[Association between depression and death in the maintenance hemodialysis population: a cohort study].

Abstract

Objective: To identify factors associated with depressive symptoms in maintenance hemodialysis patients and to examine the relationship between these symptoms and mortality. Methods: Between January and December 2019, patients who received maintenance hemodialysis in the Blood Purification Center of the Second Affiliated Hospital of Nanjing Medical University were enrolled in a prospective cohort study. Depressive symptoms were assessed using the internationally validated patient health questionnaire-8 (PHQ-8). Sleep quality and anxiety were measured with the Pittsburgh sleep quality index (PSQI) and the generalized anxiety disorder-7 (GAD-7) scale, respectively. Follow-up continued until December 31, 2022, with all-cause mortality as the primary outcome. Ordinal logistic regression was used to identify independent predictors of depression severity. Cox proportional hazards models evaluated the association between depressive symptoms and mortality. Results: A total of 532 maintenance hemodialysis (MHD) patients completed the study. Among them, 177 (33.3%) exhibited depressive symptoms. Compared with patients without depression, those with mild or moderate-to-severe depression were older [median age: 58 (50, 66) vs. 60 (55, 65) vs. 55 (46, 64)], more likely to smoke [35.9% (51/142) vs. 40.0% (14/35) vs.26.2% (93/355)], had poorer sleep quality [PSQI: 9 (6, 13) vs. 12 (9, 17) vs. 5 (3, 9)], and higher anxiety levels [GAD-7: 1 (0, 3) vs. 3 (1, 6) vs. 0 (0, 1)], the differences among the three groups were statistically significant (all P<0.05). Ordinal logistic regression identified smoking status, history of diabetes or cardiovascular disease, hemoglobin level, PSQI score, and GAD-7 score as independent predictors of depression severity (OR=1.60, 1.80, 1.81, 0.98, 3.67, 8.67; all P<0.05). After a median follow-up of 40 (35, 44) months, 109 patients died, including 66 (60.6%) from cardio-cerebrovascular causes and 24 (22.0%) from infections. Kaplan-Meier analysis revealed significantly lower cumulative survival in the depression group compared to the non-depression group (P<0.001). Cox regression analysis demonstrated that depressive symptoms remained independently associated with all-cause mortality after adjusting for confounders (HR=1.06, 95%CI 1.00-1.13, P=0.048), with an even stronger association observed for patients with PHQ-8 scores≥2.9 (HR=1.10, 95%CI 1.03-1.16, P=0.005). However, the associations between depression and cardio-cerebrovascular mortality (P=0.111) or infection-related mortality (P=0.509) were not statistically significant. Conclusions: Depressive symptoms are prevalent among maintenance hemodialysis patients and are independently associated with increased all-cause mortality. Smoking, comorbid diabetes or cardiovascular disease, low hemoglobin level, poor sleep quality, and anxiety are risk factors contributing to depression. Maintenance hemodialysis patients with PHQ-8 scores≥3 should be considered at heightened risk for mortality.