Objective: Meta-analyses indicate a 3X increased risk of gestational diabetes (GDM) among women with polycystic ovary syndrome (PCOS), however relationships between longitudinal trajectories of body mass index (BMI) and GDM risk in PCOS remain unclear. We aimed to identify BMI trajectory groups, compare GDM prevalence across trajectory groups and assess BMI trajectories impact on GDM risk.
Methods: This is a secondary analysis from the (Australian longitudinal Study on Women’s Health) ALSWH with 8,200 women aged 18-36 across five surveys (13 years). The main outcome measure was GDM prevalence. We used latent-class growth modelling to identify distinct BMI trajectories and logistic regression to assess GDM risk.
Results: 575 women (7.0%, 95% CI 6.5-7.6 %) reported PCOS. Among women with ≥1 live pregnancy, 15.1% developed GDM vs. 6.0% controls (p<0.001). Three distinct BMI trajectories were identified over the 13 year follow-up: low-stable (LSG) (63.4% women), moderately-rising (MRG) (29.2%) and high-rising (HRG) (7.5%). These were defined as: LSG-average trajectory remaining within healthy range; MRG-curvilinear trajectory commencing in healthy and terminating in overweight range and HRG-curvilinear trajectory starting and terminating in obese range. Women with PCOS were more likely to belong to MRG and HRG groups (OR 1.8, 95% CI 1.5-2.2 & OR 4.2, 95% CI 3.2-5.4). The GDM prevalence in PCOS differed significantly across trajectory groups (9.4% vs 20.0% vs 21.0%, p=0.02). After adjusting for BMI trajectories, age and demographic factors, women with PCOS were twice as likely to develop GDM compared to controls (OR 2.3, 95% CI 1.6-3.2).
Conclusion: Women with PCOS have higher rates of weight gain, yet PCOS remains an independent predictor of GDM irrespective of BMI trajectories over reproductive years. This aligns with the non-BMI dependent inherent insulin resistance in PCOS highlighting need for aggressive universal GDM screening in PCOS, independent of BMI and weight gain.