On a global scale, estimates indicate that 5 to 7 million patients have end-stage renal disease (ESRD), a critical condition necessitating renal replacement therapies, particularly dialysis [1, 2]. Coronary artery disease (CAD) constitutes a predominant component of cardiovascular pathology in this patient population, with a staggering prevalence of 42%, a rate that is several times higher than that identified in patients without renal failure [3, 4]. ESRD patients are not only more likely to experience multivessel coronary artery involvement, but they also present with more advanced stages of the disease and remarkably worse clinical outcomes relative to individuals with intact renal function [4]. Hence, traditional risk factors for CAD, particularly those related to lipid metabolism, undergo notable alterations in ESRD, with dysregulated lipid profiles and an accelerated progression of atherosclerosis [5]. Moreover, the pathogenesis of CAD in ESRD is further complicated by the presence of non-traditional risk factors, including accumulation of uremic toxins, endothelial dysfunction, vascular calcification, and chronic inflammatory states, all of which significantly amplify the cardiovascular burden in such patients [6]. These multifactorial challenges emphasize the necessity of identifying and validating innovative prognostic markers, providing more precise risk stratification and prognostic insights for ESRD patients who also suffer from CAD.

Among various hematologic indices, the red blood cell distribution width (RDW), a measure of the variability in red blood cell (RBC) size, has garnered clinicians’ attention as a potential marker of erythrocyte heterogeneity and an indirect indicator of systemic inflammation [7]. Although RDW has traditionally been employed in the differential diagnosis of anemia, an elevation of this biomarker in the context of ESRD has emerged as a robust and independent prognostic indicator. Increased RDW has been consistently associated with adverse clinical outcomes, such as escalated mortality and cardiovascular events [8]. Furthermore, hypoalbuminemia is a marker of both mortality and morbidity in patients undergoing dialysis, with lower albumin levels being associated with elevated cardiovascular risk and poorer overall prognosis [9, 10, 11, 12]. The red cell distribution width-to-albumin ratio (RAR) is an innovative biomarker of systemic inflammation and oxidative stress in ESRD patients [13]. However, data on the association of RAR with long-term cardiovascular outcomes, including all-cause mortality, cardiovascular mortality and major adverse cardiovascular events (MACEs), remain limited in patients with ESRD and concomitant CAD.

In this study, we aimed to elucidate the link between RARs and key clinical endpoints, including all-cause mortality, cardiovascular mortality, and MACEs, in a cohort of dialysis patients with CAD. Moreover, we attempted to identify the incremental prognostic value of integrating the RAR into established clinical risk scoring systems, including the Global Registry of Acute Coronary Events (GRACE) and Gensini scores, to determine whether this novel biomarker improves cardiovascular risk prediction. This study was performed based on data acquired from a large, multicenter cohort in China.

To our knowledge, the CRUISE study has been the largest retrospective cohort to date investigating the prognostic utility of the RAR in CAD patients receiving dialysis. The outcomes of the current investigation indicated a strong association between the highest RAR tertile and increased risks of all-cause mortality, cardiovascular mortality, and MACEs. Multivariate Cox proportional hazards modeling further confirmed that elevated RARs independently predicted these adverse clinical outcomes. Additionally, RCS analysis revealed a linear association between RAR and the risk of these endpoints, emphasizing the strength and consistency of the RAR as a prognostic biomarker. Moreover, integrating RAR into the Gensini and GRACE scoring models notably improved their predictive accuracy for mortality outcomes, reflecting the utility of the RAR as an adjunct to established risk stratification tools.

The RDW, a routinely assessed hematological parameter, serves as an indicator of RBC volume heterogeneity. RDW elevation in dialysis-dependent chronic kidney disease (CKD) patients can be attributed to multifactorial processes, most notably systemic inflammation and malnutrition, both of which are prevalent in this population. Uremic toxins, exposure to dialysate, and access-related infections act as major contributors to these conditions [14]. Inflammatory cytokines and oxidative stress disrupt erythropoiesis, resulting in an elevation in the proportion of immature RBCs, thereby escalating RDW. Additionally, lipid peroxidation impairs RBC deformability and further elevates RDW [15, 16]. When RDW exceeds a threshold of 14%, it impairs the efficiency of oxygen transport, enhances blood viscosity, and exacerbates the risk of myocardial ischemia and infarction, thereby contributing to adverse cardiovascular events [17].

Chronic systemic inflammation, a hallmark of dialysis-dependent CKD, is also responsible for appetite suppression and impaired nutrient intake via pro-inflammatory cytokines, resulting in hypoalbuminemia and protein-energy wasting [18]. Albumin, a negative acute-phase reactant with significant antioxidant and anti-inflammatory properties, has a remarkable function in maintaining vascular integrity and modulating inflammatory responses. Low serum albumin levels exhibited a notable association with CAD progression and serve as a marker of poor cardiovascular outcomes in this patient population [19, 20, 21, 22]. Consistently, the current study showed that patients in the highest RAR tertile exhibited both elevated RDW and diminished albumin levels, reflecting the multifactorial nature of these biomarkers in CAD pathogenesis in ESRD.

The RAR, a composite inflammatory index, may provide a more comprehensive reflection of the systemic inflammatory burden than either RDW or albumin alone. This index has demonstrated significant prognostic value in various cardiovascular/cerebrovascular diseases [23, 24]. Notably, Kimura et al. [13] pointed out that a high RAR was predictive of the progression to ESRD, as well as of increased mortality and enhanced frequency of cardiovascular events in CKD patients. Similarly, studies analyzing the MIMIC-III database and intensive care unit cohorts have identified the RAR as an independent predictor of mortality (i.e., short-/long-term) [25, 26]. Consistently, Liu et al. [27] found that the RAR was strongly associated with an elevated risk of cardiovascular disease. The higher the RAR, the greater the all-cause mortality and cardiovascular mortality rates [27].

Notably, we observed that the highest RAR tertile (Tertile 3) comprised a greater proportion of patients receiving peritoneal dialysis relative to hemodialysis. The function of dialysis modality in modulating the risk of CAD remains an area of active investigation and debate. Hemodialysis can induce a variety of pathophysiological alterations, involving volume shifts, blood pressure fluctuations, myocardial stunning, and electrolyte imbalances, all of which have the potential to exacerbate CAD and escalate arrhythmic risk [28]. On the other hand, the glucose-rich dialysate used in peritoneal dialysis may lead to accumulation of advanced glycation end-products, metabolic syndrome, and insulin resistance, all of which are known to promote atherogenesis and increase the risk of cardiovascular events. However, to date, no definitive evidence has conclusively favored one dialysis modality over another in terms of reducing the incidence of CAD or boosting cardiovascular outcomes in this patient population [29, 30].

Subgroup analysis indicated a potential interaction between RAR and clinical outcomes in AMI vs. non-AMI patients, with a more pronounced association observed among patients with acute myocardial infarction. This differential prognostic impact likely reflects the augmented systemic and local inflammatory response observed in AMI, which activates intricate cellular death pathways, including apoptosis, ferroptosis, and autophagy. The enhanced prognostic value of the RAR in the AMI subgroup is in concordance with previous studies that have highlighted its predictive significance in patients facing acute cardiovascular events [19, 25, 26].

As a composite inflammatory index, RAR offers a more holistic view on systemic inflammation and nutritional status, potentially overcoming the limitations of RDW and albumin when used in isolation. Future studies should investigate the underlying biological mechanisms, particularly in the context of dialysis-dependent patients. The pathophysiology of dialysis, characterized by chronic inflammation, oxidative stress, and uremic toxin accumulation, likely influences the relationship between RAR and adverse cardiovascular outcomes. A deeper exploration into how RAR interacts with these factors can provide insights into why it functions as a robust predictor of mortality and cardiovascular events in this unique population. Moreover, examining interactions of the RAR with dialysis modalities (e.g., hemodialysis vs. peritoneal dialysis) will help identify whether its prognostic value varies according to the type of renal replacement therapy.

This study has several strengths and limitations. A key strength is its focus on the association between the RAR and long-term outcomes in dialysis patients with CAD. We have integrated an extensive range of inpatient data, as well as the GRACE and Gensini scores, all of which are crucial for accurate risk stratification in this high-risk cohort. However, several limitations warrant consideration. First, the retrospective nature inherently raises concerns regarding confounding factors and potential selection bias, which could affect the robustness of the findings. The reliance on historical medical records limits our ability to control for unknown or unmeasured variables, thus introducing an element of uncertainty. Additionally, the retrospective design may limit the ability to account for evolving confounders, such as fluctuations in the RAR over time or changes in treatment protocols during follow-up, especially post-discharge medication adherence. Longitudinal data collection may enable a more comprehensive understanding of how the RAR changes with disease progression and how it interacts with other clinical and laboratory factors over time. Therefore, the results of this study should be interpreted with caution, and prospective studies are recommended to validate the findings and assess the robustness of the RAR as a prognostic marker in dialysis patients with CAD. Second, although the RDW is widely accepted in clinical practice to assess anemia, our study lacked data on other principal nutritional and hematological parameters, including folate, iron, and vitamin B12, which may influence the RAR and provide additional insights into mechanistic pathways. The absence of these variables may undervalue the multifactorial nature of the prognostic potential of the RAR. Third, we focused exclusively on baseline RAR values measured at the time of hospital admission, neglecting the potential prognostic implications of longitudinal fluctuations in the RAR over the course of dialysis treatment. Longitudinal assessments of the RAR may shed light on its dynamics as a biomarker for disease progression and response to therapeutic interventions. Future research integrating serial RAR measurements alongside time-dependent analysis will be essential for delineating the full prognostic capacity of the RAR. Fourth, CAD severity in this cohort was determined using the Gensini score. This approach was efficient in assessing the extent of coronary artery involvement; however, it may have lacked the granularity provided by other scoring systems (e.g., the SYNTAX score). The SYNTAX score, exhibiting a more comprehensive assessment of lesion complexity and vascular anatomy, may provide a more detailed perspective on CAD severity and its long-term prognostic impact. Future studies incorporating both the Gensini and SYNTAX scores will enable a more detailed assessment of CAD complexity and its influence on clinical outcomes in this patient population. Future prospective studies are essential to further confirm the prognostic value of the RAR in dialysis patients with CAD. These studies should aim to collect data from a more diverse, multi-center cohort and evaluate how RAR dynamics (as opposed to static baseline measures) influence long-term outcomes. Validation in other populations, including those undergoing different forms of renal replacement therapy, will be crucial for assessing the generalizability of the predictive power of the RAR. Additionally, large-scale prospective studies may help address potential selection biases inherent to retrospective analyses.

Our study provides compelling evidence that an increased RAR is independently associated with an elevated risk of all-cause mortality and cardiovascular mortality in dialysis-dependent patients with CAD. The integration of the RAR into the Gensini and GRACE risk models remarkably amplified their predictive accuracy, reflecting the utility of the RAR as a potent biomarker for prognosis in this high-risk population. Given its widespread availability, simplicity, and strong prognostic value, the RAR has emerged as a promising tool for routine clinical usage in evaluating dialysis patients with CAD. It holds noticeable potential for improving risk stratification, guiding clinical decision-making, and ultimately enhancing patient outcomes in this vulnerable cohort.

AIC, Akaike Information Criterion; AMI, Acute Myocardial Infarction; ANOVA, Analysis of Variance; BIC, Bayesian Information Criterion; CABG, Coronary Artery Bypass Grafting; CAD, Coronary Artery Disease; CI, Confidence Interval; CRUISE-R, Coronary Revascularization in Patients on Dialysis in China - Retrospective; CKD, Chronic Kidney Disease; eGFR, Estimated Glomerular Filtration Rate; ESRD, End-Stage Renal Disease; GRACE, Global Registry of Acute Coronary Events; HDL-C, High-Density Lipoprotein Cholesterol; HR, Hazard Ratio; IDI, Integrated Discrimination Improvement; IQR, Interquartile Range; LVEF, Left Ventricular Ejection Fraction; MACEs, Major Adverse Cardiovascular Event(s); NRI, Net Reclassification Improvement; PCI, Percutaneous Coronary Intervention; RAR, Red cell distribution width-to-Albumin Ratio; RBC, Red Blood Cell; RCS, Restricted Cubic Spline; RDW, Red Blood Cell Distribution Width; SD, Standard Deviation; STROBE, Strengthening the Reporting of Observational Studies in Epidemiology.

The datasets involved in the current study (CRUISE-R) are unavailable publicly owing to restrictions on data sharing imposed by ethical considerations, patient confidentiality agreements, and institutional policies. However, access to these data may be granted upon formal request to the corresponding author, subject to reasonable terms and conditions, contingent upon adherence to appropriate ethical review and approval processes, and subject to necessary data privacy and security protocols.

XZ: Conceptualization, Project administration, Formal analysis, Validation, Writing—Original Draft, Data curation. EX: Investigation, Resources, Data curation, Writing—Original Draft. ZZ: Investigation, Methodology, Data curation, Writing—Original Draft. DS: Investigation, Data curation, Writing—Original Draft. SS: Investigation, Data curation, Writing—Original Draft. HH: Investigation, Data curation, Writing—Original Draft. WL: Investigation, Writing—Original Draft. LC: Resources, Visualization, Supervision, Writing—review and editing. HZ: Resources, Visualization, Methodology, Supervision, Writing—review and editing. JZ: Conceptualization, Project administration, Supervision, Funding acquisition, Writing—review and editing. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was carried out in accordance with the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of The China–Japan Friendship Hospital (Protocol No. 2022-KY-075-1). Consent to participate statement: The necessity of informed consent was eliminated because of the study’s observational, non-invasive nature, with measures taken to preserve patients’ anonymity and confidentiality.

The authors extend their sincere gratitude particularly to participants and their families who contributed to the study, as well as to the nursing, research, and clinical teams across all hospitals involved. Their dedication, professionalism, and support were indispensable for the successful execution of this research, and their contributions are deeply appreciated.

This research was funded by the National High Level Hospital Clinical Research Funding (grant number of 2024-NHLHCRF-YS-01), Capital’s Funds for Health Improvement and Research (grant number of 2022-1-4062), the Beijing Research Ward Construction Clinical Research Project (grant number of 2022-YJXBF-04-03).

The authors declare no conflict of interest.

During the manuscript preparation process, the authors utilized ChatGPT as an AI-powered language tool to assist in refining the readability, coherence, and clarity of the text. Following the AI-generated enhancements, the authors rigorously reviewed, revised, and edited the manuscript.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/RCM47939.