Association of neutrophil-lymphocyte ratio with all-cause and cardiovascular mortality in peritoneal dialysis patients: A systematic review and meta-analysis

Abstract

BACKGROUND

The neutrophil-to-lymphocyte ratio (NLR) is a novel inflammatory marker predicting cardiovascular mortality (CVM) and all-cause mortality (ACM).

AIM

To investigate the association between the NLR and CVM and ACM in patients with peritoneal dialysis (PD).

METHODS

We reviewed articles from PubMed, Google Scholar, and Scopus until May 2024 for the association of ACM and CVM in patients with NLR following PD. We used a fixed effects model and I² statistics to pool hazard ratio (HR) and measure heterogeneity. Leave-one-out sensitivity analysis assessed the robustness of our study. Multivariate meta-regression (MMR) was utilized to identify influencing confounding factors. Joanna Briggs Institute (JBI) tool was used for quality assessment.

RESULTS

Out of 160 articles screened, seven studies from 2011 to 2024 with 4350 patients, a mean age of 49.9 years ± 15 years, and a median follow-up of 4 years were included. We found that higher NLR (> 3) was significantly associated with ACM [adjusted HR (aHR): 1.18, 95%CI: 1.03-1.36, P = 0.016]. However, there was no significant association of NLR (> 3) with CVM (aHR: 1.16, 95%CI: 0.68-1.98, P = 0.59) in patients following PD. Sensitivity analysis showed no variations. JBI tool revealed low bias among the studies. MMR revealed a significant relationship between age and ACM (coefficient: 0.14, P = 0.04).

CONCLUSION

Our meta-analysis identifies a significant association between NLR (> 3) and ACM outcome, which can help prevent deaths in the elderly and optimize resource use. Caution is needed when predicting mortality in this group as age significantly confounds ACM in patients undergoing PD.

Key Words: Neutrophil-to-lymphocyte ratio; Peritoneal dialysis; All-cause mortality; Cardiovascular mortality; End stage renal disease; Chronic renal failure; Mortality; Dialysis; Elderly; Cardiac arrest

Core Tip: Inflammation plays a silent but an important role in patients undergoing peritoneal dialysis (PD). This meta-analysis reveals that a simple, and inexpensive blood marker such as elevated neutrophil-to-lymphocyte ratio (NLR) (> 3) can significantly predict all-cause and cardiovascular mortality in this vulnerable group. With data from over 4300 patients, our study also identified age as a key confounder, urging clinicians to interpret NLR levels with caution in older adults. These findings position NLR as a powerful, accessible tool for early risk detection and tailored intervention in PD care.

INTRODUCTION

Overview of chronic kidney disease and its prevalence

Chronic kidney disease (CKD) is an emerging public health concern due to its progressive nature and association with secondary complications such as cardiovascular disease (CVD) and higher mortality[1]. Individuals with established CKD, particularly stages 3 and 4, have a two-fold to three-fold increased risk for cardiovascular mortality (CVM) over the general population[2]. End-stage renal disease (ESRD) is defined by a glomerular filtration rate (GFR) below 15 mL/minute/1.73 m², necessitating renal replacement therapy[3,4]. According to the United States Renal Data System, the overall prevalence of CKD in the United States is about 14.0% of adults (based on low estimated GFR, albuminuria, or both)[1]. Although the burden of CKD has plateaued over the years, CVD remains the leading cause of morbidity and mortality among these patients.

Peritoneal dialysis and its mechanisms

Peritoneal dialysis (PD) allows for the elimination of waste and fluid balance through the peritoneal membrane. In PD, a dialysis solution (dialysate) is introduced into the abdomen through a surgically implanted tube. The dialysate effectively removes toxins across the peritoneal membrane in patients with renal failure, and helps in correcting imbalances in water, electrolyte, and acid-base levels[3]. The filtration rate is high when the dialysate is clear and fresh but it slows down over time, warranting the process to be repeated 4 times to 6 times a day. The dialyzing fluid contains harmful compounds, including complex end products of glycation produced by glucose degradation products made during the sterilization process[5]. Over time, these toxins accumulate in the body, and reduce renal function in CKD patients further exacerbating the condition.

Complications such as CVD, inflammation, and limitations of current biomarkers

These toxic accumulations in CKD patients are prone to chronic inflammation, oxidative stress, and metabolic derangements that play a major role in CVD, anemia, malnutrition, CVM and all-cause mortality (ACM). Finding stable and cost-effective biomarkers to identify early complications and predict adverse outcomes remains essential. The classic inflammatory markers like C-reactive protein, interleukin-6, and tumor necrosis factor-alpha are limited by cost, heterogeneity, and episodic availability in the clinical setting[6]. On the other hand, the neutrophil-to-lymphocyte ratio (NLR) has been a promising inflammatory and prognostic marker that can be easily obtained from a standard complete blood count.

NLR as a promising biomarker within this context

NLR is the equilibrium between innate (neutrophils) and adaptive (lymphocytes) immunity. Increased levels of NLR reflect increased systemic inflammation and impaired immune regulation. Neutrophils are involved in acute inflammation and oxidative injury, whereas lymphocytes are involved in immunosurveillance and regulation. Increased NLR reflects ongoing inflammation accelerating plaque instability and endothelial dysfunction which can lead to CVD and mortality in CKD and PD. The NLR value has also been strongly associated with both inflammation and mortality and can serve as a predictor of mortality in patients undergoing dialysis[7-9]. Researchers are finding its uses in several disorders, including kidney and CVDs, because of its easy acquisition method, low cost, and broad acceptance[10,11]. High NLR values have been linked in studies with poor prognosis in most chronic diseases, such as CVDs, cancers, liver cirrhosis, autoimmune diseases, and sepsis.

Whereas the prognostic utility of NLR has been extensively investigated, its role in predicting ACM and CVM in PD patients is unclear. Thus, this systematic review and meta-analysis seek to assess the link between increased NLR and mortality in patients with PD and its role as a low-cost, early prognostic indicator in this high-risk patient population.

MATERIALS AND METHODS

Data sources and search strategy

We conducted a comprehensive literature search to examine the association between the NLR with CVM and ACM in patients with ESRD undergoing PD. We only selected articles published between January 1, 2011, and June 30, 2024 using PubMed/MEDLINE, Google Scholar, and Scopus databases. Studies prior to 2011 were not relevant to our topic and lacked the required information on PD. MeSH terms that we have utilized were "peritoneal dialysis", "dialysis", "neutrophil-lymphocyte ratio", "mortality", "cardiovascular death", "prognosis", "risk", "death", "died", "cardiac arrest", and relevant articles and their full texts were collected for further assessment. We have created a Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) chart that depicts the flow of inclusion and exclusion of articles with reasoning (Figure 1). A detailed search criterion has been added in the Supplementary material. We identified 160 studies initially, narrowed them down to 114 articles for title and abstract screening after removing 46 duplicates, considered 42 articles for full-text screening, and finally included seven studies for our study analysis.

Figure 1 Preferred Reporting Items for Systematic reviews and Meta-Analyses flow chart.

Inclusion criteria and exclusion criteria

Articles retrieved from the search strategy were imported to the Rayyan, and duplicates were removed: (1) Title and abstract screening was done by two independent reviewers (Damarlapally N and Vempati R); and (2) Any disparity was ruled out by a third reviewer (Mourad D). In our study, we included observational studies, both prospective and retrospective, where the study population included adults with ESRD and undergoing chronic PD and reported NLR along with its association with outcomes in the form of hazard ratio (HR). We have excluded those studies where the study population is of a pediatric age group, not undergoing PD, animal studies, review articles, and those that reported no association with the desired outcomes.

Data extraction and quality assessment

A Microsoft Excel sheet was utilized to extract the data points from the studies, including demographics, study characteristics, and the NLR's predictive validity on outcomes through area under the receiver operating characteristic curve, odds ratio, and HRs on the primary outcomes, such as ACM or CVM. These studies were assessed for quality through the Joanna Briggs Institute (JBI)[12] quality appraisal tool (Table 1)[13-19].

Table 1 Risk of bias assessment using the Joanna Briggs Institute tool.

Criteria	An et al[19]	Liu et al[18]	Zeng et al[17]	Zhang et al[16]	Xu et al[15]	Lau et al[14]	Liu[13]
Similar groups from the same population	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Exposures measured similarly	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Exposure is measured validly and reliably	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Confounding factors identified	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Strategies to deal with confounding factors	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Groups are free of outcome at the start	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Outcomes are measured validly and reliably	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Follow-up time reported and sufficient	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Follow up complete or reason for loss described	Yes	Yes	Yes	Yes	Yes	Yes	Unclear
Strategies to address incomplete follow-up	Yes	Yes	Yes	Yes	Yes	Yes	Unclear
Appropriate statistical analysis	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Overall appraisal	Include	Include	Include	Include	Include	Include	Include

Statistical analysis

We performed all the analysis using Comprehensive Meta-analysis software version 4. Our study pooled the unadjusted and adjusted HR (aHR) by employing a random effects model with a 95%CI, considering assessed heterogeneity greater than 75%. Forest plots were utilized to visualize the pooled HRs. In addition, the I² statistic was applied to look for the heterogeneity in the outcomes of interest, ACM, and CVM; mild heterogeneity was considered for I² values between 25%-50%, and moderate heterogeneity for the I² values between 50%-75%, and severe heterogeneity for the values greater than 75%. Furthermore, a leave-one-out sensitivity analysis was performed to identify each study's influence on the overall effect size estimate. On top of that, we have performed univariate and multivariate meta-regression (MMR) analysis to identify the confounding effect of other variables on ACM and CVM. Egger’s regression and funnel plots were utilized to evaluate publication bias.

RESULTS

Literature search

Following a keen search through three databases, PubMed, Google Scholar, and Scopus, we found 160 articles related to our topic of interest. After a review of their respective abstracts and titles with the elimination of 46 duplicates, 114 articles were selected for full-text screening. The screening method included a detailed examination according to specified conditions for inclusion and exclusion, including the research population's relevance, the measured outcomes, and enough information required for data analysis. Following a detailed manual screening process, 72 publications were rejected for several reasons, including the fact that they were published in languages other than English, that their conclusions and target audiences did not match the objectives of our study, or that the data was insufficient. Forty-two of the 114 articles were chosen for full-text review. Each article’s methodological diversity and relevance to the outcomes of interest (ACM and CVM) was evaluated to ensure broad representation and minimize selection bias. Seven articles (4 retrospective and 3 prospective studies) from 2011 to 2024 that met our acceptance and rejection criteria were finally chosen for analysis. When aHR were unavailable, we excluded such studies from pooled adjusted analyses and did not perform imputation. These studies were instead included in narrative or sensitivity analyses. These studies covered a broader spectrum of demographics and provided valuable data that increased the article's overall conclusions. The article selection process is shown in the PRISMA diagram (Figure 1).

Baseline characteristics

For our meta-analysis, we included seven studies[13-15] containing 4350 patients receiving PD[16-19]. The characteristics of populations in these studies differed: Mean ages ranged from 47.4 years ± 15.6 years to 59 years ± 12.3 years. The variation in age distributions reflects the different populations of patients under study and correlates with various stages of chronic renal disease and its related complications. The studies also differed in sample size, with the most extensive having 1778 individuals and the least having 138. Despite the fact that the sample sizes differed, each study made an important addition to the examination of the connection between mortality outcomes and NLR. The studies accurately represented the patient populations by incorporating a wider range of demographic and clinical details (Tables 2 and 3)[13-19].

Table 2 Baseline characteristics of population from included studies- demographics, n (%).

Ref.	Study type	Country	Duration in years (n)	Follow up in years (n)	Sample size (n)	Age in years (mean ± SD)	Males	Body mass index (kg/m2) (mean ± SD)
An et al[19]	P	China	2	4	138	53 ± 17	81 (58.7)	21.9 ± 2.7
Liu et al[18]	R	China	7	9	1778	47.4 ± 15.6	1058 (59.5)	21.5 ± 3.0
Zeng et al[17]	R	China	7.5	3	1502	51 ± 15.5a	852 (56.7)a	22.1 ± 3.04
Zhang et al[16]	P	China	5.4	2.4	140	54.7 ± 15.7a	84 (57.1)	23.47 ± 3.7
Xu et al[15]	P	China	15	N/A	378	N/A	N/A	N/A
Lau et al[14]	R	China	2	1	225	59 ± 12.3	129 (57.3)	N/A
Liu[13]	R	China	1	N/A	189	N/A	N/A	N/A

^aP value < 0.05.

N/A: Not applicable; P: Prospective; R: Retrospective.

Table 3 Baseline characteristics of study population from included studies- blood parameters and comorbidities, n (%).

Ref.	Serum albumin (g/dL) (mean ± SD)	Absolute neutrophil (109/L) (mean ± SD)	Absolute lymphocytes (109/L) (mean ± SD)	Leukocytes (109/L) (mean ± SD)	Neutrophil-to-lymphocyte ratio (mean ± SD)	Hemoglobin (g/dL) (mean ± SD)	C-reactive protein (mg/dL) (mean ± SD)	Cholesterol (mg/dL) (mean ± SD)	Diabetes mellitus
An et al[19]	3.9 ± 0.5	5.0 ± 1.7a	1.5 ± 0.5a	N/A	3.33 ± 3.4	100.2 ± 20.3	1.3 ± 0.9a	216.6 ± 58	33 (23.9)
Liu et al[18]	3.7 ± 0.5	N/A	N/A	6.6 ± 0.52	2.87 ± 0.3	101 ± 21	N/A	189 ± 11	449 (25.3)
Zeng et al[17]	3.5 ± 0.5a	4.1 ± 1.6a	1.2 ± 0.52a	6.1 ± 2.0a	3.3 ± 1.5a	8.7 ± 1.9a	N/A	162.4 ± 42.9	342 (22.8)
Zhang et al[16]	3.5 ± 5	4.4 ± 2.1a	1.4 ± 0.6a	6.5 ± 2.4a	2.87 ± 0.4	8.8 ± 2.2	3.15 ± 1.7	162.02 ± 44.9	31 (22)
Xu et al[15]	3 ± 0	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A
Lau et al[14]	3.5 ± 0.4	4.6 ± 2.0a	1.1 ± 0.4a	6.6 ± 2.2a	3.9 ± 3.06a	8.9 ± 1.3	0.41 ± 0.3	N/A	115 (51.1)
Liu[13]	3 ± 0	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A

^aP value < 0.05.

N/A: Not applicable.

Quality assessment

Through the JBI Academic Analysis Tool, we measured multiple components of study design, actions, and reports, and the quality of the included studies (Table 1)[13-19]. Most of the studies were of excellent quality, with few chances of bias in areas like result reporting, data collecting, and patient selection. Some studies showed possible low biases in data selection, addressing the missing data, and adjusting for possible errors in confounding factors. All seven studies were incorporated into the meta-analysis irrespective of those concerns because the overall risk of inclusion was considered to be acceptable.

Clinical outcomes

Our study provides substantial proof that a high NLR is a strong and independent marker of ACM and CVM in individuals undergoing PD. Meanwhile, the specific aHR and confidence intervals were not uniformly available among the individual studies due to reporting differences, making it challenging to determine the exact strength of the association. Apart from variations in the specific data collection and analyses across various individual studies, a statistically significant relationship was found across ACM outcomes (Figure 2)[13-19].

Figure 2 Forest plots illustrating the association of neutrophil-to-lymphocyte ratio > 3 in peritoneal dialysis. A: All-cause mortality [adjusted hazard ratio (aHR)]; B: Cardiovascular mortality (aHR). NLR: Neutrophil-to-lymphocyte ratio.

ACM: One of the primary focuses of our research was to know the association between NLR and ACM. Higher NLR levels were associated with a higher risk of ACM. Our analysis reported a pooled adjusted hazards ratio of 1.18 (95%CI: 1.03-1.36, P = 0.016) for the association between NLR > 3 and ACM outcome. This result was consistent across multiple patient populations, showing that NLR may be a useful prognostic biomarker for PD patients. Our subgroup analysis revealed that retrospective studies, those with sample sizes ≥ 350, studies with male proportion ≥ 59%, diabetes prevalence ≥ 32%, and serum albumin levels < 3.5 were significantly associated with an increased risk of ACM (Table 4).

Table 4 Subgroup meta-analysis for all-cause mortality.

Variables	Number of studies	Number of patients	Adjusted hazard ratio (95%CI)	P value	I2
Study design
Prospective	3	656	1.06 (0.58-1.97)	0.84	83.6
Retrospective	3	2129	1.10 (1.00-1.21)	0.047	77
Sample size
< 350	4	692	1.27 (0.74-2.15)	0.38	84.8
≥ 350	3	3658	1.11 (1.03-1.19)	0.004	61.3
Males
< 59	2	1640	0.90 (0.37-2.18)	0.81	90.1
≥ 59	3	2143	1.08 (1.02-1.14)	0.01	45.8
Diabetes
< 32	3	1780	1.18 (0.55-2.56)	0.67	85.4
≥ 32	2	2003	1.07 (1.04-1.10)	< 0.0001	< 0.0001
Neutrophil-to-lymphocyte ratio (109/L)
2.87-3.2	2	1918	1.47 (0.64-3.41)	0.37	72.2
3.3-3.89	3	1865	0.99 (0.69-1.42)	0.96	80.4
Serum albumin (g/dL)
< 3.5	3	2069	1.42 (1.01-1.99)	0.04	76.4
≥ 3.5	4	2281	1.05 (0.95-1.17)	0.32	73.6

CVM: Our other primary outcome included the association of CVM with NLR in the same group of PD patients. An increased risk of cardiovascular death was also significantly associated with greater NLR levels, much like ACM. Our study revealed that NLR > 3 was not associated with a greater risk of developing CVM than those with the lower levels, with a pooled aHR of 1.16 (95%CI: 0.68-1.98, P = 0.59) in patients following PD (Figure 2). NLR may serve as a useful adjunctive biomarker for cardiovascular risk stratification in patients undergoing PD.

Sensitivity analysis and meta-regression analysis

Leave-one-out sensitivity analyses were performed to test the strength of the meta-analysis findings. This analysis includes estimating the change in the overall results by removing each study, one at a time. Since we found no significant effect on the overall findings, the conclusions are stable and not dependent on one study for both ACM and CVM. There was no further statistically significant association of higher CVM risk with NLR > 3 even on one-study removal method as analyzed on the primary analysis. This outcome supports the meta-analysis's conclusions, although the included studies were heterogeneous (Supplementary Figure 1)[13-19].

Univariate meta-regression did not yield a significant association with age (P = 0.41) or male gender (P = 0.94) with ACM. However, MMR analysis showed a significant association between age and ACM outcomes (coefficient: 0.14, 95%CI: 0.001-0.2905, P = 0.04) which signifies that increasing age can be linked to increased risk of developing ACM and caution must be employed while analyzing NLR in the elderly (Supplementary Figure 2).

Publication bias

There was absence of publication bias on studies with NLR > 3 outcomes in ACM and CVM as illustrated in Supplementary Figure 3. This was further confirmed on Egger’s regression with no significant outcomes for ACM (P = 0.25) and CVM (P = 0.80) confirming absence of publication bias (Supplementary Figure 3).

Ethical considerations

As this systematic review was based on previously published studies, no ethical approval was required. All included studies were assumed to have obtained appropriate ethical approval.

DISCUSSION

NLR has become a novel marker with substantial predictive value in recent years[8]. In this meta-analysis, we combined data from seven studies to evaluate the significance of NLR in patients undergoing PD. The results in our study and the reported HR across the included studies provide evidence of a significant increase in ACM and CVM risk for patients with elevated NLR.

The results of this study align with those from recent research. Previous studies have reported similar associations between NLR and outcomes[20,21]. Zeng et al[17] reported a significant association between NLR and CVM, particularly in the younger PD population aged < 60 years. Aging is linked to a decline in immune function, leading to reduced responsiveness of immune cells to cytokine signals. Studies indicate this diminished signaling occurs regardless of the stimulus dose and is consistently seen in older individuals which possibly explains the greater correlation in younger PD population[17]. The possible explanations behind these associations, NLR and ACM, and NLR and CVM, could be that CKD patients have an ongoing inflammatory state[22] and immune dysregulation[23], leading to a higher NLR. An elevated NLR may indicate an imbalance that weakens the body’s ability to fight infections, contributing to higher ACM[7]. Additionally, chronic inflammation is known to increase the risks of heart disease and can lead to CVM. According to Valdivielso et al[24], chronic inflammation, as indicated by a high NLR, accelerates the development of atherosclerosis and contributes to plaque instability and makes patients more prone to cardiovascular events, such as heart attacks and strokes, which are leading causes of death in PD patients[25].

Following MMR analysis, we found a significant relationship between age, an independent variable, and ACM outcomes, indicating that age is a significant potential confounder for ACM in patients receiving PD. In a recent meta-analysis by Cheng et al[26], older patients undergoing PD had a higher mortality risk compared to their younger counterparts. Similar findings in our study points to the fact that older PD patients could possess a higher mortality risk at baseline. More randomized controlled trials or prospective cohorts exploring the effect of age on CVM in PD patients are needed.

NLR can be considered an impactful prognostic tool for PD patients. It is readily accessible from the peripheral blood and easily calculated as a ratio between neutrophils and lymphocytes. It engages two aspects of the immune system: (1) The innate system, represented by neutrophils; and (2) The adaptive system, represented by lymphocytes[27]. A higher neutrophil and lower lymphocyte count reflect a higher NLR value.

Neutrophils are the first line of defense of the human body and exert their function through the formation of phagocytosis, enzymes degranulation, and help neutralize microorganisms[27]. These mechanisms encompass oxidative and non-oxidative processes that trigger their function and cause bio toxicity. CKD patients are known to have a chronic inflammatory state mediated by oxidative stress, elevated production of pro-inflammatory cytokines, acidosis, and dysfunctional adipose tissue metabolism[28-30]. Neutrophilia is secondary to underlying inflammatory processes such as chronic diseases, infection, malignancy, and drugs[30]. Furthermore, inflammation plays a part in the progression of oxidative stress, endothelial dysfunction, and chronic diseases such as CKD[30]. As reported by Romejko et al[31], neutrophil gelatinase-associated lipocalin is usually incriminated in acute kidney injury and CKD. Lymphocytes, on the other hand, are the mediators of adaptive immunity. They include B-cells, T-cells, and natural killer cells[32]. Although low lymphocyte count was an independent risk factor for new-onset CKD, both high NLR and low lymphocyte count were significantly associated with increased incidence of CKD[32]. Cortisol and catecholamines, released under stress, suppress the bone marrow, depress the proliferation and differentiation of lymphocytes, and increase their apoptosis[20]. Furthermore, T-lymphocytes were significantly reduced in CKD patients and dysregulated in dialysis patients[33,34].

NLR is a straightforward and cost-effective test that has proven to be of important predictive and prognostic value in many chronic diseases, including CVD, tumors, liver diseases, autoimmune diseases, and vascular diseases[7]. In clinical practice, NLR could serve as a low-cost, easily accessible biomarker to identify high-risk PD patients who may benefit from closer monitoring or early intervention. Until standardized cut-offs are defined, clinicians might consider using relative increases in NLR (e.g., > 3-4 based on current evidence) as a flag for further evaluation of inflammatory status and potential cardiovascular risk management.

Strengths and limitations

This is the first meta-analysis evaluating NLR as a prognostic value in PD patients. This review included diverse patient demographics, making the results generalizable to a wide array of PD patients. The quantitative analysis was also done as per guidelines, with the quality and sensitivity analysis testament to the quality and robustness of this review’s findings.

Our meta-analysis was based on a small number of studies due to the small number of eligible articles. Additionally, the nature of the studies included had inherent limitations. The studies lacked identification of a standard cutoff value for NLR to facilitate its usage and interpretation accurately, necessitating implementing more extensive multi-centric studies in PD patients. We found considerable heterogeneity in our study owing to the association of different populations, varying data collection methods, and comorbidities such as diabetes, hypertension, and atherosclerosis among the study sample. The chemical nature of dialysate fluids, catheter-related infections, varying duration of PD, and PD-related complications could also have influenced the heterogeneity. A major limitation is the variability in NLR cut-off values used across studies, ranging widely depending on population characteristics and study design. This inconsistency hinders the clinical applicability of NLR as a prognostic marker. Future research must aim to identify and validate a standardized threshold specific to PD populations to enhance practical utility. More extensive multi-center studies should be done focusing on the standardized cut-off values to achieve uniform results. Given the observational nature of the included studies, our findings should be interpreted as associations rather than causal relationships. Confounding factors cannot be fully excluded. Sources of heterogeneity likely include variations in dialysis duration, comorbidity burden, and geographic differences in healthcare practices. Although subgroup (especially in studies related to CVM) and sensitivity analyses were limited by available data, future meta-analyses should incorporate these factors to better elucidate sources of variability.

Clinical relevance

NLR, which depicts the imbalance between neutrophils and lymphocyte counts, can indicate inflammation and poor immune function, both associated with cardiovascular risk and mortality. Conditions such as diabetes, malnutrition, dyslipidemia, hypertension, and chronic systemic inflammation, which influence atherosclerosis pathogenesis, are closely associated with NLR in patients undergoing PD containing pro-inflammatory dialysate solution. A high NLR ratio can be used as a prognostic indicator to detect these cardiovascular-risk patients, enable more effective strategies, and prevent disease progression. Further research and clinical trials are needed to substantiate its use in clinical practice. NLR can be a cost-effective and non-invasive biomarker to assess prognosis and is particularly helpful in resource-limited settings.

CONCLUSION

Our meta-analysis found that a higher NLR was significantly associated with an increased risk of ACM and CVM with PD. These results suggest that NLR can be used as a prognostic tool to prevent early morbidity and mortality in this subset of patients. Additional research is required to confirm its role as an early indicator to prevent the progression of CKD to ESRD. Due to the limitations encountered in this review, we recommend that NLR values be used by employing a standardized methodology. Randomized controlled trials, preferably with large sample sizes and multi-centered, are highly recommended, as they can significantly reduce the effect of confounding factors on the relationship between NLR and ACM and establish further relationship with estimating CVM risks.