ORIGINAL ARTICLE

Follicular uid antimullerian hormone (AMH) does

not predict IVF outcomes in polycystic ovary syndrome
patients
Bushra Abu-Fakher
a,
*
, Faizeh Al-Quobaili
b
, Marwan Alhalabi
c,d
a
Faculty of Pharmacy, Damascus University, Damascus, Syria
b
Department of Clinical Biochemistry and Microbiology, Faculty of Pharmacy, Damascus University, Syria
c
Department of Embryology and Reproductive Medicine, Faculty of Medicine, Damascus University, Syria
d
Assisted Reproduction Unit, Orient Hospital, Damascus, Syria
Received 28 October 2012; accepted 9 December 2012
Available online 4 February 2013
KEYWORDS
Follicular uid;
Antimullerian hormone;
AMH;
IVF;
PCOS
Abstract Introduction: AMH, is a member of the transforming growth factor (TGF-b) produced
by granulosa cells with the highest expression being in small antral follicles. Our objective is to eval-
uate the predictive value of follicular AMH levels as an indicator of IVF outcomes in PCOS
patients.
Material and methods: A total of 63 patients undergoing the IVF trail were recruited for this pro-
spective case-control study. The patients were classied into three groups: Group I: 43 patients with
Polycystic ovary syndrome (PCOS). Group II: 20 normo-ovulatory patients were recruited for the
IVF trail because of the male infertility factor (control group). Group III: consists of 33 patients
from group I recruited to determine the effect of controlled ovarian hyperstimulation (COH) on
serum AMH levels. Serum AMH levels were performed on day 3 of the menstrual cycle whereas
the follicular uid AMH level has been evaluated at the time of oocyte retrieval by enzyme linked
immune-sorbent assay (AMH Gen II ELISA kit, Beckman Coulter, Inc. USA). AMH concentra-
tions were adjusted to its protein content, to avoid possible bias due to FF volume variability.
The relationship between follicular uid AMH (FF AMH) levels and the number of oocytes
retrieved, number of mature oocytes, number of embryos, fertilization and clinical pregnancy rate
were assessed. The demographic data were similar between two groups as regards age and BMI.
Results: Both serum AMH levels and FF AMH levels were signicantly higher in PCOS patients
than in the controls, the values for serum levels were (5.07 3.39 ng/ml, 3.07 2.09 ng/ml) respec-
tively and for follicular uid levels were (67 50.7 ng/g
*
, 38.5 44.4 ng/g) respectively. Serum
*
Corresponding author. Tel.: +963 932749900.
E-mail address: ph.bushra@outlook.com (B. Abu-Fakher).
Peer review under responsibility of Middle East Fertility Society.
Production and hosting by Elsevier
Middle East Fertility Society Journal (2013) 18, 110114
Middle East Fertility Society
Middle East Fertility Society Journal
www.mefsjournal.org
www.sciencedirect.com
1110-5690 2013 Middle East Fertility Society. Production and hosting by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.mefs.2012.12.005
AMH levels showed a positive correlation with oocytes retrieved and mature oocyte number, but
there was no signicant correlation with good quality embryo number, fertilization rate or clinical
pregnancy rate in PCOS patients. On the other hand FF AMH levels do not show correlation with
any of the IVF outcomes mentioned above in PCOS patients. Interestingly, the interval change in
serum AMH levels between baseline AMH levels and levels after COH had a signicant predictive
value of pregnancy.
Conclusions: Serum AMH levels can predict the number of the retrieved oocytes and mature
oocytes, whereas FF AMH may not be a valuable predictor for IVF outcomes in PCOS patients.
2013 Middle East Fertility Society. Production and hosting by Elsevier B.V. All rights reserved.
1. Introduction
Polycystic ovary syndrome is the most frequent cause of
anovulatory infertility and hyperandrogenism in young women
(1). It affects upto 10% of reproductive age women (2). PCOS
is characterized by a clustering of hyperandrogenism, hyperin-
sulinemia, hypersecretion of LH, menstrual dysfunction, hir-
sutism, infertility and pregnancy and neonatal complications
(3). Although PCOS patients are typically characterized by
producing an increased number of oocytes, they are often of
poor quality, leading to lower fertilization, cleavage and
implantation rates, and a higher miscarriage rate in PCOS pa-
tients undergoing IVF treatment (4).
AMH is a glycoprotein of the transforming growth factor-
beta (TGF-b) super family. AMH is produced by granulosa
cells from 36 weeks of gestation until menopause, with the
highest expression being in small antral follicles. AMH pro-
duction gradually declines as follicles grow, once follicles reach
a size at which they are dominant it has largely disappeared. Its
removal from these larger follicles appears to be an important
requirement for dominant follicle selection and progression to
ovulation as AMH has an inhibitory role in the ovary, reduc-
ing both primordial follicle initiation and follicle sensitivity to
FSH by the inhibition of aromatase. It is for this reason that
AMH is a focus of interest in polycystic ovary syndrome (5).
Women with PCOS have elevated serum and FF AMH lev-
els versus those of normal controls (6). Recent study suggests
that increased FF AMH in women with PCOS may have
harmful consequences on oocyte quality and maturation, via
an unclear molecular mechanism, but does not have an effect
on pregnancy rates (7), whereas other investigations reveal that
women with PCOS who have lower FF AMH levels have sim-
ilar rates of oocyte maturation, fertilization and embryonic
development compared with normo ovulatory women (8).
To determine whether AMH could play a role in the success
rate of IVF- embryo transfer, this study aims to conrm the
relationship between serum and intrafollicular AMH levels,
and to determine the relationship of intrafollicular AMH to
IVF outcomes in PCOS patients and normal ovulatory women.
2. Materials and methods
A prospective case control study was performed in a total of 43
infertile PCOS patients and 20 normo-ovulatory patients. The
diagnosis of PCOS was based on Rotterdam criteria 2003, the
association of at least two of the three following criteria:
1) Ovulatory disturbance, mainly oligomenorrhea or
amenorrhea.
2) Hyperandrogenism as dened either clinically by hirsut-
ism (modied Ferriman and Gallwey score > 6), or
severe acne/seborrhea, and/or biologically by a testoster-
one serum level greater than 0.7 ng/ml and/or D4-andro-
stenedione greater than 2.2 ng/ml.
3) More than 12 follicles in the 29 mm range in each ovary
with peripheral distribution at ultrasound and/or an
ovarian volume higher than 10 ml were found. Ultra
sound examination was performed with a 7.5-MHz
transvaginal transducer.
All the control population underwent the IVF trail because of
the male factor. Exclusion criteria for the controls were a his-
tory of menstrual disturbances, absence of one of the ovaries,
abnormal serum levels of prolactin or androgens (i.e. serum
testosterone above 0.7 ng/ml and/or D4-androstenedione
above 2.2 ng/ml) and any hormonal treatment during the
3 months before the study.
The patients followed a long protocol for controlled ovar-
ian hyperstimulation which was initiated with the oral contra-
ceptive pill (OCP) on day three or ve of the cycles and
continued with pituitary downregulation by a gonadotropin-
releasing hormone (GnRH) agonist consisting of 0.05 mg of
buserelin in the mid-luteal phase (day 21) of the cycle, given
daily until the day of human chorionic gonadotropin (hCG)
administration.
On day 23 of the new cycle, ovarian stimulation was
started with an injection of recombinant FSH or human men-
opausal gonadotropin (hMG). The gonadotropin dosage was
adjusted according to the follicular growth, which was moni-
tored by ultrasound. The patients were given 10 000 IU of
hCG when at least three follicles became more than 17
18 mm. The oocytes were retrieved 35 h after hCG administra-
tion and IVF/ICSI followed by embryo transfer was per-
formed 3 days after oocyte retrieval.
Blood samples were collected on day three of the menstrual
cycle. On the day of oocyte retrieval, the follicular uid was
aspirated from all follicles (1420) mm, and then it was centri-
fuged at 4000 rpm for 10 min and stored at 80C. All follic-
ular uid samples were examined on the same day to measure
AMH levels using enzyme linked immunosorbent assay (Beck-
man Coulter, Inc. USA). AMH concentrations were adjusted
to its protein content, to avoid possible bias due to FF volume
variability, by the Bradford assay. The assay was performed in
a 96 well plate using Bradfords reagent. Suitable dilutions of
bovine serum albumin (BSA) 1 lg/ml were used as standard
solutions. All FF samples were diluted 250 times, the absor-
bance of standard and unknown solutions were determined
using a spectrophotometer. The concentrations of AMH in fol-
Follicular uid antimullerian hormone (AMH) does not predict IVF outcomes in polycystic ovary syndrome patients 111
licular uid were presented as nanogram per gram (ng/g) of
protein.
3. Statistical analysis
The data were analyzed with MedCalc

v12.2.1. Data were ex-

pressed as mean standard deviation. Comparison of the
means was performed by the MannWhitney U test while
the degree of association between continuous variables was
calculated by the Spearmans correlation coefcient.
A p value <0.05 was considered statistically signicant.
4. Results
Demographic characteristics of each group (age, BMI, serum
AMH, FF AMH, retrieved oocyte number, mature oocyte
number, embryos number and fertilization rate are shown in
Table 1.
As expected PCOS patients had signicantly higher serum
AMH levels than controls (5.07 3.9 ng/ml, 3.07 2.09 ng/
ml) respectively (p < 0.01). On the other hand FF AMH levels
were signicantly higher in PCOS patients compared to the
controls (67 50.7 ng/g, 38.5 44.4 ng/g) respectively
(p < 0.05). Serum AMH levels were negatively correlated to
age in the control group (r =0.45, p = 0.04) but not in
PCOS patients. (r is Spearman correlation coefcient).
In PCOS patients, serum AMH levels were positively corre-
lated to the number of oocytes retrieved (r = 0.39, p = 0.01)
and the number of mature oocytes (r = 0.334, p = 0.03) (Fig-
ure 1) but not to the number of total embryos or the fertilization
rate, whereas follicular uid AMH levels were not signicantly
correlated to any of the IVF outcomes mentioned above.
Similar results were found in normo ovulatory women; ser-
um AMH levels were positively correlated to oocytes retrieved
(r = 0.513, p = 0.02) mature oocytes (r = 0.615, p = 0.002)
(Figure 1) and embryo number (r = 0.625, p = 0.002).
FF AMH levels were not signicantly correlated to any of
the IVF outcomes as well. Both serum and FF AMH levels
had not a signicant predictive value of pregnancy. But inter-
estingly the interval change between baseline serum AMH lev-
els and levels after controlled ovarian hyperstimulation (values
are shown in Table 2) had a signicant predictive value of
pregnancy.
The predictive potency of the AMH interval change was
tested by the Receiver operating characteristic (ROC) proce-
dure. The area under the curve was 0.697. Several cutoff values
of AMH interval change were analyzed in terms of specicity
and sensitivity from the ROC curve data (Figure 2).
Table 3. Showed that the best adjustment between specic-
ity (92%) and sensitivity (54%) was obtained with a cutoff va-
lue of 54.7%.
5. Discussion
The nding of increased AMH levels in both follicular uid
and the serum of PCOS patients compared with controls can
be explained with the fact that PCOS patients have increasing
number of small antral follicles which are the major sites of
producing AMH and further more by the nding of Pellat
et al. who demonstrated that AMH production was on average
75 times higher per granulosa cell from anovulatory polycystic
ovaries than from normal cells (9). Our nding is in agreement
with the study of Fallat et al. who reported that the level of
AMH in the follicular uid and serum of patients with PCOS
undergoing IVF was signicantly higher than that in patients
with endometriosis or pelvic adhesions (10). We found a neg-
ative correlation between age and serum AMH only in the con-
trol group whereas FF AMH levels remain unaffected by age,
this nding is in agreement with both Das M et al. (11) and
Pigny et al. (6).
We found that serum AMH levels in PCOS patients can
predict oocyte quantity and quality (positive correlation with
mature oocytes) and this nding is in accordance with both
Arabzadeh et al. (12) and Aleyasin et al. (13).
No correlation between FF AMH and any of IVF out-
comes mentioned before were found in the PCOS group, this
is partially in agreement with Desforges who suggests that in-
creased FF AMH in women with PCOS may have harmful
consequences on oocyte quality but does not have an effect
Figure 1 Relationship between serum AMH on day three and
the number of mature oocytes in both PCOS and control groups.
Table 1 Clinical and biological features of patients.
Controls
(n = 20)
PCOS patients
(n = 43)
p-value
Age (year) 29.2 6.2 29.2 5.2 NS
BMI (kg/m
2
) 25.5 3.9 26.3 6.2 NS
sAMH (ng/ml) 3.07 2.09 5.07 3.39 0.004
FF AMH (ng/g) 38.5 44.4 67 50.7 0.02
Retrieved oocyte no. 14.4 8.2 20.1 11.2 0.05
Mature oocyte no. 10.1 5.3 13.3 8.6 0.04
Total embryos no. 8 4.3 13.2 9.1 0.04
Fertilization rate 84.2 19.4 83.8 21.7 NS
*NS = non-signicant; Values are expressed as the mean SD.
112 B. Abu-Fakher et al.
on pregnancy rates (7). Mashiach found that women with
PCOS who have lower FFAMH levels have similar rates of oo-
cyte maturation, fertilization, and embryonic development
compared with ovulating women (14). On the other hand, con-
icting results were obtained from Pabuccu 2009 who demon-
strated better fertilization, implantation and clinical pregnancy
rates in the PCOS group with highest FFAMH concentration
compared with any other group with a lower concentration (8).
In normo-ovulatory women, we found that serum AMH
levels could be a good predictor for oocyte quantity and qual-
ity (positive correlation with maturation rate) and furthermore
for embryo number, and this is in agreement with Vanrooij
et al. 2002 who demonstrated that serum AMH levels are a
good predictor of retrieved oocytes (15) and with Arabzadah
2010 who found that serum AMH levels can predict both the
quantity and quality of oocytes (12). FF AMH levels do not
relate to any of the IVF outcomes in PCOS patients and this
is in agreement with Yilmaz 2012 who showed that there is
no signicant correlation between FF AMH and IVF out-
comes in patients underwent IVF for male infertility or unex-
plained infertility(16). Fanchin et al. suggest that FF AMH
concentrations are only strongly and positively associated with
oocyte quality and implantation rates, but not with fertiliza-
tion rate, and embryo morphology in normal ovulation (17)
and this conict can be explained by the differences in the pa-
tients groups since Fanchin et al. studied non PCOS infertile
patients who underwent the IVF trial.
There are conicting results about the relation between
AMH levels and pregnancy, the most recent studies demon-
strate that Clinical pregnancy rates were lower in the high
AMH group than that of the low and average groups but this
difference was only close to statistical signicance Xi et al.
2012 (18), whereas Yilmaz 2012 found that FF AMH levels
can predict the recovery of oocytes but not oocyte quality, em-
bryo quality or pregnancy in non-obese non-hyperandrogene-
mic PCOS patients (16). Desforges 2009 demonstrated that the
AMH levels were signicantly lower in patients who began a
pregnancy (7).
So, we concluded that AMH may not be a valuable predic-
tor for outcomes success in PCOS patients and normo-ovula-
tory women undergoing IVF. Other randomized control
studies are needed.
6. Conict of interest
None.
References
(1) Franks S. Polycystic ovary syndrome. N Engl J Med
1995;333:85361.
(2) Norman R, Dewailly D, Legro R, Hickey T. Polycystic ovary
syndrome. Lancet 2007;370:68597.
(3) Moran L, Teede H. Metabolic features of the reproductive
phenotypes of polycystic ovary syndrome. Hum Reprod Update
2009;15:47788.
(4) Boomsma CM, Fauser BCJM, Macklon NS. Pregnancy compli-
cations in women with Polycystic ovary syndrome. Semin Reprod
Med 2008;26:7284.
(5) Pellatt L, Rice S, Mason H. Anti-Mullerian Hormone and
polycystic ovary syndrome: a mountain too high. Reproduction
2010;139:82533.
(6) Pigny P, Merlen E, Robert Y, Cortet-Rudelli C, Decanter C,
Jonard S, Dewailly D. Elevated serum level of Anti-Mu llerian
hormone (AMH) in polycystic ovarysyndrome: relationship to the
ovarian follicle excess and to the follicular arrest. J Clin
Endocrinol Metab 2003;88:595762.
(7) Desforges-Bullet V, Gallo C, Lefebvre C, Pigny P, Dewailly D,
Jonard CJ. Increased anti mullerian hormone and decreased FSH
levels in follicular uid obtained in women with polycystic ovaries
at the time of follicle puncture for in vitro fertilization. Fertil
Steril 2010;94:198204.
Table 2 Serum AMH interval change.
Baseline serum AMH levels
(ng/ml)
Serum AMH after COH
(ng/ml)
Interval change
(%)
PCOS patients
a
(n = 33) 5.4 3.4 1.9 1.3 60.8 20.9
Controls
a
(n = 20) 3.07 2.09 1.04 1.06 68.7 16.6
a
mean SD.
Figure 2 ROC curve for AMH interval change.
Table 3 ROC curve data.
AMH interval change (%) Specicity (%) Sensitivity (%)
36.8 20.8 100
47.1 37.5 96
54.7 54.1 92
68.4 58.3 48
80.4 91.6 28
Follicular uid antimullerian hormone (AMH) does not predict IVF outcomes in polycystic ovary syndrome patients 113
(8) Pabuccu R, Kaya C, Caglar GS, Oztas E, Satiroglu H. Follicular-
uid anti-Mullerian hormone concentrations are predictive of
assisted reproduction outcome in PCOS patients. Reprod Biomed
Online 2009;19:6317.
(9) Pellatt L, Hanna L, Brincat M, Galea R, Brain H, Whitehead S,
Mason H. Granulosa cell production of anti-Mullerian hormone
is increased in polycystic ovaries. J Clin Endocrinol Metab
2007;92:2405.
(10) Fallat ME, Siow Y, Marra M, Cook C, Carrillo A. Mullerian-
inhibiting substance in follicular uid and serum: a comparison of
patients with tubal factor infertility, polycystic ovary syndrome,
and endometriosis. Fertil Steril 1997;67:9625.
(11) Das M, Gillot D, Saridogan E, Djahanbakhch O. Anti-Mullerian
hormone is increased in follicular uid from unstimulated ovaries in
womenwithpolycysticovarysyndrome. HumReprod2008;23:21226.
(12) Arabzadeh S, Hossein G, Rashidi B, Hosseini M, Zeraati H.
Comparing serum basal and follicular uid levels of anti-
Mu llerian hormone as a predictor of in vitro fertilization
outcomes in patients with and without polycystic ovary syn-
drome. Ann Saudi Med 2010;30:4427.
(13) Aleyasin A, Aghahoseini M, Mokhtar S, Fallahi P. Anti-
mullerian hormone as a predictive factor in assisted reproductive
technique of polycystic ovary syndrome patients. Acta Med Iran
2011;49(11):71520.
(14) Mashiach R, Amit A, Hasson J, Amzalzg S, Almog B, Ben-Yosef
D, et al. Follicular uid levels of anti-Mullerian hormone as a
predictor of oocyte maturation, fertilization rate, and embryonic
development in patients with polycystic ovary syndrome. Fertil
Steril 2010;93:2299302.
(15) Van Rooij IA, Broekmans FJ, Te Velde ER, Fauser BC, Bancsi
LF, Jong FH, Themmen AP. Serum AMH levels: a novel measure
of ovarian reserve. Hum Reprod 2002;17:306571.
(16) Yilmaz N, Uygur D, Dogan M, Ozgu E, Salman B, Mollamahm-
utoglu L. The effect of follicular antimullerian hormone levels of
non-obese, non-hyperandrogenemic polycystic ovary syndrome
patients on assisted reproduction outcome. Gynecol Endocrinol
2012;28(3):1625.
(17) Fanchin R, Mendez Lozano DH, Frydman N, Gougeon A, di
Clemente N, Frydman R, Taieb J. Anti-Mullerian hormone
concentrations in the follicular uid of the preovulatory follicle
are predictive of the implantation potential the ensuing embryo
obtained by in vitro fertilization. J Clin Endocrinol Metab
2007;92:1796802.
(18) Xi W, Gong F, Lu G. Correlation of serumAnti-Mu llerian hormone
concentrations on day 3 of the in vitro fertilization stimulation cycle
with assisted reproduction outcome in polycystic ovary syndrome
patients. J Assist Reprod Genet 2012;29(5):397402.
114 B. Abu-Fakher et al.