Racial and ethnic differences in the association of metabolic syndrome with prostate-specific antigen levels in U.S. Men: NHANES 2001-2006

Background: The interplay between metabolic syndrome (MetS) and race/ethnicity on prostate specific antigen (PSA) still remains unclear. The three major definitions of MetS (National Cholesterol Education Program [ATP III], International Diabetes Federation [IDF], and World Health Organization [WHO]) differ in number and type of components, which are linked to race/ethnicity. We proposed to investigate the associations of three major definitions of metabolic syndrome with PSA levels, and to determine whether these associations vary by race and ethnicity. Methods: We used measurements of serum PSA levels (ng/mL) in 3528 adult men > 40 years from the National Health and Nutrition Examination Survey 2001-2006. MetS was defined using ATP III, IDF, and WHO criteria. Racial/ethnic groups included non-Hispanic blacks (NHBs), non-Hispanic whites (NHWs), and Mexican Americans (MAs). We computed geometric mean PSA levels from weighted multiple linear regression models. Results: Men classified with MetS-ATP III and MetS-WHO had lower PSA concentrations relative to MetSfree counterparts, 0.77 versus 0.88 ng/mL (p= 0.01) and 0.79 vs 0.87 ng/mL (p =0.03), respectively. Stratifying these associations by race and ethnicity showed that non-Hispanic whites classified with MetS-ATP III (0.76 vs. 0.89 ng/mL) and MetS-WHO (0.80 vs. 0.88 ng/mL), and Mexican Americans classified with MetS-WHO (0.75 vs. 0.90 ng/mL) had lower PSA concentrations relative to MetS-free counterparts (p <0.01, p =0.04, and p =0.02, respectively). Conclusions: MetS was inversely associated with serum PSA levels and this association varied by definition of MetS and race/ethnicity. Fasting glucose of MetS-ATP III was the only component independently associated with PSA, suggesting that all components together have a greater influence on PSA levels than do components by themselves. Our findings highlight the implication when making inferences under the assumption that one definition of MetS fits all men without considering their race and ethnicity.