Association of Anti-Müllerian Hormone on Oocyte Maturation, Fertilization, and Pregnancy Rates in Patients under Assisted Reproductive Technology Cycles: A Cross-Sectional Study.

Infertility and the inability of a person to perform the process of reproduction to have children are naturally considered painful experiences of life. The psychological and social impacts can increase its importance, and become a crisis for the affected person ( 1 ). Infertility is defined as the inability of a couple to conceive after one year of unprotected intercourse. About 85-90% of healthy young couples become pregnant within one year and often within six months of the attempt to conceive. Thus, infertility occurs in approximately 10-15% of couples ( 2 ). The cause of infertility can be related to the female (about 40%), male (about 40%) or both (about 10%), and in about 10% of couples, the cause of infertility is unknown ( 3 ). Clinical knowledge and technological advances have the greatest impact on infertility in recent years. These include the introduction of in vitro fertilization (IVF) and other assisted reproductive technologies (ART). One of the most important factors in the success of ART is the number of oocytes produced by the ovaries following hormonal stimulation ( 4 ). One of the most important limiting factors in the success of ART is poor ovarian response, which is seen in 10 to 15% of women who undergo IVF ( 5 ). Therefore, it is necessary to check the ovarian reserve before performing ART ( 6 ). Today, with the advancement of reproductive medicine, a large part of research has focused on ovarian reserve. These are generally aimed at: i. Improving the safety of ovarian stimulation methods by identifying patients with high responsive and are at higher risk of ovarian hyperstimulation syndrome (OHSS), ii. Improving the efficiency of ovarian stimulation methods by adjusting the stimulation dose, and iii. Using ovarian reserve as a tool to predict the outcome of IVF treatment and identify young women who have low ovarian reserve identify young women who have low ovarian reserve (similar to menopausal status) ( 7 ). Anti-müllerian hormone (AMH) is one of the hormones that has recently been considered as a marker for predicting ovarian response before ART ( 8 , 9 ). AMH is a chain dose glycoprotein from the family of cell growth and differentiation factors, which is secreted from the testicular Sertoli cells in the foetus during pregnancy, and prevents the transformation of molars into the uterus and other molars. Serum AMH level is related to age and gender. AMH levels decrease in men after they attain sexual maturity, whereas, in women, AMH cannot be measured before puberty. However, during puberty due to its secretion from the granulosa cells of the growing ovarian follicles, serum AMH levels increase and are followed by a decrease during reproductive years. After menopause, due to the depletion of the growing follicles, the AMH serum level is very low ( 10 ). After puberty, AMH values are relatively constant throughout the month and are not associated with menstruation. Therefore, AMH can be measured on any day of the menstrual cycle. AMH levels decrease with increasing age, and this is associated with decreased ovarian function. The level of this hormone was found to be low when IVF programs did not respond well to ovarian stimulant drugs ( 11 ). In addition, serum AMH levels were strongly correlated with the number of follicles before treatment and the number of oocytes recovered during ovarian stimulation ( 9 ). Recent studies suggest that AMH levels are an indicator of the ART success rate ( 12 - 14 ). However, other studies did not find it as a predictor of ART ( 9 , 15 , 16 ). There are few reports of the clinical importance of AMH levels measured within the late follicular phase amid ovarian stimulation ( 17 - 19 ). The aim of this study is to evaluate the relationship between AMH levels and cycle outcomes in ART methods in infertile patients by using a linear regression method.

In study, serum AMH levels were measured from the records of 1000 infertile couples to examine its relationship with the number of unknown mature eggs, the number of fertilised oocytes, and the percentage of mature eggs. Table 1 shows the demographic, laboratory, and clinical variables. AMH levels, gender, age, number of COC, M2 oocytes and 2PN as well as endometrial thickness, cause of infertility, protocol used, embryo quality and number of embryos transferred had a statistically significant association with ultrasound result (P<0.01). The type of ART was marginally significant (P=0.071) and the maturation rate was not significant (P=0.786). In Table 2, we tested the mean by ANOVA (not shown) and median via the non-parametric, Kruskal-Wallis’s test. Although we applied both tests to our data, the reported P value is based on the non-parametric test. The median of M2, 2PN and maturation rate of the oocytes were statistically significant in terms of serum AMH levels (P<0.001). The results indicated that increases in M2, 2PN and oocyte maturation rate corresponded to an increase in serum AMH levels. Baseline characteristic of the study population a ; Values are given as mean ± standard deviation (SD) and compared for both groups using the t test, b ; Median (1 st Qu., 3 rd Qu.) and compared for both groups using the Wilcoxon test, c ; Number/denominator (percentage) and compared for both groups using the chi-square test, AMH; Anti-müllerian hormone, IVF; In vitro fertilization, ICSI; Intracytoplasmic sperm injection, COC; Cumulus-oocyte complex, ART; Assisted reproductive technologies, M2; Metaphase II, PCOS; Polycystic ovary syndrome, and 2PN; Two pronuclei. The results of the linear regression of log (AMH) of each predictor (crude), as well as an adjusted model, are shown in Table 3. For each unit log (total dose gonadotropin), the AMH growth rate decreased by 30% [rate ratio=0.70; 95% confidence interval (CI)=0.59, 82]. AMH levels increased by 3% for each unit increase in 2PN (rate ratio=1.03; 95% CI=1.01, 1.05) and a 23% increase for each unit increase in log (oestradiol) (rate ratio=1.23; 95% CI=1.16, 1.30). However, AME decreased by 2% for total embryos (rate=0.98; 95% CI=0.96, 0.99), and increased by 3% for the duration of stimulation (rate ratio=1.03; 95% CI=1.001, 1.05). In comparison to zero embryos transferred, the AMH level increased by 19% of the basal level with one (rate ratio=1.19; 95% CI=1.03, 1.38), 18% with two (rate ratio=1.18; 95% CI=1.03, 1.35), and 23% with three transferred embryos (rate ratio=1.23; 95% CI=1.05, 1.44). Assessment of the infertility protocol indicated that the AMH growth rate for the long agonist protocol was 2.96 times higher than the microdose protocol (rate ratio=2.96; 95% CI=2.47, 3.56). The AMH growth rate for the antagonist protocol was 3.05 times higher than the microdose protocol (rate ratio=3.05; 95% CI=2.57, 3.61). Overall, infertility significantly decreased the AMH grow rate for male factor by 38% (rate ratio=0.62; 95% CI=0.54,0.72), endometriosis by 30% (rate ratio=0.70; 95% CI=0.55,0.89), unknown infertility cause by 65% (rate ratio=0.65; 95% CI=0.54,0.78), and ovarian factor by 28% (rate ratio=0.72; 95% CI=0.59,0.87) in comparison with polycystic ovary syndrome (PCOS). Average of M2 oocytes, 2PN and oocyte maturation rate in different groups based on serum AMH levels 1 ; ANOVA (Kruskal-Wallis) test of median, * ; Median (1 st Qu., 3 rd Qu.), AMH; Anti-Müllerian hormone, M2; Metaphase II, and 2PN; Two pronuclei. Regression models with log(AMH) as response variables a ; Adjusted for all of variables, CI; Confidence interval, ART; Assisted reproductive technology, AMH; Anti-müllerian hormone, IVF; In vitro fertilization, ICSI; Intracytoplasmic sperm injection, COC; Cumulus-oocyte complex, M2; Metaphase II, and * ; P<0.05 is considered statistically significant.

We used a multiple linear regression method in this study to evaluate the relationship between serum AMH levels and cycle outcomes in ART methods in infertile patients. We found a statistically significant association between serum AMH levels and mean M2, 2PN, and oocyte maturation rate. Our findings agreed with previous studies that examined the ability of AMH and the antral follicle count to predict the number of oocytes retrieved ( 8 ) The results showed that with increasing serum AMH levels, the M2, 2PN, and oocyte maturation rates also increased. The lowest and highest oocyte counts were observed in individuals with serum AMH levels less than 1 and more than 3.5, respectively. Consistent with our study, Wu et al. ( 20 ) conducted a prospective study and observed a positive correlation between the number of oocytes obtained and AMH serum levels on the third day. Other studies also found that patients with low AMH levels had lower oocyte counts ( 12 , 21 , 22 ) We also evaluated serum AMH levels on the success of ART treatment in infertile patients, in addition to female age, the total dose of gonadotropin, COC, M2, 2PN, oestradiol, the total number of embryos, duration of stimulation, and type of ART. Even though a known link exists between higher oocyte yield and reduced miscarriage rate ( 23 ) and a higher live birth rate ( 24 ), the evidence on the link between AMH and qualitative ART outcomes is mixed ( 25 ). According to a systematic review and meta-analysis of the evidence on the predictive potential of AMH for implantation and clinical pregnancy in women who undergo ART, AMH has some association with implantation and clinical pregnancy, although its predictive ability is limited ( 26 ). This pattern is also supported by the findings of another meta-analysis ( 27 ). When comparing women with low expected ovarian reserve to women with unknown ovarian reserve, the authors found that women with low expected ovarian reserve had a greater, albeit still slight, predictive accuracy (diagnostic OR of 4.63 vs. 2.48, respectively). This remarkable finding would need to be confirmed in larger investigations. A favourable correlation between AMH and clinical pregnancy was reported by Wang et al. ( 28 ), Hazout et al. ( 29 ), Kwee et al. ( 30 ), and Wunder et al. ( 13 ). Other studies, probably due to their small sample size and subsequent low power, did not find a link between AMH and clinical pregnancy ( 18 , 31 , 32 ). AMH appears to be a poor predictor of qualitative ART outcomes such as implantation, pregnancy, and live birth when viewed collectively. This indicates that factors other than ovarian reserve (as measured by AMH) are likely to influence pregnancy probabilities. Endometrial receptivity, sperm/egg genetics, stimulation protocol, and transfer method are all possible factors ( 33 ). Several studies have found a positive relationship between serum AMH and oocyte quality ( 11 , 32 , 34 - 37 ), whereas others did not ( 15 , 23 , 38 , 39 ). This study has some limitations such as lack of additional information about life birth rate information. The sample size was small. We are used one Infertility Center. It is recommended to use several treatment Infertility Centers.

AMH is an excellent marker of ovarian reserve because it is strongly associated with follicle count and exhibits low cyclical variability and decline during reproductive life. It has a relatively stable expression throughout the menstrual cycle and is an attractive determinant of ovarian activity. In addition, serum-based AMH may provide more prognostic value for clinical pregnancy than other markers currently available for ART. This suggests that the new marker, AMH, may better reflect ovarian function than normal hormones. The cycle stability and predictive power of AMH make AMH the most characteristic hormonal prognostic marker of ovarian response in ART.

In this cross-sectional study, we studied data obtained from 1000 infertile couples who referred for IVF or intracytoplasmic sperm injection (ICSI) at the Research and Clinical Centre of Yazd Infertility Clinic, Yazd, Iran from February 2016 to April 2016. The study was conducted according to the current version of the Declaration of Helsinki and approved by the Ethics Committee of Yazd University of Yazd (IR.YAZD.REC.1401.019). Infertile women with at least one year infertility who underwent IVF or ICSI and did use drugs that interfered with their AMH levels were included. Women who were married for less than 1.5 years were excluded. Demographic data and data that pertained to the cause of underlying infertility were obtained and serum AMH levels, number and quality of oocytes, percentage of mature oocytes, number of fertilised oocytes, and endometrial thickness were measured. The formed embryos were evaluated quantitatively and qualitatively. The number of cumulus-oocyte complexes (COC), two pronuclei oocytes (2PN), and metaphase II (M2) oocytes were also evaluated. The type of ART (IVF or ICSI) and the protocols were documented. The outcome was AMH with three main predictors: M2, 2PN, and oocyte maturation rate. Written informed consent was obtained from all probands before inclusion, and their personal data remained confidential. Patients were classified into three groups according to their AMH levels (less than 1, between 1 and 3.5, and more than 3.5 ng/ml). The formed embryos were classified according to their quality into groups: A (equal blastomeres, without fragments), B (unequal blastomeres, without fragments), C (unequal blastomeres, fragments less than 10%), and D (unequal blastomeres, fragments approximately 10%). The data for the AMH variables, the total dose of gonadotropin and oestradiol were skewed to the right; therefore, the logarithm of the variables provided a better fit for the statistical assumptions. For this reason, our response variable was log (AMH). In order to find the effective variables in the linear regression model and remove the unimportant variables, we used Akaike’s Information Criterion to fit the best model. After the application of the stepwise method, the independent variables that affected the response variable were processed in the final model. The quantitative data are presented as mean ± standard deviation (SD) and the qualitative data are shown as median (interquartile range) or number (percentage). The t test and ANOVA were used for normal data and nonparametric (e.g., Wilcoxon) for non-normal data. The Kolmogorov-Smirnov test for normality was applied to all continuous outcomes. The proportion data were analysed for the qualitative data using Pearson’s chi-squared tests. Multiple linear regression was used on logged AMH serum levels and some appropriate predictors. In addition, we used Kruskal-Wallis’s test with the AMH category (AMH ≤1, 13.5) was considered as the primary explanatory variable on the three different outcomes of M2, 2PN, and oocyte maturation rate. All analyses were performed using R version 3.6.2 (R Core Team 2019).