We performed a unique case-control study in which we assessed differences between Black and White women with PCOS in common reproductive and metabolic abnormalities in PCOS, as well as comparison of the differences between them and racially matched controls without PCOS. We found that Black women with PCOS are very similar reproductively to White women with PCOS, but that metabolically there were many factors that were more favorable in Blacks including lipid levels, serum transaminases, estimated VO2 max, bone density, fat distribution and a higher perceived overall quality of life. However results were mixed in that some parameters were not favorable in Blacks with PCOS such as increased fasting insulin levels. When we further examined the differences between cases and controls to see if PCOS is more severe in Blacks or Whites compared to a racially matched reference group, we found that Black women with PCOS did show some reproductive differences. These included lower estradiol levels, and there were also metabolic differences such as higher systolic blood pressure and lower glucose levels. Overall, our results are reassuring, that race per se in Black and White women with PCOS does not significantly influence the phenotype.
Our study overall is convincing that the full phenotype of PCOS appears very similar in Blacks and Whites, and against expectation, perhaps more favorable in Blacks. This is in line with a recent case-control study that showed lower triglyceride and higher HDL-C levels in Blacks with PCOS (2
). Some of our favorable findings such as the lower serum creatinine levels in Blacks may also be clinically insignificant. Other older studies suggest a worse phenotype in Blacks including a large randomized multicenter trial in women with PCOS, the Pregnancy in Polycystic Ovary Syndrome (PPCOS) trial. This trial noted that Black women at baseline tended to have a higher BMI and higher fasting glucose levels than Whites in the study (7
). Similar findings of increased BMI and diabetes prevalence in Blacks compared to Whites were noted in a study of women with PCOS from Boston and Iceland (8
). Within a large, community-based population of women with PCOS receiving health care in Northern California, Blacks were noted to have increased BMI and a higher prevalence of hypertension than Whites (9
). Like the PPCOS trial (7
), and unlike the other studies (2
), we excluded diabetes in all participants and found no differences in BMI between Black and White women with PCOS. These aforementioned studies also lacked a normal racial control group unlike our study.
The strengths of the study include the rigid inclusion and exclusion criteria to define both cases with PCOS and controls without it, the racial stratification of both cases and controls, and the thorough and consistent phenotyping of all the subjects. Limitations of our study include potential disparities between the two sites, which could not be assessed due to the design of our study with one site recruiting mainly Whites and another mainly Blacks. We were also unable to match control subjects on the basis of race, or to match control women with normal cycles to women with PCOS on the basis of weight or age, although we adjusted for this in our analyses. This is a common difficulty in such trials given the high prevalence of obesity in the U.S. PCOS population. Because these were women participating in other clinical studies, they may not be representative of the larger population who may be less likely to participate in drawn out demanding clinical trials For example, women with PCOS who are metabolically challenged may prefer pills (as in PPCOS) to lifestyle changes (as in our study). Finally we may have found more disparity between races if we had chosen the broader Rotterdam criteria for PCOS, as there is definitely more heterogeneity between the larger phenotypic spectrum these criteria capture.(20
We conclude that the racial disparities found in other conditions such as type 2 diabetes may not be as common in Black women with PCOS (and no diabetes), and they may also have, at least as young women, more favorable risk profiles for metabolic disease. This may ultimately be reassuring that PCOS diagnosed according to strict criteria may identify a comparable reproductive phenotype in women of different races. In the future, our findings suggests that Black and White women with PCOS could be pooled together for such baseline analyses, although further prospective studies should gauge the impact of race on treatment effects.