RESUMO
The study aimed to examine the sex specific association of obesity with cortisol metabolism in a sample of older community dwelling people. The cross-sectional analysis included 394 men and 375 women (aged 65-90 years) of the population-based KORA-Age study. Multivariable regression analyses were employed to examine the association between cortisol samples (serum and salivary samples of morning after awakening (M1), 30 min later (M2) and at late night (LNSC)). Obesity was calculated as waist-to-hip ratio (WHR) and body mass index (BMI). Cortisol levels were not significantly different between obesity measures except for elevated serum cortisol (P = 0.02) levels in individuals with a low WHR. Higher M1 levels were especially apparent in women with normal BMI. Serum cortisol levels were inversely related to WHR (P = 0.004) and CARAUC was inversely associated with BMI (P = 0.007). Sex-stratified analytic models revealed that both obesity measures showed a non-linear association with cortisol diurnal pattern (M1/LNSC) in men. Impaired cortisol patterns emerged at both very ends of the body weight distribution. These findings do not support a cortisol driven obesity etiology in an older population and even point to an inverse association of body weight with cortisol levels. Differences of cortisol secretion patterns in generalized and abdominal fat distribution were marginal.