Prevalence of anxiety and depression and its comorbidities in patients with chronic kidney disease on hemodialysis and peritoneal dialysis

Introduction: Patients with chronic kidney disease (CKD) who perform renal replacement therapy (RRT) are subject to a higher prevalence of mood disorders. Objective: The aim of this study is to compare the prevalence of anxiety and depression in patients on hemodialysis (HD) and peritoneal dialysis (PD), taking into account comorbidities that may contribute to this. Methods: The study was done in Ponta Grossa with CKD patients, using Beck Depression and Anxiety Inventory (BDI and BAI) and the Hospital Anxiety and Depression Scale (HADS). Results: We studied 155 patients, 128 in the HD group and 27 in PD. In the first, depression was found in 22.6% of patients in the BDI and 9.3% in HADS, and anxiety 25.7% in the BAI and 11.7% in the HADS. In the PD group, 29.6% of patients had depression in the BDI and 14.8% in HADS, and anxiety 11.1% in the BAI and none in HADS. Conclusion: The hemodialysis or peritoneal dialysis did not influence the prevalence of anxiety and depression in patients with CKD. AbstrAct


IntroductIon
Chronic kidney disease (CKD) is considered a public health problem worldwide.It is defined by kidney tissue injury (with or without a decrease in glomerular filtration rate) and/or a decrease in kidney function over a period of three or more months.When the glomerular filtration rate (GFR) is below 15 ml/min/1.73m 2 , the patient is in the terminal stage or dialysis, requiring renal replacement therapy (RRT), dialysis or transplant as alternative treatments. 1 According to the Brazilian Dialysis Census performed in 2010, the number of patients on dialysis has increased gradually from 42,695 in 2000 to 92,091 in 2010. 2 The estimated number of patients who started treatment in 2010 was 18,972.Of these, 90.6% were receiving hemodialysis (HD) and 9.4% were on peritoneal dialysis (PD).
Patients submitted to RRT are subject to reduced quality of life compared to the general population and a have higher prevalence of mood disorders.
The relationship between quality of life is inversely proportional to the prevalence of anxiety and depression, according to Muñoz. 3 This condition may represent an increase in morbidity and mortality in dialysis patients, as well as compromising treatment compliance and downgrading their immune and nutritional status, both because of symptoms of depression or anxiety as per associated symptoms -such as loss of concentration, loss of motivation, sleep disorders, fatigue, depressed mood and difficulty understanding information. 4,5everal questionnaires have been developed to assess anxiety and depression symptoms.Among them, some of the most used are Beck's Depression and Anxiety Inventories; and the Hospital Anxiety and Depression Scale (HADS), both validated, including patients with CKD. 6 The Beck Depression Inventory (BDI) was developed by Aaron T. Beck, and is performed to detect depressive symptoms and their intensity in patients aged over 13 years.The scale consists of 21 items, and the intensity of each one varies according to the degree of symptom severity, being rated from 0 to 3 (0 corresponds to mild or no symptoms; 3 corresponds to severe symptoms). 7The final score interpretation is given as follows: absence of depression between 0-13; mild depression between 14-19; moderate between 20-28; and severe depression between 29-63.BDI is simple to use; it can be used for various types of populations and its items correspond to the DSM-IV; therefore, it can be used both in research and in the clinical setting.Some limitations include the possible overlap between depression symptoms and other medical conditions, and the care that must be taken in performing diagnosis based only on questionaries. 8he Beck Anxiety Inventory (BAI) was developed to measure anxiety in adult patients.It can be used in different populations and is easy to deploy and interpret.Some limitations to its use are the lack of studies on the influence of other comorbidities, and patients with panic disorder also score high on the BAI.Thus, with the BDI, BAI has 21 items that assess symptom intensity from 0 (absent) to 3 (severe symptoms; almost unbearable). 9,10A score of 0-7 indicates minimum level of anxiety; 8-15 mild anxiety, 16-25 moderate and 26-63 severe anxiety.
The Hospital Anxiety and Depression Scale (HADS) was developed to identify potential new cases of anxiety and depression in adults.HADS differs from other scales because it contains items that address symptoms of anxiety and depression associated with physical illness (such as weight loss, insomnia, fatigue, headache and dizziness) to prevent interference from somatic disorders in scale scoring. 11It contains 14 items related to emotional and cognitive aspects of depression and anxiety, with seven items for each subscale.Each item is graded 0 to 3, indicating symptom intensity or frequency.The total score ranges from 0-42, and 0-21 for each subscale.The higher the score, the more severe the symptoms; 0-7 indicates the absence of significant symptoms; mild symptoms between 8-10; 11-15 and 16-21 for moderate and severe symptoms, respectively.
Although quality of life is affected both in HD and in PD, studies indicate that there are differences between them. 12-14HD patients have higher rates of depression compared to patients on PD, because the patient has to stay continuously connected to the machine during hemodialysis, restricting their daily activities and independence.In addition, suicide rates are higher among hemodialysis patients.PD patients have more autonomy, flexibility and control with fewer restrictions in their diet and social lives, which contributes to a better quality of life.
There are few studies comparing the prevalence of mood disorders among patients on HD and PD, associated to the comorbidities presented by them.The aim of this study is to compare the prevalence of anxiety and depression in two distinct groups: patients submitted to HD and PD, taking into account the presence of diabetes mellitus, CKD cause, dialysis time, prior transplant or not, registration into the transplantation queue, amaurosis, prior amputations and the use of beta-blockers and antidepressants.

Methods
This study was observational cross-sectional, held at the Renal Replacement Therapy unit of the Santa Casa de Misericordia hospital in Ponta Grossa, PR -Brazil, with all patients with CKD undergoing treatment at the time of the study.The project was approved by the Research Ethics Committee (COEP) from the State University of Ponta Grossa (UEPG) under number 05107812.5.0000.0105and the Research Evaluation Committee (COAP) from the Santa Casa de Misericordia hospital of Ponta Grossa.
155 patients were interviewed, 128 in the group of patients on hemodialysis and 27 on the peritoneal dialysis group.The researchers interviewed patients after signing the Consent Form, using the Beck Depression and Anxiety Inventory and the Hospital Scale of Anxiety and Depression.The questionnaires were used with all patients orally.Exclusion criteria were: known psychiatric disorders except for depression and anxiety, and refusal to participate.
Statistical analysis was carried out by comparing the mean scores between patients on hemodialysis and peritoneal dialysis using the Student's t-test, because all distributions were normal, and the chi-square test.Furthermore, multiple regressions were performed with each of the scores for anxiety and depression (BDI, BAI, HADS-Depression and HADS-Anxiety) and their correlation with the type of dialysis, age, gender, diabetes mellitus, duration of RRT, registration in the transplantation queue, use of antidepressants and beta-blockers.Results with p-value less than 0.05 (5%) were considered significant.results 155 patients were studied, 128 in the group of patients on HD and 27 on PD.In the HD group, the mean age was 54.96 ± 12.76 years; 70 (54.7%)were men; 37 (28.9%) had diabetes mellitus; 8 (6.3%) patients had undergone prior kidney transplantation and 47 (36.7%) were in the queue to receive transplants.No patient had previous amputations and 1 (0.8%) had amaurosis.With respect to medication, 6 (4.8%) patients used antidepressants and 15 (11.8%) were under beta-blockers.The median time to initiation of treatment with HD was 5.02 ± 3.86 years and the most prevalent cause of CKD was hypertension, in 52 (40.6%) patients.Depression of any intensity was found in 29 (22.6%)patients in the BDI and 12 (9.3%) in the HADS.Anxiety of any intensity was found in 33 (25.7%) patients in the BAI and 15 (11.7%) in the HADS.
In the PD group, the mean age was 56.48 ± 14.18 years; 12 (44.4%)patients were men, 13 (48.2%)had diabetes mellitus, 3 (11.1%)had undergone prior kidney transplantation and 8 (29.6%) were in the transplant queue.No patient had previous amputations or blindness and no patient used antidepressants or beta-blockers.The median time to treatment onset with peritoneal dialysis was 4.56 ± 2.65 years and the most prevalent cause of CKD was diabetes mellitus in 11 (40.7%)patients; followed by hypertension in 10 (37%) patients.Patient characteristics are summarized on Table 1.Depression of any intensity was found in 8 (29.6%) patients according to the BDI and in 4 (14.8%) in the HADS.Anxiety of any intensity was found in 3 (11.1%)patients in the BAI and none when the HADS was used.The prevalence of anxiety and depression is summarized on Table 2, and the mean scores of the questionnaires are summarized on Table 3.
The data that showed statistical significance with anxiety, in both questionnaires, were its negative correlation with age and, with the BAI it was the positive correlation with the use of antidepressants.As far as depression is concerned, the presence of diabetes mellitus had a positive correlation in both the BDI and the HADS, and the use of beta-blockers had negative correlation with the BDI.Gender, type of dialysis, RRT duration time and enrollment in the kidney transplantation queue showed no statistically significant relationship with anxiety or depression.The correlation coefficients between the risk factors evaluated and the scores of BDI, BAI and HADS are depicted on Table 4.

dIscussIon
][17]   However, according to the HADS, results showed conflicting data with those in the literature -with lower prevalence of both anxiety and depression.The items evaluated for each questionnaire are distinct, and the BDI assesses a wider variety of symptoms, including symptoms related to physical illness, such as weight loss, insomnia and fatigue.
HADS can be used as a screening test, but the diagnosis should be made based on clinical criteria, according to the DSM-IV TR, for the institution of effective treatment and improved quality of life for the patient. 1814]19 Both according to the mean value and by multiple regression analysis, the difference in prevalence between HD and PD was not significant.Depression levels found were higher in PD patients, while anxiety levels were higher in HD patients, but without statistical significance.Higher levels of anxiety in patients who received HD can be explained because they need to stay connected to the machine for several hours a week, restricting their independence and autonomy.Moreover, they are subjected to the stress of hospital visits every two or three days, transportation to the hospital, having to share time with other patients, restricted diet and inability to make long trips.These factors could contribute to a higher prevalence of depression in these patients, but our study showed no such result.Patient age was negatively correlated with anxiety scores, suggesting that older patients have a lower prevalence of anxiety.Studies show that the older the patient, the higher the prevalence of somatic symptoms, decreased quality of life, restrictions in social life and higher depression rates. 19,20Corroborating this study, Bayat et al. 21found no correlation between depression and patient age, as well as gender.However, Theofilou 19 showed that women have poorer mental health, with a higher prevalence of somatic symptoms and social dysfunction relative to men among HD patients, and higher rates of anxiety among PD patients.Their study suggests that men and women may differ in specific aspects of the questionnaires, such as suicidal ideal for instance, which could explain the difference between the scores found in different types of questionnaires that assess various clinical aspects of the same disease.In our study, patient gender had no significant correlation with the prevalence of anxiety or depression.
Diabetes mellitus was positively correlated with BDI and HADS depression scales.6][17] Depression is associated with hyperglycemia and an increased risk for complications from diabetes, which may explain this finding. 22This relationship is very important, and many patients with CKD suffer from diabetes.
RRT duration had no significant correlation with anxiety or depression.Cukor et al. 23 suggest that depression and anxiety run different courses in HD patients.Patients who remained depressed after 16 months of follow-up showed a decrease in quality of life and higher levels of depression.These patients fell into three patterns of disease: some patients had not been diagnosed with depression and had mild symptoms after followup; other patients had intermittent symptoms of depression and showed moderate levels in the second stage, and the third group remained with severe symptoms of depression.Anxiety does not follow this pattern, with no significant differences among patients with chronic intermittent anxiety or chronic depression after follow-up.However; the prevalence of anxiety associated with depression was higher after 16 months of follow up.A limitation of our study is to have the patient follow-up to diagnose possible progression or remissions in symptoms of anxiety and depression over time.Ginieri-Coccossis et al. 13 found a reduction of mental health, social relationships and quality of life in patients undergoing HD for more than four years, but this relationship was not present in PD patients.
Interesting findings were related to the medication studied.Beta-blocker use was negatively correlated with the BDI, suggesting that it would have a protective effect against depression.
Beta-blockers, especially the fat-soluble ones, have been associated with depression since the late 60's. 24Lipid solubility determines the degree of beta-blocker penetration on the blood-brain barrier, leading to possible side effects on the central nervous system such as depression, lethargy, nightmares and confusion.Propranolol is very soluble, whereas metoprolol has moderate lipid solubility.Water-soluble drugs, such as atenolol, have a longer half-life and cause fewer CNS side effects. 25Furthermore, pindolol has been used as an antidepressant enhancer with primary action on serotonergic receptors, but the studies are inconclusive. 26A limitation of our study was the non-selection of the type of beta-blocker used, making it impossible to have a more thorough analysis on which specific drug could have a protective effect on depression.
The use of antidepressants was positively correlated with the BAI, suggesting that patients who use antidepressants have a higher prevalence of anxiety, or that its use can increase anxiety.Anxiety disorders are responsive to various types of antidepressants, especially selective serotonin reuptake inhibitors and reuptake inhibitors of serotonin and norepinephrine, including being recommended their continued long-term use in patients who responded to medical therapy in an acute fashion. 27However, even with first-line drugs, only one third of patients have remission of their anxiety. 28,29The present study did not assess the type of antidepressant used, use onset and whether the patient had therapeutic response, which may contribute to the result that patients who use antidepressants have shown a higher prevalence of anxiety.
In addition to the aforementioned limitations, we can also add the small number of patients responding to the questionnaires, because the study was performed in a RRT unit only, with a limited number of patients.More studies are needed to correlate the probable risk factors for the development of anxiety and depression in patients with CKD under different forms of RRT.

conclusIon
Depression and anxiety disorders are highly prevalent mood disorders among patients undergoing RRT, so they should be properly diagnosed and treated, to improve the quality of life of patients with CKD.We found no statistically significant difference between the dialysis modalities performed (HD or SD).Age, diabetes mellitus and the use of antidepressants or beta-blockers appear to influence the prevalence of anxiety and depression.

tAble 1
Baseline characteristics Of the patients assessed * Mean (standard deviation); Student t-test; Chi-square test.* Chi-square test.

tAble 3
Bai, Bdi and hads scOre mean values in patients under hemOdialysis and peritOneal dialysis*

tAble 4
cOrrelatiOn cOefficient Between the risk factOrs assessed and the Beck and hads scOres