Abstract
Abstract Background Polycystic ovary syndrome (PCOS) is associated with adverse clinical outcomes that may differ according to PCOS phenotype. Methods Using UK Biobank data, we compared the incidence of type 2 diabetes (T2D), metabolic dysfunction associated steatotic liver disease, cardiovascular disease (CVD), hormone-dependent cancers, and dementia between PCOS participants and age- and body mass index-matched controls. We also compared multiorgan (liver, cardiac, and brain) magnetic resonance imaging (MRI) data and examined the impact of PCOS phenotype (hyperandrogenic and normoandrogenic) on these outcomes. Results We included 1008 women with PCOS (defined by diagnostic codes, self-reported diagnoses, or clinical/biochemical features of hyperandrogenism and a/oligoCmenorrhoea) and 5017 matched controls (5:1 ratio); median age, 61 years, body mass index, 28.4 kg/m². Adjusted Cox proportional hazard modeling demonstrated PCOS participants had greater incident T2D [hazard ratio (HR) 1.47; 95% confidence interval (CI), 1.11-1.95] and all-cause CVD (1.76; 1.35-2.30). No between-group differences existed for cancers or dementia. Liver MRI confirmed more PCOS participants had hepatic steatosis (proton density fat fraction >5.5%: 35.9 vs 23.9%; P = .02) and higher fibroinflammation (corrected T1 721.4 vs 701.5 ms; P = <.01) vs controls. No between-group difference existed for cardiac (biventricular/atrial structure and function) or brain (grey and white matter volumes) imaging. Normoandrogenic (but not hyperandrogenic) PCOS participants had greater incident all-cause CVD (1.82; 1.29-2.56) while hyperandrogenic (but not normoandrogenic) PCOS participants were more likely to have hepatic steatosis (8.96 vs 6.04 vs 5.23%; P = .03) with greater fibroinflammation (776.3 vs 707.7 vs 701.9 ms; P=<.01). Conclusion Cardiometabolic disease may be increased in PCOS patients with a disease phenotype-specific pattern. </p>