Obstructive sleep apnea as a predictive indicator for in vitro fertilization and embryo transfer outcomes in patients with polycystic ovary syndrome: a prospective cohort study.

The study encompassed 360 infertile patients diagnosed with PCOS who underwent IVF treatment utilizing the GnRH-ant protocol from January 2019 to September 2022. Among these individuals, a total of 354 patients underwent fresh or thawed embryo transfer (Fig.  1 ). The prevalence of OSA among the 360 infertile PCOS patients was 30.0% (108/360). Categorically, with the majority classified as mild (80.6%), followed by moderate (15.7%), and severe cases (3.7%). Fig. 1 Flow chart of patient enrollment. Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; HAST: home sleep apnea test Flow chart of patient enrollment. Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; HAST: home sleep apnea test Patients with PCOS&OSA had significantly higher BMI compared to PCOS patients (Table  1 ). There were no statistically significant differences were observed between the two groups in terms of age, type of infertility, and duration of infertility. Table 1 Comparison of baseline characteristics, reproductive and metabolic profile between PCOS and PCOS&OSA patients Characteristics PCOS ( n  = 252) PCOS&OSA ( n  = 108) p Age 31.0 ± 3.7 31.3 ± 3.6 0.857 BMI 23.6 ± 3.4 28.0 ± 4.4 < 0.001 Infertility type  Primary 176(69.8%) 77(71.3%) 0.782  Secondary 76(30.2%) 31(28.7%) Duration of infertility (yrs) 3(2) 3(3) 0.107 Menstruation cycle  Regular 54(21.4%) 20(18.5%) 0.531  Irregular 198(78.6%) 88(81.5%) Reproductive endocrine characteristics  AMH (ng/ml) 6.87(5.30) 5.37(4.02) < 0.001  T(nmol/l) 0.71(0.47) 0.86(0.50) 0.029  AND(nmol/l) 10.0(5.9) 10.7(7.9) 0.244  B-FSH(mIU//ml) 5.46(2.20) 5.68(2.10) 0.416  B-LH(mIU/ml) 3.9(7.7) 3.3(8.5) 0.787  B-E2(pmol/l) 152(74) 156.5(76.5) 0.553  PGN(nmol/l) 1.12(0.70) 1.06(0.70) 0.324 Metabolic profile  SBP (mmHg) 120(16) 124(16) 0.006  DBP (mmHg) 75(13) 79(15) 0.001  HbA1c(%) 5.4(0.4) 5.6(0.4) < 0.001  HOMA-IR 2.09(1.90) 3.97(4.06) < 0.001  TG (mmol/l) 1.13(0.9) 1.48(1.02) < 0.001  T-CHO (mmol/l) 4.62(1.02) 4.755(1.11) < 0.438  HDL-C (mmol/l) 1.31(0.41) 1.16(0.23) <0.001  LDL-C (mmol/l) 2.84(0.86) 2.96(1.02) < 0.084  MS 27(10.7%) 34(31.5%) < 0.001 Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; BMI: body mass index; AMH: anti-Mullerian hormone; T: testosterone; AND: androstenedione; B-FSH: basic follicle-stimulating hormone; B-LH: basic luteinizing hormone; B-E2: basic estradiol; P: progesterone; SBP: systolic blood pressure; DBP: diastolic blood pressure; HbA1c: glycosylated hemoglobin; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; TG: triglyceride; T-CHO: total cholesterol; HDL-C: high-density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol; MS: metabolic syndrome Comparison of baseline characteristics, reproductive and metabolic profile between PCOS and PCOS&OSA patients Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; BMI: body mass index; AMH: anti-Mullerian hormone; T: testosterone; AND: androstenedione; B-FSH: basic follicle-stimulating hormone; B-LH: basic luteinizing hormone; B-E2: basic estradiol; P: progesterone; SBP: systolic blood pressure; DBP: diastolic blood pressure; HbA1c: glycosylated hemoglobin; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; TG: triglyceride; T-CHO: total cholesterol; HDL-C: high-density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol; MS: metabolic syndrome Compared to the PCOS group, patients in the PCOS&OSA group exhibited significantly lower serum AMH levels (6.39 ± 4.83 ng/ml vs. 7.53 ± 4.24 ng/ml, p  < 0.05) and higher T levels (1.07 ± 0.69 nmol/L vs. 0.98 ± 0.57 nmol/L, p  < 0.05) compared to PCOS patients. However, no significant differences were found in baseline levels of E2 (196.6 ± 270.8 pmol/l vs. 170.0 ± 81.5 pmol/l, p  = 0.157), PGN(1.54 ± 3.54 nmol/l vs. 1.27 ± 0.67 nmol/l, p  = 0.257), AND (11.76 ± 6.10 vs. 10.71 ± 5.00, p  = 0.133), LH (5.53 ± 5.31 mIU/ml vs. 5.73 ± 4.86 mIU/ml, p  = 0.719), or FSH (5.71 ± 1.94 mIU/ml vs. 5.73 ± 2.37 mIU/ml, p  = 0.949) between the two groups (Table  1 ). Metabolic disorders are commonly associated with PCOS. As it is shown in Table  1 and Supplementary Table 1, metabolic abnormalities were more pronounced in PCOS & OSA patients compared to PCOS patients. Among PCOS patients, those with OSA exhibited significantly higher blood pressure (systolic blood pressure: 125.49 ± 14.72 vs. 120.75 ± 12.66 mmHg; diastolic blood pressure: 79.36 ± 11.75 vs. 74.77 ± 9.88 mmHg; p  < 0.05 for both). Moreover, PCOS&OSA patients also demonstrated worse glucose metabolism parameters, including higher levels of HbA1c (5.59 ± 0.13 vs. 5.42 ± 0.39%, p  < 0.05), OGTT-60 (13.12 ± 9.10 vs. 10.35 ± 6.67 mmol/L, p  < 0.05), OGTT-120 (11.1 ± 7.15 vs. 9.09 ± 6.40 mmol/L, p  < 0.05), INS-0 (23.36 ± 8.96 vs. 13.65 ± 13.04 nIU/ml, p  < 0.05), INS-30 (23.36 ± 8.96 vs. 13.65 ± 13.04 nIU/ml, p  < 0.05), INS-60 (134.85 ± 115.35 vs. 96.96 ± 89.21 nIU/ml, p  < 0.05), INS-120 (173.88 ± 211.96 vs. 95.28 ± 100.18 nIU/ml, p  < 0.05), as well as HOMA-IR (5.27 ± 4.52 vs. 3.04 ± 3.10, p  < 0.05), compared to patients without OSA. With regards to lipid metabolism, PCOS&OSA patients had higher levels of total triglycerides (1.80 ± 1.21 vs. 1.39 ± 0.96 mmol/L, p  < 0.05) and lower levels of total HDL (1.20 ± 0.28 vs. 1.32 ± 0.30 mmol/L, p  < 0.05) when compared to PCOS patients without OSA. Furthermore, the prevalence of metabolic syndrome was significantly higher in PCOS&OSA patients compared to PCOS patients without OSA (31.5% vs. 10.7%, p  < 0.05). When considering the data related to IVF, the administration of Gn was found to be higher in PCOS&OSA patients compared to PCOS patients, as indicated by the dosage (1953.77 ± 861.05 vs. 2354.97 ± 1009.11 IU, p  < 0.05) and duration (10.39 ± 2.43 vs. 11.07 ± 2.32 days, p  < 0.05) (Table  2 ). However, no significant differences were observed in E2, LH, and HCG levels on the trigger day between the two groups. In terms of IVF outcomes, the total number of oocytes retrieved was significantly lower in PCOS&OSA patients. In IVF-ICSI cycles, PCOS&OSA patients demonstrated a reduced total fertilization rate and a downward trend in the 2PN fertilization rate, although the difference in 2PN rate was not statistically significant( p  = 0.390). There were no statistical difference in MII rate or good quality embryo rate between the two groups. However, there were no significant differences identified in the rates of total fertilization, 2PN and good quality embryo in IVF-RT between PCOS and PCOS&OSA patients (Table  2 ). Table 2 Comparison of IVF cycle characteristics between PCOS and PCOS&OSA patients PCOS ( n  = 252) PCOS&OSA ( n  = 108) p Total amount of Gn dosage(IU) 1953.77 ± 861.05 2354.97 ± 1009.11 < 0.001 Duration of stimulation (d) 10.39 ± 2.43 11.07 ± 2.32 0.016 Hormone levels on HCG day  E2(pmol/l) 13784.66 ± 9543.49 12472.48 ± 8466.61 0.287  LH(mIU/ml) 2.61 ± 2.46 2.96 ± 3.44 0.351  PGN(nmol/l) 2.41 ± 1.60 2.43 ± 1.64 0.905 Number of retrieved oocytes 17(13) 13(11) < 0.001 Insemination method  Conventional IVF 154(61.1%) 69(64.5%) 0.436  ICSI 91(36.1%) 35(32.7%)  Half-ICSI 7(2.8%) 3(2.8%) Conventional IVF Total fertilization rate (%) 74.3% 75.4% 0.459 2PN rate (%) 61.1% 59.7% 0.390 Good quality embryo rate 33.1% 35.6% 0.128 ICSI Rate of MII oocyte obtained 74.2% 77.2% 0.107 Total fertilization rate (%) 76.6% 69.8% 0.006 2PN rate (%) 67.7% 62.7% 0.067 Good quality embryo rate 50.4% 50.2% 0.962 Endometrial thickness on the trigger day(mm) 10.2 ± 0.10 10.3 ± 0.18 0.462 Transfer strategy Fresh ET 162(65.7%) 56(54.9%) 0.81  Frozen ET 85(34.2%) 46(45.1%) Days of transferred embryo  D3 181(73.0%) 81(77.1%) 0.122  D5/6 67(27.0%) 24(22.9%) Number of embryos transferred  One 74(29.8%) 31(29.5%) 0.416  Two 174(70.2%) 74(70.5%) Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; HCG: human chorionic gonadotropin; E2: estradiol; LH: luteinizing hormone; P: progesterone; Gn: gonadotropin; IVF: in vitro fertilization; ICSI: intracytoplasmic sperm injection; PN: pronuclear; ET: embryo transfer Comparison of IVF cycle characteristics between PCOS and PCOS&OSA patients Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; HCG: human chorionic gonadotropin; E2: estradiol; LH: luteinizing hormone; P: progesterone; Gn: gonadotropin; IVF: in vitro fertilization; ICSI: intracytoplasmic sperm injection; PN: pronuclear; ET: embryo transfer We conducted a follow-up study on the pregnancy outcomes of patients undergoing IVF treatment. Among the 360 patients included in the study, a total of 353 patients underwent embryo transfer. The detailed pregnancy outcomes for the IVF-ET/FET patients are presented in Table  3 . Among the PCOS&OSA patients, a total of 105 patients underwent embryo transfer, with 35 patients (33.3%) achieving a live birth outcome after the first transfer. In contrast, out of the 248 PCOS patients underwent embryo transfer, and 128 patients (51.6%) achieved a live birth. The presence of OSA was associated with significantly lower rates of biochemical pregnancy (49.5% vs. 63.3%; RR = 0.841; p  < 0.05), clinical pregnancy (44.8% vs. 69.1%; p  < 0.05), and live birth (33.3% vs. 51.6%; RR = 0.829; p   0.05) between the PCOS&OSA and PCOS patients. Table 3 Comparison of pregnancy outcomes after IVF between PCOS and PCOS&OSA patients PCOS ( n  = 248) PCOS&OSA ( n  = 105) p Relative risk CI(95%) Lower Upper Biomedical pregnancy 157(63.3%) 52(49.5%) 0.016 0.841 0.726 0.975 Clinical pregnancy 149(60.1%) 47(44.8%) 0.008 0.829 0.719 0.957 Miscarriage 25(10.1%) 11(10.5%) 0.911 1.013 0.806 1.273 Live birth 128(51.6%) 35(33.3%) 0.002 0.804 0.794 0.921 Cumulative live birth within 18 months 156(62.9%) 48(45.7%) 0.003 0.807 0.697 0.936 Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; CI: confidence interval Comparison of pregnancy outcomes after IVF between PCOS and PCOS&OSA patients Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; CI: confidence interval The cumulative live birth rate of PCOS&OSA patients was found to be significantly lower than that of PCOS patients during the 18-month follow-up period after enrollment (45.7% vs. 62.9%, p  < 0.05, Table  3 ). Additionally, the Kaplan-Meier survival analysis revealed a statistically significant difference in the cumulative live birth rate between the two groups, with a hazard ratio of 0.65 ( p  < 0.05, Fig.  2 ). Fig. 2 Kaplan-Meier curve of cumulative live birth rate within 18 months. Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; HR: Hazard ratio Kaplan-Meier curve of cumulative live birth rate within 18 months. Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; HR: Hazard ratio To assess the association between OSA and the decline of the live birth rate in the first transfer cycle, a multifactorial binary logistic regression analysis was conducted. This analysis adjusted for factors including age, BMI, fresh or thawed embryo transfer, and the number and days of embryos transferred. The results showed that OSA was independently associated with a decline in the live birth rate (RR 0.599, p  < 0.05, Fig.  3 ). We further plotted ROC curves based on AHI and cumulative live birth within 18 months, with an area under the curve(AUC) of 5.74(CI 0.513–0.636) and a cutoff value of 5.05 (Fig.  4 ). Fig. 3 Logistic regression analysis on live birth rate of PCOS patients undergoing IVF. Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; RR: relative risk; CI: confidence interval; BMI: body mass index; ET: embryo transfer Logistic regression analysis on live birth rate of PCOS patients undergoing IVF. Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; RR: relative risk; CI: confidence interval; BMI: body mass index; ET: embryo transfer Fig. 4 ROC by Ahi. Abbreviations: ROC: receiver operating characteristic; AHI: apnea-hypopnea index; AUC: area under the curve ROC by Ahi. Abbreviations: ROC: receiver operating characteristic; AHI: apnea-hypopnea index; AUC: area under the curve As reproductive endocrine and metabolic abnormalities are prominent clinical manifestations of PCOS&OSA patients, we performed a univariate analysis to explore the relationship between these factors and live birth in IVF. The results indicate that the live birth rate in the first transfer cycle was negatively correlated with BMI, HbA1c, and HOMA-IR ( p  < 0.05, Table  4 ). Table 4 Univariate analyses of factors associated with live birth Live birth t/χ2 /U/H p Yes( n  = 163) No( n  = 190) AMH (ng/ml) 1.04 ± 0.62 0.97 ± 0.52 0.623 0.534 T (nmol/l) 7.03 ± 4.80 7.33 ± 3.97 1.165 0.245 BMI(kg/m 2 ) 25.45 ± 3.25 24.22 ± 4.15 2.761 0.006 SBP (mmHg) 122.49 ± 13.25 121.80 ± 13.90 0.478 0.633 DBP (mmHg) 76.68 ± 10.88 75.93 ± 11.08 0.642 0.521 HbA1c(%) 5.53 ± 0.36 5.41 ± 0.41 2.975 0.003 TG (mmol/l) 1.54 ± 1.04 1.52 ± 1.11 0.184 0.854 HDL-C (mmol/l) 1.28 ± 0.31 1.29 ± 0.28 0.339 0.735 HOMA-IR 3.13 ± 3.51 4.16 ± 3.89 2.592 0.010 MS 21(12.9%) 38(20.0%) 3.192 0.074 Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; AMH: anti-Mullerian hormone; T: testosterone; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; HbA1c: glycosylated hemoglobin; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; MS: metabolic syndrome Univariate analyses of factors associated with live birth Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea; AMH: anti-Mullerian hormone; T: testosterone; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; HbA1c: glycosylated hemoglobin; HOMA-IR: Homeostasis Model Assessment of Insulin Resistance; TG: triglyceride; HDL-C: high-density lipoprotein cholesterol; MS: metabolic syndrome Subsequent investigations aimed at evaluating the influence of OSA on obstetric complications in PCOS patients are detailed Table  5 . A total of 204 patients achieved live birth, of whom had available follow-up pregnancy data. There were no significant differences in twin pregnancy rate and cesarean section rate between PCOS and PCOS&OSA patients. PCOS &OSA patients have a higher risk of gestational diabetes(GDM)( p  < 0.05). Whereas PCOS&OSA patients had a increased likelihood of preterm birth, gestational diabetes mellitus (GDM), hypertensive disorders of pregnancy (HDP), placental abruption, premature rupture of membranes, postpartum hemorrhage, macrosomia, and low birth weight infants, these differences were not statistically significant. Table 5 Comparison of obstetric complications between PCOS and PCOS&OSA patients PCOS ( n  = 145) PCOS&OSA ( n  = 46) p Twin pregnancy 16.0%(25/156) 20.8%(10/48) 0.644 Cesarean section 64.7%(101/156) 70.8%(34/48) 0.435 Preterm birth 16.0%(25/156) 25.0%(12/48) 0.158 Gestational diabetes mellitus 14.1%(22/156) 27.1%(13/48) 0.037 Hypertensive disorders of pregnancy 11.5%(18/156) 18.8%(9/48) 0.197 Premature rupture of membranes 5.8(9/156) 10.4%(5/48) 0.265 Postpartum hemorrhage 3.8%(6/156) 4.2%(2/48) 0.920 Newborn birth weight 3070(168) 3049(55) 0.57 Macrosomia 2.6%(4/156) 6.3%(3/48) 0.220 Low birth weight babies 12.8%(20/156) 14.6%(7/48) 0.753 Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea Comparison of obstetric complications between PCOS and PCOS&OSA patients Abbreviations: PCOS: polycystic ovary syndrome; OSA: obstructive sleep apnea

The study protocol was approved by the Ethics Committee of the Third Hospital of Peking University (2018S2-002) in accordance with the 1975 Declaration of Helsinki. All patients involved in the project provided informed consent and their information, clinical data, and samples were used with their full knowledge and consent. This study enrolled 928 infertile patients with PCOS, aged 22 to 38, from January 2019 to September 2022 at the Reproductive Medical Center of Peking University Third Hospital. PCOS was diagnosed using the Rotterdam criteria of 2003 [ 17 ], which required the fulfillment of at least two out of the following three criteria: (1) Oligo- and/or anovulation (OA); (2) Hyperandrogenism (HA); (3) Polycystic ovaries (PCOM), as well as the exclusion of other potential causes such as congenital adrenal hyperplasia, androgen‐secreting tumors, and Cushing’s syndrome. PCOS cases were further categorized into four phenotypes based on clinical features: phenotype A, characterized by OA, HA, and PCOM; phenotype B, displaying OA and HA without PCOM; phenotype C, exhibiting HA and PCOM; and phenotype D, meeting the criteria of OA and PCOM only. All participants were newly diagnosed and excluded if they had a history of smoking or alcohol consumption, hypertension or diabetes, abnormal thyroid function, or use of oral contraceptives or other hormonal drugs within 3 months prior to enrollment. A subset of 360 patients, who had been assessed by a fertility specialist, met the criteria for receiving IVF treatment. Indications for IVF include: (1) Gamete transport disorders caused by various factors in the female; (2) Ovulation disorders; (3) Endometriosis; (4) Male oligoasthenospermia; (5) Unexplained infertility [ 18 ]. All of these patients underwent their initial IVF cycle using the Gonadotropin-releasing hormone-antagonist (GnRH-ant) protocol. The selection of fertilization methods, such as conventional IVF, intracytoplasmic sperm injection (ICSI), or half-ICSI, was based on the quality of the male partner’s sperm. All women underwent ovarian stimulation using GnRH antagonist protocols. Ovarian stimulation begins on cycle day 2 nd –3 th using recombinant FSH or human menopausal gonadotropin (hMG). Antagonist begins when the diameter of lead follicle reaches 12–14 mm or serum estradiol (E2) exceeds 600–800 pg/mL. Final oocyte maturation was triggered by administering 250 µg of human chorionic gonadotropin (hCG) when ultrasound examination revealed at least two dominant follicles measuring ≥ 18 mm in diameter. Oocyte Retrieval was performed 36 h after hCG injection [ 19 ]. Sperm analysis adhered to the WHO Laboratory Manual for the Examination and Processing of Human Semen (5th edition, 2010) and incorporated advanced functional assessments to evaluate semen quality comprehensively. The diagnosis of OSA in this study was made according to the 2016 edition of the American Academy of Sleep Medicine (AASM) Guidelines [ 20 ]. These guidelines require either polysomnography (PSG) or out-of-center sleep monitoring (OCST) to confirm the presence of respiratory events occurring at a rate of ≥ 5 breaths/hour during the testing period. In addition, clinical manifestations associated with OSA and comorbidities must also be considered. These manifestations include nonrestorative sleep, malaise or insomnia, sleep snoring, and disrupted breathing. Alternatively, a diagnosis can be made if PSG or OCST confirms the occurrence of ≥ 15 respiratory events per minute during the monitoring period. All study participants visited the respiratory department of Peking University Third Hospital, where they received sleep counseling and underwent OSA screening using Apnealink Plus, a device manufactured by ResMed Ltd. ( Australia). During sleep, data was recorded on various parameters including nasal airflow, snoring, arterial oxygen saturation, pulse, and respiratory effort. The software automatically calculated the apnea-hypopnea index (AHI) based on the recorded results, and a respiratory physician made the final diagnosis. According to the AASM guideline, the severity of OSA is classified as mild, moderate, or severe based on AHI thresholds of ≥ 5, ≥15, and ≥ 30, respectively [ 20 ]. The participants’ demographic information, sex hormone levels, and metabolic profiles were obtained from electronic medical records. These data included age, body weight, height, blood pressure, type and duration of infertility, menstrual cycle, and reproductive endocrine and metabolic markers such as follicle stimulating hormone (FSH), luteinizing hormone (LH), anti-Mullerian hormone (AMH), testosterone (T), androstenedione (AND), progesterone (PGN), estradiol (E2), prolactin (PRL), oral glucose tolerance (OGTT), insulin releasing test, glycosylated hemoglobin (HbA1c), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C). Insulin resistance was assessed using the Homeostasis Model Assessment of Insulin Resistance (HOMA-IR), calculated as fasting insulin (mIU/L) multiplied by fasting glucose (mmol/L) and divided by 22.5. Metabolic syndrome was defined according to the Chinese Diabetes Society [ 21 ] recommendations, requiring the presence of three or more of the following criteria: (1) Abdominal obesity, defined as waist circumference ≥ 85 cm for women, (2) Hyperglycemia, indicated by fasting blood glucose ≥ 6.1 mmol/L or 2-hour blood glucose ≥ 7.8 mmol/L after glycemic load, or a previous diagnosis and treatment of diabetes mellitus, (3) Hypertension, defined as blood pressure ≥ 130/85 mmHg, or a previous diagnosis and treatment of hypertension, and (4) Fasting triglyceride (TG) levels ≥ 1.70 mmol/L. The assessment of IVF outcomes comprised several factors: the dosage and duration of gonadotropins (Gn), E2, PGN and LH on trigger day, the number of oocytes retrieved, the oocyte maturation rate (within ICSI cycles), the fertilization rate, and the rates of two pronuclei (2PN) and good quality embryos. Oocyte maturation rate refers to the ratio mature MII oocytes to total number of oocytes retrieved. Fertilization rate is defined as the one or two or multiple PN divided by total number of oocytes retrieved. For ICSI cycles, it is calculated as the number of one or two or multiple PN divided mature MII oocytes. The rate of 2PN represent normal fertilization rate, which is calculated as the percentage of 2PN to total number of oocytes retrieved in conventional fertilization or MII number in ICSI. Quality of embryos is graded according to the morphology assessment criteria of the Istanbul Embryo Evaluation Symposium (Istanbul consensus workshop on embryo assessment: Proceedings of an expert meeting). The rate of good quality embryos refers to the ratio of good quality embryos on day 3 to 2PN. Live birth was defined as the presence of at least one newborn baby being born alive. Cumulative live birth referred to the first instance of live birth subsequent to either fresh or frozen embryo transfer within a single oocyte retrieval cycle. Statistical Package for Social Sciences (SPSS, version 27.0, SPSS Inc., Chicago, IL) and R language (R Foundation for Statistical Computing, version 4.2.1, Vienna, Austria) were utilized for all statistical analyses in this study. Categorical variables were described using percentages, while continuous variables that followed a normal distribution were described using mean and standard deviation (SD). Non-normally distributed continuous variables were described using median and interquartile range (IQR). Differences between continuous variables were assessed using analysis of variance (ANOVA) or nonparametric Kruskal-Wallis test, and differences between categorical variables were assessed using Pearson χ2 test. To identify variables that may affect IVF pregnancy outcomes, a multifactorial logistic regression analysis model was used, incorporating evidence from the relevant literature. The results of the logistic regression were presented as relative risk (RR) with 95% confidence intervals (CI). A p-value of < 0.05 was considered statistically significant.

In conclusion, this study revealed the negative impact of OSA on the clinical outcome of IVF in patients with PCOS. Univariate and multifactorial regression underscored an independent association of OSA with biochemical pregnancy, clinical pregnancy, and live birth, which may be related to the aggravation of abnormalities of glycolipid metabolism by OSA in patients with PCOS. Screening for and treating OSA in patients with PCOS may improve the outcome of pregnancy. It is recommended that all PCOS patients undergo OSA evaluation, and those with moderate to severe OSA should be referred to respiratory medicine for further evaluation and treatment.

A mounting body of research has identified a comparable range of comorbidities between patients with PCOS and those with OSA. Notably, the prevalence of OSA in patients with PCOS is significantly higher, typically ranging from 2 to 9.74 times greater than that of women without PCOS [ 11 ]. Recognizing the importance of this association, the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome recommended that PCOS patients undergo screening for OSA [ 7 ]. In accordance with our previous study [ 12 ], incidence of OSA among patients in this study reached 30%, based on a cohort of 360 PCOS patients undergoing IVF treatment. Furthermore, OSA correlated with adverse pregnancy outcomes in patients with PCOS. Specifically, we observed significant reductions in biochemical pregnancy rate, clinical pregnancy rate, live birth rate and cumulative live birth rate within 18 months of enrollment in patients with PCOS&OSA. There was also an increasing trend of incidence of obstetric complications among PCOS & OSA patients compared to PCOS. Managing OSA in women of reproductive age could be an effective strategy for improving reproductive outcomes, making it a crucial aspect of public health. The mechanisms underlying the association between OSA and poor pregnancy outcomes in women with PCOS have not been fully understood. OSA is characterized by intermittent hypoxia (IH) and increased oxidative stress (OS), both of which can result in various metabolic disturbances [ 22 ].This study found that the increased prevalence of OSA in PCOS patients may be associated with reproductive endocrine and metabolic abnormalities, consistent with previous studies [ 11 , 12 ]. Hypoxia and oxidative stress are pivotal drivers of reproductive and metabolic dysregulation. Hypoxia impairs ovarian function by activating hypoxia-inducible factor 1-alpha (HIF-1α), which suppresses AMH expression in granulosa cells—disrupting follicle recruitment and depleting ovarian function [ 23 – 25 ]. Simultaneously, HIF-1α upregulates steroidogenic enzymes(e.g., CYP17A) in theca cells, amplifying androgen production and exacerbating HA [ 26 ]. Oxidative stress, driven by excessive reactive oxygen species (ROS), further compounds these effects by inducing apoptosis and fibrosis of ovarian granulosa cells and diminishing AMH levels [ 27 ], while also promoting ovarian androgen synthesis [ 28 ]. These intertwined pathways extend beyond reproductive dysfunction to metabolic syndrome: hypoxia and oxidative stress exacerbates insulin resistance and dyslipidemia by suppressing adiponectin and activating lipogenic pathways, while ROS-driven mitochondrial dysfunction and chronic inflammation perpetuate obesity and hypertension [ 29 , 30 ]. Collectively, hypoxia and oxidative stress create a vicious cycle, reducing fertility through AMH depletion and HA while simultaneously worsen metabolic disturbance. HA is a hallmark feature of PCOS, and appears more severe among PCOS&OSA patients according to our findings, which are consistent with previous studies [ 11 , 12 , 31 ]. Clinically, hyperandrogenemia in PCOS correlates with poor fertilization rates, embryonic defects, and miscarriage [ 32 ]. Androgens, while essential for early follicle development, impair oocyte quality when excessive in PCOS. Excess androgens disrupt mitochondrial function and elevate oxidative stress in oocytes, reducing ATP production and damaging DNA [ 33 – 35 ]. They also impair communication between granulosa cells and oocytes leading to abnormal follicular development and reduced oocyte quality [ 33 , 36 , 37 ]. Several studies have found that HA is associated with early pregnancy miscarriage, preterm birth and obstetric complications including GDM, HDP and low birth weight babies [ 38 – 41 ]. Androgens may affect endometrial receptivity through androgen receptors(ARs) and acting as estrogen precursors [ 42 ]. Adequate androgens are necessary for placenta development, whereas excess androgens in women with PCOS alter placental morphology and angiogenesis [ 43 ]. Thus, HA may play a role in adverse pregnant outcomes among PCOS&OSA patients. AMH serves as a widely-used clinical indicator for assessing ovarian function. Our study revealed a decrease in AMH levels among PCOS&OSA patients compared to PCOS patients, aligning with previous research findings. Moreover, PCOS&OSA patients exhibited a prolonged duration and higher dosage of Gn use during COH, while the total number of retrieved oocytes decreased, suggesting impaired ovarian response. This implies that OSA may diminish ovarian reserve function in PCOS patients. Buratini’s study also observed that reduced AMH levels were associated with abnormal follicular development and decreased oocyte quality [ 44 ]. Furthermore, previous studies have indicated that PCOS patients with OSA experienced a decline in the number of good quality and transferable embryos during IVF, as well as a decrease in fertilization rate, good quality embryos, and delayed blastocyst formation during conventional and ICSI-IVF in women exposed to hypoxia [ 16 , 45 ]. In conjunction with prior research, our study suggests that AMH levels are reduced and ovarian responsiveness is impaired in PCOS&OSA patients, indicating that OSA may impact follicular development and contribute to a decline in oocyte quality. In the endometrium, AMH receptor signaling, essential for decidualization, is disrupted under hypoxic conditions, leading to embryo implantation [ 46 ]. Collectively, hypoxia and oxidative stress synergistically impair AMH’s dual roles in sustaining oocyte quality and endometrial receptivity. In addition, recent studies have shown that AMH plays a role in hypothalamic-pituitary-ovarian(HPO) axis regulation [ 47 ]. GnRH reduces serum AMH levels, which is negatively correlated with the increase in gonadotropins [ 48 ]. AMH and GnRH act in a reciprocal manner. OSA leads to reduced AMH levels in PCOS patients, which may further affect reproductive endocrine levels and pregnancy outcomes through abnormal HPO axis function. In PCOS&OSA patients, an interesting paradox arises: elevated androgen levels are accompanied by reduced AMH levels. Elevated levels of testosterone and other androgens stimulate ovarian follicles, affecting follicular development and AMH secretion. Testosterone and DHEA regulates granulosa cell function and upregulates the expression of AMH [ 49 , 50 ]. However, in the case of PCOS&OSA, AMH levels decreased despite the persistence of hyperandrogenemia. The pathological mechanism of this phenomenon remains unclear, and one possible explanation is that oxidative stress due to intermittent hypoxia caused by OSA negatively affects ovarian function. In studies in mice, hypoxia was found to lead to increased levels of autophagy and decreased fertility in granulosa cells [ 51 ]. The complex interactions between HA, AMH and OSA deserve further study. Metabolic abnormalities are a prominent clinical manifestation in patients with PCOS and OSA. Our study revealed that the coexistence of PCOS and OSA was associated with an increased risk of obesity and insulin resistance, consistent with previous studies conducted in our center and other related research [ 12 , 22 ]. Further analysis indicated that elevated BMI, higher rates of IR, and MS were linked to adverse pregnancy outcomes during IVF, aligning with findings from numerous related studies [ 19 , 52 , 53 ]. We hypothesize that OSA may exacerbate metabolic abnormalities in PCOS patients, consequently contributing to poor pregnancy outcomes in IVF. This effect may be mediated through disruption of the hypothalamic -pituitary -gonadal axis caused by metabolic disorder [ 54 ]. IR causes oxidative stress and disrupts mitochondrial function in mouse oocytes, leading to reduced oocyte quality associated with decreased fertilization and an arrest of embryo development [ 55 ]. Similarly IR leads to a decrease in the rate of oocyte maturation and blastocyst formation [ 56 ]. Chronic inflammation caused by MS, including IR, obesity and hyperlipidemia, result in impaired endometrial receptivity and displacement of the window of implantation [ 57 , 58 ]. IR and HA synergize to cause excess production of reactive oxygen species (ROS) in placenta, mitochondrial dysfunction, and disturbed superoxide dismutase-1 (SOD1) and Keap1/Nrf2 antioxidant responses leading to fetal loss in the rat [ 59 ]. And several studies have been proposed that MS, including IR and obesity, may contribute to a higher risk of developing GDM [ 60 – 62 ]. Our findings suggest that the concurrent rise of OSA and MS incidence may be associated with hypoxia and OS, resulting in poorer pregnancy outcomes. Therefore, OSA screening is essential for all patients, with prompt evaluation and management of identified cases. The strength of this study lies in its provision of detailed clinical data with a relatively large sample size, which allows for the characterization of the population under investigation, including the examination of biochemical markers associated with glucose and lipids, as well as the assessment of IVF outcomes. However, there are several limitations that warrant acknowledgment. Firstly, this particular study exclusively focused on PCOS patients who underwent IVF treatment, without the inclusion of a healthy control group. Secondly, due to the limited sample size, this study combined the analysis of pregnancy outcomes between fresh and thawed cycle transplants. Also, larger sample sizes are needed to examine the influence of obstetric complications and other conditions on IVF. To enhance the reliability and generalizability of the findings, future research should encompass larger sample sizes and multicenter studies, which can provide physicians with more precise and comprehensive information. Thirdly, while polysomnography (PSG) remains the gold standard for diagnosing obstructive sleep apnea (OSA), we justify the use of HSAT in this context based on following reasons. The enrolled patients were infertile PCOS patients from the reproductive center who urgently required fertility treatment. They showed no significant symptoms of OSA. HSAT was more convenient and cost-effective, with high patient compliance [ 63 ]. Furthermore, literature supports HSAT’s good sensitivity [ 64 ]. Patients diagnosed with moderate-to-severe OSA during screening were referred to respiratory specialists for further evaluation and treatment.

Polycystic ovary syndrome (PCOS) is the most prevalent endocrinopathy in women of childbearing age. It is characterized by menstrual disorders, hyperandrogenemia, metabolic dysfunction, and infertility, with a global prevalence of 7–15% [ 1 ] and 7.8% in China [ 2 ]. PCOS demonstrates a nuanced age-related risk profile, with peak diagnosis in early adulthood (20–30 years old) due to symptomatic manifestations like hyperandrogenism(HA) and oligo-anovulation [ 2 ], though metabolic and cardiovascular risks persist or escalate with advancing age [ 3 , 4 ]. PCOS is a leading cause of female infertility, contributing to 80% of cases of anovulatory infertility [ 5 ]. Clinically, PCOS presents with complex phenotypes, and numerous factors can exacerbate its symptoms, increase complications, and worsen prognosis. In recent years, it has been found that the prevalence of obstructive sleep apnea (OSA) is significantly higher among PCOS patients [ 6 ]. The 2023 International Evidence-Based Guideline for PCOS identifies OSA as one of its prominent features [ 7 ]. OSA　is a common sleep breathing disorder manifested by snoring with apnea and excessive daytime sleepiness. The prevalence of OSA in 2016 among the general population ranges from 9 to 38%, with lower rates in women (6–19%) compared to men [ 8 ]. OSA and PCOS share a number of common comorbidities [ 9 ], drawing increasing attention from experts in respiratory and reproductive health. Several studies have demonstrated a significantly higher prevalence of OSA in patients with PCOS compared to the general population, often with more severe symptoms [ 10 , 11 ]. In a previous study conducted at our center, we found a 40% prevalence of OSA in infertile PCOS patients, suggesting an association with reproductive and metabolic disorders [ 12 ]. Infertility is a significant concern for women with PCOS, and in-vitro fertilization (IVF) treatment plays a crucial role in improving reproductive outcomes for these patients [ 13 ]. PCOS-associated infertility arises from a combination of hormonal, metabolic, and ovarian microenvironment disruptions [ 14 ]. Recent research has revealed a connection between OSA and female infertility, with OSA being associated with preterm labor, miscarriage, and increased pregnancy complications [ 15 ]. Notably, a study by Zhang et al. demonstrated that OSA negatively influenced clinical pregnancy rate in a small sample of infertile PCOS patients undergoing IVF treatment [ 16 ]. However, the impacts of PCOS complicated with OSA (referred to as PCOS&OSA) on fertility and pregnancy outcomes remain understudied. This study aims to explore the relationship between PCOS&OSA and reproductive outcomes, contributing to the growing body of knowledge on the complex interplay between OSA and reproductive health. The findings of this study may help enhance clinical management and treatment strategies for PCOS patients undergoing IVF.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1