Background It remains unclear whether polycystic ovary syndrome (PCOS) is an independent risk factor for pregnancy complications in women undergoing assisted reproductive technology (ART) treatment. For the integrative treatment of PCOS patients, it is still important to investigate the pregnancy outcomes of PCOS patients after adjusting for potential biases, such as body mass index, embryo quality and endometrial preparation method. Methods This retrospective cohort study ultimately included a total of 336 PCOS patients who conceived after single thawed blastocyst transfer in the PCOS group and 2,325 patients in the control group from January 2018 to December 2020. A propensity score matching (PSM) model was used, and 336 PCOS patients were matched with 336 patients in the control group. Results Before PSM, no differences in the miscarriage rate, pregnancy complication rate, preterm birth rate, or live birth rate were found between the PCOS group and the control group. After PSM, the late miscarriage rate of the PCOS group was significantly higher than that of the control group (3.3% vs. 0.6%, P = 0.040), although the early miscarriage rates were similar (14.0% vs. 13.7%). The rates of pregnancy complications, preterm birth and live birth in the PCOS group were comparable to those in the matched control group (P = 0.080, P = 0.105, P = 0.109, respectively). The neonatal weights of male infants and female infants were similar between the two groups (P = 0.219, P = 0.169). Subgroup analysis showed that PCOS patients with homeostasis model assessment of insulin resistance (HOMA-IR) levels ≥ 2.49 had a significantly increased risk of preterm birth compared with those with HOMA-IR levels < 1.26 and 1.26 ≤ HOMA-IR levels < 2.49 (26.0% vs. 6.0% vs. 9.8%, P = 0.005). PCOS patients with total testosterone levels ≥ 0.7 ng/ml had a higher early miscarriage rate but a lower late miscarriage rate than those with total testosterone levels < 0.7 ng/ml (29.4% vs. 12.3%, 0% vs. 3.6%, respectively, P = 0.032). Conclusions PCOS is an independent risk factor for late miscarriage in patients conceived after a single thawed blastocyst transfer, even after adjusting for biases. Among PCOS patients, insulin resistance and hyperandrogenism are associated with a higher risk of preterm birth and early miscarriage, respectively.

ObjectiveTo evaluate HIF2α polymorphisms and glucose metabolism in a group of women with polycystic ovary syndrome (PCOS) or unexplained infertility (UI).PatientsThe infertile group (n=148) consisted of 96 women with PCOS, 52 women with UI, and176 women without infertility as a healthy control group.InterventionWe genotyped 29 single nucleotide polymorphisms (SNPs) of HIF2α by using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS)-based genotyping technology. The genetic associations were analyzed statistically.Main outcome measuresAllele frequency, genotype distribution and haplotype analyze of the HIF2α polymorphisms were performed. Body mass index (BMI), waist circumference, uric acid (UA), high-sensitivity C-reactive protein (hsCRP), lipids, glucose and insulin tolerance - were also measured.ResultsInfertile women with PCOS had a higherBMI and waist circumference, elevated hsCRP and uric acid (UA) levels, impaired glucose tolerance, and increased levels of plasma insulin compared to UI patients and healthy women. SNP analysis of HIF2α revealed that the allele and genotype frequencies of rs4953361 were significantly associated with infertile women with PCOS. Haplotype analysis of the HIF2α polymorphism identified haplotypes TGG and TGA as being associated with infertile women with PCOS. Women with the AA genotype of rs4953361 had a significantly higher BMI and post load plasma glucose and insulin levels than those of women with the GG genotype.ConclusionInfertile women with PCOS more commonly have metabolic disturbances than those with UI. This is the first study to report an association between HIF2α polymorphisms (rs4953361) and the risk of infertile women with PCOS, not UI, in Han Chinese population. These results require replication in larger populations.In this observational study, we did not report the results of a health care intervention on human participants. The study was approved by the Human Research Ethics Committee of the First Affiliated Hospital of Chongqing Medical University. Clinical data and peripheral blood samples were collected only after explaining the objectives of the study and obtaining a signed informed consent form.