http://informahealthcare.

com/jmf
ISSN: 1476-7058 (print), 1476-4954 (electronic)

J Matern Fetal Neonatal Med, Early Online: 1–6

! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.928858

ORIGINAL ARTICLE

Early indicators of cervical insufficiency assessed using magnetic

resonance imaging of the cervix during pregnancy
Viviane Vieira Francisco Habib1, Edward Araujo Júnior2, Sue Yasaki Sun1, Dirceu Faggion Júnior2, Rosiane Mattar2,
Jacob Szejnfeld1, and Sergio Aron Ajzen1
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by McMaster University on 10/17/14

1
Department of Imaging Diagnostics and 2Department of Obstetrics, Paulista School of Medicine, Federal University of São Paulo (EPM-UNIFESP),
São Paulo, SP, Brazil

Abstract Keywords
Objective: To establish the main characteristics of the cervix in pregnant women with cervical Cervical insufficiency, magnetic resonance
insufficiency, by means of magnetic resonance imaging (MRI). imaging, pregnancy
Methods: A prospective observational case-control study was conducted among 59 pregnant
women with cervical insufficiency and 10 normal pregnant women, between their 10th and History
28th weeks. The parameters analyzed in the MRI examinations were: precise identification of
the cervix; presence of hyposignal at the internal orifice of the cervix; loss of definition of the Received 13 April 2014
periendocervical stromal zone (PESZ); presence of hyposignal content inside the amniotic sac Revised 24 May 2014
(sludge sign) and anatomical and functional biometry of the cervix. Accepted 25 May 2014
Published online 27 June 2014
For personal use only.

Results: Peripheral hyposignal was found in 41 (85.4%) and loss of definition of the PESZ was
observed in 36 pregnant women (73.5%) with cervical insufficiency. Sludge was observed in 46
pregnant women with cervical insufficiency, and this was seen on MRI in 27 cases (58.7%). The
mean anatomical and functional lengths of the cervix on MRI in the pregnant women with
cervical insufficiency were 3.5 ± 0.8 cm (0.8–4.9 cm) and 28.7 ± 6.3 mm (9–41 mm). None of the
normal pregnant women presented hyposignal loss of the PESZ and the sludge sign.
Conclusion: MRI may be useful for evaluating the cervix and for early identification of signs of
cervical insufficiency during pregnancy.

Introduction elasticity of the cervix are necessary, in order to diminish the

high rates of prematurity. These rates have not diminished
Prematurity is the main cause of neonatal death worldwide,
over the last 100 years [1–3]. Ideally, the assessment method
affecting 13 million newborns every year. Premature infants
should be rapid, simple, dynamic and low-risk for the
who survive are at risk for serious complications including
pregnancy.
cerebral palsy, respiratory morbidity, mental retardation,
Ultrasonography (US) is a well-recognized method for
cardiovascular diseases and cancer [1]. These abnormalities
evaluating the cervix during pregnancy. Well-established
have been studied for decades, but prematurity still presents
criteria for such evaluations now exist, and measurement of
high prevalence in the USA and worldwide.
the cervix is the variable most used and has the lowest inter-
The most notable studies have affirmed that there are
observer variability. In this context, studies have recom-
multiple risk factors for prematurity and that these are related
mended that the cervical gland area (CGA) should be
to lifestyle, smoking, infection, malnutrition, genetic factors,
evaluated, although such investigations still need to be
multiparity, late abortion, repeated abortion, rapid delivery,
expanded [4].
short cervix, operative post-manipulation and uterine malfor-
Magnetic resonance imaging (MRI) has been used since
mation [1]. This multiplicity of factors seems to culminate in
the 1990s for evaluating fetal malformations. Given excellent
a common denominator consisting of dilation, shortening and
tissue contrast resolution, no artifacts related to fetal bone
disappearance of the cervix and preterm delivery labor [2].
shadowing, and better image quality in cases of maternal
New technologies for elucidating these early alterations,
obesity and oligohydramnios. MRI has gained acceptance as
the physiopathology of the process and the microstructure and
an accurate and safe method to access pregnancy complica-
tions such as placenta accreta, acute abdominal pain and
Address for correspondence: Prof. Edward Araujo Júnior, PhD, pelvic neoplasms [5].
Department of Obstetrics, Paulista School of Medicine, Federal
In a small number of studies, MRI of the cervix has shown
University of São Paulo (EPM-UNIFESP), Rua Carlos Weber, 956,
apto. 113 Visage, Vila Leopoldina, São Paulo – SP, Brazil CEP 05303- a specific morphological features, which may represent access
000. Tel/Fax: +55-11-37965944. E-mail: araujojred@terra.com.br to the microstructure of the cervix, through signal intensity
 2 V. V. F. Habib et al. J Matern Fetal Neonatal Med, Early Online: 1–6

(hydration), spectroscopy (metabolites) and precise evalu- Table 1. List of technical parameters and MRI sequences.
ation of the biometry and individualization of the stromal
Sequences
zone [4,5].
The diagnosis of cervical insufficiency is difficult to Technical parameters TSE HASTE TURBO FISP
establish, and the main criteria are defined during the pre- RT (ms) 4.3 1000 1000
gestational phase. Examinations such as a hysterosalpingo- ET (ms) 2.2 121 121
graphy and proof candle 8 present low negative predictive FOV (mm) 150 250 250
Matrix 360 360 360
value [6]. Patients with two or more abortions can be traced Flip angle ( ) 70 70 70
and monitored weekly with ultrasound in an attempt to Thickness (mm) 3 3 3
identify the shortening of the cervix. Therefore, the objective Gap 4 4 4
of this study was to establish an examination protocol using
RT, repetition time; ET, echo time; FOV, field of view (parameter that
high-resolution MRI of the cervix, which would be fast and allows the image resolution to be increased in the study area); Flip
objective, with the capacity to identify early signs of cervical angle, acquisition angle of the sequence; Gap, interval between slices.
insufficiency among pregnant women.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by McMaster University on 10/17/14

Methods
A prospective observational case–control study was con-
ducted in the Department of Imaging Diagnostics of Paulista
School of Medicine, Federal University of São Paulo (EPM-
UNIFESP), between November 2009 and November 2012.
Fifty-nine women between their 10th and 28th weeks of
pregnancy were studied: 49 with cervical insufficiency and 10
with normal pregnancies (controls). The project was approved
by UNIFESP’s Research Ethics Committee under the number
1120/11, and the pregnant women who agreed to voluntarily
participate signed a consent statement.
For personal use only.

All the pregnant women were sent for MRI examination

with a protocol directed towards evaluating the cervix. The
indication for examinations was based on their gestational
history, with emphasis on two previous late-stage abortions,
as well as on the diagnosis of cervical insufficiency that was
made during the pre-gestational period. Ten normal pregnant
women, without any clinical or obstetric disease or any risk
factor for cervical insufficiency, were also selected to form
the control group. The control group was selected by patients
with no history of miscarriages previously, they did the MRI
for other reasons and we added the cervix protocol. The
control group made the MRI exams by abdominal pain (4) and
suspected ovaries cysts (6).
The inclusion criteria for the study were that the women
should have singleton pregnancies with a diagnosis of cervical
insufficiency made by physicians in the Department of
Obstetrics, EPM-UNIFESP, based on the patients’ obstetric
histories and on digital cervical examination, and would
undergo cerclage. The exclusion criterion was claustrophobia.
MRI examinations of the cervix were performed using the
1.5 Tesla Sonata Maestro Class high-field magnet (Siemens,
Erlangen, Germany). The images were acquired using a
12-channel body coil. The sequences used were turbo spin-
echo (TSE), gradient echo (GE) and half-Fourier single-shot
(HASTE) and were acquired in the three planes (coronal,
sagittal and axial). Table 1 shows the MRI sequences used to
image the pregnant cervix.
The MRI evaluation was performed by identifying the
cervix in the most central slice, in which it was possible to
characterize the endocervical canal along its entire length, on
fast HASTE sequences. In acquiring the sequences, care was
taken to correctly identify the endocervical canal on all the
central images and, if it did not appear because of the angle
Figure 1. Anatomical (dark blue arrow) and functional (light blue arrow)
biometric measurements on the cervix from magnetic resonance
imaging, using T2-weighted TSE sequences.
used, an erroneous location or presence of artifacts, a new
sequence was acquired, oriented parallel to the endocervical
canal.
The images were analyzed by two radiologists (V. V. F. and
J. S.) with 10 and 35 years of experience of abdominal MRI,
respectively. The evaluations were done independently, and a
consensus was reached in cases of discordance. After
acquisition, the images were sent to a workstation, where
the following parameters were evaluated: length of the cervix
(anatomical biometry); cervical sign over its entire length;
evaluation of the periendocervical stromal zone (PESZ); the
periendocervical hyposignal which is best indentified at the
turbo FISP sequences and it demonstrates motility, loss of
collagen hydration of these tissues and biometry of the
endocervical canal surrounded by the well-defined stromal
zone (functional biometry). The length of the anatomical
cervix was measured by tracing out a continuous straight line
between the internal and external orifices in the sagittal plane,
while the length of the functional cervix was measured by
tracing out a continuous line in the well-defined PESZ,
starting in the region of the external orifice (Figure 1). In
addition, presence of hyposignal content inside the amniotic
 DOI: 10.3109/14767058.2014.928858 Cervical insufficiency on magnetic resonance imaging 3

sac at the internal orifice of the cervix (sludge signal) was
evaluated.
The women were followed up throughout their pregnancies
and, in some cases, the MRI was repeated. Also in some case,
their condition evolved with worsening of the degree of
cervical insufficiency.
The data were input to a spreadsheet in the Excel 2010
software (Microsoft Corp., Redmond, WA) and were analyzed
using the Statistical Package for the Social Sciences (SPSS)
software, version 19.0 for Windows (SPSS Inc., Chicago, IL).
All the variables were first analyzed descriptively. For the
variables of quantitative (numerical) nature, some summary
measurements were calculated, such as the mean, median,
minimum and maximum values and standard deviation,
and single-dimensional dispersion graphs were constructed.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by McMaster University on 10/17/14
insufficiency, according to the mean values for maternal age
and functional biometry. Fisher’s exact test was used to
compare the control group with the patients with cervical
insufficiency, according to the presence of the turbo FISP
sign, loss of the PESZ and presence of the sludge sign. The
Kolmogorov–Smirnov test was used to verify normal distri-
bution with regard to gestational age, maternal age, anatom-
ical biometry and functional biometry in the control group
and the patients with cervical insufficiency. In all the
analyses, the significance level used was p50.05.
Results
Table 2 presents the summary measurements for characteriz-
ing the sample of pregnant women with cervical insufficiency
and the normal pregnant women (controls). Periendocervical

The variables of qualitative (categorized) nature were

analyzed by calculating the absolute and relative (%)
frequencies and constructing bar graphs.
Inferential analyses were performed with the aim of
confirming or refuting evidence that had been found in the
descriptive analysis. For this, the Mann–Whitney test was
used to compare the control group with the patients with
cervical insufficiency, according to gestational age and
biometry. Student’s t test for independent samples was used
to compare the control group with the patients with cervical
hyposignal was found in 41 pregnant women (85.4%) and loss
of definition of the PESZ was observed in 36 (73.5%) with
cervical insufficiency (Figure 2). The sludge sign was
observed in 46 pregnant women with cervical insufficiency,
and this was present on MRI in 27 (58.7%; Figure 3). The
mean anatomical and functional lengths of the cervix seen on
MRI in the pregnant women with cervical insufficiency were
3.5 ± 0.8 cm (0.8–4.9 cm) and 28.7 ± 6.3 mm (9–41 mm),
respectively.

Table 2. Summary measurements on the general characteristics of the pregnant women in the control group and the group with
For personal use only.

cervical insufficiency.

Groups
Characteristics Measurements CI Control Total p value
Gestational age (weeks) Mean 13.5 13.6 13.5 0.137*
Median 13.0 13.5 13.0
Minimum 10.0 12.0 10.0
Maximum 28.0 16.0 28.0
SD 3.0 1.3 2.7
Age (years) Mean 31.3 30.6 31.2 0.693y
Median 32.0 32.0 32.0
Minimum 18.0 21.0 18.0
Maximum 42.0 38.0 42.0
SD 5.4 5.4 5.4
Periendocervical hyposignal No 7 (14.6%) 10 (100%) 17 (29.3%) 50.001z
Yes 41 (85.4%) – 41 (70.7%)
Total 48 (100%) 10 (100%) 58 (100%)
Loss of PESZ No 13 (26.5%) 10 (100%) 23 (39.0%) 50.001z
Yes 36 (73.5%) – 36 (61.0%)
Total 49 (100%) 10 (100%) 59 (100%)
Sludge signal No 19 (41.3%) 10 (100%) 29 (51.8%) 0.001z
Yes 27 (58.7%) – 27 (48.2%)
Total 46 (100%) 10 (100%) 56 (100%)
Anatomical Biometry (cm) Mean 3.5 4.3 3.6 50.001*
Median 3.6 4.3 3.6
Minimum 0.8 3.8 0.8
Maximum 4.9 4.8 4.9
SD 0.8 0.3 0.8
Functional Biometry (mm) Mean 28.7 42.6 31.2 50.001y
Median 30.0 42.0 31.5
Minimum 9.0 38.0 9.0
Maximum 41.0 48.0 48.0
SD 6.3 3.0 7.9

CI, cervical insufficiency; PESZ, periendocervical stromal zone; SD, standard deviation.
*Mann–Whitney test; yStudent’s t test for independent samples; zFisher’s exact test.
 4 V. V. F. Habib et al. J Matern Fetal Neonatal Med, Early Online: 1–6

Figure 2. Magnetic resonance image of the cervix of pregnant woman at 12 weeks of gestational age with cervical insufficiency. Note the periorificial
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by McMaster University on 10/17/14

hypointense and loss of stromal zone at the sagittal TURBO FISP sequence. The functional and anatomical lines were represented in (a).
For personal use only.

Figure 3. (a) Pregnant woman with 13 weeks of gestational age and cervical insufficiency. Observed periorificial hypointense and loss of definition of
periendocervical stromal zone (b) obtained 1 week after the previous one, with hypointense material within of amniotic sac (sludge signal).

In comparing the profiles of the group with cervical
insufficiency and the control group, we noted that they had
the same profile with regard to gestational age (p ¼ 0.137) and
maternal age (p ¼ 0.693). However, they differed in relation
to periendocervical hyposignal in the turbo FISP sequence
(p50.001), loss of the PESZ (p50.001) and presence of the
sludge sign (p ¼ 0.001), which occurred more frequently
among the pregnant women with cervical insufficiency than
among the controls. On the other hand, the pregnant women
with cervical insufficiency presented lower values for
anatomical biometry (p50.001) and functional biometry
(p50.001) than seen in the controls.
Discussion
This study sought to establish and characterize the main
morphological criteria in MRI of the cervix, among patients
known to present cervical insufficiency for whom cerclage
was indicated, in comparison with normal pregnant women,
with regard to the morphological context of elasticity,
hydration and contractility.
It has now been proven that a short cervix in the second
trimester is the best predictor of cervical insufficiency but,
over the last two decades, controversy regarding its treatment
has persisted. It remains uncertain whether pregnant women
with a short cervix would benefit from treatment with
progesterone or cerclage and whether, if they were to choose
treatment, the chances of premature delivery would be
reduced [1].
Few published papers on use of MRI during pregnancy for
precise evaluation of the cervix are available in the literature
[5,7–12]. Such studies have attempted to demonstrate the
appearance of the cervix through its signals and pelvic
examination protocols, without taking the gestational state
into consideration. In our study, we sought to develop a fast
protocol in which the pregnant woman would feel comfort-
able, in apparatus with a high field, excellent gradient and
high resolution that would identify the signals and the
contractility of the cervix through fast turbo FISP sequences.
Although preliminary, our study was the first to access the
cervix using signs described through MRI, in terms of both
morphology and signal intensity, thereby enabling better
knowledge of the cervix in pregnant women with and without
cervical insufficiency.
At the end of the 1980s, transvaginal US was introduced as
a promising method for studying the female internal genital
 DOI: 10.3109/14767058.2014.928858
organs, in comparison with US via the suprapubic or vaginal
palpation route [12–15]. According to some studies, transva-
ginal US is an objective, safe and non-invasive method that
enables faithful examination of the cervix and lower segment
of the uterus, especially during the second trimester of
pregnancy, thereby increasing the screening for preterm
delivery [4,5].
MRI has been shown to be superior to US with regard to
resolution and tissue contrast [10,16]. It is a promising
method for evaluating the stromal zone and makes it possible
to understand the physiological and non-physiological alter-
ations of the cervix during pregnancy [8–10,15,16]. It has also
been reported that MRI is the only examination capable of
supplying information on the cervical stroma, thus enabling
greater understanding of the physiological modifications of
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by McMaster University on 10/17/14
the cervix [9]. The cervix is usually formed by the cervical
stroma, which extends from the external orifice to the internal
orifice, as a band of hyposignal in the most proximal portion
of the endocervical canal, followed by a band of isosignal or
hypersignal immediately adjacent to this, which corresponds
to the endocervical glands and generally has continuous
thickness and a homogenous pattern, without any well-
defined failures. This characterization corresponds to the
CGA that has been described for US, and the PESZ that was
defined in the present study [10–13].
The qualitative variations in the cervical signal correlate
with the components of the endocervix. The cervix is a
For personal use only.
fibromuscular stroma and 15% of the stroma is composed of
smooth musculature, located on the periphery of the cervix
[4,5]. Collagen, glandular epithelium and mucus predominate
inside the cervix [4,5]. It is likely that women with cervical
insufficiency present alterations in these components, thus
leading to imaging abnormalities such as hyposignal (dehy-
dration, fibrosis and diminished mucus and collagen) and
contractions, relating to cervical insufficiency. Our study
showed hyposignal, loss of the PESZ and functional alter-
ations that were probably correlated with such abnormalities,
which had not been described previously. According to an
observational study on 91 pregnant patients between their
35th and 40th weeks, there is a significant correlation and a
directly proportional relationship between signal intensity and
the number of weeks until the start of delivery labor. For this
reason, we believe that the MRI signal, with sequences well
directed towards pregnancy is a fundamental criterion in this
evaluation, relating to prematurity [8].
In a novel manner, we studied the cervical signal in fast
gradient echo sequences, with HASTE-locating T2 weighting
and turbo FISP. These sequences were performed using
Siemens equipment, but this study can also be reproduced
using equipment from other manufacturers, such as the
FIESTA sequence (General Electric) and BTFE (Philips). The
great advantages of these sequences are their rapidity, lower
number of artifacts, greater comfort for the pregnant woman
and possibility of making a dynamic evaluation on the cervix.
We observed that the length of the cervix was directly
related to presence or absence of cervical insufficiency.
However, the absolute number of 3.6 cm for the anatomical
biometry in pregnant women with cervical insufficiency is not
considered to be abnormal [17]. On the other hand, the mean
for the functional biometry among the patients with cervical
Cervical insufficiency on magnetic resonance imaging 5
insufficiency was 2.8 cm, which was significantly lower than
among the patients without cervical insufficiency, for whom
the mean was 4.2 cm. This may be a better predictor for
preterm delivery. We took a short cervix to be 2.5 cm between
the 20th and 24th weeks, which is the maximum measurement
described in the literature [8]. Presence of a short cervix is
certainly compatible with cervical insufficiency, but in
pregnant women with a normal cervix, other signs need to
be sought, even before the shortening of the cervix, as was
demonstrated in some cases in our study. However, for this
analysis, it would be necessary to increase the sample size,
both for cases with cervical insufficiency and for controls, and
also to make an inter-observer assessment of these
measurements.
In agreement with our study, a study conducted in 2005
analyzed the biometric modifications and characteristics of
the endocervical tissue in second-trimester pregnant women
and found that there were modifications to the biometry and
cervical signal as the gestational age increased and with
increasing presence of early vaginal delivery. The study group
in the 2005 study had a slightly older gestational age than in
our study, but the mean length of 3.5 cm was similar to the
finding in our study [10].
Most studies that have compared MRI and US for
evaluating cervical biometry have not presented any statistical
differences between them [4,5]. This is understandable, given
the good assessment of the cervix that can be made
endovaginally. However, in these studies, MRI was performed
using pelvic protocols that were unsuitable for the gestational
phase, which impaired their performance [4,5]. High-field
equipment enables access to the molecular and functional
characteristics of the cervix, with excellent definition for
analysis on the signal intensity in the different sequences.
A prospective study on 100 pregnant women who were
admitted due to preterm labor at between the 18th and 34th
weeks classified the identification of the stromal zone as high,
intermediate or low, based on differences in MRI signal
intensity in the stromal zone in relation to the endocervical
canal and the remainder of the cervix. These authors
considered that their method was feasible, but that it had
low sensitivity and no advantages in relation to US, especially
with regard to cost [18]. Differently, our study evaluated
pregnant women with cervical insufficiency and compared
them with a control group. Moreover, we analyzed pregnant
women at a gestational age of around 13 weeks, i.e. outside of
the delivery labor period, since we believed that with
advancing gestational age and in situations of contractions,
there could be an alteration to the cervix that would modify
the assessment of cervixes with a propensity to cervical
insufficiency. We found greater sensitivity than in the
abovementioned study, through using fast sequences indicated
at an earlier gestational period.
The present study is a pioneer in identifying pregnant
women with cervical insufficiency preselected for cerclage, in
comparison with a group of pregnant women without risk
factors for cervical insufficiency. Furthermore, this study
created a protocol directed towards studying the cervix during
the first half of pregnancy.
In summary, we demonstrated that the main signs of
cervical insufficiency seen on cervical MRI in pregnant
 6 V. V. F. Habib et al.
women were periendocervical hyposignal and lack of defin-
ition of the PESZ, along with functional biometry lower than
anatomical biometry. Further studies with larger samples are
needed in order to confirm our findings.
Declaration of interest
The authors declare no conflicts of interests. The authors
alone are responsible for the content and writing of this
article.
References
1. Feltovic H, Hall TJ, Berghella V. Beyond cervical length: emerging
technologies for assessing the pregnant cervix. Am J Obstet
Gynecol 2012;207:345–54.
2. Gravett MG, Rubens CE, Nunes TM; GAPPS Review Group.
Global report on preterm birth and stillbirth (2 of 7): discovery
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by McMaster University on 10/17/14
science. BMC Pregn Childbirth 2010;10:S2.
3. Romero R. Vaginal progesterone to reduce the rate of preterm birth
and neonatal morbidity: a solution at last. Womens Health (Lond
Engl) 2011;7:501–4.
4. Brandão RS, Pires CR, Souza E, et al. [Biometric analysis of
uterine cervix during pregnancy using transvaginal ultrasonography
and magnetic resonance imaging]. Radiol Bras 2008;1:235–9.
5. Levine D, Hulka CA, Ludmir J, Li W, Edelman RR. Placenta
accreta: evaluation with color Doppler US, power Doppler US, and
MR imaging. Radiology 1997;205:773–6.
6. Matijević R, Grgić O, Klobucar A. [Diagnosis and therapy of
cervical insufficiency]. Acta Med Croatica 2003;57:287–94.
7. Oláh KS. The use of magnetic resonance imaging in the assessment
of the cervical hydration state. Br J Obstet Gynaecol 1994;101:
For personal use only.
255–7.
J Matern Fetal Neonatal Med, Early Online: 1–6
8. Chan YL, Lam WW, Lau TK, et al. Cervical assessment by
magnetic resonance imaging – its relationship to gestational age
and interval to delivery. Br J Radiol 1998;71:155–9.
9. Rae DW, Smith FW, Templeton AA. Magnetic resonance imaging
of the human cervix: a study of the effects of prostaglandins in the
first trimester. Hum Reprod 2001;16:1744–7.
10. House M, O’Callaghan M, Bahrami S, et al. Magnetic resonance
imaging of the cervix during pregnancy: effect of gestational
age and prior vaginal birth. Am J Obstet Gynecol 2005;193:
1554–60.
11. Hricak H, Chang YC, Cann CE, Parer JT. Cervical incompetence:
preliminary evaluation with MR imaging. Radiology 1990;174:
821–6.
12. Scoutt LM, McCauley TR, Flynn SD, et al. Zonal anatomy of the
cervix: correlation of MR imaging and histologic examination of
hysterectomy specimens. Radiology 1993;86:159–62.
13. Berghella V, Berghella M. Cervical length assessment by ultra-
sound. Acta Obstet Gynecol Scand 2005;84:543–4.
14. Gomez R, Galasso M, Romero R, et al. Ultrasonographic exam-
ination of the uterine cervix is better than cervical digital
examination as a predictor of the likelihood of premature delivery
in patients with preterm labor and intact membranes. Am J Obstet
Gynecol 1994;171:956–64.
15. Carr DB, Smith K, Parsons L, et al. Ultrasonography for cervical
length measurement: agreement between transvaginal and transla-
bial techniques. Obstet Gynecol 2000;96:554–8.
16. Smith FW, Adam AH, Phillips WD. NMR imaging in pregnancy.
Lancet 1983;1:61–2.
17. Brown R, Gagnon R, Delisle MF, et al. Cervical insuffi-
ciency and cervical cerclage. J Obstet Gynecol Can 2013;35:
1115–27.
18. de Tejada BM, Faltin DL, Kinkel K, et al. Magnetic resonance
imaging of the cervix in women at high risk for preterm delivery.
J Matern Fetal Neonatal Med 2011;24:1392–7.