The role of adipocytokines on depressive symptoms of patients with chronic kidney disease

Abstract

Objective: The aim of this study is to evaluate depression and anxiety scores among chronic kidney disease (CKD) patients and to search the changes of serum concentrations of adipokines with respect to emotional disturbances of CKD patients. Patients and methods: 150 patients recruited into this cross-sectional analytic study. Study groups were control, hemodialysis, predialysis, peritoneal dialysis and kidney transplantation groups. Fasting morning serum leptin, ghrelin, acylated ghrelin, neuropeptide Y, adiponectin, resistin levels of all of the groups were measured using ELISA (Sandwich) method. A screening interview based on the Structured Clinical Interview for DSM-IV and self-report scales (The Beck Depression [BDI] and The Beck Anxiety Inventory [BAI] and Brief Symptom Inventory [BSI] which is self report scales) were administered and conducted by a trained interviewer. Results: BDI scores were significantly higher in hemodialysis and predialysis groups compared to control group (p = 0.009). Somatization sub scores of BSI were significantly higher in hemodialysis and peritoneal dialysis groups compared to control group (p = 0.041). Also positive symptom distress index scores of BSI were significantly higher in hemodialysis and transplantation groups compared to control group (p = 0.047). BDI score were significantly negatively correlated with duration of education (r = −0.165, p = 0.045), positively correlated with presence of protein energy wasting (r = 0.198, p = 0.016), and resistin levels (r = 0.233, p = 0.004). Conclusion: CKD patients had higher BDI, BSI-somatization, BSI-positive symptom distress index scores compared to control group. High serum resistin levels, presence of protein energy wasting might have a role in development of depressive disorders of patients with chronic kidney disease.

• Adiponectin
• anxiety disorders
• chronic kidney disease
• depression
• ghrelin
• leptin
• resistin

Introduction

Depressive disorders are 1.5–4 times more prevalent in medically ill patients than in the general populations. This kind of disorders develops from the interaction among multiple pathophysiological, psychological, and socioeconomic stressors that chronic illness imposes on the individual. Symptoms of clinical depression affect approximately 25% patients on hemodialysis, 40% of patients on peritoneal dialysis, 37% of chronic kidney disease patients not requiring dialysis. It can be associated with low quality of life and increased mortality and more rapid glomerular filtration rate decrements. Depression is also a risk factor for poor outcomes after renal transplantation.1–3

Adipocytes secrete diverse bioactive substances, including peptide hormones, cytokines, growth and complement factors, which perform essential regulatory functions related to energy balance, satiety and immunity, glucose and lipid metabolism, reproduction, cardiovascular functions. These substances are named as adipokines. Leptin, resistin, adiponectin are some of adipokines which act in a autocrine, endocrine, and paracrine way to influence several biological functions. For instance, leptin, an anorexigenic adipokine, signals satiety. On the other hand, neuropeptide Y, produced by hypothalamus and adipose tissue, together with ghrelin, produced by stomach and involved in short term regulation of feeding and long term regulation of energy metabolism, are both orexigenic hormones.4–9 In addition to hypothalamic satiety center, receptors for leptin, ghrelin, resistin and neuropeptide Y have been identified in limbic areas of the brain that regulate emotion.10,11

Among the most prevalent mental illnesses, depression is increasing in the western world and these peptides were reported to play a role in emergence of emotional disturbances like depression. For example; leptin was found to induce antidepressive state in mice.11,12 Likewise, neuropeptide Y has been found to have an impact in the pathomechanism of both anxiety and depression. Low neuropeptide Y levels have been reported in major depressive disorder.13,14 Adiponectin which is an adipocytokine with anti-inflammatory and antiatherogenic effects was also found to be reduced by 30% in premenopausal women with major depressive disorder.15 Resistin, a low grade inflammation marker, was studied and found to be correlated with atypical depressive symptoms.16 Gastric hormone ghrelin, besides stimulating appetite, acts as a neuropeptide, participates in sleep-wake regulation and appears to be related to memory and to be involved in pathophysiology of central nervous system disorders, especially depression.17

In patients with chronic kidney disease, the serum levels of these peptides differ from people with normal renal functions. Moreover, changes in serum levels of these peptides in chronic kidney disease patients have been known to be involved with protein energy wasting. Among end stage renal disease patients with wasting, low serum ghrelin and high leptin levels are correlated with highest mortality risk.18 Neuropeptide Y levels are reported to be increased in patients with end stage renal disease and high neuropeptide Y levels are regarded to be associated with left ventricular hypertrophy.19 Like neuropeptide Y, as kidney functions decline, circulating resistin and serum levels of adiponectin are increased.20,21 However whether alterations in these peptide levels are related with the exhibition of depression among chronic kidney disease patients is not clear.

The aim of this study is to evaluate depression and anxiety scores among predialysis, hemodialysis, peritoneal dialysis and kidney transplantation patients and to search the changes of serum concentrations of these adipocytokines and ghrelin in these patients with respect to emotional disturbances. To the best of our knowledge, this is the first study about this information in the literature.

Patients and methods

A total of 150 patients were included into the study. Study groups were as follows: Group 1, the control group (consisting of 30 patients with normal kidney functions), Group 2, hemodialysis group (30 patients on chronic hemodialysis programme who never underwent a transplant), Group 3, predialysis group (30 patients with chronic kidney disease stage 4 or 5), Group 4, peritoneal dialysis group (30 patients on chronic peritoneal dialysis programme who never underwent a transplant) and group 5, kidney transplant group (30 patients with functioning kidney allograft). Inclusion criteria were being more than 18-years-old and willingness to recruit into the study. Patients with infection, high C-reactive protein levels, prior history of transplantation were excluded. The study was conducted between January 2012 and august 2012. Blood samples (5 mL) of hemodialysis patients were obtained from the arterial line of the hemodialysis before the start of morning session. All of the patients had fasted overnight before the blood samples were taken in the next morning. Blood samples were centrifuged at 3000 × g for 15 min for separation of plasma. Aliquots of plasma samples were stored at −80 °C until analyzed.

Serum leptin, ghrelin, acylated ghrelin, neuropeptide Y, adiponectin, resistin levels of all of the groups were measured using ELISA (Sandwich) method with commercial ELISA kit manufactured by DRG (lot number: 45K032, Catalog number: EIA-2395, Germany), EIA (Competitive) method with commercial kit manufactured by Phoenix (lot number: 603182, catalog number: EK-031-30, CA), EIA (Sandwich) method with commercial kit manufactured by EIA (lot number: 0438348-1, catalog number: 10006306, France), ELISA (Competitive) method with commercial kit manufactured by USCN (lot number: L120604048, catalog number: E90879Hu, China), Platinum ELISA (Sandwich) method with commercial kit manufactured by eBioscinence (lot number: 7298504, catalog number: BMS2032, North America), and Platinum ELISA (Sandwich) method with commercial kit manufactured by eBioscinence (lot number: 74391005, catalog number: BMS2040, North America) respectively. Common biochemical parameters, including urea, creatinine, albumin were measured in all patients and controls, according to standard laboratory methods in the routine clinical laboratory. Previous records of hemodialysis group’s single pool Kt/V calculations determined from the pre- and postdialysis blood urea nitrogen (BUN) levels and the pre- and postdialysis weights by using the 2 BUN method described by Daugirdas JT22 were collected. Previous records of protein equivalent of total nitrogen appearance of hemodialysis patients which were calculated using the formula nPCR = (0.0136 × [Kt/V × ([predialysis BUN + postdialysis BUN] ÷ 2)]) + 0.25123 were collected. Previous records of dialysis adequacy of peritoneal dialysis patients which were assessed via calculations of urea clearance Kt/V from 24 h urine and dialysate.24 Previous records of protein equivalent of nitrogen appearance (PNA) of peritoneal dialysis patients which were estimated with 24 h urine and dialysate collection [the equation is: 6.25 × (Volume_urine × Concentration of Urea_urine) +(Volume_dialysate × Concentration of Urea_dialysate) + 1.81 +[0.031 × lean body weight, kg])25] were collected. Creatinine clearance and urea clearance of predialysis and renal transplant patients were recorded.26 The glomerular filtration rate of predialysis and transplant groups were estimated by both MDRD Formula27 and by taking average of creatinine and urea clearances.28 Daily protein intake of predialysis patients and transplant patients were estimated from 24 h urine collections [6.25 (urine urea nitrogen + 30 mg/kg)].29

All of the groups were assessed by using a semistructured interview for sociodemographic data. The diagnosis of alcohol and drug dependence was based on psychiatric examination according to DSM-IV criteria, while all other psychiatric disorders were diagnosed with a screening interview based on the Structured Clinical Interview for DSM-IV, Clinical Version (SCID-CV) for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition30,31 Turkish version32 conducted by a trained interviewer (researcher 2). Other instruments are self-report scales. The Beck Depression,33 and The Beck Anxiety Inventory34 which are self report scales, were administered and collected by the same interviewer. One of the researchers supervised patients who had difficulty in reading and understanding the inventories, helped subjects in filling the inventories if needed.

Brief Symptom Inventory (BSI) is a 53-item assessment tool used extensively to assess global psychological distress, which is determined by the individual’s score on a global severity index.35 The global severity of illness index for each subject is obtained by averaging the 53 symptom ratings. The measure has nine specific subscales (somatization, obsessive-compulsive, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, psychoticism). The brief symptom inventory was adapted to Turkish by Hisli-Sahin and Durak.36

Beck Depression Inventory (BDI), is an instrument that assesses the presence and severity of depression. The 21 items of the inventory were selected to represent symptoms commonly associated with a depressive disorder. Item categories include mood, pessimism, crying spells, guilt, self-hate and accusations, irritability, social withdrawal, work inhibition, sleep and appetite disturbance, and loss of libido.

Beck Anxiety Inventory (BAI), is a 21 item self report questionnaire with a focus on somatic anxiety symptoms, such as heart pounding, nervousness, inability to relax, and dizziness or light-headedness. Items are rated on a 4 point scale ranging from 0 (not at all) to 3 (severely: I could barely stand it). Validity and reliability studies of the Turkish have been performed by Ulusoy et al.37

The patients were aware of the study and signed an informed consent after reading the document. The study protocol was approved by the Ethics Committee of the School of Medicine at Karadeniz Technical University.

Statistical analysis

Associations among these variables were analyzed. All variables analyzed were considered to have a normal distribution as assessed by Kolgomorov-Sminorv test. Results are expressed as means ± SD (Standard deviation). Statistical analyses were performed with SPSS version 20.1 software package (SPSS Institute Inc., 2011). Data were presented as mean ± SD for normally distributed values (at Kolmogorov–Smirnov test) and median [IQ range] for non normally distributed values. Differences between groups were established by One-Way ANOVA followed by Dunnett test for normally distributed values and by Kruskal–Wallis analysis for nonparametric values. Pearson's correlation coefficient was searched to examine the relation between variables. Before testing correlations, all non-normally distributed values were log-transformed to better approximate normal distributions. Spearman’s correlation coefficient was used for non normally distributed values. Statistical significance was defined as p was <0.05.

Results

A total of 150 patients (89 male, 61 female) with mean age of 45.4 ± 15.9 years, were included into the study. Each of the 5 groups was composed of 30 subjects. Mean age of control group was 33.4 ± 9.4 years, of hemodialysis group was 49.8 ± 16.1 years, of predialysis group was 58.4 ± 15.2 years, of peritoneal dialysis group was 39.1 ± 13.4 years, of transplantation group was 47.1 ± 13.1. Difference between the groups was significant (p = 0.0001). Length of education years of the groups (16.6 ± 5.8 years, 7.6 ± 4.3 years, 5.8 ± 4.8 years, 9.0 ± 4.1 years, 8.4 ± 4.1 years for control, hemodialysis, predialysis, peritoneal dialysis and transplantation groups respectively) was significantly different (p = 0.0001). The sociodemographic data of the subjects are presented in Table 1. There were significantly more current or ex smokers among patients with chronic kidney disease regardless of their renal replacement therapy (hemodialysis, predialysis, peritoneal dialysis and transplantation groups) than those in control group (Table 1). Anxiety scores of the groups were not statistically different (Table 2). Hemodialysis and predialysis groups had significantly higher depression scores than control group (p = 0.009). Significantly higher scores for brief symptom inventory-somatization subscale were present in hemodialysis and peritoneal dialysis groups than those in control group (p = 0.041). Hemodialysis and transplantation patients had significantly higher scores for BSI-positive symptom distress index compared to those in control group. Present and past psychiatric diagnoses were also inquired in all patients. Four of (13.3%) group 5, 10 of (33.3%) group 4 and three of (10%) group 3, none of the control and transplant groups had a psychiatric disorder history (Table 3). The difference between the groups was statistically significant (p = 0.001). In present psychiatric diagnosis, two of (6.7%) control, hemodialysis, predialysis and peritoneal dialysis groups and three of (10%) transplantation group had anxiety disorder, one of (3.3%) control and predialysis groups, two of (6.7%) hemodialysis group, five of (16.7%) peritoneal dialysis group and none of transplantation group had depressive disorder. The difference was insignificant (Table 3).

Table 1. The sociodemographic data of the subjects.

Variables	Control Group mean ± SD	Hemodialysis Group mean ± SD	Predialysis Group mean ± SD	Peritoneal Dialysis Group mean ± SD	Transplantation Group mean ± SD	p Value
Age (years)	33.4 ± 9.4^a,b,c	49.8 ± 16.1^x	58.4 ± 15.2^y,z	39.1 ± 13.4	47.1 ± 13.1	0.0001
Education time (years)	16.6 ± 5.8^a,b,c,d	7.6 ± 4.3	5.8 ± 4.8	9.0 ± 4.1	8.4 ± 4.1	0.0001
	n (%)	n (%)	n (%)	n (%)	n (%)
Marital Status married	18 (60)	22 (73.3)	25 (83.3)	20 (66.7)	22 (73.3)	NS
Gender Men	12 (40)	20 (66.7)	18 (60)	16 (53.3)	23 (76.7)	NS
Literate	30 (100)^b	27 (90)	21 (70)^y,z	29 (96.7)	29 (96.7)	0.0001
Presence of current or  past Smoking Habitus	5 (16.7)^a,b,d,c	19 (63.3)	14 (46.7)	13 (43.3)	15 (50)	0.012

Note: ^a: group 1 versus group 2, ^b: group 1 versus group 3, ^c:group 1 versus group 5, ^d: group 1 versus group 4, ^x:group 2 versus group 4, ^y: group 3 versus group 4, ^z:group 3 versus group 5.

Table 2. BAI, BDI, and BSI scores of the groups (mean ± SD).

Variables	Control Group	Hemodialysis Group	Predialysis Group	Peritoneal Dialysis Group	Transplantation Group	p Value
BAI	6.26 ± 5.1	6.61 ± 4.68	8.86 ± 6.32	7.83 ± 8.79	8.03 ± 6.19	NS
BDI	5.77 ± 4.79^a,d	11.64 ± 7.81	10.52 ± 6.99	9.73 ± 9.04	7.23 ± 5.78	0.009
SOM	1.65 ± 2.26^a,b	4.0 ± 3.4	3.72 ± 3.1	3.9 ± 4.2	3.1 ± 3.6	0.041
OBS	3.52 ± 3.4	3.61 ± 3.61	3.21 ± 2.82	3.23 ± 4.00	3.43 ± 3.48	NS
INS	2.03 ± 2.48	1.93 ± 1.94	1.86 ± 2.10	2.17 ± 2.82	2.60 ± 2.16	NS
DEP	1.97 ± 2.44	2.43 ± 2.77	2.66 ± 3.00	3.20 ± 5.15	2.17 ± 2.44	NS
ANX	2.16 ± 2.75	1.86 ± 2.22	2.48 ± 2.41	2.80 ± 4.48	3.03 ± 3.87	NS
PHOB	1.26 ± 2.27	1.68 ± 2.31	1.38 ± 1.88	1.57 ± 2.16	1.70 ± 2.61	NS
HOS	2.16 ± 2.81	2.32 ± 2.28	2.14 ± 2.8	2.90 ± 3.60	2.53 ± 3.32	NS
PAR	2.52 ± 2.37	2.46 ± 2.47	2.66 ± 2.81	3.43 ± 3.51	2.77 ± 2.52	NS
PSY	0.87 ± 1.63	1.50 ± 2.0	1.45 ± 1.76	1.73 ± 2.79	0.87 ± 1.22	NS
AI	1.29 ± 1.53	1.93 ± 1.96	2.14 ± 1.64	2.33 ± 2.64	1.33 ± 1.88	NS
GSII	0.46 ± 0.78	0.45 ± 0.33	0.47 ± 0.33	0.52 ± 0.54	0.44 ± 0.38	NS
PST	14.0 ± 11.8	14.8 ± 9.9	15.2 ± 9.1	15.7 ± 11.2	13.9 ± 9.1	NS
PSDI	1.23 ± 0.48^a,c	1.64 ± 0.67	1.43 ± 0.42	1.40 ± 0.66	1.73 ± 1.20	0.047

Notes: ^a: Control group versus Hemodialysis group, ^b: Control group versus Peritoneal Dialysis group, ^c: Control group versus Transplantation group, ^d: Control group versus Predialysis group.

BAI: Beck Anxiety Inventory, BDI: Beck Depression Inventory, SOM: somatization, OBS: Obsessive compulsivity, INS: Interpersonal Sensitivity, DEP: Depression, ANX: Anxiety, PHOB: Phobic Anxiety, HOS: Hostility, PAR: Paranoid Ideation, PSY: Psychoticism, GSI: Global Severity of illness Index, PSDI: Positive Symptom Distress Index, PST: Positive Symptom Total.

Table 3. Present and past psychiatric diagnoses of the groups.

	Control Group	Hemodialysis Group	Predialysis Group	Peritoneal Dialysis Group	Transplantation Group
	n (%)	n (%)	n (%)	n (%)	n (%)	p Value
Past history Anxiety disorder	0 (0)	0 (0)	2 (6.7)	2 (6.7)	3 (10)
Depressive disorder	0 (0)	0 (0)	1 (3.3)	7 (23.3)	0 (0)
Axis 2	0 (0)	0 (0)	0 (0)	0 (0)	1 (3.3)
Alcohol abuse	0 (0)	0 (0)	0 (0)	1 (3.3)	0 (0)
Presence of pathology	0 (0)^a.b	0 (0)^c,d	3 (10)^e	10 (33.3)^f	4 (13.3)	0.001
Present diagnosis Anxiety disorder	2 (6.7)	2 (6.7)	2 (6.7)	2 (6.7)	3 (10)
Depressive disorder	1 (3.3)	2 (6.7)	1 (3.3)	5 (16.7)	0 (0)
Axis 2	0 (0)	0 (0)	0 (0)	0 (0)	1 (3.3)
Alcohol abuse	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Presence of pathology	3 (10)	4 (13.3)	3 (10)	7 (23.4)	4 (13.3)	NS

Note: ^a: group 1 versus group 4, ^b: group 1 versus group 5, ^c: group 2 versus group 4, ^d: group 2 versus group 5, ^e: group 3 versus group 4, ^f: group 4 versus group 5.

There was no significant relation between serum leptin levels and BDI, BAI, BSI scores of the groups (Table 4). Serum leptin levels were found to be significantly negatively correlated with male gender (r = −0.551; p = 0.0001), educational level (r = −0.188; p = 0.021), literacy (r = −0.334; p = 0.0001), being current or past smoker (r = −0.236; p = 0.004), while serum leptin levels were found to be positively correlated with age (r = 0.202; p = 0.013) and being married (r = 0.281; p = 0.0001) and body mass index higher than 23 kg/m² (r = 0.397, p = 0.0001). Both being married and age were found to be positively correlated with body mass index higher than 23 kg/m² (r = 0.409; p = 0.0001; r = 0.416; p = 0.0001 respectively) on the other hand educational level was found to be negatively correlated with body mass index higher than 23 kg/m² (r = −0.228; p = 0.002). Being a part of current smoker was found to be negatively correlated with having mid arm muscle circumference more than 50th percentile (r = −0.178; p = 0.029).

Table 4. Significant correlations with BAI, BDI, and BSI scores.

Variable	BAI	BDI	SOM	OBS	INS	D	ANX	PHOB	HOS	PAR	PSY	AI	GSII	PST	PSDI
Serum ghrelin level				r = 0.175*		r = 0.163*		r = 0.172*	r = 0.226**				r = 0.187*
Serum resistin level		r = 0.233***	r = 0.268***
Adiponectin												r = 0.208**
Serum leptin levels
Serum neuropeptide  Y levels
Age			r = 0.171*
Protein intake  (nPNA)	r = 0.201*			r = 0.280**				r = 0.235*	r = 0.278**				r = 0.267**		r = 0.200*
Literacy							r = 0.211**		r = 0.220**
Smoking status			r = 0.209**	r = 0.177*			r = 0.218**		r = 0.189*				r = 0.205*	r = 0.191*
Duration of Education		r = −0.165*	r = −0.189*												r = −0.220**
Presence of PEW		r = 0.198*										r = 0.210**
Presence of BFR 10%  of body weight									r = −0.243**	r = −0.203*	r = −0.199*			r = −0.162*
Serum albumin level				r = 0.171*								r = −0.209*
Duration of Education  among PD patients			r = −0.423*					r = −0.432*
Transplantation duration													r = 0.365*
Creatinine clearance of  transplant patients					r = 0.395*

Note: BAI: Beck Anxiety Inventory, BDI: Beck Depression Inventory, SOM: somatization, OBS: Obsessive compulsivity, INS: Interpersonal Sensitivity, DEP: Depression, ANX: Anxiety, PHOB: Phobic Anxiety, HOS: Hostility, PAR: Paranoid Ideation, PSY: Psychoticism, GSI: Global Severity of illness Index, PSDI: Positive Symptom Distress Index, PST: Positive Symptom Total. *p < 0.05, **p < 0.01, ***p < 0.001.

This study had given significant positive correlations between serum ghrelin levels and obsessive compulsivity (OBS), depression (D), phobic anxiety (PHOB), hostility (HOS) and global severity of illness index (GSII) scores of BSI (Table 4). There were significant positive correlations between resistin levels and age (r = 0.244; p = 0.003), smoking status of patients (r = 0.235; p = 0.004), BDI and BSI-somatization scores. While there were significant negative correlations between resistin levels and duration of education (r = −0.372; p = 0.0001).

There were significant negative correlations between adiponectin levels and male gender (r = −0.275; p = 0.001), duration of education (r = −0.227; p = 0.005), while significant positive correlations were obtained between adiponectin levels and BSI-additional items and past history of depressive disorder (r = 0.234; p = 0.004). Serum neuropeptide Y levels was observed to be significantly negatively correlated with only past history of depression (r = −0.176; p = 0.032). Age was found to be negatively correlated with depressive disorder (r = −0.198; p = 0.016); and positively correlated with somatization scores (r = 0.171; p = 0.037). There was significant positive correlation between literacy and anxiety and hostility scores (Table 4). Chronic kidney disease patients (hemodialysis, peritoneal dialysis, predialysis and transplantation groups) had significantly higher percentage of current or past smoking history than control group (p = 0.012). Smoking status was found to be positively correlated with somatization, obsessive compulsivity, anxiety, hostility, global severity of illness index, positive symptom total scores (Table 4) and being male gender (r = 0.433 and p = 0.0001). Past history of depressive disorder was found to be positively correlated with present diagnosis of depressive disorder (r = 0.464; p = 0.0001). There was significant negative correlation between present diagnosis of depressive disorder and age (r = −0.198; p = 0.016) and being male (r = −0.168; p = 0.041).

There was significant negative correlation between duration of education and BDI, somatization, positive symptom distress index scores (Table 4). This study had given significant positive correlation between BDI scores and presence of PEW (Table 4). Positive correlations were observed between protein intake and BAI, obsessive compulsivity, phobic anxiety, hostility, global severity of illness index, positive symptom distress indexes of BSI (Table 4). Interpersonal sensitivity was only found to be positively correlated with creatinine clearance of transplant patients (r = 0.395; p = 0.031).

Presence of depressive disorder among predialysis patients was negatively correlated with duration of education (r = −0.375; p = 0.041) While presence of depressive disorder among peritoneal dialysis patients were found to be negatively correlated with mid arm muscle circumference (r = −0.355; p = 0.054) and having body mass index higher than 23 kg/m² (r = −0.360; p = 0.049).

Discussion

Chronic kidney disease (CKD) is very important public health burden leading to multisystemic involvement including depressive disorders which is the most common psychiatric disorder in dialysis patients.38,39 In our study subjects, when groups compared in means of past history of psychiatric diagnosis, 10% of predialysis group, 33.3% of peritoneal dialysis group and 13.3% of renal transplantation group had a diagnosis of any psychiatric disease in the past (p = 0.001). But no significant difference in present psychiatric diseases between the groups was found. In spite of absence of any psychiatric disease in control and hemodialysis groups in the past, 10% of control and 13.4% of hemodialysis groups had present depressive or anxiety disorder at the time of evaluation which was not statistically significant from those of other groups.

Anxiety, includes all conditions of indefinite fear and psychic disorders dominated by fear. Even though there was no significant difference with regard to present depressive or anxiety disorders of the groups, we found significantly higher BDI scores in hemodialysis and peritoneal dialysis groups compared to control group. In spite of some other studies,40 BAI scores of the groups were not statistically different in our study. This might be the result of cultural acceptance, social support, coping defense mechanisms and greater use of reappraisal to regulate emotions among the study subjects. However, both peritoneal and hemodialysis patients had higher scores in BSI-Somatization than control group. Novaković M had found higher somatization scores among hemodialysis patients with Balkan endemic nephropathy.41 Gillanders S and colleagues had put forward that hemodialysis patients who used suppression which is one of two emotion regulation strategies (the other is reappraisal) were associated with greater levels of depression, somatization and greater dissatisfaction with the time spent dealing with their kidney disease.42

We found that both hemodialysis and transplant patients had higher positive symptom distress index scores compared to control group. Symptom distress in dialysis patients was found by Gamondi C and colleagues to be important, but underestimated and undertreated.43 Based on our data, hemodialysis and transplant patients appear to be significantly more distressed than control group.

Depression has been demonstrated to be associated with less patient well-being and kidney disease management including poor drug adherence, higher mortality and hospitalization rate, reduced treatment compliance and a poor nutritional status.44 The etiology of depressive disorders in CKD patients is multiple, including psychological, physical and socioeconomic state. Leptin, adiponectin, neuropeptide Y, ghrelin and resistin were reported to be involved in pathophysiology of depressive disorders among subjects with normal kidney function.11,13,15,16,17 To our knowledge the role of adipocytokines in etiology of depressive disorders of CKD patients have not been explored yet.

We found no correlation of leptin levels with depressive or anxiety or psychosocial distress disorders of chronic kidney disease patients. As expected, leptin was only found to be in significant correlation with parameters which were significantly related with BMI like age, being married, educational level, smoking status. High body fat ratio was found to be negatively correlated with BSI-hostility, paranoid ideation, psychoticism scores. While smoking status which was significantly more common among CKD patients than control, was found to be positively correlated with BSI-somatization, OBS, anxiety, hostility, global severity of illness index; educational level was negatively correlated with BDI, BSI-somatization, positive symptom distress index scores. Emotion regulation refers to the psychological strategies people use to cope with stressors. According to our study, education seems to help people in finding psychological strategies to cope stressors.

Bilgic and colleagues reported association of malnutrition inflammation score with the presence of depression.45 Likewise we also found significant positive correlation of protein energy wasting with Beck Depression Inventory scores. However, protein intake was positively associated with perceived stress in our study.

Serum ghrelin was found to be in positive correlation with OBS, depression, phobic anxiety, hostility and GSII scores of BSI in our study. The link between stress and feeding behavior which can be bidirectional has been familiar for a long time. In parallel with our study, plasma ghrelin levels were found to be enhanced under conditions of physiological stress, and ghrelin has been suggested to play an important role in stress-induced food reward behavior in animal models.46 Ghrelin seems to reduce anxiety after acute stress.

We did not observe any significant relationship of neuropeptide Y, acylated-ghrelin and adiponectin with BDI, BAI, or BSI scores. While resistin was significantly positively correlated with scores of Beck Depression Inventory and Brief symptom Inventory-somatization scores. Lehto SM and colleagues had reported that resistin levels of subjects without kidney disease were correlated with atypical depressive symptoms.47 We also think there might be a link between resistin and depressive disorders of patient with CKD based on our results. More studies are needed to enlighten this issue.

Limitations of this study

We acknowledge that even though groups were not different with respect to gender and marital status of subjects, average duration of education was significantly lower; mean of age and presence of past/current smoking habitus were significantly higher in groups with chronic kidney disease patients compared to control group involving healthy volunteers. So, groups were not homogenous with respect to age, smoking status and educational level which might influence the depressive symptoms. More studies with homogenous groups with respect to these variables are needed before extracting final conclusion. Lastly, acylated ghrelin levels were only studied in half of the groups because of insufficiency of kits.

Conclusion

In conclusion, more attention should be paid to the interpersonal relations of CKD patients. Taken together, our results suggest that patients with chronic kidney disease had higher BDI, BSI-somatization, BSI-positive symptom distress index scores than volunteers with normal renal function. High serum resistin levels and presence of protein energy wasting might have a role in development of depressive disorders of patients with chronic kidney disease. Furthermore, education and protection from smoking habitus might be important for psychological well being of chronic kidney disease patients. The interaction between depressive disorders of CKD patients and adipocytokines needs to be studied more extensively, in order to assess better approaches to healthcare for these individuals.

Declaration of interest

This work was supported by the Scientific Research grant of Karadeniz Technical University. We declare no conflicts of interest and also that the results presented in this paper have not been published previously in whole or part, except in abstract format.

Acknowledgements

The authors thank very much to their hemodialysis and peritoneal dialysis nurses Asiye Aydin and Meryem Bakkaloglu for their careful and devoted follow-up of patients.