Diurnal and stress reactive cortisol profiles are altered when adjusting for sex hormones

Abstract

Sex differences in diurnal and stress reactive cortisol profiles are presumed to depend on sex hormone variations. Surprisingly few studies, however, adjust for within-sex variations of testosterone, estradiol, and progesterone. In a study of 204 healthy adults, we assessed whether cortisol profiles would differ when adjusting or not for within-sex concentrations of sex hormones. Diurnal cortisol was sampled on two days at (1) awakening, (2) 30-min thereafter, (3) 14h00, (4) 16h00, and (5) bedtime. Reactive cortisol was sampled using 6 saliva samples measured every 10-min as part of the Trier Social Stress Test that generally affects men more than women. Sex hormones were collected for the diurnal protocol at baseline before the psychosocial stressor. Repeated-measures analysis of covariance controlled for key covariates in analyses unadjusted or adjusted for sex hormones. Results for diurnal cortisol showed that women had higher morning cortisol levels in adjusted but not unadjusted analyses. By contrast, men had higher reactive cortisol than women only in unadjusted analysis, but this was attenuated when adjusting for sex hormones and especially testosterone among both sexes. Collectively, our results reveal that adjusting for sex hormones alters cortisol profiles in a manner that may otherwise be obscured in psychoneuroendocrine studies that presume within-sex hormone status or rely on self-reports.