1Senior consultant, Department of reproductive medicine, Jindal IVF and Sant Memorial Nursing Home, India
2Director, Department of reproductive medicine, Jindal IVF and Sant Memorial Nursing, India
3Ex resident of Department of reproductive medicine, Jindal IVF and Sant Memorial Nursing Home, India
4Senior embryologist, Department of reproductive medicine, Jindal IVF and Sant Memorial Nursing Home, India
5Ex-assistant Professor, Department of Obstetrics and Gynaecology, Post Graduate Institute of Medical Education and Research, India
Submission: April 10, 2018;Published: May 31, 2018
*Corresponding author: Dr Swati Verma, Senior consultant, Jindal IVF center, sector 20 D, Chandigarh, India, Tel: 9646004459;
How to cite this article: Swati V, Umesh N J, Sonia G, Sanjeev M, Bharti J. Outcome of In Vitro Fertilization in Women with Discordant Values of Anti Mullerian Hormone and Antral Follicle Count. Glob J Reprod Med. 2018; 4(5): 555648. DOI:10.19080/GJORM.2018.04.555648.
Objective: To evaluate the IVF outcome & compare the Metaphase 2 (M2) oocytes rate in in women showing discordance between AMH and AFC.
Design: Retrospective analysis
Settings: Women undergoing IVF cycle
Patients: Total of 807 women undergoing first IVF cycle receiving standard antagonist protocol for IVF treatment was included.
Interventions: All women of study group were classified into three groups depending upon their AFC count and for each AFC group, AMH range was identified where best pregnancy outcome was obtained.
Main Outcome measures: IVF outcome in women with discordant AFC & AMH.
Results: For each AFC group, AMH cut off values were different according to their respective percentiles. Only 20.07% ( 162/807) women among three AFC groups were observed showing concordance between AMH and AFC. Within each AFC group women showing concordance between AMH and AFC were associated with better IVF outcomes. Despite higher M2 rates reported with higher AMH values in each AFC category, lower pregnancy rates were reported.
Conclusion: Discordance between AFC and AMH they may have adverse impact on outcome of IVF.
Among various proposed biomarkers, Antral Follicle Count (AFC) and Anti Mullerian Hormone (AMH) are said to be most favorable in predicting ovarian reserve and response to stimulation.Both generally have good correlation with each other and are often used interchangeable. There is a growing evidence of consistent association of these biomarkers with ovarian response, number of retrieved oocytes and live birth rate [1-4]. Nonetheless discordance between the two does occur and has been discussed in recent trials [5,6]. The discordance between AFC and AMH raises a doubt regarding the usefulness of AMH or possibility of some laboratory error . It may also suggest
existence of some subtle ovarian pathology, not demonstrated by
ultrasonography done for assessment of AFC. A very limited data
exits on IVF outcome in women with discordance between AMH
and AFC. Therefore we conducted index review among women
undergoing In-Vitro Fertilization (IVF) at our center with the
a) To define the discordance between AMH and AFC in
women undergoing IVF
b) To study the IVF outcome in women showing
discordance between AMH and AFC
c) To compare the Metaphase 2 (M2) oocytes rate in
women with discordant AMH and AFC in relation to their IVF
This retrospective analysis was conducted from January 2012
to December 2015 at an infertility centre of North India. There
were 1754 women who underwent IVF cycle during this period.
After fulfilling inclusion criteria 807 women were found eligible
for the study (Figure 1). The women included in the analysis were
with their age below 35 years, day two serum follicle stimulating
hormone (S.FSH) levels below 10IU/ml, first IVF cycle with
self oocytes, received ovarian stimulation with standard GnRH
antagonist protocol along with fresh embryo transfer done. Most
of the women with polycystic ovaries (PCO) were excluded from
the study as fresh embryo transfer was not possible to avoid
risk of OHSS. The women with PCO who underwent fresh ET
were included in the analysis. Although this was a retrospective
analysis and did not involve any active intervention on patients,
approval was taken from the institutional review board.
The ovarian reserve was assessed by AFC, serum AMH and
day 2 S.FSH as a routine. Total number of AFC was measured on
day 2 of cycle by transvaginal ultrasound scan and all the follicles
ranging from 2 to 9 mm were included. AMH measurement
was done using the AMH Gen II ELISA kit .The assay kit has a
sensitivity of 0.08ng/ml, and intra- and inter-assay coefficients
of variation of less than 5.4 and 5.6 respectively.
Standard GnRH antagonist protocol for ovarian stimulation
was followed in all women. Starting dose of injection
Gonadotrophins was decided according to the value of AFC, AMH
and S. FSH. Ultrasound guided follicular monitoring was done
and injection GnRH antagonist. 25mg S/C was added once one
or more follicles reaches 13-14mm in size. Dosages of injection
Gonadotrophins were adjusted according to ovarian response.
Injection urinary hCG 10000 I.U was given as oocyte maturation
trigger in the patients who were not at risk of OHSS. Women with
more than 12 follicles of 16 mm and serum oestradiol levels more
than 3500pg/ml were given modified oocytes maturation trigger
with injection decapeptyl 0.2mg subcutaneous. Oocyte retrieval
was done 34-36 hours after the maturation trigger. All women
received injection progesterone 50 mg intramuscular daily for
luteal phase support, started after oocyte retrieval. Women
who received modified maturation trigger were administered
injection urinary hCG 1500I.U on the day of oocyte retrieval and
tablet oestradiol valerate 2mg, 8 hourly along with standard
luteal phase support. All women were subjected to embryo
transfer with two good quality embryos on day 2 of oocyte
Study group were categorized into three groups according to
AFC. Group 1: ≤7 (n=196), Group II: 8-15 (n=425), Group III: >15
(n= 186). For each AFC group (≤7, 8-15 and >15), entire range
of AMH values were binned at 5 percentile values using SPSS 16.
Total 20 intervals of AMH values over whole range of data were
obtained. IVF outcome (pregnancy yes and no) over various AMH
intervals was plotted as histogram. An interpolation line was
drawn to demonstrate the trend using two moving average of
Excel. Change of trend in IVF outcomes was noted by observing
the graph and corresponding cut off values for AMH were defined.
According to the cut off AMH values three subgroups i.e.
low discordant, concordant and high discordant were obtained
within each AFC group. IVF outcomes in each of three groups
were analyzed comparing three sub groups using Pearson Chi
Square test and the level of significance was calculated. The twotailed
value of P < 0.05 was considered statistically significant.
For each AFC group, AMH cut off values were different
according to their respective percentiles (Figure 2-4). Only
20.07% (162/807) women among three AFC groups were
observed showing concordance between AMH and AFC. One third
of women i.e. 33.7% (272/807) included in the study had higher
AMH values (high discordance) whereas 46.22% (373) had lower AMH values (low discordance) than expected according to
their AFC (Table 1). The three subgroups within each AFC group
were comparable for age, duration and aetiology of infertility
and indications for IVF. Within each AFC group women showing
concordance between AMH and AFC were associated with better
IVF outcomes (Table 2). Pregnancies in women in Group 1 and
Group III with both high and low discordant AMH values were
comparable (42% Vs 43.9% and 30% Vs 32.6%). Women with
AFC 8-15 but either high AMH i.e. Group II with high discordance
had lowest pregnancy rate (25%). Although two other AFC
Groups i.e.1 and III with high discordant AMH values did not
follow the same trend, pregnancy rate was 43.9% and 32.6% and
it was higher compared with their respective low AMH subgroup
women (Figure 2-3). Highest PR was observes in women of
Group 1 with concordance AMH values (59.6%). Women with
low discordant AMH values of all three AFC groups were reported
with lower (Group I, II and III; 42%, 39% and 30% respectively).
Women in Group III with high discordance (this group mainly
consisted of PCOS women) had PR 32.6%; indicating that higher
AFC with higher AMH also did not have higher conception rate
when embryos were transferred in the same cycle.
Mean M2 oocytes rate for pregnancy positive versus negative
women among three groups is shown in Figure 5 & 6. In the
women with concordant AMH values mean M2 oocytes rate for
IVF outcome positive vs. negative were in Group I, II and III were
4.33 Vs 2.98; p value.028, 8.8 vs. 6.8 p value .032 and 10.11vs
9.06; p value .048 respectively. The women of all subgroups
with positive IVF outcome demonstrated higher number of M2
rate compared to negative IVF outcome and difference was also
statistically significant. The number of M2 oocytes also showed
rising pattern corresponding to the rise with AMH values within
each AFC category.
AMH has emerged as reliable indicator and autonomous
marker of ovarian function because of its strong correlation
with AFC and operator independency. Earlier there were various
controversies regarding AMH values in relation to laboratory
assays. But now a well-established assay Gen II EUSA is
available to measure AMH; as standard techniques worldwide.
Comparable performance of AMH and AFC in IVF treatment has
been documented and may be explained to certain extent by
the fact that source of AMH is granulosa cells of antral follicles
[8,9]. Various studies have described discordance between AMH
and S.FSH in women undergoing assisted reproduction. But so
far, to the best of our knowledge, there is hardly any literature
describing discordance among AMH and AFC and their clinical
implications .Our study is first of its kind to demonstrate the
concordant AMH values according to outcome of IVF .We can
not imply the same AMH cut off values to all the women with
different AFC. Therefore it is relevant to find out the individual
cut off values for different AFC groups, thereafter to correlate the
ovarian response and IVF outcome.
Index study showed a significant proportion, amounting
to 79.93%, of women undergoing IVF treatment at the centre
were showing discordant between AMH and AFC as classified
by their percentiles according to their IVF outcome .There
was wide fluctuation of percentile values of concordance of
AHM for each AFC group, which indicate that same AHM value for different AFC group women may not generate same result.
Although there was linear correlation of number of M2 oocytes
and good quality embryos with AMH values but IVF outcome did
not follow the same trend. When IVF outcome (pregnancy yes
and no) was plotted over various AMH intervals as histogram,
we found highest pregnancy rate in all concordant groups. Low
and high AMH values in all AFC categories were associated with
poorer pregnancy rate. For better IVF outcome, there should be a
correlation between AFC and AMH. Though this is retrospective
data analysis, strength lies in the fact that discordant value of
AMH for every AFC group was described. This is in contrast to
study by Raymond Li et al.  in which the same AMH range
was generalised for the whole study group .
Current study described AMH range for each AFC category
where best IVF results were obtained for our centre. Highest
pregnancy rates were observed in all three AFC categories
showing concordance between AFC and AMH values. Low and
high AMH values in all AFC categories were associated with
poorer pregnancy rate. M2 rate increased linearly with rise of
AMH values in each category. Despite higher M2 rates reported
with higher AMH values in each AFC category, lower pregnancy
rates were reported. Women with negative pregnancy results
showed significantly low numbers of M2 oocytes when compared
with positive pregnancy within each AFC and AMH category.
We hypothesised that discordance between the two may have
adverse impact on outcome of IVF. Accumulation of more data is
needed to further validate our study.