OBJECTIVE: To determine the risk of metabolic complications, primarily metabolic syndrome, in all polycystic ovary syndrome (PCOS) phenotypes compared with control subjects. DESIGN: Retrospective chart review. SETTING: University practice. PATIENT(S): Women with PCOS (Rotterdam definition; n = 258) and women without PCOS seen during the same time period for an annual exam used as controls (n = 110). INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Metabolic syndrome. RESULT(S): Three PCOS phenotypes had a significantly higher prevalence of metabolic syndrome compared with the control subjects: oligomenorrhea/oligo-ovulation (O) + hyperandrogenism (H) + polycystic ovaries (P), age-adjusted odds ratio [OR] 6.3 (95% confidence interval 2.1-18.9); O+H, OR 7.8 (2.2-27.5); and H+P, OR 8.2 (2.3-29.3). There was no significant difference in the prevalence of metabolic syndrome between women with O+P and control subjects, even in obese women. The prevalence of insulin resistance and glucose intolerance was not significantly different between POCS phenotypes CONCLUSION(S): The risk of metabolic syndrome may vary among the four phenotypes of PCOS based on the Rotterdam criteria. This new information may be of relevance in counseling women with PCOS although larger studies may be needed to validate our findings.
OBJECTIVE: To determine the risk of metabolic complications, primarily metabolic syndrome, in all polycystic ovary syndrome (PCOS) phenotypes compared with control subjects. DESIGN: Retrospective chart review. SETTING: University practice. PATIENT(S): Women with PCOS (Rotterdam definition; n = 258) and women without PCOS seen during the same time period for an annual exam used as controls (n = 110). INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Metabolic syndrome. RESULT(S): Three PCOS phenotypes had a significantly higher prevalence of metabolic syndrome compared with the control subjects: oligomenorrhea/oligo-ovulation (O) + hyperandrogenism (H) + polycystic ovaries (P), age-adjusted odds ratio [OR] 6.3 (95% confidence interval 2.1-18.9); O+H, OR 7.8 (2.2-27.5); and H+P, OR 8.2 (2.3-29.3). There was no significant difference in the prevalence of metabolic syndrome between women with O+P and control subjects, even in obesewomen. The prevalence of insulin resistance and glucose intolerance was not significantly different between POCS phenotypes CONCLUSION(S): The risk of metabolic syndrome may vary among the four phenotypes of PCOS based on the Rotterdam criteria. This new information may be of relevance in counseling women with PCOS although larger studies may be needed to validate our findings.