Serum androgen concentrations and subclinical measures of cardiovascular disease in men and women

•Large-scale cohort study with LC-MS/MS-based sex hormone measurements.•Total testosterone was positively associated with baseline IMT in women.•Androstenedione was associated with baseline fractional shortening in both genders.•SHBG was associated with prevalent and incident plaques in women.•Observed associations were sex-specific.

ObjectivesMost of the observed associations of androgens and sex hormone-binding globulin (SHBG) with subclinical cardiovascular disease (CVD) stem from selected study samples with immunoassay-based hormone measurements. Thus, we used a large population-based sample with total testosterone (TT) and androstenedione (ASD) concentrations measured by liquid chromatography-tandem mass spectrometry.DesignData of 2140 individuals (mean age: 60,8 years) from the cohort Study of Health in Pomerania were assessed at baseline and 5-year follow-up.MethodsMultivariable regression models were implemented to assess cross-sectional and longitudinal associations of TT, free testosterone (fT), ASD, SHBG and dehydroepiandrosterone-sulphate (DHEAS) with measures of subclinical CVD including intima media thickness (IMT), carotid plaques, left ventricular mass (LVM), fractional shortening (FS), relative wall thickness (RWT), and left ventricular geometry.ResultsCross-sectional analyses yielded an association of TT with IMT in women (β-coefficient per log unit increase: 0.02; 95% CI: 0.007; 0.45) and ASD with FS in both sexes (men: β-coefficient: −2.94; 95% CI: −4.75; −1.12; women: β-coefficient: 1.64; 95% CI: 0.55; 2.73). In longitudinal analyses, DHEAS was positively associated with FS change (β-coefficient: 2.34; 95% CI: -0.59; 4.08). In women, SHBG was positively associated with incident plaques (Q1 vs. Q3 (Ref.): β-coefficient: 1.35; 95% CI: 1.04; 1.74). In both sexes, longitudinal analyses showed no consistent association of TT with subclinical CVD.ConclusionsDespite several sex-specific associations of androgens and SHBG with subclinical CVD, the present representative study for the age group ≥45 years among men and women from the general population detected no consistent associations in longitudinal analyses.