Neuronal damage and inflammatory biomarkers are associated with the affective and chronic fatigue-like symptoms due to end-stage renal disease

Subjects

The present study involved a sample of seventy patients with ESRD and 46 normal controls. Everyone had experienced an acute kidney injury (AKI) before their current condition, ultimately leading to the development of end-stage renal failure. As a result, all patients were receiving ongoing treatment through the use of continuous dialysis. The study recruited participants from the Dialysis Unit located at Al-Hakeem General Hospital in Najaf Governorate during the period spanning from March to May of 2022. Patient assessment was conducted by considering their complete medical history, including the existence of any systemic disorders. The patients were diagnosed by a senior physician in accordance with the 10^th edition of the International Statistical Classification of Diseases and Related Health Problems (ICD-10-CM), specifically under the diagnosis code N18.6. All subjects were under constant medication with folic acid or iron and folate formula (Fefol^®) along with calcium carbonate, epoetin alfa (Eprex^®), and heparin. The study’s control group comprised 46 individuals who were in good health and were selected from the same geographical region as the patients. The control group was matched to the patients in terms of age and gender. Neither the patients nor the controls exhibited any lifetime axis-1 diagnoses of prior neuropsychiatric disorders, such as major depressive episodes, anxiety disorders, schizophrenia, bipolar disorder, psycho-organic disorders, substance use disorders (excluding nicotine dependence), and chronic fatigue syndrome. All study participants exhibited no medical conditions apart from ESRD. This includes, but is not limited to, diabetes mellitus, inflammatory diseases, Parkinson’s or Alzheimer’s disease, stroke, oncologic disorders, and (auto)immune disorders.

The patients or their first-degree relatives provided written informed consent. The present study has been approved by the institutional ethics board of the University of Kufa (Document No.2229T / 2022) and the Najaf Health Directorate, Training, and Human Development Center (Document No.20123 / 2022). The research was carried out in compliance with both Iraqi and international regulations on ethics and privacy. Furthermore, it was conducted under the ethical principles outlined in the World Medical Association Declaration of Helsinki. In addition, it is noteworthy that our Institutional Review Board (IRB) adheres to the International Guideline for Human Research Protection, which is mandated by the Declaration of Helsinki, the Belmont Report, the CIOMS Guideline, and the International Conference on Harmonization in Good Clinical Practice (ICH-GCP).

Clinical assessments

A senior psychiatrist employed a semi-structured interview technique to gather sociodemographic and clinical information. The severity of fibromyalgia and chronic fatigue was assessed by a senior psychiatrist using the FibroFatigue (FF) scale (Zachrisson et al., 2002), while the degree of anxiety was evaluated using the Hamilton Anxiety Rating Scale (HAMA) (Hamilton, 1959). The participants completed the Beck Depression Inventory (BDI)-II rating scale concurrently (Beck et al., 1996). The study employed the total FF, HAMA, and BDI-II rating scales scores and used the items of these scales to derive scores for distinct symptom subdomains. The study measured pure depressive symptoms by computing the total score of BDI-II items feelings of sadness, discouragement about the future, a sense of failure, dissatisfaction, guilt, self-disappointment, self-criticism, suicidal thoughts, crying, loss of interest, difficulty making decisions, a negative self-image, and work inhibition. The construct of pure anxiety was operationalized as the summed score of the HAMA items anxious affect, tension, fears, and anxious conduct exhibited during a structured interview. The present study operationalized pure physiosomatic symptoms as a z-unit-weighted composite score based on FF, HAMA, and BDI-II items: muscle pain, muscle tension, fatigue, autonomic symptoms, gastrointestinal symptoms, headache, and flu-like malaise (all from the FF scale); fatigue and loss of libido (from the BDI-II scale); and somatic muscular, somatic sensory, cardiovascular symptoms, respiratory symptoms, gastrointestinal symptoms, and autonomic symptoms (all from the HAMA scale). The pure FF score was the summation of z scores for muscle pain, muscle tension, fatigue (FF items), somatic muscular (HAMA), and fatigue (BDI).

The diagnosis of Tobacco Use Disorder (TUD) was made following the criteria outlined in the DSM-IV-TR. The calculation of Body Mass Index (BMI) involved the division of body weight in kilograms (kg) by the square of length in meters (m).

Measurements

Blood samples were collected from the participants after overnight fasting, specifically between 8:00 a.m. and 9:00 a.m. Blood was collected into plain and EDTA tubes. Following the separation process, the serum samples were subsequently allocated into Eppendorf® tubes and kept frozen until thawed for assay. The spectrophotometric measurement of serum albumin, urea, creatinine, total serum protein, copper, and zinc was conducted using kits provided by Spectrum Diagnostics Co. (Cairo, Egypt). The CRP latex slide test manufactured by Spinreact^® (Barcelona, Spain) was utilized for conducting CRP assays on human serum. Commercial ELISA sandwich kits provided by Melsin Medical Co. (Jilian, China) and Sunlong Biotech Co., Ltd (Zhejiang, China) were used to measure GFAP, NFL, MBP, nestin, IBA1, S100B, IL-6, and IL-10. We computed two z unit-weighted composite scores: a) an inflammatory composite as z CRP + z IL-6 + z IL-10 – z zinc (z_inflammation); and b) an intermediate filament protein composite as z transformation of nestin (z NES) + z GFAP + z NFL (z_interfilament). The estimated GFR (eGFR) was calculated by using the Modification of Diet in Renal Disease Study equation (Levey et al., 2007): eGFR =175 x (S.Cr)^-1.154 x (Age)^-0.203 x 0.742 [if female] x 1.212 [if Black].

Statistical Analysis

The analysis involved the use of Pearson’s product-moment correlation coefficients to examine the correlations between continuous variables. The study utilized analysis of variance (ANOVA) to investigate the associations among groups and clinical or biomarker data. The evaluation of statistical associations between categorical variables was conducted through contingency table analysis (χ²-test). Multiple regression analysis was employed, utilizing a manual method, to evaluate the most important predictors of clinical rating scale scores. Biomarkers were utilized as explanatory variables, while allowing for the potential influence of confounding variables such as age, sex, BMI, and TUD. In addition to utilizing a manual regression technique, an automated technique was also employed, whereby model entry and removal were determined based on p-values of 0.05 and 0.10, respectively. The regression model was evaluated by computing the F-statistic, degrees of freedom, and p-values. Additionally, the proportion of variance accounted for by the model (R²), standardized β-coefficients, t-statistics, and p-values were determined for each predictor. Furthermore, an assessment was conducted on the variance inflation factor and tolerance to detect any potential collinearity or multicollinearity issues. The presence of heteroskedasticity was assessed using the White and modified Breusch-Pagan tests for homoscedasticity. All tests were two-tailed, at p=0.05. Principal component (PC) analysis was performed to delineate the first PC extracted from a set of variables (e.g. HAMD. BDI and FF scores). To be accepted as a meaningful and validated construct, all loadings on this first PC should be greater than 0.7, and this PC should account for more than 50% of the total variance in the data. Moreover, the Kaiser-Meyer-Olkin (KMO) metric should be greater than 0.6 (a test for the factorability of the correlation matrix), the anti-image matrix should be adequate, and Bartlett’s test of sphericity should be significant. All statistical analyses were carried out utilizing IBM, SPSS for Windows version 28 (IBM-USA). The primary statistical analyses in this study were the multiple regression analyses delineating the effects of the biomarkers on the physio-affective phenome of ESRD. A priori power analysis (G*Power 3.1.9.4) for linear multiple regression analysis, with an effect size of 0.176 (equivalent to 15% explained variance), alpha=0.05, power=0.8, and 4 covariates, indicates that the minimum sample size should be 73.

Construction of PCs of renal functions, dialysis, and clinical domains

Using PCA, we have developed two PCs that reflect kidney function and dialysis characteristics. First, we were able to extract a first PC (PC_kidney) from urea, creatinine, eGFR, TSP, and albumin (KMO=0.878, Bartlett’s test of sphericity χ²=368.872, df=10, p=0.001, VE=72.53%, all loadings > 0.783). Second, we extracted the first PC from the number of dialyses per week, the total number of dialyses, and the last day patients underwent dialysis, labeling it PC_dialysis (KMO=0.613, Bartlett’s test of sphericity χ²=113.336, df=3, p=0.001, VE=69.79%, all loadings > 0.774). Consequently, we computed a z composite score using PC_kidney + PC_dialysis (labeled: z_diakidney). We were able to extract one validated PC (labeled PC_phenome) from the total FF, HAMA, and BDI scores (KMO=0.730, Bartlett’s test of sphericity χ²=311.368, df=3, p=0.001, VE=88.27%, all loadings > 0.908).

Comparison of demographic and neuropsychiatric scores between ESRD and controls

Table 1 presents the sociodemographic and clinical data of ESRD patients and the healthy control group. There was no significant difference observed between the two groups in terms of age, sex ratio, BMI, TUD, residency, and employment. The single/married ratio was slightly different between both groups. z_diakidney and all clinical subdomain scores (FF-total, HAM-A total, BDI-total, pure depression, anxiety, physiosomatic, FF, and PC_phenome) were significantly higher in ESRD patients than in controls.

Comparison in serum biomarkers between ESRD and control groups

The biomarker levels in both patient groups and the healthy controls are presented in Table 2. Serum levels of NFL, MBP, nestin, z_interfilament, CRP, IL-10, copper, and z_inflammation were significantly higher in patients than in controls. A significant decrease in serum zinc was recorded in ESRD patients compared with controls. No significant differences were established in serum IBA1, S100B, GFAP, or IL-6 between ESRD patients and control groups. After FDR p correction, all significant differences listed in Table 2 remained significant, except MBP (FDR p =0.062).

Correlation matrix

The results of the correlation calculations between clinical scores and the measured biomarkers are presented in Table 3. Z-diakidney was significantly correlated with pure depression, anxiety, physiosomatic, and FF scores, z_interfilament, NFL, nestin, and the z_inflammation index. The pure depression score showed a significant correlation with anxiety, physiosomatic and FF scores, z_interfilament, nestin, S100B, and z_inflammation score. The pure anxiety symptom score is correlated significantly with pure physiosomatic and FF scores, z_interfilament, nestin, and z_inflammation. The physiosomatic score showed a significant correlation with the FF score, z_interfilament, NFL, nestin, and z_inflammation. The FF score is significantly correlated with z_interfilament, NFL, and z_inflammation. CRP (r=0.288, p=0.002, n=116) and z_inflammation (r=0.235, p=0.011, n=116) were significantly associated with z_interfilament.

Prediction of the clinical symptom domains

Table 4 shows the results of different stepwise multiple regression analyses with clinical symptom scores as dependent variables and biomarkers and confounding variables as independent variables. Firstly (denoted “a” in Table 4), we entered only the single neural variables, and secondly (denoted “b” in Table 4) we also entered z_inflammation, z_interfilament, copper, and background variables. Regression #1.a shows that 14.7% of the variance in the pure depression score was explained by NFL, MBP, and S100B. Regression #1.b shows that the model explained 53.4% of the variance in the pure depression score using z_inflammation, copper, S100B, z_interfilament (all positively associated), education (negatively associated), and sex as explanatory variables. Regression #2.a shows that 8.9% of the variance in the pure anxiety symptoms score was explained by nestin. Regression #2.b shows that a significant part of the pure anxiety variance (39.5%) is explained by z_inflammation, copper, and nestin. Regression #3.a shows that 11.3% of the variance in the pure physiosomatic symptom scores was explained by NFL and nestin. Regression #3.b shows that the model explained 46.0% of the variance in the pure physiosomatic score using z_inflammation, copper, and z_interfilament. Figure 1 shows the partial regression of the pure physiosomatic symptoms scores on the z_interfilament composite. In Regression #4, 36.7% of the variance in the FF score can be explained by the regression on copper, z_inflammation, and z_interfilament. Regression #5.a examines the effects of the neuronal markers on the phenome and shows that 12.5 % of the variance in PC_phenome was explained by nestin and NFL. Regression #5.b entered neuronal markers as well as other biomarkers and confounding variables and shows that a considerable part of the variance in the phenome score (50.4%) was explained by the regression on z_inflammation, copper, z_interfilament, and S100B. Figure 2 presents the partial regression of the PC_phenome score on the composite z interfilament score. Figure 3 shows the partial regression of the phenome score on the z_inflammation composite.

• Download figure
• Open in new tab

Figure 1.

Partial regression of the pure physiosomatic symptoms on z_interfilament, a z composite score based on the interfilament damage biomarkers nestin, glial fibrillary acidic protein, and neurofilament light chain

• Download figure
• Open in new tab

Figure 2.

Partial regression of the phenome of end-stage renal disease (comprising depression, anxiety and fibro-fatigue symptoms) on z_interfilament, a z composite score based on the interfilament damage biomarkers nestin, glial fibrillary acidic protein, and neurofilament light chain

• Download figure
• Open in new tab

Figure 3.

Partial regression of the of the phenome of end-stage renal disease (comprising depression, anxiety and fibro-fatigue symptoms) on z_inflammation, a composite based on 4 inflammatory biomarkers.

The physio-affective phenome of ESRD

The first major finding of the current study is the highly significant increase in the total and subdomain scores of neuropsychiatric symptoms (anxiety, depression, and FF) in ESRD, and that we were able to extract one principal component from those symptom domain scores. These results extend those of previous publications showing a high prevalence of neuropsychiatric disorders among ESRD patients including depression (Elzeiny and El-Emary, 2023, Ibrahim et al., 2023, Cohen and Kimmel, 2018, Savitha et al., 2020), anxiety (Semaan et al., 2018, Qawaqzeh et al., 2023, Khoury et al., 2023), and fatigue (Burdelis and Cruz, 2023, Molfino et al., 2023). In those patients, fatigue (Bossola et al., 2015), depression (Lopes et al., 2002), and anxiety (Schouten et al., 2019) are considered to be independent risk factors for increased number of hospitalizations, morbidity, and mortality. Affective disorders due to ESRD are well-known risk factors that impact disabilities, the health, and the quality of life of patients on hemodialysis treatment (Cukor et al., 2007). These studies show that to lower morbidity and mortality in ESRD patients, neuropsychiatric symptoms need to be evaluated, integrated into the severity assessment, and considered while treating ESRD patients.

In addition, our findings indicate that deteriorating kidney function and the number of dialysis treatments are substantially associated with the severity of the physio-affective phenome. Therefore, it is possible to conclude that neurotoxic factors related to ESRD and/or dialysis treatment may be responsible for the interrelated increases in depression, anxiety, and fatigue-physiosomatic symptoms.

Biomarkers of ESRD

The second major finding of our study is that NFL, nestin, a composite based on the three intermediate filament proteins (NFL, nestin, and GFAP), CRP, IL-10, and copper are increased, and zinc decreased in patients with ESRD as compared with the control group. In addition, deterioration of kidney function and an increase in the number of dialysis interventions were significantly associated with the intermediate filament composite, particularly NFL and nestin.

Increased NFL, a neuro-axonal damage biomarker (Watanabe et al., 2019), reflects ongoing and acute disease processes in patients with neuroinflammatory illnesses (Cantó et al., 2019, Srpova et al., 2021). As such, our results indicate a potential connection between increasing neurotoxicity due to renal failure, aberrations in neuronal cell communication and cell structure, and astrocyte functions as indicated by increased NFL levels (Wang et al., 2017b, Maes, 2022a). Nestin is another intermediate filament protein that is expressed by neuronal progenitor cells, including those in the adult brain; and that mediates radial growth of the axon (Hendrickson et al., 2011). Nestin-expressing neurons are mainly observed in brain regions involved in cognitive functions suggesting that nestin plays a key role in neuroplasticity. Re-expression of nestin in reactive astrocytes and glial scars may be triggered by injuries such as traumatic and ischemic brain lesions (Kernie et al., 2001, Liberto et al., 2004, Gilyarov and physiology, 2008). However, nestin is not exclusively expressed in neurons but is also expressed by kidney cells (Su et al., 2008), especially in the hypoxic regions of the kidney and following tubulointerstitial injuries (Sakairi et al., 2007, Tomioka et al., 2010). Increased expression of nestin has been reported during the regeneration of injuries to skeletal muscles (Oikawa et al., 2010).

Another intermediate filament found in the cytoskeleton of astrocytes is GFAP, which regulates astrocyte activity, cell structure, and communication between cells (Nedergaard et al., 2003, Wang et al., 2017a, Maes et al., 2023). In response to CNS damage and reactive astrogliosis, GFAP expression is elevated. According to Halford et al. (2017), damaged astrocytes can leak GFAP into the neuronal regions (Halford et al., 2017), and GFAP may be translocated into the peripheral circulation through the breakdown of the BBB (Hol and Pekny, 2015, Yang and Wang, 2015). So, the presence of GFAP in plasma may indicate increased neurotoxicity, neuronal, and glial cell damage and may even predict astrocyte cell death (Yang and Wang, 2015, Halford et al., 2017). Although increases in the two other intermediate filament proteins were observed in ESRD, no significant changes could be established in serum GFAP.

The increase in MBP in ESRD patients has not previously been studied. In cases of traumatic brain injury, benign and malignant intracranial tumors, CNS infections, and cerebrovascular accidents, serum levels of MBP are typically elevated. Increased concentrations serve as a specific indicator of demyelination (Wąsik et al., 2020). MBP is extremely basic and carries highly positive charges (Harauz et al., 2009) and it directly interacts with acidic components on the external side of the neuronal membrane to generate neurotoxicity. Therefore, it is probable that MBP plays a role in the etiology of CNS illnesses characterized by myelin destruction (Zhang et al., 2014).

Increased neuro-axis biomarkers are associated with the physio-affective phenome

The third major finding of this study is that the intermediate filament-based composite, and NFL, MBP, and S100B are significantly associated with the different symptom domains (depression, anxiety, physiosomatic, and FF scores) as well as with the physio-affective phenome of ESRD. In addition, the serum levels of intermediate filaments are elevated in ESRD and, in conjunction with S100B, are associated with the severity of neuropsychiatric symptoms, independent of the effects of inflammatory mediators and copper.

S100B is predominantly expressed by astrocytes and plays a role in neurite extension, astrocytosis, and axonal proliferation; however, its expression is increased in the adult brain following neuronal damage (Gerlach et al., 2006). During brain injury, astrocytes secrete S100B that may leak into the systemic circulation, which explains the higher serum levels in patients with neurodegenerative disorders, brain injuries, and breakdown of the blood-brain barrier (BBB) (Wang and Bordey, 2008, Chaves et al., 2010, Krishnan et al., 2020, Langeh and Singh, 2021).

Our results extend those of previous reports. For example, increased serum levels of neuronal damage biomarkers, namely PDGFR, NFL, GFAP, and phosphorylated tau protein 217, were observed in MDD patients as compared with controls (Al-Hakeim et al., 2023). Serum levels of NFL and S100B are significantly higher in MDD than in controls (Arora et al., 2019, Chen et al., 2022). NFL serum levels are inversely related to cognitive functions in depressed patients (Bavato et al., 2021). Furthermore, serum GFAP levels may rise with increasing MDD severity (Steinacker et al., 2021), and elevations in plasma GFAP were observed in individuals with neuropsychiatric illnesses who lack brain radiological abnormalities (Esnafoglu et al., 2017).

Interestingly, we detected that the inflammatory composite, indicating an acute phase response in our ESRD patients (e.g., increased CRP and lowered zinc) was significantly associated with increases in the intermediate filament proteins. This is in agreement with a previous report that increased tryptophan catabolites (TRYCATs) in patients undergoing dialysis are associated with depression and fatigue (Malhotra et al., 2017). TRYCATs are induced by pro-inflammatory cytokines and many TRYCATs have neurotoxic effects (Almulla et al., 2022). Previous work also found that severe AKI induces inflammation in association with functional changes in the brain (Liu et al., 2008). Previously, we observed that, in major depression, inflammatory mediators were significantly associated with an increase in GFAP, NFL, and phosphorylated tau protein 217 (Al-Hakeim et al., 2023). The associations between intermediate filament proteins and S100B, inflammatory mediators, and neuropsychiatric symptoms due to ESRD suggest that the severity of kidney dysfunctions and/or the number of dialysis interventions cause inflammation and consequently neuro-affective toxicity leading to neuropsychiatric symptoms.