = 0.013) were found to be predictive for obtaining clinical pregnancy. Treatment protocols were not found
to be predictive for obtaining clinical pregnancy (p > 0.05). Conclusion: Treatment protocols were not found to be predictive for obtain-
ing clinical pregnancy. Patient’s age, antral follicle number, endometrial thickness at hCG day, and picked-up oocyte counts directly ef-
fect the pregnancy rates. Long protocol affects these factors positively can be preferred in younger patients with higher antral follicle
numbers.

Key words: IVF hyperstimulation protocol; IVF outcome; Advanced age.

Introduction docrine disorders (diabetes mellitus, thyroid disorders, and adre-

Approximately fifteen percent of couples cannot have
children despite they want and this poses a problem [1].
Today women generally wait to have children until they
have a better social level. When fertility capacity change
with women age is analysed, it showed a decrease of 31%
in 35-39 compared to 20-24 years of age. It was found to
decrease in higher rates in older women [2]. Rapidly in-
creased loss of follicles, decreased oocyte quality, and re-
productive aging after 35 years of age show low responses
to assisted reproductive techniques [3].
The aim of this study was to find the most effective ovar-
ian hyperstimulation protocol without knowing the re-
sponses at first admittance in patients older than 35 years of
age, but follicle stimulating hormone levels did not increase
in high amounts in whom an anxiety was present about the
treatment response.
Materials and Methods
This is a retrospective study of 390 patients followed between
2004 and 2012 in Gazi University In Vitro fertilization Unit. All
of the patients were given IVF treatment for any of the infertility
causes (tuboperitoneal or male factor, endometriosis, anovulation,
unexplained infertility), older than 35 years, serum follicle stim-
ulating hormone level < 10 IU/L, and with no co-existing en-
nal and pituitary gland diseases). Patients were divided into three
groups according to ovarian hyperstimulation protocols used: pa-
tients given long protocol (n=181), antagonist protocol (n=71),
and micro dose flare-up (n=138) protocols. Effects of the proto-
cols used on clinical pregnancy and live birth rates were analysed.
Approval of the local ethics committee from “Gazi University
Clinical Research Ethics Committee” was taken before the study
was begun.
GnRH agonist long protocol: one mg daily leuprolide (1 mg)
was given for at least 14 days from subcutaneous route. Go-
nadotropin treatment was begun if the serum estradiol level was
lower than 50 pg/ml at the mid-luteal phase (on the 21st day) of
the cycle prior to gonadotropin was begun and menstrual bleed-
ing occurred. Leuprolide acetate treatment was decreased to a
dose of 0.5 mg/day and continued with the same dose during the
gonadotropin treatment.
Micro-dose flare-up protocol: oral contraceptive treatment con-
sisting ethinyl estradiol 0.03 mg + levonorgestrel 0.150 mg daily
was given between days 1 and 21 of the prior cycle of go-
nadotropin treatment. Leuprolide acetate 40 micrograms bid from
subcutaneous route was begun after two days from oral contra-
ceptive drug was stopped. Gonadotropin stimulation was begun at
a suitable dose after the day leuprolide acetate was begun. Pitu-
itary suppression was continued with the same dose during go-
nadotropin stimulation until the day of hCG.
GnRH antagonist protocol: gonadotropin stimulation was
begun on the third day of the cycle with the appropriate dose if the
cystic lesion was not found with ultrasonography without pitu-
itary suppression treatment. Cetrorelix 0.25 mg/day from subcu-

Revised manuscript accepted for publication May 15, 2014

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog1957.2017
722 P. Telli Celtemen, N. Bozkurt, M. Erdem, M.B. Celtemen, A. Erdem, M. Öktem, R.O. Karabacak

Table 1. — Epidemiologic and stimulation characteristics.

Long protocol (n=181) Antagonist protocol (n=71) Micro-dose protocol (n=138) p-value
Age (mean ± SD) 37.1±1.9a,b 39.5±3.2a 38.9±3.0b <0.001
Antral follicle count 7 (0-20)a,b 4 (0-16)a 4 (0-16)b <0.001
Male age (mean ± SD) 40.1±4.8 41.2±5.9 40.5±5.4 0.356
Body mass index (mean ±SD) 25.5±4.2 25.3±3.8 25.1±3.8 0.703
Duration of infertility (month) 90 (6-288)a 48 (6-276)a,c 72 (6-324)c <0.001
Baseline FSH (IU/L) 6.3 (2.0-9.9) 6.5 (1.4-9.7) 6.6 (1.4-9.9) 0.122
Serum E2 on day 3 (pg/dL) 47 (10-664) 45 (10-252) 48 (7.7-854) 0.666
≥ 17 mm follicle count 3 (0-12)a,b 2 (0-7)a 2 (0-9)b <0.001
Serum E2 on day of hCG (mean) 1564 (133-10210)a.c 1112 (100-5120)c 1204 (92-4844)a <0.001
Duration of gonadotropin stimulation (mean) 10 (6-17)a,b 9 (5-16)a,c 11 (5-20)b,c <0.001
Total dose of FSH (mean) 2387.5 (1050-6300)b 2125 (900-3875)c 3000 (900-6300)b,c <0.001
hCG day endometrial thickness (mean) 11 (7-19)a,b 10 (6-15)a 10 (6-17)b <0.001
a: The difference between long protocol and antagonist treatment groups was statistically significant (p < 0.001).
b: The difference between long protocol and micro-dose treatment groups was statistically significant (p < 0.001).
c: The difference between micro-dose and antagonist treatment groups was statistically significant (p = 0.005).

Table 2. — Laboratory and pregnancy outcomes.

Long protocol (n=181) Antagonist protocol (n=71) Micro-dose protocol (n=138) p-value
Oocytes retrieved 9 (0-47)a,b 4 (0-34)a 4 (0-42)b <0.001
MII oocytes 6 (0-36)a,b 3 (0-29)a 3 (0-37)b <0.001
Fertilized oocyte number 5 (0-28)a,b 2 (0-19)a 3 (0-29)b <0.001
Embryos transferred 3 (0-5)a,b 2 (0-5)a 2 (0-5)b 0.004
Implantation rate (%) 13.5±21.5b 13,9±27.0 7.3±17.9b 0.009
Clinical pregnancy rate 59 (32.6%)b 19 (26.8%) 24 (17.4%)b 0.009
Live birth rate 40 (22.1%)b 12 (16.9%) 11 (8.0%)b 0.003
Number of cancelled cycles (%) 21.0 28.2 31.2 0.108
a: The difference between long protocol and antagonist treatment groups was statistically significant (p < 0.05).
b: The difference between long protocol and micro-dose treatment groups were statistically significant (p < 0.05).

taneous route was begun when the dominant follicle reached a 14-
mm diameter and given with gonadotropins until the day of hCG.
Gonadotropin treatment dose was ordered due to ovarian re-
sponse. Statistical analyses were made by SPSS 11.5. Descriptive
statistics were shown as mean ± standard deviation or median
(minimum - maximum). Categorical variables were shown as
number of cases and percentages. Difference of means between
groups was evaluated with Student’s t-test and One way variance
analysis. Difference of medians were evaluated with Mann Whit-
ney U test and Kruskal Wallis test. Factors having statistically sig-
nificant effect on pregnancy at univariate analysis or thought to
have effect on pregnancy were evaluated with multivariate logis-
tic regression analysis. Statistical significance level was accepted
as p < 0.05.
Results
Comparison of demographics, endocrinologic variables,
and stimulation characteristics of these three stimulation
groups are summarized in Table 1. Baseline characteristics
as body mass index, male age, and third day estradiol and
FSH levels were similar among all three groups. However
age was lower (p < 0.001), and number of antral follicles,
mature follicle number ≥ 17 mm, estradiol and endometrial
thickness at hCG day was higher in long protocol group at
a statistically significant level (p < 0.001). Statistically sig-
nificant difference was found between gonadotropin treat-
ment duration and doses among groups (p < 0.001). The
shortest stimulation duration and least gonadotropin usage
were seen in antagonist protocol group.
Laboratory and pregnancy results are summarized in
Table 2. Picked-up oocyte number, fertilized oocyte num-
ber, and as a result transferred embryo numbers were sim-
ilar in antagonist protocol and micro-dose protocol groups,
but they were higher in long protocol group at a statisti-
cally significant level (p < 0.005). Implantation rate was
similar in long protocol and antagonist protocol groups and
higher than micro-dose protocol group (p < 0.05). Cycle
cancel rates were higher in micro-dose protocol group but
the difference was not statistically significant among the
three groups (21,0%, 28,2%, and 31,2% in long, antago-
nist, and micro-dose protocol groups, respectively; p =
0.108). Clinical pregnancy and live birth rates were highest
in long protocol group and lowest in micro-dose protocol
group. The difference between long and micro-dose proto-
col groups was statistically significant (p < 0.05).
 What is the best initial cycle IVF protocol for patients over 35 years old? 723

Table 3. — Multivariate logistic regression analysis of all amount of gonadotropin usage, and a lower cost in weak
probable significant factors for distinction of clinical preg- responsive patients in the literature [7,8]. Despite these
nant and non-pregnant group within study group. data, Malmusi et al. found in their study that total go-
Parameter Odds 95% Confidence p-value nadotropin dose used in micro-dose flare-up protocol was
ratio interval lower than antagonist protocol [9]. In the present study con-
Lower limit Upper limit
sistent with the literature, gonadotropin dose used and du-
Age 0.848 0.755 0.953 0.006
Long protocol 1.110 0.587 2.099 0.748 ration was lower in antagonist protocol than the long
Antagonist 1.783 0.828 3.840 0.139 protocol and were highest in micro-dose flare-up protocol
AFC 1.094 1.019 1.175 0.013 at the statistical significant level.
Total oocyte count 1.062 1.018 1.108 0.005 Malmusi et al. [9] found total and mature oocyte counts
Endometrium thickness 1.186 1.051 1.338 0.006 were higher in GnRH agonist group than GnRH antagonist
Sixth day estradiol 1.000 1.000 1.001 0.470 group. Prapas et al. showed that more oocytes had been ob-
tained with agonist protocol, but metaphase II oocyte counts
were similar with antagonist protocol [10]. Craft et al. found
higher oocyte counts and higher pregnancy rates when they
Multivariate logistic regression analysis in study popula- compared GnRH antagonist protocol results with patients’
tion is summarized in Table 3. According to this analysis; prior GnRH agonist cycles [11]. It was also shown that
picked-up oocyte count (p = 0.005), endometrium thickness lower counts of oocytes had been obtained with antagonist
at hCG day (p = 0.006), age (p = 0.006), and antral follicle protocol compared to micro-dose protocol in prospective
numbers (p. =0.013) were found to be predictive for ob- studies [9, 12]. However there was no difference of picked-
taining clinical pregnancy. Treatment protocols were not up oocyte and mature oocyte counts between GnRH ago-
found to be predictive for obtaining clinical pregnancy (p > nist and antagonist protocols in a meta-analysis published
0.05). in 2011 [13]. In the present study it was shown that higher
number of oocytes were obtained with long protocol con-
sistent with literature data [7, 8] and showed that picked-up
Discussion
oocyte counts directly effects the pregnancy rates.
Today demographic data show that women postpone hav- The GnRH agonist treatment was proposed to increase
ing children. It is a reality that natural fecundity decreases endometrial receptivity by decreasing nitric oxide syn-
with increasing age, but no interpretation can be made thethase levels and implantation success with micro-dose
about after what age it is impossible to have children. flare-up protocol was attributed to this proposal [14]. De-
In this study long protocol data was more successful than creased growth factor synthesis is believed to decrease es-
antagonist protocol and micro-dose protocol was the weak- trogen levels and cause insufficient endometrial growth for
est one regarding clinical pregnancy, and live birth rates in implantation in GnRH antagonist treated patients [15]. Mal-
controlled ovarian hyperstimulation applied patients at first sumi et al. also found higher implantation rates in micro-
admission, older than 35 years of age, and FSL level < 10 dose flare-up protocol supporting this knowledge [9]. There
IU. Further statistical analysis showed none of the protocols was no significant difference between implantation rates
had direct effect on pregnancy. between long and antagonist protocols in the study of Pra-
The best ovarian stimulation protocol in advanced age pas et al. [10]. Endometrial thickness was similar in antag-
patients must have acceptable cycle cancel rates, highest onist and micro-dose protocols and higher in long protocol,
numbers of the highest quality mature oocyte count, rea- and it was found to have direct effect on predicting preg-
sonable duration and costs, suitable endometrium for im- nancy in the present study. Implantation rates were similar
plantation, and maximum pregnancy and live birth rates. in long and antagonist protocol, but lower in microdose
Choice of the best protocol in advanced age patients is still flare-up protocol group at the statistically significant level.
controversial because of the heterogeneity of the treatment The negative effect of high dose gonadotropin used on en-
protocols used and patients’ clinical features. dometrium might cause low implantation rates but similar
Several GnRH agonist protocols were attempted in which transferred embryo numbers and endometrial thickness in
dose and timing were different. It was thought that high microdose flare-up and antagonist protocol groups.
dose gonadotropin treatment in weak responsive patients Antagonist protocol was suggested to be related with in-
might stimulate follicular development and decrease cycle creased pregnancy success in older patients [8]. On the con-
cancel rates in late 1980s [4]. However contrasting opin- trary, there are some data that show that long protocol was
ions were proposed after a short time [5]. It was proposed more successful in initial cycle compared to antagonist pro-
that decreasing gonadotropin requirement with decreasing tocol [10] and it was suggested that pregnancy rates were
GnRH agonist dose might be more rational for increasing higher with microdose protocol [12, 13, 16]. There was no
oocyte numbers [4, 6]. difference between protocols in several studies and
Antagonist protocol showed to have a short duration, less Cochrane 2010 review, but pregnancy rates were fewer in
724 P. Telli Celtemen, N. Bozkurt, M. Erdem, M.B. Celtemen, A. Erdem, M. Öktem, R.O. Karabacak
antagonist protocol [3, 8, 9, 12, 17-24].
Clinical pregnancy and live birth rates were found to be
higher in long than antagonist protocols and were signifi-
cantly lower in micro-dose flare-up than other protocols in
first cycle of controlled ovarian hyperstimulation treatment
in the present study. These results might be related with
long protocol that was preferred in relatively young patients
regarding the treatment duration and pregnancy rates might
be higher in this group of patients. However further statis-
tical analysis showed none of the protocols was directly
predictive when adjusted for age between groups.
Cycle cancel rates were similar among all three groups.
Antagonist protocol was suggested to be with a less cycle
cancel rate in weak responsive patients in some prospec-
tive studies [8]; however Prapas et al. suggested the con-
trary [10]. Cycle cancel rates were found lower in
antagonist protocol comparing same patient’ prior GnRH
agonist cycle results in two retrospective studies [11]. How-
ever two different meta-analyses published in 2006 and
2011 showed there were no differences detected in cycle
cancel rates between GnRH antagonist and agonist proto-
cols [12, 25].
Conclusion
The best stimulation protocol in advanced age patients
must have acceptable cycle cancel rates, potential highest
number of the best quality oocytes, acceptable costs and
duration, convenient endometrium for implantation, and
maximum pregnancy and live birth rates. Choice of the best
treatment protocol in older patients is still controversial due
to heterogeneity of the treatment protocols used and clini-
cal features of the patients in the literature.
Treatment protocols were not found to be predictive for
obtaining clinical pregnancy. It was concluded that patient
age, antral follicle numbers, endometrium thickness at hCG
day, and picked-up oocyte counts directly affect the preg-
nancy rates. Long protocol affecting these factors positively
can be preferred in younger patients with higher antral fol-
licle numbers. However antagonist protocol can be preferred
in older patients with low antral follicle numbers because
of the shorter stimulation time and due to less gonadotropin
usage. Success rates might be increased and cycle cancel
rates might be decreased with a proper treatment protocol
chosen individually. Therefore controlled ovarian hyper-
stimulation protocol must be chosen regarding critical fac-
tors like individual features, endocrinologic factors, and age
for achieving the highest success rates for each patient.
Acknowledgments
The authors thank the clinicians, embryologists, and
nurses of the Gazi University IVF Center.
References
[1] Cetinkaya M.B., Siano L.J., Benadiva C., Sakkas D., Patrizio P.: “Re-
productive outcome of women 43 years and beyond undergoing ART
treatment with their own oocytes in two Connecticut university pro-
grams”. J. Assist. Reprod. Genet., 2013, 30, 673.
[2] Dunson D.B., Colombo B., Baird D.D.: “Changes with age in the
level and duration of fertility in the menstrual cycle”. Hum. Reprod.,
2002, 17, 1399.
[3] Pandian Z., McTavish A.R., Aucott L., Hamilton M.P., Bhattacharya
S.: “Interventions for ‘poor responders’ to controlled ovarian hyper
stimulation (COH) in in-vitro fertilisation (IVF)”. Cochrane Data-
base Syst. Rev., 2010, 1, CD004379.
[4] Hofmann G.E., Toner J.P., Muasher S.J., Jones G.S.: “High-dose fol-
licle stimulating hormone (FSH) ovarian stimulation in low-respon-
der patients for IVF”. J. In Vitro Fert. Embryo Transf., 1989, 6, 285.
[5] Klinkert E.R., Broekmans F.J., Looman C.W., Habbema J.D., te
Velde E.R.: “Expected poor responders on the basis of an antral fol-
licle count do not benefit from a higher starting dose of go-
nadotrophins in IVF treatment: a randomized controlled trial”. Hum.
Reprod., 2005, 20, 611.
[6] Garcia-Velasco J.A., Isaza V., Requena A., Martínez-Salazar F.J.,
Landazábal A., Remohí J., Pellicer A., Simón C.: “High doses of
gonadotrophins combined with stop vs. non stop protocol of GnRH
analogue administration in low responder IVF patients: a prospec-
tive randomized controlled trial”. Hum. Reprod., 2000, 15, 2292.
[7] Al-Inany H., Aboulghar M.: “GnRH antagonist in assisted repro-
duction: a Cochrane review”. Hum. Reprod., 2002, 17, 874.
[8] Marci R., Caserta D., Dolo V., Tatone C., Pavan A., Moscarini M.:
“GnRH antagonist in IVF poor-responder patients: results of a ran-
domized trial”. Reprod. Biomed Online, 2005, 11, 189.
[9] Malmusi S., La Marca A., Giulini S., Xella S., Tagliasacchi D.,
Marsella T.,Volpe A.: “Comparison of a gonadotropin-releasing hor-
mone (GnRH) antagonist and GnRH agonist flare-up regimen in
poor responders undergoing ovarian stimulation”. Fertil. Steril.,
2005, 84, 402.
[10] Prapas Y., Petousis S., Dagklis T., Panagiotidis Y., Papatheodorou
A., Assunta J., Prapas N.: “GnRH antagonist versus long GnRH ag-
onist protocol in poor IVF responders:a randomized clinical trial”.
Eur. J. Obstet. Gynecol. Reprod. Biol., 2013, 166, 43.
[11] Craft I., Gorgy A., Hill J., Menon D., Podsiadly B.: “Will GnRH an-
tagonists provide new hope for patients considered ‘‘difficult re-
sponders’’ to GnRH agonist protocols?” Hum. Reprod., 1999, 14,
2959.
[12] Demirol A., Gurgan T.: “Comparison of microdose flare-up and an-
tagonist multiple-dose protocols for poor-responder patients: ran-
domized study”. Fertil. Steril., 2009, 92, 2.
[13] Pu D., Wu J., Liu J.: “Comparisons of GnRH antagonist versus
GnRH agonist protocol in poor ovarian responders undergoing IVF”.
Hum. Reprod., 2011, 26, 2742.
[14] Wang J., Zhou F., Dong M., Wu R., Qian Y.: “Prolonged go-
nadotropin releasing hormone agonist therapy reduced expression of
nitric oxide synthase in the endometrium of women with en-
dometriosis and infertility”. Fertil. Steril., 2006, 85, 1037.
[15] Hernandez E.R.: “Embryo implantation and GnRH antagonists: em-
bryo implantation: the Rubicon for GnRH antagonists”. Hum. Re-
prod., 2000, 15, 1211.
[16] Tarlatzis B.C., Zepiridis L., Grimbizis G., Bontis J.: “Clinical man-
agement of low ovarian response to stimulation for IVF: a system-
atic review”. Hum. Reprod. Update, 2003, 9, 61.
[17] Detti L., Williams D.B., Robins J.C., Maxwell R.A., Thomas M.A.:
“A comparison of three downregulation approaches for poor re-
sponders undergoing in vitro fertilization”. Fertil. Steril., 2005, 84,
5
[18] Cheung L.P., Lam P.M., Lok I.H., Chiu T.T., Yeung S.Y., Tjer C.C.,
Haines C.J.: “GnRH antagonist versus long GnRH agonist protocol
in poor responders undergoing IVF: a randomized controlled trial”.
Hum. Reprod., 2005, 20, 616.
 What is the best initial cycle IVF protocol for patients over 35 years old? 725

[19] De Placido G., Mollo A., Clarizia R., Strina I., Conforti S., Alviggi [24] Devesa M., Martinez F., Coroleu B., Tur R., Gonzalez C., Rodriguez
C.: “Gonadotropin-releasing hormone (GnRH) antagonist plus re- I., Barri P.N.: “Poor prognosis for ovarian response to stimulation: re-
combinant luteinizing hormone vs. a standard GnRH agonist short sults of a randomised trial comparing the flare-up GnRH agonist pro-
protocol in patients at risk for poor ovarian response”. Fertil. Steril., tocol vs. the antagonist protocol”. Gynecol. Endocrinol., 2010, 26,
2006, 85, 247. 509.
[20] Tazegul A., Gorkemli H., Ozdemir S., Aktan T.M.: “Comparison of [25] Franco J.G. Jr., Baruffi R.L., Mauri A.L., Petersen C.G., Felipe V.,
multiple dose GnRH antagonist and minidose long agonist protocols Cornicelli J., et al.: “GnRH agonist versus GnRH antagonist in poor
in poor responders undergoing in vitro fertilization: a randomized ovarian responders: a meta-analysis”. Reprod. Biomed. Online, 2006,
controlled trial”. Arch. Gynecol. Obstet., 2008, 278, 467. 13, 618.
[21] Wang B., Sun H.X., Hu Y.L., Chen H., Zhang N.Y.: “Application of
GnRH-antagonist to IVF-ET for patients with poor ovarian re-
sponse”. Zhonghua Nan Ke Xue, 2008, 14, 423.
[22] Kahraman K., Berker B., Atabekoglu C.S., Sonmezer M., Cetinkaya
Corresponding Author:
E., Aytac R., Satiroglu H.: “Microdose gonadotropin-releasing hormone P. TELLI CELTEMEN, M.D.
agonist flare-up protocol versus multiple dose gonadotropin-releasing- Gazi Universitesi
hormone antagonist protocol in poor responders undergoing intracyto- Kadın Hastalıkları ve Dogum ABD.
plasmic sperm injection-embryo transfer cycle”. Fertil. Steril., 2009, 06100 Yenimahalle/Ankara (Turkey)
91, 2437. e-mail: pinartelli@gmail.com
[23] Liu X.Q., Wang W.F., Tan D.X.: “Clinical outcomes of gonadotropin-
releasing hormone antagonist used in poor ovarian responders”. Wei
Chuang Yi Xue, 2009, 4, 657.
CEOG Clinical and Experimental
Obstetrics & Gynecology

Effect of aerobic exercises versus foot reflexology

on post-menopausal depression

A.M. Eman1, M.E. Mahmoud2

1
Department of Women Health; 2Department of Basic Science, Faculty of Physical therapy, Kafr el shiekh University, Cairo (Egypt)

Summary
Purpose: Aim of this study was to compare between aerobic exercise and foot reflexology effects on treatment of post-menopausal
depression. Materials and Methods: Forty post-menopausal women complaining of mild to moderate depression selected randomly
from outpatient clinic of neuropsychiatry, kafr Elsheikh hospital between March, 2015 and November 2015 were included in the study.
Their age ranged from 45 to 55 years and their BMI from 25 to 35 kg/m². They had no musculoskeletal or cardiovascular disorders, free
diabetes, hypertension, and no history of neurological disorders. They were divided randomly into two groups equal in number: group
A treated by aerobic exercise 40 minutes, three times/ week for four weeks while group B was treated with foot reflexology for four
weeks. Depression was evaluated by the Beck Depression Inventory (BDI) before and after the program for both groups. Results: The
obtained results showed a statistically significant decrease (p < 0.05) in depression in both groups; when both groups were compared,
a statistically expressive writing moderated the relation between intrusive thoughts. Depressive symptoms significantly decreased in in
group A compared to group B (p < 0.05). Conclusion: It can be concluded that aerobic exercise and foot reflexology are effective ad-
jacent methods in reducing post-menopausal depression, but aerobic exercise is more effective than foot reflexology.

Key words: Postmenopausal depression; Aerobic exercise; Foot reflexology.

Introduction
Menopause is the end of reproductive period in female life,
and it is manifested by the permanent end of menstruation
lasting at least 12 months. Menopause is not a disease, it is a
normal stage in female life. Even so, the physical and psy-
chological effect of menopause can affect woman’s life, sap
her energy, and disturb the female psychological state [1].
Menopausal symptoms usually begin before the one-year
of last menses. They include irregular menstruation, infertil-
ity, dyspareunia due to dryness of vagina , hot flashes, sleep
disturbance, disturbed mood, trunk obesity, thinning hair, and
breast atrophy [2]. Menopause may increase feelings of sad-
ness and episodes of depression in some women.
About 8% and 15% of menopausal women have some form
of depression [3]. Menopausal depression may be due to ab-
normal change in levels of hormones in the body; during
menopause, the estrogen, progesterone, and androgen levels
are constantly fluctuating. These hormones drop, especially
estrogen, which may be the cause of sadness in females. Fe-
males with a severe drop in mood, results in depression [4].
Depression is a mental disorder that is characterized by low
mood, accompanied by low self-esteem, and loss of interest
or pleasure in normally enjoyable activities. It is also called
major depressive disorder, clinical depression, and major de-
pression. Depression has adverse effects on the woman’s fam-
ily relationships, work, sleeping, appetite, and general
condition [5].
The side-effects of noradrenergic and specific serotonergic
antidepressant are drowsiness, increased appetite, and weight
gain [6]. Between 30% and 50% of individuals treated with
antidepressant do not show improvement [7, 8]. Aerobic ex-
ercise has been prescribed for the treatment of a variety of
medical disorders as hyperlipidemia, osteoarthritis, fi-
bromyalgia, and diabetes. In addition, exercise improves psy-
chological state [9, 10]; it has also an improvement in a
variety of psychiatric conditions, especially in depression
[11]. The depressed woman with regular exercise may receive
positive feedback from the others and an increased sense of
self-worth. Exercise may act as a diversion from negative
thoughts [12, 13].
Social contacts may be an important mechanism, and phys-
ical activity may have physiological effects such as changes
in endorphins and monoamine concentrations [14, 15]. The
aim of treatment by reflexology is to promote harmony of
mind, body, and soul [16]. It is considered a safe, noninva-
sive, and inexpensive form of healthcare, used by the major-
ity of the population especially in children, very elderly,
terminally ill patients, and pregnant women [17].

Revised manuscript accepted for publication February 1, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3577.2017
 A.M. Eman, M.E. Mahmoud 727

Figure 1. — Appropriate
foot reflexology points for
patients in group B.

Materials and Methods point in exterior edge of the planter aspect left foot and liver re-
flex point opposite of the spleen at right foot; adrenal gland re-
Forty post-menopausal women that complaining of depression
flex point (5) is located between reflex points (6) and (4). Kidney
and selected randomly from outpatient clinic of neuropsychiatry,
reflex point (6) is almost between width of planter aspect of four
kafr Elsheikh hospital, were diagnosed by physician with mild and
toes, under base of both feet, and in the center of the foot. In ad-
moderate depression. Their ages ranged from 45 to 55 years and
dition to breast reflex point from ankle joint, width-wise line
their BMI was from 25 to 35 kg/m². They had no musculoskeletal
(wrinkle) to the junction of the toes. These cones are on the basis
or cardiovascular disorders, diabetes, hypertension, and no history
of sole division in the center of diaphragm line, and autonomic
of neurological disorders. All subjects assigned an informal con-
nervous system reflex point (7) located at soles of both feet be-
sent form before beginning the study. Cairo University ethical
tween heels and bases of the toes [19] (Figure 1). Reflexology
committee approved the study (P.T.REC/01200112). The patients
sessions were provided for 30 minutes (each foot 15 minutes)
were divided randomly into two groups equal in number. Assess-
twice a week for four weeks.
ment of all subjects in both groups (A and B) was carried out be-
All statistical measures were performed using the Statistical
fore and after the treatment program via the Beck Depression
Package for Social Science (SPSS) program version 18. The cur-
Inventory (BDI-II) questionnaire [18]. It is a 21-question multiple-
rent test involved two independent variables. The first was the
choice self-report inventory to assess depression state. Each an-
tested group which had two levels (group A and group B). The
swer is scored on a scale value from 0 to 3. Higher total scores
second was the training periods which had two levels (pre and
indicate more severe depressive symptoms. The standardized cut-
post). The one dependent ordinal variable was BDI. Accordingly
offs used differ from the original: 0–13: minimal depression,14–
non-parametric tests “Wilcoxon Signed Rank tests” were used to
19: mild depression, 20–28: moderate depression, and 29–63:
compare between pre- and post-tests for BDI questionnaire for
severe depression. Group A consisted of 20 post-menopausal
each group and “Mann-Whitney tests” were conducted to com-
women treated by aerobic exercise, in form of three sessions a
pare BDI questionnaire between both groups in the pre- and post-
week. Each session included 40-minute walking, (divided into ten
tests with an alpha level of 0.05.
minute warm-up, 20 minutes of exercise, and ten minutes of cool-
ing down). Exercise was set at 60-70% HR max as HR max = 220
– age.
Group B consisted of 20 post-menopausal women treated by Results
foot reflexology. In this group, all subjects were instructed briefly
and clearly about the nature of treatment and its value in order to There was no statistical significant difference in the mean
gain their confidence and co-operation throughout the study pe- values of age (57.07 ± 4.94), weight (81.33 ± 8.63), height
riod. Each patient was advised to wear light and comfortable (160.07 ± 3.54), and BMI (31.67 ± 3.09) between group A,
clothing and assume a relaxed supine laying position in a quiet and group B (56.47 ± 3.78, 82.87 ± 6.27, 160.47 ± 2.03,
room. First, whole of the sole was washed with warm water then
and 32.14 ± 2.51, with t-test = 0.374, –0.557, –0.380, and
massage protocol involved a combination of five minutes of light
stroking and light pressure using the whole hand to plantar and –0.464; p-value = 0.600, 0.582, 0.707, and 0.647, respec-
dorsal surfaces for each foot. Reflexology intervention was ap- tively (Table 1).
plied by using a combination of finger pivot and thumb walking Table 2 presents the comparison of the median scores of
techniques to the base of the foot and the toes that correspond to BDI questionnaire in the pre- and post-tests that were 20
the reflex points. The pressure was exerted on related and speci- and 10, respectively in the group A. Statistical analysis
fied zones with special concentration on the following points: gen-
ital zone which included ovarian reflex point (1) and uterine reflex using the non-parametric Wilcoxon Signed Rank tests re-
point (2) which are located at both feet under the lateral and me- vealed that there was a significant decrease in the BDI
dial malleolus, respectively; reflex point (3) represents pituitary questionnaire in the post-test in group A (Z = –4.379, p =
gland, exactly in the planter aspect of the center of hallux (big 0.000). Meanwhile, the median score of BDI questionnaire
toe) of both feet; solar plexus point (4) represents spleen reflex
728 Effect of aerobic exercises versus foot reflexology on post-menopausal depression

Table 1. — Demographic features of the two studied treatment of clinical depression. Additionally, exercise has
groups. a moderate reducing effect on state and trait anxiety and
Group A (n= 20) Group B (n= 20) t-value p-value can improve physical self-perceptions, and in some cases
Age (years) 57.07 ± 4.94 56.47 ± 3.78 0.374 0.600 (NS) global self-esteem. Also there is now good evidence that
Weight (kg) 81.33 ± 8.63 82.87 ± 6.27 –0.557 0.582 (NS) aerobic and resistance exercise enhances mood states, and
Height (cm) 160.07 ± 3.54 160.47 ± 2.03 –0.380 0.707 (NS) weaker evidence that exercise can improve cognitive func-
BMI (kg/m2) 31.67 ± 3.09 32.14 ± 2.51 –0.464 0.647 (NS) tion (primarily assessed by reaction time) in older adults
[23]. The results of the present study agree with Conn who
stated that exercise is an effective non-pharmacological
Table 2. — Median score, U, Z, and P values of the Beck therapy to reduce depressive symptoms among those liv-
Depression Inventory questionnaire (BDI) pre- and post- ing with depression, with a moderate standardized mean re-
tests in both groups. duction when compared to those who do not exercise [24].
BDI Median score Z-value p-value
Pre Post Elshamy et al. concluded that aerobic exercise with anti-
Group A 20 10 –4.379 0.000* depressive drug produced substantial improvement in mood
Group B 21 14 –4.293 0.000* of post- menopausal women with major depressive disor-
U-value 280 62 ders than antidepressants alone [25] .
Z-value –0.633 –4.912 Castren also agreed with the present results as he found
p-value 0.527 0.000 that physical activity and exercise help depressed patients
and promoted quicker and better relief from depression
[26]. Aerobic exercises help antidepressants and psy-
chotherapy work better and many find that walking, for ex-
in the pre- and post-tests were 21 and 14, respectively, in ample, is of great help; the reasons for improvement in this
group B. Statistical analysis using the non-parametric study may be related to the fact that exercise produces
Wilcoxon Signed Rank tests revealed that there was a sig- higher levels of chemicals in the brain, such as dopamine,
nificant decrease in the BDI questionnaire in the post-test serotonin, and norepinephrine. In general this leads to im-
in group B (Z = –4.293, p = 0.000). provements in mood and sleep disturbance, which is effec-
Considering the effect of the tested group (first inde- tive in countering depression [27].
pendent variable) on BDI questionnaire, Mann-Whitney Reflexology has been used as an alternative or comple-
tests revealed that the median score of the pre-test between mentary therapy to relieve stress and tension, improve the
both groups revealed that there was no significant differ- blood supply, and promote homeostasis [28]. Another
ence between both groups (U = 280, Z = –0.633, and p = possible theory to be taken into account in the effect of
0.929), while the median score of the post-test between reflexology is a specific form of foot massage in which it
both groups showed a significant reduction in BDI ques- is believed that areas in the feet and hands correspond to
tionnaire in favor of group A (U = 0.62, Z = –4.912, and p the glands, organs, and other parts of the body [29]. Also,
= 0.000*) (Table 2). reflexology produces a relaxing effect by relieving ten-
sion and stress related to physical problems. This relax-
ation affects the autonomic response, which, in turn,
Discussion
affects the endocrine, immune, and neuropeptide systems
Menopause always occurs during women’s midlife, dur- [30]. The finding of the present study agree with Nancy et
ing their late 40s or early 50s, and signals the end of the al. who stated that reflexology can be used to decrease
fertile phase. The functional disorders often significantly anxiety and pain in patients with cancer [31]. These re-
speed up the menopausal process and create more signifi- sults were supported by Rapaport et al. who also stated
cant health problems, both physical and emotional, for the that mood disorders had been reduced by relaxation train-
affected woman [20]. Common menopausal symptoms in- ing in patients suffering from depressive disorders [32].
clude menstrual irregularities, hot flash, night sweats, mood The results also agree with Ahn 2006 who stated that foot
swing, headache, insomnia, vaginal dryness, urinary prob- reflexology is effective in relieving of pain and depres-
lems, weight gain, memory and cognitive change, and fa- sion [33]. Finally, the psychological explanation states that
tigue. The dangerous symptoms are heavy bleeding, heart reflexology is simply a method of showing care and con-
palpitation, depression, and high blood pressure [21]. cern for patients. It has a positive effect on well-being and
The interaction of physical fitness and mental well-being quality of life, stress, anxiety, and pain [34].
has been increasingly recognized in psychiatry. In the
meanwhile, solid evidence has emerged that regular exer-
cise is associated with therapeutic effects in depressive pa-
tient and other psychiatric disorders [22]. Sufficient
evidence now exists for the effectiveness of exercise in the
 A.M. Eman, M.E. Mahmoud
References
[1] Chattha R.: “Treating the climacteric symptoms in Indian women
with an integrated approach to yoga therapy: a randomized control
study”. Menopause, 2008, 15, 862.
[2] Col N.F.: “In the clinic: menopause”. Ann. Intern. Med., 2009, 150,
ITC4.
[3] Soares CN.: “Perimenopause-related mood disturbance: an update
on risk factors and novel treatment strategies available”. In: Meeting
Program and Abstracts. Psychopharmacology and Reproductive
Transitions Symposium. American Psychiatric Association 157th An-
nual Meeting. May 1-6, 2004. New York, NY. Arlington, VA: Am.
Psychiatric Publish., 2004, 51.
[4] Bromberger J.T., Schott L.L., Kravitz H.M., Sowers M., Avis N.E.,
Gold E.B., et al.: “Longitudinal change in reproductive hormones
and depressive symptoms across the menopausal transition: results
from the Study of Women’s Health Across the Nation (SWAN)”.
Arch. Gen. Psychiatry, 2010, 67, 598.
[5] Barlow D.H.: “Abnormal psychology: an integrative approach”. 5th
ed. Belmont, CA: Thomson Wadsworth, 2005, 248.
[6] Stimmel G.L., Dopheide J.A., Stahl S.M.: “Mirtazapine; An antide-
pressant with noradrenergic and specific serotonergic effects”. Phar-
macotherapy, 1997, 17, 10.
[7] Baghai T., Moller H., Rupprecht R.: “Recent progress in pharmaco-
logical and non-pharmacological treatment options of major depres-
sion”. Curr. Pharmaceutical Design, 2006, 12, 503.
[8] Ruhé H.G., Huyser J., Swinkels J.A., Schene A.H.: “Switching an-
tidepressants after a first selective serotonin reuptake inhibitor in
Major Depressive Disorder”. J. Clin. Psychiatry, 2006, 67, 1836.
[9] Saulicz M., Saulicz E., Myśliwiec A., Wolny T., Linek P., Knapik
A.: “Effect of a 4-week Nordic walking training on the physical fit-
ness and self-assessment of the quality of health of women of the
perimenopausal age”. Prz. Menopauzalny, 2015, 14, 105.
[10] Ghorbani F., Heidarimoghadam R., Karami M., Fathi K., Minasian
V., Bahram M.E.: “The effect of six-week aerobic training program
on cardiovascular fitness, body composition and mental health
among female students”. J. Res. Health Sci., 2014, 14, 264.
[11] Carneiro L.S., Fonseca A.M., Vieira-Coelho M.A., Mota M.P., Vas-
concelos-Raposo J.: “Effects of structured exercise and pharma-
cotherapy vs. pharmacotherapy for adults with depressive symptoms:
a randomized clinical trial”. J. Psychiatr. Res., 2015, 71, 48-55.
[12] Lepore S.J.: “Expressive writing moderates the relation between in-
trusive thoughts and depressive symptoms”. J. Pers. Soc. Psychol.,
1997, 73, 10307.
[13] Mynors Wallis L.M., Gath D.H., Baker F.: “Randomised controlled
trial of problem solving treatment, antidepressant medication, and
combined treatment for major depression in primary care”. BMJ,
2000, 320, 2630.
[14] Thornen P., Floras J.S., Hoffman P., Seals D.R.: “Endorphins and
exercise: physiological mechanisms and clinical implications”. Med.
Sci. Sports Exerc., 1990, 22, 41728.
[15] Leith L.M.: “Foundations of exercise and mental health”. Morgan-
town, WV: Fitness Information Technology, 1990, 1744.
[16] Holt J., Acharya J.: “The effectiveness of foot reflexology in induc-
ing ovulation: a sham-controlled randomized trial”. Fertil. Steril.,
2009, 91, 2514.
[17] Wyatt A., Sikorskii M., Rahbar D. Victorson A., You M.: “Health-re-
lated quality-of-life outcomes: a reflexology trial withpatients with
advanced stage breast cancer”. Oncol. Nurs. Forum, 2012, 39, 568.
729
[18] McPherson A., Martin C.R.: “A narrative review of the Beck De-
pression Inventory (BDI) and implications for its use in alcohol de-
pendant population”. J. Psychiatr. Ment. Health Nurs., 2010, 17, 19.
[19] Fard S., Dolatian M., Heshmat, R., Majd H. “The effect of foot re-
flexology on premenstrual syndrome”. Archives Des Sciences, 2012,
65, 1661.
[20] Walker M.L., Herndon J.G.: “Menopause in nonhuman primates?”
Biol. Reprod., 2008, 79, 398.
[21] Gass MlS.: “The 5-minute obstetrics & gynecology consult”.
Philadelphia: Lippincott Williams&Wilkins, 2008, 292.
[22] Broocks A., Aherndt U., Sommer M.: “Physical training in the treat-
ment of depressive disorders”. Psychiatr. Prax., 2007, 34, 300.
[23] Fox K.R.: “The influence of physical activity on mental well-being”.
Public Health Nutr., 1999, 2, 411.
[24] Conn V.S.: “Depressive symptom outcomes of physical activity in-
terventions: meta-analysis findings”. Ann. Behav. Med., 2010, 39,
128.
[25] El-Shamy F., El-kosery S., El Bigawy A.: “Effect of aerobic exer-
cises on postmenpousal women with major depression”. Br. J. Appl.
Sci. Technol., 2013, 3, 214.
[26] Castren E.: “Is mood chemistry?” Nat. Rev. Neurosci., 2005, 6, 241.
[27] Blumenthal J.A., Babyak M.A., Moore K.A., Craighead W.E., Her-
man S., Khatri P., et al.: “Effects of exercises training on patients
with major depression”. Arch. Intern. Med., 1999, 159, 2349.
[28] Jones J., Thomson P., Irvine K., Leslie S.J.: “Is there a specific he-
modynamic effect in reflexology? A systematic review of random-
ized controlled trials”. J. Altern. Complement. Med., 2013, 19, 319.
[29] Micozzi M.S.: “Fundamentals of complementary and alternative
medicine”. 1st ed. New York: Churchill Livingston, 1996.
[30] White A.R., Williamson J., Hart A., Ernst E.: “A blinded investiga-
tion into the accuracy of reflexology charts”. Complement. Ther.
Med., 2000, 8, 166.
[31] Nancy L.N. Stephenson S.P. Weinrich R.N., Tavakoli A.S.: “The ef-
fects of foot reflexology on anxiety and pain in patients with breast
and lung cancer”. Oncol. Nurs. Forum, 2000, 27, 1
[32] Rapaport A., Mark H., Clary V., Cathryn C. ,Fayyad F., Rana W., En-
dicott G., Jean R.: “Quality-of-Life Impairment in Depressive and
Anxiety Disorders”. Clin. Obstet. Gynecol., 2005, 162, 1171.
[33] Ahn S.A.: “The effects of foot reflexology on pain and depression of
middle-aged women with osteoarthritis”. Korean J. Rehabil. Nurs.,
2006, 9, 25.
[34] Dobbs B.Z.: “Alternative health approaches”. Nursing Mirror, 1985,
160, 41.
[35] Green V.L., Alexandropoulou A., Walker M.B., Walker A.A., Sharp
D.M., Walker L.G., Greenman J.: “Alterations in the Th1/Th2 balance
in breast cancer patients using reflexology and scalp massage”. Exp.
Ther. Med., 2010, 1, 97.
Corresponding Author:
M.E. EMAN, M.D.
Department of Women Health
Faculty of Physical Therapy
Kafr el shiekh University
Cairo (Egypt)
e-mail: Emanabdelfatah123@yahoo.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Significance of growth differentiation factor 15

in primary ovarian insufficiency: inflammatory,
biochemical, and hormonal correlates

S.Y. Tunc1, N.Y. Goruk2, E. Agacayak1, M.S. Icen1, F.M. Findik1, H. Kusen3, M.S. Evsen1,
H. Yuksel4, T. Gul1
Department of Obstetrics & Gynecology, Dicle University School of Medicine, Diyarbakir
1
2 Department of Obstetrics & Gynecology, Memorial Hospital, Diyarbakir
3 Department of Obstetrics & Gynecology, Sirnak State Hospital, Sirnak
4 Department of Medical Biochemistry, Dicle University School of Medicine, Diyarbakir (Turkey)

Summary
Purpose: To investigate the levels of growth differentiation factor-15 (GDF-15) in primary ovarian insufficiency (POI) and to eval-
uate its correlation with hormonal, biochemical, and inflammatory indicators. Materials and Methods: This comparative, cross-sec-
tional study was carried out in 60 cases consisting of 30 healthy controls (mean age: 29.2 ± 5.0 years) and 30 patients with POI (mean
age: 28.9 ± 6.8 years). Two groups were compared in terms of serum levels of glucose, lipids, thyroid stimulating hormone (TSH), fol-
licle stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), prolactin (PRL), GDF-15, and neutrophil-lymphocyte ra-
tios (NLR). Correlation between GDF-15 and NLR with these variables was sought. Results: Serum levels of FSH (p < 0.001), LH (p
< 0.001), NLR (p < 0.001) and TSH (p = 0.020) were increased significantly in POI group. In POI patients, a correlation was detected
between levels of GDF-15 levels and PRL (p = 0.049). Conclusion: The authors suggest that NLR can serve as a promising marker for
diagnosis and follow-up of POI, whereas GDF-15 seems not to have such a potential.

Key words: Primary ovarian insufficiency; Growth differentiation factor-15; Neutrophil lymphocyte ratio; Inflammation.

Introduction younger than 30 years and 1:100 in women before 40 years

Primary ovarian insufficiency (POI) is characterized with
the absence, non-functionality or early depletion of the
ovarian reserve that may in turn result in infertility. Its rel-
evance has been increased recently attributed to the delayed
age of motherhood in developed countries [1]. It is typi-
cally accompanied with primary or secondary amenorrhea
for at least four months in women younger than 40 years of
age with menopausal serum FSH levels > 40 IU/L obtained
on two intervals at least one month apart and estradiol (E2)
levels < 50 pg/ml [2]. Even though serum levels of anti-
Müllerian hormone can serve as an indicator of POI since
it reflects of the state of follicular senescence, there are no
single screening tests that can predict a woman’s repro-
ductive lifespan at the moment [3, 4]. Due to the contin-
uum of impairment of ovarian function, with no specific
endpoint for this process, use of the term POI seems to be
more appropriate than the term premature ovarian failure
[5]. Actually, hypergonadotropic hypogonadism, premature
ovarian failure, and ovarian dysgenesis can be included
within the context of POI.
The incidence of POI for women younger than 20 years
of age is 1:10,000 whereas it rises to 1:1,000 in women
[6]. Therefore, POI is claimed to be one of the leading
causes of female infertility [1]. POI presents with a variety
of symptoms due to low levels of estroid hormones and sec-
ondary to diminution of ovarian function. Thus, symptoms
including hot flashes, sleep disturbances, decreased mental
concentration, loss of sexual desire, and vaginal dryness
can be seen. Moreover, long-term consequences of hypoe-
strogenism including higher risk for osteoporosis or car-
diovascular disease can occur [7]. In addition to hormone
replacement therapy, professional and family support is
necessary to eliminate the adverse effects of POI on emo-
tional health, stress, social life, and profession. Etiology of
POI may be linked with iatrogenic reasons, environmental
factors, viral infections, metabolic factors, autoimmune dis-
eases, and genetic alterations [8]. Mostly, the origin is id-
iopathic and it occurs without warning symptoms in many
cases [9].
The role of autoimmunity in POI and the presence of au-
toimmune oophoritis has been shown in POI patients with
adrenal autoimmunity. Such autoimmunity may occur due
to antigens common to both organs composed of steroido-
genic cells. The mechanism that causes and induces ovar-

Revised manuscript accepted for publication November 30, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3478.2017
 S.Y. Tunc, N.Y. Goruk, E. Agacayak, M.S. Icen, F.M. Findik, H. Kusen, M.S. Evsen, H. Yuksel, T. Gul 731

ian autoimmunity and inflammation is still obscure. Nev- Table 1. — Comparative overview of demographic, bio-
ertheless, POI may be associated with ovarian tissue dam- chemical, hormonal, and inflammatory markers in primary
age attributed to viral infection or other injury, which may ovarian insufficiency and control groups.
in turn induce the ovary to become antigenic [10]. Variable Control group POI group p-value
The systemic inflammatory response can be evaluated Age (years) 29.2±5.0 28.9±6.8 0.864
with the neutrophil to lymphocyte ratio (NLR) which has BMI (kg/m2) 24.1±4.2 25.9±4.1 0.094
been suggested as an inexpensive and widely available Marital status (M/S) 26/4 22/8 0.197
marker [11]. Yildirim et al. suggested that NLR, which is an Smoking habit (Y/N) 10/20 11/19 0.787
Gravidity 1.0-2.3 1.0-2.0 0.412
inexpensive and readily available marker, could have a
Glucose (mg/dl) 89.8±18.0 96.3±17.5 0.165
promising role as a marker for POI [12]. Total cholesterol (mg/dl) 170.6±46.9 184.5±25.4 0.159
Growth differentiation factor 15 (GDF-15), is a member Triglycerides (mg/dl) 110.9±37.1 124.4±25.6 0.106
of the human transforming growth factor-α superfamily. FSH (IU/L) 5.98±1.95 60.09±20.01 <0.001*
The placenta is the only tissue that expresses a substantial LH (IU/L) 6.35±3.72 38.07±21.21 <0.001*
amount of GDF-15 under physiological circumstances and E2 (pg/ml) 52.69±39.98 33.75±39.70 0.071
GDF-15 presumably plays a role at the maternal-fetal in- PRL (ng/ml) 14.15-7.49 14.00-9.45 0.594
terface. It is supposed to promote fetal survival via sup- TSH (mIU/L) 1.60±0.83 2.41±1.63 0.020*
pression of the production of proinflammatory cytokines GDF-15 (ng/L) 371.19±333.59 531.17±393.14 0.142
NLR 1.84-1.65 7.60-8.65 <0.001*
within the uterus. Notably, GDF-15 displayed immuno-
suppressive effects via inhibition of the proliferation of pe- POI: primary ovarian insufficiency; BMI:body-mass index; M: married;
S: single;Y: yes; N: no; NLR: neutrophil-lymphocyte ratio;
ripheral blood mononuclear cells [13]. FSH: follicle stimulating hormone; LH: luteinizing hormone; E2: estradiol;
As far as we know, any association between POI and lev- PRL: prolactin; TSH: thyroid stimulating hormone; GDF-15: growth differentia-
els of GDF-15 has not been investigated in the medical lit- tion factor-15; *statistically significant.
erature yet. The aim of the current study was to evaluate
levels of GDF-15 in POI and to investigate the hormonal,
inflammatory and biochemical correlates.
Biochemical parameters were analyzed by using an autoana-
lyzer. Plasma glucose levels were measured using the glucose
Materials and Methods oxidase method. Prolactin (4–15.2 ng/ml), FSH (1–8 IU/L),
Study design luteinizing hormone (LH) (1–12 IU/L), TSH (0.3–4 mIU/ml),
This cross-sectional, comparative trial was implemented in the and E2 (7.63–42.6 pg/ml) levels were analyzed by an im-
Obstetrics and Gynecology Department of the present institution. munoassay analyzer.
Before the study, approval of local Institutional Review Board and
written informed consent of all participants were obtained. Growth differentiation factor 15 immunoassay
Thirty healthy controls and 30 women diagnosed with POI In accordance with the method described by Kempf et al.,
were recruited. Participants in these two groups were matched GDF-15 level in plasma was measured by an immunoradiomet-
for age, gravidity, and body-mass index (BMI). Women with pri- ric sandwich assay by using a polyclonal, affinity chromatogra-
mary or secondary amenorrhea before 40 years of age, a normal phy-purified goat anti-human GDF-15 IgG antibody [14]. All
karyotype of 46XX, and follicle stimulating hormone (FSH) lev- analyses were performed in duplicate and clinical data were
els > 40 IU/L in at least two consecutive measurements were se- blinded to the laboratory. The detection limit of the assay was 20
lected. Patients with secondary causes of POI, such as surgery, ng/L, intra-assay imprecision was 10.6% or less, and inter-assay
chemotherapy or radiotherapy, and chromosomal abnormalities imprecision was 12.2% or more [14].
were excluded from the study. Women in the control group re-
ported to no pre-existing medical or obstetric conditions, such as Statistical analysis
chronic or acute inflammatory diseases, such as collagen vascu- Data was analysed by means of “SPSS Statistics 20” program.
lar diseases, infections, cardiovascular diseases, diabetes melli- Normal distribution of variables was assessed with Kol-
tus or renal diseases. Descriptive data including age, BMI, mogorov-Smirnov test and parametric tests were used for vari-
smoking habit, and marital status were recorded. ables with normal distribution, while non-parametric tests were
utilized for variables without normal distribution. Two depend-
Serum studies ent groups were compared with Independent-Samples t- and
Peripheral venous blood samples from drawn from antecubital Mann-Whitney U-tests. Correlation between variables with nor-
veins after bed rest in semirecumbent position for one hour sub- mal distribution was evaluated with Pearson Correlation test,
sequent to an overnight fasting period. All of the collected blood while Spearman’s Rho test was used for assessment of variables
samples were centrifuged at 4,000 rpm and +4°C for ten minutes that do not display normal distribution. Pearson Chi-square test
and they were transferred into Eppendorf tubes. After storage of was used for comparison of categorical variables. Quantitative
samples at room temperature for an hour, samples were kept at - variables were expressed as mean, standard deviation, median,
80°C in deep freeze until analysis was carried out. Complete and interquartile range. Confidence interval was 95% and a p-
blood count, serum glucose level, lipid profile (including total value < 0.05 was accepted as statistically significant.
cholesterol and triglycerides), and levels of thyroid stimulating
hormone (TSH) were studied. NLR was calculated for POI and
control groups.
732 Significance of growth differentiation factor 15 in primary ovarian insufficiency: inflammatory, biochemical, and hormonal correlates

Table 2. — Correlation of GDF-15 and NLR to demographic, biochemical, hormonal and inflammatory markers in pri-
mary ovarian insufficiency and control groups.
Variable Control group POI group
GDF-15 NLR GDF-15 NLR
r-value p-value r-value p-value r-value p-value r-value p-value
Age (years) -0.329 0.224 -0.176 0.353 0.160 0.398 -0.075 0.696
BMI (kg/m2) -0.133 0.483 0.088 0.644 0.103 0.590 0.141 0.498
Gravidity -0.080 0.673 0.124 0.515 0.281 0.133 -0.054 0.777
Glucose (mg/dl) -0.019 0.921 -0.083 0.665 -0.043 0.820 -0.142 0.453
Total cholesterol (mg/dl) -0.199 0.292 -0.273 0.145 0.188 0.319 0.215 0.253
Triglycerides (mg/dl) -0.290 0.120 -0.030 0.877 0.115 0.546 -0.019 0.921
FSH (IU/L) -0.048 0.800 -0.211 0.263 0.195 0.303 -0.108 0.570
LH (IU/L) -0.070 0.715 0.011 0.953 -0.101 0.595 0.031 0.871
E2 (pg/ml) -0.121 0.524 0.308 0.098 -0.058 0.760 0.084 0.658
PRL (ng/ml) -0.022 0.910 0.190 0.315 0.362 0.049* -0.160 0.397
TSH (mIU/L) 0.222 0.238 0.225 0.232 -0.113 0.553 0.154 0.417
POI: primary ovarian insufficiency; NLR: neutrophil-lymphocyte ratio; BMI: body-mass index; FSH: follicle stimulating hormone; LH: luteinizing hormone;
E2: estradiol; PRL: prolactin; TSH: thyroid stimulating hormone; GDF-15: growth differentiation factor-15; * statistically significant.

Results mune, metabolic dysfunction, infectious, and iatrogenic

Demographic and laboratory data derived from control and
POI groups are demonstrated in Table 1. As can be seen, no
significant differences were detected between two groups in
terms of age, gravidity, smoking habit, BMI, marital status, as
well as serum levels of glucose, cholesterol, triglycerides, pro-
lactin (PRL) and GDF-15. Notably, serum levels of FSH (p <
0.001), LH (p < 0.001), NLR (p < 0.001) and TSH (p = 0.020)
were increased significantly in POI group. However, serum
TSH levels were within normal levels in both groups.
As presented in Table 2, results of the correlation analy-
sis yielded that there was no correlation between any of the
variables under investigation and levels of GDF-15 and
NLR. In POI patients, a correlation was detected between
levels of GDF-15 levels and PRL (p = 0.049). However, no
correlation could be established between NLR and other
parameters. In POI group, patients with smoking habit had
remarkably higher levels of GDF-15 (p = 0.037).
In the control group, no difference was detected between
smokers and non-smokers in terms of NLR (p = 0.355).
Similarly, there was no difference between smokers and
non-smokers in POI group with respect to NLR (p = 0.683).
Discussion
The current study was implemented to assess NLR and serum
levels of GDF-15 in patients with POI and to evaluate the in-
flammatory, biochemical, and hormonal correlates. The present
results imply that GDF-15 levels were not altered in POI but
NLR can be a useful marker for diagnosis and follow-up.
POI has been associated with three potential mechanisms
including a congenital decrease in primordial follicles, ac-
celerated follicular atresia, and an inability to recruit pri-
mordial follicles. However, etiology underlying POI
remains unexplained for the vast majority of cases. Poten-
tial etiologies for POI can be divided into genetic, autoim-
categories. The most common autoimmune disorder linked
with POI is thyroiditis and a strong association was sug-
gested between POI and autoimmune polyendocrine syn-
drome [15]. Inflammatory basis for POI has been recently
investigated by Yildirim et al. and they suggested that NLR
may be a significant promising marker before presentation
or in the early stages of POI and may be useful for devel-
oping appropriate fertility treatment options [11]. The pres-
ent results are consistent with their data indicating that NLR
may have a potential as a marker for diagnosis and screen-
ing of POI. However, they found that NLR was lower in
POI patients compared to controls. Controversially, the
present authors found that NLR was higher in POI. The rea-
son for this difference may be either linked with genetic or
environmental conditions or may be attributed to the dis-
tinct types of inflammation that may be involved in POI.
Remembering that there are currently no standardized tests
for identification of POI, NLR may constitute a practical
and inexpensive alternative tool for diseases related to
chronic low-grade inflammation [16].
Biomarkers can aid in exploration of new targets for ther-
apy and may define risk groups for individualized therapy.
GDF-15 increases in cancer as well as in acute inflammation
and it is induced by the tumor suppressor gene p53. There-
fore, it may be a downstream target of pathways regulating
cell cycle arrest and apoptosis and thus important for pro-
liferation, invasion, metastases, and treatment resistance in
cancer [17]. Inflammation is recognized as a hallmark of
cancer and owing to the finding that NLR was claimed as a
discriminator between myomas and sarcomas [18].
To the best of the present authors’ knowledge, this is the
first report focussing on the levels of GDF-15 in POI. Al-
though the present authors could not demonstrate any associ-
ation between POI and GDF-15, the complex process of
inflammation and interaction with many variables hinder
 S.Y. Tunc, N.Y. Goruk, E. Agacayak, M.S. Icen, F.M. Findik, H. Kusen, M.S. Evsen, H. Yuksel, T. Gul
making straightforward conclusions. Further trials investi-
gating molecular and inflammatory basis of diseases should
be designed in a multi-centric fashion on larger series to
achieve more accurate outcomes.
The present authors did not observe any difference be-
tween controls and POI patients in terms of GDF-15, how-
ever, interestingly POI patients with smoking habit had
higher levels of GDF-15 compared to POI patients that do
not smoke. Therefore, GDF-15 may be involved in processes
linked with carcinogenesis rather than the type of inflam-
mation involved in pathogenesis of POI. Elucidation of the
precise association between GDF-15 and inflammation cas-
cade warrants further randomized, controlled trials on larger
series.
The present results remind that early diagnosis and treat-
ment of POI require further investigation and role of in-
flammatory process in pathogenesis of POI may extend
beyond ovarian autoimmunity. Understanding the molecu-
lar and inflammatory basis of POI is crucial for development
of appropriate anti-inflammatory treatment, which may pro-
vide preservation of ovarian function. Not only autoimmu-
nity, but also infectious or other types of ovarian injury may
result in insufficiency of immune regulation and give rise to
loss of tolerance to components of ovarian tissue [12].
As a recently described marker of systemic inflammation,
NLR is used for diagnosis and follow-up of malignancies in
gynecological practice [11, 16]. It has been demonstrated that
increased NLR was associated with greater pathology [19].
Although previous publications has shown that lymphocy-
tosis was linked with chronic inflammation and autoimmu-
nity [20], the present authors noted that NLR was increased
in POI representing a relative dominance of neutrophils over
lymphocytes. Miyake et al. have shown that CD8 T lym-
phocytes were diminished and total lymphocyte count was
increased in POI [21]. Although the present authors have not
analyzed lymphocyte subset counts, overt diminution of CD8
T lymphocytes may be one of the explanations of increased
NLR in POI. Furthermore, this finding may remind a possi-
ble role of infectious injury in development of POI.
Lack of analysis for lymphocyte subgroups, small sample
size, cross-sectional study design, and data derived form the
experience of a single institution comprise the main limita-
tions of the current study. Thus, associations and interpreta-
tions must be made with caution.
To conclude, results of the current study indicate that NLR
can serve as a promising marker for diagnosis and follow-
up of POI, whereas GDF-15 seems not to have such a po-
tential. Understanding the molecular and inflammatory basis
of POI is mandatory for development of more effective
modes of diagnosis and treatment.
References
[1] Fortuño C., Labarta E.: “Genetics of primary ovarian insufficiency: a
review”. J. Assist. Reprod. Genet., 2014, 31, 1573.
[2] Nelson L.M.: “Clinical practice. Primary ovarian insufficiency”. N.
733
Engl. J. Med., 2009, 360, 606.
[3] Visser J.A., Schipper I., Laven J.S., Themmen AP.: “Anti-Mullerian
hormone: an ovarian reserve marker in primary ovarian insufficiency”.
Nat. Rev. Endocrinol., 2012, 8, 331.
[4] De Vos M., Devroey P., Fauser B.C.: “Primary ovarian insufficiency”.
Lancet, 2010, 376, 911.
[5] Welt C.K.: “Primary ovarian insufficiency: a more accurate term for
premature ovarian failure.” Clin. Endocrinol. (Oxf.), 2008, 68, 499.
[6] Coulam C.B., Adamson S.C., Annegers J.F.: “Incidence of premature
ovarian failure”. Obstet. Gynecol., 1986, 67, 604.
[7] Aubuchon M., Santoro N.: “Lessons learned from the WHI: HRT re-
quires a cautious and individualized approach”. Geriatrics, 2004, 59,
22.
[8] van Kasteren Y.M., Hundscheid R.D., Smits A.P., Cremers F.P., van
Zonneveld P., Braat D.D.: “Familial idiopathic premature ovarian fail-
ure: an overrated and underestimated genetic disease?” Hum. Reprod.,
1999, 14, 2455.
[9] Beck-Peccoz P, Persani L.: “Premature ovarian failure”. Orphanet J
Rare Dis., 2006, 1, 9.
[10] Hoek A., Schoemaker J, Drexhage H.A.: “Premature ovarian failure
and ovarian autoimmunity”. Endocr. Rev., 1997, 18, 107.
[11] Ilhan M., Ilhan G., Gök A.F., Bademler S., Verit Atmaca F., Ertekin C.:
“Evaluation of neutrophil-lymphocyte ratio, platelet-lymphocyte ratio
and red blood cell distribution width-platelet ratio as early predictor of
acute pancreatitis in pregnancy”. J. Matern. Fetal Neonatal Med., 2015,
Jun 24, 1. [Epub ahead of print]
[12] Yildirim G., Tokmak A., Kokanal M.K., Sarkaya E., Zungun C., Inal
H.A., Yilmaz F.M., Ylmaz N.: “Association between some inflamma-
tory markers and primary ovarian insufficiency”. Menopause, 2015,
22, 1000.
[13] Corre J., Hébraud B., Bourin P.: “Concise review: growth differentia-
tion factor 15 in pathology: a clinical role?” Stem Cells Transl. Med.,
2013, 2, 946.
[14] Kempf T., Horn-Wichmann R., Brabant G., Peter T., Allhoff T., Klein
G., et al.: “Circulating concentrations of growth-differentiation factor
15 in apparently healthy elderly individuals and patients with chronic
heart failure as assessed by a new immunoradiometric sandwich assay”.
Clin. Chem., 2007, 2, 284.
[15] Cox L., Liu J.H.: “Primary ovarian insufficiency: an update”. Int. J.
Womens Health, 2014, 6, 235.
[16] Kirbas A., Biberoglu E., Daglar K., Iskender C., Erkaya S., Dede H.,
Uygur D., Danisman N.: “Neutrophil-to-lymphocyte ratio as a diag-
nostic marker of intrahepatic cholestasis of pregnancy”. Eur. J. Obstet.
Gynecol. Reprod. Biol., 2014, 180, 12.
[17] Trovik J., Salvesen H.B., Cuppens T., Amant F., Staff A.C.: “Growth
differentiation factor-15 as biomarker in uterine sarcomas”. Int. J. Gy-
necol. Cancer, 2014, 24, 252.
[18] Kim H.S., Han K.H., Chung H.H., Kim J.W., Park N.H., et al.: “Neu-
trophil to lymphocyte ratio for preoperative diagnosis of uterine sarco-
mas: a case-matched comparison”. Eur. J. Surg. Oncol., 2010, 36, 691.
[19] Keskin Kurt R., Okyay A.G., Hakverdi A.U., Gungoren A., Dolap-
cioglu K.S., Karateke A., Dogan M.O.: “The effect of obesity on in-
flammatory markers in patients with PCOS: a BMI-matched
casecontrol study”. Arch. Gynecol. Obstet., 2014, 290, 315.
[20] Shah A., Diehl L.F., St Clair E.W.: “T cell large granular lymphocyte
leukemia associated with rheumatoid arthritis and neutropenia”. Clin.
Immunol., 2009, 132, 145.
[21] Miyake T., Sato Y., Takeuchi S.: “Implications of circulating autoanti-
bodies and peripheral blood lymphocyte subsets for the genesis of pre-
mature ovarian failure”. J. Reprod. Immunol., 1987, 12, 163.
Corresponding Author:
S.Y. TUNC, M.D.
Department of Obstetrics & Gynecology
Dicle University School of Medicine
Silvan Street
21280 Sur/Diyarbakır (Turkey)
e-mail: drsenemtunc@hotmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Clinicopathological changes of perinatal mortality

during the last 20 years in a tertiary hospital of Greece

C. Goudeli1, L. Aravantinos2, D. Mpotsis2, G. Creatsas2, A. Kondi-Pafiti3

1 Department of Gynaecology, St. Savvas Anticancer-Oncologic Hospital, Athens

2nd Department of Obstetrics and Gynecology Aretaieion Hospital, University of Athens

2
3 Department of Pathology, Aretaieio Hospital, University of Athens Medical School, Athens (Greece)

Summary
Introduction: Perinatal period is the period that includes fetuses weighing > 500 grams (22nd week of gestation) and newborns aged up to
seven days. Perinatal mortality is one of the earliest quantitative measurements of quality in obstetric care and affects approximately 0.5%
to 1% of all pregnancies. Aim: The purpose of this study was the identification, classification, and frequency of causes of perinatal mortal-
ity in premature infants during 20 years (1992-2012) in a tertiary Maternity Hospital in Athens. Materials and Methods: This was a retro-
spective study based on Pathology Department record and contains autopsy findings of fetuses, newborns, and membranes of the period
1992-2012 in conjunction with clinical information. The authors excluded pharmaceutical miscarriage and those containing vague variables.
The total population birth to the mentioned years in this Hospital was 23,703 and there were 278 deaths. The authors used the classification
system of ReCoDe (2005) which best suited the present data. Changes in perinatal death cause were estimated and compared every five
years during this period (1993-1997, 1998-2002, 2003-2007, and 2008-2012) and also divided according to the following gestational ages:
22-27, 28-31, 32-36, and 37-43 weeks using the SPSS 19.0. Results: Perinatal mortality was reduced up to 72.3% during these years. The
vast majority of stillbirths were in their 22-27 week of gestation. Almost half of the fetal deaths were caused by fetal abnormalities, while in
78% the placenta had a main or secondary role. A detailed description of embryo-membranes and clinical status of the mother was per-
formed. Finally the authors identified 15 newborns who had reached the 28th day of their life, of which 12 (80%) were premature. The ma-
jority were females and the mean age of the mothers was 28 years. Seven out of 12 newborns died of fetal problems, while three out of 12
due to intrapartum pathology. Conclusion: Pathogenesis of perinatal mortality is often unclear and associated to multiple causes. Impressive
reduction of neonatal mortality has been realized during recent years due to the developments in obstetric and neonatal intensive care, but
still many improvements are needed to be done.

Key words: Perinatal mortality; Stillbirth; Gestational age; Fetal death in Greece; ReCoDe; Cause of fetal death.

Introduction
Perinatal mortality is one of the earliest quantitative
measurements of quality in obstetric care. Perinatal deaths
affect approximately 0.5% to 1% of all pregnancies [1].
Stillbirth or fetal death is death prior to the complete ex-
pulsion or extraction from the mother of a product of con-
ception, irrespective of the duration of pregnancy; the death
is indicated by the fact that after such separation the fetus
does not breathe or show any other evidence of life, such as
beating of the heart, pulsation of the umbilical cord or def-
inite movement of voluntary muscles [2].
Perinatal period commences at 22 completed weeks (154
days) of gestation (the time when birth weight is normally
500 grams and body length 25 cm crown-heel and ends
seven completed days after birth. More than 3.3 million still-
births and over three million early neonatal deaths are esti-
mated to take place every year. In the 2000, over 6.3 million
perinatal deaths occurred worldwide: almost all of them
(98%) occurred in developing countries and 27% in the least
developed countries. In developed countries, where inter-
ventions have largely eliminated excess early neonatal mor-
tality, over six out of ten perinatal deaths are stillbirths [3].
Assignment of a probable cause of death is important to de-
velop interventions for stillbirth prevention. There are cur-
rently at least 32 classification systems of stillbirth, many of
which have been developed for different purposes. They have
different categories for classifying causes, numerous defini-
tions for relevant conditions, and varying levels of complex-
ity. After reviewing the existing systems – Wigglesworth,
Aberdeen, NICE, TULIP, Nordic-Baltic etc - the authors se-
lected the ReCoDe 2005. The hierarchy started from condi-
tions affecting the fetus and moved outward in simple
anatomical groups, which were subdivided into pathophysi-
ological conditions. The analysis of secondary codes provided
further insight into the conditions leading to death [4].
The aim of this retrospective study was to assess any
changes in cause-specific fetal death rates in the population
of a specific tertiary care unit reflecting the level of perina-
tal care of this region. It was possible to conduct such a study
at this institution because all fetal deaths occurring in the last
two decades have been analyzed and the vast majority has
had complete postmortem examination. It is important to

Revised manuscript accepted for publication February 10, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3593.2017
 C. Goudeli, L. Aravantinos, D. Mpotsis, G. Creatsas, A. Kondi-Pafiti 735

note that no other study with this time duration, parameters, Table 1. — Fetal death characteristics.
and large population has ever been realized in Greece [5]. Fetal deaths (n=278)
n (%)
Gestational week
Materials and Methods 22nd – 27th 170 (61.2)
The purpose was to study changes in gestational-age-specific 28th – 31th 28 (10.1)
risk of fetal death among all pregnancies after 22 weeks of gesta- 32nd – 36th 53 (19.1)
tion in a central-university hospital of Athens from 1993 to 2012. 37th – 43rd 27 (9.7)
The study was a population-based registry study. The authors ex- Maternal age (years)
cluded stillbirths less than 22 weeks of gestation and therapeutic < 25 38 (17)
abortions-terminated pregnancies. Any additional births that 25-30 74 (33.2)
lacked information on potentially confounding variables were also 31-35 71 (31.8)
excluded. As a result, 23,520 births were included 36-40 27 (12.1)
Collected data included medical and obstetric history, maternal > 40 13 (5.8)
and fetal characteristics, and birth details. The authors utilized the Gender
ReCoDe classification system 2005, because after reviewing the lit- Males 134 (52.1)
erature, it was the system that provided the largest number of cases Females 123 (47.9)
classified according to the information of the present data. Autopsy Premature labor
(gross and microscopic examination of all organs) was performed No 27 (9.7)
in all cases. Yes 251 (90.3)
Perinatal mortality, early neonatal mortality, and stillbirth rate Birth weight (grams), mean (SD) 1029.3 (849.8)
were calculated according to the World Health Organization def- Weight percentile, mean (SD) 29.8 (26.4)
initions [2]. Gestational age was calculated from the last men-
Placenta
strual period; ultrasound dating measurements were given priority
Abnormal 185 (78.4)
when they were available. In few cases of 1990s during which the
Normal 51 (21.6)
gestational age was not provided, the authors used the crown-
rump length. It is, however, unlikely that change in estimation of Umbilical Cord
gestational length has significantly biased the present results be- Abnormal 77 (27.7)
cause gestational age was grouped and term pregnancies were de- Normal 201 (72.3)
fined as 37 weeks of gestation or above. Any overestimation of Multiple pregnancy
term pregnancies by using last menstrual period for prediction of No 118 (67.0)
term would rather underestimate than overestimate the reduction Yes 58 (33.0)
in fetal death rate at term in the later time periods of this study. Way of miscarriage
The authors used the age at the estimated date of fetal death rather Stillbirth 263 (94.6)
than age at delivery because delivery sometimes occurred many Newborn 15 (5.4)
days after death. To determine whether the fetus was small for
gestational age (SGA), the authors compared fetal weight with
the mean birth weight for the infants born at the same gestational
age.
The study population consisted of 278 stillbirths of occurring
23,520 births in a tertiary care unit during 1993-2012. Data were 35 years. Among the perinatal deaths, 134 were males
collected from the file of Pathology Department of the hospital. (52.1%). Most miscarriages occurred in premature labors
The population served was mainly white and from all socioeco- (< 37th gestational week) with the percentage being 90.3%.
nomic classes. Mean birth weight was 1029.3 (SD = 849.8) grams and
Quantitative variables were expressed as mean values (SD),
while qualitative variables were expressed as absolute and relative
mean weight percentile was 29.8 (SD = 26.4). In 78.4% of
frequencies. Rates of fetal death per 1,000 ongoing pregnancies the miscarriages, there was a problem in the placenta and in
were calculated in total sample and by year of delivery. Relative 27.7% a problem in the umbilical cord. Also, 33.0% of the
risks of fetal death and their 95% confidence intervals (CIs) were miscarriages occurred with multiple pregnancy and 94.6%
calculated for each time period, with 1993-1997 as the reference were stillbirths.
period. Statistical significance was set at p < 0.05 and analyses
The perinatal mortality decreased by 72.3%, from 21.3
were conducted using SPSS statistical software (version 19.0).
per 1,000 births in the years 1993-1997 to 6.3 per 1,000
births in the years 2008-2012 (Figures 1 and 2). The rela-
tive risk of fetal death in 1998-2002 was 0.72 (95% CI:
Results
0.53–0.97) and significant lower comparing births during
A total of 23,520 births in the present hospital from 1993 1993-1997 (Table 2). Also, the relative risk of fetal death in
to 2012 were included in the study. In the study period, 278 2003-2007 was 0.44 (95% CI: 0.31–0.61) and significant
fetal deaths (11.8‰) occurred. Characteristics of the fetal lower comparing births during 1993-1997, while for the pe-
deaths are presented in Table 1. The majority of fetal deaths riod 2008-2012, it was 0.29 (95% CI: 0.20–0.41) and sig-
occurred between 22nd and 27th week of gestation (61.2%). nificant lower comparing births during 1993-1997.
Also, 33.2% of the miscarriages occurred in mothers aged In the total sample, most common cause for miscarriage
from 25 to 30 years and 31.8% in mothers aged from 31 to was regarding problems of the fetus (reported in 44.6% of
736 Clinicopathological changes of perinatal mortality during the last 20 years in a tertiary hospital of Greece

Table 2. — The number of fetal deaths and relative risks gestation, where most common cause reported was as-
(RR) with 95% CI of fetal death according to year of de- phyxia (37.0%).
livery.
Year Fetal deaths Births Miscarri- RR (95% CI)*
N (%) N (%) ages (%) Discussion
1993-1997 86 (30.9) 3959 (17.0) 21.3 1.00‡ In the western world, the stillbirth rate has declined since
1998-2002 85 (30.6) 5453 (23.5) 15.3 0.72 (0.53–0.97) the 1950s [6]. The decline was most pronounced early in
2003-2007 59 (21.2) 6227 (26.8) 9.4 0.44 (0.31–0.61)
this period [7]; indeed increasing stillbirth rates have been
2008-2012 48 (17.3) 7603 (32.7) 6.3 0.29 (0.20–0.41)
Total 278 (100.0) 23242 (100.0) 11.8 - reported since 2001 [8]. While national and international
attention, statistics, and interventions focus on live infants,
*RR (95% confidence interval). ‡ indicates reference category.
stillborn infants have largely been overlooked. However,
these deaths do matter to the mother, family, society, and to
the healthcare system [3].
the cases), followed by problems in the placenta (19.4%) Although social factors exert the main influence on the
(Table 3). Similar were the results when causes were re- outcome of a birth, as societies advance, good medical care
ported according to year of delivery and by gestational age, tends to play a greater role. New technologies are not neces-
except for cases occurring between 37th and 43rd week of sarily beneficial, as sex-selection procedures and inappro-

Table 3. — Fetal deaths due to miscellaneous causes in total sample, and according to year of delivery and gestational week.
Total sample Year Gestational week
(n=278) 1993-1997 1998-2002 2003-2007 2008-2012 22nd–27nd 28th–31st 32nd–36th 37th–43rd
n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%)
Amniotic fluid 10 (3.6) 7 (8.1) 1 (1.2) 0 (0.0) 2 (4.2) 9 (5.3) 0 (0.0) 0 (0.0) 1 (3.7)
Chorioamnionitis 8 (2.9) 6 (7) 1 (1.2) 0 (0.0) 1 (2.1) 8 (4.7) 0 (0.0) 0 (0.0) 0 (0.0)
Oligohydramnios 1 (0.4) 0 (0.0) 0 (0.0) 0 (0.0) 1 (2.1) 1 (0.6) 0 (0.0) 0 (0.0) 0 (0.0)
Polyhydramnios 1 (0.4) 1 (1.2) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (3.7)
Fetus 124 (44.6) 33 (38.4) 33 (38.8) 37 (62.7) 21 (43.8) 77 (45.3) 12 (42.9) 32 (60.4) 3 (11.1)
Lethal congenital anomaly 47 (16.9) 13 (15.1) 13 (15.3) 14 (23.7) 7 (14.6) 32 (18.8) 2 (7.1) 11 (20.8) 2 (7.4)
Infection 25 (9) 6 (7) 6 (7.1) 9 (15.3) 4 (8.3) 22 (12.9) 1 (3.6) 2 (3.8) 0 (0.0)
Non-immune hydrops 3 (1.1) 1 (1.2) 1 (1.2) 1 (1.7) 0 (0.0) 2 (1.2) 0 (0.0) 1 (1.9) 0 (0.0)
Iso-immunisation 2 (0.7) 1 (1.2) 1 (1.2) 0 (0.0) 0 (0.0) 1 (0.6) 1 (3.6) 0 (0.0) 0 (0.0)
Fetomaternal haemorrhage 1 (0.4) 0 (0.0) 1 (1.2) 0 (0.0) 0 (0.0) 1 (0.6) 0 (0.0) 0 (0.0) 0 (0.0)
Twin-twin transfusion 3 (1.1) 0 (0.0) 0 (0.0) 1 (1.7) 2 (4.2) 2 (1.2) 1 (3.6) 0 (0.0) 0 (0.0)
Fetal growth restriction 43 (15.5) 12 (14) 11 (12.9) 12 (20.3) 8 (16.7) 17 (10) 7 (25) 18 (34) 1 (3.7)
Intrapartum 22 (7.9) 13 (15.1) 7 (8.2) 1 (1.7) 1 (2.1) 0 (0.0) 3 (10.7) 9 (17) 10 (37)
Asphyxia 22 (7.9) 13 (15.1) 7 (8.2) 1 (1.7) 1 (2.1) 0 (0.0) 3 (10.7) 9 (17) 10 (37)
Mother 5 (1.8) 0 (0.0) 4 (4.7) 0 (0.0) 1 (2.1) 4 (2.4) 0 (0.0) 1 (1.9) 0 (0.0)
Hypertensive diseases in pregnancy 3 (1.1) 0 (0.0) 3 (3.5) 0 (0.0) 0 (0.0) 3 (1.8) 0 (0.0) 0 (0.0) 0 (0.0)
Lupus/ antiphospholipid syndrome 1 (0.4) 0 (0.0) 1 (1.2) 0 (0.0) 0 (0.0) 1 (0.6) 0 (0.0) 0 (0.0) 0 (0.0)
Other 1 (0.4) 0 (0.0) 0 (0.0) 0 (0.0) 1 (2.1) 0 (0.0) 0 (0.0) 1 (1.9) 0 (0.0)
Placenta 54 (19.4) 9 (10.5) 19 (22.4) 11 (18.6) 15 (31.3) 35 (20.6) 9 (32.1) 7 (13.2) 3 (11.1)
Abruption 24 (8.6) 8 (9.3) 5 (5.9) 5 (8.5) 6 (12.5) 14 (8.2) 6 (21.4) 4 (7.5) 0 (0.0)
Praevia 1 (0.4) 0 (0.0) 0 (0.0) 1 (1.7) 0 (0.0) 1 (0.6) 0 (0.0) 0 (0.0) 0 (0.0)
Placental insufficiency/infarction 27 (9.7) 1 (1.2) 14 (16.5) 5 (8.5) 7 (14.6) 18 (10.6) 3 (10.7) 3 (5.7) 3 (11.1)
Other 2 (0.7) 0 (0.0) 0 (0.0) 0 (0.0) 2 (4.2) 2 (1.2) 0 (0.0) 0 (0.0) 0 (0.0)
Trauma 1 (0.4) 1 (1.2) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (3.7)
Iatrogenic 1 (0.4) 1 (1.2) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 1 (3.7)
Umbilical Cord 20 (7.2) 8 (9.3) 8 (9.4) 2 (3.4) 2 (4.2) 11 (6.5) 1 (3.6) 3 (5.7) 5 (18.5)
Prolapse 2 (0.7) 0 (0.0) 2 (2.4) 0 (0.0) 0 (0.0) 1 (0.6) 0 (0.0) 0 (0.0) 1 (3.7)
Constricting loop or knot 9 (3.2) 6 (7) 1 (1.2) 2 (3.4) 0 (0.0) 4 (2.4) 0 (0.0) 1 (1.9) 4 (14.8)
Velamentous insertion 3 (1.1) 2 (2.3) 0 (0.0) 0 (0.0) 1 (2.1) 1 (0.6) 1 (3.6) 1 (1.9) 0 (0.0)
Other 6 (2.2) 0 (0.0) 5 (5.9) 0 (0.0) 1 (2.1) 5 (3.0) 0 (0.0) 1 (1.9) 0 (0.0)
Uterus 1 (0.4) 0 (0.0) 0 (0.0) 0 (0.0) 1 (2.1) 1 (0.6) 0 (0.0) 0 (0.0) 0 (0.0)
Rupture 1 (0.4) 0 (0.0) 0 (0.0) 0 (0.0) 1 (2.1) 1 (0.6) 0 (0.0) 0 (0.0) 0 (0.0)
Unclassified 41 (14.7) 15 (17.4) 13 (15.3) 8 (13.6) 5 (10.4) 33 (19.4) 3 (10.7) 1 (1.9) 4 (14.8)
No information available 25 (9) 13 (15.1) 9 (10.6) 1 (1.7) 2 (4.2) 22 (12.9) 3 (10.7) 0 (0.0) 0 (0.0)
No relevant condition identified 16 (5.8) 2 (2.3) 4 (4.7) 7 (11.9) 3 (6.3) 11 (6.5) 0 (0.0) 1 (1.9) 4 (14.8)
 C. Goudeli, L. Aravantinos, D. Mpotsis, G. Creatsas, A. Kondi-Pafiti
Figure 3. — Female stillbirth (30 wog) with cat-eye syndrome-
chromosomal anomaly (inv.dup.22), congenital abnormalities, and
intraventricular communication combined with placental
hematoma.
priate assisted reproduction show. The way they contribute to
adverse pregnancy outcomes is not captured in current meth-
ods of collecting, analyzing or presenting perinatal data [2].
In the western world, perinatal mortality has declined
since the mid-19th century from 26-43 per 1,000 births to a
level of five to ten per 1,000 births in the first decade of the
21st century. Also, the fetal death rate has been reported to
decline. In Europe, the fetal death rate after 28 weeks of
737
Figure 1. — Perinatal mortality
during the last 20 years.
Figure 2. — Perinatal mortality
during the last 20 years.
gestation has declined from 25-45 to three to five per 1,000
births form 1940 to 2000 [3]. During our study period the
fetal death rate declined in pregnancies lasting longer than
22 weeks and the decline was more prominent in pregnan-
cies at term.
The most common cause of death was due to a fetal
problem (44.6%) and especially a lethal congenital anom-
aly, infection or fetal growth restriction (Figures 3 and 4).
According to European Surveillance of Congenital Anom-
alies (EUROCAT), the prevalence of chromosomal anom-
alies is 3.6 per 1,000 births, contributing 28% of stillbirths
and 48% of all terminations of pregnancy following pre-
natal diagnosis. Congenital heart defects are the most
common non-chromosomal anomalies, followed by limb
defects, anomalies of urinary system, and nervous system
defects [9]. In the present study, infections have been re-
ported to account for 9%, while 10-25% of fetal deaths
are attributed to congenital infections in developed coun-
tries [10]. On the other hand, fetal growth restriction is a
condition of fetal death related to many parameters and
reaches 15.5% of perinatal mortality of the present data.
While 70% of small fetuses are small for normal reasons
and not at risk, 30% are pathologically small at risk for nu-
merous complications. There are no randomized trials ad-
dressing the timing of delivery of IUGR fetus in the late
preterm or early-term period, taking under consideration
factors such as non-stress testing, fetal movement, interval
738 Clinicopathological changes of perinatal mortality during the last 20 years in a tertiary hospital of Greece
Figure 4. — Male stillbirth (27 wog). Non-immune hydrops with
no other pathology.
growth, amniotic fluid volume, etc [11, 12].
In the present study, 33% of the miscarriages occurred in
multiple pregnancy, 94.6% were automatic (stillbirth), and
the rest live births. The potential causes of fetal death in
multiple gestations are numerous and include virtually
every obstetric complication, including placental insuffi-
ciency, abruption, preeclampsia, and preterm labor. Other
problems are unique to multiple gestations, especially in
cases of monochorionic placentation, such as twin-twin
transfusion syndrome, cord enlargement, and twin-reverse
arterial perfusion [10].
Assisted reproductive technologies (ART) have a high rate
of multiple gestations (31% in USA) and yet there are con-
cerns about the risk for adverse pregnancy outcomes. In as-
sociation with maternal age, ART is accused of spontaneous
abortion, ectopic pregnancies, chromosomal abnormalities,
imprinting disorders, prematurity, birth defects, IUGR,
preeclampsia, and placental abruption. As a result of com-
mon use of in vitro fertility, -2.5% of French infants in 2006-
future initiatives are needed to characterize ART as the main
cause or the substrate of a fetal abnormality [13-15].
Many cases of fetal death, especially at term, are attrib-
uted to umbilical cord accidents. Thus the demonstration
of cord occlusion, fetal hypoxia, and the exclusion of other
causes is required to confirm the diagnosis. In the present
Figure 5. — Placenta of male stillbirth (29 wog) infected by CMV.
Maturation abnormalities and micro-calcifications can be noted.
population, umbilical abnormality was the main cause for
20% of perinatal deaths, the majority of which referred to
constricting loop or knot.
In 78.4% of the miscarriages, there was pathology in the
placenta and in 19.4%, it was the main cause of death.
Among the placental anomalies, abruption and placental in-
sufficiency or infractions were the most common. The pla-
centa can be considered the diary of pregnancy; after death,
it remains viable for several days. The value of examining the
placenta for determining or excluding a cause of death in
stillbirths is evident and varies from 28-85% (Figure 5).
Thus, placental causes of death have been found in up to 60%
of perinatal mortality cases and 64% of intrauterine fetal
deaths depending on the classification system [16].
As a result of membranes’ or fetus’ malfunction, the pres-
ent authors estimated amniotic fluid pathology which
reached 3.6% of the perinatal mortality. A variety of other
disorders such as uterine rupture (0.4%) and trauma (0.4%)
were more rare conditions of death.
Several maternal medical disorders are associated with an
increased risk for fetal death. It is debatable as to whether
these conditions are causal or risk factors, because most af-
fected women deliver live infants. It is estimated that ma-
ternal diseases play a role in 10% of perinatal mortality [17].
Hypertensive disease during pregnancy is the most common
cause of maternal disease which led to fetal predicament and
death (1.1%) in the present study.
Delivery-related perinatal death is defined as intra-
partum stillbirth or neonatal death that is unrelated to con-
genital abnormality. The causes of intrapartum stillbirth
indicate that most of these stillbirths are caused by events
that will occur only during labor and delivery, such as as-
phyxia in the present population (7.9%). Similarly, having
excluded deaths because of congenital abnormality, most
neonatal deaths are due to intrapartum events or the effects
 C. Goudeli, L. Aravantinos, D. Mpotsis, G. Creatsas, A. Kondi-Pafiti
of prematurity [18, 19].
Finally, in each classification of stillbirth population, there
is a percentage of unclassified/unexplained fetal deaths; in the
present case this accounted for 14.7% (61% of which was due
to no information provided). In the majority of studies, many
cases remain unexplained even after extensive evaluation.
Losses later in gestation (third trimester) are more likely to be
unexplained than losses earlier in gestation. Such losses are
strongly associated with IUGR as well as most of the previ-
ously described risk factors [20]. Hence, these deaths are
largely linked to maternal health conditions before pregnancy,
complications of pregnancy, such as preeclampsia, and pla-
cental dysfunction, without being able to establish a cause [21].
It is likely that the causes of fetal deaths differ accord-
ing to the length of gestation [22]. However, infections are
assumed to be linked to mid-pregnancy fetal deaths, [23]
whereas fetal deaths at or near term may be caused by pre-
eclampsia, placental abruption, fetal growth restriction or
complications during delivery [24]. Developed-country his-
torical data suggest that, as smaller and sicker babies sur-
vive, an increasing number of small babies are registered.
In the present population, the majority of the fetal deaths
occurred between 22nd and 27th week of gestation (61.2%),
while the intrapartum asphyxia was noticed from 32nd to
43rd weeks of gestation (86%). Also, 33.2% of the miscar-
riages occurred in mothers aged from 25 to 30 years and
31.8% in mothers aged from 31 to 35 years. Hence, knowl-
edge of changes in gestational-age-specific mortality over
time may reveal underlying causes of fetal death and also
be an evaluation of obstetric care [25].
Consistent demographic factors for fetal death induce
race, low socioeconomic status, inadequate prenatal care,
less education, and advanced maternal age [6]. This fact is
proven in many studies including in the present study; after
the end of 2008 when the economic and humanitarian cri-
sis began in Greece, it is interesting to note a small rise of
fetal mortality in the present Hospital (Figure 2).
Conclusion
The perinatal mortality indicator plays an important role
in providing the information needed to improve the health
status of pregnant women, new mother, and newborns. That
information allows decision-makers to identify problems,
track temporal and geographical trends, disparities, and as-
sess changes in public health policy and practice.
Common causes and risk factors for fetal death include
chromosomal abnormalities, placental disorders, and intra-
partum asphyxia. Clinicians should encourage families to
allow a thorough investigation of potential causes of fetal
deaths to facilitate emotional closure, to assess recurrence risk.
Prospective surveillance can result in the timely delivery
of a fetus at risk from an unfavourable intrauterine envi-
ronment. This is now assisted by technological methods,
whereas problems as prematurity and fetal growth restric-
tion still remain unrecognized antenatally.
739
References
[1] Gordon C.S.: “Estimating risk of perinatal death”. AJOG, 2005, 192,
17.
[2] World Heath Organization: “Neonatal and perinatal mortality: country,
regional and global estimates”, 2006. Available at: http://apps.who.int/
iris/bitstream/10665/43444/1/9241563206_eng.pdf
[3] Sarfraz A.A., Samuelsen S.O., Esklid A.: “Changes in fetal death dur-
ing 40 years-different trends for different gestational ages: a popula-
tion-based study in Norway”. BJOG, 2010, 118, 488.
[4] Gardosi J., Kady S.M., McGeown P., Francis A., Tonks A.: “Classifi-
cation of stillbirth by relevant condition at death (ReCoDe): population
based cohort study”. BMJ, 2005, 331, 1113.
[5] Pantou K., Drougia A., Krallis N., Hotoura E., Papassava M., An-
dronikou S.: “Perinatal and neonatal mortality in Northwest Greece
(1996-2004)”. J. Matern. Fetal Neonatal Med., 2010, 23, 1237.
[6] Fretts R.C.: “Etiology and prevention of stillbirth”. Am. J. Obstet. Gy-
necol., 2005, 193, 1923.
[7] Kalter H.: “Five-decade international trends in the relation of perinatal
mortality and congenital malformations: stillbirth and neonatal death
compared”. Int. J. Epidemiol., 1991, 20, 173.
[8] Smith G.C., Fretts R.C.: “Stillbirth”. Lancet, 2007, 370, 1715.
[9] Dolk H., Loane M., Garne E.: “The prevalence of congenital anom-
alies in Europe”. Adv. Exp. Med. Biol., 2010, 20, 349.
[10] Silver R.M.: “Fetal death”. Obstet. Gynecol., 2007, 109, 153.
[11] Henry L.G.: “Timing delivery of the growth-restricted fetus”. Semin.
Perinatol., 2011, 35, 262.
[12] Baschat A.A., Galan H.L., Ross M.: “Fetal growth restriction”. Ob-
stetrics: Normal and Problem Pregnancies, 2007, 5, 771.
[13] Monnier P., Pilar Velez M., Forges T.: “The impact of assisted repro-
ductive technologies on perinatal outcomes”. Pregnancy Disorders and
Perinatal outcomes, 2012, 2, 20.
[14] Sunderman S., Chang J., Flowers L.: “Assisted reproductive technol-
ogy surveillance- United States 2006”. MMWR Surveill. Summ., 2009,
58, 1.
[15] Agence de la biomedicine: “Bilan des activites de procreation et géné-
tique humaines en France 2007”.
[16] Korteweg F.J., Erwick J.J., Holm J., Ravise J., Meer J., Veeger N., Tim-
mer A.: “Diverse placental pathologies as the main causes of fetal
death”. Obstet. Gynecol., 2009, 114, 809.
[17] Kristensen J., Vestergaard M., Wisborg K., Kesmodel U., Secher N.J.:
“Pre-pregnancy weight and the risk of stillbirth and neonatal death”.
BJOG, 2005, 112, 403.
[18] Information and Statistics Division NHS Scotland: “Scottish perinatal
and infant mortality report 2000”. Edinburgh: ISD Scotland Publica-
tions, 2001.
[19] Stubblefield P.G., Berek J.S.: “Perinatal mortality in term and post-term
births”. Obstet. Gynecol., 1980, 56, 678.
[20] Laube D.W., Schauberger C.W.: “Fetomaternal bleeding as a cause foe
“unexplained” fetal death”. Obstet. Gynecol., 1982, 60, 649.
[21] Mmbaga B., Lie R.T., Olomi R., Johnson M., Olola O., Daltveit A.K.:
“Causes of perinatal death at a tertiary care hospital in Northern Tan-
zania 2000-2010: a registry based study”. BMC Pregnancy Childbirth,
2012, 12, 139.
[22] Fretts R.C., Boyd M.E., Usher R.H., Usher H.A.: “The changing pat-
tern of fetal death, 1961-1988”. Obstet. Gynecol., 1992, 79, 35.
[23] Goldenberg RL, Kirby R, Culhane JF. Stillbirth: a review. J. Matern.
Fetal Neonatal Med., 2004, 16, 79.
[24] Walsh CA, Vallerie AM, Baxi LV. Etiology of stillbirth at term: a 10-
year cohort study. J. Matern. Fetal Neonatal Med., 2008, 21, 493.
[25] Macdorman M.F., Kirmeyer S.: “Fetal and perinatal mortality”. Natl.
Vital Stat. Rep., 2009, 57, 1.
Corresponding Author:
C. GOUDELI, M.D.
117 Perikleous Street
Athens Attiki 15233 (Greece)
e-mail: cgoud10@yahoo.gr
CEOG Clinical and Experimental
Obstetrics & Gynecology

Roles of high-risk human papilloma virus (HR-HPV)

E6/E7mRNA in triaging HPV16/18 cases

L. Liu, Y.M. Chen, Q.Y. Zhang, C.Z. Li

Department of Obstetrics and Gynecology, Shandong Provincial Hospital Affiliated to Shandong University, Jinan (China)

Summary
Objective: This study aimed to investigate the roles of high-risk human papilloma virus (HR-HPV) E6/E7mRNA in triaging patients
negative for intraepithelial lesion (NILM) accompanied with HPV16/18 infection. Materials and Methods: A total of 514 female pa-
tients simultaneously underwent cytological assay, HPV-DNA assay, E6/E7mRNA detection, and colposcopic biopsy were selected. The
study endpoint was the histologically confirmed high-grade cervical intraepithelial neoplasia (CIN) II or higher(II+). Results: The pos-
itive rates of E6/E7mRNA in histopathologically confirmed cervicitis, CIN I, CIN II, CIN III, and cervical cancer were 53.4%, 66.7%,
89.9%, 91.4%, and 100%, respectively. Among the patients that underwent the colposcopic biopsy due to cytological NILM plus
HPV16/18 infection, the positive predictive values of HPV16/18 and E6/E7mRNA towards high-grade cervical lesions were 21.5% and
40.4%, respectively, and the comparison between these two factors showed statistically significant difference (p < 0.05). The negative
predictive value of E6/E7mRNA was 97.8%. Conclusions: E6/E7mRNA showed good triaging effects towards the patients with cyto-
logical NILM plus HPV16/18 infection and could significantly reduce colposcopy and biopsy.

Key words: Cervical cancer; Screening; E6/E7mRNA; HPV.

Introduction intraepithelial lesion (NILM) plus HPV16/18 infection,

In 1970s, German virologist ZurHausen firstly proposed
the assumption that human papilloma virus (HPV) was
closely related with the onset of cervical cancer; a large
number of studies had shown that HPV infection, espe-
cially persistent high-risk HPV infection, was the main
reason of the most cases of cervical precancerous lesions
and cervical cancer [1, 2]. HPV-DNA detection could in-
crease the detection rate towards high-grade cervical in-
traepithelial neoplasia (CIN) and cervical cancer [3], but
80% of high-risk virus infections were transient [4], and
most new infections could be extinct in two years [5].
Only 1% of the high-risk virus infected women would
gradually develop into cervical cancer [6]. This caused
the fact that though the sensitivity of HPV-DNA detection
was high, its specificity was relatively low [7]; therefore,
it might result in unnecessary colposcopy and biopsy in
many women [8], thus increasing patients’ economic bur-
dens and psychological burdens. The clinical diagnosis
and treatment urgently required one screening method
with high specificity and sensitivity towards cervical can-
cer [9], and the E6/E7mRNA detection is one of the re-
search hotspots. Study had shown that the overexpression
of E6/E7 oncoprotein was closely related to the progres-
sion risk of cervical diseases [10]; meanwhile, it also had
important significance in triaging high-risk cervical virus
infections [11]. This study analyzed the expressions of
E6/E7mRNA in the patients with cytological negative for
aiming to investigate its roles and values in triaging high-
level cervical lesions.
Materials and Methods
A total of 514 female patients simultaneously underwent cyto-
logical assay, HPV-DNA assay, E6/E7mRNA detection, and col-
poscopic biopsy in Wenzhou People’s Hospital from April 2014 to
August 2015 and were selected, among which 93 patients with cy-
tological NILM plus HPV16/18 infection. All the study subjects
had no history of CIN, cervical cancer, pelvic radiation therapy,
total hysterectomy, or present pregnancy.This study was conducted
in accordance with the declaration of Helsinki and with approval
from the Ethics Committee of Shandong University. Written in-
formed consent was obtained from all participants.
One cervical brush was rotated three to five circles at the cer-
vical squamous columnar junction area; the brush was then fully
rinsed in ThinPrep cell preservation solution to maximally collect
the cells sampled,.The solution was then sent to the cell lab for
programmed processing; the cytological diagnosis was performed
by professional gynecological pathologists, and the result deter-
mination referred to the modified TBS classification system
(2001).
HPV-genotype microarray detection system, PE5700 gene
amplifier, HbriMax medical rapid nucleotide hybridization in-
strument, and HPV amplification genotype detection kit were
utilized. DNA extraction kit used the Cape medical rapid nu-
cleotide hybridization instrument as the platform and then high-
throughput detected the 21 HPV subtypes (accounting for 95%
of HPV infection) on the nucleotide probe-fixed low-density
microarray film using the principle of flow-through hybridiza-

Revised manuscript accepted for publication February 29, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3613.2017
 L. Liu, Y.M. Chen, Q.Y. Zhang, C.Z. Li
Table 1. — Relationship between E6/E7mRNA-positive
rates and pathological results.
Index Cases E6/E7mRNA-positive rate (%)
Normal/inflammation 245 131 (53.47%)
CIN I 87 58 (66.67%)*
CIN II 89 80 (89.89%)*#
CIN III 81 74 (91.36%)*#
Cervical cancer 12 12 (100%)*#Δ
* Compared with normal or inflammation, significant difference;
# compared with group CIN I, statistically significant difference;
Δ compared with group CIN II, statistically significant difference.
tion, including 13 kinds of high-risk subtypes (16, 18, 31, 33,
35, 39, 45, 51, 52, 56, 58, 59, and 68), five kinds of low-risk
subtypes (6, 11, 42, 43, and 44), and three kinds of common
subtypes in Chinese populations (53, 66, and CP8304).
The Quanti Virus HPV E6/E7mRNA diagnosis kit was used to
detect 14 kinds of HR-HPV (HPV 16, 18, 31, 33, 35, 39, 45, 51,
52, 56, 58, 59, 66, and 68) using the branched DNA technology.
The patients with abnormal colposcopic results then underwent
cervical biopsy; the pathological results were diagnosed by two
professional pathologists. The diagnostic criteria referred to the
WHO classification criteria.
SPSS17.0 statistical software was used for the statistical pro-
cessing. The counting data were performed using the χ2 test, with
p < 0.05 considered as statistically significant. The positive and
negative predictive values were calculated using traditional con-
tingency table.
Results
The detection results of E6/E7mRNA in a variety of
pathological lesions are shown in Table 1; it could be seen
that along with the increased levels of the cervical lesions,
the positive rates of E6/E7mRNA gradually increased,
among which the intergroup comparison between CIN II
and CIN III groups, as well as between CIN III and cervi-
cal cancer groups, which showed no statistically significant
difference (p > 0.05). The comparison of the positive rates
of E6/E7mRNA between CIN II− and CIN II+ groups
showed statistically significant difference (p < 0.05). The
data are shown in Tables 1 and 2.
These was a total of 93 patients with cytological NILM plus
HPV16/18 infection, including 47 cases with E6/E7mRNA-
positive rate and 46 negative cases; 20 cases were patholog-
ically diagnosed as CIN II+. The positive predictive value of
HPV16/18 was 21.51% (20/93); the positive predictive value
of E6/E7mRNA was 40.43% (19/47). The difference between
the two group was statistically significant (21.51%, 40.43%,
χ2 = 20.15, p < 0.05); the negative predictive value of
E6/E7mRNA was 97.83% (45/46, Table 3).
Discussion
Lu et al. have shown that only when the number of the
host cell-integrated HR-HPV reached a certain level, the
741
Table 2. — Relationship between E6/E7mRNA-positive
rates and pathological results.
Item Number of patients Positive rate of E6/E7mRNA(%)
CIN II−a 332 189 (56.93%)
CIN II+b 182 166 (91.21%)
c2 64.66
P 0.00
a
CIN II−, including normal cervix, inflammation, and CIN I;
b
CIN II+, including CIN II, CIN III, and cervical cancer.
Table 3. — Relationships between E6/E7mRNA and differ-
ent cervical lesions in patients with NILM plus HPV16/18
infection.
Group Cases E6/E7mRNA
Positive Negative
CIN II+ 20 19 1
CIN II− 73 28 45
E6/E7 oncogene could be overexpressed, and then high-
level lesions even cervical cancer might develop [12].
Therefore, detecting the expression of the E6/E7 oncogene
could triage whether the cervical high-risk virus was per-
sistent or transient [13, 14]; according to the central dogma,
the expression of the E6 and E7 oncogenes could be real-
ized through detecting the transcription of E6/E7mRNA in
the HPV oncogene [15].
The present study aimed to investigate whether the de-
tection of E6/E7mRNA using the branched DNA technol-
ogy could effectively triage the patients with cytological
NILM plus HPV16/18 infection. The results showed that
if these patients further underwent the E6/E7mRNA detec-
tion triaging, the positive predictive value towards the high-
grade cervical lesions could be significantly improved;
meanwhile, its negative predictive value was also high, and
it could effectively reduce the colposcopy and biopsy, thus
alleviating patients’ mental and financial burdens.
The results of this study also showed that the positive
expression rate of E6/E7mRNA was gradually increased
with the increasing of the histopathological levels, con-
sistent with Ratnam et al. and Coquillard et al. [16, 17].
In particular, the positive rate of E6/E7mRNA in CIN II+
was significantly higher than CIN II−, indicating that the
E6/E7mRNA expression was closely related with the cer-
vical lesions, and its high expression could facilitate the
occurrence and development of cervical precancerous le-
sions.
Rijkaart et al. [18] found that 8% of the patients with
normal cytological results plus cervical high-risk virus in-
fection would eventually develop into high-level cervical
lesions (CIN II+). Seventy percent of the patients with
cervical cancer would be accompanied with the HPV16
or 18 infection, among which the cervical squamous cell
742 Roles of high-risk human papilloma virus (HR-HPV) E6/E7mRNA in triaging HPV16/18 cases
carcinoma was usually accompanied with the HPV16 in-
fection [19]; therefore, in patients older than 30 years,
even if their cytological results were normal, once the
HPV16/18 infection was found, the colposcopy and
pathological biopsy should also by recommended. The re-
sults of this study revealed that among the 93 patients with
cytological NILM plus HPV16/18 infection, 20 patients
underwent high-grade cervical changes (CIN II+), and the
positive predictive value of HPV16/18 towards the high-
grade cervical lesions was only 21.51%. If these patients
further underwent the E6/E7mRNA triaging, the present
authors found that the positive predictive value of
E6/E7mRNA in these patients was 40.43%, and the neg-
ative predictive value was 97.83%. The positive predic-
tive value of E6/E7mRNA was significantly higher than
that of HPV16/18; more importantly, E6/E7mRNA had a
high negative predictive value, i.e. if the E6/E7mRNA test
result was negative, the patients would rarely develop
high-grade cervical lesions, therefore, these patients could
be recommended outpatient follow-up. This would greatly
reduce the colposcopy and biopsy. Hence, the
E6/E7mRNA detection could be an effective triaging ap-
proach in patients with cytological NILM plus HPV16/18
infection. The results of this study were similar to those by
Perez and Rijkaart et al. [11, 20].
Limitations of the present study are the following: first,
the case number was small, therefore a larger sample size
is required to further confirm the above findings. Second,
through the E6/E7mRNA detection, one case of CIN II
was misdiagnosed: this patient was 45-years-old and un-
derwent cervical conization. Therefore, if the E6/E7mRNA
test results in young CIN II patients with fertility require-
ments are negative, could they have the follow-up obser-
vation? Meanwhile, if the E6/E7mRNA test results in the
patients with negative cytological or histopathologic test
results are positive, would they be more likely to develop
to CIN II+ in future? These queriea require further in-depth
studies.
Acknowledgments
This study was supported by the Shandong Provincial
Science and Technology Development Projects
(2014GSF118048).
References
[1] Waldstrøm M., Ornskov D.: “Clinical performance of a human pa-
pillomavims messenger RNA test (Aptima HPV Assay) on residual
material from archived 3-year-old PreservCyt samples with low-
grade squamous intraepithelial lesion”. Arch. Pathol. Lab. Med.,
2011, 135, 1052.
[2] Baussano I., Elfström K.M., Lazzarato F., Gillio-Tos A., De Marco
L., Carozzi F., et al.: “Type-specific human papillomavirus biologi-
cal features: validated model-based estimates”. PLoS One, 2013, 8,
e81171.
[3] Matah M., Sareen S.: “Detection of HPV by PCR-A novel step in
the prevention of cancer cervix”. J. Obstet. Gynecol. India, 2012,
62, 188.
[4] Mehlhorn G., Obermann E., Negri G., Bubendorf L., Mian C., Koch
M., et al.: “HPV L1 detection discriminates cervical precancer from
transient HPV infection: a prospective international multicenter
study”. Mod. Pathol., 2012, 26, 967.
[5] Jaisamrarn U., Castellsagué X., Garland S.M., Naud P., Palmroth J.,
Del Rosario-Raymundo M.R., et al.: “Natural history of progression
of HPV infection to cervical lesion or clearance: analysis of the con-
trol arm of the large, randomized PATRICIA study”. PLos One,
2013, 8, e79260.
[6] Josefsson A.M., Magnusson P.K., Ylitalo N., Sørensen P., Qwar-
forth-Tubbin P., Andersen P.K., et al.: “Viral load of human pa-
pilloma virus 16 as a detetmination for development of cervical
carcinoma in situ: a nested case-control study”. Lancet, 2000, 355,
2189.
[7] Burger E.A., Kornør H., Klemp M., Lauvrak V., Kristiansen I.S.:
“HPV mRNA tests for the detection of cervical intraepithelial neo-
plasia: a systematic review”. Gynecol. Oncol., 2011, 120, 430.
[8] Schiffman M., Solomon D.: “Screening and prevention methods for
cervical cancer”. JAMA, 2009, 302, 1809.
[9] Franco E.L., Cuzick J., Hildesheim A., de Sanjosé S.: “Chapter 20:
Issues in planning cervical cancer screening in the era of HPV vac-
cination”. Vaccine, 2006, 24, S3/171.
[10] Tropé A., Sjøborg K., Eskild A., Cuschieri K., Eriksen T., Thoresen
S., et al.: “Performance of human papillomavims DNA and mRNA
testing strategies for women with and without cervical neoplasia”. J.
Clin. Microbiol., 2009, 47, 2458.
[11] Perez Castro S., Iñarrea Fernández A., Lamas González M.J.,
Sarán Diez M.T., Cid Lama A., Alvarez Martín M.J., et al.:
“Human papillomavirus (HPV) E6/E7 mRNA as a triage test after
detection of HPV 16 and HPV 18 DNA”. J. Med. Virol., 2013,
85, 1063.
[12] Lu X., Lin Q., Lin M., Duan P., Ye L., Chen J., et al.: “Multiple-in-
tegrations of HPV16 genome and altered transcription of viral onco-
genes and cellular genes are associated with the development of
cervical cancer”. PLoS One, 2014, 9, e97588.
[13] Möckel J., Quaas J., Meisel H., Endres A.S., Schneider V.: “Human
papillomavirus E6/E7 mRNA testing has higher specificity than liq-
uid-based DNA testing in the evaluation of cervical intraepithelial
neoplasia”. Anal. Quant. Cytol. Histol., 2011, 33, 311.
[14] Cattani P., Zannoni G.F., Ricci C., D’Onghia S., Trivellizzi I.N., Di
Franco A., et al.: “Clinical performance of human papillomavirus
E6 and E7 mRNA testing for high-grade lesions of the cervix”. J.
Clin. Microbiol., 2009, 47, 3895.
[15] Dockter J., Schroder A., Hill C., Guzenski L., Monsonego J., Gia-
chetti C.: “Clinical performance of the APTIMA HPV Assay for the
detection of high-risk HPV and high-grade cervical lesions”. J. Clin.
Virol., 2009, 45, S55.
[16] Ratnam S., Coutlee F., Fontaine D., Bentley J., Escott N., Ghatage
P., et al.: “Clinical performance of the PreTect HPV-Proofer
E6/E7mRNA assay in comparison with that of the Hybrid Capture 2
test for idertification of women at risk of cervical cancer”. J. Clin.
Microbiol., 2010, 48, 2779.
[17] Coquillard G., Palao B., Patterson B.K.: “Quantification of intracel-
lular HPV E6/E7mRNA expression increases the specificity and pos-
itive predictive value of cervical cancer screening compared to HPV
DNA”. Gynecol. Oncol., 2011, 120, 89.
[18] Rijkaart D.C., Berkhof J., van Kemenade F.J., Coupe V.M., Hes-
selink A.T., Rozendaal L., et al.: “Evaluation of 14 triage strategies
for HPV DNA-positive women in population-based cervical screen-
ing”. Int. J. Cancer, 2012, 130, 602.
[19] Kang W.D., Kim C.H., Cho M.K., Kim J.W., Cho H.Y., Kim Y.H.,
et al.: “HPV-18 is a poor prognosticfactor, unlike the HPV viral load,
in patients with stage IB-IIA cervical cancer undergoing radical hys-
terectomy”. Gynecol. Oncol., 2011, 121, 546.
[20] Rijkaart D.C., Heideman D.A., Coupe V.M., Brink A.A., Verheijen
R.H., Skomedal H., et al.: “High-risk human papillomavirus
(hrHPV) E6/E7 mRNA testing by pretest HPV-Proofer for detection
 L. Liu, Y.M. Chen, Q.Y. Zhang, C.Z. Li 743

of cervical high-grade intraepithelial neoplasia and cancer among Corresponding Author:

hrHPV DNA-positive women with normal cytology”. J. Clin. Mi- C.Z. LI, M.D.
crobiol., 2012, 50, 2390. Department of Obstetrics and Gynecology
Shandong Provincial Hospital
Affiliated to Shandong University
Longitude 5, Latitude 7, no. 324
Jinan 250000, China
e-mail: zhangzhonglidc@163.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

The relevance of fascial surgical repair in the management

of pelvic organ prolapse (POP)

F. Nobili, A. Lukic, I. Puccica, M. Vitali, M. Schimberni, F. Manzara, A. Frega, B. Mossa, M. Moscarini†, D. Caserta
Surgical and Medical Department of Translational Medicine, Sant’Andrea Hospital, Faculty of Medicine and Psychology,
University of Rome “la Sapienza”, Rome (Italy)

Summary
Purpose: To evaluate the anatomical and functional outcomes and post-operative compliance of fascial surgical repair in the man-
agement of pelvic organ prolapse (POP). Materials and Methods: The authors analyzed 147 patients before and after surgical treatment
for POP analyzing pre- and post-operative symptoms. Patients were divided into two groups: group A patients who underwent vaginal
hysterectomy, associated with anterior, posterior, and/or both vaginal repair; group B that underwent only anterior and/or posterior sur-
gical vaginal correction. Results: The average time of post-operative hospitalization was significantly longer in group A than in group
B (p = 0.019). However group A showed a better outcome in terms of days after surgery regarding post voiding residual <100 cc (p =
0.039). During follow-up, urinary incontinence improved (p= 0.001), whereas pelvic pressure, regular bowel function, and improvement
in sexual activity were not significant (p > 0.05). Conclusions: Currently we do not have a surgical procedure which can be considered
the best for treating prolapse, so it seems that the best option is a personalized surgery tailored for each patient.

Key words: Pelvic organ prolapse; Urinary incontinence; POP surgery; POP-Q; Quality of life; Mesh; Hysterectomy; Sexual function.

Introduction
Pelvic organ prolapse (POP) is a disorder that affects over
50% of women aged over 79 years and 10% of those be-
tween 30 and 39 years [1]. The increase of average life ex-
pectancy highlights the importance of POP in terms of
prevention and management. Among patients referred to
the present Department, over 50% of them presented with
an anatomical alteration of the pelvis, but only 3-6% re-
ported associated symptoms that compromised quality of
life. When conservative therapies, physiotherapy or vaginal
pessaries can no longer control symptoms, surgical correc-
tion is the treatment most frequently used. Pelvic floor sur-
gery is a functional surgery, which must seek to recover the
quality of life of women while not always coinciding with
anatomical healing. In the last 20 years the use of prosthetic
surgery (mesh) for surgical correction of prolapse, which
had raised hopes of a better outcome in terms of durability
than fascial surgery, does not seem to have achieved the ex-
pected improvements, while some reported severe compli-
cations [2]. Traditional fascial surgery, based on the ability
of original vaginal tissue to repair itself, played an impor-
tant role in the treatment of prolapse and is now being re-
considered as the first choice when conservative treatment
is no longer conclusive.
The aim of this study was to evaluate the anatomical and
functional outcomes and postoperative compliance of fas-
cial surgery.
Materials And Methods
The study was conducted at the Department of Surgery and Medi-
cine and Translational Medicine, Sant’Andrea Hospital, Faculty of
Medicine and Psychology at “La Sapienza” University of Rome, be-
tween January 2009 and December 2015. The study was reviewed and
approved by the Institutional Review Board and was conducted in ac-
cordance to the Helsinki Declaration. Patients were divided into two
groups according to the type of surgery: group A patients who were
subjected to vaginal hysterectomy (with or without salpingo-oophorec-
tomy), associated with anterior, posterior, and/or both vaginal correc-
tion; group B that had had only anterior and/or posterior surgical
vaginal correction without vaginal hysterectomy. Three surgeons, ex-
perts in vaginal surgery, performed all surgical treatments. Among pa-
tients referred to the present Department, 147 were enrolled in the study.
Inclusion criteria were the presence of symptomatic genital prolapsed
or prolapse of grade III or IV according to the classification of POP-Q
examination and patients who had undergone vaginal hysterectomy or
plastic surgery of the vaginal walls. A questionnaire (P-QOL, Prolapse
- Quality of Life, edited version 4) was performed from two to six years
after surgery [3]. The data was processed using SPSS software version
21.0. For numeric variables, the authors verified the normal distribution
with the application of the Kolmogorov-Smirnov test (K-S). In case of
normal distribution (K-S test value of p > 0.05) we proceeded by the
application of parametric tests (Student’s T) to verify the significance
between the mean values. Otherwise, with values of p < 0.05 for the K-
S test, non-parametric tests were applied (Mann-Whitney U test). The
presence of association between nominal variables was evaluated
through the application of the Chi-Square test. McNemar’s test was
applied in order to verify the existence of significant differences in di-
chotomous data (presence/absence of a symptom) before and after sur-
gery, and then to assess its effectiveness.

Revised manuscript accepted for publication November 14, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog4001.2017
 F. Nobili, A. Lukic, I. Puccica, M. Vitali, M. Schimberni, F. Manzara, A. Frega, B. Mossa, M. Moscarini, D. Caserta 745

Table 1. — Features of study population. patients enrolled had a symptomatic genital prolapse
n Min-max Mean (SD) Median (pelvic pressure, urinary incontinence, irregular bowel
Current age (years) 146 40-87 68.3 (8.8) 68.0 function, and sexual activity). The surgical technique cho-
Age at surgery (years) 147 36-83 64.4 (8.9) 64.0 sen was based on the vaginal compartment involved (ante-
Weight (kg) 136 44-99 65.1 (9.6) 65.0 rior, posterior, and apical), age, symptoms, and above all
Height (m) 138 1.45-1.80 1.60 (0.06) 1.60 the patients requirements. There were no significant differ-
Max birth weights (gr) 56 1750-5000 3612.5 (595.6) 3600.0 ences between the mean values of the variables, current
Menopausal age 133 38-58 50.5 (3.9) 51.0
age, age at time of surgery, BMI, maximum birth weight,
and age at menopause and parity in the two groups (Table
2). Post-operative features analyzed were: time of hospital
stay, post-operative temperature higher than 37.5°C, post-
Results
voiding residual greater than 100 cc, and time of catheter
The characteristics of the study population are summa- removal. The average time of post-operative hospitaliza-
rized in Table 1. Concerning the BMI, out of 136 patients tion was significantly greater in group A (6.8 days) than in
(11 patients were missing for incomplete data), 70 patients group B (6.2 days)(p = 0.019). Moreover group A showed
(51.5%) were normal weight, 40 (29.4%) were overweight, a better outcome in terms of days after surgery with post-
21 (15.4%) were obese, and five (3.7%) were underweight. voiding residual < 100 cc (p = 0.039). The results are sum-
Regarding parity, 83 patients (56.5%) had had two deliver- marized in Table 3.
ies, 27 patients (18.4%) three deliveries, 22 patients Five intra- and peri-operative complications (4%) were
(15.0%) one delivery, 11 patients (7.6%) more than four observed in group A. Out of these five, one was an acci-
deliveries, and four patients (2.7%) were nulliparous. The dental injury of the bladder that was immediately repaired;
average birthweight was 3,612.5 grams (n=56; SD=595.6). in two cases it was necessary to perform a laparoscopy in
The average age and median of menopausal women were order to repair a lesion of the ovarian pedicle and to remove
respectively 50.5 (SD=3.9) and 51.0 years (Table 1). The a patch. One patient experienced vaginal bleeding requiring
most common co\morbidities among these patients were suture (within 12 hours), and the last patient had a pelvic
hypertension (43.5%; n=64), followed by hypothyroidism hematoma which resolved itself spontaneously.
(16.3%; n=24), chronic obstructive pulmonary disease The Chi-Square test showed no significance (p > 0.05)
(COPD) (8.8%; n=13), and diabetes (85.4%; n=8). The 147 between the type of surgery and the comorbidities (hyper-
patients were divided into two groups: group A (n=121; tension, hypothyroidism, diabetes, and COPD).
82.3%) and group B (n=26; 17.7%) depending on the type Out of 147 patients, 119 were subjected to a question-
of surgery, to evaluate the accuracy of surgical indication, naire for the follow-up (25 patients were lost, two did not
post-operative course, and early and late complications. All give their consent, and one patient had died). The aim of

Table 2. — Mean of demographic and physical variables in the two different groups of patients.
Group A mean (SD) Group B mean (SD) p-value K-S p-value Mann Whitney or Student’s t-test
Current age (years) 68.7 (8.7) 66.3 (9.2) < 0.05 0.493
Age at surgery (years) 64.9 (8.7) 62.2 (9.6) < 0.05 0.444
Weight (kg) 65.8 (9.8) 61.7 (9.0) < 0.05 0.069
Height (m) 1.61 (0.06) 1.59 (0.07) < 0.05 0.119
Max birth weights (gr) 3618.5 (598.6) 3585 (612.4) > 0.05* 0.874
Age at menopause 50.7 (3.9) 49.7 (4.3) < 0.05 0.195
Childbirths number 2.2 (1.1) 2.2 (0.7) < 0.05 0.571
*Student t-test was applied .

Table 3. — Mean of post-surgical variables in the two different groups of patients.

Group A mean (SD) Group B mean (SD) p-value K-S p-value Mann Whitney
or Student’s t-test
Total period of hospitalization (days) 10.5 (3.2) 10.2 (5.3) < 0.05 0.153
Period of post-operative hospitalization (days) 6.8 (1.9) 6.2 (3.0) < 0.05 0.019
Days with fever > 37.5°C 1.84 (1.3) 2.00 (0.9) < 0.05 0.407
Catheter removal, post-operative day 3.4 (1.1) 3.5 (1.4) < 0.05 0.989
Days with post-voiding residual > 100 cc 2.8 (2.3) 5.0 (1.7) < 0.05 0.039
746 The relevance of fascial surgical repair in the management of pelvic organ prolapse (POP)

Table 4. — Symptoms before and after surgery in the whole sample.

Post
Yes No Total McNemar p-value
Pelvic pressure Pre Yes 5 (4.2%) 114 (95.8%) 119 (100.0%)
No — — — —*
Total 5 (4.2%) 114 (95.8%) 119 (100.0%)
Urinary incontinence Pre Yes 19 (38.8%) 30 (61.2%) 49 (100.0%)
No 9 (12.9%) 61 (87.1%) 70 (100.0%) 0.001
Total 28 (23.5%) 91 (76.5%) 119 (100.0%)
Regular bowel function Pre Yes 89 (98.9%) 1 (1.1%) 90 (100.0%)
No 0 (0.0%) 29 (100.0%) 29 (100.0%) 1.000
Total 89 (74.8%) 30 (25.2%) 119 (100.0%)
Improvement in sexual activity Pre Yes — — —
No 39 (76.5%) 12 (23.5%) 51 (100.0%) —*
Total 39 (76.5%) 12 (23.5%) 51 (100.0%)
*McNemar test was not been applied because there was only one answer mode in pre-operative time, respectively, “no” for pelvic pressure and “yes” for im-
provement in sexual activity.

Table 5. — Symptoms before and after surgery in group A.

Post
Yes No Total McNemar p-value
Pelvic pressure Pre Yes 1 (1.0%) 96 (99.0%) 97 (100.0%)
No — — — —*
Total 1 (1.0%) 96 (99.0%) 97 (100.0%)
Urinary incontinence Pre Yes 16 (42.1%) 22 (57.9%) 38 (100.0%)
No 8 (13.6%) 51 (86.4%) 59 (100.0%) 0.016
Total 24 (24.7%) 73 (75.3%) 97 (100.0%)
Regular bowel function Pre Yes 75 (98.7%) 1 (1.3%) 76 (100.0%)
No 0 (0.0%) 21 (100.0%) 21 (100.0%) 1.000
Total 75 (77.3%) 22 (22.7%) 97 (100.0%)
Improvement in sexual activity Pre Yes — — —
No 34 (81.0%) 8 (19.0%) 42 (100.0%) —*
Total 34 (81.0%) 8 (19.0%) 42 (100.0%)
*McNemar test was not applied because there was only one answer mode in pre-operative time, respectively, “no” for pelvic pressure and “yes” for improvement
in sexual activity.

Table 6. — Symptoms before and after surgery in group B.

Post
Yes No Total McNemar p-value
Pelvic pressure Pre Yes 4 (18.2%) 18 (81.8%) 22 (100.0%)
No — — — —*
Total 4 (18.2%) 18 (81.8%) 22 (100.0%)
Urinary incontinence Pre Yes 3 (27.3%) 8 (72.7%) 11 (100.0%)
No 1 (9.1%) 10 (90.9%) 11(100.0%) 0.039
Total 4 (18.2%) 18 (81.8%) 22 (100.0%)
Regular bowel function Pre Yes 14 (100.0%) 0 (0.0%) 14 (100.0%)
No 0 (0.0%) 8 (100.0%) 8 (100.0%) 1.000
Total 14 (63.6%) 8 (36.4%) 22 (100.0%)
Improvement in sexual activity Pre Yes — — —
No 5 (55.6%) 4 (44.4%) 9 (100.0%) —*
Total 5 (55.6%) 4 (44.4%) 9 (100.0%)
*McNemar test was not been applied because there was only one answer mode in pre-operative time, respectively, “no” for pelvic pressure and “yes” for im-
provement in sexual activity.
 F. Nobili, A. Lukic, I. Puccica, M. Vitali, M. Schimberni, F. Manzara, A. Frega, B. Mossa, M. Moscarini, D. Caserta
the questionnaire was to detect the presence of pelvic pres-
sure, urinary incontinence, regular bowel function, and im-
provement of sexual activity, in order to compare
symptoms before and after surgery (Table 4). The McNe-
mar test cannot be applied for pelvic pressure and for im-
provement of sexual activity because in both cases the
variable relating to pre-surgery presents only one answer
mode. All patients had resolved pelvic pressure and 23.5%
had restarted sexual activity. Sexual activity was not in-
vestigated in 68 patients because it was not possible to eval-
uate. Concerning urinary incontinence, McNemar showed
a statistically significant difference (p = 0.001) between pa-
tients who had been cured (61.2%) or had improved
(38.8%) after surgery.
Regular bowel function did not show any significant dif-
ferences (p = 1.000) before and after surgery (Table 4). The
trend was similar in the two groups (Tables 5 and 6). The
average level of satisfaction on a visual scale from 1 to 10
among all patients was 8.7 (SD=2.0); the Mann-Whitney
U test showed an average value significantly higher in
group A (8.9) than in group B (7.7) (p = 0.002).
Discussion
POP is a common condition and various factors con-
tribute to its onset. In this study most of the women were in
menopause, in line with reports in medical literature that
ageing affects the quality of muscle-fascial tissue of the
pelvic floor [4]. It is due to the loss of estrogen receptors on
the surface of pelvic tissues which can cause a condition of
hypotrophy-atrophy that can induce the development of
prolapse [5]. 82.3% of patients were multiparous, while
15% had one delivery, and only 2.7% were nulliparous. As
shown in medical literature, each vaginal delivery can dam-
age the pelvis, which is the main risk factor in relation to
parity [4]. It is shown that in a group of women of the same
age that pelvic floor disorders are more common in multi-
parous than in nulliparous women, confirming the role
played by obstetric trauma [6]. The presence of genetic al-
terations of connective tissue of endopelvic fascia and vagi-
nal wall, together with comorbidities (chronic bronchitis,
hypertension or diseases requiring long-term treatment with
corticosteroids) explain the need for surgical treatment for
prolapse in nulliparous women [4, 7]. Moreover, Memon
et al. [8] highlight a higher fetal size among risk factors;
Viktrup et al. [9] identify a higher head circumference of
the fetus as a cause of urinary incontinence onset. In the
present study, the average birth weight was greater than the
average standard values.
It is well known that the removal of the uterus involves
a longer operative time and moderate blood loss increasing
the time spent at hospital [10]. In fact, in the present sam-
ple, the group of women that underwent vaginal hysterec-
tomy had also a significantly longer post-operative
hospitalization. In group B, women took longer in order to
747
be able to urinate spontaneously. The accuracy of plastic
vaginal surgery and the stress to which tissue is exposed
during the operation influences the time required by the pa-
tient before being able to void spontaneously [11]. Pelvic
pressure disappeared especially in patients subjected to
vaginal hysterectomy. Withagen et al. [12] showed that also
repair of cystocele using mesh, if associated with a hys-
terectomy, improves pelvic pressure [13]. Sexual activity
significantly improved in 76.5% of patients. The present
result is confirmed by the literature that shows how recov-
ering an appropriate vaginal function depends on the si-
multaneous restoration of anatomical and neurovascular
factors [14]. In the present study 12.8% of women had a de
novo urinary incontinence. Prolapse and stress urinary in-
continence (SUI) can occur simultaneously, but many
women showed an underestimated incontinence with an in-
creased risk of developing de novo SUI after surgical pro-
lapse repair. Thus, performing an anti-incontinence
procedure at the time of prolapse repair is an effective way
in reducing the risk of hidden SUI postoperatively [15]. The
analysis of the 13 cases of failure highlighted a pre-opera-
tive diagnostic error resulting in an unsatisfactory surgical
result. In eight out of 13 cases (61.4%), a prolapse of cen-
tral or anterior compartment was incorrectly diagnosed. In
the remaining cases, the failure was due to the presence of
several comorbidities (TIA, dementia) at the time of sur-
gery or years later. In fact, even in the study of Marschalek
et al. [10], patients who had maintained their uterus after
surgery developed a new prolapse of the central compart-
ment. Based on the low rate of surgical complications, the
small number of recurrences and patient satisfaction, in the
present authors’ opinion, fascial surgery still plays a rele-
vant role in the treatment of POP.
Conclusions
It is well known that only symptomatic prolapse must be
operated. Nevertheless, age of patient, general health con-
dition, location and number of anatomical defects, sever-
ity of associated symptoms, and nature of employment
must be considered in order to decide the most appropriate
procedure. Since there is still no technique considered to
be the gold standard for prolapse, the authors believe it is
necessary to always perform conservative surgery, to de-
termine the timing and to personalize the surgery according
to the needs of each patient.
References
[1] Nygaard I., Barber M.D., Burgio K.L., Kenton K., Meikle S., Schaf-
fer J., et al.: “Prevalence of symptomatic pelvic floor disorders in
US women”. JAMA, 2008, 300, 1311.
[2] Department of Health and Human Services Food and Drug Admin-
istration: “Reclassification of surgical mesh for transvaginal pelvic
organ prolapse repair and surgical instrumentation for urogyneco-
logic surgical mesh procedures; designation of special controls for
748 The relevance of fascial surgical repair in the management of pelvic organ prolapse (POP)
urogynecologic surgical mesh instrumentation”. Federal Register,
2014, 79, 24634.
[3] Digesu G.A., Khullar V., Cardozo L., Robinson D., Salvatore S.: “P-
QOL: a validate questionnaire to assess the symptoms and quality
of life of women with urogenital prolapse”. Int. Urogynecol. J., 2005,
16, 176.
[4] Barber M.D., Maher C.: “Epidemiology and outcome assessment of
pelvic organ prolapsed”. Int. Urogynecol. J., 2013, 24, 1783.
[5] Chow D., Rodriguez L.V.: “Epidemiology and prevalence of pelvic
organ prolapsed”. Curr. Opin. Urol., 2013, 23, 293.
[6] Bozkurt M., Yumru A.E., Sahin L.: “Pelvic floor disfunction and and
effects of pregnancy and mode of delivery on pelvic floor”. Taiwan
J. Obstet. Gynecol., 2014, 53, 452.
[7] Kawasaki A., Corey E.G., Laskey R.A., Weidner A.C., Siddiqui N.Y.,
Wu J.M.: “Obesity as a risk for the recurrence of anterior vaginal
wall prolapsed after anterior colphorraphy”. J. Reprod. Med., 2013,
58, 195.
[8] Memon H.U., Handa V.L.: “Vaginal childbirth and pelvic floor dis-
orders”. Womens Health, 2013, 9, 265.
[9] Viktrup L., Lose G., Rolff M., Barfoed K.: “The symptom of stress
incontinence caused by pregnancy or delivery in primiparas”. Ob-
stet. Gynecol., 1992, 79, 945.
[10] Marschalek J., Trofaier M.L., Yerlikaya G., Hanzal E., Koelbl H.,
Ott J., Umek W.: “Anatomic outcomes after pelvic-organ-prolapse
surgery comparing uterine preservation with hysterectomy”. Eur. J.
Obstet. Gynecol. Reprod. Biol., 2014, 183, 33.
[11] Van Der Ploeg J.R., Oude Rengerink K.O., Van Der Steen A., Van
Leeuwen J.H.S., Stekelenburg J., Bongers M.Y.: “Transvaginal pro-
lapse repair with or without the addition of a midurethral sling in
women with genital prolapse and stress urinary incontinence: a ran-
domized trial”. BJOG, 2015, 122, 1022.
[12] Withagen M.I., Milani A.L., De Leeuw J.W., Vierhout M.E.: “De-
velopment of de novo prolapse in untreated vaginal compartments
after prolapse repair with and without mesh: a secondary analysis of
a randomized controlled trial”. BJOG, 2012, 119, 354.
[13] Lykke R., Blaakaer J., Ottesen B., Gimbel H. “The indication for
hysterectomy as a risk factor for subsequent pelvic organ prolapse re-
pair”. Int. Urogynecol. J., 2015, 26, 1661.
[14] Jha S., Gray T.: “A systematic review and meta-analysis of the im-
pact of nativetissue repair for pelvic organ prolapse on sexual func-
tion”. Int. Urogynecol. J., 2015, 26, 321.
[15] Siddiqui N.Y., Edenfield A.L.: “Clinical challenges in the manage-
ment of vaginal prolapse”. Int. J. Womens. Health, 2014, 6, 83.
Corresponding Author:
A. FREGA, M.D.
Surgical and Medical Department of Translational Medicine
Sant’Andrea Hospital, Faculty of Medicine and Psychology
University of Rome “la Sapienza”
Via di Grottarossa, 1039
00189 Rome (Italy)
e-mail: antonio.frega@uniroma1.it
CEOG Clinical and Experimental
Obstetrics & Gynecology

Inherited thrombophilia and thromboprophylaxis:

a retrospective analysis of pregnancy outcomes in 106 patients

H. Alptekin1, N. Alptekin2, R. Selimoğlu1, T. Cengiz1, S. Barış3

1 Department of Obstetrics and Gynecology, 2 Department of Pediatrics, 3 Department of Medical Genetics,
Mevlana University Faculty of Medicine, Konya (Turkey)

Summary
Objective: Low-molecular-weight heparin (LMWH) and low-dose aspirin (LDA) given in combination were evaluated in females with
five commonly inherited thrombophilia polymorphisms to address unexplained recurrent pregnancy loss (RPL). Materials and Methods:
After excluding other causes of RPL, 106 of 183 females suffering RPL and diagnosed with inherited thrombophilia were studied along
with 62 healthy, age-matched control subjects carrying one or more pregnancies successfully (no gestational complications or abortion).
Test patients were given a combination of LMWH and LDA. All participants were screened for five thrombophilic mutations: factor V Lei-
den G1691A, prothrombin (FII) A20210G, PAI-1 4G/5G insertion/deletion, and two methylenetetrahydrofolate reductase (MTHFR) poly-
morphisms (C677T and A1298C). Results: With thromboprophylaxis, 73 of 84 (86.9%) pregnancies succeeded, representing a significant
increase in the rate of live births (vs. 232 prior losses). Of the five test panel mutations, three or more (homozygous and/or heterozygous)
were observed in 48 test patients (45.3%), whereas only three control subjects (4.8%) were similarly affected (p < 0.05). Frequencies of all
five mutations were significantly higher in test patients (vs. controls), with PAI-1 4G/5G and MTHFR (C677T and A1298C) identified via
binary logistic regression as independent correlates of habitual abortion. Conclusion: The risk of RPL increases with three or more ho-
mozygous or heterozygous genotypes in inherited thrombophilia, especially with PAI-1 4G/5G and MTHFR (C677T and A1298C). As in
acquired thrombophilia, LMWH/LDA combination treatment may increase live birth rates in patients with inherited thrombophilia.

Key words: Inherited thrombophilia; Thromboprophylaxis; Recurrent pregnancy loss; Low-molecular-weight heparin; Low-dose aspirin.

Introduction frequently in patients with inherited thrombophilia. The

Recurrent pregnancy loss (RPL), defined as at least
three or more (two in some studies) consecutive preg-
nancy losses (usually in the first trimester), is among the
most common causes of female infertility, affecting 1–2%
of women of reproductive age [1]. Because the etiologies
of RPL vary widely, clinical investigations may be exten-
sive and costly. Given that the study of the fetus/embryo
is not feasible in this context, studies are limited to
parental analysis. Once endocrine disorders (i.e., ovarian
dysfunction, thyroid dysfunction, hypopituitarism, and di-
abetes), uterine malformations, chromosomal abnormali-
ties, inflammatory diseases (especially systemic lupus
erythematosus), and infectious diseases are excluded, one
of the most common factors is inherited thrombophilia
[2]. In fact, 50–65% of women with a history of unex-
plained pregnancy loss suffer from inherited or acquired
thrombophilia and may benefit from thromboprophylaxis
[3, 4].
Pathophysiologic mechanisms in thrombophilia involve
placental microcirculatory thrombosis with a heightened
hypercoagulable state during pregnancy [5]. As a result,
venous thromboembolism, preeclampsia, intrauterine
growth retardation, and fetal loss are apt to occur more
latter is attributable to a distinct group of genetic muta-
tions, most of which are inherited as autosomal dominant
traits and lead to hypercoagulable states, namely factor V
Leiden (FVL) G1691A, prothrombin (FII) G20210A,
plasminogen activator inhibitor type 1 (PAI-1) 4G/5G in-
sertion/deletion polymorphism, and hyperhomocysteine-
mia with methylenetetrahydrofolate reductase (MTHFR)
C677T and A1298C mutations. Protein S, protein C, and
antithrombin deficiencies may also result in hypercoagu-
lable states. As acquired thrombophilic defects, antiphos-
pholipid antibodies [lupus anticoagulant (LA), antiβ2
glycoprotein-1 antibodies (antiβ2GP1 Abs), and anticar-
diolipin antibodies (aCL)] are now regarded as important,
treatable causes of RPL. In such instances, antithrombotic
therapies have helped to promote successful pregnancies
[6, 7].
The relationship between acquired thrombophilia and
RPL is well established, but studies of RPL developing in
inherited thrombophilia report conflicting results with re-
spect to the efficacy of thromboprophylaxis [8]. Hence,
the present authors’ intent was to examine the effects on
live birth rates of low-molecular weight heparin (LMWH)
and low-dose aspirin (LDA) given in combination to pa-

Revised manuscript accepted for publication January 26, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3556.2017
750 H. Alptekin, N. Alptekin, R. Selimoğlu, T. Cengiz, S. Barış
tients with five commonly inherited thrombophilic poly-
morphisms and RPL. Age-matched healthy subjects with
no history of abortions or gestational complications were
selected as controls.
Materials and Methods
Patients registering two or more lost pregnancies (early or late
miscarriages or stillbirths) for no reason other than throm-
bophilia qualified for the study. All prior lost pregnancies were
verified by positive hCG (urine or serum) plus ultrasound or
uterine curettage with histologic confirmation. Pregnancy loss
was viewed as early (gestational week 13+6) or late (gestational
weeks 14–21+6) miscarriage, whereas stillbirth was equated
with pregnancy loss after 22 weeks of gestation. Infant survival
beyond the 28-day neonatal period constituted a live birth.
Exclusion criteria were extraneous presumptive etiologies as
follows: 1) abnormal blood karyotype in either partner; 2) lethal
fetal defect; 3) infectious disease during pregnancy; 4) known
erythroblastosis fetalis; 5) autoimmune disease (antiphospho-
lipid antibodies or idiopathic thrombocytopenic purpura); 6)
trauma during pregnancy; 7) tobacco consumption (≥ ten ciga-
rettes/day); 8) abnormal uterine anatomy by hysterosalpingog-
raphy, hysteroscopy, or uterine hydrosonography; 9) endocrine
disorder (thyroid dysfunction, hyperprolactinemia, luteal insuf-
ficiency, polycystic ovary syndrome, or diabetes mellitus); 10)
obesity; 11) single abortion; 12) history of epilepsy; 13) renal
or hepatic insufficiency; and 14) thrombocytopenia. In fact, any
medical disease with a potential to impact the outcome of preg-
nancy was grounds for exclusion.
Ultimately, 106 out of 183 females suffering RPL and diag-
nosed with inherited thrombophilia were recruited in the study.
Sixty-two healthy, age-matched women with one or more suc-
cessful pregnancies, no gestational complications (intrauterine
growth restriction, stillbirth, or abruptio placentae), and no abor-
tions were enrolled in the study as control subjects. All partici-
pants were seen as outpatients in the Department of Obstetrics
and Gynecology at Mevlana University in Turkey, between 2012
and August 30, 2015
Patients in the study group were prescribed a combination of
subcutaneous LMWH (enoxaparin sodium 0.4 ml, 4,000 IU,
once daily) and oral LDA (100 mg/day). Once starting enoxa-
parin (early in the fourth week of amenorrhea after positive preg-
nancy test), patients injected their abdomens or shoulders
systematically, and platelets were checked at each weekly visit
for heparin-induced thrombocytopenia. All patients were given
folic acid tablets (400 mcg) daily until week 13 of gestation. Pa-
tients with threatened abortion symptoms such as vaginal bleed-
ing, lower abdominal pain, and subchorionic hematoma at
ultrasonography also received oral micronized progesterone
(100 mg, twice daily until week 12 of gestation). Physical ex-
aminations were carried out during the first visit to determine
body mass index (BMI) and blood pressure. Patients were mon-
itored at weeks 6, 8, 11, 14, 18, 24, 28, 32, 36, and 38 of gesta-
tion. During antenatal visits, patients underwent routine obstetric
ultrasound and laboratory investigations. LDA was abandoned at
week 36 of gestation, but enoxaparin was continued until the
first signs of labor.
To qualify for the study, other potential etiologic factors were
excluded. A total of 168 participants were tested for five throm-
bophilic mutations: FVL (G1691A), FII (A20210G), PAI-1
4G/5G, and MTHFR (C677T and A1298C). All tests for an-
tiphospholipid antibodies were negative. A blood DNA purifica-
tion kit was utilized to extract genomic DNA from
EDTA-anticoagulated whole blood samples. Real-time poly-
merase chain reaction (PCR) conducted was used to determine
thrombophilic genotypes other than PAI-1. PAI-1 4G/5G I/D was
assayed using “allele-specific amplification” PCR (Muetze et al.)
based on internal control primers specific for the 4G allele (for-
ward: 5’-TGC AGC CAG CCA CGT GAT TGT CTA G-3’; re-
verse: 5’-AAG CTT TTA CCA TGG TAA CCC CTG GT-34G and
5’- GTC TGG ACA CGT GGG GA-3’) and for the 5G allele (for-
ward: 5’-TGC AGC CAG CCA CGT GAT TGT CTA G- 3’, re-
verse: 5’-AAG CTT TTA CCA TGG TAA CCC CTG GT-3’, and
5’-GTC TGG ACA CGT GGG GG-3’) [9]. As such, females car-
rying these specific alleles (heterozygous or homozygous) were
considered to have inherited thrombophilia.
The primary measure of outcome was live birth rate, defined
as survival beyond the 28-day neonatal period. Preeclampsia,
abruption placentae, premature delivery (at 24–37 weeks), gesta-
tional age at birth (weeks), birth weight (grams), intrauterine
growth restriction (birth weight < 2 SD), and adverse effects of
therapy, such as bleeding, thrombocytopenia (platelet count <
100,000 / dl), and primary postpartum hemorrhage (> 500 ml
blood loss), were assessed as secondary outcomes.
The prevalence of polymorphisms in test patients and control
subjects were compared using the two-tailed Fisher’s exact test
and a Chi-square test; logistic regression was applied to analyze
the effects of each polymorphism. The Pearson’s chi-square test
was instrumental, with significance set at p < 0.05. Only one out-
come, live birth rate, was investigated. No secondary outcomes
were statistically analyzed. Standard software was utilized for all
calculations (SPSS v20.0), expressing data as the mean ± SD.
Results
Of 168 participants, 106 had suffered RPL and had in-
heritable thrombophilia constituted study group; whereas
the remaining 62 patients who had successful delivery,
without history of abortion constituted the control group.
The average age of the subjects was 27.9 ± 5.7 (range, 18–
40) years in test patients and 29.4 ± 5.8 (range, 19–39)
years in control subjects, which did not differ significantly
(p = 0.1). In the test group, prior pregnancy losses and live
births totalled 232 (mean, 2.8 ± 1.3, range, 2–10) and 38
(single child, 27; two children, eight), respectively. Sev-
enty-one patients (66.9%) had no live children. Outcomes
prior to enrollment were poor, with live births in approxi-
mately 14% of pregnancies. Although 22 of the test patients
were not actually pregnant, thrombophilia panels were run
due to habitual abortion. These particular patients were ex-
cluded from the live birth rate calculations, because they
did not receive thromboprophylaxis. A history or early
pregnancy loss was recorded in 104 test patients (98.1%),
whereas 12 patients (11.3%) had suffered late pregnancy
loss and five patients (4.7%) had experienced stillbirths.
With respect to control subjects, 25 were nulliparous and 37
were multiparous. There were no miscarriages, and there
were a total of 97 live children.
The first pregnancies following treatment (one twin preg-
nancy and one with RPL linked to portal vein thrombosis)
in 73 of 84 patients (86.9%) with a history of RPL were live
births. No stillbirths/neonatal deaths or fetal anomalies were
 Inherited thrombophilia and thromboprophylaxis: a retrospective analysis of pregnancy outcomes in 106 patients 751

Table 1. — Patient characteristics and outcomes of pregnancies at time of referral and after thromboprophylaxis in test
females with recurrent pregnancy loss (RPL) and control subjects.
Characteristic RPL (n=106) Controls (n=62) p value
Mean ± SD Range n (%) Mean ± SD Range n (%)
At time of referral
Age 27.9±5.7 18–40 29.4±5.8 19–39 0.10
Mean number of EPLa 2.6±1.2 2–8 104 (98.1%) -
Mean number of LPLb 0.1±0.4 0–2 12 (11.3%) -
Mean number of stillbirths/ neonatal deaths 0.06±0.3 0–2 5 (4.7%) -
Total pregnancy lossc 2.8±1.3 2–10 -
Total live births 38 97
Live birth rate, unadjusted 38/270 (14%) 97/97 (100%)
≥ 4 previous losses 24 (22.6%) -
≥ 1 previous live birth 35 (33%) 62 (100%)
After thromboprophylaxisd
Live birth rate, unadjusted 73/84 (86.9%)
Stillbirth/neonatal death -
EPL 11/84 (13.1%)
LPL -
Gestational age (weeks) 38.1±2.5 34–41 82 (97.6%) 38.3±2.7 33–41 61 (98.3%) 0.90
Birth weight (g) 3224±362 3,245±351 0.70
a
Early pregnancy losses (EPL), bLate pregnancy losses (LPL), cAll pregnancy losses, d22 non-pregnant patients with a history of habitual abortions were excluded.

identified in any of the test patients given thromboprophy- Table 2. — Frequencies of genotypic polymorphisms in pa-
laxis. First-trimester abortions recurred in 11 patients tients with recurrent pregnancy loss (RPL) and control sub-
(13.1%), and one pregnancy was terminated due to cystic jects.
hygroma. Karyotyping of products of conception was per- Gene RPL Controls p value*
formed in seven of 12 miscarriages [normal, four; abnor- (n=106) (n=62)
mal, one (48 XY, +13, +15); failed cultures, two]. No FVL (G1691A)
significant differences were observed between birth weights Normal homozygous G/G 80 (75.4%) 56 (90.3%) 0.018
Heterozygous G/A 23 (21.6%) 4 (6.5%)
registered for test patients and controls (3224 ± 362.8 and
Mutant homozygous A/A 3 (2.8%) 2 (3.2%)
3245 ± 351.8 grams, respectively; p = 0.7). The details of FII (G20210A) (prothrombin)
demographics and outcomes of prior pregnancies are sum- Normal homozygous G/G 93 (87.7%) 60 (96.8%) 0.047
marized in Table 1. Heterozygous G/A 13 (12.3%) 2 (3.2%)
Analysis of DNA, isolated from peripheral blood sam- Mutant homozygous A/A 0 0
ples collected in EDTA and carried out specifically for MTHFR (C677T)
identifying FVL (G1691A), FII (G20210A), MTHFR Normal homozygous C/C 49 (46.2%) 37 (59.7%) 0.092
(C677T and A1298C), and PAI-1 4G/5G I/D polymor- Heterozygous C/T 43 (40.5%) 20 (32.3%)
phisms, is provided for all study participants (Table 2). Mutant Homozygous T/T 14 (13.2%) 5 (8.1%)
MTHFR (A1298C)
In binary comparisons, FVL (G1691A), FII (G20210A),
Normal homozygous A/A 26 (24.5%) 38 (61.3%) 0.000
PAI-1 4G/5G, and MTHFR (C677T and A1298C) differed Heterozygous A/C 65 (61.3%) 17 (27.4%)
significantly in test patients (vs. controls). Binary logistic Mutant Homozygous C/C 15 (14.1%) 7 (11.3%)
regression analysis identified PAI-1 4G/5G and MTHFR PAI-1 4G/5G I/D polymorphism
(C677T and A1298C) as independent correlates of habit- Normal 5G/5G 21 (19.8%) 30 (48.4%) 0.000
ual abortion. Heterozygous 4G/5G 57 (53.7%) 21 (33.9%)
Homozygous and/or heterozygous mutations were ob- Mutant Homozygous 4G/4G 28 (26.4%) 11 (17.7%)
served in all 102 test patients (54 homozygous mutations in *Homozygous and heterozygous groups analyzed together.
one or two thrombophilia panels) and in 58 of 62 control
subjects (23 homozygous mutations in one or two throm-
bophilia panels). However, three or more homozygous
and/or heterozygous mutations (in panel of five) were ob- patients who had live births, did not differ significantly (p
served in 48 test patients (45.3%), but in only three control > 0.05) (Table 3).
subjects (4.8%) (p < 0.05), constituting a significant dif-
ference. Coexistence of mutations in the 11 test patients, No heparin-induced thrombocytopenia or allergies were
for whom thromboprophylaxis failed and in the other 73 seen or recorded during the trial. Test patients experienced
752 H. Alptekin, N. Alptekin, R. Selimoğlu, T. Cengiz, S. Barış
Table 3. — Zygosity of factor V Leiden (G1691A), FII
(G20210A), PAI-1 4G/5G, and MTHFR (C677T and
A1298C) in patients with recurrent pregnancy loss (RPL)
and control subjects.
RPL (n=106) Controls (n=62) p value
Homozygosity (mutant)
None 52/106 (49.1%) 39/62 (62.9%) 0.211
1 48/106 (45.3%) 21/62 (33.9%)
2 6/106 (5.7%) 2/62 (3.2%)
Homozygosity and/or heterozygosity
None - 4/62 (6.5%) 0.000
1 17/106 (16%) 30/62 (48.4%)
2 41/106 (38.7%) 25/62 (40.3%)
3 32/106 (30.2%) 3/62 (4.8%)
4 12/106 (11.3%) - safe and effective means of venous thromboembolic pro-
5 4/106 (3.8%) - phylaxis/treatment during all stages of pregnancy [18].
minor vaginal bleeding during first, second, and third
trimesters, although such bleeding was largely a first
trimester event and was mild (11.3%), without any need for
blood transfusion. Subsequently, LMWH was discontinued
until bleeding ceased completely. Any skin reactions, bruis-
ing, and itching at injection sites were resolved by switch-
ing sites. None of the neonates suffered hemorrhagic
disease, intracranial hemorrhage, or thrombocytopenia.
Discussion
Acquired thrombophilia is a feature of antiphospholipid
syndrome (APS), which occurs as a primary or secondary
event in systemic lupus erythematosus. Diagnosis of APS
is based on the presence of LA, antiβ2GP1, and aCL anti-
bodies, and the use of LMWH/LDA together in this con-
text usually enhances both fetal and mother outcomes [10].
Although the pathogenic mechanisms of antiphospholipid
antibodies have been described [11], there is no real con-
sensus on the pathophysiology or treatment of inherited
thrombophilia. Once Dahlback et al. defined activated pro-
tein C (APC) resistance, Bertina et al. found that a single
mutation in the factor V gene (known as factor V Leiden)
was responsible. FVL is an autosomal-dominant mutation,
with 12–15% prevalence in various populations [12, 13].
The risk of thrombosis is three- to eight-fold higher with
heterozygous FVL status, increasing to 80-fold in ho-
mozygous states [13, 14]. APC resistance accounts for al-
most 50% of patients with inherited thrombophilia [12].
In a study conducted by Ivanov et al., the prevalence of
4G/5G I/D was substantially higher in females with RPL
(41.8%) vs. controls (26.8%) (OR=1.96, 95% CI: 1.05–
3.69; p = 0.034) [15]. Subrt et al. also showed that the PAI-
1 4G/4G homozygous genotype increases the risk of RPL,
independent of a positive antiphospholipid antibody test
[16]. The present study similarly identified MTHFR
(C677T and A1298C) and PAI-1 4G/5G mutations as in-
dependent correlates of RPL (OR = 2.41, 95% CI: 1.13–
5.12; p = 0.022; OR = 7.81, 95% CI: 3.49–17.4; p = 0.000,
and OR = 6.67, 95% CI: 2.88–15.41; p = 0.000, respec-
tively). According to Habibovic et al., FVL (G1691A), FII
(G20210A), MTHFR (C677T) mutations, and RPL share
no associations, but the combined mutations may be linked
to recurrent miscarriages [17].
The number of approaches for treating pregnant women
with known thrombophilia has evolved over the years with
varying success, including immunoglobulins, aspirin, and
glucocorticoids. Unfortunately, none of these treatments
have been very effective. Greer and Nelson-Piercy proved
that LMWH does not cross the placenta, which makes it a
Low-level LMWH as antithrombotic therapy was endorsed
at the Seventh American College of Chest Physicians
(ACCP) Conference on Antithrombotic and Thrombolytic
Therapy [19]. Without therapeutic intervention, the per-
centage of pregnancies resulting in live births is only 20–
28% [20]. No maternal and fetal serious side effects related
to combined LMWH/LDA were observed during the course
of the present study, and a live birth rate of 86.9% was
achieved. Prior to this trial, only 38 live births in 270 preg-
nancies (14%) were accrued in our test patients.
Mitic et al. achieved success in 29 of 38 pregnancies
through thromboprophylaxis, reflecting significant im-
provement (76%) over 81 prior pregnancy losses [21]. In
addition, Giancotti et al. reported that LMWH or LMWF
plus aspirin increased the rate of live births effectively in
patients with RPL and positive thrombophilic scans [22].
Taken together with the present results (i.e., 86.9% live birth
rate) in a broader group of patients, it appears that the com-
bination treatment of LMWH/LDA is a promising thera-
peutic approach for women with inherited thrombophilia
and RPL.
In the Live-Enox study, pregnancy outcomes and drug
safety were assessed in females with thrombophilia and his-
tories of RPL who were given enoxaparin (40 mg and 80
mg daily doses). Regardless of dose, outcomes and safety
proved equivalent, culminating in live birth rates of 84%
and 78%, respectively [20]. Use of LMWH early as throm-
boprophylaxis also brought success in reducing early and
late spontaneous abortions for at least 50% of patients in
another study [23]. Likewise, the rate of live births in-
creased significantly (by 9.76-fold) in females given
LMWH compared with placebo in the Qublan et al. study
(RR = 9.76, 95% CI: 1.31–72.86; p = 0.03) [24]. Finally,
when patients given aspirin only were compared with oth-
ers taking LMWH, a statistically significant increase in live
birth rate was documented by Gris et al. (RR = 3.0, 95% CI:
2.10–4.28; p < 0.00001) [25]. In all these trials, the use of
LMWH in pregnant women with inherited thrombophilia
proved superior to control interventions (placebo and as-
pirin) in terms live birth rates achieved.
 Inherited thrombophilia and thromboprophylaxis: a retrospective analysis of pregnancy outcomes in 106 patients
When Laskin et al. compared female patients given
LMWH and aspirin with other patients given aspirin alone,
they found no significant differences in live birth rates (RR
= 1.01, 95% CI: 0.8–1.26; p = 0.95) [26]. Their findings
were consistent with two other studies by Tan et al. [27]
and Kovalevsky et al. [28], neither of which reported sig-
nificant differences in treatment efficacies. In the present
test patients, however, thromboprophylaxis increased the
live birth rate significantly. Furthermore, three or more mu-
tations (homozygous and/or heterozygous) out of five in a
thrombophilia test panel coexisted at a significantly higher
rate in patients with RPL than in control subjects. PAI-1
4G/5G and MTHFR (C677T and A1298C) mutations in
particular were identified as independent variables, corre-
lating with habitual abortion. In patients with a history of
RPL, DNA screening for thrombophilia is therefore advis-
able. Infarction and impairment of chorionic villous vas-
cularization are putative risk factors in inherited
thrombophilia [29]. By initiating thromboprophylaxis at the
earliest opportunity, the chances of a healthy pregnancy
may increase. Indeed, the conflicting outcomes of these
studies are perhaps due to discrepancies in the onset of
treatment.
The benefit of thromboprophylaxis in this setting is
best evaluated using a control group where treatment is
withheld. For ethical reasons, however, this scenario is
unacceptable but nevertheless detracted from the present
study. On the same basis, it is impossible to avoid folic
acid and progesterone in patients with threatened abor-
tion co-therapies that are routinely used for antenatal
care. Hence, some participants were exposed to a multi-
plicity of therapies, in addition to anticoagulants, cloud-
ing final therapeutic assessment. Still, the present
comparison of live birth rates determined both before and
after thromboprophylaxis does provide some index of
treatment efficacy.
In conclusion, the data herein indicate that with three or
more mutations (homozygous or heterozygous) out of the
five that are frequently seen in inherited thrombophilia, es-
pecially PAI-1 4G/5G and MTHFR (C677T and A1298C),
the risk of RPL is heightened. As in acquired throm-
bophilia, LMWH/LDA given in combination to patients
with RPL and inherited thrombophilia may increase live
birth rates. A multicenter collaboration (i.e., randomized,
placebo-controlled study) is needed to corroborate the pres-
ent findings and establish an evidence-based treatment pro-
tocol.
Acknowledgments
The authors especially thank Drs. Fatih Kayhan and
Saniye Çimen for their technical assistance and dedication,
and for sharing their advice during the preparation of this
manuscript.This study was approved by the hospital re-
search ethics committee (26857650/023).
753
References
[1] Farquharson R.G., Jauniaux E., Exalto N.: “Updated and revised
nomenclature for description of early pregnancy events”. Hum. Re-
prod., 2005, 20, 3008.
[2] D'Uva M., Micco P.D., Strina I., Placido G.D.: “Recurrent pregnancy
loss and thrombophilia”. J. Clin. Med. Res., 2010, 2, 18.
[3] Sarig G., Younis J.S., Hoffman R., Lanir N., Blumenfeld Z., Brenner
B.: “Thrombophilia is common in women with idiopathic pregnancy
loss and is associated with late pregnancy wastage”. Fertil. Steril.,
2002, 77, 342.
[4] Visser J., Ulander V.M., Helmerhorst F.M., Lampinen K., Morin-
Papunen L., Bloemenkamp K.W., et al.: “Thromboprophylaxis for
recurrent miscarriage in women with or without thrombophilia.
HABENOX: a randomised multicentre trial”. Thromb. Haemost.,
2011, 105, 295.
[5] Kujovich J.L.: “Thrombophilia and pregnancy complications”. Am
J. Obstet. Gynecol., 2004, 191, 412.
[6] Kutteh WH.: “Antiphospholipid antibody-associated recurrent preg-
nancy loss: treatment with heparin and low-dose aspirin is superior
to low-dose aspirin alone”. Am. J. Obstet. Gynecol., 1996, 174, 1584.
[7] Rai R, Regan L.: “Obstetric complications of antiphospholipid anti-
bodies”. Curr Opin Obstet Gynecol., 1997, 9, 387.
[8] de Jong PG, Goddijn M, Middeldorp S.: “Antithrombotic therapy
for pregnancy loss”. Hum Reprod Update., 2013, 19, 656.
[9] Muetze S, Eggermann T, Leeners B, Birke C, Kuse S, Ortlepp JR, et
al.: “The 4G/5G polymorphism in the plasminogen activator in-
hibitor-1 gene is not associated with HELLP syndrome”. J. Thromb.
Thrombolysis, 2009, 27, 141.
[10] Bates S.M., Greer I.A., Middeldorp S., Veenstra D.L., Prabulos A.M.,
Vandvik P.O., et al.: “VTE, thrombophilia, antithrombotic therapy,
and pregnancy: Antithrombotic Therapy and Prevention of Throm-
bosis, 9th ed: American College of Chest Physicians Evidence-Based
Clinical Practice Guidelines”. Chest, 2012, 141, e691S.
[11] Nunez-Alvarez C.A., Cabiedes J.: “Pathogenic mechanisms of the
anti-phospholipid antibodies”. Reumatol. Clin., 2011, 7, 72.
[12] Dahlback B.: “Physiological anticoagulation. Resistance to activated
protein C and venous thromboembolism”. J. Clin. Invest., 1994, 94,
923.
[13] Bertina R.M., Reitsma P.H., Rosendaal F.R., Vandenbroucke J.P.:
“Resistance to activated protein C and factor V Leiden as risk factors
for venous thrombosis”. Thromb. Haemost., 1995, 74, 449.
[14] Rosendaal F.R., Koster T., Vandenbroucke J.P., Reitsma P.H.: “High
risk of thrombosis in patients homozygous for factor V Leiden (ac-
tivated protein C resistance)”. Blood, 1995, 85, 1504.
[15] Ivanov P., Komsa-Penkova R., Ivanov I., Konova E., Kovacheva K.,
Simeonova M., et al.: “Plasminogen activator inhibitor type 1 activ-
ity in women with unexplained very early recurrent pregnancy loss”.
Akush. Ginekol., 2010, 49, 3.
[16] Subrt I., Ulcova-Gallova Z., Cerna M., Hejnalova M., Slovanova J.,
Bibkova K., et al.: “Recurrent pregnancy loss, plasminogen activa-
tor inhibitor-1 (-675) 4G/5G polymorphism and antiphospholipid an-
tibodies in Czech women”. Am. J. Reprod. Immunol., 2013, 70, 54.
[17] Habibovic Z., Zeybek B., Sanhal C., Eroglu Z., Karaca E., Ulukus M.:
“Effects of inherited thrombophilia in women with recurrent preg-
nancy loss”. Clin. Exp. Obstet. Gynecol., 2011, 38, 347.
[18] Greer I.A., Nelson-Piercy C.: “Low-molecular-weight heparins for
thromboprophylaxis and treatment of venous thromboembolism in
pregnancy: a systematic review of safety and efficacy”. Blood, 2005,
106, 401.
[19] Bates S.M., Greer I.A., Hirsh J., Ginsberg J.S.: “Use of antithrom-
botic agents during pregnancy: the Seventh ACCP Conference on
Antithrombotic and Thrombolytic Therapy”. Chest, 2004, 126, 627S.
[20] Brenner B., Hoffman R., Carp H., Dulitsky M., Younis J., Investiga-
tors L-E.: “Efficacy and safety of two doses of enoxaparin in women
with thrombophilia and recurrent pregnancy loss: the LIVE-ENOX
study”. J. Thromb. Haemost., 2005, 3, 227.
[21] Mitic G., Novakov Mikic A., Povazan L., Mitreski A., Kopitovic V.,
754 H. Alptekin, N. Alptekin, R. Selimoğlu, T. Cengiz, S. Barış
Vejnovic T.: “Thromboprophylaxis implementation during preg-
nancy in women with recurrent foetal losses and thrombophilia”.
Med. Pregl., 2011, 64, 471.
[22] Giancotti A., La Torre R., Spagnuolo A., D'Ambrosio V., Cerekja A.,
Piazze J., et al.: “Efficacy of three different antithrombotic regimens
on pregnancy outcome in pregnant women affected by recurrent
pregnancy loss”. J. Matern. Fetal Neonatal Med., 2012, 25, 1191.
[23] Badawy A.M., Khiary M., Sherif L.S., Hassan M., Ragab A., Abdelall
I.: “Low-molecular weight heparin in patients with recurrent early mis-
carriages of unknown aetiology”. J. Obstet. Gynaecol., 2008, 28, 280.
[24] Qublan H., Amarin Z., Dabbas M., Farraj A.E., Beni-Merei Z., Al-
Akash H., et al.: “Low-molecular-weight heparin in the treatment of
recurrent IVF-ET failure and thrombophilia: a prospective random-
ized placebo-controlled trial”. Hum. Fertil., 2008, 11, 246.
[25] Gris J.C., Mercier E., Quere I., Lavigne-Lissalde G., Cochery-Nou-
vellon E., Hoffet M., et al.: “Low-molecular-weight heparin versus
low-dose aspirin in women with one fetal loss and a constitutional
thrombophilic disorder”. Blood, 2004, 103, 3695.
[26] Laskin C.A., Spitzer K.A, Clark C.A., Crowther M.R., Ginsberg J.S.,
Hawker G.A., et al.: “Low molecular weight heparin and aspirin for
recurrent pregnancy loss: results from the randomized, controlled
HepASA Trial”. J. Rheumatol., 2009, 36, 279.
[27] Tan W.K., Lim S.K., Tan L.K., Bauptista D.: “Does low-molecular-
weight heparin improve live birth rates in pregnant women with
thrombophilic disorders? A systematic review”. Singapore Med J.,
2012, 53, 659.
[28] Kovalevsky G., Gracia C.R., Berlin J.A., Sammel M.D., Barnhart
K.T.: “Evaluation of the association between hereditary throm-
bophilias and recurrent pregnancy loss: a meta-analysis Arch. Intern.
Med., 2004, 164, 558.
[29] Said J.M., Higgins J.R., Moses E.K., Walker S.P., Borg A.J., Mona-
gle P.T., et al.: “Inherited thrombophilia polymorphisms and preg-
nancy outcomes in nulliparous women”. Obstet. Gynecol., 2010, 115,
5.
Corresponding Author:
H. ALPTEKİN, M.D.
Department of Obstetrics and Gynecology
Mevlana University Faculty of Medicine
Konya (Turkey)
e-mail: alptekinhusnu74@hotmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Is there a relationship between maternal blood type and

the incidence of gestational diabetes mellitus?
A retrospective review

A.M. Oraif, H.A. Jabar, M. Ashi, H. Al Ghanmi, A.R. Al Jarallah

Department of Obsterics and Gynecology, King Abdulaziz University, Jeddah (Saudi Arabia)

Summary
Background: Gestational diabetes mellitus (GDM) is a widely common condition that is defined as glucose intolerance of relatively dif-
ferent degrees, and it affects pregnant women. In a recent study performed in Turkey, the authors found a higher risk of GDM for the pa-
tients with blood group AB. Aim: This study was done to detect the association of blood group type and the incidence of GDM. Materials
and Methods: A retrospective study was carried out in 2016 in a group of GDM patients at King Abdulaziz University Hospital (KAUH).
Patients were identified using the electronic medical records’ system. Results: The percentages of patients with GDM for O, A, B, and AB
groups were 43.8%, 33.7%, 16.3%, and 6.2%, whereas those of control group (healthy donors) were 38.20%, 33.90%, 24.9%, and 3%, re-
spectively. In both groups, the ratio of the patients with blood group O was the highest, while the ratio of group AB was the lowest. Blood
group AB was found to be higher in the patients with GDM compared to the control group. Conclusion: Women with AB blood group might
have a higher risk of developing GDM. More studies are needed to confirm the finding in this study.

Key words: Gestational diabetes mellitus (GDM); Maternal blood type; Glucose intolerance.

Introduction
Gestational diabetes mellitus or (GDM) is a widely com-
mon condition defined as glucose intolerance of relatively
different degrees, and it specifically affects pregnant
women [1]. It is a significant complication of pregnancy
that carries a high risk of comorbidity or mortality to the
pregnant woman and her baby. GDM associated with in-
creased incidence of various conditions, such as pre-
eclampsia, hypertension, and the chance of developing
overt diabetes mellitus (DM) later in life [2].
The pathophysiology of GDM may include insulin re-
sistance or pancreatic β-cell dysfunction, as in late preg-
nancy, the requirements of insulin increases to meet the
high metabolic demands of the mother. In comparison to
healthy women, GDM patients show a consistent weak in-
sulin response to nutrients and glucose specifically [3].
Screening of GDM can be done using different methods,
includes the 50-gram oral glucose challenge test (OGCT)
which is the most used screening method for GDM [4].
Current updated screening method from the International As-
sociation of the Diabetes and Pregnancy Study Groups
(IADPSG) recommends that the patient should start with fast-
ing glucose test at first prenatal visit, followed by a two-hour
75-gram OGCT at 24 and 28 weeks gestational age when in-
dicated [5]. Moreover, the diagnosis of GDM made when
the one or more glucose values fall at or above the specified
thresholds [1]. On the other hand, ABO blood group stud-
ies confirm that there are no known diseases that may result
from lacking the expression of ABO antigen, but the sus-
ceptibility to some diseases are found to be interrelated to
patients ABO phenotype. Correlations such as the obser-
vation that gastric cancer is more common in group A in-
dividuals, whereas duodenal and gastric ulcers occur more
commonly among blood group O individuals, remain con-
flicting [6]. The aim of the present study was to detect the
potential relationship between developing GDM and blood
group.
Materials and Methods
A retrospective study was performed in a group of pregnant pa-
tients with GDM conducted at King Abdulaziz University Hospi-
tal (KAUH) from January 2014 until December 2015. Ethical
approval was obtained from King Abdulaziz University IRB and
the methods were carried out in accordance with the approved
guidelines.
By using the electronic medical records’ system, patients were
identified. Data collection included: personal data, serologically
determined blood group and Rh factor, obstetric history (parity),
and any known medical illnesses
Fifty-gram OGCT is performed routinely in the present hospi-
tal as follows: 50 grams of glucose is dissolved in 200 ml of water
and the patient is then asked to drink it in five minutes. After one
hour, a blood specimen is obtained and blood sugar levels are
tested by glucometer. If the blood sugar is greater than 140 mg, the

Revised manuscript accepted for publication April 4, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3652.2017
756 A.M. Oraif, H.A. Jabar, M. Ashi, H. Al Ghanmi, A.R. Al Jarallah

Table 1. — Comparison between GDM patients and control was a significant difference between the patients with GDM
group according to blood group. and control group in terms of distribution of ABO blood
ABO blood Control group, Patients with GDM p-value groups. Blood group AB was found to be higher in the pa-
group n (%) n (%) tients with GDM compared to the control group (p = 0.035).
O 89 (38.20%) 78 (43.8%) 0.035 Also there were no statistical differences in Rh factor dis-
A 79 (33.90%) 60 (33.7%)
tribution among GDM group and control group (Table 2).
B 58 (24.90%) 29 (16.3%)
AB 7 (3%) 11 (6.2%)
Total 233 (100%) 178 (100%) Discussion
The pathogenesis of GDM is not yet clear which has led
Table 2. — Comparison between GDM patients and control the present authors to think about the hypothesis of associ-
group according to blood group and Rh factor. ation of maternal blood group and the incidence of GDM.
ABO blood groups GDM patients Control group Unfortunately there are no sufficient research papers ad-
with Rh factor dressing this topic.
O Rh+ 74 (41.6%) 82 (35%) The present authors found in this study that patients with
O Rh− 4 (2.2%) 7 (3%) blood group AB have the highest risk of developing GDM,
A Rh+ 55 (30.9%) 74 (31.6%) which agrees with other studies from Turkey [1], Iran, and
A Rh− 5 (2.8%) 5 (2.1%) India [7], with an increase from 3% in control group to 6.2%
B Rh+ 26 (14.6%) 54 (23.1%)
in GDM patients. Although this does not agree with a study
B Rh− 3 (1.7%) 4 (1.7%)
AB Rh+ 8 (4.5%) 5 (2.1%) in Tianjin, China [8] and Thailand [9], as it appears that AB
AB Rh− 3 (1.7%) 2 (0.9%) blood group is a protective factor for GDM. Followed by O
Total 178 (100) 233 (100) blood group and A blood group.
The present authors found that blood group B is a protec-
tive factor for GDM, as there was a significant decrease
from 24.9% in the control group to 16.3% in GDM patients.
screening test is considered positive, and OGCT is used to confirm Rh factors were not associated with the development of
the diagnosis of GDM. GDM, which also agrees with the study from Turkey [10].
An initial blood sample was taken after eight to14 hours of fast- In contrast, another study states that patients with blood
ing and the patient was asked to drink 100 grams of glucose dis- group AB have increased risk of DM type 2 [1] therefore
solved in 200–400 ml water within five minutes. Blood samples we have to be more careful with screening and follow up of
were taken at one, two, and three hours. The plasma glucose con-
centration was considered normal if it was below 95 mg/dl (fasting), GDM and DM type 2 in these patients.
180 mg/dl (one hour), 155 mg/dl (two hours), and 140 mg/dl (three Not only do AB blood group patients have an increased
hours). A patient was considered to have GDM if two or more val- risk of GDM, but they also have an greater risk of increased
ues were met or exceeded. levels of serum urea and creatinine [7]; hence this suggests
The distribution of blood groups among the patients with GDM abnormal metabolic and endocrine changes in these patients.
were compared to a control group of a total of 233 healthy blood
donors who donated blood in Jeddah city (west coast of Saudi More studies are needed to study the association between
Arabia) in the year 2014. GDM and blood group because the latter is stable through-
Inclusion criteria: all pregnant women with GDM in King Ab- out life. Other risk factors should be screened in high-risk
dulaziz University hospital in the past two years. Exclusion crite- blood groups. Genetic studies are needed to clarify the as-
ria: all pregnant women with pre-existing DM or uncomplicated sociation between blood group and GDM.
pregnancy.
The Statistical Package for the Social Sciences (SPSS version
20) used to analyze data using (chi-square test). The frequency of Conclusion
occurrence of different variables calculated p value was less than
0.01. In conclusion, the present study found that there is a higher
potential susceptibility of patients with AB blood group to de-
velop GDM; hence from this perspective the present authors
Results
recommend that individuals with AB blood group should un-
A total of 178 patients were diagnosed with GDM. The dergo routine OGCT early in gestational age for the early de-
mean age of patients with GDM was 31.1 ± 5.85 years. The tection of GDM and prevention of unfavorable consequences.
percentages of patients with GDM for O, A, B, and AB
groups were 43.8%, 33.7%, 16.3%, and 6.2%, whereas
References
those of control group (233 healthy donors) were 38.20%,
33.90%, 24.9%, and 3%, respectively. In both groups, the [1] Hatice K., Abdulsamet E., Ozerhan O., Kubra E., Ali C.,Deniz A., et
al.: “The role of blood groups in the development of diabetes melli-
ratio of the patients with blood group O was the highest, tus after gestational diabetes mellitus”. Ther. Clin. Risk Manag.,
while the ratio of group AB was the lowest (Table 1). There 2015, 11, 1613.
 Is there a relationship between maternal blood type and the incidence of gestational diabetes mellitus? A retrospective review 757

[2] Angadi N., Raju V., Dakshayini B., Syed Z.: “Screening in high-risk
group of gestational diabetes mellitus with its maternal and fetal out-
comes”. Indian J. Endocrinol Metab., 2012, 16, S74.
[3] Risto K., Tapani R.: “Gestational diabetes: pathogenesis and conse-
quences to mother and offspring”. Rev. Diabet. Stud., 2008, 5, 194
[4] Donovan L., Hartling L., Muise M., Guthrie A., Vandermeer B., Dry-
den D.M.: “Screening tests for gestational diabetes: a systematic re-
view for the U.S. Preventive Services Task Force”. Ann. Intern.
Med., 2013, 159, 115.
[5] Werner E.F., Pettker C.M., Zuckerwise L., Reel M., Funai E.F., Hen-
derson J., Thung SF: “Screening for gestational diabetes mellitus:
are the criteria proposed by the international association of the Dia-
betes and Pregnancy Study Groups cost-effective?” Diabetes Care,
2012, 35, 529.
[6] Muhammad K., Hamid N., Narazah Y.: “Association of ABO blood
groups with diabetes mellitus”. Libyan J. Med., 2010, 8, 5.
[7] Seyfizadeh N., Yousefi B.: “Is there association between ABO blood
group and the risk factors of unfavorable outcomes of pregnancy?”
J. Matern. Fetal Neonatal Med., 2014, 17, 1.
[8] Zhang C., Li Y., Wang L., Sun S., Liu G., Leng J., et al.: “Blood
group AB is protective factor for gestational diabetes mellitus: a
prospective population-based study in Tianjin, China”. Diabetes
Metab. Res. Rev., 2015, 31, 627.
[9] Phaloprakarn C., Tangjitgamol S.: “Maternal ABO blood group and
adverse pregnancy outcomes”. J. Perinatol., 2013, 33, 107.
[10] Kanat M., Goksugur SB., Ozlu T., Tunckale A., Ozturk B., Ozturk
FY., et al.: “The effect of feto-maternal blood type incompatibility on
development of gestational diabetes mellitus”. Clin. Ter., 2014, 165,
e145.
Corresponding Author:
A.M. ORAIF, M.D.
King AbdulAziz University
P.O. Box 40915
Jeddah 21511 (Saudi Arabia)
e-mail: ayman_oraif@yahoo.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Clinical value of transfontanellar ultrasonography

for neonatal insular development

X.K. Chen1, S.H. Chen2, G.R. Lv2, J.H. You2, Z.K. Chen1
1 Department of Ultrasonography, Children’s Hospital of Fudan University Xiamen Branch, Xiamen Children’s Hospital, Xiamen, Fujian
2 Department of Ultrasonography, Second Affiliated Hospital of Fujian Medical University, Quanzhou, Fujian (China)

Summary
Objectives: The aims of this study were to assess the morphological characteristics and to establish ultrasonographic standards of
normal neonatal insula size using transfontanellar ultrasonography, and to evaluate the clinical value of this technique. Materials and
Methods: The authors performed transfontanellar ultrasonography in 481 single-birth cases at 28-43 weeks’ gestation. Ultrasono-
graphic examinations were performed in the parasagittal plane at the level of the insula through the anterior fontanelle, measuring
area, and perimeter of the insula. Regression analyses were used to evaluate the relationship between insula size and gestational age
(GA), and 60 cases were randomly selected for assessment of intra-observer and inter-observer reliability of ultrasonographic meas-
urements. The authors obtained standard values of normal insula size and used them to assess and follow-up 40 cases with suspected
insular malformations. Additionally, 30 late-onset neonates who were determined as small for gestational age (SGA) and 45 normally
growing neonates were examined and tested using the Neonatal Behavioral Assessment Scale (NBAS). Results: The neonatal insula
appeared as an inverted triangle, with the insular gyri extending radially in an anterior-inferior to posterior-superior direction. The
area and perimeter of the normal neonatal insula significantly increased with GA (p < 0.01 for both), and these measurements were
highly reliable. This assessment of cases with suspected insular malformation showed that five out of 40 cases presented abnormal-
ities. Late-onset SGA neonates presented a significantly smaller area and perimeter in the insula compared to controls. In addition,
the measured values of the insula significantly correlated with NBAS scores. Conclusion: Evaluation of the neonatal insula using
transfontanellar ultrasonography can be performed and is clinically useful.

Key words: Insula development; Late-onset neonates; Gestational age; Insular abnormalities.

Introduction matter in preterm infants was smaller. Although the use of

The insula, or insular lobe, is located deep within the lat-
eral fissure and accounts for only a small volume of the
cerebral hemispheres. The insula is covered by the oper-
cula, which are part of the frontal, parietal, and temporal
lobes [1]. The anterior, superior, and inferior limiting sulci,
which are located on the medial surface of the fronto-or-
bital, frontoparietal, and temporal opercula, separate the in-
sula from the opercular cortex. The insular lobe, which is
shaped like an inverted pyramid, is composed of a number
of gyri. The insula is involved in sensory, motor, cognitive,
emotive, and visceral functions [2-9]. Obsessive-compul-
sive disorder, epilepsy, Parkinson’s disease, anxiety, drug
addiction, and other neuropsychiatric disorders are associ-
ated with insular abnormalities [10-14]. Recent studies per-
formed using MRI and functional MRI (fMRI) have
provided information on the function of the insula. Gou-
sias et al. [15] used three-dimensional reconstructions of
MRI scans to measure the volume of the insula. By com-
paring the insular volumes of preterm infants with in-
trauterine growth restriction to those of normal-term
infants, Padilla et al. [16] found that the grey and white
MRI has obvious advantages for the assessment of the
brain, cranial ultrasonography is a convenient, inexpensive,
and applicable technique for bedside consultation, which
makes it invaluable for the screening of neonatal cerebral
diseases. Prenatal screening by ultrasonography can be
used to examine abnormalities in multiple brain regions and
structures such as the cerebellum, corpus callosum, sulci,
gyri, and ventricles. Brain gyri can be observed in the 17th
week of pregnancy, while the cingulate sulcus can be ob-
served in the 24th week of pregnancy. Govaert et al. [17],
who first described the use of cranial ultrasonography to
examine the insula, expounded the performance of this
technique for the examination of normal and abnormal
anatomy, but a quantitative index is still not available. Cur-
rently, there are few studies addressing normal develop-
ment of the insula in subjects that are normal for gestational
age (GA). The aims of this study were to assess the mor-
phological characteristics and to establish ultrasonographic
standards of normal neonatal insula size using trans-
fontanellar ultrasonography, and to evaluate the clinical
value of this technique.

Revised manuscript accepted for publication May 12, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3743.2017
 X. Chen, S. Chen, G. Lv, J. You, Z. Chen 759

Materials and Methods

This study was performed at the Department of Ultrasonogra-
phy of the Second Affiliated Hospital of Fujian Medical Univer-
sity from June 2012 to December 2013. The study protocol was
approved by the Ethics Committee at the Second Affiliated Hos-
pital of Fujian Medical University, and written consent was ob-
tained from the parents or legal guardians prior to enrollment in
the study.
Using systematic sampling, 481 neonatal pediatric inpatients
(285 males and 196 females) were selected for this study. The in-
clusion criteria were as follows: 1) mothers with regular menstru-
ation who knew the exact date of their last menstrual period, 2) a
singleton pregnancy with GA between 28 and 43 weeks and con-
firmed by ultrasonic estimation of the crown-rump length early in
the first trimester, 3) no medical history of congenital central nerv-
ous system disorders, 4) no other risks or medical history that
might cause fetal central nervous system diseases, and 5) the GA
estimated using ultrasonography had to be consistent with the ges- Figure 1. — Triangular shape of the insula lobe is seen in
tational weeks calculated according to the menstrual history. In ad- parasagittal insula plane at 39 gestational weeks.
dition, a sample of 75 singleton neonates, including 30 neonates
who were classified as late-onset small for gestational age (SGA)
and 45 appropriate for gestational age (AGA) infants were tested
using the Neonatal Behavioral Assessment Scale (NBAS). These All procedures followed were in accordance with the ethical
singleton neonates were between 34-37 weeks’ gestation. standards of the responsible committee on human experimenta-
All examinations were carried out using an ultrasonographic
tion (institutional and national) and with the Helsinki Declaration
machine equipped with a convex array probe with a frequency of
of 1975, as revised in 2008 (5). Informed consent was obtained
3.5 MHz, and an ultrasonographic machine equipped with a 7.5
MHz linear probe. from all patients for being included in the study.
All neonates were examined three to seven days after birth
using transfontanellar ultrasonography. The authors attempted to
maintain the newborn in a quiet state in order to avoid interfer- Results
ences during examination, while the probe was placed on the an- The neonatal insula is located on the facies medialis of
terior fontanelle. After performing routine scanning in the coronal the temporal lobe and deep within the lateral fissure. The
and parasagittal planes to exclude intracranial lesions, the mor-
phological features of the insula in the parasagittal plane were ex- insular lobe, which looks like an inverted triangle, is sepa-
amined. All images were saved to the device for off line rated from the surrounding brain lobes by the anterior, su-
measurements. The perimeter (cm) and area (cm2) of the insula perior, and inferior limiting sulci. The insular gyri are
were measured, and the same doctor obtained three separate meas- petal-shaped and extend radially in an anterior-inferior to
urements and averaged them (Figure 1). In addition, the authors posterior-superior direction. The central sulcus of the in-
identified fetal perimeter and area of insula at a specific GA (28
to 43 weeks) with a regression model. sula, which divides it into a larger anterior and a smaller
For the reliability analysis, 60 cases were collected by system- posterior insula, is clearly seen in the majority of newborns.
atic sampling. Each of the observers performed two consecutive The anterior insular lobe, which is associated with the
image collections from each subject, and images were saved for frontal lobe, is mainly composed of three short gyri. The
offline analysis. The measurements that were conducted by X.K. posterior insula, which is in close contact with the tempo-
Chen were used to evaluate intra-observer repeatability. The
measurements that were obtained by X.K. Chen and S.H. Chen
ral lobe, is mainly composed of the anterior long insular
were used to evaluate inter-observer reliability. The method of gyrus and the posterior long insular gyrus (Figure 2).
measurement was the same as that described earlier. Figures 3 and 4 show the relation between the area and
The authors screened an additional 40 newborns with suspected perimeter of the neonatal insula and gestational age in
insular malformations to evaluate if the established criteria were weeks. The area and perimeter of the normal neonatal in-
helpful in clinical diagnosis. Additionally, 30 SGA and 45 AGA
sula increased with GA. The regression equations for the
neonates were respectively screened using the above method and
evaluated by NBAS.SPSS 17.0 software package that was used to insular area and perimeter were as follows:
analyze the measurement data of normal newborns of different
GAs in order to determine the normal reference values of the in- Area (cm²) = 2.28 − 0.307GA + 0.011GA²
sular area and perimeter, which were expressed as mean ± stan- Perimeter (cm) = -2.85 + 0.316GA + 0.001GA²
dard deviation. The relation between measured values and
gestational weeks was determined using regression analysis. Data
for AGA and SGA newborns measured by two different physi- The intraclass correlation coefficient and its 95% confi-
cians were analyzed for repeatability and consistency using the dence intervals for the measurement of the area and perime-
intraclass correlation coefficient and the Bland-Altman method. P ter of the neonatal insula and the Bland-Altman analysis of
values less than 0.05 were considered statistically significant. Stu- measurements of the area and perimeter of the insula per-
dent’s t-tests for independent samples and Pearson’s χ2 tests were formed by different physicians is shown in Figures 5a and
used to compare the quantitative and qualitative data, respectively.
760 Clinical value of transfontanellar ultrasonography for neonatal insular development

Figure 2. — Parasagittal plane of insula at 40 gestation weeks. 1a Figure 3. — The area of the neonatal insula against gestational
and 1b: short gyri of insula; 2: long gyrus of insula; 3: central sul- age.
cus of insula; 4: limen of insula; 5: limiting sulci; 6: frontal lobe;
7: temporal lobe.

Figure 4. — The perimeter of the neonatal insula against gesta-

tional age.

5b. The repeatability of these measurements was very high,

regardless of whether the measurements were performed
by the same or by different physicians. As observed in the
figures, these measurements were very consistent.
In the present study, the secondary gyri of the insula were
almost visible at 28 weeks’ gestation, and became clearly
visible at 34 weeks (Figure 6). However the abnormal mor-
phology of the neonatal insula appeared as long irregular
strips without gyrus which clearly delineated the insula. In
addition, the area and perimeter of the insula were at least
two standard deviations less than the corresponding values Figure 5. — A: Bland-Altman Plots of the area of insula measured
in newborns of normal GA. Five cases had insular abnor- by different physicians; B: Bland-Altman Plots of the perimeter of
malities. In addition to insular abnormalities, three new- insula measured by different physicians.
 X.K. Chen, S.H. Chen, G.R. Lv, J.H. You, Z.K. Chen 761

Figure 6. — Development of insular gyration in insular plane. The secondary gyri of the insula are almost visible at the 28th week of
gestation, and they become much clearer at 34 and 40 weeks gestation.

Figure 7. — Ultrasono-
gram of the newborn with
abnormal insula associ-
ated with severe hydro-
cephalus at 38 gestation
weeks. a) (arrows) Insula
shows a small unclear tri-
angle lacking secondary
gyri in insular plane. (b)
Combined with severe di-
latation of the lateral ven-
tricles. LV: lateral
ventricles. c) (arrows)
The cerebellum with
clear dysplasia. (d) The
cisterna magna is se-
verely expand. CM: cis-
tema magna.

borns also exhibited bilateral ventricular enlargement with Discussion

severe ventricular dilation, expansion of the cavity of the
septum pellucidum, and unclear cerebellar structure. Fur-
thermore, one newborn showed slight dilation of the lateral
ventricles, two presented leukomalacia near the anterior
horn of the left ventricle, one had leukomalacia near the an-
terior horn of both lateral ventricles, and two showed agen-
esis of the cerebellar vermis (Figure 7).
The present results showed that SGA and AGA neonates
exhibited very different perimeters and areas of the insula.
In addition, NBAS scores of SGA neonates were signifi-
cantly lower than those of AGA newborns (Table 1).
Overall, the size of the neonatal insula increased with
GA, and the area and perimeter were parameters with a
high degree of reliability. Using these values the present
authors found five cases of insular abnormalities. An ab-
normally developed insular lobe can appear as a small tri-
angle lacking secondary gyri, an area lacking the normal
triangular shape, or an abnormal secondary gyrus that is
supported by the cerebral lateral fissure [17]. Because the
insular lobe is composed of gyri, and is surrounded by three
limiting sulci, the abnormal development of sulci and gyri
may therefore influence the size of the insula. Previous
studies have reported that insular abnormalities are com-
mon in polymicrogyria syndrome, dilated lateral ventricles,
762 Clinical value of transfontanellar ultrasonography for neonatal insular development
Table 1. — Perinatal outcome of the study groups.
SGA (n=30) AGA (n=45) p before 28 weeks are still needed. Moreover, it will be nec-
GA at birth (weeks) 35.6 ± 1.4 36.3 ± 1.1 0.58
Birthweight (grams) 1689 ± 236 1863 ± 268 < 0.01
Male gender 51.5% 48.9% 0.68
NBAS 40.6 ± 2.5 45.3 ± 2.3 0.04
Area of insular 419.2 ± 36.4 469.7 ± 49.5 < 0.01
Perimeter of insular 86.1 ± 4.9 93.5 ± 5.2 < 0.01
glutaricacidemia type II, and other diseases and etiologies
[17]. In cases of serious ventricular dilatation, the brain
parenchyma may be compressed and sometimes is not de-
tected. In addition to compression factors, serious ventric-
ular dilation with retardation of the development of cerebral
sulci and gyri may be related to other factors. For example,
a fetus with ischemic cerebrovascular disease with dilated
lateral ventricles often exhibits cerebral abnormalities such
as agyria and necrotic lesions such as periventricular leuko-
malacia [18]. Furthermore, intrauterine growth restriction
affects the development of the fetus and its brain. Dyspla-
sia of the insula in the present five cases were associated
with the presence of severely dilated ventricles or leuko-
malacia. In addition, this comparative assessment of 30
SGA and 45 AGA newborns showed that they presented
different insula size (i.e., perimeter and area). These find-
ings are in agreement with previous MRI studies [19]. Pre-
vious studies have shown that SGA neonates had a
significantly thinner and smaller cortex compared to AGA
newborns. Previous reports on the neurodevelopmental out-
comes of late-onset SGA infants have shown decreased at-
tention, sociability, self-regulation, communication,
problem-solving, and memory function scores [20-22]. All
these functions are closely related to the insula and the lim-
bic system [8]. Egaña-Ugrinovic et al. found significant as-
sociation between insular size and neurobehavioral
parameters in the NBAS test. In term SGA infants, the thin-
ner and smaller insular were, the worse the neurobehavioral
outcomes were [18]. Accordingly, in the present study,
NBAS scores of SGA infants were significantly lower than
those of AGA newborns, supporting that insular abnormal-
ities affect newborn neural function. The cause of SGA is
still unclear, although it is possible that pathogenic factors
such as chronic intrauterine hypoxia originate from the
mother, the fetus, or the placenta. SGA is the main compli-
cation in pregnant women with high blood pressure, dia-
betes, or cardiovascular disease. Through observations of
the neonatal insula at 28-43 weeks’ gestation, the present
authors found that each sulcus and gyrus of the insular lobe
was visible using transfontanellar ultrasonography after 34
weeks’ gestation, and the area and perimeter of the insula
positively correlated with increases in GA.
The present study had some limitations. Although the
present normal reference values of neonatal insular size
covered a large majority of GAs, reference ranges for GAs
essary to replicate the present findings in future studies with
a larger sample size for each GA. In addition, further re-
search is needed to explore the effects of Doppler ultra-
sonography on insular blood flow.
Although CT and MRI are more frequently used in the di-
agnosis of brain diseases, these methods have a number of
problems. They are expensive, radioactive, and extremely
inconvenient to use in serious cases and in children who
are not easy to calm. Transfontanellar ultrasonography has
the advantages of being non-invasive, quick, and conven-
ient. It provides an effective means for observing neonatal
insular development, and it can better track and assess
neonatal insular development over time.
Acknowledgments
This study was supported by grants from the Class B Pro-
gram of Education Committee of Fujian Province, Peoples
Republic of China (NO. JB12102) and the Science and
Technology Program of Quanzhou City, Fujian Province,
and Peoples Republic of China (NO. 2013Z101).
References
[1] Nieuwenhuys R., Voogd J., van Huijzen C.: “The Human Central
Nervous System”. Berlin: Springer Verlag, 1988, 13.
[2] Etkin A., Wager T.D.: “Functional neuroimaging of anxiety: A meta-
analysis of emotional processing in PTSD, social anxiety disorder,
and specific phobia”. Am. J. Psychiatry, 2007, 164, 1476.
[3] Stein M.B., Simmons A.N., Feinstein J.S., Paulus M.P.: “Increased
amygdala and insula activation during emotion processing inanxi-
ety-prone subjects”. Am. J. Psychiatry, 2007, 164, 318.
[4] Cauda F., D’Agata F., Sacco K., Duca S., Geminiani, G., Vercelli
A.: “Functional connectivity of the insula in the resting brain”. Neu-
roimage, 2011, 55, 8.
[5] Pugnaghi M., Meletti S., Castana L., Francione S., Nobili L., Mai
R., et al.: “Features of somatosensory manifestations induced by in-
tracranial electrical stimulations of the human insula”. Clin. Neuro-
physiol., 2011, 122, 2049.
[6] Stephani C., Fernandez-Baca Vaca G., Maciunas R., Koubeissi M.,
Lüders H.O.: “Functional neuroanatomy of the insular lobe”. Brain
Struct. Funct., 2011, 216, 137.
[7] Nieuwenhuys R.: “The insular cortex: A review”. In: Hofman M.A.,
Falk D. (eds). Progress in brain research. Amsterdam: Elsevier,
2012, 123.
[8] Augustine J.R.: “Circuitry and functional aspects of the insularlobe
in primates including humans”. Brain Res. Brain Res. Rev., 1996,
22, 229.
[9] Craig A.D.: “Interoception and emotion: A neuroanatomicalperspec-
tive”. In: Lewis M., Haviland-Jones J.M., Barrett L.F. (eds). Hand-
book of emotions. 3rd ed. New York: Guilford Publications, 2008, 272.
[10] Song A., Jung W.H., Jang J.H., Kim E., Shim G., Park H.Y., et al.:
“Disproportionate alterations in the anterior and posterior insular
cortices in obsessive-compulsive disorder”. PLoS One, 2011, 6,
e22361.
[11] Naqvi N.H., Bechara A.: “The insula and drug addiction: an intero-
ceptive view of pleasure, urges, and decision-making”. Brain Struct.
Funct., 2010, 214, 435.
[12] Vollstädt-Klein S., Loeber S., Kirsch M., Bach P., Richter A., Büh-
ler M., et al.: “Effects of cue-exposure treatment on neural cue re-
 X.K. Chen, S.H. Chen, G.R. Lv, J.H. You, Z.K. Chen
activity in alcohol dependence: a randomized trial”. Biol. Psychiatry,
2011, 69, 1060.
[13] Song X.G., Li C.F., Hu L., Ma N., Wang J., Li N. et al.: “Effect of
acupuncture on heroin cue-induced functional magnetic resonance
in heroin-addicted human subjects”. Zhen Ci Yan Jiu, 2011, 36, 121.
[14] Barrós-Loscertales A., Garavan H., Bustamante J.C., Ventura-Cam-
pos N., Llopis J.J., Belloch V., et al.: “Reduced striatal volume in
cocaine-dependent patients”. Neuroimage, 2011, 56, 1021.
[15] Gousias I.S., Rueckert D., Heckemann R.A., Dyet L.E., Boardman,
J.P., Edwards A.D., et al.: “Automatic segmentation of brain MRIs
of 2-year-olds into 83 regions of interest”. Neuroimage, 2008, 40,
672.
[16] Padilla N., Falcón C., Sanz-Cortés M., Figueras F., Bargallo N.,
Crispi F., et al.: “Differential effects of intrauterine growth restric-
tion on brain structure and development in preterm infants: A mag-
netic resonance imaging study”. Brain Res., 2011, 1382, 98.
[17] Govaert P., Swarte R., De Vos A., Lequin M.: ”Sonographic ap-
pearance of the normal and abnormal insula of Reil”. Dev. Med.
Child. Neurol., 2004, 46, 610.
[18] Chen X.K., Lv G.R., Lin H.T.: “Observation on the development of
fetal cerebral sulci by prenatal ultrasonography”. Chin. Ultrasonogr.,
2009, 2, 149.
[19] Egaña-Ugrinovic G., Sanz-Cortes M., Figueras F., Couve-Perez C.,
Gratacós E.: “Fetal MRI insular cortical morphometry and its asso-
763
ciation with neurobehavior in late-onset small-for-gestational-age
fetuses”. Ultrasound Obstet. Gynecol., 2014, 44, 322.
[20] Figueras F., Oros D., Cruz-Martinez R., Padilla N., Hernandez-An-
drade E., Botet F., et al.: “Neurobehavior in term, small-for-
gestational age infants with normal placental function”. Pediatrics,
2009, 124, e934.
[21] Eixarch E., Meler E., Iraola A., Illa M., Crispi F., Hernandez-Andrade
E., et al.: “Neurodevelopmental outcome in 2-year-old infants who
were small-for-gestational age term fetuses with cerebral blood flow
redistribution”. Ultrasound Obstet. Gynecol., 2008, 32, 894.
[22] Geva R., Eshel R., Leitner Y., Fattal-Valevski A., Harel S.: “Mem-
ory functions of children born with asymmetric intrauterine growth
restriction”. Brain Res., 2006, 1117, 186.
Corresponding Author:
X.K. CHEN, M.D.
Department of Ultrasonography
Children’s Hospital of Fudan University Xiamen Branch
Xiamen Children’s Hospital
No. 92-98 Yibin Road, Huli District
Xiamen, Fujian 361006 (China)
e-mail: xiaokanchen@126.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Arteriovenous malformations (AVM) of the corpus uteri

L. Roncati1,2, T. Pusiol2
1Department of Diagnostic and Clinical Medicine, University of Modena and Reggio Emilia, Modena
2Cervical Cancer Screening Center, Santa Maria del Carmine Hospital, Rovereto (Italy)

Summary
Introduction: The arteriovenous malformations (AVM) are sporadic lesions and they consist in a large number of tortuous, dilated,
and thick-walled vessels with different sizes. AVM can be divided into two distinct types, in relation to their site: the superficial type
and the deep type. Uterine AVM are precisely labelable as deep AVM and they are well-known in the gynaecological and radiological
practice, but rarely reported in the histopathological literature. Materials And Methods: To fill this gap, the authors have retrospectively
examined 25 cases of incidental uterine AVM from post-menopausal Caucasian women with a mean age of 68 years (range 63 and 74
years), who underwent hysterectomy with bilateral salpingo-oophorectomy for uterine prolapse. Surprisingly, in the anamnestic records,
all the patients suffered from dysmenorrhea during their life. Moreover, they have reviewed about 300 cases of uterine AVM, reported
in the gynaecological and radiological English literature, correlating them with their observations. Results: From their results, it emerged
that the use of the term ‘acquired’ to describe uterine AVM should be avoided, because all the present lesions show a malformative mor-
phology, related to a developmental disorder. Since no criteria exist to differentiate between AVM and placental bed sub-involution, a
descriptive nomenclature should be preferred in the radiologic terminology. Conclusion: AVM should be routinely remarked in the
histopathological reports, because their presence could be correlated with an explainable history of dysmenorrhea. Even if emboliza-
tion remains an acceptable form of treatment in order to avoid hysterectomy in those patients presenting with heavy life-threatening
bleeding, a vasoconstrictive therapy could be considered when other possible causes of disabling dysmenorrhea are excluded and the
presence of AVM at high flow has been ascertained by eco-colour Doppler.

Key words: Arteriovenous malformation (AVM); Uterus; corpus uteri; Histology; Dysmenorrhea; Vasoconstrictors.

Introduction Materials and Methods

The arteriovenous malformations (AVM) are infrequent
lesions and they consist of a large number of tortuous, di-
lated, and thick-walled vessels with different sizes, show-
ing at least a focal presence of the internal elastic lamina
[1]. The AVM can be divided into two distinct types, in re-
lation to their site. Deep AVM are commonly located in
head and neck region, limbs, gastrointestinal tract, central
nervous system or deep soft tissues. The superficial type, in
contrast, is usually located in the skin of face or neck
among middle-aged or elderly adults, it is much smaller,
and it is not associated with any appreciable circulatory dis-
turbance [1]. In clinical practice, the diagnosis of uterine
AVM can be achieved by radiological procedures. The le-
sions appear as hypoechoic winding spaces inside the my-
ometrium with a low impedance and a high flow on colour
Doppler examination [2, 3]. Angiography may be a further
diagnostic procedure to ascertain their presence [4], while
their embolization has been proposed as an effective and
safety therapeutic solution [5]. Here, the authors studied a
25-case series of deep AVM of the corpus uteri, correlating
their observations with a critical review of the terminology
in the medical literature, while providing also a possible
non-invasive therapeutic solution for symptomatic patients.
We enrolled in this study 25 cases of incidental uterine AVM
from post-menopausal Caucasian women with a mean age of 68
(range 63 and 74) years, who underwent hysterectomy with bilat-
eral salpingo-oophorectomy for uterine prolapse at the Santa
Maria del Carmine Hospital in Rovereto (Italy), from 2005 to
2015. During anamnesis, all the patients suffered from dysmen-
orrhea during their life. We applied clear and reproducible inclu-
sion criteria in the diagnostic phase. Firstly, we have considered
only AVM lesions characterized by thick-walled vessels, haphaz-
ardly arranged in myometrium and perimetrium. Secondly, the le-
sions had to involve more than the 75% of the myometrium, until
they reached the basal endometrium. The surgical specimens were
fixed in neutral-buffered formalin for at least 24 hours and then
paraffin embedded. The tissue section, obtained from the paraffin
blocks, were routinely stained with Haematoxylin and Eosin
(H&E).
Results
On macroscopic examination, the uterus measured 7×
4.5×3 cm on average, while the ovaries appeared slightly
enlarged in relation to age in every case. The most sig-
nificant finding was a red-brown appearance of the cor-
pus, which showed a myriad of vascular channels by
sagittal sectioning (no lesion involved the cervix uteri).
In four patients, small sub-serosal leiomyomas were
found. On microscopic examination, the myometrium

Revised manuscript accepted for publication May 8, 2017

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog4260.2017
 L. Roncati, T. Pusiol 765

Figure 1. — In a panoramic view, MAV appears as a large amount

of thick-walled engorged vessels of medium and large size anas-
tomosed to each other, surrounding the endometrium and involv-
ing more than the 75% of the myometrium and the perimetrium
(A, H&E, ×2.5). The vessels are arterial (circle) or venous (aster- Figure 2. — An exemplificative ultrasound scan of the uterus (U),
isks) in nature (B, H&E, ×5); they reach the basal endometrium performed by a 6.5-MHz transvaginal probe, which shows multi-
(arrows), which is atrophic (C, H&E, ×5). At higher magnifica- ple anechoic round and oval areas, 5 mm in maximum diameter
tion, the vascular structures exhibit degenerative changes, such as (arrows), into the myometrium at level of the uterine corpus/fun-
tunica media calcification (yellow arrows), intimal proliferation dus (D1 and D2).
with fibrosis (black arrows), and mucoid degeneration (blue ar-
rows) of the wall (D, H&E, ×10).

radiological reports can be noted. Insun et al. associated

AVM with massive operative bleeding in a 47-year-old pa-
tient and histologically documented irregular-shaped ves-
and both ovarian hila were affected by a diffuse prolifer-
sels grouped within the myometrium [7]. Chien et al.
ation of medium and large sized arteries and veins, in
described another case of uterine AVM rupture suggesting
close association, with a malformative nature (Figure 1,
that it can be the cause of copious and unexplained vaginal
panels A and B). The endometrium was atrophic and sur-
bleeding, sometimes difficult to treat [8]. Kasznica et al.
rounded by the underlying myometrial vessels (Figure 1,
reported in a 34-week stillborn female fetus the presence
panel C). The arteries showed striking degenerative
of vascular channels evenly distributed throughout the en-
changes, including intimal proliferation, fibrosis, and me-
tire myometrium and the basal endometrium [9]. In a hys-
dial calcific sclerosis, also called Mönckeberg’s arte-
terectomy specimen, Busmanis et al. described the
riosclerosis (Figure 1, panel D). The veins exhibited
histological features of AVM, associated with florid my-
intimal thickening too.
ometritis in a 67-year-old Chinese woman [10]. Ciani et al.
reported a haemorrhagic myometrial nodule of 20 mm in
Discussion diameter, in a 56-year-old woman with grade III uterine
prolapse. The lesion microscopically corresponded to a tan-
Uterine symptomatic AVM is associated with recurrent
gle of intermingled hyperplastic arteries and veins of inter-
pregnancy loss, menorrhagia, menometrorrhagia or abnor-
mediate size. The diagnosis was ‘acquired AVM with
mal heavy bleeding after invasive procedure. Its possible
massive endometrial stromal component’ [11]. Brown et al.
presence should be always considered in patients present-
described a case of uterine AVM in a 29-year-old woman
ing with abnormal heavy uterine bleeding and negative
with gross and microscopic documentation. The patient had
human chorionic gonadotropin values [6]. Moreover, the
been treated with methotrexate for non-metastatic gesta-
consequences of a curettage in case of undiagnosed AVM
tional trophoblastic disease [12]. To the present authors’
can be life-threatening [6]. In all patients of this series, hys-
knowledge, about 300 cases of AVM have been reported in
terectomy was performed for uterine prolapse and the AVM
gynaecological and radiological English literature, where
of the uterine corpus was an incidental finding. In the as-
they are curiously classified as congenital or acquired [13-
sessment of uterine AVM, an evident discrepancy between
15]. The first should be intended due to a disorder in the
few histological studies and numerous gynaecologic and
embryonic vascular development, while the second as the
766 Arteriovenous malformations (AVM) of the corpus uteri
result of a previous uterine tissue damage, such as tro-
phoblastic disease, incomplete miscarriage, caesarean sec-
tion, dilation and curettage, uterine infection, myomata,
endometriosis, and exposure to diethylstilboestrol [13-15].
From the present series, the term acquired AVM should be
avoided, because all the lesions show a malformative mor-
phology. They are in fact composed of distinctive thick-
walled vessels and not simple thin-walled capillaries, as
observed in course of AVM in other districts [16]. More-
over, they are haphazardly arranged in the myometrium
and perimetrium, involving more than the 75% of the my-
ometrium, until they reach the basal endometrium. Tim-
merman et al. conducted a prospective study regarding 30
consecutive patients with uterine AVM diagnosed by ul-
trasonography and eco-colour Doppler imaging [17].
Spectral analysis of the vessels showed the presence of low
impedance and high-velocity flow. Eight patients required
embolization of uterine arteries and three of them had true
AVM. In six cases, molar pregnancy was discovered and
embolization of uterine arteries was not necessary. Tro-
phoblast tissue was detected in six cases and unilateral em-
bolization of uterine arteries was performed in two cases.
The authors concluded that a conservative management is
possible in more than two-thirds of patients. Moreover, in
their study, no AVM without prior pregnancy was encoun-
tered in the examined patients [17]. This data supports the
present authors’ thesis, suggesting that many cases of
AVM, labelled as acquired, could simply represent a sub-
involution of the placental bed. At the present time, no cri-
teria exist to differentiate between AVM and placental bed
sub-involution, based on ultrasonography with colour
Doppler imaging. For this reason, a descriptive terminol-
ogy, such as ‘arteriovenous shunt’ or ‘arteriovenous fis-
tula’, should be preferred in the radiological practice
(Figure 2).
Conclusion
AVM should be always remarked in the histopatholog-
ical reports because their presence can be correlated with
an unexplainable history of dysmenorrhea, regardless of
past pregnancies. Therefore, when other possible causes
of dysmenorrhea are excluded and the presence of AVM
at high flow has been ascertained by eco-colour Doppler,
a vasoconstrictive therapy could be a viable approach in
those patients affected by disabling pain. A low-dose use
of methylergometrine, an analogue of the alkaloid er-
gonovine present in ergot, could be proposed to these pa-
tients; methylergometrine is in fact a well-known
vasoconstrictor used as salt of maleic acid (methyler-
gonovine maleate) in obstetrics and gynaecology to stop
uterine bleeding [18]. However, patients presenting with
heavy life-threatening bleeding must be immediately
treated; embolization remains an acceptable form of treat-
ment in order to avoid hysterectomy.
References
[1] Fleming H., Östör A.G., Pickel H., Fortune D.W.: “Arteriovenous
malformations of the uterus”. Obstet. Gynecol., 1989, 73, 209.
[2] Sugiyama T., Honda S., Kataoka A., Komai K., Izumi S., Yakushiji M.:
“Diagnosis of uterine arteriovenous malformation by colour pulsed
Doppler ultrasonography”. Ultrasound Obstet. Gynecol., 1996, 8, 359.
[3] Jain K.A., Jeffrey R.B. Jr, Sommer F.G.: “Gynecologic vascular ab-
normalities: diagnosis with Doppler US”. Radiology, 1991, 178, 549.
[4] Bottomley J.P., Whitehouse G.H.: “Congenital arteriovenous mal-
formations of the uterus demonstrated by angiography”. Acta Ra-
diol. Diagn., 1975, 16, 43.
[5] Wang Z., Chen J., Shi H., Zhou K., Sun H., Li X., et al.: “Efficacy
and safety of embolization in iatrogenic traumatic uterine vascular
malformations”. Clin. Radiol., 2012, 67, 541.
[6] Narang H.K., Puri M., Patra S., Trivedi S.S.: “Arterio-venous mal-
formations of uterus - diagnostic and management dilemmas”. J. Ob-
stet. Gynaecol., 2015, 35, 632.
[7] Insun K., Seung Y.H., Sang A.Y., Lee K.W.: “Arteriovenous malfor-
mation of the uterus: a cause of massive operative bleeding”. J. Ko-
rean. Med. Sci., 1991, 6, 187.
[8] Chien S.C., Lo C.Y., Wei M.C.: “Rupture of uterine arteriovenous
malformation as a cause of severe menorrhagia”. Taiwan J. Obstet.
Gynecol., 2007, 46, 314.
[9] Kasznica J., Nisar N.: “Congenital vascular malformation of the
uterus in a stillborn: a case report”. Hum. Pathol., 1995, 26, 240.
[10] Busmanis I., Ong C.L., Tan A.C.: “Uterine hemorrhage in a
menopausal female associated with an arteriovenous malformation
and myometritis”. Pathology, 2000, 32, 220.
[11] Ciani S., Merino J., Vijayalakhsmi S., Nogales F.F.: “Acquired uter-
ine arteriovenous malformation with massive endometrial stromal
component”. Histopathology, 2005, 46, 234.
[12] Brown J.V. 3rd, Asrat T., Epstein H.D., Oglevie S., Goldstein B.H.:
“Contemporary diagnosis and management of a uterine arteriove-
nous malformation”. Obstet. Gynecol., 2008, 112, 467.
[13] Goyal S., Goyal A., Mahajan S., Sharma S., Dev G.: “Acquired uterine
arteriovenous malformation developing in retained products of concep-
tion: a diagnostic dilemma”. J. Obstet. Gynaecol. Res., 2014, 40, 271.
[14] Maleux G., Timmerman D., Heye S., Wilms G.: “Acquired uterine
vascular malformations: radiological and clinical outcome after tran-
scatheter embolotherapy”. Eur. Radiol., 2006, 16, 299.
[15] Beller U., Rosen R.J., Beckman E.M., Markoff G., Berenstein A.:
“Congenital arteriovenous malformation of the female pelvis: a gy-
necologic perspective”. Am. J. Obstet. Gynecol., 1988, 159, 1153.
[16] Zizzo M., Roncati L., Colasanto D., Manenti A.: “Pancolorectal
varices superimposed on arteriovenous malformations: A life-threating
complex disease”. Clin. Res. Hepatol. Gastroenterol., 2016, 40, 75.
[17] Timmerman D., Wauters J., Van Calenbergh S., Van Schoubroeck D.,
Maleux G., Van Den Bosch T., Spitz B.: “Color Doppler imaging is
a valuable tool for the diagnosis and management of uterine vascular
malformations”. Ultrasound Obstet. Gynecol., 2003, 21, 570.
[18] Fujimoto M., Takeuchi K., Sugimoto M., Maruo T.: “Prevention of
postpartum hemorrhage by uterotonic agents: comparison of oxy-
tocin and methylergometrine in the management of the third stage of
labor”. Acta Obstet. Gynecol. Scand., 2006, 85, 1310.
Corresponding Author:
L. RONCATI, M.D., Ph.D
Department of Diagnostic and Clinical Medicine
and of Public Health, Division of Pathology
University of Modena and Reggio Emilia
Policlinico Hospital
Viale del Pozzo, 71
I-41124 Modena (Italy)
e-mail: emailmedical@gmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Investigation of the relationship between fear

of childbirth and social supports of pregnant women
in the third trimester in Turkey

S. Ertekin Pinar1, O. Duran Aksoy1, B. Cesur1, D. Bilgic1, G. Daglar1, E. Guler2

1 Faculty of Health Sciences, Cumhuriyet University, Sivas; 2 Faculty of Nursing, 9 Eylul University, İzmir
3 Private Güven Hospital, Ankara (Turkey)

Summary
Purpose: To investigate the relationship between fear of childbirth and social supports of pregnant women in the third trimester. Ma-
terials and Methods: The sample of this cross-sectional study comprised 302 pregnant women who were admitted to the gynecology
and obstetrics clinic of a state hospital in Turkey. Data were collected with the Personal Information Form, Wijma Delivery Expect
ancy/Experience Questionnaire (W-DEQ-A), and Multidimensional Scale of Perceived Social Support (MSPSS). Results: While no re-
lationship was determined between the mean total scores obtained from the W-DEQ-A and MSPSS scales (p > 0.05), statistically sig-
nificant positive correlations were determined between family and friends (r = 0.206, p = 0.000), family, and significant others (r = 0.193,
p = 0.001), and friends and significant others (r = 0.156, p = 0.006) subscales of the MSPSS (p < 0.05). Conclusion: As the social sup-
port received from the family increased so did the support from friends and significant others, and as the support received from signif-
icant others increased so did the support from friends.

Key words: Fear of childbirth; Mental health; Pregnancy; Social support.

Introduction of the fear of childbirth are the possibility of not arriving at

Pregnancy and labor are two of the most important nor-
mal physiological events in a woman’s life [1]. They are
the milestones and natural crises of life affecting women
physiologically, mentally, and socially [2, 3]. While a
woman’s mental state and lifestyle affect the course of the
pregnancy, pregnancy itself creates significant reflections
on her mental-emotional life [3]. Changes experienced dur-
ing pregnancy vary from one trimester to another. The most
prominent feelings during the first trimester are ambivalent
feelings towards being pregnant. In the second trimester,
while these feelings decline, biological ties with the fetus
are felt more deeply and closely. During the third trimester,
as the birth approaches, negative feelings accompanied by
anxiety increase, the woman becomes more sensitive about
issues regarding herself and the baby, she becomes more
dependent on others, and fear of childbirth increases [4].
One of the factors that affect anxiety, one of the problems
experienced most in the third trimester, is the fear of child-
birth [5]. Fear of childbirth is reported to be associated with
cesarean delivery or prolonged labor traumatic life events,
relationship with the partner, depression, anxiety, and low
self-esteem [1, 4, 6, 7]. In a study, the fear of childbirth
was found to be associated with the health of the baby, the
process and type of the delivery, the spouse’s approach, and
attitudes displayed by hospital staff [8]. Among the causes
the hospital in time for delivery, possibility that some things
may go wrong, possibility of doing something wrong, or
possibility that the baby or the mother suffers injuries or
dies during delivery, baby’s gender, the environment where
the delivery occurs, episiotomy, a change in the mode of
delivery, interference with vaginal delivery, pain and suf-
fering, staying alone in an unfamiliar environment, and lack
of social support [6, 8-11]. In addition, primiparous women
may experience fear of childbirth due to such reasons as
loss of control, uncertainty, and considering that they can-
not deliver a baby, whereas multiparous women may suffer
fear of childbirth due to the history of stillbirth and com-
plications experienced in previous deliveries [10, 12].
In the literature it is reported that about 5-20% of the
pregnant women suffer fear of childbirth and that serious
weakness is the kind of fear in 6% of them [5, 12-14]. It
is reported that while in 57.3% of the cases, women suf-
fer fears due to misapplications by health personnel dur-
ing delivery, in 75% of the cases, fear stems from medical
staff and hospital environment [10]. In a study by Fen-
wick et al., 48% of the women had moderate and 26% had
intense levels of fear of childbirth [15]. While antenatal
R
This study was presented as an oral presentation at the 3rd National-2,
International Midwifery Congress, November 20-23, Antalya, Turkey.

Revised manuscript accepted for publication April 4, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3642.2017
768 S. Ertekin Pinar, O. Duran Aksoy, B. Cesur, D. Bilgic, G. Daglar, E. Guler
fears may lead to distress and pain, a negative birth expe-
rience, administration of high levels analgesia, insomnia,
and emergency cesarean during delivery, may cause an in-
creased risk of severe mood disorders, such as depression
in the postpartum period [1, 5, 6]. Akdolun et al. deter-
mined that 66.3% of the women having fear of childbirth
and 84.6% of the women having extreme fear of child-
birth during the second trimester, suffered postpartum
mental distress [9]. Therefore, due to fear of childbirth,
many women avoid normal delivery and prefer elective
cesarean section [7, 10, 16]. One of the most important
factors causing anxiety during pregnancy and affecting
coping with the fear of childbirth suffered, especially dur-
ing the third trimester, is the lack of social support the
pregnant woman receives from other people [17, 18].
All interpersonal relationships having an important place
in the lives of people and providing them with emotional,
physical, and cognitive support are defined as the “social
support system” which help protect health [19]. The social
support system is a strong resource which contributes to
the solution, prevention, and treatment of sociological and
psychological problems of an individual, and to his/her
coping with challenges. Social support involves emotional
support which refers to empathy, concern, love, and trust,
and tangible support refers to help with household chores
and child care, and informational support refers to infor-
mation and assistance that can be used to cope with prob-
lems [16, 18]. Social support is an important determinant of
self-sufficiency, adaptation to the maternal role, and satis-
faction with baby care. It is reported that if a woman re-
ceives sufficient social support, she can have a non-
problematic pregnancy, adopt the maternal role quickly, and
have fewer postpartum problems. Positive social support
reduces the effects and complications of stress, depression,
and anxiety that may arise during pregnancy and after de-
livery [16, 17, 20].
Early determination of fear that pregnant women are
likely to suffer can help health professionals to plan
healthcare they provide. Women suffering from anxiety
and fear during pregnancy are provided support that all
health personnel are knowledgeable about the physiology
and psychology of pregnancy, and issues likely to arise
during pregnancy play an important role in the identifica-
tion, prevention, and early intervention of pregnancy-re-
lated problems, reduce the negative effects of these
problems on both maternal and neonatal health, and im-
prove preventive mental health services. In addition, if
mental health problems such as the blues, depression, and
psychosis that women suffer in the postpartum period are
to be reduced, and suicides are to be prevented by deter-
mining women with suicidal intent, it is important to de-
termine fear of childbirth and social support. Therefore,
the study was conducted to investigate the relationship be-
tween fear of childbirth and social support of women in
the third trimester of pregnancy.
Materials and Methods
The sample of this descriptive and cross-sectional study com-
prised 302 pregnant women who were admitted to the gynecology
and obstetrics clinic of a state hospital in Sivas, a province, in
Turkey between July 1, 2013 and September 1, 2013. The women
were in the third trimester of pregnancy, and they and the fetuses did
not have any diagnosed health problems. Data were collected with
the Personal Information Form, a version of the Wijma Delivery
Expectancy/Experience Questionnaire (W-DEQ-A) and Multidi-
mensional Scale of Perceived Social Support (MSPSS) scale. Per-
sonal Information Form was developed by the researchers through
a literature review and has 30 items related to pregnant women’s
W-DEQ-A is a 33-item Likert-type scale was developed by Wijma
et al. to measure stress and fear during labor [21]. Each item is
scored from 0 to 5. The minimum and maximum possible scores
to be obtained from the scale are 0 and 165, respectively. The cut-
off point of the scale is 84. The higher the score obtained from
the scale, the higher the level of the stress and anxiety is. The scale
was adapted to Turkish by Körükçü et al. [22]. Cronbach’s alpha
coefficient of the original scale was found to be 0.89 for the total
group [22]. In this present study, the Cronbach’s alpha coefficient
was calculated as 0.88. MSPSS scale was developed by Zimet et
al. [23]. The scale’s reliability and validity in Turkey was con-
ducted by Eker and Arkar in 1995 [24]. The scale consists of 12
items and is a seven-point Likert-type scale ranging from “very
strongly disagree” (1) to “very strongly agree (7). The scale has
three subscales referring to support sources: family, friends, and
significant others. Each of the subscales consists of four items. The
minimum and maximum possible scores obtainable from each sub-
scale are 4 and 28, respectively. The minimum and maximum pos-
sible total scores obtainable from the scale are calculated by
summing the scores from the subscales and are 12 and 84, respec-
tively. High scores obtained from the scale indicate that the level of
perceived social support is high. In Eker and Arkar’s study, the
scale’s reliability coefficient ranged from 0.80 to 0.95 indicating
that the scale has a high level of consistency [24]. In this present
study, Cronbach’s alpha coefficient of the scale was calculated as
0.81.
After the women who met the inclusion criteria and accepted to
participate in the study were informed about the purpose of the
study, their informed consents were obtained. Data collection tools
were filled in by the researchers through face-to-face interviews.
It took approximately 15-20 minutes to fill in the forms. The data
were analyzed using the SPSS 14.00 software package, frequency
distribution, Pearson correlation analysis, t-test. and ANOVA. P-
value of < 0.05 was considered as significant.
Before the study was conducted, approvals were obtained from
Cumhuriyet University Health Services Research Hospital Ethics
Committee and from the hospital where the study was to be con-
ducted. The purpose of the study was explained to the individuals
who agreed to participate in the study by researchers and their in-
formed consents were obtained. The study was conducted in ac-
cordance with the Declaration of Helsinki.
Results
The mean age of the participants was 26.26 ± 4.84 years.
Their mean age at first marriage was 21.12 ± 3.51. The
mean numbers were 2.05 ± 1.21 for pregnancies, 0.83 ±
0.96 for live births, 0.79 ± 0.89 for living children, and 0.22
± 0.54 for abortions.
In this present study, 51.7% of the participants were in
the age group of 26-35 years, 29.1% were primary school
Investigation of the relationship between fear of childbirth and social supports of pregnant women in the third trimester in Turkey 769

Table 1. — Descriptive characteristics of pregnant women. Table 2. — Fear of birth and social support scores of preg-
Descriptive characteristics n % nant women.
Age (years) MSPSS Min-max X ± SD
19–25 135 44.7 Family 4−28* (4−28)** 26.00 ± 4.36
26–35 156 51.7 Friend 4−28*(4−28)** 20.34 ± 8.79
36 and above 11 3.6 A special person 4−28*(4−28)** 23.35 ± 7.47
Education level Total 18-84*(12*84)** 69.69 ± 14.18
Primary school 88 29.1 W-DEQ-A 4−130*(0−165)** 56.91 ± 23.81
Secondary school 76 25.2
* Taken from the scale of pregnant women, minimum and maximum points.
High school 88 29.1 ** Scale’s minimum and maximum points.
University 50 16.6
Employment status
Working 36 11.9
Table 3. — The relationship between perceived social sup-
Not working 266 88.1
Perception of the economic situation port and birth fears in pregnant.
Good 114 37.8 Variables Family Friend A special Total
person MSPSS
Middle 184 60.9
Bad 4 1.3 W-DEQ-A r = -0.017 r = 0.005 r = -0.033 r = -0.019
Responsible for taking care of family p = 0.772 p = 0.926 p = 0.567 p = 0.739
Yes 91 30.1 Family r = 0.206 r = 0.193 r = 0.537
No 211 69.9 p = 0.000* p = 0.001* p = 0.000*
Smoking status Friend r = 0.156 r = 0.766
Smoker 38 12.6 p = 0.006* p = 0.000*
Non-smoker 264 87.4 A special person r = 0.683
Smoking status of husband p = 0.000*
Smoker 180 59.6 * p < 0.05.
Non-smoker 122 40.4
Total 302 100.0

scores obtained from the MSPSS scales (p > 0.05); however,

graduates, 29.1% were high school graduates, 88.1% were
unemployed, 87.4% were non-smokers, 59.6% were mar-
ried to smokers, and 69.9% did not have to provide care to
any other family member or relative. While 37.8% of them
perceived their economic status as good, 60.9% perceived
it as moderate (Table 1).
In this present study, 41.4% of the participants were
primiparous, 96% conceived spontaneously, 94.7% had de-
sired pregnancies, 75.2% had planned pregnancies, 79.8%
planned to have a normal vaginal delivery, 91.7% had reg-
ular checkups, 88.4% received no previous pregnancy-re-
lated education, 53.3% had no problems in their previous
pregnancies, 54.6% had no problems during their previous
labor or postpartum periods, 93% had someone to help and
support them during pregnancy, and 94% had someone to
help and support them after delivery.
While the mean total score was 56.91 ± 23.81 (min-max:
4-30) for the W-DEQ-A scale, it was 69.69 ± 14.18 (min-
max: 18-84) for the MSPSS scale. The mean scores obtained
from the subscales of MSPSS were as follows: 26.00 ± 4.36
(min-max: 4-28) for the family subscale, 20.34 ± 8.79 (min-
max: 4-28) for the friends subscale, and 23.35 ± 7.47 (min-
max: 4-28) for the significant others subscale (Table 2).
In the present study, no statistically significant relation-
ships were determined between the mean total scores ob-
tained from the W-DEQ-A scale and total and subscale
correlations between the mean scores obtained from the fam-
ily and friends, family and significant others, and friends and
significant others subscales of the MSPSS (p < 0.05) were
statistically significantly positive (p < 0.05). Correlations be-
tween the mean total MSPSS score and mean scores for the
family, friends, and significant others subscales were also
statistically significantly positive (p < 0.05) (Table 3).
Discussion
The period of pregnancy and childbirth during which sev-
eral physiological and psychological changes are experi-
enced is important since it requires adaption to new and
different roles. In this present study conducted to investigate
the relationship between fear of childbirth and social support
of pregnant women in the third trimester, the total mean score
of the W-DEQ-A scale was calculated as 56.91 ± 23.81. out
of 165 points when the cut-point of the W-DEQ-A scale was
taken as 84 (11.2% fear of childbirth). The present study
showed that the participants’ level of fear of childbirth was
low. The results obtained from other studies using the same
scale related to the fear of childbirth ranging between 9.1%
and 15.8% and thus were close to the results of the present
study [7, 14, 25, 26]. Fear of childbirth is often associated
with the thought that the woman will not be completely in-
dependent and will suffer pain, loss of control, negative per-
ception of childbirth, depression, lack of confidence in the
healthcare team, previous birth experiences, the woman’s
770 S. Ertekin Pinar, O. Duran Aksoy, B. Cesur, D. Bilgic, G. Daglar, E. Guler
personality traits, low self-esteem, problems with the hus-
band, daily intense stressors and lack of social support, and
they may lead to negative birth expectations, cesarean sec-
tions, increases in pain during delivery, postpartum depres-
sion, stress and anxiety [6, 7, 10, 11, 25, 27, 28]. Babacan
Gümüş et al. reported that mothers who suffered fear of
childbirth during pregnancy and/or had problems during the
postpartum period had higher levels of depression [29]. Ak-
dolun Balkaya et al. determined that more than half of the
women who suffered fear of childbirth during the second
trimester continued to have mental distress in the postpar-
tum period [9]. Therefore, the low level of fear of childbirth
determined in the present study is important for the preven-
tion and reduction of adverse conditions that may arise dur-
ing pregnancy and in the postpartum period. Contrary to the
present results, the results of Melender’s study conducted in
329 pregnant women indicated that 78% of them had fears
related to pregnancy or childbirth, or both, and that their neg-
ative moods played a part [11]. In Akdolun et al.’s study of
184 pregnant women, 98.4% of the participants suffered fear
of childbirth [9].
Social support is an important factor that reduces anxiety,
stress, and fear of childbirth during pregnancy and delivery.
Social support also plays an important role in motivating a
person to cope with the problems [2, 17]. In the present study,
while the mean total MSPSS score was 69.69 ± 14.18 out of
84 points, the mean scores for the family, friends, and signif-
icant others subscales of the MSPSS were 26.00 ± 4.36, 20.34
± 8.79, and 23.35 ± 7.47 out of 28 points, respectively. Ac-
cording to the results of this present study, pregnant women
seem to have high levels of social support, which may ac-
count for their low level of fear of childbirth. In addition, the
result that social support from friends and significant others
increased as the social support the pregnant women received
from the family increased, that social support from friends
and the family increased as the social support from signifi-
cant others increased, and that social support from significant
others and the family increased as the social support from
friends increased, may account for their low level of fear of
childbirth. Supportive relationships play an important role in
promoting health, preventing health-related problems and
stress, protecting an individual from the effects of stress,
changing the perception of events where there exists stress,
helping the person in cases where he/she suffers difficulties,
and increasing their coping capacity [16, 19]. If a pregnant
woman receives high levels of social support from the spouse,
family, friends, and other people, and if she perceives that this
support is sufficient, she may experience less physical and
psychological problems [30]. In the literature, it is stated that
social support has a positive effect on pregnancy, helps
women to perceive labor as a more positive experience and
facilitates their transition to the maternal role [18, 31].
Gözüyeşil et al. found that while 72.9% of the pregnant
women received most of the support they needed from their
husbands, 25.6% received it from their relatives [30].
Conclusions
The participants had low levels of fear of childbirth and
high levels of perceived social support. There was no rela-
tion between their fear of childbirth and perceived social
support. As the social support received from the family in-
creased, so did the support from friends and significant oth-
ers, as the support received from significant others
increased, so did the support from the family and friends,
and as the support received from friends increased, so did
the support from the family and significant others.
A limitation of the present study is that the results ob-
tained from it are applicable only to the study sample and
cannot be generalized. Based on the results, it is recom-
mended that pregnant women should be provided with in-
formation and counselling on, including: physiological and
psychological changes that occur during pregnancy and
childbirth, having desired and planned pregnancy before
they become pregnant, pregnant women should be helped
to develop their skills to cope with challenges such as anx-
iety and stress, reduce their fear of childbirth by determin-
ing risk factors regarding fears of childbirth during the
pre-pregnancy period, receive support from the family,
friends, and significant others during pregnancy by pro-
viding education and counselling, have a positive childbirth
experience, and to suffer from fear of childbirth as little as
possible.
References
[1] Otley H.: “Fear of childbirth: understanding the causes, impact and
treatment”. Br. J. Midwifery, 2011, 19, 215.
[2] Büyükkayacı Duman N., Koçak C.: “The effect of social support on
state anxiety levels during pregnancy”. Soc. Behav. Pers., 2013, 41,
1153.
[3] Kuğu N., Akyüz G.: “Psychical situation in pregnancy”. Cumhuriyet
Medical Journal, 2001, 23, 61.
[4] Poikkeus P., Saisto T., Unkila-Kallio L., Punamaki R-L., Repokari
L., Vilska S., et al.: “Fear of childbirth and pregnancy-related anxi-
ety in women conceiving with assisted reproduction”. Obstet. Gy-
necol., 2006, 108, 70.
[5] Hofberg K., Ward M.R.: “Fear of pregnancy and childbirth”. Post-
grad. Med. J., 2003, 79, 505.
[6] Kitapçıoğlu G., Yanıkkerem E., Sevil Ü., Yüksel D.: “Fear of child-
birth and the postpartum period: A scale development and validation
study”. Journal of Adnan Menderes University Medical Faculty,
2008, 9, 47-54.
[7] Nieminen K., Stephansson O., Ryding E.L.: “Women’s fear of child-
birth and preference for cesarean section a cross-sectional study at
various stages of pregnancy in Sweden”. Acta Obstet. Gynecol.
Scand., 2009, 88, 807.
[8] Khosravy F., Shahoei R., Nasab L.H., Ranaei F., Abdolahi M.: “Fears
associated with pregnancy and childbirth among Kurdish women in
Iran”. Life Sci J., 2013, 10, 367.
[9] Akdolun Balkaya N., Vural G., Eroğlu K.: “Investigating problems
related to the health of mothers and their babies resulting from risk
factors determined during pregnancy”. Düzce University, Institute of
Health Sciences, 2014, 4, 6.
[10] Dimitrova D.S.: “Experience of fear of childbirth amongst pregnant
women in Bulgaria”. WebmedCentral Psychology, 2012, 3,
WMC003905.
[11] Melender H.L.: “Experiences of fears associated with pregnancy and
Investigation of the relationship between fear of childbirth and social supports of pregnant women in the third trimester in Turkey
childbirth: a study of 329 pregnant women”. Birth, 2002, 29, 101.
[12] Fenwick J., Gamble J., Creedy D.K., Buist A., Turkstra E., Sneddon
A., et al.: “Study protocol for reducing childbirth fear: A midwife-
led psycho-education intervention”. BMC Pregnancy and Childbirth,
2013, 13, 190.
[13] Størksen T.H., Eberhard-Gran M., Garthus-Niegel S., Eskild A.:
“Fear of childbirth; the relation to anxiety and depression”. Acta Ob-
stet. Gynecol. Scand., 2012, 91, 237.
[14] Zar M., Wijma K., Wijma B.: “Relations between anxiety disorders
and fear of childbirth during late pregnancy”. Clin. Psychol. Psy-
chother, 2002, 9, 122.
[15] Fenwick J., Gamble J., Nathan E., Bayes S., Hauck Y.: “Pre-and
postpartum levels of childbirth fear and the relationship to birth
outcomes in a cohort of Australian women”. J. Clin. Nurs., 2009,
18, 667.
[16] Daş Z.: “Psychosocial and cultural dimensions pregnancy”. In:
Taşkın L. (ed). Birth and women’s health nursing. Ankara: Sistem
Offset Printing, 2009. [In Turkish]
[17] Mermer G., Bilge A., Yücel U., Çeber E.: “Evaluation of perceived
social support levels in pregnancy and postpartum periods”. Journal
of Psychiatric Nursing, 2010, 1, 71.
[18] Melender H.L.: “Fears and coping strategies associated with preg-
nancy and childbirth in Finland”. J. Midwifery Womens Health, 2002,
47, 256.
[19] Ardahan M.: “Social Support and Nursing”. Atatürk University Jour-
nal of Nursing High School, 2006, 9, 68.
[20] Vırıt O., Akbaş E., Savaş H.A., Sertbaş G., Kandemir H.: “Associa-
tion between the level of depression and anxiety with social support
in pregnancy”. Archives of Neuropsychiatry, 2008, 45, 9.
[21] Wijma K., Wijma B., Zar M.: “Psychometric aspects of the W-DEQ;
a new questionnaire for the measurement of fear of childbirth”. J.
Psychosom. Obstet. Gynecol., 1998, 19, 84.
[22] Korukcu O., Kukulu K., Firat M.Z.: “The reliability and validity of
the Turkish version of the Wijma Delivery Expectancy/Experience
Questionnaire (W-DEQ) with pregnant women”. J. Pers. Assess.,
1998, 52, 30.
771
[23] Zimet G.D., Dahlem N.W., Zimet S.G., Farley G.K.: “The Multidi-
mensional Scale of Perceived Social Support”. J. Pers. Assess., 1988,
52, 30.
[24] Eker D., Arkar H.: “Factorial structure, validity, and reliability of re-
vised form of the Multidimensional Scale of Perceived Social Sup-
port”. Turkish Journal of Psychology, 1995, 10, 45.
[25] Spice K., Jones S.L., Hadjistavropoulos H.D., Kowalyk K., Stewart
S.H.: “Prenatal fear of childbirth and anxiety sensitivity”. J. Psy-
chosom. Obstet. Gynaecol., 2009, 30, 168.
[26] Stoll K., Hall W.: “Vicarious birth experiences and childbirth fear:
does it matter how young Canadian women learn about birth?” J.
Perinat. Educ., 2013, 22, 226.
[27] Hildingsson I., Nilsson C., Karlstrőm A., Lundgren I.: “A longitudi-
nal survey of childbirth-related fear and associated factors”. J. Ob-
stet. Gynecol. Neonatal Nurs., 2011, 40, 532.
[28] Saisto T., Halmesmäki E.: “Fear of childbirth: a neglected dilemma”.
Acta Obstet. Gynecol. Scand., 2003, 82, 201.
[29] Babacan Gümüş A.B., Keskin G., Alp N., Özyar S., Karsak A.: “The
prevalence of postpartum depression and associated variables”. New
Symposium Journal, 2012, 50, 145.
[30] Gözüyeşil E.Y., Şirin A., Çetinkaya Ş.: “Depression of the pregnants
and the analysis of the factors which affect the depression”. Journal
of Fırat Health Services, 2008, 3, 39.
[31] Molzan Turan J., Bulut A., Nalbant H., Ortaylı N., Erbaydar T.:
“Challenges for the adoption of evidence-based maternity care in
Turkey”. Soc. Sci. Med., 2006, 62, 2196.
Corresponding Author:
S. ERTEKİN PINAR, M.D.
Cumhuriyet University
Faculty of Health Sciences
Midwifery Department
58140 Sivas (Turkey)
e-mail: sepinar09@gmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Prevalence of congenital malformations during pregnancy

in China: screening by ultrasound examination

L.J. Kong#, L. Fan#, G.H. Li, W.Y. Zhang

Department of Obstetrics, Beijing Obstetrics and Gynecology Hospital, Capital Medical University, Beijing (China)

Summary
Purpose: To assess the prenatal prevalence of congenital malformations and the different types and to determine rate of perinatal mor-
tality. Design: Cross-sectional study. Setting: Tertiary care centre. Materials and Methods: During the reporting period from August 2013
to September 2015, 6,432 ultrasound examinations were conducted in 2,832 pregnant women, out of whom 2,689 deliveries occurred in
the referral center. Results: The authors diagnosed 119 cases with 154 congenital malformations (isolated: 82.35% cases; complex: 17.65%
cases). The prenatal prevalence of congenital malformations was 54.38 for each 1,000 pregnancies, whereas the birth prevalence was 51.15
for each 1,000 births. The perinatal death rate was 35.29% (complex 73.68% and isolated 26.51%). The average maternal age of pregnant
women was 29.94 years. Overall, the most widely observed congenital malformations involved circulatory system (20.78%), followed by
musculoskeletal system (16.23%), followed by nervous system (12.34%), eye, ear, face, and neck (11.04%), cleft lip and cleft palate
(7.79%), digestive system (7.79%), genital organs (6.49%), chromosomal abnormalities (5.84%), urinary system (4.55%), others (3.89%),
and respiratory system (3.25%). Conclusion: The present study demonstrated that majority of perinatal deaths were due to complex con-
genital malformations. In turn the most common malformations included congenital heart diseases, neural tube defects, cleft lip/cleft palate,
and polydactyly.

Key words: Congenital malformations; Fetal anomalies; Pregnancy; Prenatal diagnosis; Ultrasound.

Introduction recognize diverse causes as well as pathogenetic pathways

Congenital malformations also referred to as birth defects
or congenital anomalies/disorders are functional and/or
structural, as well as single or multiple abnormalities of
morphogenesis in body or organs that occur in utero and
can be antenatal, during child birth or later. They result in
long-term incapacity/disability, with greater influence on
healthcare organization, society, individuals, and family.
Although the cause of nearly half of the malformations can-
not be determined, the other risk factors or causes are as
follows: demographic and socioeconomic factors, infec-
tions, genetic as well as environmental factors, and nutri-
tional status during maternity. As per recent stats,
congenital anomalies contribute to 2.76 million deaths dur-
ing initial four weeks of birth every year globally [1-3]. The
worldwide prevalence is approximately 2% to 3% [4,5]. As
per world health statistics, there was one death per 1,000
live births due to congenital malformations globally in 2000
to 2013 (ranging from 5% to 7%) among children aged
below five years. In China, the estimated deaths due to con-
genital malformations ranged from 6% to 13% from 2000
to 2013 [6].
Proof of congenital anomalies at birth begins a complex
clinical procedure focused to amend diagnostic definition,
clinical/prognostic assessment, including genetic counsel-
ing, and treatment choice. Majority of congenital anomalies
These authors contributed equally.
#
irrespective of similar phenotypic pattern [2, 3, 7]. Hence
the diagnostic process is time-consuming and troublesome
and may require long follow ups, including but not limited
to imaging, phenotype analysis, anamnesis, and laboratory
tests.
Ultrasound examination is advantageous in the early dis-
covery of congenital anomalies. In populations at low risk,
17% to 35 % of sensitivity and 99% of specificity, can be
observed while in populations at high risk, more than 90%
sensitivity can be noted [8, 9]. The specific use of 3D ul-
trasound in assessment of skeletal, limb, and facial structure
anomalies was demonstrated in reviews conducted by
Timor-Tritsch et al. and Goncalves et al. [10, 11]. Prenatal
diagnosis of congenital malformations is critical for the
suitable counseling of parents regarding special needs, in
utero interventions, voluntary pregnancy termination when
required, intimation to neonatology team for proper care,
delivery in the appropriate centre, and future prevention [9,
12]. The types as well as prevalence of congenital anom-
alies vary from one country to that of another and in turn
from one region to that of another.
As per the present authors’ knowledge, none of the stud-
ies focused on determination of the prenatal prevalence of
congenital malformations in China. The major goal of the
present study was to assess the prenatal prevalence and of

Revised manuscript accepted for publication June 8, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3788.2017
 L.J. Kong, L. Fan, G.H. Li, W.Y. Zhang 773

congenital malformations and their different types, and to Table 1. — Maternal demographic characteristics (n=119).
determine rate of perinatal mortality. Characteristics Total (n=119)
Maternal age, years Mean (SD) 29.94 (7.81)
Median (range) 28 (18-49)
Materials and Methods Age category < 25 years, n (%) 44 (37)
This single-centre cross-sectional observational study was per- 25-35 years, n (%) 41 (34.5)
formed at a tertiary hospital, China over a period of two years, > 35 years, n (%) 34 (28.6)
from August 2013 to September 2015. Gender Male, n (%) 65 (54.6)
Every pregnant women attended in the setting had a routine Female, n (%) 52 (43.7)
ultrasound examination at gestation period between 18 and 20 Unknown, n (%) 2 (1.7)
weeks or later at scheduled follow-up visit. In the present study, Parity 0, n (%) 84 (70.6)
pregnant women with issues of congenital malformations with 1, n (%) 22 (18.5)
prenatal diagnosis at or who were referred to hospital with con- ≥ 2, n (%) 13 (10.9)
genital malformations were included. Included patients were ex- Smoking Yes, n (%) 21 (17.6)
amined, diagnosed, and properly counseled by specialists. The
No, n (%) 98 (82.4)
majority of the patients had more than one ultrasound examina-
Drinking Yes, n (%) 16 (13.4)
tion. All the required maternal data, congenital malformation de-
tails, delivery data, neonatal results, etc, collected during No, n (%) 103 (86.6)
diagnosis were recorded in an information sheet. Data gathering Family history Yes, n (%) 18 (15.1)
and follow up were conducted in all the units, wards, and labo- No, n (%) 101 (84.9)
ratories, whenever possible. Consanguinity Yes, n (%) 27 (22.7)
A spontaneous abortion or miscarriage was noted if pregnancy No, n (%) 92 (77.3)
loss occurred before 20 gestation weeks. Stillbirths were enlisted
SD = standard deviation.
at 22 weeks or older gestational age. Neonatal death was noted
up to 28 days of newborn child life. Perinatal mortality encom-
passes the fetal or neonatal death from 22 gestation weeks to 28
days of newborn child life. As per International Statistical Clas-
Results
sification of Diseases and Related Health Problems 10th Revision
(ICD-10) 2010 [13], congenital malformations were classified During the reporting period, 6,432 ultrasound examina-
based on the system/organ involved (circulatory, digestive, mus- tions were conducted for 2,832 pregnant women out of
culoskeletal, nervous, respiratory, as well as urinary systems, ear,
whom 2,689 deliveries happened in the referral center. The
face, eye and neck, cleft lip and cleft palate, genital organs, and
others). Malformations were considered as isolated when a sin- remaining patients were alluded for delivery in respective
gle system was involved, whereas complex when involvement of referral centers and a few were lost to follow up.
more than one system was observed. Out of 2,832 pregnant women, 119 cases were diagnosed
Prenatal prevalence was computed from the aggregate number with 154 congenital malformations (isolated: 98 [82.35%]
of pregnant women, whereas birth prevalence as well as the peri- cases and complex: 21 [17.65%] cases). The prenatal preva-
natal mortality were assessed from the number of deliveries. The
Institutional ethics committee approval was acquired for the pres- lence of congenital malformations was 54.38 for each 1,000
ent study and patient confidentiality was strictly maintained. pregnancies. Of 119 cases, six (5.04%) patients experienced
spontaneous abortion or miscarriage, 102 (85.71%) patients
delivered in the referral center, whereas 11 (9.24%) patients

Figure 1. — Prenatal prevalence of congeni-

tal malformation based on system or organ
involved is depicted.
774 Prevalence of congenital malformations during pregnancy in China: screening by ultrasound examination

Table 2. — Prenatal congenital malformations (CMs) diagnosed by ultrasound.

Type of CM based on CM Type of CM based on CM
system or organ involved Total N % Prenatal system or organ involved Total N % Prenatal
prevalence prevalence
(per 1,000 (per 1,000
pregnancies) pregnancies)
Nervous system 19 12.34 6.71 Cleft lip and cleft palate 12 7.79 4.24
Neural tube defects 14 9.09 4.94 Cleft lip 8 5.19 2.82
Anencephalus and similar 4 2.59 1.41 Cleft palate with cleft lip 4 2.59 1.41
Encephalocele 4 2.59 1.41 Cleft palate 4 2.59 1.41
Spina bifida 6 3.89 2.12 Digestive system 12 7.79 4.24
Hydrocephalus 3 1.95 1.06 Esophageal atresia 3 1.95 1.06
Microcephaly 2 1.29 0.71 Duodenal atresia or stenosis 2 1.29 0.71
Eye, ear, face and neck 17 11.04 6.00 Atresia of small intestine 2 1.29 0.71
Eye 7 4.55 2.47 Imperforate anus 1 0.65 0.35
Anophthalmos 3 1.95 1.06 Hiatus hernia 2 1.29 0.71
Microphthalmos 2 1.29 0.71 Atresia of large intestine 2 1.29 0.71
Congenital cataract 1 0.65 0.35 Genital organs 10 6.49 3.53
Congenital glaucoma 1 0.65 0.35 Hypospadias 4 2.59 1.41
Ear 5 3.25 1.77 Undescended testicle 2 1.29 0.71
Macrotia 2 1.29 0.71 Indeterminate sex 2 1.29 0.71
Microtia 2 1.29 0.71 Absence of ovary 1 0.65 0.35
Pointed ear 1 0.65 0.35 CM of uterus and cervix 1 0.65 0.35
Face and neck 5 3.25 1.77 Urinary system 7 4.55 2.47
Sinus, fistula and cyst of brachial cleft 2 1.29 0.71 Congenital hydronephrosis 1 0.65 0.35
Otocephaly 1 0.65 0.35 Cystic kidney disease 2 1.29 0.71
Pterygium colli 1 0.65 0.35 Renal agenesis 2 1.29 0.71
Macrostomia 1 0.65 0.35 CM of kidney/urinary system 2 1.29 0.71
Circulatory system 32 20.78 11.30 Musculoskeletal system 25 16.23 8.83
Cardiac chambers and connections 10 6.49 3.53 Polydactyly 8 5.19 2.82
Common arterial truncus 4 2.59 1.41 Syndactyly 2 1.29 0.71
Single ventricle 2 1.29 0.71 Congenital diaphragmatic hernia 2 1.29 0.71
Transposition of great vessels 4 2.59 1.41 Musculoskeletal dysplasia 2 1.29 0.71
Cardiac septa 10 6.49 3.53 Gastroschisis 3 1.95 1.06
Ventricular septal defect 2 1.29 0.71 Omphalocele 2 1.29 0.71
Tetralogy of Fallot 4 2.59 1.41 Congenital scoliosis 1 0.65 0.35
Atrioventricular septal defect 3 1.95 1.06 Limb reduction defects 1 0.65 0.35
Unspecified 1 0.65 0.35 Talipes 4 2.59 1.41
Pulmonary and tricuspid valves 3 1.95 1.06 Chromosomal abnormalities 9 5.84 3.18
Pulmonary valve stenosis 1 0.65 0.35 Down syndrome 4 2.59 1.41
Tricuspid atresia 2 1.29 0.71 Patau syndrome 3 1.95 1.06
Aortic and mitral valves 6 3.89 2.12 Edwards syndrome 2 1.29 0.71
Aortic valve atresia/stenosis 2 1.29 0.71 Others 6 3.89 2.12
Mitral valve anomalies 1 0.65 0.35 Moebius syndrome 1 0.65 0.35
Hypoplastic left heart syndrome 3 1.95 1.06 CM of breast 2 1.29 0.71
Absence of aorta 2 1.29 0.71 CM of integument 2 1.29 0.71
Coarctation of aorta 1 0.65 0.35 Fetal alcohol syndrome 1 0.65 0.35
Respiratory system 5 3.25 1.77 Total 154 54.38
Malformations of nose 3 1.95 1.06
Malformations of lung 2 1.29 0.71

were either alluded for delivery in respective referral cen-
ters or were lost to follow up. During the study period,
2,522 patients delivered with 129 congenital malforma-
tions. The birth prevalence of congenital malformations
was 51.15 for each 1,000 births.
The perinatal death rate was 35.29% (36 stillborn and
neonatal deaths out of 102 cases). The perinatal death rate
with complex congenital malformation was 73.68% (14 out
of 19 patients) and isolated congenital malformation was
26.51% (22 out of 83 patients).
The average maternal age (SD) of pregnant women was
29.94 (7.81) years. The fetuses were 65 (54.6%) males, 52
(43.7%) females, and two (1.7%) were of unknown sex.
The maternal demographic characteristics including age,
age category, gender of fetus, parity, smoking, drinking,
family history, and consanguinity are presented in Table 1.
 L.J. Kong, L. Fan, G.H. Li, W.Y. Zhang
Overall, the most widely observed congenital malforma-
tions involved circulatory system (n=32, 20.78%) with a
prenatal prevalence of 11.3 for each 1,000 pregnancies. In
the circulatory system, defects related to cardiac chambers
and connections and cardiac septa were mostly seen (6.49%
each). The second common congenital malformations in-
volved musculoskeletal system (n=25, 16.23%) followed
by nervous system (n=19, 12.34%), eye, ear, face and neck
(n=17, 11.04%), cleft lip and cleft palate (n=12, 7.79%),
digestive system (n=12, 7.79%), genital organs (n=10,
6.49%), chromosomal abnormalities (n=9, 5.84%), urinary
system (n=7, 4.55%), others (n=6, 3.89%), and respiratory
system (n=5, 3.25%), respectively. Further information re-
garding prenatal congenital malformations is demonstrated
in Table 2. Prenatal prevalence of congenital malformation
based on system or organ involved is depicted in Figure 1.
Discussion
Since the emergence of ultrasonography in the decade of
1960, there was a rise in the total sum of pregnancy imag-
ing studies. Considerable improvements in magnification
imaging as well as signal processing enhanced the capacity
to envision anatomy of embryos and fetuses. Variations with
respect to critical practice regarding frequency and ultra-
sonography performance during pregnancy exist. Develop-
ments in imaging, for example magnetic resonance imaging
and echocardiography, have adjoined to early fetal evalua-
tion in certain specific cases. The capabilities to hearten a
high risk pregnant woman regarding ordinary fetal discov-
eries, and to give exhaustive counseling/guidance with the
choice to terminate in instances of unequivocally suspected
deadly or significant anomalies, have moved to the advance
in gestational age from the prenatal diagnosis [14].
Anatomy in addition to pathophysiology of fetus can con-
trast from that of the infant, pediatric as well as adult pop-
ulation, hence a reasonable comprehension and learning of
this is crucial. Experience of personnel, imaging, and ma-
ternal attributes impact reporting accuracy of fetal malfor-
mations, specifically in late first trimester. Imaging skills
and high expenses with regards to equipment as well as
time enhance identification rates. The existence of related
malformations and risk factors for fetal anomalies (e.g.,
family history, smoking, drinking, consanguinity, obesity,
etc) paves for more focused attention, thereby enhancing
the accuracy of results [14-17]. Various systematic reviews
and studies demonstrated greater identification rates of
anomalies prior to 24 weeks, particularly major malforma-
tions. Many organizations inferred that ultrasonography at
previable 2nd trimester ought to be routinely offered and
specific guidelines have to be followed [8, 14, 18, 19]. A
study conducted by Reddy et al. stressed on the fact that
no less than one ultrasound study ought to be offered to
every single pregnant women somewhere around 18 and
20 weeks [15]. Early finding of extensive variety of fetal
775
malformations can be determined via transvaginal and/or
transabdominal examinations during 11 to 14 gestation
weeks [16, 20-22].
The perinatal death rate was higher with complex con-
genital malformations, was firmly associated to maternal age,
and the multifaceted nature of the malformations. The pres-
ent study was clinically based and hence does not constitute
the actual prevalence in China. This information ought to in-
vigorate future research and coordinated effort for more pre-
cise and outright reporting of congenital malformations in
China. The prenatal prevalence of congenital malformations
was 54.38 for each 1,000 pregnancies whereas the birth
prevalence was 51.15 for each 1,000 births. The most widely
observed congenital malformations involved circulatory sys-
tem, followed by musculoskeletal and nervous systems, and
so on. The rates of consanguinity, smoking, drinking, and
family history are 22.7%, 17.6%, 13.4%, and 15.1%, re-
spectively. The determination of congenital malformations
has enhanced significantly in the previous few years which
is majorly attributed to innovations in ultrasound technology
and skilled personnel. This may clarify the expanded num-
ber of cases as of now being analyzed contrasted with the
past, which thus mirrors a greater prenatal prevalence of con-
genital malformations.
In studies specific to prevalence of congenital malforma-
tions in China demonstrated the following results: average
incidence reported from 1997 to 2011 in Henan province of
China was 86.2 cases per 10,000 births with majority of neu-
ral tube defects [23]; data from 2000 to 2010 in Hainan
province of China showed rising trend of birth defects preva-
lence with 77.99 cases in 2000 to 98.93 cases per 10,000
births in 2010 (polydactyly, cleft lip, hydrocephalus, con-
genital heart diseases, and limb defects as the most common
malformations) [24]; average incidence reported from 2009
to 2010 in five countries/cities in Gansu province was 7.49%,
with most common defects being congenital heart disease,
neural tube defects/pigmented nevus, and hydrocephalus
[25]; prevalence of 156.1 cases per 10,000 births was re-
ported in population based survey performed from 2005 to
2008 in Inner Mangolia, China (with higher prevalence of
neural tube defects and congenital heart disease) [26]; data
from 2006 to 2012 in Tongzhou district of Beijing showed
prevalence of 12.62% birth defects (congenital heart dis-
eases, polydactyly, cleft lip/palate, neural tube defects, and
external ear malformation as the most defects) [27]. The
present study nearly showed similar results with congenital
heart diseases, neural tube defects, cleft lip/cleft palate, and
polydactyly as the frequent malformations.
There is developing proof of connection between prenatal
maternal ecological exposures and raised risk of congenital
malformations. All the more particularly, many epidemiolog-
ical studies showed that exposure during pregnancy to envi-
ronmental factors, such as air pollutants (e.g. NO2, PM10,
and SO2), tobacco smoke, and contaminated water is alto-
gether connected with an increased risk for congenital mal-
776 Prevalence of congenital malformations during pregnancy in China: screening by ultrasound examination
formations [28-31]. Maternal age, consanguinity, drinking,
smoking and family history, parity, socioeconomic status, etc
were also considered to be risk factors for congenital malforma-
tions in many studies [23-27]. The higher prenatal prevalence in
current study can be attributed to aforementioned factors.
It is noteworthy that not all pregnant women who were di-
agnosed by ultrasonography had delivered in the referral cen-
ter and hence the false positive, as well as false negative results
are not known. However, current results demonstrated the ul-
trasound as robust diagnostic tool with skilled personnel.
Conclusion
The present study demonstrated that majority of perinatal
deaths were due to complex congenital malformations. The
most frequently reported congenital malformations involved
circulatory system followed by musculoskeletal system and
nervous system. In turn the most common malformations in-
cluded congenital heart diseases, neural tube defects, cleft
lip/cleft palate, and polydactyly. Continued efforts are required
in order to improve maternal as well as child healthcare via in-
formation sharing, health education, and preventive measures.
References
[1] World Health Organization. Congenital anomalies. Available at:
http://www.who.int/mediacentre/factsheets/fs370/en/
[2] Corsello G., Giuffrè M.: “Congenital malformations”. J. Matern.
Fetal Neonatal Med., 2012, 25, 25.
[3] Twinning P., McHugo J.M., Pilling D.W.: “Textbook of fetal abnor-
malities”. London: Churchill Livingstone, 2000.
[4] Smith WD.: “Classification, nomenclature, and naming of morpho-
logic defects”. J. Pediatr., 1975, 87, 162.
[5] Kalter H.: “Five-decade international trends in the relation of peri-
natal mortality and congenital malformations: Still birth & neonatal
death compared”. Int. J. Epidemiol., 1991, 20, 173.
[6] World Health Organization: “World Health Statistics 2015”. Available
at: http://apps.who.int/iris/bitstream/10665/170250/1/9789240694439
_eng.pdf
[7] Jones K.L.: “Smith’s recognizable patterns of human malforma-
tions”. Philadelphia: WB Saunders Company, 1997.
[8] American College of Obstetricians and Gynecologists: “ACOG Prac-
tice Bulletin No. 101: Ultrasonography in pregnancy”. Obstet. Gy-
necol., 2009, 113, 451.
[9] Sallout B., Obedat N., Shakeel F., Mansoor A., Walker M., Al-Badr
A.: “Prevalence of major congenital anomalies at King Fahad Med-
ical City in Saudi Arabia: a tertiary care centre-based study”. Ann.
Saudi Med., 2015, 35, 343.
[10] Timor-Tritsch I.E., Monteagudo A.: “Three and four-dimensional ul-
trasound in obstetrics and gynecology”. Curr. Opin. Obstet. Gy-
necol., 2007, 19, 157.
[11] Gonçalves L.F., Lee W., Espinoza J., Romero R.: “Three- and 4-di-
mensional ultrasound in obstetric practice. Does it help?” J. Ultra-
sound Med., 2005, 24, 1599.
[12] Ewigman B., Crane J., Frigoletto F., LeFevre M., Bain R., McNel-
lis D.: “Effect of prenatal ultrasound screening on perinatal out-
come”. N. Engl. J. Med., 1993, 329, 821.
[13] World Health Organization: “International Statistical Classification
of Diseases and Related Health Problems 10th Revision (ICD-10)
Version for 2010”. Available at: http://apps.who.int/classifications/
icd10/browse/2010/en#/XVII
[14] Rayburn W.F., Jolley J.A., Simpson L.L.: “Advances in ultrasound
imaging for congenital malformations during early gestation”. Birth
Defects Res. A. Clin. Mol. Teratol., 2015, 103, 260.
[15] Reddy U.M., Abuhamad A.Z., Levine D., Saade G.R.: “Fetal Imaging
Workshop Invited Participants. Fetal imaging: executive summary of a
Joint Eunice Kennedy Shriver National Institute of Child Health and
Human Development, Society for Maternal-Fetal Medicine, American
Institute of Ultrasound in Medicine, American College of Obstetricians
and Gynecologists, American College of Radiology, Society for Pedi-
atric Radiology, and Society of Radiologists in Ultrasound Fetal Imag-
ing Workshop”. Am. J. Obstet. Gynecol., 2014, 210, 387.
[16] Timor-Tritsch I.E., Bashiri A., Monteagudo A., Arsland A.A.: “Qualified
and trained sonographers in the US can perform early fetal anatomy scans
between 11 and 14 weeks”. Am. J. Obstet. Gynecol., 2004, 191, 1247.
[17] Syngelaki A., Chelemen T., Dagklis T., Allan L., Nicolaides K.H.:
“Challenges in the diagnosis of fetal non-chromosomal abnormalities
at 11–13 weeks”. Prenat. Diagn., 2011, 31, 90.
[18] Whitworth M., Bricker L., Neilson J.P., Dowswell T.: “Ultrasound for
fetal assessment in early pregnancy”. Cochrane Database Syst Rev.,
2010, 4, CD007058.
[19] Cargill Y., Morin L., Bly S., Butt K., Denis N., Gagnon R., et al.: “Con-
tent of a complete routine second trimester obstetrical ultrasound ex-
amination and report”. J. Obstet. Gynaecol. Can., 2009, 31, 272.
[20] Economides D.L.: “Early pregnancy screening for fetal abnormalities”.
Ultrasound Obstet. Gynecol., 1999, 13, 81.
[21] den Hollander N.S., Wessels M.W., Niermeijer M.F., Los F.J., Wladimiroff
J.W.: “Early fetal anomaly scanning in a population at increased risk of ab-
normalities”. Ultrasound Obstet. Gynecol., 2002, 19, 570.
[22] Souka A.P., Pilalis A., Kavalakis Y., Kosmas Y., Antsaklis P., Antsak-
lis A.: “Assessment of fetal anatomy at the 11–14-week ultrasound ex-
amination”. Ultrasound Obstet. Gynecol., 2004, 24, 730.
[23] Xia L., Sun L., Wang X., Yao M., Xu F., Cheng G., et al.: “Changes in
the Incidence of Congenital Anomalies in Henan Province, China, from
1997 to 2011”. PLoS One, 2015, 10, e0131874.
[24] Fan L., Gong T., Cao X., Du Y.: “Epidemiologic characteristics of birth
defects in the Hainan Province from 2000 to 2010, China”. Birth De-
fects Res. A. Clin. Mol. Teratol., 2013, 97, 750.
[25] Du W.Y., Pei L.Y., Ma R.L., Wu S., Jiang D.M., Ma Q., et al.: “Present
situation of congenital defects in five counties (cities) of Gansu province
in 2009 2010”. Zhonghua Liu Xing Bing Xue Za Zhi, 2013, 34, 1402.
[26] Zhang X., Li S., Wu S., Hao X., Guo S., Suzuki K., Yokomichi H., et al.:
“Prevalence of birth defects and risk-factor analysis from a population-
based survey in Inner Mongolia, China”. BMC Pediatr., 2012, 12, 125.
[27] Yu J.R., Jin L., Xiao L.H., Jin L.: “Prevalence of birth defects in the
Tongzhou District of Beijing between 2006 and 2012”. Zhongguo Dang
Dai Er Ke Za Zhi, 2014, 16, 1133.
[28] Leonardi-Bee J., Britton J., Venn A.: “Secondhand smoke and adverse
fetal outcomes in nonsmoking pregnant women: a meta-analysis”. Pe-
diatrics, 2011, 127, 734.
[29] Vrijheid M., Martinez D., Manzanares S., Dadvand P., Schembari A.,
Rankin J., et al.: “Ambient air pollution and risk of congenital anom-
alies: a systematic review and meta-analysis”. Environ. Health Per-
spect., 2011, 119, 598.
[30] Hwang B.F., Jaakkola J.J., Guo H.R.: “Water disinfection by-products
and the risk of specific birth defects: a population-based cross-sectional
study in Taiwan”. Environ. Health., 2008, 7, 23.
[31] Ngo A.D., Taylor R., Roberts C.L.: “Paternal exposure to Agent Orange
and spina bifida: a metaanalysis”. Eur. J. Epidemiol., 2010, 25, 37.
Corresponding Author:
W.Y. ZHANG, M.D.
Department of Obstetrics
Beijing Obstetrics and Gynecology Hospital
Capital Medical University, no.251
Yaojiayuan Road, Chaoyang District
Beijing 100026 (China)
e-mail: zhangweiyuan56@hotmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Does increase in body mass index effect primary

dysmenorrhea?

M. Temur1,3, U. Gök Balci2, Y. A. Güçlü2, B. Korkmaz3, P.Ö. Özbay4, N. Soysal2, Ö. Yilmaz1,

T. T. Yilmazer2, T. Çift3, K. Öngel2
Department of Obstetrics and Gynecology, Manisa Merkezefendi State Hospital, Manisa
1
2 Department of Family Medicine, Tepecik Research and Training Hospital, İzmir
3Department of Obstetrics and Gynecology, Bursa Yüksek İhtisas Research and Training Hospital, Bursa
4Department of Obstetrics and Gynecology, Aydın Obstetrics and Pediatrics Hospital, Aydın (Turkey)

Summary
Purpose: In the present study, the aim was to evaluate the relationship between obesity and dysmenorrhea and the effects of socio-
demographic features on it. Materials and Methods: A total of 303 women were included in the study .Grading of severity of dysmen-
orrhea was made based on Verbal Multidimensional Scoring System (VMSS). Results: When correlations between severity of
dysmenorrheic symptoms and patients were assessed, there was a statistically significant difference between the rates of chronic dis-
ease in the dysmenorrhea groups and the rates of dysmenorrhea history in the family (p = 0.037 and p = 0.008, respectively). There was
a statistically significant difference in the mean body mass index (BMI) in the dysmenorrhea grades (p < 0.001). The mean BMI for those
without dysmenorrhea was higher than those with mild or moderate dysmenorrhea. Those with severe dysmenorrhoea had a significantly
higher mean BMI than those with mild dysmenorrhea (p <0.001, p = 0.002, and p = 0.009, respectively). There was a statistically sig-
nificant difference in dysmenorrheal grades and BMI groups (p = 0.002). The severity of dysmenorrhoea in those with a BMI of 30 and
above was greater than those of mild and moderate ones. Conclusion: The main underlying cause of dysmenorrhea may not be obesity,
but it may be one of the correctible predisposing factors. Balanced diet and trying to decrease one’s BMI within normal limits may lower
the incidence of dysmenorrhea.

Key words: Dysmenorrhea; Obesity; Body mass index.

Introduction
Dysmenorrhea is a cyclic pain felt especially on supra-
pubic and pelvic region during menstrual period. Dysmen-
orrhea includes two types: primary and secondary
dysmenorrhea. Primary dysmenorrhea is a cyclic pain stem-
ming from intrinsic uterine factors. Though etiology of pri-
mary dysmenorrhea is not fully understood, prostaglandins
and especially PGF alpha has been held responsible from
the development of dysmenorrheic pain. PGF2 alpha in-
duces myometrial contractions and ischemia leading to
emergence of dysmenorrheic pain. Intrinsic as well as emo-
tional factors, as depression and anxiety, induce primary
dysmenorrhea. However, secondary dysmenorrhea can
occur as a result of intrapelvic pathologies as endometrio-
sis, adenomyosis, and ovarian cyst and application of an
intrauterine device [1, 2].
Primary dysmenorrhea is a relatively prevalent problem in
women of their reproductive ages. Dysmenorrhea adversely
effects daily work and routine activities with resultant mil-
lion dollars of material damage [3]. Worldwide studies on
dysmenorrhea report its prevalence as ranging between 28
and 65.7 percent [4, 5]. However in Turkey, its incidence
has been reported to vary between 58% and 89% [6-8].
Obesity is defined as an excessive accumulation of fat in
the body. Obesity is an important risk factor for chronic dis-
eases as diabetes, cardiovascular diseases, and cancer. Es-
pecially in recent years, it is an important health problem
with increasing importance. Nowadays, nearly 300 million
obese women have been detected worldwide. According to
2008, data nearly 1.4 billion adults are overweight and ap-
proximately 300 million of them are obese [9].
Body mass index (BMI) is a simple measurement tool
used worldwide for the classification of obesity. BMI is
easily calculated based on body weight and height. BMI is
estimated by dividing body weight by the square of body
height. BMI is categorized according to this classification
as follows: BMI < 18.5 kg/m2, underweight; BMI: 18.50-
24.99 kg/m2, normal range; BMI: 25.00-29.99 kg/m2, over-
weight; BMI: 30.00-34.99 kg/m2, obese class I; BMI:
35.00-39.99 kg/m2, obese class II; BMI ≥ 40.00 kg/m2,
obese class III [9].
In the present study, the authors aimed to evaluate the re-
lationship between obesity and dysmenorrhea and the ef-
fects of sociodemographic features on dysmenorrhea.

Revised manuscript accepted for publication March 13, 2017

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3595.2017
778 M. Temur, U. Gök Balci, Y. A. Güçlü, B. Korkmaz, P.Ö. Özbay, N. Soysal, Ö. Yilmaz, T. T. Yilmazer, T. Çift, K. Öngel

Table 1. — General features of the patients.

Age (years) Ort. ± SD (min-max) 32.3±9.0 (13-57)
Education level, n (%) No 5 (1.7)
Primary 83 (27.4)
Middle 36 (11.9)
High 93 (30.7)
University 86 (28.4)
Economical level, n (%) Poor 23 (7.6)
High 59 (19.5)
Moderate 221 (72.9)
Smoking history, n (%) 88 (29.0)
Chronic disease, n (%) 90 (29.7)
Parity Ort. ± SD (min-max) 1.0-1.2 (0-8)
Parity, n (%) Nulliparity 136 (44.9)
Primiparity 57 (18.8)
Multiparity 110 (36.3)
Type of birth, n (%) Vaginal birth 4 (1.3)
Cesarean birth 163 (53.8) Figure 1. — The relationship between dysmenorrhea grade and
No 136 (44.9) body mass index.
BMI (kg/m2), Ort. ± SD (min-max) 28.2±8.7 (14.3-64.3)
BMI (kg/m2), n (%) < 18.5 14 (4.6)
18.5-24.9 128 (42.2)
lation was applied when conditions were not provided. Statistical
25-29.9 53 (17.5) significance level of alpha was accepted as p < 0.05.
30≤ 108 (35.6)
Dysmenorrhea history in family, n (%) 65 (21.5)
Dysmenorrhea grade, n (%) No 112 (37.0) Results
Mild 80 (26.4)
Moderate 67 (22.1) Three hundred three women with a mean age of 32.3 ±
Severe 44 (14.5) 9.0 years were included in the study. The general charac-
teristics of the pregnants are summarized in Table 1.
Of the women 26.4% had mild, 22.1% had moderate, and
Materials and Methods 14.5% had severe dysmenorrhea; 37% of the women had
A total of 303 women (age range, 15-44 years) in their repro- no dysmenorrhea (Table 1). There was no statistically sig-
ductive age with regular menstruation periods who consulted to nificant difference in the mean age, education and eco-
Obesity Clinic of Izmir Tepecik Training and Research Hospital nomic levels, smoking rates, and parity types of the
between February 2013 and June 2013 were included in the study. dysmenorrhea groups (p = 0.607, p = 0.164, p = 0.600, p =
Using a questionnaire form prepared by the investigators, so- 0.172, and p = 0.888, respectively).
ciodemographic characteristics of the female patients were
recorded. In the outpatient clinic, body weight and height were There was a statistically significant difference between
measured and recorded in the questionnaire file. BMIs were cal- the rates of chronic disease in the dysmenorrhea groups and
culated by dividing body weight in kg by square of height in me- the rates of dysmenorrhea history in the family (p = 0.037
ters and obesity was classified based on this formula. Grading of and p = 0.008, respectively). Those with severe dysmenor-
severity of dysmenorrhea was made based on Verbal Multidi- rhea had a higher rate of chronic illness (33.9% mild,
mensional Scoring System (VMSS) [10].
(A) Mild dysmenorrhea is defined as painful menstruation with
22.5% moderate 22.4%, and severe 43.2%). Those with
no limitation of normal activity, with infrequent requirement of moderate dysmenorrhea had a higher rate of dysmenorrhea
analgesics and no systemic complaints. (B) Moderate dysmenor- history in the family (history of dysmenorrhea in the fam-
rhea is defined as menstrual pain affecting daily activities, with re- ily 14.3%, mild 21%, moderate 35.8%, and severe 18.2%)
quirement of analgesics for pain relief and few systemic (Table 2).
complaints. (C) Severe dysmenorrhea is defined as menstrual pain
There was a statistically significant difference in the
with severe limitation of daily activities, poor response to anal-
gesics, and apparent systemic complaints like vomiting, fainting, mean BMI in the dysmenorrhea grades (p <0.001). The
etc. mean BMI for those without dysmenorrhea was higher than
SPSS 15.0 was used for statistical analysis. Descriptive statis- those with mild and moderate dysmenorrhea. Those with
tics, number, and percentage for categorical variables, mean, stan- severe dysmenorrhea had a significantly higher mean BMI
dard deviation, minimum, and maximum were given for than those with mild dysmenorrhea (p <0.001, p = 0.002,
numerical variables. Because the numerical variables did not sat-
isfy the normal distribution condition, multiple group compar- and p = 0.009, respectively). There was a statistically sig-
isons were made with the Kruskal Wallis test independently. nificant difference in dysmenorrheal grades and BMI
Subgroup analyzes were done by Mann Whitney U test and in- groups (p = 0.002). The severity of dysmenorrhea in those
terpreted by Bonferroni correction. Comparisons of ratios in with a BMI of 30 and above was greater than those with
groups were made with Chi Square Analysis. Monte Carlo simu- mild and moderate ones (Table 3, Figure 1).
 Does increase in body mass index effect primary dysmenorrhea? 779

Table 2. — The relationship between dysmenorrhea severity and studied parameters.

Dysmenorrhea grade
No Mild Moderate Severe
Ort. ± SD Ort. ± SD Ort. ± SD Ort. ± SD
Age (years) 32.8 ± 9.1 31.8 ± 8.7 32.9 ± 9.3 31.0 ± 9.0 0.607
n % n % n % n % p
Education level No 2 1.8 1 1.3 2 3 0 0 0.164
Primary 33 29.5 20 25 19 28.4 11 25
Middle 19 17 5 6.3 6 9 6 13.6
High 26 23.2 24 30 27 40.3 16 36.4
University 32 28.6 30 37.5 13 19.4 11 25
Economical level Poor 6 5.4 5 6.3 9 13.4 3 6.8 0.600
High 25 22.3 14 17.5 12 17.9 8 18.2
Moderate 81 72.3 61 76.3 46 68.7 33 75
Smoking history 28 25 20 25 22 32.8 18 40.9 0.172
Chronic disease 38 33.9 18 22.5 15 22.4 19 43.2 0.037
Parity Nulliparity 49 43.8 39 48.8 26 38.8 22 50 0.888
Primiparity 20 17.9 15 18.8 14 20.9 8 18.2
Multiparity 43 38.4 26 32.5 27 40.3 14 31.8
Dysmenorrhea history in family 16 14.3 17 21.3 24 35.8 8 18.2 0.008

Table 3. — The relationship between dysmenorrhea grade and body mass index.
Dysmenorrhea grade
No Mild Moderate Severe
Ort. ± SD Ort. ± SD Ort. ± SD Ort. ± SD
BMI (kg/m2) 30.9 ± 10.0 25.6 ± 7.5 26.0 ± 5.8 29.4 ± 8.8 <0.001
n % n % n % n % p
BMI (kg/m2) < 18.5 5 4.5 2 2.5 5 7.5 2 4.5 0.002
18.5-24.9 33 29.5 47 58.8 32 47.8 16 36.4
25-29.9 18 16.1 14 17.5 13 19.4 8 18.2
≥ 30 56 50.0 17 21.3 17 25.4 18 40.9
Kruskal-Wallis (Mann-Whitney U test).

Discussion demonstrate that dysmenorrhea still continues to be one of

In the pathophysiology of dysmenorrhea, together with
withdrawal of progesterone, release of prostaglandins and
leukotrienes from arachidonic acid is very important. In-
flammatory response mediated by these prostaglandins
(mainly PGF2) and leukotrienes induces vasoconstriction
and myometrial contractions leading to development of is-
chemia and pain [11].
Dysmenorrhea is one of the most frequent gynaecologi-
cal causes of presentation to hospital. In the literature its
incidence ranges between 43 and 90 percent [12, 13]. In a
study performed on a group of young girls in Italy, the in-
cidence of primary dysmenorrhea was detected as 85% and
in another study in Mexico its incidence was reported as
48.4% [14, 15]. The incidence of dysmenorrhea in Turkey
changes between 63 and 70 percent [8, 16]. In this study
the rate of dysmenorrhea was 63%, which is comparable
with studies performed in this country. These outcomes
the prevalent gynaecologic problems among female popu-
lation. Differences in the incidence of dysmenorrhea among
countries might stem from differences in geographic, ge-
netic, nutritional factors, and varying BMIs.
In studies performed, a correlation between family his-
tory and dysmenorrhea has been detected. In studies per-
formed in compliance with many literature studies, family
history can significantly affect prevalence of dysmenorrhea
[1, 7]. In the present study, dysmenorrhea was more com-
mon in those who had dysmenorrhea history in the family.
In recent studies, cytokine gene expressions in peripheral
blood of dysmenorrheic patients were analysed and dys-
menorrhea has been assertedly related to alterations in the
balance between proinflammatory cytokines and TGF beta
superfamily. Increases in menstrual phase proinflammatory
cytokines (IL1B, TNF, IL6, and IL8) and decreases in some
TGF-β superfamily members (BMP4, BMP6, GDF5,
GDF11, LEFTY2, NODAL, and MSTN) were detected
780 M. Temur, U. Gök Balci, Y. A. Güçlü, B. Korkmaz, P.Ö. Özbay, N. Soysal, Ö. Yilmaz, T. T. Yilmazer, T. Çift, K. Öngel
[17].
Adipose tissue is not only a source of energy, but it also
functions as an active endocrine organ, with an ability to
secrete various cytokines, peptides originating from adi-
pose tissue. As observed in many studies, adipose tissue se-
cretes increased amounts of various cytokines (TNF alpha,
IL 6, and IL 8), prostaglandins (PGI2 and PGF2), and
proinflammatory cytokines [18]. In some literature studies,
a correlation was found between BMI and dysmenorrhea.
However, in some other studies, though debatable, obesity
has been indicated as one of the predisposing factors for
dysmenorrhea [19]. In a study by Harlow et al., and Ju et al.
a correlation was found between higher BMI and dysmen-
orrhea [20, 21]. Origin of this finding may be related to the
role played by cytokines in the pathophysiology of primary
dysmenorrhea and increased release of cytokines from ex-
cess adipose tissue in patients with higher BMI concur-
rently with primary dysmenorrhea. Moreover, in some
studies a correlation was found between decreased BMI
and dysmenorrhea. However in these study populations
obese patients were limited in number and these studies
yielded scarce data for the evaluation of the impact of obe-
sity on dysmenorrhea [22, 23], which investigated the as-
sociation between obesity and dysmenorrhea in adolescent
girls, the study group with BMI above 25 kg/m2 consisted
of 25.26% of the study population. In this study, although,
a statistically significant difference between obese and non-
obese patients as for dysmenorrhea could not be demon-
strated, the obese group experienced dysmenorrheic
symptoms more frequently [24]. In a study by Haidari et
al., the authors indicated the presence of a correlation be-
tween dysmenorrhea and some anthropometric parameters
including waist-hip ratio and waist circumference [25]. In
a study by Neal et al., the authors detected that diets with
low-fat content decreased concentrations of sex hormone-
binding globulin and body weights without resultant alle-
viation of dysmenorrheic symptoms [26]. However, in their
study the percentage of patients with BMI over 25 and 30
kg/m2 was higher than those found in many other studies.
Dysmenorrhea was statistically significantly less frequent
in the group of patients with BMIs above 30 kg/m2 (p <
0.05). This condition is strongly related to the fact that pri-
mary dysmenorrhea is associated with ovulatory cycles and
also it may be correlated with non-observance of ovulatory
cycles in women with BMI over 30 kg/m2 [15]. In the pres-
ent study, there was more severe dysmenorrhoea with a
BMI of 30 and over. This may be due to increased secretion
of dysmenorrhoea from fat cells. Interestingly, there was
no dysmenorrhoea in those with a BMI of 30 or more. This
can be explained by the presence of more ovulation defects
in the excess of adipose tissue. Indeed when compared with
the group with BMI lower than 30 mg/kg2, in the group
with higher BMI, a statistically significantly increased in-
cidence of dysmenorrhea was detected (p < 0.05) This find-
ing demonstrates potential effect of cytokines produced in
adipose tissue on dysmenorrhea.
In the literature the relationship between BMI and dys-
menorrhea is debatable. The common characteristic of the
studies which asserted a correlation was that they detected
higher incidence of dysmenorrhea in study groups with
lower and higher BMIs, rather than those with normal BMI.
The main underlying cause of dysmenorrhea may not be
obesity, but it may be one of the correctible predisposing
factors. Therefore, a balanced diet and trying to decrease
one’s BMI within normal limits may lower the incidence
of dysmenorrhea.
Acknowledgement
The authors are grateful to Dr. Özgür Yılmaz for his valu-
able contributions to the statistical analysis. This study did
not receive any specific grant from any funding agency in
the public, commercial or not-for-profit sector. This study
was supported by Muzaffer Temur (first author).
References
[1] Dorn L.D., Negriff S., Huang B., Pabst S., Hillman J., Braverman
P., Susman E.J.: “Menstrual symptoms in adolescent girls: associa-
tion with smoking, depressive symptoms, and anxiety”. J. Adolesc.
Health, 2009, 44, 237.
[2] Wallace S., Keightley A., Gie C.: “Dysmenorrhoea”. Obstet. Gy-
naecol., 2010, 12, 149.
[3] Coco AS.: “Primary dysmenorrhea”. Am. Fam. Physician, 1999, 60,
489.
[4] Burnett M.A., Antao V., Black A., Feldman K., Grenville A., Lea R.:
et al.: “Prevalence of primary dysmenorrhea in Canada”. J. Obstet.
Gynaecol. Can., 2005, 27, 765.
[5] Rodrigues A.C., Gala S., Neves Â., Pinto C., Meirelles C., Frutuoso
C.: et al.: “Dysmenorrhea in adolescents and young adults: preva-
lence, related factors and limitations in daily living”. Acta Med.
Port., 2011, 24, 383.
[6] Nur N., Sumer H.: “Prevalence of dysmenorrhea and related risk fac-
tors in adolescents”. Surekli. Tip. Egitimi. Dergisi., 2008, 7, 27.
[7] Polat A., Celik H., Gurates B., Kaya D., Nalbant M., Kavak E., et al.:
“Prevalence of primary dysmenorrhea in young adult female uni-
versity students”. Arch. Gynecol. Obstet., 2009, 279, 527.
[8] Unsal A., Ayranci U., Tozun M., Arslan G., Calik E.: “Prevalence of
dysmenorrhea and its effect on quality of life among a group of fe-
male university students”. Ups. J. Med. Sci., 2010, 115, 138.
[9] World Health Organisation: “Global database on body mass index”.
Available at: http://apps.who.int/bmi/index.jsp?introPage=intro_3.
html
[10] Andersch B., Milsom I.: “An epidemiologic study of young women
with dysmenorrhea”. Am. J. Obstet. Gynecol., 1982, 144, 655.
[11] Harel Z.: “Dysmenorrhea in adolescents and young adults: etiology
and management”. J. Pediatr. Adolesc. Gynecol., 2006, 19, 363.
[12] Svanberg L., Ulmsten U.: “The incidence of primary dysmenorrhea
in teenagers”. Arch Gynecol., 1981, 230, 173.
[13] Jamieson D.J., Steege J.F.: “The prevalence of dysmenorrhea, dys-
pareunia, pelvic pain and irritable bowel syndrome in primary care
practices”. Obstet. Gynecol., 1996, 87, 55.
[14] Balbi C., Musone R., Menditto A., Di Prisco L., Cassese E., D’Ajello
M., et al.: “Influence of menstrual factors and dietary habits on men-
strual pain in adolescence age”. Eur. J. Obstet. Gynecol. Reprod.
Biol., 2000, 91, 143.
[15] Ortiz M.I., Rangel-Flores E., Carrillo-Alarcón L.C., Veras-Godoy
H.A.: “Prevalence and impact of primary dysmenorrhea among Mex-
 Does increase in body mass index effect primary dysmenorrhea?
ican high school students”. Int. J. Gynaecol. Obstet., 2009, 107, 240.
[16] Aykut M., Gunay O., Gun I., Tuna R., Balcı E., Ozdemir M., et al.:
“Biyolojik, sosyo-demografik ve nutrisyonel faktörlerin dismenore
prevelansına etkisi”. Erciyes Tıp. Dergisi., 2007, 29, 393.
[17] Ma H., Hong M., Duan J., Liu P., Fan X., Shang E., et al.: “Altered
cytokine gene expression in peripheral blood monocytes across the
menstrual cycle in primary dysmenorrhea a case-control study”.
PLoS One, 2013, 8, e55200.
[18] Wang Y.Y., Lin S.Y., Chuang Y.H., Chen C.J., Tung K.C., Sheu W.H.:
“Adipose proinflammatory cytokine expression through sympathetic
system is associated with hyperglycemia and insulin resistance in a rat
ischemic stroke model”. Am. J. Physiol. Endocrinol. Metab., 2011, 300,
E155.
[19] Lefebvre G., Pinsonneault O., Antao V., Black A., Burnett M., Feld-
man K.: “Primary dysmenorrhea consensus guideline”. J. Obstet.
Gynaecol. Can., 2005, 27, 1117.
[20] Harlow S.D., Park M.: “A longitudinal study of risk factors for the
occurrence, duration and severity of menstrual cramps in a cohort of
college women”. Br. J. Obstet. Gynaecol., 1996, 103, 1134.
[21] Chauhan M., Kala J.: “Relation between dysmenorrhea and body
mass index in adolescents with rural versus urban variation”. J. Ob-
stet. Gynaecol. India, 2012, 62, 442.
781
[22] Tangchai K., Titapant V., Boriboonhirunsarn D.: “Dysmenorrhea in
Thai adolescents: prevalence, impact and knowledge of treatment”.
J. Med. Assoc. Thai., 2004, 87, 69.
[23] Kamacı M., Önder Y., Akman N.: “Adölesanlarda primer dis-
menorenin vücut kitle indeksiyle ilişkisi” Van. Tıp. Dergisi., 1997, 4,
154.
[24] Kaur K.: “Obesity and dysmenorrhea in young girls: is there any
link”. Hum. Biol., 2014, 3, 214.
[25] Haidari F., Akrami A., Sarhadi M., Shahi M.M. : “Prevalence and
severity of primary dysmenorrhea and its relation to anthropometric
parameters”. J. HAYAT., 2011, 17, 70.
[26] Barnard N.D., Scialli A.R., Hurlock D., Bertron P.: “Diet and sex-
hormone binding globulin, dysmenorrhea, and premenstrual symp-
toms”. Obstet. Gynecol., 2000, 95, 245.
Corresponding Author:
M. TEMUR, M.D.
Manisa Merkezefendi Hospital
Department of Obstetrics and Gynecology
45020, Merkezefendi, Manisa (Turkey)
e-mail: drmuzaffer@yahoo.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Maternal serum soluble CD40 ligand concentration

as a predictor of preeclampsia at first trimester

F. Hatiboğlu1, S. Kumbasar2, B.A. Şık3, E. Sever2, M. Temur2, S. Salman4, Ö. Çot2, A. Özcan5, F. Yazıcıoğlu5
Department of Obstetrics and Gynecology, Adıyaman University School of Medicine, Adıyaman
1

Department of Obstetrics and Gynecology, Sakarya Research and Education Hospital, Sakarya
2
3 Department of Obstetrics and Gynecology, Istanbul Aydın University School of Medicine, Istanbul
4 Department of Obstetrics and Gynecology, Gaziosmanpaşa Taksim Research and Education Hospital, Istanbul
5 Department of Obstetrics and Gynecology, Süleymaniye Research and Education Hospital, Istanbul (Turkey)

Summary
Aim: The aim of this study was to investigate the use of serum soluble CD40 ligand (sCD40L) concentration values, measured be-
tween 11+0 and 13+6 weeks of pregnancy, in the prediction of preeclampsia development and to determine the presence of a statisti-
cally significant difference. Materials and Methods: sCD40L concentrations of 202 cases who were admitted to the present hospital
for routine control between 11+0 and 13+6 gestational weeks were measured and antenatal follow up was performed until delivery.
Results: Among 202 patients who completed gestational period, 172 subjects developed no preeclampsia, while two cases had severe
and 28 cases had mild preeclampsia (30 subjects in total). sCD40L level was detected as 4212.35 ± 3366.46 pg/ml in normotensive
pregnant cases, while it was 5244.63 ± 3633.27 pg/ml in the patients with preeclampsia. There was no statistically significant differ-
ence between preeclamptic and normotensive patient group in terms of sCD40L concentrations (p < 0.05). Conclusion: The authors
revealed that the mean sCD40L did not significantly increase during first trimester in the patients with preeclampsia, while it showed
a tendency to increase in these cases. They believe that other than sCD40L concentration values, consideration of other patient-related
factors such as some parameters including S endoglin, and uterine artery pulsatility may provide more successful results in the pre-
diction of preeclampsia. Therefore, prospective, randomized, and controlled studies are required to investigate the importance of
sCD40L concentrations in the prediction of preeclampsia during first trimester.

Key words: sCD40L, Preeclampsia, First-trimester screening.

Introduction damage caused by increased inflammatory response begin-

Preeclampsia is one of the leading causes of maternal
and fetal morbidity and mortality in underdeveloped and
developing countries has a prevalence of 3-4% [1].
Preeclampsia and gestational hypertension (GH) are seen
in about 8-10% of primigravidas. The etiopathology of
preeclampsia is not fully understood despite extensive in-
vestigations [2]. In the studies on the etiopathogenesis of
preeclampsia, different mechanisms such as endothelial
dysfunction, inflammatory processes, oxidative stress, and
derangement in renin-angiotensin system, prostaglandins,
nitric oxide, endothelins, genetic predisposition, and im-
munological factors have been proposed [3]. All of these
cause vasoconstriction and blood pressure increases [3].
The main pathology underlying preeclampsia is decreased
or absent trophoblastic invasion from maternal spiral ar-
teries causing endothelial damage in uteroplacental and
systemic circulation, which ends up with abnormal pla-
centation [3]. Although abnormalities of placentation
occur during 10th -16th gestational weeks, the clinical signs
and symptoms appear in the second and third trimester [1].
The main pathology in preeclampsia is endothelial cell
ning in the early stages of pregnancy. In the pathogenesis of
preeclampsia, various molecules such as vascular endothe-
lial growth factor (VEGF), placental growth factor (PIGF),
soluble VEGF-receptor1, tumor necrosis factor (TNF), and
serum soluble CD40 ligand (sCD40L) have been proposed
to play role in endothelial cell dysfunction and inflammatory
response [4]. Flow cytometric studies involving the patients
with preeclampsia showed that thrombocytes were over ac-
tivated [5]. Increased thrombocyte activation during first
and second trimester indicates increased risk for preeclamp-
sia [5]. It was shown that increase in CD63, which indicates
increased thrombocyte activation in the first trimester, was
an independent risk factor in the development of preeclamp-
sia [5]. Activated thrombocytes induce the release of medi-
ators triggering inflammatory response in leukocytes and
endothelial cells. sCD40L produced by activated thrombo-
cytes will bind to CD40 on endothelial cells inducing ex-
pression of tissue factor that plays an important role in the
inflammatory response [6]. When the present authors ex-
amined the studies on thrombocyte activation, they noticed
that the main molecule responsible for the development of

Revised manuscript accepted for publication February 10, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3596.2017
 F. Hatiboğlu, S. Kumbasar, B.A. Şık, E. Sever, M. Temur, S. Salman, Ö. Çot, A. Özcan, F. Yazıcıoğlu 783

inflammation and initiating thrombocyte activation was Table 1. — Demographical features, serum biochemistry,
sCD40L. For this reason, based on the fact that endothelial and ultrasonography results of the patients.
cell damage begins in the first trimester, they measured the Features Normotensive Preeclamptic p* value
concentration of sCD40L, which is synthesized as a result of pregnant subjects subjects
(n=172) (n=30)
increased thrombocyte function, as a factor triggering en- mean±ss /n (%) mean±ss /n (%)
dothelial cell damage. The measurements were performed Age (years) 28.3±5.9 29.1±6.3 0.48
during the first trimester with the combined test.The aim of Gravida 1.9±1.1 2.1±1.0 0.55
this study was to investigate whether serum sCD40L meas- Parity 1.2 ± 0.9 1.5 ± 1.4 0.60
urement in the first trimester could be a novel determinant BMI (kg/m2) 28.3 ±3.3 29.3± 6.7 0.59
in prediction of preeclampsia. Gestational week
12.3 ± 0.8 12.1 ± 0.7 0.22
according to CRL
Gestational age
Materials and Methods 38.0±2.8 38.4±1.8 0.53
at delivery
The study included antenatal follow up of 480 subjects who Cesarean delivery 98(%57) 19(%63) 0.63**
were admitted to Perinatology Department of Istanbul Süley- Birth weight of 3209.77± 3310.17±
0.40
maniye Gynecology and Obstetrics Education and Research Hos- newborn (grams) 594.55 674.95
pital for the first trimester ultrasound scans. The subjects were Mean sCD40L 4212.35 ± 5244.63 ±
0.11
informed about the study and informed consent form was ob- (pg/ml) 3366.46 3633.27
tained. The authors planned a retrospective-cohort study here.
t-test; **chi-square, CRL: crown rump length,
*
Initial admission evaluation was performed between 11+0 and
Values are mean standard deviation or n (%).
13+6 gestational weeks. A detailed history including subjects’ age,
BMI, parity, past medical history (such as diabetes mellitus,
chronic hypertension, thrombophilia, antiphospholipid syn-
drome), drugs used (antihypertensive drugs, steroids, insulin, be-
tamimetics, aspirin, anticoagulants, antiepileptics, antidepressants, Results
antithyroids, thyroxin, anti-inflammatory drugs), and conception First-trimester screening test and sCD40L concentration
method (spontaneous, ovulation induction, IVF) were taken. The
measurement of 480 subjects were performed and the cases
enrollment criteria were intact pregnancy in 11th -14th gestational
week and presence of no detected anomaly in 11th -14th gestational were retrospectively evaluated. Cases whose pregnancy out-
week. The cases with multiple pregnancy, chronic hypertension, comes could not be obtained (273 patients), were excluded
diabetes mellitus, thrombophilia, molar pregnancy, and history of from the study. Also five cases were excluded due to fetal
drug use were excluded. Antenatal follow up of the cases was death or abortus before 24th week of gestation. A total of 278
made by perinatology outpatient clinic until delivery. The data on patients were excluded. Among 202 cases with known re-
pregnancy outcomes were obtained from the records of the pres-
ent hospital. All reported obstetrical outcomes or pregnancy re- sults of 11th -14th week screening, sCD40L concentration and
lated hypertension cases were assessed to determine whether the pregnancy outcome, 172 (85%) cases had no preeclampsia,
condition was preeclampsia. while two patients developed severe and 28 cases had mild
Following the collection of blood samples from 480 pregnant preeclampsia, 30 cases (14.9%) in total. Table 1 shows de-
women for the first trimester screening test during 11th -14th week mographical features and pregnancy outcomes of the pa-
first trimester ultrasound scan, they were centrifuged at 1,000xg
cycle for 30 minutes to separate serum and then stored in sealed tients. There was no statistically significant difference
tubes at -20°C freezer. sCD40L levels were measured in all sam- between preeclamptic and normotensive pregnant cases with
ples by using enzyme-linked immunosorbent assay (ELISA) regard to their gestational week, gravida status, number of
method. parity, body mass index, gestational age at delivery, birth
The mean systolic and diastolic blood pressure values during weight (p > 0.05) (Table 1). None of the patients enrolled in
pregnancy, urinary protein excretion in 24-hour urine, total weight
the study were smokers and had pregnancy via assisted re-
gain, time of delivery, mode of delivery and birth weight, antihy-
pertensive drugs used during pregnancy, history of hospitaliza- productive technique (IVF). sCD40L level was detected to be
tion due to hypertension, history of severe headache and 4212.35 ± 3366.46 pg/ml in normotensive pregnant subjects,
abdominal pain in the hospital, and need for magnesium sulfate while it was 5244.63 ± 3633.27 pg/ml in preeclamptic cases.
antepartum treatment, were examined. There was no statistically significant difference between
For all subjects enrolled in the study, gestational week at de-
preeclamptic and normotensive subjects in terms of serum
livery, mode of delivery, birth weight, and presence of preeclamp-
sia (classified as mild and severe according to Sibai criteria) were soluble CD40L concentrations (p > 0.05) (Table 1).
determined. Among 30 preeclamptic cases, only two received
SPSS 10.0 package software was used for the assessment of the MgSO4 treatment. None of the subjects had seizure or vi-
data. Normal distribution of numerical data was evaluated by Stu- sual findings. Two subjects were hospitalized due to
dent’s t-test, Fisher exact test, and chi-square tests were used for preeclampsia. One preeclamptic patient received alfamed in
the comparison of the groups and p < 0.05 was accepted as sta-
tistically significant. the last month, five subjects received nidilat, while one pa-
tient received nidilat plus alfamed in the last month of preg-
nancy, and one subject received nidilat from the 20th week
(Table 2).
784 Maternal serum soluble CD40 ligand concentration as a predictor of preeclampsia at first trimester

Table 2. — Clinical characteristics of the women with

preeclampsia.
Preeclamptic
pregnant woman
(n=30) mean ±ss
Systolic blood pressure (mm Hg) 148±7.7
Diastolic blood pressure (mm Hg) 101 ± 5.4
Proteinuria in 24 hours (mg/day) 564± 382
Clinical presentation n (%)
Severe hypertension (>160/110 mm Hg) 2 (6)
Symptoms of end-organ involvement 1 (3)
Intrauterine growth restriction 1 (3)
Proteinuria >5 g / 24 hours 2 (6)
Eclampsia -
Headache 6 (20)
Changes in vision -
Magnesium sulfate antepartum treatment 2 (6)
Edema 16 (53)
Outpatient treatment (alfametildopa, nifedipin) 8 (26) Figure 1. — Distribution of sCD40L concentration in preeclamp-
- Alfametildopa (in the last month) 1 tic and normotensive subjects.
- Nifedipin 5
- Alfametildopa + nifedipin (in the last month) 1
- Nifedipin (from the 20th week) 1

Discussion
Table 3. — Comparison of pathological serum sCD40L
concentrations of preeclamptic and normotensive subjects. Since maternal and fetal morbidity of preeclampsia is
Normal Pathological Fisher Exact
high, many studies have been conducted on this issue. Early
sCD 40L sCD 40L p value identification and close follow up of high-risk patients play
n % n % an important role in preventing complications. The patho-
Normotensive genesis of preeclampsia is still unclear. The cause of ma-
167 97.1 5 2.9
pregnant woman ternal clinical symptoms seen in preeclampsia is systemic
Preeclamptic
27 90.0 3 10 0.099 endothelial cell dysfunction, while the causes of endothe-
pregnant woman
lial cell damage are unknown. There are many studies
showing that maternal and fetal inflammatory response
play role in the pathogenesis of preeclampsia [5]. Increased
inflammatory markers in the serum during first and second
Cut-off pathological value of sCD40L was 11,000 pg/ml trimester indicate increased risk for preeclampsia. It was
obtained by adding +2 SD to the mean value of normal shown that TNF, soluble TNF-alpha receptor, TNF-alpha
pregnancy. There was no statistically significant difference receptor 1 and 2, interleukin 2 (IL-2), sCD40L, and leuko-
between normal and preeclamptic groups with regards to cyte activation were increased in preeclampsia [7, 8].
number of cases with values above cut-off pathological CD40L is a trimeric transmembrane protein of tumor necro-
value (p > 0.05) (Table 3). sis factor family. CD40-CD40L is present in leukocytes,
However, the mean sCD40L concentration was found endothelial cells, smooth muscle cells, and activated throm-
higher in preeclamptic patients (5244.63 ± 3633.27 pg/ml) bocytes [9]. The CD40-CD40L was first identified on acti-
when compared to normotensive cases (4212.35 ± 3366.46 vated T cells. In addition to its expression on T cells, CD40
pg/ml) (Figure 1). In normotensive group, minimum level is expressed by macrophages, dentritic cells, and mono-
of sCD40L concentration was found to be 846 pg/ml and cytes and its interaction with CD40L leads to the synthesis
maximum level was 7,578 pg/ml. However, in preeclamp- of pro-inflammatory cytokines, such as TNF-α ,interleukin
tic group, minimum level of CD40 concentration was 1,611 (IL)-1, and IL-6. Furthermore, CD40L expression has been
pg/ml and the maximum level was 8,877 pg/ml. Figure 1 reported in mast cells, eosinophils, and basophils; mast
shows sCD40L concentration in preeclamptic and nor- cells and basophils induce IgE production by B cells
motensive pregnant cases. through the activation of the CD40 receptor by CD40.
Commonly, these reports support the important role of
CD40-CD40L system in allergic reactions, inflammation,
and humoral immunity [10]. Ninety percent of circulating
sCD40L originates from activated thrombocytes. sCD40L
 F. Hatiboğlu, S. Kumbasar, B.A. Şık, E. Sever, M. Temur, S. Salman, Ö. Çot, A. Özcan, F. Yazıcıoğlu
induces coagulation by increasing the synthesis and ex-
pression of tissue factor from endothelial cells and mono-
cytes [11]. Studies showed that thrombocyte activation
occurred during first trimester and the clinical picture of
preeclampsia appeared weeks and months later [12, 13].
Activated thrombocytes play a role in acute and chronic in-
flammation by degranulating and activating monocytes [14,
15]. Thrombocyte activation and aggregation take place
after release of some mediators such as thromboxane A2
and sCD40L [16]. Thrombocytes express CD40L in their
membranes during activation. Afterwards, CD40L leaves
the membrane and switches to soluble form. sCD40L can
be measured in the blood [17, 18]. sCD40L plays role in
thrombocyte activation and stabilization of arterial throm-
bus. sCD40L concentration was found to be increased in
chronic inflammatory diseases such as cystic fibrosis, in-
flammatory bowel diseases, and systemic lupus erythe-
matosus [19]. Thromboxane A2 increases inflammation
and endothelial damage by causing vasoconstriction and
thrombocyte aggregation. Resultant endothelial damage
plays an important role in the pathogenesis of preeclamp-
sia. Thrombocyte activation was found to be more promi-
nent in the patients with pregnancy-related complications
such as preeclampsia and intrauterine growth retardation
when compared to normal pregnancies and non-pregnant
individuals [20]. Harlow et al. measured sCD40L concen-
tration in preeclamptic individuals, pregnant subjects with
gestational and essential hypertension, normotensive preg-
nant subjects, and non-pregnant individuals and revealed
that it was significantly higher only in the subjects with
preeclampsia [21]. The results of other groups were simi-
lar [20]. Lukanov et al. measured the concentrations of
CD40L and CD62P located on thrombocyte surface and
CD40L on monocyte surface in non-pregnant, preeclamp-
tic, and normotensive pregnant subjects and detected sig-
nificantly higher levels only in preeclamptic patients when
compared to other groups [22]. In the study by Alacacıoğlu
et al., sCD40L concentration was found to be higher in
preeclamptic subjects than that of normotensive pregnant
subjects [23].
A study by Erez et al. showed higher concentrations of
sCD40L in pregnant women when compared to non-preg-
nant subjects. In this study, sCD40L concentration was
found to be higher in the subjects who were in labour when
compared to women who were not in labour [24].
Inwald et al. suggested that sCD40L induces leukocyte
activation by causing CD62P expression and the release of
alpha granule and dense granules [25]. As a result, CD40-
CD40L interactions and sCD40L are regarded as respon-
sible for increased thrombocyte activation and
inflammation seen in preeclampsia [26].
Azzam et al. analyzed 11 women with mild preeclamp-
sia, 11 women with severe preeclampsia, and six women
with HELLP syndrome and compared with 13 normoten-
sive pregnant women as a control group. They found that
785
sCD40L concentration and the platelet surface expression
of CD40L was significantly higher in women with mild and
severe preeclampsia and HELLP syndrome compared with
normal pregnancy group [27].
Wu et al. investigated the role of CD40/CD40L in the
pathogenesis of preeclampsia. Maternal serum was obtained
from 20 patients with preeclampsia (PE group) as well as
20 healthy pregnant women (control group). The human um-
bilical endothelial cell line was cultured in the presence of
maternal serum, after which cell growth and apoptosis were
assessed by MTT and flow cytometry analysis. As compared
to human umbilical endothelial cell line cells treated with
control sera, those treated with preeclampsia sera had altered
morphology, decreased cell growth, increased apoptosis, and
greater CD40/CD40L protein and mRNA expression. They
asserted that PE sera may induce endothelial cell damage
possibly through increased CD40/CD40L expression [28].
sCD40L concentration has been measured in non-preg-
nant individuals, preeclamptic, and in normotensive preg-
nant subjects in many studies [29]. However, the present
authors did not find any study in the literature in which
sCD40L concentration was measured in the first trimester.
In the present study, the authors measured sCD40L , one of
the most important markers of thrombocyte activation, dur-
ing the first trimester double screening test of 202 subjects,
and investigated the place of mean sCD40L concentration in
the prediction of preeclampsia. In the literature, there are
some studies in which first trimester thrombocyte activation
has been measured by using flow cytometry and various in-
dicators. A study in which CD63 was used, CD63 increase
in the first trimester was found to be an independent risk
factor for the development of preeclampsia [26].
Conclusion
In the present study, the authors revealed that the mean
sCD40L was not significantly increased in preeclamptic
cases during first trimester, while it showed a tendency to
increase in these cases No statistically significant differ-
ence was detected between preeclamptic and normotensive
pregnant subjects in terms of sCD40L concentration. Other
than sCD40L concentration, addition of other patient-re-
lated factors may be beneficial in obtaining more success-
ful results in the prediction of preeclampsia.
References
[1] Papageorghio A.T., Yu CK., Bindra R., Pandis G., Nicholaides K.H.,
Fetal Medicine Foundation Second Trimester Screening Group:
“Multicenter screening for preeclampsia and fetal growth restriction
by transvaginal uterine artery Doopler at 23 weeks of gestation”. Ul-
trasound. Obstet. Gynecol., 2001, 18, 441.
[2] Bronses I.A.: “Morphological changes in uteroplacental bed, in preg-
nancy hypertension”. Clin. Obstet. Gynaecol.,1977, 4, 573.
[3] Sibai M.D.: “Diagnosis, prevention, and management of eclampsia”.
Obstet. Gynecol., 2005, 105, 402
[4] Romero R., Nien J.K., Espinoza J., Todem D., Fu W., Chung H., et
786 Maternal serum soluble CD40 ligand concentration as a predictor of preeclampsia at first trimester
al.: “A longitudinal study of angiogenic (placental growth factor)
and anti-angiogenic (soluble endoglin and soluble vascular en-
dothelial growth factor receptor -1) factors in normal pregnancy and
patients destined to gestational age neonate”. J. Matern. Fetal.
Neonatal. Med., 2008, 21, 9.
[5] Konijnenberg A., Stokkers E.W., Van der Post J.A., Schaap M.C.,
Boer K., Bleker O.P., et al.: “Extensive platelet activation in
preeclampsia compared with normal pregnancy: Enhanced expres-
sion of cell adhesion molecules”. Am. J. Obstet. Gynecol., 1997, 176,
461.
[6] Mellembakken J.R., Solum N.O., Ueland T., Videm V., Aukrust P.:
“Increased concentrations of soluble CD 40 ligand, RANTES and
GRO-alpha in preeclampsia – possible role of platelet activation”.
Thromb. Haemost., 2001, 86, 1272.
[7] Carr D.B., McDonald G.B.,Brateng D., Desai M., Thach C.T., East-
erling T.R.: “The relationship between hemodynamics and inflam-
matory activation in vomen at risk for preeclampsia”. Obstet.
Gynecol., 2001, 98, 1109.
[8] Del C.I., Cruz M.A., Zhang H., Lopez J., Afshar-Kharghan V.:
“Platelet activation leads to activation and propagation of the com-
plement system”. J. Exp. Med., 2005, 201, 871.
[9] Aruffo A., Hollenbaugh D., Wu L.H., Ochs H.D.: “The molecular
basis of X-linked agammaglobulinemi,hyper-IgM syndrome, and se-
vere combined immunodeficiency in humans”. Curr. Opin. Hema-
tol., 1994, 1, 12.
[10] Van Kooten C., Banchereau J.: “CD40-CD40 ligand’’. J. Leukoc.
Biol., 2000, 67, 2.
[11] Cipollone F., Feri C., Desideri G., Paloscia L.,Materazzo G., Mas-
cellanti M. “Preprocedural level of soluble CD40L is predictive of
enhanced inflammatory response and restenosis after coronary an-
gioplasty”. Circulation, 2003, 108, 2776.
[12] Bar J., Ben –Haroush A., Lahav J., Sullivan M.: “Interaction between
platelets and cytokines-a possible role in the pathogenesis of
preeclampsia”. Vasc. Dis. Prev., 2004, 1, 101.
[13] Janes S.L., Kyle P.M., Redman C., Goodall A.H.: “Flow Cytometric
detection of activated platelets in pregnant women prior to the de-
velopment of pre-eclampsia”. Throm. Haemost.,1995, 74, 1059.
[14] Mause S.F., Von H.P., Zernecke A.,Koenen R.R., Weber C.: “Platelet
microparticles:a transcellular delivery system for rantes promoting
monocyte recruitment on endothelium. ”Arterioscler. Thromb. Vasc.
Biol., 2005, 25, 1512.
[15] Nijm J., Wikby A., Tompa A., Olsson A.G., Jonasson L.: “Circulat-
ing levels of proinflammatory cytokines and neutrofil-platelet ag-
gregates in patients with coronary arter disease”. Am. J. Cardiol.,
2005, 95, 452.
[16] Mulvihill N.T., Foley J.B.: “Inflammation in acute coronary syn-
dromes”. Heart, 2002, 87, 201.
[17] Henn V., Slupsky J.R., Grafe M.,Anagnostopoulos I., Forster
R.,Muller –Berghaus G.: “CD40 ligand on activated platelets trig-
gers an inflammatory reaction of endothelial cells”. Nature, 1998,
391, 591.
[18] Henn V.,Steinbach S., Buchner K., Presek P.,Kroczek R.A: “The in-
flammatory action of CD40 ligand (CD154) expressed on activated
human platelets is temporally limited by coexpressed CD40”. Blood,
2001, 98, 1047.
[19] Goules A, Tzzzioufas A.G., Manousakis M.N., Kirou K.A.,Crow
M.K., Routsias J.G.: “Elevated levels of soluble CD40 ligand
(sCD40L) in serum of patients with systemic autoimmune disease”.
J. Autoimmun., 2006, 26, 11165.
[20] Ahlawat S., Pati H.P., Bhaatla N., Fatima L., Mittal S.: “Plasma
platelet aggregating factor and platelet aggregation studies in pre-
eclampsia”. Acta. Obstet. Gynecol. Scand.,1996, 75, 428.
[21] Harlow F.H., Brown M.A., Brighton T.A., Smith S.L., Trickett A.E.,
Kwan Y.L., Davis G.K.: “Platelet activation in the hypertensive dis-
orders of pregnancy”. Am. J. Obstet. Gynecol., 2002, 187, 688.
[22] Lukanov T.H., Bojinova S.I., Popova V.S., Emin A.L., Veleva
G.L., Gecheva S.P., Konova E.I.: “Flow cytometric investigation of
CD40-CD40 ligand system in preeclampsia and normal pregnancy”.
Clin. Appl. Thromb. Hemost., 2010, 16, 306.
[23] Alacacioglu I., Ozcan M.A., Piskin O., Yuksel F., Alacacioglu A.,
Demirkan F., et al.: “Increased concentration of soluble CD40 ligand
in preeclampsia”. Clin. Appl. Thromb. Hemost., 2007, 13, 201.
[24] Erez O., Romero R., Hoppensteadt D., Fareed J., Chaiworapongsa T.,
Kusanovic J.P., et al.: “ Premature labor: a state of platelet activa-
tion? ” J. Perinat.Med., 2008, 36, 377.
[25] Inwald D.P., McDowall A., Peters M.J., Callard R.E., Klein N.J.:
“CD40 Is constitutively expressed on platelets and provides a novel
mechanism for platelet activation”. Circ. Res., 2003, 92, 1041
[26] Konijnenberg A., Van der Post J.A., Mol B.W., Schaap M.C.,
Lazarov R., Bleker O.P.: “Can flow cytometric detection of platelet
activation early in pregnancy predict the occurence of preeclamp-
sia.Aprospective study”. Am. J. Obstet. Gynecol., 1997, 177, 434
[27] Azzam H.A., Abousamra N.K., Goda H., El-Shouky R., El-Gilany
A.H.: “The expression and concentration of CD40 ligand in normal
pregnancy, preeclampsia, and hemolytic anemia, elevated liver en-
zymes and low platelet count (HELLP) syndrome”. Blood Coagul.
Fibrinolysis, 2013, 24, 71.
[28] Wu C.F., Huang F.D., Sui R.F., Sun J.X.: “Preeclampsia serum up-
regulates CD40/CD40L expression and induces apoptosis in human
umbilical cord endothelial cells”. Reprod. Biol. Endocrinol., 2012,
10, 28
[29] Qın W., Lı C.P., Zhan Y., Ye Y.H., Cuı X.N.: “The correlation be-
tween soluble plasma cd40-cd40 ligand and morbidıty of preeclamp-
sia”. Acta Academiae. Medicinae. Qingdao. Universitatis, 2011, 5,
396.
Corresponding Author:
S. KUMBASAR, M.D.
Department of Obstetrics and Gynecology
Sakarya University School of Medicine
Sakarya Research and Education Hospital
Sakarya (Turkey)
e-mail: doktor1977@hotmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Case Reports

Abnormal bending of the umbilical cord due to adhesion

of the cord to the placenta

Tatsuya Ishiguro1,2, Takao Ishiguro1

1 Department of Obstetrics and Gynecology, Ladies Clinic Ishiguro, Ara-machi, Sanjo-city, Niigata
2 Department of Obstetrics and Gynecology, Niigata University Medical and Dental Hospital, Asahimachi-dori, Chuo-ku, Niigata (Japan)

Summary
Background: Although cord abnormalities can cause fetal distress, there are many cases of fetal distress caused by unknown factors.
Case: The mother was a 27-year-old Japanese woman. The umbilical cord was attached to nearly the center of the placenta, which was
smoothly delivered. Macroscopically, at the site of cord attachment to the placenta, the cord appeared partially flattened and adhered to
the placenta, resulting in abnormal bending of the cord. Pathological examination of the cord and placenta, including the site of adhe-
sion, did not show any remarkable findings. Therefore, the adhesion might have caused temporary bending of the cord, which resulted
in fetal distress. Conclusion: The authors encountered a rare case of abnormal adhesion of the umbilical cord to the placenta that caused
fetal distress. The presence of abnormalities of the placenta and umbilical cord should be macroscopically examined immediately after
delivery, even when only mild fetal distress is noted.

Ker words: Umbilical cord bending; Fetal distress.

Introduction second stage of labor, a severe variable deceleration pattern with

Nuchal cord, cord strictures, and velamentous insertion
are some of the major cord abnormalities that can cause
fetal distress, and there are many cases of fetal distress
caused by unknown factors [1]. Ultrasonography during
pregnancy can detect many abnormalities of the umbilical
cord and the placenta, such as placental hematoma, lake,
and cyst [1-4]. The early detection of these abnormalities
allows for appropriate management during pregnancy to
minimize the risk of life-threatening complications in the
mother and fetus. The authors encountered a unique case
of abnormal adhesion of the umbilical cord to the placenta
that caused fetal distress after the first stage of labor.
Case Report
A 27-year-old Japanese woman (primigravida) visited the pres-
ent clinic at 32 weeks of gestation. She was spontaneously preg-
nant and no remarkable abnormalities were noted during
pregnancy at previous hospitals in the United States and Japan.
Routine ultrasonography did not detect any abnormality of the
fetus, umbilical cord, or placenta. At 40 weeks of gestation, she
was admitted to the present clinic owing to onset of labor. How-
ever, after 36 hours, an ecbolic was used and oxytocin was ad-
ministered because of feeble labor pain. At the first stage of labor,
cardiotocography showed that the fetal heartbeat had a mild vari-
able deceleration pattern with a minimum heartbeat of 90
beats/min; however, the heartbeat recovered. Subsequently, at the
a minimum heartbeat of 70 beats/minute was noted. Using vac-
uum extraction, a female baby was delivered. Her birth weight
was 3,025 grams, and Apgar scores were 8 points at one minute
and 9 points at five minutes. In the umbilical artery, the pH was
7.255, the base excess was -5.1 mmol/l, and the hemoglobin con-
centration was 16.8 g/dl
The oval meconium-stained yellow placenta measured 20×16.5
×1.5 cm, with a weight of 510 grams. The umbilical cord, which
measured 52 cm in length and 1.2 cm in diameter without abnor-
mal cord coiling, was attached to nearly the center of the placenta,
which was smoothly delivered. Macroscopically, at the site of
cord attachment to the placenta, the cord appeared partially flat-
tened and adhered to the placenta, resulting in abnormal bending
of the cord (Figure 1). The flat part of the cord was easily detached
from the placenta. Pathological examination of the cord and pla-
centa, including the site of adhesion, showed chorioamnionitis of
Grade 3, funisitis of Grade 2, and narrowing of the intervillious
space without massive infarction; the umbilical cord had normal
Wharton’s jelly around three umbilical blood vessels. Therefore,
the adhesion might have caused temporary bending and com-
pression of the cord, which was enhanced by uterine contraction;
this event resulted in fetal distress.
Discussion
Fetal distress due to cord abnormalities is commonly en-
countered during pregnancy. Cord abnormalities related to
morphology, coiling, placental insertion, number of ves-
sels, and diameter can be associated with perinatal compli-

Revised manuscript accepted for publication May 9, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3727.2017
788 Tatsuya Ishiguro, Takao Ishiguro

Figure 1. — Macroscopic
image of the placenta and
umbilical cord. The flat sur-
face of the umbilical cord
(black dotted line) adheres to
the surface of the placenta
(yellow dotted line). The as-
terisk indicates the site at
which bending of the cord
was noted.

cations [5]. A unique mucopolysaccharide-rich membrane References

known as Wharton’s jelly cushions the umbilical blood ves-
sels, preventing disruption of the flow due to compression
or bending caused by fetal movements and uterine con-
traction [6]. The reduction of this jelly leads to fetal growth
retardation due to hypoplasia of the umbilical vessels, and
a lean umbilical cord is associated with fetal distress [1, 6].
In most pregnant cases, the umbilical cord inserts to near
the center of the placenta. Abnormal cord insertion includ-
ing marginal cord and velamentous cord insertion, is asso-
ciated with fetal growth retardation, intrauterine fetal
demise, and neonatal demise [1]. In the present case, unique
abnormal adhesion of the umbilical cord to the placenta
was found after delivery. Unfortunately, the cause and
mechanism of this adhesion was unclear because the patho-
logical examination showed only chorioamnionitis and fu-
nisitis. To the present authors’ knowledge, no similar cases
have been reported so far. Nevertheless, a healthy baby
without intrauterine growth restriction was born because of
only temporal cord bending during labor.
The detection of abnormal adhesion of the umbilical cord
to the placenta during pregnancy, as in the present case, is
difficult, even with the use of high-performance ultra-
sonography. The presence of abnormalities of the placenta
and umbilical cord should be macroscopically examined
immediately after delivery, even when only mild fetal dis-
tress is noted.
[1] Moshiri M., Zaidi S.F., Robinson T.J., Bhargava P., Siebert J.R.,
Dubinsky T.J., Katz D.S.: “Comprehensive imaging review of
abnormalities of the umbilical cord”. Radiographics, 2014, 34,
179.
[2] Doehrman P., Derksen B.J., Perlow J.H., Clewell W.H., Finberg H.J.:
“Umbilical artery aneurysm: a case report, literature review, and
management recommendations”. Obstet. Gynecol. Surv., 2014, 69,
159.
[3] Bowman Z.S., Kennedy A.M.: “Sonographic appearance of the pla-
centa”. Curr. Probl. Diagn. Radiol., 2014, 43, 356.
[4] Abramowicz J.S., Sheiner E.: “Ultrasound of the placenta: a sys-
tematic approach. Part I: Imaging”. Placenta, 2008, 29, 225.
[5] Predanic M.: “Sonographic assessment of the umbilical cord”. Don-
ald School J. Ultrasound Obstet. Gynecol., 2009, 3, 48.
[6] Di Naro E, Raio L., Cromi A., Giocolano A.: “Sonographic assess-
ment of the umbilical cord”. Donald School J. Ultrasound Obstet.
Gynecol., 2012, 6, 66.
Corresponding Author:
TATSUYA ISHIGURO, M.D.
Department of Obstetrics and Gynecology
Niigata University Medical and Dental Hospital
1-757, Asahimachi-dori, Chuo-ku
Niigata 951-8510 (Japan)
e-mail: tishigur@med.niigata-u.ac.jp
CEOG Clinical and Experimental
Obstetrics & Gynecology
Umbilical endometriosis: a rare case of spontaneous
cutaneous umbilical endometriosis
M. Kalinderis, U. Singh
Department of Obstetrics & Gynaecology, Darent Valley Hospital, Dartford, Kent (United Kingdom)
Summary
Umbilical endometriosis is an extremely rare type of endometriosis, with an incidence of 0.5% to 1% of all endometriosis cases. This
report describes a case of umbilical endometriosis that developed in the absence of previous abdominal or uterine surgery. Clinical di-
agnosis of umbilical endometriosis is difficult and the differential diagnosis of umbilical lesions can be confusing, thus a high level of
clinical suspicion is required.
Key words: Villar’s nodule; Endometriosis; Umbilical lesion.
Introduction
Endometriosis is defined by the presence of endometrial
glands and stroma outside the endometrial cavity. It is a com-
mon condition that affects about 10% of women of repro-
ductive age and 35–50% of women with pelvic pain and
infertility [1]; however its cause remains currently poorly un-
derstood. The classic triad of symptoms dysmenorrhea, dys-
pareunia, and dyschezia is highly suggestive of
endometriosis. Endometriosis is an estrogen-dependent dis-
ease that can be pharmacologically or/and surgically treated
[2].
Endometriosis is dominantly found in the pelvis, but can
also develop outside the uterus or ovary and is called ec-
topic or extra-gonadal/extragenital endometriosis. Extra-
gonadal endometriosis, although rarely described, its seed
can be found almost throughout the female body including
bowel, bladder, lungs, brain, umbilicus, and surgical scars
[3]. This is a case report study describing a cutaneous um-
bilical endometriosis in a surgically naïve young woman.
Case Report
A 41-year-old Asian woman initially presented to the surgical
department complaining of umbilical non-reducible painful
swelling, associated with purulent and bloody discharge over an
18-month period. She was previously treated with meticulous hy-
giene and courses of antibiotics without any improvement.
After clinical examination, the surgeon suspected an umbilical
hernia. Under general anesthesia, exploration of umbilicus was
offered and dissection down to the sheath and around the umbili-
cus was performed. A small hernia was repaired and the affected
area was excised and was sent for histological examination. The
abdominal defect was repaired with Vicryl one on J-shape needle
Revised manuscript accepted for publication July 18, 2016
Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3839.2017
in two layers and the umbilicus was reconstructed. The postoper-
ative period was uneventful and the patient was discharged home
the following day. After the histology results that reported fibro-
adipose tissue, with endometriosis associated with haemorrhage
and chronic inflammation, the patient was referred to gynaecology
clinic complaining of recurrent discharge from the umbilicus.
The patient had two uncomplicated vaginal deliveries with no
history of subfertility. Her menstrual cycles were regular, how-
ever associated with excruciating painful and heavy periods. She
also reported deep dyspareunia. She did not report any intermen-
strual or postcoital bleeding and she was up to date with the
smears. She did not provide history of urinary or bowel symp-
toms. The past medical history was unremarkable and she was
surgically naïve before the excision of the granulomatous lesion.
From detailed history umbilical bleeding was reported to in-
crease during menstruation. On physical examination abdomen
was soft, not tender. Gynaecological examination revealed an an-
teverted, non-tender, mobile, and normal-sized uterus. Adnexal
masses were not palpable on either side and no nodules were felt
in the uterosacral ligaments or in the rectovaginal space. Gynae-
cological ultrasound scan was performed in which the anteverted
uterus appeared normal in size and shape. The myometrium ap-
peared heterogeneous and contained a small myometrium cyst;
the appearance was suggestive of adenomyosis. The endometrium
was regular in outline and within normal limits measuring 8.3
mm. Both ovaries appear sonographically normal. The right ovary
measured 3×2.8×2.6 cm and contained a 2-cm normal dominant
follicle. The left ovary measured 2.4×3×2.8 cm. No adnexal cyst
or mass was seen, as well as no free fluid.
Discussion
Umbilical endometriosis is a rare manifestation of en-
dometriosis, with an estimated incidence of 0.5–1% of all
endometriosis cases and up to 30–40 % of cases with cuta-
neous endometriosis, only 15% of which is associated with
790 M. Kalinderis, U. Singh
pelvic endometriosis [4]. Umbilical endometriosis is also
known as Villar’s nodule as is attributed to a physician who
first described the disease in 1886 [5]. From 1996 to 2007,
Victory et al. found only 122 reported cases of umbilical
endometriosis worldwide [6]. It can be divided into primary
and secondary. Primary umbilical endometriosis arises de
novo. Spontaneous or primary umbilical endometriosis, oc-
curring without any previous abdominal or uterine surgery,
is extremely rare. The development of secondary umbili-
cal endometriosis more often appears after a surgery, espe-
cially laparoscopic surgical procedures involving the
umbilicus [4]. The present patient presented with a negative
surgical history, which rendered the pathophysiology of the
disease even more enigmatic.
Multiple theories have been proposed as causes of en-
dometriosis. Sampson in 1927 was the first to propose the
retrograde menstruation or implantation theory suggesting
that during menstruation blood refluxes through the fal-
lopian tubes and endometrial tissue implants into the nearby
organs. However, this theory has been disputed in the past
since retrograde menstruation occurs in 76–90% of women
with patent fallopian tubes and only a minority will develop
the disease; this theory cannot also explain the occurrence
of endometriosis in pre-pubertal girls, newborns or males.
A direct transportation of endometrial cells via blood or
lymph vessels or even through surgical manipulations has
also been proposed [1]. According to the theory of meta-
plasia, endometriosis originates from extrauterine cells that
abnormally transform into endometrial cells. The Coelomic
metaplasia theory proposes metaplasia of peritoneal
mesothelial tissue cells into endometrial cells. Hormones,
especially estrogen, are thought to stimulate this transfor-
mation. The embryonic rest theory proposes that the pres-
ence of cells of Müllerian origin within the peritoneal
cavity could be induced to form endometrial tissue when
subjected to the appropriate stimuli [1]. Inflammation and
oxidative stress may contribute to the pathogenesis of en-
dometriosis, as well as apoptosis suppression and survival
of endometrial cells, immunological dysfunction, and ge-
netic predisposition. Moreover specific attention has been
paid to the role of stem cells through endometrial self-gen-
eration in specific niches of the endometrium [6].
In the development of spontaneous umbilical en-
dometriosis, as in the case presented, some etiologies have
been proposed, but none of them can entirely explain the
appearance of the tumor. It is possible that the umbilicus,
considered a physiologic scar, could have tropism for en-
dometrial cells [7]. It has been also suggested that isolated
umbilical endometriosis may arise through metaplasia of
urachal remnants [8]. The theory of lymphatic or vascular
transportation and reimplantation at the umbilicus could
also be possible.
Umbilical endometriosis is commonly found in the re-
productive age group. The mean age of diagnosis has been
reported to be 37.7 years with the youngest being 23 years
[5]. Typical symptoms involve cyclical swelling, pain, and
bleeding from umbilicus, however there are case reports
were the main symptom is not bleeding [9, 10]. Clinically,
it presents as a reddish-brown painful nodule with cyclic
variations in size with or without bleeding that many coin-
cide or not with patient’s menstrual cycle. Its size ranges
from 0.5-3 cm, but larger masses have been described, as
well [11]. The diagnosis is primarily clinical, but often puz-
zling and misguiding. Umbilical endometriosis can be mis-
diagnosed in 20-50% of the cases. The differential
diagnoses of umbilicus endometriosis should include
mainly melanoma, umbilical metastases (Mary Sister
Joseph Nodule), and pyogenic granuloma. Lipoma, ab-
scess, omphalitis, cyst, hernia, various granulomas, urachal
lesions, nodular melanoma, primary or metastatic carci-
noma, and keloid and residual embryonic tissue should be
considered in the differential diagnosis, as well [8]. A
biopsy is mandatory especially when melanoma cannot be
ruled out. In a case of umbilical endometriosis maligniza-
tion can occur, however the risk of malignancy is reassur-
ingly low. Dermoscopy may be used as an auxiliary tool,
however the diagnosis is often made incidentally by histo-
logical examination after surgical exploration and excision
of the lesion.
Umbilical endometriosis treatment involves medical
management such as progestational drugs, danazol, and
GnRH analogues or surgical excision. Medical treatment
could potentially be a pre-treatment option especially in
cases of large tumours. However, surgical excision with
complete wide resection of the lesion with free margins and
minimal spillage is considered as a first line treatment. If
the umbilicus cannot be preserved then reconstruction is
recommended in conjunction with plastic surgeons. Mesh
is required in cases of abdominal wall defects, however the
possibility of mesh contamination by endometriosis needs
further evaluation. Follow up following excision is para-
mount and is recommended every six months for a period
of two years. The patient should be informed about the risk
of malignancy and local recurrence.
Conclusion
Cutaneous endometriosis is a rare skin pathology that
may present to the gynaecologist, general surgeon, derma-
tologist or plastic surgeon. This report describes a case of
umbilical endometriosis that developed in the absence of
previous abdominal or uterine surgery. Clinical diagnosis is
difficult and umbilical endometriosis may go unrecognized
because of its rarity, leading to multiple medical visits and
a delayed diagnosis. A definite diagnosis can only be estab-
lished by histopathological examination. Its pathogenesis
remains currently uncertain. Since pelvic endometriosis may
be present, referral to a gynaecologist is recommended in
every case. Increased clinical surveillance and suspicion is
required for the diagnosis of endometriosis, a multifaceted
 Umbilical endometriosis: a rare case of spontaneous cutaneous umbilical endometriosis
and enigmatic disease with variable appearance.
Acknowledgements
The authors thank Dr. Kallirhoe Kalinderi for the critical
reviewing of the manuscript
References
[1] Vercellini P., Viganò P., Somigliana E., Fedele L.: “Endometrio-
sis: pathogenesis and treatment”. Nat. Rev. Endocrinol., 2014, 10,
261.
[2] Kodaman P.H.: “Current strategies or endometriosis management”.
Obstet. Gynecol. Clin. North Am., 2015, 42, 87.
[3] Agarwal N., Subramanian A.: “Endometriosis morphology, clinical
presentations and molecular pathology”. J. Lab. Physicians, 2010, 2,
1.
[4] Pramanik S.R., Mondal S., Paul S., Joycerani D.: “Primary umbili-
cal endometriosis: a rarity”. J. Hum. Reprod. Sci., 2014, 7, 269.
[5] Victory R., Diamond M.P., Johns D.A.: “Villar’s nodule: a case re-
port and systematic literature review of endometriosis externa of the
umbilicus”. J. Minim. Invasive Gynecol., 2007, 14, 23.
791
[6] Sourial S., Tempest N., Hapangama D.K.: “Theories on the patho-
genesis of endometriosis”. Int. J. Reprod. Med., 2014, 2014, 179515.
[7] Attia L., Ben Temime R., Sidhom J., Sahli A., Makhlouf T., Chachia
A., et al.: “A case of cutaneous endometriosis development on an
abdominal scar”. Tunis Med., 2010, 88, 841.
[8] Frischknecht F., Raio L., Fleischmann A., Dreher E., Luscher K.P.,
Mueller M.D.: “Umbilical endometriosis”. Surg. Endosc., 2004, 18, 345.
[9] Chatzikokkinou P., Thorfinn J., Angelidis I.K., Papa G., Trevisan G.:
“Spontaneous endometriosis in an umbilical skin lesion”. Acta Der-
matovenerol. Alp. Panonica Adriat., 2009, 18, 126.
[10] Dessy L.A., Buccheri E.M., Chiummariello S., Gagliardi D.N., On-
esti M.G.: “Umbilical endometriosis, our experience”. In Vivo, 2008,
22, 811.
[11] Latcher J.W.: “Endometriosis of the umbilicus”. Am. J. Obstet. Gy-
necol., 1953, 66, 161.
Corresponding Author:
M. KALINDERIS M.D., PhD
Department of Obstetrics & Gynaecology
Darent Valley Hospital
Darent Wood Road
Dartford, Kent DA2 8DA (United Kingdom)
e-mail: m.kalinderis@hotmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Successful single-port laparoscopic management

of abdominal pregnancy in the Douglas pouch

X. Yang, K. Ma
Department of Obstetrics and Gynecology, Beijing Tsinghua Changgung Hospital Medical Center, Tsinghua University, Beijing (China)

Summary
Abdominal pregnancy is estimated to account for 1.3% of all ectopic pregnancies. Laparoscopic treatment of intact and even ruptured
abdominal pregnancies is becoming more common. Here, the authors report the first single-port laparoscopic resection of an abdomi-
nal pregnancy, providing a detailed description of the procedures. At laparoscopy, a dense tissue clot was found implanted in the Dou-
glas pouch with no signs of tubal abortion. The diagnosis of abnormal pregnancy was made and the tissue was removed successfully.
In conclusion, we could consider that single-port laparoscopic surgery for an abdominal pregnancy is a feasible surgical technique.

Key words: Abdominal pregnancy; Ectopic pregnancy; Single-port laparoscopy.

Introduction
Ectopic pregnancy, the implantation of a fertilized ovum
outside the endometrial cavity, occurs in 1.5–2% of all
pregnancies [1]. Abdominal pregnancy is estimated to ac-
count for 1.3% of all ectopic pregnancies and has a high
associated mortality rate of 5–6% [2]. This rate may be high
because early diagnosis is difficult, and the condition is
often not discovered until the crisis of internal bleeding oc-
curs. The most common site of implantation in ectopic
pregnancies is the fallopian tube, with an incidence of up to
95% [1]. The first case of implantation in the Douglas
pouch, which is extremely rare, was described by Galabin
in 1896 [3], and most cases are treated by laparotomy.
With the increasing availability of advanced equipment
and expertise, laparoscopic treatment of intact and even
ruptured abdominal pregnancies is becoming more com-
mon. In the past several years, many gynecologic surgeons
attempted to improve cosmetic results and reduce postop-
erative hospital stay after laparoscopic surgery. Transum-
bilical single-port access laparoscopic surgery is carried out
through a small incision in the umbilicus, resulting in a vir-
tually invisible scar. This kind of single-port system has
made many gynecological surgeries feasible. Here, the au-
thors report the first single-port laparoscopic resection for
an abdominal pregnancy, providing a detailed description
of the procedures.
Case Report
A 30-year-old woman, gravid 0, para 0, was referred to the pres-
ent outpatient clinic with a suspected diagnosis of ectopic preg-
nancy. Her last menstrual period was five weeks ago. Her initial
β-hCG level was 5474 mIU/ml. Her vital signs were stable at pres-
entation, with a hemoglobin level of 11.7 g/dl. Transvaginal ul-
trasound showed a sac-like mass (1.7×1.6 cm) located in the
posterior Douglas pouch (Figure 1), with no signs of intrauterine
pregnancy. An MRI image of the abdomen and pelvis showed the
extrauterine gestational sac (1.5×1.5 cm) in the Douglas pouch
(Figure 2). The serum β-hCG level increased to 6,800 mIU/ml
two days later. After consultation, the patient consented to a la-
paroscopic examination. A single-port laparoscopic operation was
planned to prevent further enlargement of the gestational sac and
the risk of rupture.
The operation was completed with the patient under general
anesthesia and in the dorsal lithotomy position. After sterile cov-
erage, the authors inserted a uterine manipulator into the uterine
cavity and used a single incision laparoscopic surgery (SILS) port.
After a 2.5-cm skin incision was made at the umbilicus, the sub-
cutaneous fat tissues were dissected and the peritoneum was
opened. A SILS port wound retractor (30 mm) was placed in the
peritoneal cavity covering the skin to the peritoneum (Figure 3).
The SILS port has three access ports (one 12-mm port and two
five-mm ports) as well as a gas inlet. The authors used a rigid, 30°,
ten-mm laparoscope, conventional rigid straight instruments, and
pre-bent laparoscopic instruments (Figure 4).
At laparoscopy, a dense tissue clot (about two cm) was found
implanted in the Douglas pouch, with mild bleeding. Bilateral
oviducts and ovaries were intact and grossly normal with no signs
of tubal abortion (Figure 5). The tissue was removed using grasp-
ing forceps and hydro-dissection. After removal of necrotic tis-
sue, a peritoneal defect (approximately 3×2 cm, and 0.5 cm deep)
was seen in the Douglas pouch, with active bleeding. Bipolar elec-
trocauterization was used to achieve hemostasis, and the peri-
toneal defect was tamponaded with hemostatic material to stop
the bleeding.
The patient recovered rapidly and was discharged on the second
postoperative day. Her serum β-hCG levels reduced to 2,107
mIU/ml on the first postoperative day and to 522 mIU/ml two
days later. On the 15th postoperative day, this level reduced to nor-

Revised manuscript accepted for publication November 16, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3463.2017
 X. Yang, K. Ma 793

Figure 1. — Transvaginal ultrasound showing a gestational sac- Figure 2. — Axial MRI of the abdomen and pelvis obtained using
like mass in the Douglas cul-de-sac (black arrow). the True FISP sequence showing the extrauterine gestational sac
(black arrow), empty uterus (asterisk), and bilateral ovaries (white
arrows).

Figure 3. — The SILS port was inserted through the incision. Figure 4. — External view during single-port access transumbil-
ical surgery.

Figure 5. — Laparoscopic view of abdominal pregnancy. The Figure 6. — Postoperative wound on the umbilicus is an incision
mass containing the ectopic pregnancy can be seen in the Dou- of approximately 18 mm (one month after surgery).
glas pouch (black arrow).
794 Successful single-port laparoscopic management of abdominal pregnancy in the Douglas pouch
mal (< 5 mIU/ml). The pathology report confirmed intact tro-
phoblasts. Transvaginal ultrasound showed slight pelvic effusion
in the Douglas pouch one month after surgery and the postopera-
tive wound on the umbilicus was only a skin incision of approx-
imately 18 mm (Figure 6).
Discussion
Most ectopic pregnancies occur in the ampullary segment
of the fallopian tube. However, they may also occur within
the interstitial portion of the fallopian tube, in the uterine
cervical canal, between the leaves of the broad ligament,
in the ovary, within a scar from a cesarean section, or in the
abdomen. These unusual ectopic pregnancies are difficult to
diagnose and are associated with significant morbidity and
mortality [1, 4].
Abdominal pregnancy is the rarest form of ectopic preg-
nancy, occurring in 1.3% cases, and its associated mortal-
ity rate is seven times higher than that in non-abdominal
cases [2]. The reported sites of abdominal pregnancy are
the Douglas pouch, posterior uterine wall, uterine fundus,
infundibulopelvic ligaments, anterior abdominal wall,
omentum, liver, spleen, lesser sac, and diaphragm [4, 5].
Abdominal pregnancies are classified as either primary or
secondary. It may be difficult to differentiate between these
classes, because the original site of implantation cannot be
accurately determined. Studdiford set forth the following
criteria for the diagnosis of primary abdominal pregnancy:
both tubes and ovaries are normal, with no evidence of re-
cent or remote injury; absence of any uteroperitoneal fis-
tula; and presence of a pregnancy related exclusively to the
peritoneal surface and young enough to eliminate the pos-
sibility of secondary implantation following primary im-
plantation in the tube [6]. As the original site of
implantation may be difficult to determine, it has been sug-
gested that a true primary abdominal pregnancy can be di-
agnosed only when the gestational age is less than ten
weeks [7]. Studdiford’s criteria were later modified by
Friedrich and Rankin as follows: a pregnancy of less than
12 weeks’ histologic gestation, whose trophoblastic attach-
ments are related solely to a peritoneal surface; grossly nor-
mal tubes and ovaries; and absence of any uteroperitoneal
fistula [8, 9]. The present case fulfills both the original and
modified criteria for primary abdominal pregnancy in the
Douglas pouch, and the pathology report confirmed intact
trophoblasts without the degenerative changes that are
found in the tissues of tubal abortion.
The traditional management of abdominal pregnancy in-
volves laparotomy with removal of the embryo with or
without placental tissue. Laparoscopic management has not
been used often, because controlling hemorrhagia can be
difficult because of trophoblastic invasion of the retroperi-
toneal vasculature. The reported instances of laparoscopic
management are associated with a shorter operative time
and reduced blood loss [5]. This approach is cost effective
and should be the treatment of choice, except in cases of
extensive intraperitoneal bleeding, intravascular compro-
mise, or poor visualization of the pelvis at the time of la-
paroscopy, for which laparotomy cannot be avoided.
Recently, gynecologic surgeons have attempted to per-
form single-port transumbilical laparoscopic surgery, and
this is rapidly evolving and under active investigation as a
routine surgery. In the present clinic as well, the authors
have performed many single-port laparoscopic operations,
including adnexectomy and cystectomy for adnexal cysts
and salpingectomy for tubal pregnancies. To our knowl-
edge, this is the first report of successful single-port la-
paroscopic resection for an early abdominal pregnancy.
Although abdominal pregnancy is an exceptional condition,
in a patient with clinical findings suggesting ectopic preg-
nancy, if both the uterus and adnexa are found to be normal
during laparoscopic exploration, unusual locations such as
the Douglas pouch or retroperitoneum should be carefully
examined. The authors used bipolar electrocauterization
and tamponading with a hemostatic material to achieve
complete hemostasis in the peritoneal defect after removal
of the placenta. However, hemostasis at surgery may also
be achieved with the use of vasopressin, bipolar electrodes,
and monopolar scissors.
References
[1] Barnhart K.T.: “Ectopic pregnancy”. N. Engl. J. Med., 2009, 361,
379.
[2] Martin J.N. Jr., Sessums J.K., Martin R.W., Pryor J.A., Morrison
J.C.: “Abdominal pregnancy: current concepts of management”. Ob-
stet. Gynecol., 1988, 71, 549.
[3] Galabin A.L.: “Primary abdominal pregnancy”. Br. Med. J., 1903, 1,
664.
[4] Lee J.W., Sohn K.M., Jung H.S.: “Retroperitoneal ectopic preg-
nancy”. Am. J. Reprod., 2005, 184, 1600.
[5] Chetty M., Elson J.: “Treating non-tubal ectopic pregnancy”. Best
Pract. Res. Clin. Obstet. Gynecol., 2009, 23, 529.
[6] Studdiford W.E.: “Primary peritoneal pregnancy”. Am. J. Obstet. Gy-
necol., 1942, 44, 487.
[7] Makinen J.: “Histologically verified primary peritoneal pregnancy
with implantation in the sigmoid mesenterium”. Eur. J. Obstet. Gy-
necol. Reprod. Biol., 1986, 22, 171.
[8] Friedrich E.G., Rankin C.A.: ‘Primary pelvic pregnancy”. Obstet.
Gynecol., 1968, 31, 649.
[9] Joong S.S., Young J.M., Seung R.K., Kyung T.K., Hyung M., Youn
Y.H.: “Primary peritoneal pregnancy implanted on the uterosacral
ligament: a case report”. J. Korean Med. Sci., 2000, 15, 359.
Corresponding Author:
K. MA, M.D.
Department of Obstetrics and Gynecology
Beijing Tsinghua Changgung
Hospital Medical Center, Tsinghua University
No.168 Li Tang Road, Dongxiaokou Town
Changping District, Beijing (China)
e-mail: markpolo126@126.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

A case of disseminated intravascular coagulation developed

after surgical management of corpus luteal hemorrhage in
a patient with Klippel-Trenaunay syndrome

S.E. Han, Y.H. Kim, S.C. Kim, J.K. Joo, D.S. Suh, K.H. Kim, K.S. Lee
Department of Obstetrics and Gynecology, Medical Research Instutite, Pusan National University Hospital,
Pusan National University School of Medicine, Busan (Korea)

Summary
Klippel-Trenaunay syndrome (KTS) is a complex congenital disorder characterized by a triad of varicose veins, cutaneous capillary
malformation, and hypertrophy of bone and/or soft tissue. KTS may be associated with massive hemorrhage or coagulopathy that be a
life-threatening situation. Although women in reproductive age are at risk of ruptured corpus luteum with active arterial bleeding, if it
managed properly, the development of serious complications, such as disseminated intravascular coagulation (DIC) rarely develops.
However, in case of patient with vascular malformation, there is possibility of unexpected complication occurrence such as DIC. The
authors report a case of a 29-year-old female with KTS who presented with corpus luteal hemorrhage and which lead to DIC, despite
adequate surgical and medical treatment.

Key words: Klippel-Trenaunay syndrome; Congenital vascular anomaly; Corpus luteum; Hemoperitoneum.

Introduction Case Report

Klippel-Trenaunay syndrome (KTS) is a rare congenital
vascular malformation characterized by the clinical triad of
bony or soft tissue hypertrophy, usually affecting one ex-
tremity; hemangiomas and/or lymphangiomas, and vari-
cosities or venous malformations [1]. Vascular
malformation of KTS can involve several organs and be a
source of significant morbidity and even mortality [2].
Clinical manifestations of KTS range from occult to mas-
sive, life-threatening hemorrhage [3]. Several cases of life-
threatening gastrointestinal bleeding in KTS patients have
been reported [3-6]. In contrast, large venous malforma-
tions can be associated with low-grade consumptive coag-
ulopathy [7, 8].
Spontaneous hemoperitoneum may occur in various gy-
necological conditions. The most common gynecological
causes of spontaneous hemoperitoneum in women of child-
bearing age are ectopic pregnancy and ruptured corpus
luteal cyst [9]. The treatment of spontaneous hemoperi-
toneum of gynecological causes is well established. How-
ever in patients with coagulopathy, the management has to
be planned differently.
The authors present a case of a 29-year-old female with
KTS who presented with a spontaneous hemoperitoneum
by corpus luteal hemorrhage and subsequent life-threaten-
ing disseminated intravascular coagulation (DIC) after
ovarian cystectomy.
A 29-year-old female presented to the emergency room with
sudden onset abdominal pain. She had a history of multiple ad-
missions for cutaneous large hemangioma, presenting mainly as
hemangioma bleeding that started at the age of 13 and diagnosed
KTS. She had regular menstruations since the age of 13, lasting
three to five days, occurring every 28 to 30 days. She presented
with tachycardia (pulse rate 112 beats per minute), abdominal dis-
tension, and moderate rigidity on the lower quadrants on palpation
on physical examinations. In addition, blood pressure (BP) was
90/60 mmHg and general appearance was acute ill-looking but alert
and oriented. Laboratory results showed anemia (hemoglobin 7.6
g/dl) and slight elevation of PT (INR 1.40) and aPTT time (44.6
sec). Transabdominal ultrasound revealed a moderate amount of
fluid in the abdomen and in the pouch of Douglas and a seven-cm
sized right adnexal mass with signs of peripheral vasculatization.
An abdomen CT scan showed hemoperitoneum due to rupture of
hemorrhagic cyst in right adnexa and diffuse hemangioma with ve-
nous malformation in left side of body (Figure 1).
The patient was taken to the operating room with a presumptive
diagnosis of a ruptured hemorrhagic cyst. She underwent ex-
ploratory laparotomy via Pfannenstiel’s skin incision on right side
of lower abdomen. There was a massive hemorrhage due to cor-
pus luteum cyst rupture on the surface of the right ovary. Both fal-
lopian tubes, left ovary, and uterus appeared to be normal. During
the surgery, an amount of hemorrhagic fluid was evacuated from
the abdominal cavity. Hemostatic electric coagulation was applied
to the bleeding site and right ovarian wedge resection was per-
formed. In addition, five units of packed red blood cell and five
units of fresh frozen plasma were transfused during the surgery.
Postoperative course was complicated by hemoperitoneum,
worsening coagulopathy, sepsis, and DIC. The enhanced CT

Revised manuscript accepted for publication December 9, 2015

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3494.2017
796 S.E. Han, Y.H. Kim, S.C. Kim, J.K. Joo, D.S. Suh, K.H. Kim, K.S. Lee
scan showed active contrast leakage in pelvic cavity and hemo-
peritoneum around uterus (Figure 2). The authors attempted to
manage hemorrhage in regards to the treatment of DIC. How-
ever, active arterial bleeding in cystectomy site was observed on
CT follow-up. The follow-up secondary laparotomy was decided
on the next day. A right oophorectomy was performed and the
bleeding site was sutured to manage hemorrhage. Immediate
postoperative hemoglobin was 7.9 g/dl and platelet count was
60,000/ul and the patient received massive transfusion after sur-
gery due to chronic DIC with multiple hemangioma in KTS syn-
drome. Although active bleeding was controlled and stable vital
sign was maintained after the re-operation, the patient was trans-
Figure 1. — Abdominal computed tomography scan. (A) Hem-
orrhagic cyst in right adnexa (arrow, 6.8-cm sized mass). (B) Dif-
fuse hemangioma in left side of body. Soft tissue thickening
associated vascular malformation in the abdominopelvis wall. (C)
Coronal view; hemoperitoneum due to rupture of hemorrhagic
cyst.
ferred to the department of internal medicine because of contin-
uous bleeding in the drain site due to DIC, and was eventually
discharged after three months.
Discussion
KTS is a rare congenital disorder with variable clinical
presentation related to malformations of blood and lymph
vessels, and disturbed growth of bone and soft tissue [1].
KTS is a complex congenital syndrome which is associated
with life threatening complications like bleeding in geni-
 A case of disseminated intravascular coagulation developed after surgical management of corpus luteal hemorrhage in a patient etc.
Figure 2. — Abdominal computed tomography scan (arrow indi-
cates active contrast leakage in the right ovary and in the left
pelvic cavity).
tourinary system, spleen, liver, gastrointestinal tract, or cen-
tral nervous system [10]. Several cases of life-threatening
situations due to bleeding in KTS patients have been re-
ported. There are several treatments such as sclerotherapy,
excision of vasricose vessel, and conservative therapy that
are attempted in the patients, yet, no appropriate treatment
has been determined. In the previous reports, organ in-
volvements of hemangioma were reported, not corpus
luteal hemorrhage in KTS [3, 6]. Corpus luteum hemor-
rhage may occur spontaneously or often triggered by coitus,
trauma, exercise, or vaginal examination. The risk of hem-
orrhagic complications of ovulation begins on the ovula-
tion day and extends throughout corpus luteal life span,
which is 14 days without pregnancy.
The treatment of corpus luteal hemorrhage includes both
conservative and surgical management. Conservative man-
agement is successful in the majority of patients with rup-
tured ovarian cysts with hemoperitoneum. However,
surgical intervention is absolutely indicated in presence of
low blood pressure and a large amount of hemoperitoneum
[11]. Surgical management includes the following: electro-
cauterization of the ovarian surface, cystectomy, wedge re-
section, and ovarian reconstruction.
The present authors performed ovarian cystectomy and
confirmed the absence of active bleeding before leaving the
operation room. After the first surgery, hemorrhage was ob-
served and it was considered due to the active bleeding
caused by DIC. After taking the volume of bleeding and
transfusion into consideration, coagulopathy in KTS might
be the main cause of bleeding.
In ovarian cystectomy, the suture in ovarian stroma is
required and this may increase the risk of bleeding in the
797
presence of DIC compared to when oophorectomy is per-
formed.
In the cases of women with coagulopathy, we have to con-
sider more bleeding preventive operative technique such as
oophorectomy, not cystectomy. Preserving ovarian function
must be considered in the surgery of young women who
have coagulopathy. However we have to bear in mind that
DIC can develop regardless of volume of bleeding in the
operation of the patient with coagulopathy.
References
[1] Jacob A.G., Driscoll D.J., Shaughnessy W.J., Stanson A.W., Clay
R.P., Gloviczki P.: “Klippel-Trenaunay syndrome: spectrum and
management”. Mayo Clin Proc., 1998, 73, 28.
[2] Wilson C.L., Song L.M., Chua H., Ferrara M., Devine R.M., Dozois
R.R., et al.: “Bleeding from cavernous angiomatosis of the rectum in
Klippel-Trenaunay syndrome: report of three cases and literature re-
view”. Am. J. Gastroenterol., 2001, 96, 2783.
[3] Wang Z.K., Wang F.Y., Zhu R.M., Liu J.: “Klippel-Trenaunay syn-
drome with gastrointestinal bleeding, splenic hemangiomas and left
inferior vena cava”. World J. Gastroenterol., 2010, 16, 1548.
[4] Herman R., Kunisaki S., Molitor M., Gadepalli S., Dillman J.R.,
Geiger J.: “Rectal bleeding, deep venous thrombosis, and coagu-
lopathy in a patient with Klippel-Trenaunay syndrome”. J Pediatr
Surg., 2012, 47, 598.
[5] Samo S., Sherid M., Husein H., Sulaiman S., Yungbluth M., Vainder
J.A.: “Klippel-Trenaunay Syndrome Causing Life-Threatening GI
Bleeding: A Case Report and Review of the Literature”. Case Rep.
Gastrointest. Med., 2013, 2013, 813653.
[6] Thosani N., Ghouri Y., Shah S., Reddy S., Arora G., Scott L.D.:
“Life-threatening gastrointestinal bleeding in Klippel-Trenaunay
syndrome”. Endoscopy. 2013, 45, E206.
[7] Mason K.P., Neufeld E.J., Karian V.E., Zurakowski D., Koka B.V.,
Burrows P.E.: “Coagulation abnormalities in pediatric and adult pa-
tients after sclerotherapy or embolization of vascular anomalies”.
AJR Am. J. Roentgenol., 2001, 177, 1359.
[8] Mazoyer E., Enjolras O., Laurian C., Houdart E., Drouet L.: “Coag-
ulation abnormalities associated with extensive venous malforma-
tions of the limbs: differentiation from Kasabach-Merritt syndrome”.
Clin. Lab. Haematol., 2002, 24, 243.
[9] Fiaschetti V, Ricci A, Scarano AL, Liberto V, Citraro D, Arduini S,
et al. Hemoperitoneum from corpus luteal cyst rupture: a practical
approach in emergency room. Case Rep. Emerg. Med., 2014, 2014,
252657.
[10] Oduber C.E., van der Horst C.M., Hennekam R.C.: “Klippel-Tre-
naunay syndrome: diagnostic criteria and hypothesis on etiology”.
Ann. Plast. Surg., 2008, 60, 217.
[11] Kim J.H., Lee S.M., Lee J.H., Jo Y.R., Moon M.H., Shin J., et al.:
“Successful conservative management of ruptured ovarian cysts with
hemoperitoneum in healthy women”. PLoS One, 2015, 10,
e0142287.
Corresponding Author:
J.K. JOO, M.D. Ph.D.
Department of Obstetrics and Gynecology
Pusan National University Hospital
179 Gudeok-ro, Seo-gu Busan-si
Busan (Korea)
e-mail: jongkilj@hanmail.net
CEOG Clinical and Experimental
Obstetrics & Gynecology

Breast capillary hemangioma at the tail of Spencer: a rare entity

A. Bothou1, I. Grammatikakis2, N. Evangelinakis2, C. Eftichiadis3, G. Iatrakis1, S. Zervoudis1

1 Rea Maternity Hospital and Technological Educational Institute (TEI), University of Athens, Athens
2 3rd Department of Obstetrics and Gynecology, University of Athens, “Attikon” Hospital, Athens
3 Department of Pathology, General Hospital of Attica (KAT), Athens (Greece)

Summary
A palpable breast lump is a frequent clinical finding and preoperative evaluation varies depending on its localization and character-
istics. Vascular tumors are rarely diagnosed especially regarding the tail of Spencer region. In general, they appear oval-shaped with well-
circumscribed margins, but their echostructure varies, and it might be quite difficult for the breast specialist to differentiate it from
complex cysts, fibroadenomas or some carcinomas. The authors describe a rare location of breast hemangioma with mammographic char-
acteristics that were suspicious for malignancy. There were no identifiable risk factors, no familial history of breast lumps, and patient
did not mention the intake of hormonal therapy. The lump was evaluated by mammography and breast ultrasonography, whereas due
to the high vascularity of the nodule, the decision not to perform fine-needle aspiration (FNA) was made and an excisional biopsy was
performed. The histological result was “breast capillary hemangioma”.

Key words: Breast hemangioma; Rare breast lump; Capillary hemangioma; Benign breast lump; Oval-shaped breast lump; Heman-
gioma.

Introduction
A palpable breast lump is a frequent clinical finding and
preoperative evaluation varies depending on its localiza-
tion and characteristics. Vascular tumors are rarely diag-
nosed especially regarding the tail of Spencer region [1].
Benign vascular masses are even rare [2], and in most cases
these vascular tumors are malignant. The authors report a
case of a capillary hemangioma at the tail of Spencer re-
gion evaluated by mammography and ultrasonography. Be-
cause the atypic and heterogenic aspect of the lump,
fine-needle aspiration (FNA) was avoided and an excision
biopsy was performed.

Case Report
Patient, 42-years-old, presented to the Breast Unit of the present
hospital complaining of a palpable mass over the left breast which
appeared a few months ago. The lump was painless, flexible, not
blocked, and its margins appeared to be irregular. There were no
signs of inflammation of the mass, it did not retract or ulcerate the
above skin, whereas the architecture of the breast was not affected.
The patient had never performed a mammography. The family his-
tory of the patient was free regarding malignancies, while patient’s
history included dyslipidemia under statins treatment from the last
year. The patient underwent a digital mammography, which con-
firmed the findings of clinical examination. In particular it de-
scribed a mass with irregular margins that displaces normal
parenchyma that was not in contact with pectoral muscles, without
any microcalcifications, and normal architecture of the rest of the
breast. A lymph node was also enlarged in the left axilla (Figure 1).
Since magnification mammography did not delineate the structure
of the mass (serous or solid content) and did not add any important Figure 1. — Mammography of the left breast.

Revised manuscript accepted for publication February 1, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3586.2017
 A. Bothou, I. Grammatikakis, N. Evangelinakis, C. Eftichiadis, G. Iatrakis, S. Zervoudis
Figure 2. — Magnification mammography of the palpable mass of
the left breast.
details (Figure 2), an ultrasound was performed. The ultrasound
revealed a 2.4×1.1 cm mass, non-homogenous, hyperechoic with
mild rim, significant vascularity with low resistance indices, dis-
cernible blood flow inside it, and no acoustic shadowing (Figure 3).
Furthermore, despite the irregular margins of the mass, it seemed
to be clearly separated from the adjacent parenchyma (well-cir-
cumscribed). Due to the high vascularity of the nodule, the deci-
sion no to perform FNA was made and an excisional biopsy was
performed. Furthermore, despite high probability that the mass was
benign, patient was requested to remove the breast nodule with no
further investigative procedures.
Grossly an encapsulated globular mass 8×5×4cm was dissected
through an elliptical incision over the palpable mass. Nodule was
identified over pectoral muscle approximately one cm beneath skin
and it was excised with macroscopically free surgical margins (Fig-
ure 4). Due to the small size of the excised specimen, no graft was
required for the reconstruction, but with glandular flaps. Single su-
tures were placed in the subcutaneous tissue and subcuticular suture
was placed at the skin. Excised nodule was sent to the pathologist.
Adequate adherent breast tissue was also removed, though without
any apparent macroscopic findings. Hematoxylin and Eosin stain
revealed dilated vascular channels congested with erythrocytes
lined with endothelial cells (Figure 5). No signs of atypia were iden-
tified in all sections studied. Endothelial cells of the blood vessels
exhibited eosinophilic cytoplasm, and in some sections there was
infiltration of the mass by eosinophils and lymphocytes. Few lu-
mens were thrombosed, hemosiderin was identified, whereas spin-
dle cells were in close proximity and in between there were spaces
containing small amount of blood. Thus, the diagnosis of capillary
hemangioma was made. No local recurrence is detected two years
799
Figure 3. — Ultrasound of the palpable mass of the left breast.
Figure 4. — Macroscopic image of the excised lump.
and seven months after diagnosis and excision.
Discussion
Vascular tumors of the breast are divided in two basic cat-
egories: angiosarcomas (that are more common) and he-
mangiomas [3]. The incidence of hemangiomas of the breast
varies between 1.2% (discovered in mastectomies for breast
cancer) and 11% (in forensic autopsies) [4, 5]. Heman-
giomas are benign tumors that are rarely found in the breast,
but sometimes they have been found on microscopy of
biopsy material when there are other indications [6, 7]. In
the vast majority of hemangiomas, the diagnosis is based on
history and clinical examination [8]. However, in case of a
breast hemangioma, ultrasonography and biopsy are neces-
sary. Due to their potentially suspicious morphology, there
are challenging in diagnosis [9, 10]. Pathologists identify
four different types of hemangiomas: perilobular, parenchy-
mal, subcutaneous, and venous. In the present case it was a
parenchymal hemangioma composed of dilated blood ves-
sels filled with erythrocytes, and based on their size, it was
a capillary hemangioma. Most reported breast hemangiomas
are cavernous and many times contain calcifications, which
800 Breast capillary hemangioma at the tail of Spencer: a rare entity
Figure 5. — Pathologic sections of the lesion with Hematoxylin and Eosin stain.
usually represent phleboliths. Usually, capillary heman-
giomas of the breast remain clinically occult [11]. In the vast
majority they appear oval-shaped with well-circumscribed
margins, but their echostructure varies, and it may be quite
difficult for the radiologist to differentiate them from com-
plex cysts and fibroadenomas [12-14], or medullary or tu-
bular carcinomas. Many of them are characterized by
phleboliths, which appear in mammography as microcalci-
fications, making interpretation of findings more difficult
[12].
Thorough examination of case reports published, reveals
that in almost all cases the diagnosis of hemangioma is set
after surgical treatment, whereas preoperative diagnosis is
extremely rare [15]. Of particular interest is the fact that
most hemangiomas appear in mammography as lobular le-
sions and in ultrasound as hypoechoic masses that in both
cases have well-circumscribed edges [16]. Dynamic con-
trast-enhanced magnetic resonance imaging (DCE-MRI) is
a diagnostic tool with high positive and negative predictive
values that could be proven very helpful in the preoperative
diagnosis of hemangiomas [17]. Although first results ap-
pear promising, the high cost and low availability of this
exam limit its use.
Considering all types of breast hemangiomas, cavernous
is the most common one, with only a few reports in medical
literature for the remaining types. The issue of whether these
masses should be excised or followed up remains quite con-
troversial. The most important element is to distinguish
which of these entities could possibly be angiosarcoma, in
order to perform a more aggressive treatment [2, 18]. Of
particular interest is the fact that FNA or biopsies and par-
tial excision exhibit increased risk of hemorrhage or
hematomas. In the present case, the authors describe a rare
location of breast hemangioma with mammographic char-
acteristics that were suspicious for malignancy. There were
no identifiable risk factors and patient did not mention the
intake of hormonal therapy (which is associated with sudden
increase of the dimensions of such lumps).
Conclusion
To summarize, vascular tumors should always be in-
cluded in differential diagnosis of breast masses, especially
when blood flow is identified inside them. In the present
case, no phleboliths were identified, that in many occasions
help with the diagnosis. In case that biopsy is performed,
there is an increased risk of hemorrhage and should there-
fore a complete excision of the mass should be preferred in
order to exclude the possibility of malignancy.
 A. Bothou, I. Grammatikakis, N. Evangelinakis, C. Eftichiadis, G. Iatrakis, S. Zervoudis
References
[1] Hoda S.A., Cranor M.L., Rosen P.P.: “Hemangiomas of the breast
with atypical histological features: further analysis of histological
subtypes confirming their benign character”. Am J Surg. Pathol.,
1992, 16, 553.
[2] Mariscal A., Casas J.D., Balliu E., Castella E.: “Breast hemangioma
mimicking carcinoma”. Breast, 2002, 11, 357.
[3] Kim S.M., Kim H.H., Shin H.J., Gong G., Ahn S.H.: “Cavernous
haemangioma of the breast”. Br. J. Radiol., 2006, 79, e177-80.
[4] Rosen P.P., Ridolfi R.L.: “The perilobular hemangioma. A benign
microscopic vascular lesion of the breast”. Am. J. Clin. Pathol.,
1977, 68, 21.
[5] Lesueur G.C., Brown R.W., Bhathal P.S.: “Incidence of perilobular
hemangioma in the female breast”. Arch. Pathol. Lab. Med., 1983,
107, 308.
[6] Dener C., Sengul N., Tez S., Caydere M.: “Haemangiomas of the
breast”. Eur. J. Surg., 2000, 166, 977.
[7] Kawatra V., Lakshmikantha A., Dhingra K.K., Gupta P., Khurana N.:
“A rare coexistence of concurrent breast hemangioma with fi-
broadenoma: a case report”. Cases J., 2009, 15, 7005.
[8] Metry D. “Evaluation and diagnosis of infantile hemangiomas”. Up-
ToDate, 2015.
[9] Brodie C., Provenzano E.: “Vascular proliferations of the breast”.
Histopathology, 2008, 52, 30-44.
[10] Aurello P., Cicchini C., Mingazzini P.: “Hemangioma of the breast:
an unusual lesion without univocal diagnostic findings”. J. Exp. Clin.
Cancer Res., 2001, 20, 611.
[11] Schäfer F.K., Biernath-Wuepping J., Eckmann-Scholz C., Order
B.M., Mathiak M., Hilpert F., et al.: “Rare benign entities of the
breast-myoid hamartoma and capillary hemangioma”. Geburtshilfe
Frauenheilkd., 2012, 72, 412.
801
[12] Siewert B., Jacobs T., Baum J.K.: “Sonographic evaluation of sub-
cutaneous hemangioma of the breast”. AJR Am. J. Roentgenol., 2002,
178, 1025.
[13] Mesurolle B., Sygal V., Lalonde L., Lisbona A., Dufresne M.P.,
Gagnon J.H., Kao E.: “Sonographic and mammographic appearances
of breast hemangioma”. AJR Am J Roentgenol., 2008, 191, W17.
[14] Adwani A., Bees N., Arnaout A., Lanaspre E.: “Hemangioma of the
breast: clinical, mammographic, and ultrasound features”. Breast J.,
2006, 12, 271.
[15] Tilve A., Mallo R., Pérez A., Santiago P.: “Breast hemangiomas: cor-
relation between imaging and pathologic findings”. J. Clin. Ultra-
sound., 2012, 40, 512.
[16] Funamizu N., Tabei I., Sekine C., Fuke A., Yabe M., Takeyama H.,
Okamoto T.: “Breast hemangioma with difficulty in preoperative di-
agnosis: a case report”. World J. Surg. Oncol., 2014, 12, 313.
[17] Yang L.H., Ma S., Li Q.C., Xu H.T., Wang X., Wang E.H.: “A sus-
picious breast lesion detected by dynamic contrast-enhanced MRI
and pathologically confirmed as capillary hemangioma: a case re-
port and literature review”. Korean J. Radiol., 2013, 14, 869.
[18] Kinoshita S., Kyoda S., Tsuboi K., Son K., Usuba T., Nakasato Y., et
al.: “Huge cavernous hemangioma arising in a male breast”. Breast
Cancer, 2005, 12, 231.
Corresponding Author:
S. ZERVOUDIS, M.D.
Rea Hospital
383-Suggrou Avenue
Palaio Faliro 17564 (Greece)
e-mail: szervoud@otenet.gr
CEOG Clinical and Experimental
Obstetrics & Gynecology

A very rare case of ectopic intramural pregnancy after IVF-ET

B. Bechev, M. Konovalova
Department of Obstetrics and Gynecology, Dr. Shterev Hospital, Sofia (Bulgaria)

Summary
This case report shows the successful use of early medical treatment of ectopic intramural pregnancy with ultrasound-guided la-
paroscopic methotrexate (Mtx) injection. A 28-year old woman, gravida 1, para 0, with a history of laparoscopic cystectomy of an en-
dometriotic cyst of the left ovary two years ago was referred to the present clinic. The patient was treated with IVF-ET because of tubal
occlusion. Transvaginal ultrasonography showed an ectopic gestational sac (GS) with presence of yolk sac and embryo with a heart-
beat in the posterior uterine wall, completely surrounded by myometrium. The authors successfully treated her with US-guided injec-
tion of Mtx into the GS cavity by laparoscopic approach.

Key words: Intramural pregnancy; Laparoscopic approach; IVF-ET; Methotrexate.

Introduction
Intramural pregnancy is the rarest form of ectopic preg-
nancy. More than 95% of ectopic pregnancies involve the
fallopian tubes. Other ectopic sites of implantation are less
frequent. In the literature the present authors found only 18
cases of intramural pregnancies. Intramural pregnancy
refers to pregnancy implanted within the myometrium,
which is completely surrounding the gestational sac (GS)
with no communication to the endometrial cavity. These
cases frequently are complicated by uterine rupture and
hemorrhage, so early diagnosis and treatment are very im-
portant. The causes of intramural pregnancy remain un-
clear, however, uterine trauma and disease are considered
to increase the risk of the incidence. Other predisposing risk
factors include cesarean section and adenomyosis. The eti-
ology of intramural pregnancy is unclear and may result
from assisted reproductive technology and/or defective mi-
gration of an implanting pregnancy.
yolk sac were scanned in the posterior wall of the uterus. The sac
was completely surrounded by myometrium (Figure 1). Because of
the increasing size of the GS up to 12.0 mm and the increasing lev-
els of hCG up to 8,466 IU/l laparoscopic procedure was performed
on day 27 after ET. During laparoscopy a round thickening of the
posterior uterine wall with size around 15.0 mm, suspicious for ec-
topic GS was detected. Under US guidance the GS was injected
through the abdomen with 17G needle with two ml methotrexate
(Mtx) (Figure 2). Two days after the procedure, the levels of hCG
began to decrease. The patient was discharged from the unit the next
day after the operation. The hCG levels were checked weekly. On
day 50th after the operation, hCG was negative.
Discussion
Intramural ectopic pregnancy refers to a GS within the
myometrium without any connection with the uterine cav-
ity and the fallopian tubes. In the literature almost all re-

Case Report
A 28-year old woman, gravida 1, para 0, was admitted for a pre-
sumed ectopic pregnancy. She had a history of laparoscopic cys-
tectomy of an endometriotic cyst of the left ovary. During the
previous operation, partial salpingectomy of the left tube was per-
formed because of hydrosalpinx and an isthmic obstruction of the
right tube was diagnosed. An IVF-ET procedure because of the bi-
lateral tubal obstruction was performed. There was no pregnancy
detected after the fresh embryo transfer (ET). One year later frozen
ET (FrET) was performed. Total three embryos frozen on day 3
were thawed and transferred with cryo-survival rate 80%, 80%, and
100%. Endometrium was prepared with estrogen and progesterone
with 10.3-mm thickness on the day of ET. After positive pregnancy
test the patient was examined by ultrasound on day 25 after ET. En-
larged and asymmetric uterus and an ectopic GS with presence of Figure 1. — US image of intramural gestational sac.

Revised manuscript accepted for publication February 29, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3608.2017
 B. Bechev, M. Konovalova
Figure 2. — Laparoscopic injection of methotrexate into the in-
tramural gestational sac under US guidance.
ported cases were treated by laparotomy or with conser-
vative medical treatment - Mtx and potassium chloride [1].
There are only two reported cases with laparoscopic treat-
ment of the intramural pregnancy. The first case ended
with laparoscopic hysterectomy and the second treated the
condition with more conservative approach - laparoscopic
resection of the gestational sac and preserving of the uterus
[2]. The present is an extremely rare case of laparoscopic
conservative treatment of intramural pregnancy with US-
guided Mtx injected in the ectopic gestational sac.
The causes of intramural pregnancy still remain unclear
because of the rareness of the condition. Since the present
patient had a history of endometriosis and probable adeno-
myosis, the authors believe that this is the most reasonable
factor for the intramural pregnancy. Deep adenomyosis has
enough endometrial tissue and it can also respond to estro-
gen and progesterone stimulation. The microscopic sinus
tract related with adenomyosis facilitates the migration of
803
the embryo into the uterine wall.
Intramural pregnancy is a life-threatening condition with
high risk of severe hemorrhage. Its early detection allows con-
servative treatment and preserving the fertility of the patient.
References
[1] Khalifa Y., Redgment C.J., Yazdani N., Taranissi M., Craft IL.: “In-
tramural pregnancy following difficult embryo transfer”. Hum. Re-
prod., 1994, 9, 2427.
[2] Tucker S.W.: “Laparoscopic management of an intramural preg-
nancy”. J. Am. Assoc. Gynecol. Laparosc., 1995, 2, 467.
[3] Bernstein H.B., Thrall M.M., Clark W.B.: “Expectant management
of intramural ectopic pregnancy”. Obstet. Gynecol., 1999, 42, 2.
[4] Lu H.F., Sheu B.C., Shih J.C., Chang Y.L., Torng P.L., Huang S.C.:
“Intramural ectopic pregnancy. Sonographic picture and its relation
with adenomyosis”. Acta Obstet. Gynecol. Scand., 1997, 76, 886.
[5] Falfoul A., Jadoui A., Bellasfar M., Kaabar N., Hamdoun L., Ben
Zineb N., et al.: “Intramural pregnancy: a case report. J. Gynecol.
Obstet. Biol. Reprod. (Paris), 1992, 21, 641.
[6] Buster J.E., Pisarska M.D.: “Medical management of ectopic preg-
nancy”. Clin. Obstet. Gynecol., 1999, 42, 23.
[7] Lee G.S., Hur S.Y., Kown I., Shin J.C., Kim S.P., Kim S.J.: “Diag-
nosis of early intramural ectopic pregnancy”. J. Clin. Ultasound,
2005, 33, 190.
[8] Hamilton C.J., Legarth J., Jaroudi K.A.: “Intramural pregnancy after
in vitro fertilization and embryo transfer”. Fertil. Steril., 1992, 57,
215.
[9] Ginsburg K.A., Quereshi F., Thomas M., Snowman B.: “Intramural
ectopic pregnancy implanting in adenomyosis”. Fertil. Steril., 1989,
51, 354.
[10] Fait G., Goyert G., Sundareson A., Pickens A. Jr.: “Intramural preg-
nancy with fetal survival: case history and discussion of etiologic fac-
tors”. Obstet. Gynecol., 1987, 70, 472.
Corresponding Author:
B. BECHEV, M.D.
Department of Obstetrics and Gynecology
Dr. Shterev Hospital
Hr. Blagoev 25-31
1330 Sofia (Bulgaria)
e-mail: bobbybe4@abv.bg
CEOG Clinical and Experimental
Obstetrics & Gynecology

A rare cause of intractable tachycardia during caesarean section:

acute cannabis use

B. Tuncali
Department of Anaesthesiology, Baskent University Zubeyde Hanim Practice and Research Centre, Izmir (Turkey)

Summary
Persistent, unexplained perioperative tachycardia during caesarean section may be a challenge for the anaesthesiologist because the
differential diagnosis is large and it may negatively impact patient outcome. Cannabis is the most common recreational drug generally
used for its hallucinogenic properties in pregnancy. Although the cardiovascular effects of cannabis is dose-dependent, acute effects of
low doses induce euphoria, tachycardia, and anxiety. However, the majority of pregnant patients with a history of drug addiction hide
or deny it due to feelings of shame and guilt or legal concerns. The authors present a case of persistent perioperative tachycardia dur-
ing caesarean section under combined spinal epidural anaesthesia (CSEA) in a drug-addictive pregnant who received cannabis-con-
taining cigarettes six hours before her admission to the hospital.

Key words: Perioperative tachycardia; Cannabis; Caesarean section; Combined spinal epidural anaesthesia.

Introduction midazolam, 50 µg of fentanyl, and incremental doses of propofol (a

Perioperative tachycardia is a common event which may
be associated with poor outcome during anaesthesia [1]. To
determine the cause is not always simple because the etiol-
ogy may be of multiple origins [2]. Herein, the authors
present a case of persistent perioperative tachycardia in a
parturient who underwent an emergency caesarean section
under combined spinal epidural anaesthesia.
Case Report
A 26-years-old pregnant patient was admitted to the obstetric
clinic due to preterm labour at 37 weeks of gestation. She had no
medical or surgical history, except for iron and multivitamin sup-
plementation during pregnancy, and her blood pressure was
141/76 mmHg with a heart rate of 119/minute. Laboratory tests in-
cluding the whole blood analysis, serum glucose levels, thyroid
function, and coagulation tests were normal. An emergency cae-
sarean section under combined spinal epidural anaesthesia
(CSEA) was planned.
On arrival into the operating room, monitoring including non-in-
vasive blood pressure, electrocardiogram, and pulse oximetry re-
vealed a blood pressure of 136/78 mmHg, sinus rhythm with a heart
rate of 128/minute, and peripheral oxygen saturation of 97%, re-
spectively. Respiratory rate was 20/minute. After obtaining intra-
venous access, a CSEA was performed in the sitting position with
1.7 ml 0.5 % hyperbaric bupivacaine, followed by placement of an
epidural catheter. At the 14th minute after CSEA, the sensory block
level reached T4 level and the operation was commenced. Mean-
while, her blood pressure and heart rate were 113/68 mm Hg and
126/minutes, respectively. A baby girl with a weight of 3,680 grams
was delivered at the sixth minute after the skin incision with APGAR
scores of 9 and 10 at one and five minutes, respectively. She ex-
pressed no pain or discomfort at the operation site. Following the in-
jection of ten IU oxytocin, she received a total of three mg of
total dose of 110 mg) in order to provide and maintain a Ramsay se-
dation score of 3-4, because her tachycardia was attributed to anxi-
ety. The surgical procedure lasted 38 minutes with a body
temperature of 36.5-36.7°C and stable hemodynamic status, except
the sinus tachycardia (heart rate 117-132/ minute) throughout the
procedure. In the operation, a total of 1,300 ml crystalloid solution
was given and the urine output was 600 ml. Postoperative pain con-
trol was provided by epidural bupivacaine 0.125% and intravenous
analgesics afterwards. Intraoperative blood glucose level was 94 g/dl.
At postoperative one hour, her haemoglobin and haematocrit levels
were normal and the pain levels were below 3 according to visual
analogue scale (VAS), but sinus tachycardia was still present with a
heart rate of 112/minute. The authors discussed their concerns for
tachycardia with the surgeon and informed the patient that advance
tests and consultations including cardiologic assessment would be
made to determine the underlying pathology. Meanwhile, she in-
formed them that she had been using cannabis-containing cigarettes
at least two days per week during the past three years, including the
same day (six hours before her admission to the hospital) but did not
tell this to the doctors due to legal concerns. At postoperative sixth
hour, her heart rate gradually decreased to 76/minute and was dis-
charged uneventfully on postoperative day 2.
Discussion
Sinus tachycardia, commonly confronted by anaesthesiol-
ogists, may be present preoperatively or occur at any time of
the perioperative period, which usually indicates that there is
a pathology in the absence of an identifiable cause. When
promptly identified and appropriately treated, it has a little ef-
fect on perioperative morbidity. However, sometimes, it may
be a great challenge for the anaesthesiologist. Because dif-
ferential diagnosis is extensive, including emotional stress,
pain, medications, drug withdrawal, undiagnosed infection,

Revised manuscript accepted for publication April 13, 2016

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3675.2017
 B. Tuncali
underlying diseases, such as hyperthyroidism and pheochro-
mocytoma, or cardiac disorders, such as pre-excitation syn-
dromes and re-entrant pathways. If not recognized or
adequately treated, these disorders may increase the risk of
anaesthesia and negatively impact outcomes [1-3]. Rose et al.
showed that both intra- and postoperatively, tachycardia for
longer than ten minutes and hypertension for longer than five
minutes were found to be associated with an increased inci-
dence of postoperative cardiac events [4]. Frolich and Caton
found that a high baseline heart rate was strongly predictive of
marked hypotension, following spinal anaesthesia in pre-hy-
drated pregnant patients [5]. The clinical significance of tachy-
cardia depends on the simultaneous blood pressure and the
cardiac rhythm of the patient. Watterson et al. showed 145
causative factors in 123 reports of tachycardia in 4,000 inci-
dents reported to Australian Incidence Monitoring Study and
found that 27% of the tachycardia events were associated with
normotension. The most common (48%) cause in those events
were related to drugs [6]. In the present patient, sinus tachy-
cardia was accompanied with normotension and was attributed
to anxiety, because there was no history of preoperative med-
ication at home or in the ward preoperatively. Additionally, pre-
and postoperative whole blood analysis did not show profound
anaemia or leucocytosis, and intraoperative blood glucose level
and body temperature were within normal limits, excluding the
anaemia, hypoglycaemia or infection. The aetiology of tachy-
cardia in the present patient was due to acute cannabis use and
consequent increase in sympathetic activity, which is consis-
tent with the study of Watterson et al. suggesting the most com-
mon causes of tachycardia in normotensive patients were
related to drugs [6].
Cannabis is obtained from the dried flowering tops and
leaves of Cannabis Sativa. It remains the most common recre-
ational drug generally used for its hallucinogenic properties in
pregnancy [7]. The acute effects of cannabis include euphoria,
tachycardia, and anxiety. The cardiovascular effects of
cannabis is dose-dependent. At low and moderate doses,
cannabis leads to an increase in sympathetic activity resulting
in tachycardia and increased cardiac output. However, at high
doses, parasympathetic activity is increased, leading to brady-
cardia and hypotension. [7, 8]. Although changes in blood
pressure have been reported with negligible clinical signifi-
cance in a healthy parturient, additive effects of cannabis and
potent inhaled anaesthetics can result in pronounced myocar-
dial depression during general anaesthesia [8]. Additionally,
adverse psychiatric and autonomic reactions to cannabis may
interfere with safe induction of anaesthesia and postoperative
recovery [8]. In a parturient with a history of acute cannabis
abuse, drugs that increase heart rate such as ketamine, pan-
curonium, atropine, and epinephrine should be avoided. Ad-
ditionally, respiratory complications including oropharyngitis,
uvular edema, and bronchospasm during general anaesthesia
have been reported [7, 8]. In the present patient, the caesarean
section under combined epidural spinal anaesthesia was un-
eventful except for the tachycardia. On the other hand, the au-
thors could have used general anaesthesia in which the
805
additive effects with potent inhaled anaesthetics, as well as the
interactions with sedative-hypnotic drugs, might have resulted
in hemodynamic instability and/or interfere with recovery.
In a cannabis addicted-parturient, cannabinoids enter the
embryo or fetus during pregnancy. Although, there appears
to be no evidence of teratogenicity, low neonatal birth weight,
increased risk of complications during labour, delay in cog-
nitive development in infants, and ten-fold increased risk of
leukaemia have been reported [8]. In the present patient,
APGAR scores and body weight of the baby were normal.
The prevalence of drug abuse in pregnancy is on the increase
worldwide [8]. Although it has been suggested that a history
of illicit drug use should be obtained routinely in the preoper-
ative assessment, the majority of pregnant patients with a his-
tory of drug addiction hide or deny it due to feelings of shame
and guilt or legal concerns [9-11]. Physicians should always be
aware that patients may not tell the truth due to various causes
including legal concerns. Anaesthesiologists should also be fa-
miliar with the effects of illicit drugs on body systems to pro-
vide a safe anaesthesia for this group of patients.
References
[1] Reich D.L., Bennett-Guerrero E., Bodian C.A., Hossain S., Winfree
W., Krol M.: “Intraoperative tachycardia and hypertension are inde-
pendently associated with adverse outcome in noncardiac surgery of
long duration”. Anesth. Analg., 2002, 95, 273.
[2] Webb R.K., Currie M., Morgan C.A., Williamson J.A., Mackay P., Rus-
sell W.J., et al.: “The Australian Incident Monitoring Study: an analy-
sis of 2000 incident reports”. Anaesth. Intensive. Care, 1993, 21, 520.
[3] Cohen S.P., Kent C.: “Pronounced unexplained preoperative tachy-
cardia heralding serious cardiac events: a series of three cases”. Can.
J. Anaesth., 2005, 52, 858.
[4] Rose D.K., Cohen M.M., DeBoer D.P.: “Cardiovascular events in the
postanesthesia care unit: contribution of risk factors”. Anesthesiology.,
1996, 84, 772.
[5] Frolich M.A., Caton D.: “Baseline heart rate may predict hypotension
after spinal anesthesia in prehydrated obstetrical patients”. Can. J.
Anesth., 2002, 49, 185.
[6] Watterson L.M., Morris R.W., Williamson J.A., Westhorpe R.N.: “Cri-
sis management during anesthesia; Tachycardia”. Qual. Saf. Health.
Care, 2005, 14, e10.
[7] Kuczkowski K.M.: “Anesthetic implications of drug abuse in preg-
nancy”. J. Clin. Anesth., 2003, 15: 382.
[8] Ashton C.H.: “Adverse effects of cannabis and cannabinoids”. Br. J.
Anaesth., 1999, 83, 637.
[9] Wong G.T.C., Irwin M.G.: “Poisoning with illicit substances: toxicol-
ogy for the anaesthetist”. Anaesthesia, 2013, 68, 117.
[10] Mills P.M., Penfold N.: “Cannabis abuse and anesthesia”. Anaesthe-
sia, 2003, 58, 1125.
[11] Van Gelder M.M., Reefhuis J., Caton A.R., Werler M.M, Druschel
C.M., Roeleveld N.: “National Birth Defects Prevention Study. Char-
acteristics of pregnant illicit drug users and associations between
cannabis use and perinatal outcome in a population-based study”.
Drug. Alcohol. Depend., 2010, 109, 243.
Corresponding Author:
B. TUNCALI, M.D.
Department of Anaesthesiology and Reanimation
Baskent University Zubeyde Hanim Practice
and Research Centre, Caher Dudayev Bulvari
Karsiyaka, Izmir 35540 (Turkey)
e-mail: tuncali.bahattin@gmail.com
CEOG Clinical and Experimental
Obstetrics & Gynecology

Confined placental mosaicism of trisomy 16 detected by

non-invasive prenatal testing and multiple abnormalities

Ting Wang#, Qin Zhang#, Haibo Li, Wei Wang

Center for Reproduction and Genetics, Nanjing Medical University Affiliated Suzhou Hospital, Suzhou (China)

Summary
This study aimed to investigate a case of confined placental trisomy 16 mosaicism (CPM16) with abnormal amniotic fluid, placen-
tal lake, and other abnormalities. Maternal serum screening was performed to assess the risk of foetal aneuploidy, and massively par-
allel sequencing was used to detect cell-free fetal DNA. The abnormalities of the fetus, amniotic fluid, and placenta were detected by
ultrasonic inspection. Maternal serum screening indicated a high risk for trisomy 21, and the non-invasive prenatal testing (NIPT) re-
sult was positive for trisomy 16. The pregnancy was terminated and karyotype analysis of fetal heart blood revealed a 46, XX karyotype.
Copy number variation (CNV) sequencing of placental tissues indicated that CPM16 is the main cause of false-positive NIPT results
and intrauterine growth retardation (IUGR) diagnoses. Combining molecular genetics technologies, such as CNV sequencing, can be
complementary, and provide an effective strategy to determine the cause of such abnormalities.

Keywords: Confined placental mosaicism; Non-invasive prenatal testing; Multiple abnormalities.

HiSeq2500 platform. The obtained sequencing reads and the

Introduction
In the past quarter-century, the main approach to fetal
aneuploidy screening has been invasive prenatal diagnosis,
such as amniocentesis or chorionic villus sampling (CVS).
Lo et al. discovered cell-free fetal DNA in the plasma of
pregnant women in 1997 [1], which resulted in the realisa-
tion of non-invasive prenatal testing (NIPT) for screening
of foetal aneuploidy [2]. However, in addition to fetal ane-
uploidy, foetal, placental, maternal or other abnormalities
have been frequently detected using both prenatal invasive
and non-invasive diagnosis [3, 4]. Some abnormalities are
technically difficult to detect by prenatal diagnosis or re-
main undetected [5], whereas other abnormalities are indi-
cators of certain diseases [6]. The current study examined
a case of confined placental trisomy 16 mosaicism with in-
trauterine growth retardation (IUGR), apparent thickening
of the placenta, placental lake and oligoamnios. This study
also investigated the biological basis for these abnormali-
ties and the effects of prenatal diagnosis.
Case Report
This study was approved by the Institutional Ethics Committee
of Suzhou Hospital affiliated to Nanjing Medical University. Writ-
ten informed consent was obtained from all participants of this
study.
NIPT was performed following standard techniques [7]. Ma-
ternal peripheral blood was collected, and DNA libraries were
constructed and subjected to massively parallel sequencing on the
human genomic sequence hg20 were aligned. Uniquely mapped
reads for Chr13, Chr16, Chr18, and Chr21 were subsequently
counted and normalised for the GC content. Data from the test
samples were compared with those from the reference sample,
and Z-scores were calculated with −3.0 < Z < 3.0 as the normal
value.
Ultrasound screening was performed repeatedly between 19
and 22 weeks (W) of gestation using the 730 Expert system with
a 2–5 MHz transabdominal convex transducer and a three-di-
mensional broadband curved array transducer (3D6-2, 2–6 MHz)
in accordance with the routine fetal ultrasound scan guideline [8].
To estimate fetal biometry and well-being, the following sono-
graphic parameters were used: biparietal diameter, head circum-
ference, abdominal circumference, femur diaphysis length, and
humerus length. Conventional ultrasound scanning was also per-
formed to evaluate the development of the fetal head, face, spine,
chest and abdomen, internal organs., and limbs. The placenta, um-
bilical cord, and amniotic fluid were also examined.
Copy number variation (CNV) sequencing was performed as
described previously [9]. Three duplicate biopsy samples from
various regions of the placental tissue were collected. A repre-
sentation of Chr16 in each sample was calculated using Chr16 se-
quence reads/total sequence reads. Levels of T16 mosaicism were
subsequently determined using the following formula: CR Chr16
in the test sample/mean CR of Chr16 in the reference sample ×
100%.
Haemothorax blood amounting to 3 ml was collected from the
aborted fetus with heparinisation and then cultured in the medium
containing phytohaemagglutinin. The G-banded karyotype was
analysed in accordance with the International System for Human
Cytogenetic Nomenclature (ISCN2013). The AI karyotyping
image analysis system was used to count 60 metaphases, and 20
karyotypes were microscopically analysed in triplicate.

These authors contributed equally to this work.

#

Revised manuscript accepted for publication January 26, 2017

Clin. Exp. Obstet. Gynecol. - ISSN: 0390-6663 7847050 Canada Inc.
XLIV, n. 5, 2017 www.irog.net
doi: 10.12891/ceog3841.2017
 Ting Wang, Qin Zhang, Haibo Li, Wei Wang 807

Table 1. — Record of pregnancy follow-up after abnormal maternal serum screening.

Sample Gestational age Prenatal/postnatal test Result/diagnosis
Maternal peripheral blood 16 W Maternal serum screen 1/190 risk of T21
cffDNA 18 W NIPT T13 negative (Z = 0.18), T18 negative (Z = -0.87)
T21 negative (Z = -0.73), T16 positive
Established fetus 19 W Ultrasound/sonogram BDP=35 mm, FL=20 mm, nasal bone=3.0 mm
Amniocytes 21 W aCGH 46, XX
Amniocytes/Maternal peripheral blood 21W STR No fetal DNA
Established fetus 21 W Ultrasound/sonogram IUGR, Severe oligoamnios
Established fetus 22 W Ultrasound/sonogram IUGR, Apparent thickening of the placenta
Severe oligoamnios, Placental lake
23 W Ultrasound/sonogram Terminate the pregnancy
Fetal heart blood TOP Karyotyping 46, XX
Multiple placental tissue TOP CNV sequencing Confined placental trisomy 16 mosaicism
(details in Table1 and Fig. 2)
W = weeks and TOP = termination of pregnancy. NIPT and calculation of Z-scores were performed with Z-scores >3 or ≤ 3 defined as abnormal.

Table 2. — The results of CNV sequencing.

Region Placental tissue CNV sequencing result
Region A Centre of placental-maternal side 47,XX,+16 [65%]/46,XX, [35%]
Region B Middle of placental-maternal side 47,XX,+16 [60%]/46,XX, [40%]
Region C Edge of placental-maternal side 46, XX
Region D Centre of placental-fetal side 47,XX,+16 [63%]/46,XX, [37%]
Region E Middle of placental-fetal side 47,XX,+16 [61%]/46,XX, [39%]
Region F Edge of placental-fetal side 47,XX,+16 [64%]/46,XX, [36%]
Region G skin of fetus 46, XX region H umbilical cord 46, XX

A 29-year-old G1P0 mother with no family history of congen-
ital anomalies, early infant deaths or consanguinity was referred
for routine prenatal screening (Table 1). At 16W of gestation, ma-
ternal serological screening indicated a relatively high risk for
T21. The mother subsequently received genetic counselling and
selected NIPT on cell-free fetal DNA as a second screening test at
18W, which yielded a positive result for T16. Routine prenatal ab-
dominal ultrasound assessment in the present centre at 19W re-
vealed that one side of the fetal nasal bone was incomplete.
Amniocentesis at 21W was conducted but failed because of low
levels of amniotic fluid. Ultrasound assessment was repeated
twice at 21W and 22W. The results showed signs of IUGR, in-
cluding apparent thickening of the placenta and severe oligohy-
dramnios with an amniotic fluid index of 4.8 cm (amniotic fluid
index of < 5 cm is considered as oligohydramnios). Meanwhile,
the placenta covered the intracervical mouth, accompanied with
cystic organisations of different sizes and a thin placenta. The tis-
sue was less than normal and showed a “boiling state” (Figure 1);
hence, the patient was diagnosed with placental lake. The pregnant
woman decided to terminate the pregnancy at 23W.
A haemothorax blood sample collected from the aborted fetus
was analysed. The result showed a 46, XX karyotype, which was
inconsistent with the result of NIPT. A previous study demon-
strated that most cases of T16 are aborted spontaneously between
8W and 15W of gestation [10]. Given the discrepancy between
the results of serological screening, NIPT, ultrasound screening,
and karyotyping of fetal heart blood, the authors suspected a preg-
nancy with placental trisomy 16 mosaicism (CPM16). To inves-
tigate this case further, they analysed the aborted placenta by CNV
sequencing of placental tissues (Tables 1 and 2). Six samples of
placental tissue (three from the maternal side and three from the
fetal side) combined with samples of the umbilical cord tissue and
skin tissue were obtained to determine the relative levels and dis-
tribution of T16 mosaicism in the placental tissue (Tables 1 and 2).
Three samples from the centre, middle. and edge of the fetal side
of the placenta indicated average levels of 63%, 61%, and 64% for
T16, respectively. However, only the centre and middle of the ma-
ternal side of the placenta showed average levels of 65% and 60%
for T16, respectively; however, the edge of the maternal side of
the placenta was normal for CNV sequencing analysis (Table 1).
Discussion
Numerous published data have suggested that NIPT ex-
hibits high accuracy in detecting fetal trisomy 13, 18, and
21, with both sensitivity and specificity of > 99% [4]. How-
ever, several studies in recent years revealed discordant re-
sults between fetal karyotyping and NIPT; such results
were attributed to false-positive results caused by confined
placental mosaicism (CPM) [6, 7]. Thus, additional strate-
gies should be proposed to confirm the results of NIPT. In
the present study, a case of CPM16 was confirmed by CNV
sequencing and karyotyping after NIPT. According to the
specific placental cell lineages exhibiting the abnormal cell
line, CPM could be categorised into three types. Placental
mosaicism can only be found in trophoblast (type I) and
chorionic stroma (type II); however, the condition can be
confined to both cell lineages (type III) [11]. Type III is
808 Confined placental mosaicism of trisomy 16 detected by non-invasive prenatal testing and multiple abnormalities
Figure. 1. — The image of ultrasound screening at 19W: appar-
ent thickening of the placenta and severe oligoamnios are ob-
served.
mostly meiotic in origin, and most reported cases with
IUGR and intrauterine fetal death are associated with
CPM16 [12]. On the basis of previous studies and the afore-
mentioned data, CPM16 in the current study was cate-
gorised under type III.
The majority of IUGR cases are associated with placen-
tal insufficiency [13]. The placenta in the current study was
thickened, and an obvious symptom of placental lake was
observed (Table 1 and Figure 1). Therefore, the authors
postulated that placental lake and other abnormalities may
be the causes of IUGR in this study. Some studies demon-
strated no apparent correlation between IUGR and CPM
[14]. Regardless, T16 is the most common trisomy found in
CPM associated with IUGR, and chromosome 16 was first
identified as one of the candidate chromosomes related to
IUGR [15]. Thus, IUGR is a potential indicator of CMP16
combined with positive NIPT results for T16.
Invasive diagnostic methods, such as amniocentesis or
CVS, have long been regarded as the gold standard for ane-
uploidy confirmation [7]. However, placental abnormality
and severe oligohydramnios may lead to failed amniocen-
tesis, and some reported cases associated with severe oligo-
hydramnios render amniocentesis technically difficult or
failed [5]. Stagnation of fetal development and abnormal
metabolism for the fetus may result in a small quantity of
fetal cells in the amniotic fluid. Hence, amniocentesis re-
mains limited with respect to aneuploidy detection, and
several strategies should be combined to confirm the re-
sults.
Although challenges still exist for NIPT because of the
false-positive results caused by CPM or occult maternal
malignancies, NIPT for aneuploidy confirmation remains
an effective strategy for prenatal screening. Meanwhile, the
discordance between the fetal karyotype and NIPT caused
by occult maternal malignancies revealed that presympto-
matic detection of tumours in pregnant women undergoing
routine NIPT may be realised [6], and further research
should be proposed in the future. Hence, NIPT and other
non-invasive detection strategies have more applications in
disease detection and diagnosis.
In conclusion, the authors investigated a case of confined
placental trisomy 16 mosaicism (CPM16) with abnormal
amniotic fluid, placental lake, and other abnormalities.
CNV sequencing analysis of the biopsy of the umbilical
cord tissue and skin tissue could be complementary and
provide an effective strategy to determine the cause of such
abnormalities.
Acknowledgements
The authors thank the families for participating in this re-
search project. This work is supported by Clinical Medi-
cine Science and Technology projects of Jiangsu province
(BL2013019), Jiangsu Provincial Health Department Sci-
entific Research Project (Q201412), and Suzhou Science
and Technology Support Program (SS201429).
References
[1] Lo Y.D., Corbetta N., Chamberlain P.F., Rai V., Sargent I.L., Red-
man C.W., Wainscoat J.S.: “Presence of fetal DNA in maternal
plasma and serum”. Lancet, 1997, 350, 485.
[2] Lau T., Cheung S., Lo P., Pursley A., Chan M., Jiang F., et al.:
“Non invasive prenatal testing for fetal chromosomal abnormali-
ties by low-coverage whole-genome sequencing of maternal plasma
DNA: review of 1982 consecutive cases in a single center”. Ultra-
sound. Obstet. Gynecol., 2014, 43, 254.
[3] Verp M.S., Bombard A.T., Simpson J..L, Elias S., Reynolds J.F., Hall
J.G.: “Parental decision following prenatal diagnosis of fetal chro-
mosome abnormality”. Am. J. Med. Genet., 1988, 29, 613.
[4] Nagamatsu T., Kamei Y., Yamashita T., Fujii T., Kozuma S.: “Pla-
cental abnormalities detected by ultrasonography in a case of con-
fined placental mosaicism for trisomy 2 with severe fetal growth
restriction”. J. Obstet. Gynaecol. Re., 2014, 40, 279.
[5] Fleischer J., Shenoy A., Goetzinger K., Cottrell C.E., Baldridge D.,
White F.V., Marwan Shinawi M.: “Digynic triploidy: utility and
challenges of noninvasive prenatal testing”. Clin. Case Rep., 2015,
3, 406.
[6] Amant F., Verheecke M., Wlodarska I., Dehaspe L., Brady P., Bri-
son N., et al.: “Presymptomatic Identification of Cancers in Preg-
nant Women During Noninvasive Prenatal Testing”. JAMA Oncol.,
2015, 1, 814.
[7] Mao J., Wang T., Wang B.J., Liu Y.H., Li H., Zhang J., et al.: “Con-
fined placental origin of the circulating cell free fetal DNA revealed
by a discordant non-invasive prenatal test result in a trisomy 18 preg-
nancy”. Clin. Chim. Acta., 2014, 433, 190.
[8] Salomon L., Alfirevic Z., Berghella V., Bilardo C., Hernandez-An-
drade E., Johnsen S., et al.: “Practice guidelines for performance of
the routine mid-trimester fetal ultrasound scan”. Ultrasound. Obstet.
Gynecol., 2011, 37, 116.
[9] Wang Y., Chen Y., Tian F., Zhang J., Song Z., Wu Y., et al.: “Ma-
ternal mosaicism is a significant contributor to discordant sex chro-
mosomal aneuploidies associated with noninvasive prenatal testing”.
Clin. Chem. 2014, 60, 251.
[10] Benn P.: “Trisomy 16 and trisomy 16 mosaicism: a review”. Am. J.
Med. Genet., 1998, 79, 121.
[11] Lestou V.S., Kalousek D.K.: “Confined placental mosaicism and in-
 Ting Wang, Qin Zhang, Haibo Li, Wei Wang 809

trauterine fetal growth”. Arch. Dis. Child Fetal, 1998, 79, F223.
[12] Wolstenholme J., Rooney D., Davison E.: “Confined placental mo-
saicism, IUGR, and adverse pregnancy outcome: a controlled retro- Corresponding Author:
spective UK collaborative survey”. Prenatal. Diag., 1994, 14, 345.
HAIBO LI, M.D.
[13] Kawakita T., Sasaki H., Hirabuki S., Asamoto A.: “Fetal growth re-
striction and reversed middle cerebral artery end-diastolic flow with
Center for Reproduction and Genetics
subchorionic placental lake”. J. Obstet. Gynaecol. Re., 2013, 39, 578. Nanjing Medical University Affiliated
[14] Kontomanolis E., Lambropoulou M., Georgiadis A., Gra- Suzhou Hospital
matikopoulou I., Deftereou T., Galazios G.: “The challenging tri- 26 Daoqian Street
somy 16: a case report”. Clin. Exp. Obstet. Gynecol.,2011, 39, Suzhou 215002 (China)
412-413. e-mail: lihaiboszh@163.com
[15] Monk D., Moore G.E.: “Intrauterine growth restriction - genetic
causes and consequences”. Semin. Fetal Neonat. M., 2004, 9, 371.
 EUROPEAN ACADEMY
OF GYNAECOLOGICAL CANCER, EAGC
EAGC

Chairman: Péter B´ósze (Hungary)

Executive Board: PIERO SISMONDI (Italy)

PIERLUIGI BENEDETTI PANICI (Italy) CLAES TROPÉ (Norway)
CARLOS F. DE OLIVEIRA (Portugal) LÁSZLÓ UNGÁR (Hungary)
ANDRÉ VAN ASSCHE (Belgium)
GIUSEPPE DE PALO (Italy)
RAIMUND WINTER (Austria)
SANTIAGO DEXEUS (Spain)
WILLIAM DUNLOP (UK)
STELIOS FOTIOU (Greece) International Advisory Board
Chairman: Antonio Onnis (Italy)
GERALD GITSCH (Austria)
A. PETER M. HEINTZ (Netherlands) HUGH ALLEN (Canada)
MICHAEL HOECKEL (Germany) CURT W. BURGER (Netherlands)
JAN JACOBS (UK) ALBERTO COSTA (Italy)
ANDRÉ GORINS (France)
JACQUES LANSAC (France)
NEVILLE F. HACKER (Australia)
TIZIANO MAGGINO (Italy)
MARIA MARCHETTI (Italy)
HARALD MEDEN (Germany)
STELIOS P. MICHALAS (Greece)
JOSEPH MONSONEGO (France)
MARIA TERESA OSORIO (Portugal)
LASZLÓ PÁLFALVI (Hungary) ULF ULMSTEN (Sweden)
SERGIO PECORELLI (Italy) JAN B. VERMORKEN (Belgium)
DENIS QUELLEU (France) GEORGE D. WILBANKS (USA)
STELIO RAKAR (Slovenia) JAN ZIELINSKI (Poland)

All questions concerning the Accademy may be sent to:

PETER BOSZE, M.D. - P.O. Box 46 - Budapest 1301 (Hungary)
Phone: +36 1 4290317 - Fax: +36 1 2752172 - E-mail: eagc@cme.hu

www.cme.hu

Administrative Office:
1301 Budapest, P.O. Box 46 - Hungary
Fax (36 1) 4290318 - E-mail: eagc@cme.hu
 ✂
ISSN: 0390-6663
CLINICAL AND EXPERIMENTAL
OBSTETRICS & GYNECOLOGY
Subscription Order Card 2017
an International Journal Founded in 1974 (ISSN 0390-6663) - Vol. XLIV. Issued bimonthly. All subscriptions are entered on a
Edited by:
A. ONNIS
calendar-year basis. Individual rate is not applicable if payment is made through an Institution.
Subscriptions are entered with prepayment only and are accepted per calendar year only but can be
backdated depending on availability. If not cancelled by the end of October, they will be tacitly consi-
dered as renewed; cancellations will not be refunded.
Discounts: 10% to book sellers and subscription agencies.

Please enter my subscription at the rate I have checked:

PAPER ISSUE ONLINE ISSUE
Institutional: 700 USD Insitutional: 450 USD
Individual: 500 USD Individual: 270 USD
I.R.O.G. CANADA, Inc. - 4900 Côte St-Luc - Apt # 212 Single copy: 150 USD Single issue: 100 USD
Montréal, Qué. H3W 2H3 (Canada)
Tel. +514-4893242 - Fax +514-4854513 - E-mail: canlux@qc.aira.com - www.irog.net
Single article: 30 USD
Payment: (USD ONLY)
Founding Editor for PDF file: online through PayPal (all credit cards)
A. Onnis for hard copy
Montréal (CND) Credit Card: Mastercard Visa Diners
Beneficiary: 7847050 Canada Inc. - 4900 Côte St-Luc, #212 - Montréal, Québec,
Editors-in-Chief Bank transfer:
H3W 2H3 Canada - Account number 00001 003402-402245 SWIFT ROYCCAT2
M. Marchetti J.H. Check
Montréal (CND) Camden, NJ (USA) N° Exp. Date

Assistant Editor Signature Date

A. Sinopoli
Toronto (CND) An invoice is issued only after payment is processed; no proforma receipts will be issued. The sub-
scription order form is available through the Montréal office (Fax +1-514-485-4513) or Padua office
(Fax +39-049-8752018) or through our website www.irog.net

✂
EUROPEAN JOURNAL
OF GYNAECOLOGICAL ONCOLOGY
Subscription Order Card 2017
Founded in 1980 (ISSN 0392-2936) - Vol. XXXVIII. Issued bimonthly. All subscriptions are entered
on a calendar-year basis. Individual rate is not applicable if payment is made through an Institution.
Subscriptions are entered with prepayment only and are accepted per calendar year only but can be
backdated depending on availability. If not cancelled by the end of October, they will be tacitly con-
sidered as renewed; cancellations will not be refunded.
Discounts: 10% to book sellers and subscription agencies.

Please enter my subscription at the rate I have checked:

PAPER ISSUE ONLINE ISSUE
Institutional: 700 USD Insitutional: 450 USD
Individual: 500 USD Individual: 270 USD
Single copy: 150 USD Single issue: 100 USD

Founding Editor Single article: 30 USD

A. Onnis Payment: (USD ONLY)
Montréal (Canada) for PDF file: online through PayPal (all credit cards)
for hard copy
Editors-in-Chief Credit Card: Mastercard Visa Diners
M. Marchetti P. Bó´sze
Montréal (Canada) Budapest (Hungary) Bank transfer: Beneficiary: 7847050 Canada Inc. - 4900 Côte St-Luc, #212 - Montréal, Québec,
H3W 2H3 Canada - Account number 00001 003402-402245 SWIFT ROYCCAT2
Associate Editor N° Exp. Date
T. Maggino
Padua (Italy) Signature Date

Assistant Editor An invoice is issued only after payment is processed; no proforma receipts will be issued. The sub-
A. Sinopoli scription order form is available through the Montréal office (Fax +1-514-485-4513) or Padua office
Toronto (CND) (Fax +39-049-8752018) or through our website www.irog.net