Effect of hyperandrogenism on pregnancy outcomes in women with polycystic ovary syndrome undergoing in vitro fertilization/intracytoplasmic sperm injection-embryo transfer

Abstract

Objective: To explore the effects of hyperandrogenism (HA) on pregnancy outcomes in women with polycystic ovary syndrome (PCOS) undergoing in vitro fertilization/intracytoplasmic sperm injection-embryo transfer (IVF/ICSI-ET). Methods: A retrospective study was conducted on infertile women with PCOS undergoing IVF/ICSI-ET from January 2017 to June 2021 in our center. Patients were divided into HA group and NON-HA group according to the levels of testosterone. Propensity score matching (PSM) was used to balance the influence of female age and IVF/ICSI-ET for patients with gonadotropin-releasing hormone (GnRH)antagonist protocol and GnRH agonist protocol, separately. After the PSM procedure, 191 cases in HA group and 382 cases in NON-HA group, were included. Hormone levels and pregnancy outcomes were compared in the two groups. Results: The female age was comparable in two groups [HA: (29.6±3.7) vs NON-HA: (29.5±3.6), P=0.665]. The basal luteinizing hormone [(10.82±6.73) vs (7.76±5.30) IU/L], testosterone [(3.27±0.97) vs (1.60±0.59) nmol/L], free androgen index (7.13 vs 2.77), anti-mullerian hormone [(11.37±5.74) vs (9.67±4.67) ng/ml], fasting glucose [(5.18±0.49) vs (5.06±0.42) mmol/L], 1h glucose [(9.34±2.42) vs (7.99±2.21) nmol/L], 2 h glucose [(7.66±2.17) vs (6.64±1.84) nmol/L], 2 h insulin [(129.81±145.49) vs (97.51±86.92) mU/L], total cholesterol [(5.35±0.89) vs (4.92±0.92) mmol/L], triglycerides [(1.55±1.28) vs (1.33±0.77) mmol/L], and low density lipoprotein cholesterol levels [(3.38±0.66) vs (3.14±0.71) mmol/L] were significantly higher in HA group, compared with NON-HA group (P<0.05). The initiated gonadotropin dose was higher in HA group than that in NON-HA group [(126.96±33.65) vs (137.60±38.12) U, P=0.001], but moderate-severe ovarian hyperstimulation syndrome (OHSS) rate was similar in two groups (P>0.05). The rates of implantation, clinical pregnancy, miscarriage, and live birth were comparable between the two groups (P>0.05). Also, in the subgroups, the rates of implantation, clinical pregnancy, live birth, and miscarriage were similar in HA group and NON-HA group. Conclusions: The risks of hormonal abnormality and glucose-lipid metabolic disorder were higher in PCOS women with HA, whereas satisfactory pregnancy outcomes could be achieved under proper ovarian stimulation undergoing IVF/ICSI-ET.