Anti-Müllerian Hormone (AMH) is A Good
Predictor for Ongoing Pregnancy in Women Undergoing IVF/ICSI in Antagonist Cycles
The Fertility Clinic, Copenhagen University Hospital, Denmark
Submission: January 02, 2019;Published: January 10, 2019
*Corresponding author: Negjyp Sopa, The Fertility Clinic, Hvidovre Hospital, Copenhagen University Hospital, Denmark
How to cite this article: Negjyp S. Anti-Müllerian Hormone (AMH) is A Good Predictor for Ongoing Pregnancy in Women Undergoing IVF/ICSI in Antagonist Cycles. J Gynecol Women’s Health. 2019: 13(5): 555872. DOI: 10.19080/JGWH.2018.13.555872
Observation cohort study, assessed pregnancy rate based only on the AMH algorithm in 888 patients/ 2148 cycles in antagonist protocol for In vitro-fertilisering/intracytoplasmatisk spermainjektion (IVF/ICSI). Patients are divided into 3 groups based on the value of AMH. The first group had AMH <12pmol/L group two between 12-32pmol/L and group three> 32pmol/L Patients with low AMH <12pmol/L (n=291 patients/764 cycles) had maximal stimulation with corifollitropin-alfa (n=473/62%) or 300 IU/day of rFSH (n=289/38%) as well 300 IU/day of HP-HMG(n=2/0,3%). Patients with AMH 12-32 pmol/L (n=401 patients/941 cycles) had standard stimulation with 150 IU/day of rFSH (n=789/84%), or HP-HMG 150 IU/day (n=103/11%) as well corifollitropin-alfa (n=49/5%). Patients with AMH>32pmol/L(n=196 patients/ 443 cycles) had minimal stimulation with 112 IU/day of HP-hMG (n=244/55%) or rFSH 112 IU/day (n=199/45%). The ongoing pregnancy rates per started cycle in the three AMH groups first cycle were 15%, 25% and 38% respectively. The chance of achieving an ongoing pregnancy was reduced in women with low AMH whereas high AMH was associated with an increased chance of achieving an ongoing pregnancy, regardless of the patient’s age. The proportion with pregnancy in the low AMH group was significantly (p< 0.001) lower than in the high AMH group. The conclusion is that among high AMH patients, has statistically significantly high ongoing pregnancy rate. However, a randomised study with enough power with our protocol should be performed.
Anti-Mullerian hormone (AMH) is a member of the TGF-beta family and is expressed by the small preantral and early antral follicles. The AMH level reflects the size of the primordial follicle pool and the best biochemical marker of ovarian function across an array of clinical situations . In adult women, AMH levels gradually decline as the primordial follicle pool declines with age . The AMH level appears to be an early, reliable, direct indicator of declining ovarian function. In patients planning IVF, an AMH level correlate with the number of oocytes retrieved after stimulation, and is the best biomarker for predicting poor and excessive ovarian response [3,4].
Unlike the day 3FSH, AMH can be measured anytime during the menstrual cycle and typically demonstrates minimal intracycle variability since the growth of small preantral follicles that express it is continuous, not cyclical. Reliable AMH kits  are readily available from multiple vendors and AMH measurements can be obtained from large clinical reference laboratories [6,7].
Anti-Müllerian hormone (AMH) is a very important parameter for adequate ovarian stimulation for women undergoing IVF/ICSI treatment [8-10]. A multicentre observational Japanese study has shown that AMH had a significantly higher predictive value for normal and high respondents than FSH and E2 . A prospective cohort study suggested that Anti-Müllerian hormone (AMH) is strongly associated with live birth rates after IVF/ICSI . The number of oocytes in a linear increase depends on an AMH level .
We implemented the use of Anti-Müllerian hormone (AMH) as a biomarker for ovarian stimulation [14-16] since 2014. Our target was to achieve around 10 oocytes, with a range of 8-14. This was done by segmenting the patients into 3 groups given the maximum, standard or low dose of FSH stimulation according to their AMH levels.
Observation cohort study, where data were retrospectively collected, by 888 patients undergoing their 2148 cycle antagonist protocol, at the Fertility Clinic Rigshospitalet, Copenhagen
University Hospital from January 2014 to June 2017. All included
patients were stimulated according to an AMH-based FSH
Patients with low AMH <12pmol/L (n=291 patients/764
cycles were stimulated with a weight-adjusted dose of Elonva®
(n=473/62%) according to labelling: women >60kg received
150μg and women <60kg received 100μg, or 300 IU/day of rFSH
(n=289/38%) as well 300 IU/ of HP-HMG (n=2/0,3%). Patients
with AMH 12-32pmol/L (n=401 patients/941 cycles) had
standard stimulation with 150IU/day of rFSH (follitropin-alfa
or -beta) (n=789/84%), or HP-HMG 150 IU/day (n=103/11%)
as well corifollitropin-alfa (n=49/5%). Patients with AMH
>32pmol/L (n=196 patients/ 443 cycles) were stimulated
with Menopur® (HP-hMG) 112 IU/day (n=244/55%) or rFSH
(follitropin-alfa or -beta)112 IU/day (n=199/45%). The FSH
dose was fixed the first seven days of stimulation. After seven
days of stimulation, the FSH dose could be adjusted based on
sonographic assessment. The standard criteria for triggering of
ovulation were at least 3 follicles ≥17mm.
Oocyte retrieval was done 36 hours after Ovitrelle (hCG)/
Suprefact®GnRH triggering, and a transfer was done at day 2
or day 5. Luteal support was given using vaginal progesterone
Lutinus® 100mg 3 times daily for 2 weeks. Lutinus® All data
were prospectively collected and recorded in an SSPS version 22
and STATA 13.1. database. The study was approved by the Danish
Patient Safety Authority (Number 3-3013-2242/1).
Baseline characteristics of the 888 patients treated in
antagonist protocol in 2148 cycle were stimulated according to an
AMH-based FSH dosing algorithm Table 1. As seen, 291 (32.8%)
were in the low AMH group, 401 (45.1%) in the intermediate and
196 (22.1%) in the high AMH group. Age, AMH and AFC were
significantly different between AMH-groups (p<0.001), whereas
weight and BMI are not significant.
Data are described with means and standard deviation, Kruskal-Wallis test and median test.
P-value < 0.05 was considered statistically significant. Statistics, comment in text.
1. Data are described with means and standard deviation, Kruskal-Wallis test and Chi-square test.
2. Statistics, comment in text.
Table 2 shows first cycle response parameters and clinical
outcome in the three AMH groups. The ongoing pregnancy rates
per started cycle in the three AMH groups were 15%, 25% and
38%, respectively (p<0.001) and per transfer 21%, 30% and
48% respectively (p<0.001). The chance of achieving an ongoing
pregnancy was reduced in women with low AMH, whereas high
AMH was associated with an increased chance of achieving an
1. Data are described with means and standard deviation, Chi-square test, Kruskal-Wallis test one-way ANOVA
2. Statistics, comment in text.
Table 3 shows response parameters and clinical outcome for
all cycles 2148. The ongoing pregnancy per started cycle in the
AMH groups were 14%, 22% and 32% respectively (p<0.001)
and per transfer 20%, 26% and 38 respectively (<0.001). In the
first cycle, the pregnancy rate is high compared to all cycles. The
pregnancy rate is low after the first cycle, but if we are based on
the AMH group, there is still significant pregnancy rate between
AMH groups. The pregnancy rate is high in the group with high
AMH and low in the group with low AMH.
1. P-value < 0.05 was considered statistically significant.
2. Statistics, comment in text.
Table 4 shows the pregnancy rate the relationship to the
first cycle and the second or more cycle based on AMH and age.
In the first cycle, age did not affect pregnancy relative to AMH
groups. The pregnancy rate was significantly high in groups
with high AMH, while in patients>35-years-old pregnancy rate
was generally low, but still significantly high in groups with
high AMH. In the second or more cycle of patients ≤35-yearsold,
age had no influence, and patients with high AMH had a high
pregnancy rate. Patients >35-years-old in second or more cycle
had no significant difference in pregnancy rate between AMH
The present study provides results from 2148 ART cycles
where a stimulation based on the AMH level FSH dosing
algorithm. The main findings of the present study were patients
with AMH levels above 32pmol/L and thus a predicted high rate
of pregnancy. Several studies based on the number of oocytes
that predict high pregnancy rates [16-19]. We know that patients
with high AMH can develop more follicles, and as a result, many
oocytes come with aspiration. It has always been difficult to keep
the balance between having a high pregnancy and minimizing
OHSS. Patients with anovulation have high AMH and tend to
have many follicles on stimulation, but they may also have a low
response if stimulated with an inadequate dose of FSH [20-22].
In our study we could show that patients with high AMH, even
some of them, do not have many oocytes nevertheless, had better
pregnancy rate in the first cycle regardless of age. In patients
>35-year-old, the rate of pregnancy decreases but is still present,
for patients with high AMH has a high pregnancy rate compared
to patients with low AMH.
Patients who had completed number two or more cycles
pregnancy is still high in the group of patients with high AMH
aged <35 years. In patients aged>35 years, with two or more
cycles we could not find different inter-group pregnancy rates,
regardless of AMH level. AMH is a very good predictor of
pregnancy prognosis. Patients with high AMH are at risk group
for OHSS therefore we choose AMH algorithm to minimize
OHSS and maintain good pregnancy rate. We have mildly
stimulated patients with AMH> 32 where some of the patients
have responded with very few follicles but the pregnancy rate
was satisfactory and none of them had OHSS [23-25]. Several
studies show that a number of oocytes 8-14 will provide better
pregnancy rate [16-18]. We agree that patients with high AMH
and young age can reach in most cases the number of oocytes we
want. Our study confirms that AMH is a really good prediction for
oocytes retrieval and pregnancy rate, especially in patients <35
years-old , whatever there is a first, second or more cycle.
Low AMH patients do not have a good pregnancy prognosis,
the number of oocytes retrieval is not high and the pregnancy
rate is low [26-30]. Our study shows that even though age
was <35 years, the patients with low AMH had significant low
pregnancy compared to patients with high AMH. AMH level
reflects pregnancy chances in all groups, we can see differences
in pregnancy rate in all AMH groups. Group of patients with
AMH <12pmol/L had poor pregnancy and group of patients with
AMH> 32pmol/L had the best pregnancy rate. Patients with AMH
12-32pmol/L normal responder stimulated with standard dose
FSH, the pregnancy rate was better than the group of patients
with low AMH, but low compared to the group of patients with
high AMH> 32pmol/L.
The main finding suggest that an AMH is a good predictor
for pregnancy prognosis in patients at the least <35 years old.
Patients with high AMH have a significantly better pregnancy
rate compared to patients with low AMH. The AMH algorithm
can be used to select the starting dose of FSH to minimize OHSS
and retains good pregnancy rate. However, a randomised study
with enough power with our protocol should be performed