Diurnal Cortisol Features and Type 2 Diabetes Risk in Patients With Hypertension and Obstructive Sleep Apnea: A Cohort Study

Abstract

Context: The hypothalamic–pituitary–adrenal (HPA) axis may be associated with type 2 diabetes (T2D); however, whether HPA axis dysfunction is associated with incident T2D remains unclear in patients with hypertension and obstructive sleep apnea (OSA). Objective: To investigate the relationship between the diurnal cortisol features and the risk of incident T2D in patients with hypertension and OSA. Methods: Participants with cortisol rhythm test at baseline in the Urumqi Research on Sleep Apnea and Hypertension cohort were enrolled. The Cox regression model was used to evaluate the relationship between ln-transformed diurnal cortisol features and T2D risk. Stratified and sensitivity analyses were also performed. Results: A total of 1478 patients with hypertension and OSA were enrolled in this study. During a median follow-up of 7.0 years, 196 participants developed T2D. Overall, a steep diurnal cortisol slope (DCS) was significantly associated with decreased T2D risk (per SD increase, HR 0.88, 95% CI 0.79-0.97, P = .014). Midnight cortisol was positively associated with increased T2D risk (per SD increase, HR 1.25, 95% CI 1.08-1.45, P = .003). Sensitivity analyses showed similar results. Neither DCS nor midnight cortisol was associated with incident T2D in the women’s subgroup or participants with mild OSA. Conclusion: Steeper DCS and higher midnight cortisol levels are associated with lower and higher T2D risks in patients with hypertension and OSA, respectively, at least in men or participants with moderate to severe OSA. Diurnal cortisol features may represent an early prevention target for diabetes in this population.