October 2010
Reviewer: Jeff C. Huffman, MD
Depressive symptoms and mortality in patients after kidney transplantation: a prospective prevalent cohort study
Novak M, Molnar MS, Szeifert L, et al
Psychosom Med 2010; 72(6):527-534
Background: There have been multiple studies looking at the links between depression and mortality in medical illnesses such as cardiovascular disease, cancer, and diabetes. However, there has been limited and mixed data about the connection between depression and mortality in patients with renal disease and almost no data about this link in patients undergoing renal transplantation. Such a question is important given the frequency of depression and substantial mortality rates in this cohort.
Methods: The population of known stable kidney transplant patients who were followed at a single transplant clinic in Budapest, Hungary were invited to participate in this prospective cohort study between August 2002 and February 2003. Enrolling subjects provided self-report information about demographics/medical illness, allowed access to their medical records for laboratory values and other information, and completed a self-report depression evaluation (CES-D). Subjects were followed until death, return to dialysis, or the end of the study (December 2007). Associations between depression were evaluated using multivariate models, and depression was considered as both a linear (CES-D score) and a dichotomous (CES-D≥18 = depression) variable in separate analyses.
Results: In total, 840 of the 1067 kidney transplant patients enrolled in and completed the study. The median CES-D score was 9 and 22% met CES-D criteria for depression. Of note, only 2% of patients were treated for depression. Subjects were followed for a median of 58 months, and 125 (14.9%) suffered mortality during the study period; 27% died from infection, 23% from cardiovascular/cerebrovascular causes, 19% from malignancies, and 31% due to other causes/unknown cause. Using depression as a linear variable, CES-D score was associated with mortality (and graft loss/return to dialysis), independent of multiple covariates including age, gender, duration of ESRD, and medical comorbidities. Using depression as a dichotomous variable (CES-D ≥18 = depression), depression was associated with greater mortality (21% vs. 13%; p=0.004) and this depression-mortality relationship remained after correction for multiple potential confounders.
Commentary: Add kidney transplant patients to the list of patient populations for whom depression appears to be an independent risk factor for death. This study appears to be a well-designed evaluation of a specific research question; they were able to get a substantial majority of all patients at the clinic to enroll, followed them for approximately 5 years, and performed careful statistical analysis that included multiple important sociodemographic, medical, and laboratory data (including C-reactive protein).
When discussing links between depression and mortality in medically-ill patients, questions around mechanism often arise. Physiologic mechanisms may play a major role perhaps particularly for those patients who died of cardiovascular causes—given the effects of depression on heart rate variability, platelet activity, inflammation, endothelial function and other physiologic process that are associated with increased risk of cardiac morbidity and mortality. Furthermore, depression has been associated with downregulating the cellular immune response, and thus may be connected with death by infection. On the other hand, it seems clear that the impact of depression on healthy behaviors—including adherence to medications and other disease-specific treatment recommendations—must play a substantial role in the link between depression and mortality in patients receiving a transplanted kidney.
The implications of this study are that depression is an important problem in kidney transplant patients, and further attention to screening and treatment of depressive disorders in this cohort is necessary. The good news is that screening for depression in medically-ill patients is easy (1-2 questions about depressed mood/anhedonia work well). Furthermore, standard treatments for depression should be effective in this cohort, keeping in mind the potential for drug-drug interactions with immunosuppressants via the cytochrome P450 3A4 isoenzyme (thus making SSRIs like citalopram or sertraline good first-line choices).