Diabetes, Active Disease, and Afternoon Serum Cortisol Levels Predict Cushing's Disease Mortality: A Cohort Study

Abstract

Context: Cushing's disease (CD) is a life-threating disease, with increased mortality in comparison with the general population. Objective: This study aimed to evaluate standardized mortality ratios (SMRs) in CD patients. We also analyzed independent risk factors related to increased mortality. Design: We conducted a longitudinal cohort study in a 3rd level specialty center, from 1979 to 2018, in patients with CD. Results: From 1375 cases with a pathology diagnosis of pituitary adenoma, 191 cases had the confirmed diagnosis of CD (14%). A total of 172 patients completed follow-up, with a mean age at diagnosis of 33 ± 11 years, female predominance (n = 154, 89.5%), majority of them with microadenoma (n = 136, 79%), and a median follow-up of 7.5 years (2.4-15). Eighteen patients (10.5%) died, 8 of them (44%) were with active CD, 8 (44%) were under remission, and 2 (11%) were under disease control. Estimated all-cause SMR = 3.1, 95% confidence interval (CI) 1.9-4.8, P < 0.001. Cardiovascular disease was the main cause of death (SMR = 4.2, 1.5-9.3, P = 0.01). Multivariate Cox regression models adjusted for potential cofounders showed that diabetes (HR = 5.2, IC 95% 1.8-15.4, P = 0.002), high cortisol levels after 1600 hours at diagnosis (3.4, 2.3-7.0, P = 0.02), and active CD (7.5, 3.8-17.3, P = 0.003) significantly increased the risk of mortality. Conclusions: Main cause of CD mortality was cardiovascular disease. Main risk factors for mortality were uncontrolled diabetes, persistently high cortisol levels after 1600 hours at diagnosis, and active disease at last follow-up.