Characteristics of Adolescents with Differing Polycystic Ovary Syndrome Phenotypes

Study Objective: To analyze clinical, metabolic, hormonal, and ultrasound characteristics of adolescents with polycystic ovary syndrome phenotypes. Design: We performed a retrospective analysis of quality improvement data. We divided patients according to phenotype on the basis of clinical or biochemical diagnosis of hyperandrogenism (HA), irregular menstruation (IM), and presence or absence of polycystic ovarian morphology (PCOM) on pelvic ultrasound (PUS) images, if obtained. The 5 resulting groups were: (1) HA/IM/normal PUS, n = 28; (2) HA/ PCOM, n = 10; (3) IM/PCOM, n = 18; (4) HA/IM/PCOM, n = 40; and (5) HA/IM/no PUS obtained, n = 80. We compared parameters between groups using the nonparametric Wilcoxon rank sum test. Setting: Boston Children's Hospital, 2012-2016. Participants: One hundred seventy-six girls and young women aged 11-25 years. Interventions: None. Main Outcome Measures: (1) Clinical, metabolic, and hormonal characteristics; and (2) PUS measurements. Results: Groups with HA had significantly higher acne scores, Ferriman-Gallwey scores, and total and free testosterone concentrations than groups without HA. Significant differences in hemoglobin A1c were found between the IM/PCOM and HA/IM/PCOM groups (5.1% vs 5.3%; P = .01) and the IM/PCOM and HA/IM/no PUS groups (5.1% vs 5.3%; P ! .01). In patients who had ultrasound performed, 49/94 (52.1%) met PCOM criteria on the basis of ovarian size, 37/94 (39.4%) on the basis of follicle number, and 27/94 (28.7%) on both; 10/94 (10.5)% had incidental findings on ultrasound, with 2 patients requiring further management. Conclusion: Limited differences in clinical, metabolic, and hormonal characteristics exist between adolescents with different phenotypes of polycystic ovary syndrome, and are mostly related to the presence or absence of HA. Of patients with ultrasound examinations, only 2 had clinically actionable incidental findings.