Menstrual dysfunction in PCOS: primarily linked to hyperinsulinemia over dysglycemia

Objective While insulin resistance (IR) underlies reduced ovulation in polycystic ovary syndrome (PCOS), the relative contributions of hyperinsulinemia versus dysglycemia to anovulatory dysfunction remain unclear. This study quantitatively assessed the relationship between menstrual disturbance and both insulin and glucose parameters in women with PCOS. Methods This retrospective cross-sectional study included 408 women diagnosed with PCOS. Participants were categorized into three groups based on self-reported menstrual cycle length: Eumenorrhea (26–34 days), Oligomenorrhea (35–90 days), and Amenorrhea (> 90 days). All participants underwent a standardized 75g oral glucose tolerance test (OGTT). Fasting plasma glucose (FPG) and insulin (FINS), along with 2-hour post-load glucose (2hPG) and insulin (2hINS) levels were measured. Insulin resistance was assessed using the Homeostatic Model Assessment (HOMA-IR). Statistical analyses compared metabolic parameters across groups using ANOVA/chi-square tests and assessed relationships using multivariable linear regression. Results Increasing menstrual cycle length was significantly associated with elevated body mass index (BMI), waist-to-hip ratio (WHR), FPG, 2hPG, FINS, 2hINS, alanine aminotransferase (ALT), low-density lipoprotein cholesterol (LDL-C), androstenedione, luteinizing hormone/follicle-stimulating hormone ratio (LH/FSH), LH, prolactin, and free androgen index (FAI) (all P < 0.05). Conversely, sex hormone-binding globulin (SHBG) and high-density lipoprotein cholesterol (HDL-C) levels decreased significantly with greater menstrual disturbance (all P < 0.05). The prevalence of insulin resistance (IR) was significantly higher in women with amenorrhea compared to those with eumenorrhea (79.9% vs. 44.6%; P < 0.001), demonstrating a progressive increase in IR risk with worsening menstrual dysfunction (Ptrend < 0.001). Following adjustment for BMI and WHR, the amenorrhea group demonstrated persistently elevated FINS (β = 2.55, 95% CI 0.47 to 4.63; P = 0.017) and 2hINS (β = 37.24, 95% CI 8.13 to 66.35; P = 0.013). No statistically significant differences in FPG or 2hPG levels were observed between the groups. Conclusion In women with PCOS, menstrual disturbance severity is independently associated with hyperinsulinemia, not dysglycemia. These findings suggest that hyperinsulinemia, rather than glucose levels, represents a key biomarker determining the severity of menstrual dysfunction in PCOS.