Depressive Symptoms and Mortality in Patients After Kidney Transplantation: A Prospective Prevalent Cohort Study MARTA NOVAK, MD, PHD, MIKLOS ZSOLT MOLNAR, MD, PHD, LILLA SZEIFERT, MD, AGNES ZSOFIA KOVACS, MD, ESZTER PANNA VAMOS, MD, PHD, REZSO ZOLLER, MD, ANDRAS KESZEI, MD, PHD, AND ISTVAN MUCSI, MD, PHD Objective: To analyze in a prospective cohort study if depressive symptoms are an independent predictor of mortality in kidney transplant recipients. Methods: Data from 840 transplanted patients followed at a single outpatient transplant center were analyzed. Sociodemographic parameters and clinical data were collected at enrollment (between August 2002 and February 2003). Participants completed the Center for Epidemiologic Studies-Depression (CES-D) scale. Depression was defined as CES-D score of ⱖ18. Data on 5-year outcomes (death censored graft loss or mortality) were collected. Results: The prevalence of depression was 22%. Mortality was higher (21% versus 13%; p ⫽ .004) in patients with versus without depression. In a multivariate Cox proportional hazard model, both the baseline CES-D score (hazard ratiofor each 1-point increase ⫽ 1.02; 95% confidence interval, 1.00 –1.04) and the presence of depression at baseline (hazard ratiopresence ⫽ 1.66; 95% confidence interval, 1.12–2.47) were significantly associated with mortality. The baseline CES-D score also significantly predicted death censored graft loss (hazard ratiofor each 1-point increase ⫽ 1.03; 95% confidence interval, 1.01–1.05). Conclusion: Depressive symptoms are an independent predictor of mortality in kidney transplanted patients. Key words: depression, mortality, kidney transplantation. CES-D ⫽ Center for Epidemiologic Studies-Depression scale; CKD ⫽ chronic kidney disease; CNI ⫽ calcineurin inhibitor; CRP ⫽ C-reactive protein; CsA ⫽ cyclosporine A; DOPPS ⫽ Dialysis Outcomes and Practice Patterns Study; GFR ⫽ estimated glomerular filtration rate; ESRD ⫽ end-stage renal disease; ESRDSI ⫽ End-Stage Renal Disease Severity Index; Hb ⫽ hemoglobin; HLA ⫽ human leukocyte antigen; HR ⫽ hazard ratio; IQR ⫽ interquartile range. INTRODUCTION epression is one of the most common psychiatric conditions in patients with end-stage renal disease (ESRD); its estimated prevalence, however, varies greatly across studies (1–5). In Phase I of the Dialysis Outcomes and Practice Patterns Study (DOPPS), depressive symptoms were assessed by two simple questions from the Kidney Disease Quality of Life-Short Form questionnaire. From those data, the authors (6) estimated the prevalence of depression around 20%. D From the Institute of Behavioral Sciences (M.N., M.Z.M., L.S., A.Z.K., R.Z., A.K., I.M.), Semmelweis University, Budapest, Hungary; Department of Psychiatry (M.N.), University Health Network, University of Toronto, Toronto, Ontario, Canada; 1st Department of Internal Medicine (M.N., E.P.V., I.M.), Semmelweis University, Budapest, Hungary, Department of Transplantation and Surgery (M.Z.M.), Semmelweis University, Budapest, Hungary; Department of Primary Care and Social Medicine (E.P.V.), Imperial College London, London, United Kingdom; and the Department of Epidemiology (A.K.), Maastricht University, Maastricht, Netherlands. Drs. Novak and Molnar contributed equally to the manuscript. Address correspondence and reprint requests to Marta Novak, MD, PhD, Department of Psychiatry, University Health Network, University of Toronto, EN 8 –212, Elizabeth Street, Toronto, Ontario, M5G 2C4 CANADA. E-mail: marta@nefros.net Received for publication May 25, 2009; revision received January 24, 2010. This study was supported, in part, by non-U.S. governmental Grants T-048767 and F-68841 from the National Research Fund (Hungarian Scientific Research Fund, OTKA), Grant 100/2006 from Committee for Health Research (ETT), the Hungarian Kidney Foundation, and the Foundation for Prevention in Medicine. Also, this paper was supported, in part, by the Janos Bolyai Research Scholarship of the Hungarian Academy of Sciences (M.N. and M.Z.M.). M.N. and M.Z.M. were recipients of the Hungarian State Eotvos Fellowship. M.N. has been supported by a grant from the Center for Integrative Mood Research, Toronto, Canada. The authors have not disclosed any potential conflicts of interest. DOI: 10.1097/PSY.0b013e3181dbbb7d Psychosomatic Medicine 72:527–534 (2010) 0033-3174/10/7206-0527 Copyright © 2010 by the American Psychosomatic Society Nearly ten thousand dialyzed patients were enrolled in the DOPPS II. In this cohort, depressive symptoms were assessed by the Center for Epidemiologic Studies-Depression (CES-D) scale, and the prevalence of depression estimated by this instrument was 43%, whereas the prevalence of physiciandiagnosed depression was only 13.9% (6). Several studies found increased mortality among depressed patients compared with their nondepressed counterparts among individuals with chronic medical conditions, such as coronary artery disease (7–9), cerebrovascular disease (10), Type 2 diabetes mellitus (11), cancers (12), and rheumatoid arthritis (13). Studies (1,2,4,14,15) investigating the association between depression and mortality in dialyzed populations yielded conflicting data that can be explained, at least in part, by the use of different diagnostic criteria and methodology. Devins et al. (16) and Christensen et al. (15), and a recent study by Wolf and Mori (17) did not find any association between psychosocial factors and survival. In a carefully designed longitudinal study, Kimmel et al. (1) demonstrated that depression as a time-varying variable was a significant predictor of mortality. The association was further confirmed by later studies (6,18,19), including the DOPPS. There are only little data available regarding the significance of depression in kidney transplanted (Tx) patients (5,20,21). One study (21) detected increased morbidity among depressed Tx patients compared with their nondepressed counterparts. Rocha et al. (20) found that Tx patients with moderate-to-severe depressive symptoms had an increased risk for negative outcome (chronic allograft nephropathy, death censored graft loss, or death) compared with nondepressed patients, but the sample size was rather small. More recently, Dobbels et al. (5), analyzing data from ⬎47,000 patients, reported that the diagnosis of depression as identified from Medicare claims was associated with a two-fold risk of death censored graft loss and death with a functioning graft. New information about the association between depressive symptoms and negative outcomes can further improve our understanding of the psychosomatic context of chronic kidney disease, and it also may point to potentially treatable factors 527 M. NOVAK et al. associated with clinical outcome. Such results may also help to formulate rationale for prospective treatment studies aimed at specific aspects of mental health concerns that patients with chronic kidney disease (CKD) may have. In this prospective, prevalent, cohort study, we wanted to determine if the severity of depressive symptoms and the presence of clinically potentially significant depression, identified by the self-reported CES-D questionnaire, are associated with increased mortality and death censored graft loss after renal transplantation. METHODS Sample of Patients and Data Collection This prevalent cohort of stable kidney Tx patients was selected by inviting all patients ⱖ18 years (n ⫽ 1067), who were regularly followed at a single kidney transplant outpatient clinic at the Department of Transplantation and Surgery at Semmelweis University, Budapest, to participate in our prospective cohort study. The baseline assessment was conducted between August 2002 and February 2003 (Transplantation and Quality of Life-Hungary Study). Our sample represents almost two thirds of the total Hungarian renal transplant population in 2002. The number of kidney transplantations at our transplant center has been stable—around 200 to 250 transplants/year over the last 5 to 7 years. The number of transplanted patients regularly followed at the center has increased from 1067 in 2002 to 1350 in 2007. Demographic data and details of medical history were collected at the time of study enrollment when information about age, gender, and etiology of CKD was obtained. Laboratory data were extracted from the patients’ charts and from the hospital’s electronic laboratory database. The following laboratory variables were tabulated: hemoglobin (Hb), serum creatinine, blood urea nitrogen, serum albumin, and serum C-reactive protein (CRP). Transplant-related data extracted from medical records included the following information: medication (including current immunosuppressive treatment), ESRD “vintage” (time elapsed since the initiation of the first treatment for ESRD), transplant “vintage” (time elapsed since the time of the transplantation). Patients completed a battery of validated questionnaires, including the CES-D scale, at the same time waiting for their regular follow-up visit at the transplant center. Estimated glomerular filtration rate (eGFR) was calculated, using the abbreviated Modification of Diet in Renal Disease study formula: eGFR (mL/min per 1.73 m2) ⫽ 186 ⫻ (SCr)⫺1.154 ⫻ (Age)⫺0.203 (⫻ 0.742 if female) (22), where SCr indicates serum creatinine. The study was approved by the Ethics Committee of the Semmelweis University. Before enrollment, all patients received detailed written and verbal information regarding the aims and protocol of the study and provided their informed consent. Self-Reported Comorbidity Information about the presence or absence of comorbid conditions (including diabetes) was obtained from the patients. Self-reported comorbidity score was calculated by summing the number of comorbid conditions the patients reported. Earlier works of our group suggested that this score correlates with mortality and provides valuable information about the overall clinical condition of the patients (23–25). Follow-Up Patients were followed from the day of the baseline visit to the time they returned to dialysis or died or until December 31, 2007. Patients were censored if they returned to dialysis or were lost to follow-up. Median, interquartile range (IQR) follow-up time, was 58 (7) months (follow-up range, 59 months). Data on outcomes were collected during the follow-up period by persons who were blinded to patients’ CES-D scores at baseline; the information was entered in our electronic database. The cause of death was extracted from the 528 charts. Two primary outcome measures were defined: 1) mortality with working graft; and 2) death censored graft loss—return to dialysis. Initiation of dialysis was based on a clinical decision made by the patient’s primary nephrologist. Assessment of Depression The Hungarian version of the CES-D scale has been prepared, according to a recommended procedure (26), and has been validated by our team in Hungarian hemodialysis and kidney transplanted patients (M. Novak et al., submitted). Internal consistency and test-retest reliability of the Hungarian version were good. Cronbach’s ␣ was 0.86 for dialysis and 0.89 for kidney transplanted patients, respectively. Test-retest correlation was 0.78. Furthermore, we found a very good fit between the original four factor structure of the CES-D scale (Depressed Affect, Positive Affect, Somatic Component, Interpersonal) and data obtained both in Hungarian hemodialysis and in Hungarian Tx patients. Finally, the CES-D score showed a moderate-strong correlation with other self-reported measures of emotional well-being/mental health. Based on these results, we suggest that the Hungarian CES-D scale is a reliable tool to measure depressive symptoms in different CKD populations. In this work, the total CES-D score was used to measure the severity of depressive symptoms. In addition, as secondary analyses, a cutoff score of 18 was used to obtain an estimate of the frequency of clinically significant depression, as suggested by Hedayati et al. (27) for patients with ESRD. Immunosuppressive Therapy Standard maintenance immunosuppressive therapy generally consisted of prednisolone, either cyclosporine A microemulsion formulation (Neoral) (CsA) or tacrolimus, combined with mycophenolate-mofetil or azathioprine or sirolimus. In certain analyses, patients were classified as calcineurin inhibitor users (CNI users) or calcineurin inhibitor nonusers (CNI nonusers) based on the use of any calcineurin inhibitor (CNI) (CsA or tacrolimus). Statistical Analysis Statistical analysis was carried out, using the SPSS 15.0 software (SPSS, Inc., Chicago, Illinois). Continuous variables were compared, using Student’s t test or the Mann-Whitney U test, and categorical variables were analyzed with the ␹2 test. To assess variables associated with the outcome measures, univariate and multivariate Cox proportional hazards analyses and KaplanMeier survival plots were used. To assess the independent association between depressive symptoms and outcomes, all variables known to be associated with mortality/death censored graft loss based on external evidence and clinical knowledge were included in the final multivariate models (5,6,24,28). The backward elimination strategy was used to identify covariates as a secondary analysis. As the number of deaths was relatively small and the number of potential covariables to adjust for was relatively high, we did the multivariate analysis in three steps. In the first model (Model 1), we adjusted for age and gender. In the next step (Model 2), we also adjusted for the number of self-reported comorbid conditions and ESRD “vintage.” In Model 3, all potentially important variables were included. Variance influence factors were used to indicate colinearity between independent variables. Proportional hazards assumptions were tested, using scaled Schoenfeld residuals. RESULTS Demographics and Baseline Characteristics of the Sample Data on depression were not available due to refusal or inappropriate completion of the questionnaire for 20% (n ⫽ 216) of the transplanted patients, and 11 more patients were lost during the follow-up period (nonparticipant group). The final sample analyzed, therefore, consisted of 840 Tx patients. There were no significant differences between participants and nonparticipants in age, gender distribution, presence of diabetes, number of self-reported comorbid conditions, eGFR, Psychosomatic Medicine 72:527–534 (2010) DEPRESSION AND MORTALITY IN KIDNEY TRANSPLANTATION TABLE 1. Characteristics of Patients With Versus Without CES-D Scores of >18 Male, % (n) Age, mean ⫾ SD (years) Marital status (married), % (n) Educational status (high school or higher), % (n) Estimated GFR, mean ⫾ SD (mL/min/1.73 m2) Diabetes, % (n) Number of comorbid conditions, median (IQR) Serum albumin, mean ⫾ SD (g/L) Hemoglobin, mean ⫾ SD (g/L) Serum CRP, median; interquartile range (mg/L) Total time with “end-stage renal disease,” mos, median (IQR) Azathioprine therapy, % (n) Mycophenolate mofetil therapy, % (n) Tacrolimus therapy, % (n) Steroid therapy, % (n) Cyclosporine therapy, % (n) Sirolimus therapy, % (n) Total Population At Baseline (n ⫽ 840) With CES-D ⱖ18 (n ⫽ 187) With CES-D ⬍18 (n ⫽ 653) p 59% (494) 49 ⫾ 13 66 (554) 54 (449) 50 ⫾ 22 17 (138) 2 (2) 42 ⫾ 3 132 ⫾ 19 2; 6 80 (71) 50% (94) 50 ⫾ 12 53 (99) 48 (89) 46 ⫾ 20 19 (34) 3 (3) 41 ⫾ 3 132 ⫾ 19 4; 8 78 (68) 61% (400) 48 ⫾ 13 70 (455) 55 (360) 51 ⫾ 22 16 (104) 2 (3) 42 ⫾ 3 132 ⫾ 19 2; 5 80 (70) .005 .08 ⬍.001 .04 .02 .26 ⬍.001 .04 .53 .11 .22 12 63 18 88 70 2 11 68 15 89 71 2 12 62 18 87 69 2 (98) (530) (149) (733) (583) (19) (20) (127) (29) (166) (133) (4) (78) (403) (120) (567) (450) (15) .37 .08 .21 .34 .34 .58 CES-D ⫽ Center for Epidemiologic Studies-Depression; SD ⫽ standard deviation; GFR ⫽ glomerular filtration rate; IQR ⫽ interquartile range; CRP ⫽ C-reactive protein. serum CRP, serum albumin, Hb, ESRD “vintage,” steroid use, and CNI use (not shown). At baseline, the mean ⫾ standard deviation age was 49 ⫾ 13 years; 59% of patients were male; and 17% suffered from diabetes mellitus among 840 stable kidney Tx patients. The mean ⫾ standard deviation calculated glomerular filtration rate was 50 ⫾ 22 mL/min/1.73 m2. Transplant “vintage” (median [IQR]) was 54 [63] months. Distribution of underlying kidney diseases in the participant group was the following: chronic glomerulonephritis was 18%, diabetic nephropathy was 15%, autosomal dominant polycystic kidney disease was 13%, chronic pyelonephritis/tubulointerstitial nephritis was 11%, hypertensive nephropathy was 8%, and other or unknown kidney disease was 35%. Seventy percent of the participants were taking CsA, 88% were on prednisolone and 63% were on mycophenolate-mofetil, 18% of the patients received tacrolimus and 12% were on azathioprine. Only 19 (2%) participant patients were given sirolimus. The median (IQR) CES-D score was 9 (11) and the prevalence of depression (CES-D score of ⱖ18) was 22%. Only 2% of patients with depression were taking antidepressant medications. The basic clinical and sociodemographic characteristics of patients with versus without CES-D scores of ⱖ18 are shown in Table 1. The proportion of males was significantly lower among depressed versus nondepressed patients. Marital status and level of education were also different between the depressed versus the nondepressed group. Patients with depression had significantly worse graft function, more self-reported comorbid conditions, and lower serum albumin (Table 1). A detailed analysis of the correlates of depression in Tx patients is published in a separate paper (29). The cohort was followed-up over a median of 58 months. Over 3521 person-years, 125 patients died (crude mortality Psychosomatic Medicine 72:527–534 (2010) rate, 28 of 1000 person-years) and over 3245 person-years of follow-up, 95 kidney grafts were lost (crude rate of death censored graft loss ⫽ 34 of 1000 person-years). During the follow-up, 23% of mortality was due to cardio- or cerebrovascular diseases, 27% to infections, 19% to malignant disorders, and 31% to other or unknown cause. We did not find any association between the cause of death and the presence of depression. Univariate Analyses of Transplanted Patients’ Mortality and Death Censored Graft Loss In univariate Cox proportional hazards models, we found a significant association between mortality and the CES-D score (hazard ratio [HR]for each 1 point increase ⫽ 1.03; 95% confidence interval [CI], 1.01–1.04). In addition, older age, male gender, the presence of diabetes mellitus, comorbidity, lower eGFR, lower serum albumin, higher serum CRP, and longer ESRD “vintage” were all significantly associated with mortality (not shown). The association between death censored graft loss and the CES-D score (HRfor each 1-point increase ⫽ 1.03; 95% CI, 1.01– 1.05) was also significant. Furthermore, the univariate Cox proportional hazards models revealed a significant association between death censored graft loss and age, initial eGFR, hemoglobin, ESRD “vintage,” and nonuse of CNI (not shown). Multivariate Analyses of Transplanted Patients’ Mortality and Death Censored Graft Loss To analyze the independent association between the CES-D score and mortality, three multivariate models were built (Table 2). In Model 1, we adjusted for age and gender. In the next step, we also adjusted for the number of self-reported 529 M. NOVAK et al. TABLE 2. Multivariate Cox Proportional Hazards Analyses of Patient Outcome Model 1 CES-D (for each 1-point increase) Depression (presence) Mortality HR 95% CI p HR 95% CI p 1.03 1.90 1.01–1.04 1.29–2.79 .001 .001 1.03 1.48 1.01–1.05 0.92–2.38 .006 .1 Model 2 CES-D (for each 1-point increase) Depression (presence) Mortality Death Censored Graft Loss HR 95% CI p HR 95% CI p 1.02 1.75 1.01–1.04 1.18–2.59 .004 .005 1.03 1.40 1.01–1.05 0.86–2.28 .01 .18 Model 3 CES-D (for each 1-point increase) Depression (presence) Death Censored Graft Loss Mortality Death Censored Graft Loss HR 95% CI p HR 95% CI p 1.02 1.66 1.00–1.04 1.12–2.47 .04 .01 1.03 1.43 1.01–1.05 0.87–2.33 .01 .15 Adjusted for: Model 1: age, gender; Model 2: Model 1 ⫹ number of self-reported comorbid conditions, total end-stage renal disease “vintage”; Model 3: Model 2 ⫹ estimated glomerular filtration rate, serum albumin, hemoglobin, serum C-reactive protein. HR ⫽ hazard ratio; CI ⫽ confidence interval; CES-D ⫽ Center for Epidemiologic Studies-Depression. comorbid conditions (including diabetes) and ESRD “vintage.” Finally, all potentially important variables associated with outcomes were included in Model 3. Accordingly, age, gender, the number of self-reported comorbid conditions, ESRD “vintage,” eGFR, serum albumin, Hb, and serum CRP were the covariables entered into the final Cox model, in addition to the CES-D score, respectively. In Model 3, the CES-D score (HRfor each 1-point increase ⫽ 1.02; 95% CI, 1.00 –1.04) was significantly associated with mortality. The CES-D score was also significantly associated with death censored graft loss (HRfor each 1-point increase ⫽ 1.03; 95% CI, 1.01–1.05). In the secondary models, where backward elimination strategy was used, the CES-D score was significantly associated both with mortality (HRfor each 1-point increase ⫽ 1.02; 95% CI, 1.00 –1.04) and death censored graft loss (HRfor each 1-point increase ⫽ 1.03; 95% CI, 1.01–1.05). The CES-D score remained a significant predictor of mortality, and death censored graft loss even if transplant “vintage,” instead of total ESRD “vintage,” was included in the models (not shown). The assumptions of proportional hazards were not violated in any of the multivariate analyses. Analyses Using the “Presence of Depression” (CES-D Score of >18) as Independent Variable In the study population, the frequency of negative outcomes was significantly higher in patients with versus without CES-D score of ⱖ18 (depression) (mortality: 21% versus 13%; p ⫽ .004; death censored graft loss: 14% versus 11%; p ⫽ .1). Patients without depression had significantly better survival chance than those with depression, as shown in the Kaplan-Meier survival plot (Fig. 1). A similar trend was seen for death censored graft loss (Fig. 2). 530 Figure 1. Kaplan-Meier survival plot: association between presence of depression and mortality, Log Rank: p ⫽ .004 In univariate Cox proportional hazards model, mortality and the presence of depression were significantly associated (CES-D score of ⱖ18) (HRpresence ⫽ 1.72; 95% CI, 1.18 – 2.51). A trend for increased risk of death censored graft loss in patients with depression was also seen (HRpresence ⫽ 1.46; 95% CI, 0.93–2.29), but this was not statistically significant. In the final multivariate Cox model (Model 3), the presence of depression at baseline significantly predicted mortality (HRpresence ⫽ 1.66; 95% CI, 1.12–2.47). The association between the presence of depression and death censored graft loss did not quite reach statistical significance (Table 2), although a trend for worse graft survival in depressed patients is suggested by the Kaplan-Meier plot (Fig. 2). When the multivarPsychosomatic Medicine 72:527–534 (2010) DEPRESSION AND MORTALITY IN KIDNEY TRANSPLANTATION Figure 2. Kaplan-Meier survival plot: association between presence of depression and graft failure, Log Rank: p ⫽ .1. iate analyses were repeated, using a CES-D cutoff score of 16, which is suggested for the general population, qualitatively similar results were obtained (not shown). We repeated the multivariate analyses, using backward elimination strategy. The presence of depression at baseline significantly predicted mortality (HRpresence ⫽ 1.63; 95% CI, 1.10 –2.42). The association between the presence of depression and death censored graft loss did not quite reach statistical significance (HRpresence ⫽ 1.49; 95% CI, 0.93–2.39). The assumptions of proportional hazards were not violated in any of the multivariate models. DISCUSSION We have found that depressive symptoms are independently associated with mortality and death censored graft loss—return to dialysis in a large sample of renal transplant recipients. Previous studies (7–12) reported increased mortality among depressed patients suffering from various chronic medical conditions as opposed to their nondepressed counterparts. Depression was also associated with mortality (30,31) and morbidity (21) in patients living with different transplanted organs. Most of the studies analyzing this question in dialysis patients also confirmed that depression was associated with negative outcome both in hemodialysis and peritoneal dialysis patients (1,6,32–37). Renal transplantation offers significant survival advantage compared with maintenance dialysis and also brings substantial emotional and psychological benefits to patients. At the same time, there are several factors that can potentially contribute to increased psychological distress in the Tx population (38,39). To date, only a few papers (5,21,38,40 – 43) assessed the presence and significance of depression in kidney Tx patients. The reported prevalence of mild-moderate or severe depression in those papers was about 22% to 25%; however, the sample size was usually small (40,42). In their recent Psychosomatic Medicine 72:527–534 (2010) paper, analyzing data from ⬎47,000 patients from the United States Renal Data System, Dobbels et al. (5) reported a cumulative prevalence of depression of 9% to 13% at 3 years post transplant, which is very similar to the prevalence of “physician-diagnosed depression” reported in the DOPPS study (6). These authors defined depression, using an algorithm to identify the code “depression not elsewhere classified” from Medicare claims. According to the authors, the prevalence of depression in their study (similar to the prevalence of physician-diagnosed depression in the DOPPS) may be less than the true incidence. We based our definition of depression on self-reported depressive symptoms, and we used a cutoff score validated for patients with CKD (27). Importantly, the prevalence of depression found in our patient group (22%) (29) was similar to other reports (40,42), which used similar methodology. To establish the independent association between the severity of depressive symptoms or the presence of clinically potentially significant depression and mortality or death censored graft loss, we built multivariate regression models. In these analyses, we considered several potentially important covariables known to be associated with poor outcome in renal transplant recipients (24,44,45). In these analyses, 1-point increase in the baseline CES-D score was associated with 2% increase in the hazard of mortality and with a 3% increase in the hazard of death censored graft loss over a 5-year follow-up period. Furthermore, the presence of depression at the baseline was associated with a 66% increase in the hazard of mortality over 5 years. Dobbels et al. (5) recently reported that depression, identified through Medicare claims, was associated with a two-fold increased risk for death censored graft loss and death in kidney transplant recipients. Our findings corroborate their findings and extend those at the same time. The fact that, in addition to the presence of depression, the severity of depressive symptoms as a continuous variable was significantly associated with outcome is a robust finding and suggests that it is not necessarily the clinical diagnosis of “major depression” but the whole continuum of depressed mood and distress that may be important. Several potential mechanisms may explain the association between depression and outcome (i.e., mortality and death censored graft loss). The first and the most plausible cause is nonadherence with treatment recommendations. Feelings of hopelessness, worthlessness, difficulties with concentration and memory, and loss of interest in daily activities may result in nonadherence behaviors that include forgetting to take pills or missing regular follow-up appointments (46 – 48). Another potential cause of death is suicide, which is relatively frequent in the dialyzed population (49) and has been reported in kidney transplant recipients as well (42,50). We did not have evidence, however, for suicide events in our patient group. Another mechanism that may link depression with negative outcome is the potential association between the presence of depressive symptoms and cardiovascular disorders. The association between psychological distress, particularly depression, and increased risk of cardiac diseases has been extensively reported 531 M. NOVAK et al. previously (8,51–55). Depression may result in loss of appetite and malnutrition. Friend et al. (56) showed that depression is associated with reduced serum albumin in dialyzed patients. Protein and energy malnutrition may also be associated with increased risk of mortality and death censored graft loss. There is also a complex interaction between depression and the immune system (57), and depression is associated with elevated levels of several proinflammatory cytokines (58 – 63). Chronic inflammation, in turn, is frequently linked to accelerated atherogenesis, cardiovascular events, and mortality (64,65). Further studies are needed to confirm these proposed hypothetical mechanisms linking depression and worse outcomes after kidney transplantation. Our work is notable for the large number of enrolled patients, long follow-up, and also for the extensive clinical and sociodemographic data collected. In the multivariate models, we adjusted for several important covariables. Several important limitations of this report should also be noted. It is possible that, in some patients, the declining overall condition (which was not captured by our comorbidity tool or declining serum albumin, for which we adjusted our models) before death could have led to increased depressive symptoms. In this study, 216 of 1067 patients refused or failed to complete the questionnaire properly, and 11 patients were lost during follow-up. These patients had similar sociodemographic characteristics compared with those who completed the questionnaire. We believe that the relatively high refusal rate does not invalidate our results. Depressive symptoms were assessed by a self-reported scale, which does not allow us to diagnose major depression. Depression is best diagnosed by an experienced clinician; the clinical diagnosis of depression cannot be based on questionnaires. Questionnaires, on the other hand, remain valuable tools to assess different aspects of a patient’s well-being, including depressive symptoms, in large-scale epidemiologic studies. We recently demonstrated that the Hungarian CES-D scale is a valid and reliable tool to measure depressive symptoms in hemodialyzed and kidney Tx patients (M. Novak et al., submitted). Furthermore, the cutoff score that we used to define clinically potentially significant depression has been validated by Hedayati et al. (27) against the Structured Clinical Interview for Depression in patients with CKD. The authors established that a cutoff score of 18 provided the best diagnostic accuracy (27). Using the same cutoff score in kidney Tx patients may provide a conservative estimate of the prevalence of depression. One important consideration, however, is that the population of the current study (Hungarian kidney transplant recipients) is potentially different from the U.S. CKD population enrolled in the validation study. When the multivariate analysis was repeated using a cutoff score of 16, which is suggested for the general population, qualitatively similar results were obtained (not shown). Finally, we suggest that in medically ill patients, such as the kidney Tx recipients, it is not only the firm clinical diagnosis of major depression but the whole continuum of depressed mood and 532 distress, feelings of hopelessness, worthlessness that may be important in determining important clinical outcomes, such as morbidity, mortality, and quality of life. Finally, we used a single baseline assessment of depressive symptoms. The psychological condition of patients may change during the follow-up, consequently, multiple measurements of these symptoms could have strengthened our results. Information about comorbid conditions was based on self-report of the patients. However, elements of the ESRDSeverity Index, a valid comorbidity questionnaire (66), were integrated into our tool. In an earlier cross-sectional analysis, the self-reported comorbidity score was also significantly correlated with several domains of the Short Form-36 Health Survey (quality of life) instrument in both dialyzed and in kidney Tx patients, and it was significantly correlated with serum albumin as well (67). Therefore, we suggest that this score provides valuable information about the patients’ overall clinical condition. Another limitation of this study is that patients from a single center were enrolled; therefore, our results cannot be generalized without further considerations. Finally, we did not have information on parameters (human leukocyte antigen mismatch, panel reactive antibodies, cold ischemic time, number of blood transfusion, acute rejection episodes, viral infections, smoking status, dyslipidemia) that may have influenced the results. Neither were additional risk factors, such as microalbuminuria-proteinuria, left ventricular hypertrophy, serum homocystein level, measured. We cannot rule out the possibility that including all those variables, or other yet unknown risk factors, would have altered our results. In summary, we showed that depressive symptoms are an independent predictor of patient survival and death censored graft loss in kidney Tx patients. We, therefore, believe that screening for depressive symptoms could be advised during the regular follow-up of kidney Tx patients. Patients who have CES-D scores above the suggested cutoff should be further evaluated by mental health professionals. Our results call for randomized studies assessing the effect of treating depression on various outcomes, including mortality and quality of life, in this patient population. We thank the patients and the staff of the Department of Transplantation and Surgery, Semmelweis University, Budapest, Hungary for their assistance in this survey. REFERENCES 1. Kimmel PL, Peterson RA, Weihs KL, Simmens SJ, Alleyne S, Cruz I, Veis JH. Multiple measurements of depression predict mortality in a longitudinal study of chronic hemodialysis outpatients. Kidney Int 2000; 57:2093– 8. 2. Christensen AJ, Ehlers SL. Psychological factors in end-stage renal disease: an emerging context for behavioral medicine research. J Consult Clin Psychol 2002;70:712–24. 3. Finkelstein FO, Finkelstein SH. Depression in chronic dialysis patients: assessment and treatment. Nephrol Dial Transplant 2000;15:1911–13. 4. Kimmel PL. Depression in patients with chronic renal disease: what we know and what we need to know. J Psychosom Res 2002;53:951– 6. 5. Dobbels F, Skeans MA, Snyder JJ, Tuomari AV, Maclean JR, Kasiske BL. Depressive disorder in renal transplantation: an analysis of Medicare claims. Am J Kidney Dis 2008;51:819 –28. Psychosomatic Medicine 72:527–534 (2010) DEPRESSION AND MORTALITY IN KIDNEY TRANSPLANTATION 6. Lopes AA, Bragg J, Young E, Goodkin D, Mapes D, Combe C, Piera L, Held P, Gillespie B, Port FK. Depression as a predictor of mortality and hospitalization among hemodialysis patients in the United States and Europe. Kidney Int 2002;62:199 –207. 7. Herrmann C, Brand-Driehorst S, Buss U, Ruger U. Effects of anxiety and depression on 5-year mortality in 5,057 patients referred for exercise testing. J Psychosom Res 2000;48:455– 62. 8. Barefoot JC, Brummett BH, Helms MJ, Mark DB, Siegler IC, Williams RB. Depressive symptoms and survival of patients with coronary artery disease. Psychosom Med 2000;62:790 –5. 9. Bush DE, Ziegelstein RC, Tayback M, Richter D, Stevens S, Zahalsky H, Fauerbach JA. Even minimal symptoms of depression increase mortality risk after acute myocardial infarction. Am J Cardiol 2001;88:337– 41. 10. Morris PL, Robinson RG, Andrzejewski P, Samuels J, Price TR. Association of depression with 10-year poststroke mortality. Am J Psychiatry 1993;150:124 –9. 11. Bruce DG, Davis WA, Starkstein SE, Davis TM. A prospective study of depression and mortality in patients with type 2 diabetes: the Fremantle Diabetes Study. Diabetologia 2005;48:2532–9. 12. Whooley MA, de Jonge P, Vittinghoff E, Otte C, Moos R, Carney RM, Ali S, Dowray S, Na B, Feldman MD, Schiller NB, Browner WS. Depressive symptoms, health behaviors, and risk of cardiovascular events in patients with coronary heart disease. JAMA 2008;300:2379 – 88. 13. Ang DC, Choi H, Kroenke K, Wolfe F. Comorbid depression is an independent risk factor for mortality in patients with rheumatoid arthritis. J Rheumatol 2005;32:1013–9. 14. Kimmel PL, Peterson RA, Weihs KL, Simmens SJ, Alleyne S, Cruz I, Veis JH. Psychosocial factors, behavioral compliance and survival in urban hemodialysis patients. Kidney Int 1998;54:245–54. 15. Christensen AJ, Wiebe JS, Smith TW, Turner CW. Predictors of survival among hemodialysis patients: effect of perceived family support. Health Psychol 1994;13:521–5. 16. Devins GM, Mann J, Mandin H, Paul LC, Hons RB, Burgess ED, Taub K, Schorr S, Letourneau PK, Buckle S. Psychosocial predictors of survival in end-stage renal disease. J Nerv Ment Dis 1990;178:127–33. 17. Wolf EJ, Mori DL. Avoidant coping as a predictor of mortality in veterans with end-stage renal disease. Health Psychol 2009;28:330 –7. 18. Diefenthaeler EC, Wagner MB, Poli-de-Figueiredo CE, Zimmermann PR, Saitovitch D. Is depression a risk factor for mortality in chronic hemodialysis patients? Rev Bras Psiquiatr 2008;30:99 –103. 19. Hedayati S, Abu Minhajuddin RT, Rush J. Depression predicts initiation of dialysis, hospitalization and death in chronic kidney disease patients. J Am Soc Nephrol 2008;19:769A. 20. Rocha G, Poli de Figueiredo CE, d’Avila D, Saitovitch D. Depressive symptoms and kidney transplant outcome. Transplant Proc 2001;33:3424. 21. Noohi S, Khaghani-Zadeh M, Javadipour M, Assari S, Najafi M, Ebrahiminia M, Pourfarziani V. Anxiety and depression are correlated with higher morbidity after kidney transplantation. Transplant Proc 2007;39: 1074 – 8. 22. Levey A, Greene T, Kusek J, Beck GJ. A simplified equation to predict glomerular filtration rate from serum creatinine. Abstr. J Am Soc Nephrol 2000;11:A0828. 23. Mucsi I, Molnar MZ, Ambrus C, Szeifert L, Kovacs AZ, Zoller R, Barotfi S, Remport A, Novak M. Restless legs syndrome, insomnia and quality of life in patients on maintenance dialysis. Nephrol Dial Transplant 2005;20:571–7. 24. Molnar MZ, Czira M, Ambrus C, Szeifert L, Szentkiralyi A, Beko G, Rosivall L, Remport A, Novak M, Mucsi I. Anemia is associated with mortality in kidney-transplanted patients—a prospective cohort study. Am J Transplant 2007;7:818 –24. 25. Molnar MZ, Novak M, Ambrus C, Kovacs A, Pap J, Remport A, Szeifert L, Mucsi I. Anemia in kidney transplanted patients. Clin Transplant 2005;19:825–33. 26. Novak M, Mah K, Molnar MZ, Ambrus C, Csepanyi G, Kovacs A, Vamos E, Zambo M, Zoller R, Mucsi I, Devins GM. Factor structure and reliability of the Hungarian version of the Illness Intrusiveness Scale: invariance across North American and Hungarian dialysis patients. J Psychosom Res 2005;58:103–10. 27. Hedayati SS, Bosworth HB, Kuchibhatla M, Kimmel PL, Szczech LA. The predictive value of self-report scales compared with physician diagnosis of depression in hemodialysis patients. Kidney Int 2006;69:1662– 8. 28. Rifkin DE, Katz R, Chonchol M, Fried LF, Cao J, de Boer IH, Siscovick DS, Shlipak MG, Sarnak MJ. Albuminuria, impaired kidney function and Psychosomatic Medicine 72:527–534 (2010) 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. cardiovascular outcomes or mortality in the elderly. Nephrol Dial Transplant [Epub ahead of print] Szeifert L, Molnar MZ, Ambrus C, Koczy AB, Kovacs AZ, Vamos EP, Keszei A, Mucsi I, Novak M. Symptoms of depression in kidney transplant recipients: a cross-sectional study. Am J Kidney Dis 55:132– 40. Prieto JM, Atala J, Blanch J, Carreras E, Rovira M, Cirera E, Espinal A, Gasto C. Role of depression as a predictor of mortality among cancer patients after stem-cell transplantation. J Clin Oncol 2005;23:6063–71. Zipfel S, Schneider A, Wild B, Lowe B, Junger J, Haass M, Sack FU, Bergmann G, Herzog W. Effect of depressive symptoms on survival after heart transplantation. Psychosom Med 2002;64:740 –7. Rocha LP, Peterson JC, Meyers B, Boutin-Foster C, Charlson ME, Jayasinghe N, Bruce ML. Incidence of posttraumatic stress disorder (PTSD) after myocardial infarction (MI) and predictors of ptsd symptoms post-MI—a brief report. Int J Psychiatry Med 2008;38:297–306. Wai L, Richmond J, Burton H, Lindsay RM. Influence of psychosocial factors on survival of home-dialysis patients. Lancet 1981;2:1155– 6. Burton HJ, Kline SA, Lindsay RM, Heidenheim AP. The relationship of depression to survival in chronic renal failure. Psychosom Med 1986;48: 261–9. Ziarnik JP, Freeman CW, Sherrard DJ, Calsyn DA. Psychological correlates of survival on renal dialysis. J Nerv Ment Dis 1977;164:210 –3. Shulman R, Price JD, Spinelli J. Biopsychosocial aspects of long-term survival on end-stage renal failure therapy. Psychol Med 1989;19: 945–54. Troidle L, Watnick S, Wuerth DB, Gorban-Brennan N, Kliger AS, Finkelstein FO. Depression and its association with peritonitis in longterm peritoneal dialysis patients. Am J Kidney Dis 2003;42:350 – 4. Baines LS, Joseph JT, Jindal RM. Emotional issues after kidney transplantation: a prospective psychotherapeutic study. Clin Transplant 2002;16:455– 60. Kimmel PL. Psychosocial factors in chronic kidney disease patients. Semin Dial 2005;18:71–2. Akman B, Ozdemir FN, Sezer S, Micozkadioglu H, Haberal M. Depression levels before and after renal transplantation. Transplant Proc 2004; 36:111–3. Christensen AJ, Ehlers SL, Raichle KA, Bertolatus JA, Lawton WJ. Predicting change in depression following renal transplantation: effect of patient coping preferences. Health Psychol 2000;19:348 –53. Soykan A, Arapaslan B, Kumbasar H. Suicidal behavior, satisfaction with life, and perceived social support in end-stage renal disease. Transplant Proc 2003;35:1290 –1. Karaminia R, Tavallaii SA, Lorgard-Dezfuli-Nejad M, Moghani Lankarani M, Hadavand Mirzaie H, Einollahi B, Firoozan A. Anxiety and depression: a comparison between renal transplant recipients and hemodialysis patients. Transplant Proc 2007;39:1082– 4. Jassal SV, Schaubel DE, Fenton SS. Predicting mortality after kidney transplantation: a clinical tool. Transpl Int 2005;18:1248 –57. Winkelmayer WC, Lorenz M, Kramar R, Horl WH, Sunder-Plassmann G. Percentage of hypochromic red blood cells is an independent risk factor for mortality in kidney transplant recipients. Am J Transplant 2004;4:2075– 81. DiMatteo MR, Lepper HS, Croghan TW. Depression is a risk factor for noncompliance with medical treatment: meta-analysis of the effects of anxiety and depression on patient adherence. Arch Intern Med 2000;160: 2101–7. Joseph JT, Baines LS, Morris MC, Jindal RM. Quality of life after kidney and pancreas transplantation: a review. Am J Kidney Dis 2003;42: 431– 45. Denhaerynck K, Dobbels F, Cleemput I, Desmyttere A, Schafer-Keller P, Schaub S, De Geest S. Prevalence, consequences, and determinants of nonadherence in adult renal transplant patients: a literature review. Transpl Int 2005;18:1121–33. Kimmel PL, Peterson RA. Depression in end-stage renal disease patients treated with hemodialysis: tools, correlates, outcomes, and needs. Semin Dial 2005;18:91–7. Washer GF, Schroter GP, Starzl TE, Weil R 3rd: Causes of death after kidney transplantation. JAMA 1983;250:49 –54. Dickens C, McGowan L, Percival C, Douglas J, Tomenson B, Cotter L, Heagerty A, Creed F. Association between depressive episode before first myocardial infarction and worse cardiac failure following infarction. Psychosomatics 2005;46:523– 8. Musselman DL, Evans DL, Nemeroff CB. The relationship of depression 533 M. NOVAK et al. 53. 54. 55. 56. 57. 58. 59. 534 to cardiovascular disease: epidemiology, biology, and treatment. Arch Gen Psychiatry 1998;55:580 –92. Glassman A. Depression and cardiovascular disease. Pharmacopsychiatry 2008;41:221–5. Lichtman JH, Bigger JT Jr, Blumenthal JA, Frasure-Smith N, Kaufmann PG, Lesperance F, Mark DB, Sheps DS, Taylor CB, Froelicher ES. Depression and coronary heart disease: recommendations for screening, referral, and treatment: a science advisory from the American Heart Association Prevention Committee of the Council on Cardiovascular Nursing, Council on Clinical Cardiology, Council on Epidemiology and Prevention, and Interdisciplinary Council on Quality of Care and Outcomes Research: endorsed by the American Psychiatric Association. Circulation 2008;118:1768 –75. Thombs BD, de Jonge P, Coyne JC, Whooley MA, Frasure-Smith N, Mitchell AJ, Zuidersma M, Eze-Nliam C, Lima BB, Smith CG, Soderlund K, Ziegelstein RC. Depression screening and patient outcomes in cardiovascular care: a systematic review. JAMA 2008;300:2161–71. Friend R, Hatchett L, Wadhwa NK, Suh H. Serum albumin and depression in end-stage renal disease. Adv Perit Dial 1997;13:155–7. Klapheke MM. Transplantation psychoneuroimmunology: building hypotheses. Med Hypotheses 2000;54:969 –78. Kalender B, Corapcioglu Ozdemir A, Koroglu G. Association of depression with markers of nutrition and inflammation in chronic kidney disease and end-stage renal disease. Nephron Clin Pract 2005;102:c115– c121. Empana JP, Sykes DH, Luc G, Juhan-Vague I, Arveiler D, Ferrieres J, Amouyel P, Bingham A, Montaye M, Ruidavets JB, Haas B, Evans A, Jouven X, Ducimetiere P. Contributions of depressive mood and circulating inflammatory markers to coronary heart disease in healthy Euro- 60. 61. 62. 63. 64. 65. 66. 67. pean men: the Prospective Epidemiological Study of Myocardial Infarction (PRIME). Circulation 2005;111:2299 –305. Musselman DL, Miller AH, Porter MR, Manatunga A, Gao F, Penna S, Pearce BD, Landry J, Glover S, McDaniel JS, Nemeroff CB. Higher than normal plasma interleukin-6 concentrations in cancer patients with depression: preliminary findings. Am J Psychiatry 2001;158:1252–7. Pace TW, Mletzko TC, Alagbe O, Musselman DL, Nemeroff CB, Miller AH, Heim CM. Increased stress-induced inflammatory responses in male patients with major depression and increased early life stress. Am J Psychiatry 2006;163:1630 –3. Miller AH, Maletic V, Raison CL. Inflammation and its discontents: the role of cytokines in the pathophysiology of major depression. Biol Psychiatry 2009;65:732– 41. Adler UC, Marques AH, Calil HM. Inflammatory aspects of depression. Inflamm Allergy Drug Targets 2008;7:19 –23. Violani C, Lucidi F, Devoto A, Lombardo C, De Santo RM. Insomnia and its comorbidities in chronic kidney disease. Semin Nephrol 2006;26: 61–3. Jylha M, Paavilainen P, Lehtimaki T, Goebeler S, Karhunen PJ, Hervonen A, Hurme M. Interleukin-1 receptor antagonist, interleukin-6, and C-reactive protein as predictors of mortality in nonagenarians: the vitality 90⫹ study. J Gerontol A Biol Sci Med Sci 2007;62:1016 –21. Craven J, Littlefield C, Rodin G, Murray M. The Endstage Renal Disease Severity Index (ESRD-SI). Psychol Med 1991;21:237– 43. Barotfi S, Molnar MZ, Almasi C, Kovacs AZ, Remport A, Szeifert L, Szentkiralyi A, Vamos E, Zoller R, Eremenco S, Novak M, Mucsi I. Validation of the Kidney Disease Quality of Life-Short Form questionnaire in kidney transplant patients. J Psychosom Res 2006;60: 495–504. Psychosomatic Medicine 72:527–534 (2010)