Increased salivary cortisol nadir is associated with a low trabecular bone score and increased vertebral fractures prevalence in post-menopausal women in the population cohort osteolaus

In glucocorticoid (GC)-induced osteoporosis, fracture risk is poorly correlated with BMD. The bone texture analysis Trabecular Bone Score (TBS) correlates with micro-architecture (MA) parameters, and to fracture risk independently of BMD and of major clinical risk factors. Lower TBS (poor bone structure) but normal BMD are found in patients with high GC levels, even in subclinical hyper-cortisolism. Cortisol production and bone turnover circadian cycles are inversely correlated, with a time lag of 4h. Cortisol circadian rhythm changes with age, with a greater area under the curve due to an increase of the nadir levels on the 1s thalf of the night. Salivary cortisol follows the same modifications with aging as serum measures. It has been hypothesized that changes in cortisol circadian cycle with aging could participate to BMD loss. No study has addressed its role on MA alteration or fracture risk. We analyzed salivary cortisol circadian rhythm correlation to bone health parameters in the OsteoLaus cohort, a population-based cohort of 1500 randomly selected Caucasian women (50 to 80 y old) living in Lausanne. 754 women had bone parameters (BMD, TBS and VFA), as well as salivary cortisol circadian rhythm measures (wake-up, 30 min after wake-up, 11 am and 8 pm). They were split in tertiles of age, and in tertiles of salivary cortisol values at 8 pm. Comparison of bone parameters from younger/lowest and older/highest tertiles was done by applying a Student t-test. Ancova analysis was used when several variables were considered. Salivary cortisol concentration at 8 pm increased with age. Hip BMD and spine TBS decreased with age (no difference in spine BMD). Comparison between lowest versus highest tertiles of salivary cortisol concentration at 8 pm showed: a) Lower TBS values (1.30 vs 1.27; p= 0.001), but not spine or hip BMD, with highest 8 pm salivary cortisol concentration, independently of age, BMI or BMD; b) Higher number of prevalent vertebral fractures grades 2 and 3 ( 8 . 1% versus 1.7%; p< 0.003) in the tertile with both the highest concentration of salivary cortisol and the lowest TBS value. c) No difference was found for the BMD parameters. d) The increased prevalence of vertebral fractures was not related to an increase in the clinical risk as calculated using the FRAX® tool, as there was no difference in the risk of major osteoporosis fractures between the mentioned salivary cortisol tertiles (p=0.52). In conclusion, highest cortisol values at 8 pm were associated with altered microarchitecture as assessed by TBS, independently of age and BMI, but not with BMD decrease. Moreover, highest cortisol values and lowest TBS values were associated with an increased prevalence of vertebral fractures, without differences in the clinical risk calculated by FRAX®. TBS may be one of the main factors determining vertebral fracture risk. It is to be determined whether salivary cortisol at 8 pm is an independen risk factor.