High anti-Müllerian hormone level as a predictor of poor pregnancy outcomes in women with polycystic ovary syndrome undergoing in vitro fertilization/intracytoplasmic sperm injection: a retrospective cohort study.

We recruited 5,844 women with infertility and PCOS. We screened 4,719 women with PCOS for the final analysis (Fig.  1 ). We divided all participants into three groups using the 25th and 75th percentile cutoffs for serum AMH concentrations. In the low AMH group, 1,198 women with PCOS had AMH levels of ≤ 4.98 ng/mL. In the average AMH group, 2,346 women with PCOS had AMH levels between 4.98 and 10.65 ng/mL. In the high AMH group, 1,175 women with PCOS had AMH levels of ≥ 10.65 ng/mL. Fig. 1 Flow chart of the study cohort population. AMH, anti-Müllerian hormone; ICSI, intracytoplasmic sperm injection; IVF, in vitro fertilization; PCOS, polycystic ovary syndrome Flow chart of the study cohort population. AMH, anti-Müllerian hormone; ICSI, intracytoplasmic sperm injection; IVF, in vitro fertilization; PCOS, polycystic ovary syndrome We compared the baseline features of the three groups (Table  1 ). Age and body mass index (BMI) significantly decreased in women from the low to the high AMH groups. The baseline FSH level decreased, whereas the baseline luteinizing hormone level increased from the low to the high AMH group. The maximal estradiol (E2) concentration was 42,043 pmol/L, 58,601 pmol/L, and 68,250 pmol/L in the low, average, and high AMH group. A significant increase was observed in the number of antral follicles from the low to the high AMH group. The initial therapeutic dose, gonadotropin duration, and gonadotropin dose significantly decreased from the low to the high AMH group, indicating a positive correlation between AMH and ovarian response to COS in women with PCOS. In addition, the endometrium on the day of hCG administration in the high AMH group was slightly thinner than that in the low and average AMH groups. No significant difference was observed in infertility factors, infertility type, or infertility years among the groups. Table 1 Baseline features of three groups Characteristics Serum AMH, ng/ml ≤ 4.98 ( n  = 1198) 4.98–10.65 ( n  = 2346) ≥ 10.65 ( n  = 1175) P -value Maternal age (years) 31.1 ± 3.7 b,c 30.7 ± 3.6 d 30.4 ± 3.5 < 0.001 BMI (kg/m 2 ) 24.23 ± 3.11 a,b 23.60 ± 3.08 d 23.25 ± 3.10 < 0.001 Baseline FSH (IU/L) 5.92 ± 1.46 a,b 5.75 ± 1.41 b 5.58 ± 1.39 < 0.001 Baseline LH (IU/L) 4.28 (3.07,6.26) a,b 5.08 (3.49–7.66) b 6.35 (4.34–9.17) < 0.001 Baseline E2 (pmol/L) 152.0 (117.0–189.0) b 155.0 (120.0–195.0) b 165.0 (129.0–204.0) < 0.001 LH/FSH ratio 0.76 (0.55–1.08) a,b 0.90 (0.65–1.30) b 1.16 (0.83–1.64) < 0.001 Infertility factors (%) 0.133  Uterine and tubal factors 29.2 (353/1198) 28.8 (683/2346) 25.7 (308/1175)  Ovulatory dysfunction 21.2 (255/1198) 22.9 (534/2346) 25.3 (293/1175)  Unexplained reasons 17.5 (202/1198) 18.1 (407/2346) 20.1 (225/1175)  Male factors 32.1 (388/1198) 30.2 (722/2346) 28.9 (349/1175) Infertility type (%) 0.731  Primary 69.1 (828/1198) 70.2 (1648/2346) 69.3 (814/1175)  Secondary 30.9 (370/1198) 29.8 (698/2346) 30.7 (361/1175)  Infertility years (years) 3 (2–5) 3 (2–5) 3 (2–5) 0.206  Follicle number 16.5 ± 5.2 a.b 19.0 ± 5.7 b 22.4 ± 6.4 < 0.001 Fertilization methods (%) 0.001  IVF 63.7 (763/1198) b 60.0 (1502/2346) b 70.1 (824/1175)  ICSI 36.3 (435/1198) b 36.0 (844/2346) b 29.9 (351/1175) Initial therapeutic dose 150.0 (137.5,187.5) a,b 150 (134.4,150) b 150 (112.5,150) < 0.001 Gonadotropin duration (d) 10.7 ± 2.2 a,b 10.6 ± 2.2 b 10.4 ± 2.5 0.005 Gonadotropin dose (IU) 1800.0 (1425.0–2400.0) a,b 1625.0 (1312.0–2175.0) b 1425.0 (1162.5–1875.0) < 0.001 Endometrial thickness on hCG day, (mm) 10.5 ± 1.6 b 10.4 ± 1.6 b 10.2 ± 1.7 < 0.001 Ovulation stimulating regimen (%) < 0.001  Antagonist protocol 85.5 (1024/1198) b 87.1 (2044/2346) b 91.6 (1074/1175)  Agonist protocol 14.5 (174/1198) 12.9 (302/2346) 8.4 (101/1175) AMH anti-Müllerian hormone, BMI body mass index, FSH follicle-stimulating hormone, LH luteinizing hormone, E2 estradiol, IVF in vitro fertilization, ICSI Intracytoplasmic sperm injection a: P  < 0.001, significant difference compared with the average AMH group b: P  < 0.001, significant difference compared with the high AMH group c: P  < 0.01, significant difference compared with the average AMH group d: P  < 0.01, significant difference compared with the high AMH group Baseline features of three groups 1800.0 (1425.0–2400.0) a,b 1625.0 (1312.0–2175.0) b 1425.0 (1162.5–1875.0) AMH anti-Müllerian hormone, BMI body mass index, FSH follicle-stimulating hormone, LH luteinizing hormone, E2 estradiol, IVF in vitro fertilization, ICSI Intracytoplasmic sperm injection a: P  < 0.001, significant difference compared with the average AMH group b: P  < 0.001, significant difference compared with the high AMH group c: P  < 0.01, significant difference compared with the average AMH group d: P  < 0.01, significant difference compared with the high AMH group We compared the characteristics and pregnancy outcomes among the three groups (Table  2 ). The ovarian stimulation syndrome (OHSS) rate was 0.9%, 0.8%, and 2.0% in the low, average, and high AMH group ( P  = 0.004). The number of oocytes retrieved and good-quality embryos per cycle significantly increased from the low to the high AMH group. No significant difference was observed in the fertilization rate, number, and embryo type transferred in fresh cycles among the groups. The fresh ET cycle rate significantly decreased from the low to the high AMH group (52.4%, 39.0%, and 27.1%, respectively; P  < 0.001). Table 2 Cycle characteristics and pregnancy outcomes of the three groups Characteristics Serum AMH, ng/ml ≤ 4.98 ( n  = 1198) 4.98–10.65 ( n  = 2346) ≥ 10.65 ( n  = 1175) P -value OHSS rates (%) 0.9 (11/1198) d 0.8 (19/2346) c 2.0 (24/1175) 0.004 Number of oocytes retrieved 14 (9–19) a,b 17 (11–23) b 20 (13–27) < 0.001 Fertilization rate 0.62 (0.45–0.75) 0.60 (0.44–0.74) 0.60 (0.46–0.73) 0.453 Number of good-quality embryos per cycle 6 (3–9) a,b 7 (3–10) b 8 (4–12) < 0.001 Number of embryos transferred in fresh ET (%) 0.642  1 10.8 (68/628) 12.0 (110/916) 10.3 (33/319)  2 89.2 (560/628) 88.0 (806/916) 89.7 (286/319) Embryo type transferred in fresh ET (%) 0.243  Cleavage embryo 96.0 (603/628) 94.1 (862/916) 94.9 (303/319)  Blastocyst 4.0 (25/628) 5.9 (54/916) 5.1 (16/319) Fresh ET cycle rate (%) 52.4 (628/1198) a,b 39.0 (916/2346) b 27.1 (319/1175) < 0.001 Clinical pregnancy rate (%) 44.9 (282/628) 43.3 (397/916) 38.9 (124/319) 0.204 Twin/multiple pregnancy rates (%) 17.0 (95/560) 16.5 (133/806) 13.6 (39/286) 0.432 Miscarriage rate (%) 11.3 (32/282) c,d 17.1 (68/397) 21.8 (27/124) 0.018 Live birth rate (%) 39.8 (250/628) c 35.9 (329/916) 30.4 (97/319) 0.017 Cumulative live birth rate (%) 55.2 (532/964) 57.1 (1002/1755) 58.1 (511/879) 0.423 ET embryo transfer, AMH anti-Müllerian hormone, OHSS ovarian hyperstimulation syndrome a: P  < 0.001, significant difference compared with the average AMH group b: P  < 0.001, significant difference compared with the high AMH group c: P  < 0.01, significant difference compared with the high AMH group d: P  < 0.05, significant difference compared with the average AMH group Cycle characteristics and pregnancy outcomes of the three groups ET embryo transfer, AMH anti-Müllerian hormone, OHSS ovarian hyperstimulation syndrome a: P  < 0.001, significant difference compared with the average AMH group b: P  < 0.001, significant difference compared with the high AMH group c: P  < 0.01, significant difference compared with the high AMH group d: P  < 0.05, significant difference compared with the average AMH group Importantly, the LBR of fresh ET was 39.8%, 35.9%, and 30.4% in the low, average, and high AMH groups, respectively. A significant intergroup difference was observed in LBR ( P  = 0.017). Post hoc tests indicated that the LBR in the high AMH group was lower than that in the low AMH group. The clinical pregnancy rate (CPR) for fresh ET was 44.9%, 43.3%, and 38.9% in the low, average, and high AMH groups, respectively. No significant difference was observed in the CPR of fresh ET among the groups ( P  = 0.204). The twin/multiple pregnancy rate for fresh ET was 17.0%, 16.5% and 13.6% in the low, average, and high AMH groups. There was no difference in the twin/multiple pregnancy rate for fresh ET among three groups ( P  = 0.432). The miscarriage rates of fresh ET were 11.3%, 17.1%, and 21.8% in the low, average, and high AMH groups, respectively. A significant intergroup difference was observed in miscarriage rate ( P  = 0.018). Post hoc tests indicated that the miscarriage rate in the average and the high AMH groups was higher than that in the low AMH group. CLBR was 55.2%, 57.1%, and 58.1% in the low, average, and high AMH groups, respectively. No significant difference was observed in CLBR among the groups ( P  = 0.423). Subsequently, we conducted multivariable logistic regression analysis to analyze the correlation between AMH levels and pregnancy outcomes of fresh ET in women with PCOS. Table 3 presents the study results. Notably, the LBR for fresh ET was decreased in the high AMH group compared with that in the low AMH group (OR: 0.629, 95% CI: 0.460–0.860). The miscarriage rate for fresh ET was increased in the average and high AMH groups compared with that in the low AMH group (average vs. low: OR: 1.592, 95% CI: 1.017–2.490; high vs. low: OR: 2.045, 95% CI: 1.152–3.633). Finally, multivariable logistic regression analysis revealed no significant association between AMH levels and CLBR (Supplementary Table 1). Table 3 Logistic regression analysis of factors that correlated with pregnancy outcomes of fresh ET Clinical pregnancy rate Miscarriage rate Live birth rate OR (95%CI) P -value OR (95%CI) P -value OR (95%CI) P -value Low AMH Ref Ref Ref Average AMH 0.929 (0.749–1.152) 0.501 1.592 (1.017–2.490) 0.042 0.824 (0.661–1.027) 0.085 High AMH 0.784 (0.581–1.058) 0.111 2.045 (1.152–3.633) 0.015 0.629 (0.460–0.860) 0.004 Maternal age (years) 1.015 (0.987–1.044) 0.288 1.040 (0.986–1.098) 0.149 1.006 (0.978–1.035) 0.687 BMI (kg/m 2 ) 0.959 (0.927–0.992) 0.017 1.058 (0.990–1.130) 0.099 0.943 (0.910–0.976) 0.001 Infertility years (years) 0.970 (0.934–1.009) 0.127 0.997 (0.928–1.070) 0.929 0.965 (0.928–1.005) 0.086 FSH 1.000 (0.936–1.069) 0.992 1.018 (0.895–1.158) 0.782 0.995 (0.930–1.066) 0.894 Follicle number 1.007 (0.990–1.024) 0.419 0.995 (0.963–1.027) 0.754 1.009 (0.992–1.027) 0.289 Ovulation stimulating regimen 0.888 (0.670–1.177) 0.409 0.798 (0.453–1.406) 0.435 0.958 (0.718–1.278) 0.769 Gonadotropin dose 1.000 (1.000–1.000) 0.122 1.000 (1.000–1.000) 0.768 1.000 (1.000–1.000) 0.183 Endometrial thickness (mm) 1.280 (1.205–1.360) 0.000 1.111 (0.992–1.243) 0.068 1.263 (1.187–1.344) 0.000 Fertilization methods (%) 1.060 (0.850–1.321) 0.606 1.047 (0.686–1.599) 0.830 1.055 (0.841–1.324) 0.642 Number of oocytes retrieved 1.003 (0.980–1.026) 0.815 0.991 (0.948–1.036) 0.680 1.005 (0.982–1.028) 0.682 Fertilization rate 1.356 (0.842–2.186) 0.210 0.617 (0.253–1.507) 0.290 1.557 (0.950–2.550) 0.079 Embryo type transferred in fresh ET (%) 1.147 (0.747–1.761) 0.530 0.932 (0.396–2.195) 0.873 1.180 (0.761–1.830) 0.459 OR Odds ratio, CI Confidence interval, AMH Anti-Müllerian hormone, BMI Body mass index, FSH Follicle-stimulating hormone, ET Embryo transfer Logistic regression analysis of factors that correlated with pregnancy outcomes of fresh ET 0.929 (0.749–1.152) 1.592 (1.017–2.490) 0.824 (0.661–1.027) 0.784 (0.581–1.058) 2.045 (1.152–3.633) 0.629 (0.460–0.860) 1.015 (0.987–1.044) 1.040 (0.986–1.098) 1.006 (0.978–1.035) 0.959 (0.927–0.992) 1.058 (0.990–1.130) 0.943 (0.910–0.976) 0.970 (0.934–1.009) 0.997 (0.928–1.070) 0.965 (0.928–1.005) 1.000 (0.936–1.069) 1.018 (0.895–1.158) 0.995 (0.930–1.066) 1.007 (0.990–1.024) 0.995 (0.963–1.027) 1.009 (0.992–1.027) 0.888 (0.670–1.177) 0.798 (0.453–1.406) 0.958 (0.718–1.278) 1.000 (1.000–1.000) 1.000 (1.000–1.000) 1.000 (1.000–1.000) 1.280 (1.205–1.360) 1.111 (0.992–1.243) 1.263 (1.187–1.344) 1.060 (0.850–1.321) 1.047 (0.686–1.599) 1.055 (0.841–1.324) 1.003 (0.980–1.026) 0.991 (0.948–1.036) 1.005 (0.982–1.028) 1.356 (0.842–2.186) 0.617 (0.253–1.507) 1.557 (0.950–2.550) 1.147 (0.747–1.761) 0.932 (0.396–2.195) 1.180 (0.761–1.830) OR Odds ratio, CI Confidence interval, AMH Anti-Müllerian hormone, BMI Body mass index, FSH Follicle-stimulating hormone, ET Embryo transfer

This retrospective cohort study included 5,844 women with PCOS who underwent IVF/ICSI at the Reproductive Center of Peking University Third Hospital between February 2017 and June 2023. The inclusion criteria were as follows: (i) women aged 20–40 years, (ii) male or tubal factor infertility, and (iii) complete clinical data. The exclusion criteria were as follows: (i) recurrent miscarriages or abnormal parental karyotyping; (ii) history of malignancies; (iii) endocrinological diseases, including endometriosis, hypothalamic tumors, and hyperprolactinemia; (iv) history of ovarian surgery; and (v) missing data regarding baseline characteristics. We included 4,719 women in the final analysis. This study was approved by the Medical Science Research Ethics Committee of Peking University Third Hospital (Reference Number M3032464). All participants provided written informed consent. A standardized controlled ovarian stimulation (COS) regimen was administered by clinicians. All patients were administered gonadotropin-releasing hormone antagonists or agonists. Ultrasonography indicated that ≥ 2 leading follicles had reached 18 mm in diameter. Clinicians administered recombinant human chorionic gonadotropin (HCG) (250 µg; Eiser, Serono, Germany), triptorelin (0.2 mg; Diphereline, Ipsen Pharma Biotech, France) combined with HCG, or triptorelin only to induce oocyte maturation on trigger day. Oocytes were retrieved 34–36 h after administration, followed by insemination via IVF or ICSI. Embryo cleavage or blastocyst transfer was performed on day 3 or 5 after oocyte retrieval. The criteria for converting to freeze-all embryos was performed as follows: The serum E2 levels ≥ 15,000 pmol/L on trigger day, oocytes retrieval numbers ≥ 20, an unsuitable endometrium, elevated progesterone levels > 6 nmol/L, and other personal reasons. These patients and patients without live births from fresh transfer subsequently received frozen-thawed embryo transfer (FET). Serum AMH concentration was measured using a fully automatic chemiluminescence immunoassay analyzer (ADVIA Centaur XP, Siemens Healthcare Diagnostics) within 6 months before COS. Baseline serum hormone levels were measured on the second day of the menstrual cycle during the COS. A live birth refers to the delivery of ≥ 1 live fetus after 28 weeks of gestation. Miscarriage refers to pregnancy loss before 28 weeks of gestation. Clinical pregnancy refers to detecting the presence of gestational sacs in the uterus via ultrasonography. Cumulative live births resulting from a living fetus (or living fetus) occur in fresh and subsequent freeze–thaw transfers of one retrieval cycle. We analyzed cumulative live birth rates (CLBRs) of women involved between February 2017 and December 2021, considering that the pregnancy outcomes of embryos retrieved in 2022 were not completely obtained. SPSS version 28.0 (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. Categorical data were presented as corresponding percentages and frequencies, continuous data with normal distribution as mean ± standard deviation, and continuous data with non-normal distribution as median (interquartile range). The chi-square test was used for categorical variables, one-way analysis of variance for continuous data with normal distribution (Bonferroni correction in the post hoc test), and the Kruskal‒Wallis test for continuous data with non-normal distribution. Multivariable logistic regression analysis was conducted to adjust for relevant confounders. The odds ratio (OR) and standard 95% confidence interval (CI) were calculated to evaluate differences. Statistical significance was set at a two-sided P -value of < 0.05.

This retrospective study investigated whether AMH affects pregnancy outcomes among women with PCOS aged 20–40 years undergoing IVF/ICSI. Excessive AMH was correlated with a decreased LBR after fresh ET and was correlated with an increased miscarriage rate during fresh ET. The correlation between AMH level and BMI has been explored in previous studies. AMH levels are reportedly inversely correlated with BMI in women with PCOS [ 14 ]. Lefebvre et al. reported that AMH was lower in women with PCOS with normal weight than in those with obesity [ 15 ]. This study indicated a negative correlation between AMH levels and BMI in women with PCOS, consistent with previous findings. However, the mechanisms underlying this correlation remain unclear. One possible explanation may be the unbalanced secretion of leptin in women with PCOS [ 16 ]. Leptin is secreted by adipocytes, which are the main components of adipose tissue [ 17 ]. Elevated leptin levels inhibit AMH expression in GCs [ 18 ]. Therefore, women with higher BMI and PCOS have more adipose tissue and secrete more leptin, leading to decreased AMH levels. Previous study found the correlation between AMH and endometrium in PCOS. Su et al. reported that the endometrium on the transfer day was thinner in women with PCOS with AMH levels of ≥ 12 ng/mL than in those with AMH levels of < 12 ng/mL [ 19 ]. The mechanism underlying the relationship between AMH and the endometrium remains controversial. Recently, Paulson et al. reported the expression of AMH and AMH type II receptors in the endometria of women with PCOS [ 20 ]. AMH inhibits the growth of endometrial cancer cells by regulating several cell pathways and cellular factors [ 21 , 22 ]. Besides, AMH expression in endometriotic tissues suppressed the viability and growth of endometriotic cells [ 23 ]. These studies provide a possible explanation that overexpressed AMH may inhibit endometrial cell viability and function in women with PCOS [ 24 ], leading to decreased endometrial thickness in women with high AMH levels. In this study, endometrial thickness on hCG day was slightly lower in the high AMH group than in the low and average AMH groups. This difference in thickness was small and clinically insignificance, which may be due to the large sample size. In our study, no significant association was observed between AMH levels and CPR. Tal et al. reported that women with PCOS and with AMH levels of > 10 ng/mL had higher CPR than those with AMH levels of ≤ 10 ng/mL [ 11 ]. In another study, CPR was decreased in the high AMH group compared with in the low AMH group [ 25 ]. Nevertheless, Su et al. reported that no significant CPR differences were observed between women with AMH levels of ≥ 12 ng/mL and those with AMH levels of < 12 ng/mL [ 19 ]. In this study, we enrolled 4,719 women with PCOS in the analysis using 4.98 and 10.65 AMH levels as the 25th and 75th percentile cutoffs, respectively. In this study, no significant difference was observed in the CPR of fresh ET among the groups. Whether AMH level affects the occurrence of miscarriage in women with PCOS remains debatable [ 26 ]. A study reported that AMH level of ≥ 3.99 ng/mL was a risk factor for early miscarriage in women undergoing IVF/ICSI [ 27 ]. In contrast, Liu et al. reported that women with AMH levels of ≥ 5.71 ng/mL had lower early miscarriage rates than women with AMH levels of ≤ 2.25 ng/mL [ 12 ]. Both studies included women with and without PCOS in their analysis. Therefore, the results did not reveal the condition of the women with PCOS. This study revealed that the miscarriage rate was higher in the high AMH group than in the low AMH group among women with PCOS undergoing IVF/ICSI. However, the mechanisms underlying this correlation remain unclear. Gleicher et al. reported that excessive AMH was correlated with spiking miscarriage rates across all ages [ 28 ]. They suggested that AMH may be an unknown “embryo quality factor.” In addition, premature delivery was twofold higher in women with PCOS than in healthy women [ 24 ]. Excessive AMH was significantly correlated with preterm birth [ 29 ]. All participants with AMH levels of > 20 ng/ML delivered preterm infants, while women who delivered at term showed AMH levels that were similar to those of healthy females [ 29 ]. AMH and its receptor have been detected in the endometrium, which could directly or indirectly affect endometrial function [ 20 ]. In addition, excessive AMH impairs myometrial maintenance, remodeling, and the inhibitory effect of contractility, preventing the fetus from developing in the uterus [ 29 ]. In addition, AMH is positively associated with the androgen concentration in women with PCOS [ 30 ]. Excessive androgen concentration in the endometrial environment impairs endometrial function and receptivity by altering the expression of factors and signals, including estrogen receptor, progesterone receptor, and proinflammatory NF-κB signaling [ 24 , 31 , 32 ]. Hyperandrogenism impairs trophoblast invasion and placental mitochondrial function, increasing the risk of miscarriage [ 33 , 34 ]. Therefore, we considered that excessive AMH may lead to increased miscarriages in women with PCOS through the above possible mechanisms. Importantly, we observed a negative correlation between AMH levels and LBR in patients with PCOS. However, no significant among-group difference was found in LBR using 4.91 and 10.88 as the 25th and 75th percentile cutoffs, respectively [ 12 ]. Tal et al. reported that excessive AMH was correlated with lower LBR using 3.32 and 8.27 as the 25th and 75th percentile cutoffs, respectively [ 13 ]. In addition, a negative relationship was observed between AMH levels and LBR in women with PCOS aged < 30 years, using 6.77 and 14.30 as the 25th and 75th percentile cutoffs, respectively [ 25 ]. In this study, we involved 4,719 women with PCOS aged 20–40 years in the analysis using 4.98 and 10.65 as the 25th and 75th percentile cutoffs, respectively. The live birth rate was significantly lower in the high AMH group than in the low AMH group. Accordingly, excessive AMH was an independent prognostic factor for reduced LBR after fresh ET in women with PCOS aged 20–40 years undergoing IVF/ICSI. This negative correlation may be owing to the pathogenic role of excessive AMH in women with PCOS. Excessive AMH correlates with hyperandrogenism, anovulation, inflammation, and metabolic abnormalities in women with PCOS [ 35 , 36 ]. Women with PCOS and AMH level of ≥ 10.65 ng/mL are more prone to miscarriage, and less likely to delivery compared with those with PCOS and AMH level of ≤ 4.98 ng/mL (miscarriage rate, high vs. low, 21.8% vs. 11.3%; LBR, high vs. low, 30.4% vs. 39.8%). These findings prompted us to exercise greater caution when administering fresh ET in women with PCOS and AMH level of ≥ 10.65 ng/mL. In our study, no significant intergroup difference was observed in CLBR. This study suggested a lower LBR in the high AMH group than in the low AMH group, whereas CLBR was comparable among the groups. Above results indicated that pregnancy outcomes in women with PCOS and high AMH levels benefited from FET. One explanation may be the increased number of good-quality embryos in the high AMH group. Overall, these findings indicate a guiding role for AMH levels in the clinical regimen of women with PCOS. This study has some limitations. First, this was a single-center study, which may have resulted in bias. However, a single center can better control the quality of IVF-ET procedures and laboratory measurements. The second limitation is its retrospective and non-randomized design. Therefore, the findings of this study should be confirmed by large-scale, high-quality prospective cohort studies. Third, additional molecular biology experiments are essential to explore the role of AMH in PCOS pathogenesis.

The live birth rate for fresh ET was decreased in the high AMH group compared with that in the low AMH group, whereas the miscarriage rate for fresh ET was increased in the average and high AMH groups compared with that in the low AMH group. However, no significant differences were observed in CLBR among the groups. Therefore, high AMH level is associated with poor pregnancy outcomes of fresh ET in women with PCOS aged 20–40 years undergoing IVF/ICSI. These findings underscore the need for caution when administering fresh ET in women with PCOS and high AMH levels, providing useful medical insights into transplant strategies in this patient subgroup.

Anti-Müllerian hormone (AMH) is synthesized by pre-antral and small antral follicles in women and belongs to the members of the transforming growth factor β superfamily [ 1 ]. AMH counteracts the growth-promoting function of follicle-stimulating hormone (FSH) in granulosa cells (GCs) by reducing follicle sensitivity to FSH [ 2 ]. Therefore, a normal AMH concentration is essential for inhibiting primordial follicle recruitment, follicular development, and regulating dominant follicle selection [ 3 ]. AMH assesses ovarian reserve and predicts ovarian response to assisted reproductive technology in healthy women [ 4 ]. Polycystic ovary syndrome (PCOS) is the most common endocrine disorder causing infertility in women of reproductive age, with a worldwide prevalence of 5%–18% [ 5 ]. PCOS phenotypes include oligo- or anovulation, hyperandrogenism, polycystic ovarian morphology, and metabolic disorders [ 6 ]. The pathological mechanisms underlying PCOS are complex. An ovarian follicular arrest is an important pathophysiological feature of PCOS, resulting from insufficient FSH secretion or FSH inhibition [ 7 ]. AMH levels in women with PCOS are 2- to-fourfold higher than those in healthy females. Excessive AMH inhibits the growth-promoting function of FSH, leading to follicular arrest in women with PCOS [ 8 ]. In addition, excessive AMH impairs aromatase catalytic activity stimulated by FSH, preventing the conversion of testosterone and androstenedione to estrogen in GCs [ 3 ]. Insufficient estradiol synthesis in the follicular microenvironment may lead to impaired follicular development in women with PCOS [ 9 ]. Therefore, excessive AMH may be associated with PCOS severity, contrary to its role in healthy females [ 10 ]. Recently, researchers have focused on the correlation between AMH levels and pregnancy outcomes in women with PCOS undergoing in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI). High AMH levels may be beneficial for clinical pregnancy in women with PCOS [ 11 ]. However, AMH levels did not correlate with pregnancy outcomes in another study involving 418 women with PCOS [ 12 ]. In contrast, Tal et al. examined 184 women with PCOS and found a negative relationship between AMH levels and the live birth rate (LBR) [ 13 ]. These findings indicate that the correlation between AMH and pregnancy outcomes in women with PCOS undergoing IVF/ICSI remains controversial. Therefore, this study included large sample size representative of the nationwide population in China. We aimed to evaluate the correlation between AMH levels and pregnancy outcomes after IVF/ICSI in women with PCOS.

Supplementary Material 1. Supplementary Material 1.