Increased risk of ASCVD in women with PCOS is only partially mediated by incident metabolic comorbidities

Abstract

Background/Synopsis

Women with polycystic ovary syndrome (PCOS) have a high prevalence of cardiovascular disease risk factors such as diabetes and dyslipidemia. Previous research suggests PCOS status may be associated with increased hazards of individual CVD events.

Objective/Purpose

To determine ASCVD risk associated with PCOS and address whether incident intermediate conditions including diabetes and hyperlipidemia mediate ASCVD risk.

Methods

We performed a retrospective cohort study using the nationwide claims database Optum Clinformatics® Data Mart (2000-2022). Women aged 18-50 with PCOS were identified through ICD9/10 codes for PCOS and matched 1:5 by age and visit month to women without PCOS. The primary outcome was composite ASCVD (coronary artery disease, angina, myocardial infarction, carotid artery disease, ischemic stroke, and transient ischemic attack). Causal mediation analyses were conducted for incident intermediate conditions (defined as hypertension, hyperlipidemia, obesity, and type 2 diabetes [T2DM]) developed after PCOS diagnosis to determine whether these conditions mediated the association between PCOS and composite ASCVD. Confounders of the exposure-outcome, exposure-mediator, and mediator-outcome associations were considered in the analysis, and interactions were also considered. Adjusted hazard was determined by controlling for race, ethnicity, education and history of smoking, infertility, depression, metabolic dysfunction associated steatotic liver disease, oral contraception and metformin use, and the other intermediate conditions which were not the outcome.

Results

420,756 unique patients with PCOS matched to 2,103,780 patients without PCOS. Mean age in both groups was 31.2 ± 7.1 years. The crude incidence per 1000 person-years of compositive ASCVD was 4.64 in women with PCOS compared to 0.79 in those without PCOS with an adjusted HR of 4.44, 95% CI 4.28-4.62. The incidence of each individual component of ASCVD was also increased. Those with PCOS had a significantly higher hazards for each of the incident intermediate conditions as well: hypertension: (aHR 1.47, 95% CI 1.45-1.48), T2DM: (aHR 2.22, 95% CI 2.19-2.25), hyperlipidemia: (aHR 1.35, 95% CI 1.33-1.36), obesity: (aHR 2.19, 95% CI 2.18-2.21). When controlling for the covariates, incident hypertension mediated 14.9% of the association between PCOS and ASCVD, T2DM mediated 9.6% of the association, hyperlipidemia mediated 12.0%, and obesity mediated 13.3% of the association. A combined mediator variable for any one of these conditions mediated 35.3% of the association.

Conclusions

In the largest longitudinal study of pre-menopausal women with PCOS residing in the United States, PCOS was independently associated with ASCVD and incident intermediate conditions partially mediated this risk. Our findings indicate that PCOS is a CVD risk enhancing condition.