OR31-02 Diagnostic Criteria for Polycystic Ovary Syndrome in Adolescents: Impact on Prevalence and Longitudinal Body Mass Index Trajectories

Abstract

Background: Polycystic ovary syndrome (PCOS) is characterised by oligo-anovulation (OA), hyperandrogenism (HA) and polycystic ovary morphology (PCO). While the Rotterdam criteria (defined as 2 out of 3 features) are the most widely used criteria in adults, controversy remains for the diagnostic criteria in adolescents as many PCOS features overlap with normal pubertal physiological changes. The 2018 international evidence-based PCOS guideline recommends modified Rotterdam criteria (OA and HA) in adolescents based on expert consensus. We aimed to 1) compare the prevalence of PCOS using original and modified Rotterdam criteria in an unselected adolescent cohort and 2) explore the association between diagnostic phenotypes and long-term body mass index (BMI) trajectories. Methods: 227 adolescent females of the Western Australian Pregnancy Cohort (Raine) Study undertook detailed PCOS assessment at the mean age of 15.3 years (mean age of menarche 12.4 years). Detailed anthropometric measurements were collected from birth until age 22 years. T-test was used for group BMI comparisons and longitudinal BMI was analysed using Generalised Estimating Equations with PCOS by time and PCOS phenotypes by time as interaction terms. Results: PCOS was diagnosed in 66 (29.1%) participants using original Rotterdam criteria versus 37 (16.3%) participants using modified Rotterdam criteria. Using modified Rotterdam criteria, participants with PCOS had higher mean group BMI than participants without PCOS from age 5 years onwards. Significant interaction was detected between PCOS and time (p<0.001) on longitudinal BMI gain where higher BMI gain was observed in participants with PCOS from age 14 years onwards. Only the modified criteria phenotype was significantly associated wth long-term BMI gain whereas other PCOS phenotypes had similar BMI trajectories as participants without PCOS (p<0.001). Conclusions: Our findings validate the PCOS guideline recommendation as modified Rotterdam criteria reduce over-diagnosis of PCOS in adolescents and accurately identify the phenotype at risk of long-term weight gain. The BMI trajectories of females with and without PCOS diverge from early childhood suggesting that metabolic dysfunction in PCOS commences early in the pre-pubertal period. Disclosures: Nothing to disclose. Funding: PCOS CRE scholarship and Research Training Program Scholarship awarded to CT; NHMRC Medical Research Future fund awarded to HT; National Heart Foundation Future Leader Fellowship awarded to LM; NHMRC early career fellowship awarded to AJ.