Depression and anxiety among patients undergoing dialysis and kidney transplantation: a cross-sectional study

ABSTRACT BACKGROUND: Depression and anxiety are the most prevalent psychological disorders among end-stage renal disease patients and are associated with various conditions that result in poorer health outcomes, e.g. reduced quality of life and survival. We aimed to investigate the prevalences of depression and anxiety among patients undergoing renal replacement therapy. DESIGN AND SETTING: Cross-sectional study in Belo Horizonte, Brazil. METHODS: Patients’ depression and anxiety levels were assessed using the Beck Inventory. The independent variables were the 36-Item Short-Form Health Survey (SF-36), Charlson Comorbidity Index and Global Subjective Assessment, along with sociodemographic and clinical characteristics. RESULTS: 205 patients were included. Depression and anxiety symptoms were detected in 41.7% and 32.3% of dialysis patients and 13.3% and 20.3% of transplantation patients, respectively. Lower SF-36 mental summary scores were associated with depression among transplantation patients (odds ratio, OR: 0.923; 95% confidence interval, CI: 0.85-0.99; P = 0.03) and dialysis patients (OR: 0.882; 95% CI: 0.83-0.93; P ≤ 0.001). Physical component summary was associated with depression among dialysis patients (OR: 0.906; 95% CI: 0.85-0.96; P = 0.001). Loss of vascular access (OR: 3.672; 95% CI: 1.05-12.78; P = 0.04), comorbidities (OR: 1.578; 95% CI: 1.09-2.27; P = 0.01) and poorer SF-36 mental (OR: 0.928; 95% CI: 0.88-0.97; P = 0.002) and physical (OR: 0.943; 95% CI: 0.89-0.99; P = 0.03) summary scores were associated with anxiety among dialysis patients. CONCLUSIONS: Depression and anxiety symptoms occurred more frequently among patients undergoing dialysis. Quality of life, comorbidities and loss of vascular access were associated factors.

Given the need for better understanding of affective disorders and associated factors in end-stage renal disease, the present study set out to 1) investigate the prevalence of depression and anxiety among patients undergoing different types of renal replacement therapy and 2) investigate the factors associated with the presence and severity of depression and anxiety symptoms. Kidney transplantation is believed to favor a better clinical condition and a daily routine that is more active and less dependent on the restrictions imposed by dialysis. Our hypothesis was that dialysis patients would present higher prevalence of depression and anxiety symptoms than a group of transplantation patients.

Study design
This investigation consisted of a cross-sectional follow-up study on participants in a cohort that had been established in 2006. The cohort included patients who were undergoing renal replacement therapy at 10 public dialysis services funded by the Brazilian public healthcare system in Belo Horizonte, Minas Gerais, Brazil 27 (Figure 1).

Participants
The initial cohort included all patients aged 18 years or over who started to undergo dialysis between January 1, 2006, and January 1, 2008, with a minimum of three months of treatment and no previous history of kidney transplantation. A total of 748 out of 2,386 patients met the selection criteria, and 36 of these patients refused to participate. Therefore, 712 patients formed the initial sample and were followed up over a non-concurrent or retrospective period (January 2006 to January 2008) and a concurrent or prospective period (January 2006 to May 2017).
The participants in the present cross-sectional study were patients who were undergoing dialysis or were surviving transplantation between February and May 2017, and whose physical condition and cognitive ability were sufficient for them to be able to complete the questionnaires. Patients who refused to participate, those who recovered their renal function and those who were referred for treatment elsewhere were excluded.

Ethical considerations
The Institutional Review Board (IRB) at the Federal University of Minas Gerais (UFMG) approved this study (under procedural no. 1.747.336), and all participants signed an informed consent form.

Measurements
Sociodemographic and clinical data were collected during structured interviews or were extracted from the medical records maintained by the participating units. Comorbidities were measured using the Charlson Comorbidity Index (CCI): this is a scoring system comprising 19 comorbidity items with assigned weights ranging from one to six, such that higher summed scores correspond to clinical conditions of greater severity. 28 Nutritional status was determined using the Subjective Global Assessment (SGA), which is a method that categorizes patients as well-nourished, suspected of being malnourished or severely malnourished, based on features of their physical examinations and clinical histories. 29 Additional groups of covariates were selected as follows: • sociodemographic: sex, age, ethnic group, marital status, religion, level of education, occupation and income. We defined alcohol use as consumption of five or more drinks (for males) or four or more drinks (for females) on a single occasion within the last 30 days.
• healthcare service (type of renal replacement therapy facility and travel time): for classification purposes, the facilities were grouped according to indicators of capacity to handle cases of increasing levels of complexity (graded from one to three), through a calculation using principal component analysis (PCA). The characteristics associated with the level of complexity with minimal variability between facilities included the type of service (outpatient or inpatient), teaching activities (yes or no) and kidney transplantation service availability (yes or no).
Depression and anxiety symptoms were assessed using the Beck Depression Inventory (BDI) and Beck Anxiety Inventory (BAI), validated for the Brazilian population. 30 The BDI and BAI are questionnaires consisting of 21 depression-related and 21 anxiety-related items to evaluate the presence and severity of symptoms over the course of the last week. Items are scored from 0 to 3, with a total summed score ranging from 0 to 63 points.
The scores indicating depression and anxiety are those greater than the cutoff of 11 and 10 points, respectively. Higher scores correspond to more severe symptoms, and the levels of depression and anxiety are graded as minimal or absent, mild, moderate or severe. The BAI and BDI tests were applied in accordance with Resolution 9 of the Brazilian Federal Psychology Council. 31 The quality-of-life assessment was based on the Portuguese version of the 36-item Short-Form Health Survey (SF-36). 32 This instrument comprises eight domains (physical functioning, role-physical, bodily pain, general health status, vitality, social functioning, role-emotional and mental health) and two summary measurements (physical and mental component summaries). Its scores can range from 0 to 100, and scores closer to 100 indicate better quality of life.
The patients were interviewed by trained health-related undergraduate students who were participating in the research project.
The interviews were conducted over the course of dialysis sessions (hemodialysis patients) or during follow-up visits (peritoneal dialysis and kidney transplantation patients).

Statistical analysis
Descriptive statistics were produced, based on frequencies for categorical variables, or on means ± standard deviations (SDs) for quantitative variables with normal distribution or medians for those with non-normal distribution. The non-paired Student t test and the Mann-Whitney test were used to compare normally and non-normally distributed quantitative variables, respectively. Pearson's chi-square test and Fisher's exact test were used to compare categorical variables. Risk factors for depression and anxiety and symptom severity were analyzed using age-and sexadjusted multivariate logistic regression models. For the logistic regression analysis, all variables that showed a significance level of 0.20 or lower were tested and only those with a significance level of 0.05 or lower were presented in the final model. Statistical analyses were performed using SPSS version 16.0.

Patient characteristics
This cross-sectional analysis included 205 patients: 130 of them were on dialysis and 75 of them had undergone transplantation.
A majority of the patients were male (52.7%), and many were married or in a de facto relationship (56.1%). The mean age was 54.5 years (SD = 12.7). In addition, many of the patients had not completed elementary education (58.2%), most did not have a job (78.5%) and most were living on some form of governmentprovided benefit (79.8%) (Tables 1 and 2

Characteristics of the patients with depressive symptoms
Univariate analysis revealed that most of the transplantation patients with depressive symptoms were women (P = 0.001), most were not married or in a de facto relationship (P = 0.02) and many had lower scores in the SF-36 domains of bodily pain (P = 0.001), social functioning (P = 0.02), role-emotional (P = 0.001) and mental health (P ≤ 0.001), and in the mental component summary (P ≤ 0.001) ( Table 1). Compared with nondepressive dialysis patients, depressed dialysis patients mostly had brown/black skin color (P = 0.01), presented more comorbidities, as shown by a higher CCI (P = 0.01), had had higher numbers of visits (P = 0.01) over the last 12 months, had shorter travel times to the healthcare service (P = 0.02) and had lower SF-36 scores ( Table 1). Continuous variables with normal distribution (Shapiro-Wilk normality test) were summarized using the mean ± standard deviation (SD) and were compared using the t test. For other quantitative variables, the median was used as a summary measurement, and the Mann-Whitney test was used for comparisons within the group.      Table 3).

Characteristics of the patients with anxiety symptoms
Univariate analysis revealed that most of the transplantation patients with anxiety symptoms were women (P = 0.01) with higher CCI (P = 0.01) and lower scores in the SF-36 domains of bodily pain (P = 0.006), role physical (P = 0.01), role emotional (P = 0.01) and mental health (P ≤ 0.001), and in the mental component summary (P = 0.004) ( Table 2).  Table 4).

Severity of depression and anxiety symptoms
Patients scoring higher than 20 points in the BDI and BAI were diagnosed as presenting moderate to severe depression or anxi-  Table 5).     The type of renal replacement therapy has been shown to be an important factor associated with mental health. In previous studies, transplantation patients were found to score lower than dialysis patients for both anxiety and depression. [34][35][36] The mental health of hemodialysis and peritoneal dialysis patients is thought to be impacted to a greater extent than that of transplantation patients because of the strict routine of dialysis sessions, along with the countless restrictions that limit these individuals' full participation in social, familial and productive activities. In contrast, kidney transplantation promotes greater wellbeing and freedom from dialysis and related restrictions and has a positive impact on self-perceived health. 1,[34][35][36][37][38][39][40] This phenomenon is particularly true for physical function- stressful life events, concurrent mental disorders and health status. 46 In the present study, greater severity of depressive symptoms was associated with worse general health status and poorer nutritional status.
The relationship between nutritional status and the severity of depressive symptoms needs to be appreciated from different perspectives. One potential explanation for this relationship is the negative impact of affective disorders on eating behavior.
However, these disorders may be concurrent with ongoing nutritional deficits and underlying disease progression. 47 The role of mental health in healthy behaviors also needs to be emphasized. This includes adequate food intake, since depression is known to interfere with eating habits and may lead to either increased or decreased appetite. 44 Additionally, depression has been positively correlated with undernourishment and poorer levels of hemoglobin, ferritin and albumin, in some end-stage renal disease studies. 47,48 The presence of bodily pain and less frequent participation in recreational activities were associated with greater severity of anxiety symptoms in the present study. Anxiety has been correlated with complaints of pain. Some studies have shown that patients with chronic pain had elevated levels of concern, tension and nervousness with regard to their illness and their general clinical condition, which influenced their perception of the painful experience. 49 On the other hand, states of pain, whether acute or chronic, favor psychological manifestations and become a factor in increasing the incidence of mood and anxiety disorders among these patients, compared with the general population. 49,50 Considering that chronic kidney disease increases the risk of having pathological conditions such as diabetes mellitus, neurological conditions, bone diseases and vascular diseases, patients undergoing renal replacement therapy are more likely to experience different types of pain of variable intensity and in a variety of locations. These patients' types of pain are associated not only with their pathological condition but also with the intercurrences and specificities of the renal treatment itself. 50 A cross-sectional study on 205 patients on dialysis showed that there was higher prevalence of mental disorders among patients with moderate or severe chronic pain than among those with mild or no pain. Severe irritability and anxiousness, and inability to cope with stress, were also more common among patients with pain than among those without pain. 51 Overall, chronic kidney disease patients participate less in recreational activities after they have started to undergo renal replacement therapies. 52 Although their reduced engagement in social activities may be partly due to their clinical status, the type of renal replacement therapy also needs to be considered, as shown by the lower scores among patients undergoing dialysis. 35 A systematic review of the literature that examined studies com- Although the data in our study were derived from a cross-sectional follow-up study on participants from another cohort that had been established in 2006, this work has made a contribution to the scarce scientific literature. Nonetheless, at the end of the follow-up, 507 patients were censored, and there were 449 deaths, representing 40.06% of the initial sample. A high mortality rate is expected among end-stage renal patients, especially in the first years of dialysis, because of several factors such as advanced age, diabetes mellitus, the underlying cause of chronic kidney disease and residence in cities with worse developmental rates. [52][53][54][55] Accordingly, the patients participating in our study were the ones who survived and therefore were in a better clinical and emotional condition.
Some limitations of the present study need to be considered. Firstly, simultaneous occurrence of end-stage renal disease and affective disorder symptoms needs to be considered. Coexistence of symptoms associated with both the uremic state and depressive mood, such as fatigue, reduced appetite, memory impairment and irritability, may occur.
Secondly, despite broad application and validation for end-stage renal disease patient populations, Beck's inventories have limitations that may interfere with making diagnoses of depression and anxiety, such as use of self-report questionnaires and inclusion of somatic symptoms that are not exclusive to emotional disorders.
Additionally, given that uremic parameters and graft function were not considered in the present study, associations between affective symptoms and actual renal function status could not be established.
Transplantation patients with impaired graft function usually have higher levels of depression or anxiety because they face the fear of having to start to undergo dialysis again.
Thirdly, the results of the present study were derived from a sub-cohort with a 10-year follow-up. Therefore, survival biases may not have been eliminated, given that the participants potentially reflected those with better health status. Among dialysis patients, long-term survivors may exhibit less evidence of depression or anxiety or may experience severely affected mental health when they have no options for transplantation.
Fourthly, the low number of patients in the renal transplantation group may have hindered possible detection of an association between anxiety and quality of life. This evaluation could not be made in the present study.
Lastly, given the cross-sectional study design, no causal links could be established, and the progression of depression and anxiety symptoms over time could not be measured. For these factors to be measured, longitudinal approaches are required.

CONCLUSION
This study revealed that depression and anxiety are common conditions among chronic kidney disease patients and that they occur more frequently among those undergoing dialysis than among those undergoing transplantation. Lower quality-of-life scores were associated with symptoms of depression in both types of renal replacement therapy. Presence of comorbidities, loss of vascular access and worse quality of life were associated with anxiety symptoms among dialysis patients, whereas none of these factors was associated with anxiety symptoms among transplantation patients. Treatment of affective disorders needs to be effectively included within the routine care provided for chronic kidney disease patients and should be maintained across the continuum of care. Further investigations are warranted to identify major risk factors and design better interventions for management, control and prevention.