Metabolic and reproductive outcomes in lean versus overweight/obese phenotypes of polyendocrine metabolic ovarian syndrome: a retrospective cohort study with long-term follow-up

Polyendocrine metabolic ovarian syndrome (PMOS; formerly polycystic ovary syndrome [PCOS]) is the most common endocrinopathy in reproductive-age women. Whether lean (BMI < 25 kg/m²) and overweight/obese (BMI ≥ 25 kg/m²) phenotypes carry materially different long-term metabolic and reproductive risks remains incompletely defined, particularly in Chinese women. The objective of this study was to compare long-term metabolic and reproductive outcomes between lean and overweight/obese PMOS phenotypes in Chinese women. We conducted a retrospective cohort study of 1 500 women (500 lean; 1 000 overweight/obese) with Rotterdam-defined PMOS enrolled between December 2020 and May 2022 at a tertiary reproductive medicine centre and followed through December 2025 (median follow-up, 4.3 years). Dual primary endpoints were incident metabolic syndrome (MetS; Chinese Diabetes Society 2017 criteria) and cumulative live-birth rate (CLBR) in the assisted reproductive technology (ART) subgroup. Analyses used Cox proportional-hazards regression, generalised estimating equations (GEE), and patient-level logistic regression. Sensitivity analyses included propensity-score matching, overlap weighting, and alternative MetS definitions. At baseline, overweight/obese women had higher BMI (29.54 ± 3.04 vs. 22.23 ± 1.50 kg/m²), higher HOMA-IR (3.90 ± 1.50 vs. 2.20 ± 0.82), and higher prevalent MetS (31.8% vs. 0.2%). Incident MetS occurred in 19.94% vs. 9.22% of lean women (adjusted hazard ratio 2.33, 95% CI 1.66–3.27; P < 0.001); incident type 2 diabetes in 7.0% vs. 2.6% ( P = 0.001). Among 973 ART patients (1 432 cycles), CLBR was 52.84% vs. 70.49% (adjusted odds ratio [aOR] 0.46, 95% CI 0.34–0.62; P < 0.001); cycle-level GEE yielded an aOR of 0.56 (95% CI 0.44–0.70). Gestational diabetes (23.49% vs. 10.06%), hypertensive disorders of pregnancy (9.44% vs. 2.83%), and macrosomia (11.65% vs. 4.40%) were all more frequent in the overweight/obese group (all P ≤ 0.001). Subgroup estimates were consistent across age, Rotterdam phenotype, and baseline insulin-resistance status. Elevated BMI in PMOS was associated with a more than two-fold risk of incident MetS and an approximately halved CLBR after ART, alongside a substantially higher burden of pregnancy complications. Lean PMOS was not metabolically benign. These findings support BMI-stratified metabolic surveillance and structured pre-conception weight management in routine PMOS care. Not applicable.