Protein carbonyl: A sensitive marker reflecting oxidative stress in prevalent hemodialysis patients

Abstract

BACKGROUND

Protein carbonyls (PCOs) are well-established biomarkers of oxidative stress, closely linked to disease progression and therapeutic responses across various pathological conditions.

AIM

To assess serum PCO titres as an indicator of oxidative damage in prevalent hemodialysis (HD) patients and to explore their association with nutritional status.

METHODS

This controlled cross-sectional study included 60 patients with end-stage renal disease (ESRD), aged 18-65 years, undergoing maintenance HD three times per week for at least six months. All participants were treated through the same high-flux dialyzer equipped with a polysulfone membrane. A control group of 30 healthy individuals was also enrolled. Serum PCO titres were measured in all participants.

RESULTS

The ESRD group exhibited significantly higher values of systolic blood pressure, serum creatinine, serum phosphate, C-reactive protein, and PCO than the control group. In contrast, hemoglobin titres and serum albumin titres were significantly diminished in ESRD patients. Receiver operating characteristic curve analysis of serum PCO demonstrated excellent discriminatory capacity, with a cut-off value of 115 pmol/mg yielding 93.3% sensitivity and 100% specificity. A significant positive correlation was found between C-reactive protein titres and both age and parathyroid hormone concentrations in the ESRD group. Additionally, nutritional status was significantly worse in ESRD patients than in controls.

CONCLUSION

Serum PCO titres were markedly elevated in ESRD patients undergoing HD, reflecting enhanced oxidative stress. Given its high diagnostic accuracy, PCO can act as a sensitive and reliable marker for oxidative stress in ESRD.

Key Words: Protein carbonyl; Marker; Oxidative stress; Hemodialysis; Reactive oxygen species; Chronic kidney disease

Core Tip: Serum protein carbonyl titres were markedly elevated in end-stage renal disease patients undergoing hemodialysis, reflecting enhanced oxidative stress. Given its high diagnostic accuracy, protein carbonyl can act as a sensitive and reliable marker for oxidative stress in end-stage renal disease.

INTRODUCTION

Uremia and hemodialysis (HD) are known to promote increased generation of reactive oxygen species while concurrently reducing antioxidant defenses in individuals undergoing HD[1,2]. Beyond the intrinsic impact of HD, several dialysis-related variables critically influence oxidative stress titres in this patient population. These include dialysis duration, iron supplementation, anemia, the presence of a central venous catheter, vascular access type and dysfunction, the nature of dialyzer membranes, HD modality, and anticoagulation practices[3].

Oxidative stress is increasingly acknowledged as a crucial element in the initiation and advancement of chronic kidney disease (CKD), and it also significantly influences renal replacement treatment. The cellular damage induced by oxidative stress is linked to the development of several systemic complications in CKD, including anemia, inflammation, atherosclerosis, hypertension, cardiovascular disease, and compromised immune responses[4].

A multifactorial chronic inflammatory state is observed in CKD, with key contributors including increased production of pro-inflammatory cytokines, oxidative stress, metabolic acidosis, recurrent or chronic infections, intestinal dysbiosis, and disruptions in adipose tissue metabolism[5]. Additionally, HD subjects are exposed to procedural stressors that compound their risk profile, such as rapid fluctuations in plasma electrolyte concentrations, hemodynamic strain resulting from intradialytic and interdialytic changes in cardiac preload, and variable blood pressure[4].

Inflammatory processes are characterized by elevated titres of mediators such as cytokines (interleukin-1, interleukin-6, tumor necrosis factor-α), adipokines, acute-phase proteins [primarily C-reactive protein (CRP)], and adhesion molecules. These biomarkers are closely associated with CKD-related complications, as confirmed by clinical investigations[6].

Oxidative stress is detectable in early stages of CKD[7] and progressively intensifies with disease progression, becoming particularly pronounced at the initiation and during maintenance of HD. Protein carbonyls (PCOs) are considered among the most validated biomarkers of oxidative stress, with demonstrated associations with both disease status and therapeutic interventions across various medical conditions[8].

The practical advantages of assessing plasma proteins, namely, ease of sampling and the relatively extended half-life of many plasma proteins, render PCO a particularly valuable biomarker for oxidative damage in the CKD population[9].

Accordingly, this study was conducted to evaluate serum PCO level as an index of oxidative injury among prevalent HD patients and to examine its correlation with nutritional status.

MATERIALS AND METHODS

This controlled cross-sectional study enrolled 60 adult patients, aged between 18 years and 65 years, all diagnosed with end-stage renal disease (ESRD) and undergoing maintenance HD thrice weekly for a minimum duration of six months. All participants received HD under standardized conditions through high-flux dialyzers equipped with polysulfone membranes. A comparison group consisting of 30 healthy individuals was also included.

The study took place from September 2022 to February 2023, following approval by the Ethical Committee of Ain Shams University, Cairo, Egypt, and in accordance with the principles outlined in the Declaration of Helsinki (No. MS 850/2022). The patient’s informed consent form has been provided. The form confirms that the patient (or legal guardian) agreed to participate and to the publication of the clinical data for scientific purposes.

Exclusion criteria included the presence of acute infections within the three months preceding enrollment, as well as comorbidities known to influence serum PCO titres, such as liver cirrhosis, chronic inflammatory conditions, malignancies (both hematologic and solid tumors), and the use of immunosuppressive medications.

All enrolled patients underwent comprehensive clinical evaluation and laboratory investigations. The latter encompassed measurements of hemoglobin, serum urea, creatinine, albumin (ALB), iron, transferrin, ferritin, calcium (Ca), phosphate (PO₄), intact parathyroid hormone (PTH), CRP, and serum PCO titres. Nutritional assessment was performed through the subjective global assessment (SGA) method. Blood samples were collected immediately before the HD session.

Measurement of serum PCO

Blood samples were collected in sterile, pre-labeled tubes and preserved at -20 °C until further analysis. Serum PCO titres (ng/mL) were quantified through a commercially available enzyme-linked immunosorbent assay kit (Human PCO ELISA Kit, CAT: E1246Hu; Bioassay Technology Laboratory, China), following the manufacturer’s instructions. The assays were performed through an enzyme-linked immunosorbent assay plate reader (DAS Plate Reader, SN 2006, Italy).

Briefly, serum samples and standards were added to pre-coated microplate wells containing monoclonal anti-human PCO antibodies and incubated under specified conditions. Calibration of the assay was achieved through a standard curve generated from serial dilutions of the provided standards, and sample concentrations were calculated accordingly.

Following incubation, biotinylated anti-human PCO antibodies were added, followed by streptavidin-horseradish peroxidase conjugate. After appropriate washing steps to remove unbound reagents, a substrate solution was introduced to produce a colorimetric reaction proportional to the PCO concentration. The reaction was terminated through an acidic stop solution, and absorbance was measured at 450 nm.

All samples were analyzed in duplicate to ensure assay reproducibility, and the intra-assay coefficient of variation was maintained below 10%. Quality control samples were included in each assay run to ensure accuracy and consistency of the measurements.

Sample size determination

Sample size was calculated through PASS 11.0 and based on a study by Dimitrijevic et al[10]. A sample of 90 subjects has 96% power to detect a difference of 0.5000 and an area under the curve of 0.7700 under the alternative hypothesis, through a two-sided Z-test at a significance level of 0.0050. The data are discrete (rating-scale) responses.

Statistical analysis

For the statistical analysis, IBM SPSS Statistics version 27 (IBM^©, Chicago, IL, United States) was employed. To assess the normality of the data distribution, both the Shapiro-Wilk test and visual inspection of histograms were used. For normally distributed quantitative data, results were presented as mean ± SD, and one-way analysis of variance was performed, followed by Tukey’s post hoc comparisons to identify significant differences between groups. In cases where data did not conform to a normal distribution, the median and interquartile range were used to summarize the data. Group comparisons for non-normally distributed data were conducted through the Kruskal-Wallis test, followed by pairwise comparisons with the Mann-Whitney U test to examine differences between specific groups. Categorical variables were reported as frequencies and percentages, and comparisons between groups were made through the χ² test. Additionally, relationships between variables were evaluated through Pearson’s correlation coefficient to explore potential associations. Diagnostic accuracy was assessed through receiver operating characteristic curve analysis. A P < 0.05 was considered statistically significant, reflecting the threshold for detecting meaningful differences or relationships in the study data.

RESULTS

There was no significant difference between the two groups regarding either of these factors. SBP was significantly elevated in the ESRD group as opposed to controls (P < 0.001; Table 1).

Table 1 Comparison of socio-demographic and clinical data between the two groups, mean ± SD/n (%).

Item		ESRD group (n = 60)	Control group (n = 30)	P value
Age (years)		51.5 ± 13.5	53 ± 12	0.5941
BMI		29.32 ± 6.4	28.83 ± 6.8	0.7401
Sex	Male	34 (56.7)	18 (60)	0.7632
	Female	26 (43.3)	12 (40)
Smoking	Smoker	13 (21.7)	6 (20)	0.8552
	Non-smoker	47 (78.3)	24 (80)
Clinical data	SBP (mmHg)	140.4 ± 14.7	125.4 ± 9.8	< 0.0011
	DBP (mmHg)	85.3 ± 7.8	84.5 ± 6.6	0.674

¹Using independent t test.

²Using χ² test.

ESRD: End-stage renal disease; BMI: Body mass index; SBP: Systolic blood pressure; DBP: Diastolic blood pressure.

In the ESRD group, the mean dialysis vintage was 5.7 ± 4.2 years, and 95.6% of patients had an A-V shunt as their access. The most common etiology of ESRD was hypertensive nephropathy (21.1%), followed by diabetic nephropathy and glomerulonephritis (12.2% each). The most important comorbidities were hypertension (70%), followed by diabetes mellitus and ischemic heart disease (23.3% each). Mean interdialytic weight gain was 2.58 ± 0.65. The most frequently administered medications were one-alpha (33.3%) and IV iron (26.7%) (Table 2).

Table 2 Clinical characteristics of included patients, mean ± SD/n (%).

Item		n = 60
Dialysis vintage (years)		5.7 ± 4.2
Access	A-V shunt	58 (95.6)
	Permcath	2 (4.4)
ESRD etiology	Analgesic nephropathy	5 (5.6)
	Diabetic nephropathy	11 (12.2)
	Glomerulonephritis	11 (12.2)
	Hypertensive nephropathy	19 (21.1)
	Obstructive uropathy	5 (5.6)
	Systemic lupus erythematosus	3 (3.3)
	Urinary tract infection	4 (4.4)
	Unknown	2 (2.2)
Comorbidities	Diabetes mellitus	14 (23.3)
	Hypertension	42 (70)
	Ischemic heart disease	14 (23.3)
	Cerebrovascular disease	1 (1.7)
	Heart failure	6 (6.7)
Interdialytic weight gain		2.58 ± 0.65
1		2 (3.3)
2		29 (48.3)
3		26 (28.9)
4		3 (3.3)
Medications used	Intravenous iron	24 (26.7)
	Ator	12 (13.3)
	Cinacalcet	8 (8.9)
	One-alpha	30 (33.3)

ESRD: End-stage renal disease.

Serum creatinine, serum PO₄, CRP, and PCO were significantly elevated in the ESRD group as opposed to controls. On the other hand, hemoglobin titres and serum ALB were significantly decreased in ESRD than in controls (Table 3).

Table 3 Comparison of laboratory data between the two groups, mean ± SD.

Item	ESRD group (n = 60)	Control group (n = 30)	P value
Hemoglobin (gm/dL)	10.83 ± 1.4	11.51 ± 2.7	0.015a,1
Creatinine (mg/dL)	11.36 ± 2.8	1.3 ± 0.5	< 0.001a,1
ALB (gm/dL)	3.88 ± 0.5	4.42 ± 0.5	< 0.001a,1
Ca (mg/dL)	9.34 ± 4.6	10.76 ± 1.7	0.071
PO4 (mg/dL)	5.23 ± 2.1	3.97 ± 0.4	0.021a,1
Urea (pre) (mg/dL)	146.6 ± 32.4	-	-
Urea (post) (mg/dL)	55.7 ± 20.7	-	-
PTH (pg/mL)	449.7 ± 301.3	-	-
Ferritin (ng/mL)	389.3 ± 253.4	-	-
CRP (mg/dL)	16.8 ± 14.3	1 ± 2.3	< 0.001a,1
PCO (pmol/mg)	239.8 ± 107.7	72.2 ± 16.7	< 0.0012

^aP value < 0.05.

¹Using Mann-Whitney test.

²Using Independent t test.

ESRD: End-stage renal disease; ALB: Albumin; Ca: Calcium; PO₄: Phosphate; PTH: Parathyroid hormone; CRP: C-reactive protein; PCO: Protein carbonyl.

Receiver operating characteristic curve analysis of serum PCO revealed excellent discriminative power at a cut-off of 115 pmol/mg, providing 93.3% sensitivity and 100% specificity (Figure 1).

Figure 1 Receiver operating characteristic curve analysis of serum protein carbonyl for predication of oxidative stress in prevalent hemodialysis patients. ROC: Receiver operating characteristic.

Analysis of the correlation between CRP and clinical and laboratory data in the ESRD group revealed significant positive correlations between serum CRP and patients’ age and PTH titres. Analysis of the correlation between PCO and clinical and laboratory data in the ESRD group revealed no significant correlations (Table 4).

Table 4 Correlation between C-reactive protein and clinical and laboratory data in end-stage renal disease group.

Item	CRP (mg/dL)		PCO (pmol/mg)
	r value	P value	r value	P value
Age (years)	0.292	0.023a	0.099	0.353
BMI	0.155	0.145	-0.014	0.899
SBP (mmHg)	-0.004	0.972	0.085	0.425
DBP (mmHg)	-0.078	0.468	0.144	0.175
Hemoglobin (gm/dL)	-0.120	0.261	-0.030	0.780
Creatinine (mg/dL)	-0.045	0.676	-0.085	0.428
ALB (gm/dL)	-0.061	0.569	-0.164	0.122
Ca (mg/dL)	0.205	0.052	-0.022	0.834
PO4 (mg/dL)	-0.028	0.795	-0.163	0.126
Urea (pre) (mg/dL)	0.033	0.757	0.186	0.079
Urea (post) (mg/dL)	0.113	0.289	0.213	0.103
PTH (pg/mL)	0.371	0.004a	-0.077	0.470
Ferritin (ng/mL)	-0.249	0.112	0.051	0.750

^aP value < 0.05.

r: Spearman correlation test; CRP: C-reactive protein; PCO: Protein carbonyl; BMI: Body mass index; SBP: Systolic blood pressure; DBP: Diastolic blood pressure; ALB: Albumin; Ca: Calcium; PO₄: Phosphate; PTH: Parathyroid hormone.

Assessment of nutritional status between the two groups revealed significantly worse nutritional status in the ESRD group as opposed to the control group. Comparison between nutritional status and PCO in the ESRD group revealed no significant associations (Table 5).

Table 5 Comparison of nutritional status between the two groups and between nutritional status and protein carbonyl in end-stage renal disease group, mean ± SD/n (%).

Item	Well	Mild	Moderate	Severe	P value
ESRD group	23 (38.3)	6 (10)	30 (50)	1 (1.7)	< 0.001a
Control group	14 (46.7)	13 (43.3)	3 (10)	0 (0)	-
PCO (pmol/mg)	245.8 ± 116.5	190 ± 57.1	249.8 ± 104.8	260	0.526

^aP value < 0.05.

ESRD: End-stage renal disease; PCO: Protein carbonyl.

DISCUSSION

Individuals with ESRD exhibit increased morbidity and mortality relative to age-matched counterparts in the general population, with cardiovascular complications and infections constituting the primary etiologys of death in this population[11].

Analysis of clinical and laboratory data showed that the ESRD group had significantly elevated SBP, serum creatinine, and serum PO₄ titres in contrast with the control group. On the other hand, hemoglobin titres and serum ALB were significantly decreased in ESRD than in controls. This was predicted as a result of chronic kidney damage that leads to failure of the excretory and endocrine functions with subsequent elevation of blood pressure, serum urea and creatinine, anemia, and hypoalbuminemia, as previously mentioned in several articles[12,13].

In the current study, assessment of CRP and PCO revealed significantly elevated titres in the ESRD group as opposed to controls. These findings are consistent with those of Oberg et al[14] who demonstrated that plasma PCO titres were markedly elevated in patients with CKD than in healthy individuals. Similarly, Colombo et al[12] reported elevated titres of oxidative stress markers in ESRD, including a significant increase in plasma advanced oxidation protein products. The same results were reported by Pavone et al[15] and Colombo et al[16]. As in ESRD patients, Pavone et al[15] and Colombo et al[16] observed a small increase in plasma PCO titres after HD as opposed to the pre-HD value.

In this study, serum PCO revealed excellent discriminative power at a cut-off of 115 pmol/mg, providing 93.3% sensitivity and 100% specificity. Dimitrijevic et al[10] conducted a study to assess the impact of oxidative stress on myocardial remodeling in ESRD. In their analysis, PCO titres were assessed in a cohort of 104 individuals undergoing HD, alongside transthoracic echocardiographic evaluation. The results showed that patients with left ventricular hypertrophy had significantly higher plasma PCO titres than those with normal ventricular geometry. There was no significant correlation between PCO and clinical and laboratory data in the ESRD group. This was contradictory to the results reported by Song et al[17] exhibited that carbonyl protein titres were correlated negatively with serum titres of ALB and transferrin. Low serum ALB titres indicate systemic inflammation and oxidative stress, as well as potential malnutrition or malabsorption. Notably, ALB serves as a key extracellular antioxidant, playing a crucial role in counteracting oxidative damage.

A novel intervention to diminish PCO titres in HD patients was reported by Ghoraba et al[18], who conducted a case-control study to explore the effect of an aerobic exercise training program during HD on oxidative stress markers, including PCO. PCO titres were significantly decreased in patients who underwent the aerobic exercise training program than in those who didn’t apply for the program.

In terms of nutritional assessment, the ESRD group exhibited markedly poorer nutritional status compared to the control group, with 38.3% of patients classified as well-nourished, 10% as mildly malnourished, 50% as moderately malnourished, and 1.7% as severely malnourished. Nonetheless, serum PCO titres and nutritional status exhibited no significant correlation within the ESRD group. However, Peng et al[19] reported more favorable nutritional outcomes, despite including an older patient population (mean age 63.47 ± 13.95 years), with 58.3% of patients well-nourished, 38.6% mildly to moderately malnourished, and only 3.1% severely malnourished. This difference may be explained by the larger sample size in their study. Lin and Hung[20] explored the association between nutritional status and HD patients through SGA. Total 96 cases were enrolled in the study, and the majority of patients (76%) had normal nutritional status, 24% had moderate nutritional status, and none had severe nutritional status. Moreover, consistent with our results, Song et al[17] reported that the SGA score was a significant predictor for serum PCO titres.

Analysis of the correlation between CRP and clinical and laboratory data in the ESRD group revealed a significant positive correlation between serum CRP and patients’ age and PTH titres.

The observed positive correlation between CRP, age, and PTH titres may reflect the complex interplay between chronic inflammation, aging, and secondary hyperparathyroidism in ESRD. Aging is associated with a progressive increase in systemic inflammatory activity, often referred to as inflammaging, characterized by elevated circulating inflammatory mediators, including CRP[21]. Additionally, elevated PTH titres in ESRD are frequently associated with increased oxidative stress, endothelial dysfunction, and vascular calcification. High PTH acts as a uremic toxin, directly contributing to vascular remodeling and damage[22]. Secondary hyperparathyroidism may further amplify inflammatory signaling pathways, contributing to oxidative damage and cardiovascular risk in this population[23]. These interrelated mechanisms may partially explain the observed associations between CRP, age, and PTH concentrations in the present study.

This study has several limitations, including a relatively small sample size and a single-center facility, which may limit the generalizability of findings. The cross-sectional design does not allow causal relationships among oxidative stress, inflammation, and nutritional status to be established; longitudinal studies are needed to confirm temporal associations. Second, although nutritional status was assessed through SGA and serum ALB, ALB may be influenced by inflammation and hydration status, limiting its specificity as an isolated nutritional marker. Additional validated nutritional tools and biomarkers could provide a more comprehensive assessment. Finally, some potential confounding factors affecting oxidative stress, such as dietary intake, physical activity, smoking status, and medication use, were not fully controlled. Larger prospective multicenter studies are recommended to confirm these findings.

CONCLUSION

Clinical and laboratory parameter comparisons between the two groups revealed significantly elevated SBP, serum creatinine, and PO₄ titres in the ESRD group compared to controls. Conversely, hemoglobin titres and serum ALB were significantly diminished in the ESRD. Evaluation of CRP and PCO titres also showed significantly elevated titres in ESRD relative to the controls. Regarding nutritional status, the ESRD group demonstrated significantly worse outcomes, with only 38.3% of patients classified as well-nourished, 10% as mildly malnourished, 50% as moderately malnourished, and 1.7% as severely malnourished. However, no significant association was found between PCO titres and nutritional status within the ESRD group.