Polycystic Ovarian Morphology and Chronic Morbidity and Mortality in PCOS

Importance: Polycystic ovary syndrome (PCOS) is a prevalent endocrine disorder associated with insulin resistance and cardiovascular risk. The long-term association of ovarian morphology subtypes with long-term morbidity and mortality remains unclear. Objective: To determine whether long-term morbidity and mortality differ among women with PCOS with or without polycystic ovarian morphology (PCOM). Design, Setting, and Participants: This prospective cohort study was conducted from 1987 to 2005 with follow-up until 2024 (37 years) at a single academic referral center where women with PCOS consented for long-term health monitoring. Of 1089 initially enrolled women, 340 women with PCOS and sonographic data at baseline were included. Data were analyzed from January to April 2024. Exposure: Ovarian morphology was defined by the presence (PCOM) or absence (non-PCOM) of 12 or more peripherally distributed 2- to 9-mm follicles. Main Outcomes and Measures: All-cause mortality and chronic morbidity, including diabetes and cardiovascular, neurologic, thyroid, respiratory, gastrointestinal, kidney, autoimmune, psychiatric, and cancer conditions were assessed. Demographics, hormone levels, and cardiovascular risk factors were collected at enrollment. Outcomes were assessed at follow-up in older age, with data analyzed using χ2 tests, t tests, and multivariate logistic regression adjusted for confounders (baseline age, body mass index, fasting insulin, lipids, and follow-up duration). Results: Among 340 women with PCOS, 189 women had PCOM (mean [SD] age at enrollment, 28.03 [5.98] years) and 151 women did not (mean [SD] age at enrollment, 32.57 [9.21] years). Women with PCOM were younger at enrollment (P < .001), with higher mean (SD) body mass index (27.20 [5.74] vs 25.31 [6.31]; P = .004), cholesterol ratios (70.64 [51.76] vs 48.36 [51.84]; P < .001), and levels of luteinizing hormone (7.28 [8.60] mIU/mL vs 3.47 [4.39] mIU/mL; P < .001), androgens (eg, total testosterone: 90.78 [37.18] ng/dL vs 52.74 [76.66] ng/dL; P < .001), fasting insulin (12.84 [9.83] μIU/mL vs 8.85 [6.33] μIU/mL; P < .001), total cholesterol (167.57 [63.32] mg/dL vs 147.1 [82.24] mg/dL; P = .01), and triglycerides (97.35 [72.57] mg/dL vs 72.57 [61.06] mg/dL; P < .001). Mean (SD) follow-up duration was shorter in women with PCOM (33.7 [2.3] years vs 35.1 [4.9] years; P < .001). Mortality rates were similar, although mean (SD) age at death was younger in women with PCOM (54.3 [11.5] years vs 70.8 [13.6] years; P = .08). Non-insulin-dependent diabetes was significantly more common in women with PCOM (45 women [23.8%] vs 14 women [9.3%]; P < .001). Hypertension was not statistically significantly different in women with PCOM. Multivariate analysis confirmed the association of PCOM with non-insulin-dependent diabetes (aOR, 2.92; 95% CI, 1.06-2.82; P = .02). No other significant differences in chronic diseases were observed. Conclusions and Relevance: In this study, the PCOM pattern was associated with early metabolic abnormalities and increased risk of type 2 diabetes but not with increased mortality or other chronic diseases.