Abstract Study question Can serum AMH levels affect pregnancy outcomes particularly live birth rate (LBR) and cumulative live birth rate (CLBR) in PCOS women receiving IVF/ICSI treatment? Summary answer Higher serum AMH indicated a decreased LBR and increased miscarriage rate in PCOS women undergoing IVF/ICSI cycle. However, AMH has no association with CLBR. What is known already AMH could reduce follicle sensitivity to FSH and acts as an inhibitor of primordial follicle recruitment, which is essential for follicle growth. Therefore, AMH has widely been suggested to reflect ovarian reserve and predict ovarian response to assisted reproductive technology (ART). Women with PCOS presented AMH level around 2 to 4-fold higher than normal female. The pathological mechanisms of PCOS are complex. Ovarian follicular arrest is regarded as an important pathophysiology of PCOS, which is a consequence of insufficient FSH secretion or inhibition FSH action. Consequently, increased AMH arise might be associated with severity and poor pregnancy outcomes of PCOS. Study design, size, duration This was a retrospective cohort study included 4719 PCOS patients undergoing IVF/ICSI treatment at the Reproductive Center of Peking University Third Hospital from February 2017 to January 2022. Participants/materials, setting, methods Participants were divided into three groups according to the cutoffs defined by the 25th and 75th percentiles of serum baseline AMH level, low-AMH group, n = 1198 (AMH ≤ 4.98 ng/mL); average-AMH group, n = 2346 (4.98 < AMH < 10.65 ng/mL); high-AMH group, n = 1175 (AMH ≥ 10.65 ng/mL). Baseline parameters and pregnancy outcomes including clinical pregnancy rate (CPR), miscarriage rate, LBR and CLBR were compared among three groups. Main results and the role of chance In our cohort, we observed a significant progressively increase in the number of oocytes retrieved and the number of good-quality embryos per cycle from low- to average- to high-AMH group. No statistical significance was observed in fertilization rate (62% vs 60% vs 60%, P = 0.453). In addition, there was no significant difference in CPR among three groups (44.8% vs 42.8% vs 36.8%, P = 0.069). The miscarriage rate was progressively increased from low, average, and high AMH group (11.2% vs 18.5% vs 20.0%, P = 0.023). Importantly, LBR in fresh transfer was progressively decreased from low, average, and high-AMH group (39.6% vs 33.9% vs 29.0%, P = 0.004). However, no statistical difference but an increasing trend was observed in the CLBR among three groups (55.2% vs 56.9% vs 58.1%, P = 0.753). After adjusting by logistic regression analysis, the results indicated that elevated AMH was still strongly associated with decreased LBR in fresh transfer. In addition, BMI, endometrial thickness, and fertilization rate also have correlations with LBR. Nevertheless, the positive relationship between AMH and miscarriage rate remained only in low and average AMH group. Finally, no significant association was shown between AMH and CLBR in logistic regression analysis. Limitations, reasons for caution Firstly, this was a single-center study and might result in bias. Another limitation was due to its retrospective and non-randomized design. Besides, our study indicated an increasing trend in CLBR without statistical significance. Therefore, further large-scale, high-quality prospective cohort studies are still needed to confirm these findings. Wider implications of the findings These findings bring into consideration the potential role of AMH as the guidance in clinical regime. Since high AMH level might act as a biomarker in predicting worse pregnancy outcomes of PCOS women with IVF/ICSI fresh cycle. Trial registration number not applicable
Read full abstract