Abstract
Background
The neutrophil-to-lymphocyte ratio (NLR) is a biomarker of systemic inflammation and has been associated with adverse outcomes in dialysis patients. Its role in predicting the future risk of peritoneal dialysis-associated peritonitis (PDAP) before clinical onset remains unclear.
Objective
To evaluate the association between pre-peritonitis NLR levels and the risk of developing PDAP.
Methods
We conducted a retrospective, matched (1:1) case–control study involving patients on peritoneal dialysis at a single center between January 2010 and October 2024. Cases were patients who developed a first episode of PDAP (diagnosed per ISPD guidelines). Controls were matched to cases on sex and age (±3 years). The exposure was NLR measured from routine blood tests during a 3-month period preceding the peritonitis event for cases, or a corresponding pre-index period for controls. Logistic regression analysis was used to assess the association.
Results
A total of 178 patients (89 cases and 89 matched controls) were included in the analysis. In conditional logistic regression models accounting for the matched design, a higher NLR was independently associated with an increased risk of PDAP. When analyzed as a continuous variable (per 1-unit increase in ln(NLR)), the fully adjusted odds ratio (OR) was 2.25 (95% confidence interval [CI]: 1.26–4.02, p = 0.006). When NLR was categorized into tertiles, patients in the highest tertile (NLR ≥ 1.24) demonstrated a consistent positive association with peritonitis risk compared to those in the lowest tertile (NLR < 0.52), with a fully adjusted OR of 11.00 (95% CI: 3.06–39.52, p < 0.001; P for trend < 0.001). Restricted cubic spline analysis revealed a significant linear dose–response relationship (P for non-linearity = 0.511). An elevated NLR (≥ median) also showed moderate discriminative ability for predicting PDAP, with an area under the curve (AUC) of 0.78 (95% CI: 0.71–0.85).
Conclusion
A higher NLR measured during a clinically stable period prior to infection is associated with an increased risk of subsequent PDAP. If validated in prospective studies, NLR may serve as a simple and readily available biomarker for risk stratification in this population.
Keywords: biomarker, inflammation, neutrophil-to-lymphocyte ratio, peritoneal dialysis, peritonitis, risk factor
1. Introduction
Peritoneal dialysis (PD) is a widely utilized renal replacement therapy for end-stage renal disease. However, its clinical course is frequently complicated by peritoneal dialysis-associated peritonitis (PDAP), a major cause of morbidity, technique failure, and mortality in this population, with mortality rates ranging from 2 to 6% (1). Beyond acute morbidity, recurrent peritonitis episodes and the chronic inflammatory state inherent to PD patients contribute to peritoneal membrane damage, fibrosis, and ultimately, PD discontinuation. Therefore, identifying modifiable risk factors for PDAP is crucial for improving patient outcomes and preserving peritoneal membrane function.
The neutrophil-to-lymphocyte ratio (NLR), a readily calculated hematologic index derived from routine complete blood counts, has emerged as a robust and accessible biomarker of systemic inflammation and immune dysregulation (2). In the context of end-stage renal disease, elevated NLR has been consistently associated with adverse outcomes. Cross-sectional and cohort studies in hemodialysis (HD) populations have linked higher NLR to increased systemic inflammation and poor survival (3). Furthermore, recent evidence suggests that NLR measured at the time of diagnosis is predictive of treatment failure in established PDAP (4). However, no studies have examined whether NLR levels prior to the onset of peritonitis are associated with the development of PDAP in patients undergoing PD.
Despite these associations, a critical gap remains in understanding whether a pre-existing, subclinical inflammatory state-as reflected by NLR-predisposes patients to the initial development of peritonitis. Specifically, no studies have investigated the association between NLR levels measured during a stable, infection-free period and the subsequent risk of incident PDAP.
To address this knowledge gap, we conducted a retrospective case–control study. Our primary objective was to evaluate whether an elevated NLR, measured up to 3 months prior to clinical presentation, is associated with an increased risk of developing PDAP in patients undergoing peritoneal dialysis.
2. Methods
2.1. Study design and setting
We conducted a retrospective, matched case–control study at the Department of Nephrology, The Fifth Affiliated Hospital of Wenzhou Medical University. The study period spanned from January 2010 to October 2024. Data were extracted from electronic medical records between November 2024 and January 2025. The study protocol was approved by the Medical Ethics Committee of the Fifth Affiliated Hospital of Wenzhou Medical University (Approval No: 2025(I)-081–01), which waived the requirement for informed consent due to the retrospective nature of the study.
2.2. Participant selection
Case Definition and Identification: We used retrospective consecutive enrollment to identify all first-episode PD-associated peritonitis cases that met the ISPD diagnostic criteria. Cases were adult patients (age ≥18 years) undergoing peritoneal dialysis (PD) who developed their first episode of PD-associated peritonitis (PDAP) during the study period. Peritonitis was diagnosed according to the International Society for Peritoneal Dialysis (ISPD) guidelines, requiring at least two of the following criteria: (1) abdominal pain with cloudy dialysate, with or without fever; (2) dialysate white blood cell count >100 cells/μL with ≥50% polymorphonuclear neutrophils; or (3) positive dialysate culture.
Control Selection: Controls were selected using 1:1 nearest-neighbor greedy matching without replacement, based on sex and age (±3 years) to minimize confounding. For each case, one control was drawn from the pool of PD patients who had not experienced peritonitis by the time of the case’s peritonitis event (the index date). If a suitable match could not be found, the case was excluded from the analysis (5).
Inclusion and Exclusion Criteria: For both cases and controls, inclusion required continuous follow-up at our center for at least 3 months prior to the index date and availability of complete baseline clinical and laboratory data. We excluded patients who met any of the following criteria: (1) peritonitis secondary to traumatic injury, intestinal perforation, tunnel/exit-site infection, or biliary tract infection; (2) current use of immunosuppressive therapy or presence of a primary immunodeficiency; (3) diagnosis of any active infectious disease (e.g., pulmonary, urinary tract, gastrointestinal) or non-infectious inflammatory condition (e.g., active autoimmune disease, inflammatory bowel disease flare) within 3 months before the index date. Data extraction was performed independently by two researchers who were blinded to case/control status. The detailed patient selection process is summarized in Figure 1.
Figure 1.
Patient enrollment flowchart.
2.3. Definitions of exposure and covariates
The primary exposure variable was the neutrophil-to-lymphocyte ratio (NLR), calculated as the absolute neutrophil count divided by the absolute lymphocyte count from a single routine peripheral blood test obtained during the predefined 3-month window.
For cases, the exposure NLR was defined as the value measured during a 3-month period immediately preceding the diagnosis of peritonitis (the index date), but excluding any laboratory tests performed after the onset of symptoms (2).
For controls, the exposure NLR was obtained from a blood test performed during a corresponding 3-month window prior to their individually assigned index date. This index date for each control was defined as the date on which they had accumulated the same duration of PD follow-up as their matched case had at the time of peritonitis onset. This approach ensured temporal comparability of exposure assessment between cases and controls.
Baseline covariates were collected from the same pre-index date 3-month window and included: demographic factors (age, sex, education level), clinical characteristics (dialysis vintage, smoking status, history of hypertension, history of diabetes mellitus, primary renal disease, body mass index), and laboratory parameters (hemoglobin, albumin, globulin, serum calcium, blood urea nitrogen, serum creatinine, parathyroid hormone, C-reactive protein, total white blood cell count, and urea clearance index [Kt/V]).
2.4. Statistical analysis
Continuous variables are presented as mean ± standard deviation if normally distributed, or as median (interquartile range) if non-normally distributed. Normality was assessed using the Shapiro–Wilk test. Categorical variables are presented as frequencies and percentages. Group comparisons between cases and controls were performed using the independent samples t-test or Mann–Whitney U test for continuous variables, and the chi-square test or Fisher’s exact test for categorical variables, as appropriate.
The association between NLR and PDAP risk was evaluated using conditional logistic regression to account for the matched design. NLR was analyzed in two ways: (1) as a continuous variable using ln(NLR), and (2) as a categorical variable based on tertiles of the NLR distribution among all participants (Q1: <0.52, Q2: 0.52–1.24, Q3: ≥1.24). The trend across tertiles was tested by modeling the tertile median as a continuous variable (6).
We constructed a series of hierarchical models to assess the robustness of the association with progressive adjustment for potential confounders: Model 1 (Crude): Unadjusted. Model 2: Adjusted for the matching variables (age, sex). Model 3: Model 2 + education level, smoking status, diabetes, hypertension, and dialysis vintage. Model 4 (Fully Adjusted): Model 3 + laboratory covariates: parathyroid hormone, hemoglobin, albumin, calcium, globulin, C-reactive protein, and white blood cell count. Results are reported as odds ratios (ORs) with 95% confidence intervals (CIs). To explore the dose–response relationship, a restricted cubic spline model with three knots was fitted for ln(NLR) in the fully adjusted model, testing for non-linearity.
Sensitivity analyses were conducted in two steps. First, a multivariable, unconditional logistic regression was performed including all variables with p < 0.20 in the bivariate analyses. Second, the discriminative performance of the neutrophil-lymphocyte ratio (NLR) was evaluated using receiver operating characteristic (ROC) curve analysis. All statistical analyses were conducted in R (version 4.4.0) and SPSS (version 26.0). A two-tailed p value < 0.05 was considered statistically significant. To avoid overfitting, we first applied least absolute shrinkage and selection operator (LASSO) regression for variable selection. Concurrently, a simple model was constructed using core variables identified: ln(NLR), dialysis vintage, albumin, and C-reactive protein (CRP). These core variables were used to develop a simplified model.
3. Results
3.1. Baseline clinical data of participants
A total of 178 peritoneal dialysis patients (89 cases with peritonitis, 89 controls) were analyzed (Table 1). Compared to controls, cases were younger (median 51.0 vs. 56.0 years; p = 0.017) and had shorter dialysis duration (42.0 vs. 59.0 months; p < 0.001). They also exhibited higher CRP (median 2.10 vs. 2.00 mg/L; p = 0.034), higher BUN (19.80 ± 7.33 vs. 17.56 ± 4.31 mmol/L; p = 0.015), lower globulin (29.44 ± 5.24 vs. 31.39 ± 4.43 g/L; p = 0.008), higher ln(NLR) (median −0.01 vs. -0.54; p < 0.001), and a higher proportion with Ca < 2.0 mmol/L (16.85% vs. 4.49%; p = 0.008). No significant differences were found in other baseline characteristics (all p > 0.05).
Table 1.
Baseline characteristics of peritoneal dialysis patients with and without subsequent peritonitis.
| Variables | Total (n = 178) | Control (n = 89) | Case (n = 89) | Statistic | P |
|---|---|---|---|---|---|
| Age, M (Q1, Q3) | 54.00 (47.25, 63.00) | 56.00 (49.00, 64.00) | 51.00 (46.00, 61.00) | Z = −2.38 | 0.017 |
| Gender, n (%) | χ2 = 0.58 | 0.448 | |||
| Female | 75 (42.13) | 35 (39.33) | 40 (44.94) | ||
| Male | 103 (57.87) | 54 (60.67) | 49 (55.06) | ||
| WBC(109/L), Mean ± SD | 7.27 ± 2.91 | 6.98 ± 2.66 | 7.56 ± 3.13 | t = −1.33 | 0.186 |
| CRP, M (Q1, Q3) | 2.00 (1.00, 5.00) | 2.00 (1.00, 3.00) | 2.10 (1.00, 7.00) | Z = −2.11 | 0.034 |
| Hemoglobin(g/L), Mean ± SD | 105.89 ± 17.65 | 107.61 ± 14.20 | 104.18 ± 20.48 | t = 1.30 | 0.196 |
| Albumin(g/L), Mean ± SD | 36.49 ± 4.67 | 36.95 ± 3.59 | 36.03 ± 5.52 | t = 1.30 | 0.194 |
| Globulin(g/L), Mean ± SD | 30.42 ± 4.93 | 31.39 ± 4.43 | 29.44 ± 5.24 | t = 2.68 | 0.008 |
| Phosphorus(mmol/L), Mean ± SD | 1.58 ± 0.50 | 1.55 ± 0.49 | 1.61 ± 0.52 | t = −0.84 | 0.405 |
| BUN(mmol/L), Mean ± SD | 18.68 ± 6.10 | 17.56 ± 4.31 | 19.80 ± 7.33 | t = −2.47 | 0.015 |
| CR(umol/L), Mean ± SD | 982.72 ± 301.51 | 1003.36 ± 283.11 | 962.09 ± 319.13 | t = 0.91 | 0.363 |
| ln(NLR), M (Q1, Q3) | −0.44 (−0.61, 0.16) | −0.54 (−0.66, −0.42) | −0.01 (−0.48, 1.42) | Z = −6.50 | <0.001 |
| Dialysis duration (Month), M (Q1, Q3) | 49.50 (27.00, 71.75) | 59.00 (40.00, 79.00) | 42.00 (18.00, 61.00) | Z = −3.78 | <0.001 |
| Primary disease, n (%) | – | 0.213 | |||
| Chronic glomerulonephritis | 108 (60.67) | 48 (53.93) | 60 (67.42) | ||
| Hypertension | 63 (35.39) | 37 (41.57) | 26 (29.21) | ||
| Diabetes or other | 7 (3.93) | 4 (4.49) | 3 (3.37) | ||
| Smoking, n (%) | χ2 = 0.03 | 0.865 | |||
| No | 131 (73.60) | 66 (74.16) | 65 (73.03) | ||
| Yes | 47 (26.40) | 23 (25.84) | 24 (26.97) | ||
| Education level, n (%) | χ2 = 3.25 | 0.197 | |||
| Illiterate | 32 (17.98) | 14 (15.73) | 18 (20.22) | ||
| Below junior high school | 62 (34.83) | 27 (30.34) | 35 (39.33) | ||
| At or above junior high school | 84 (47.19) | 48 (53.93) | 36 (40.45) | ||
| Diabetes, n (%) | χ2 = 2.07 | 0.150 | |||
| No | 165 (92.70) | 85 (95.51) | 80 (89.89) | ||
| Yes | 13 (7.30) | 4 (4.49) | 9 (10.11) | ||
| Hypertension, n (%) | χ2 = 1.50 | 0.220 | |||
| No | 43 (24.16) | 18 (20.22) | 25 (28.09) | ||
| Yes | 135 (75.84) | 71 (79.78) | 64 (71.91) | ||
| PTH (pg/ml), n (%) | χ2 = 0.80 | 0.370 | |||
| PTH < 600 | 130 (73.45) | 68 (76.40) | 62 (70.45) | ||
| PTH ≥ 600 | 47 (26.55) | 21 (23.60) | 26 (29.55) | ||
| Ca (mmol/L), n (%) | χ2 = 7.13 | 0.008 | |||
| Ca < 2.0 | 19 (10.67) | 4 (4.49) | 15 (16.85) | ||
| Ca ≥ 2.0 | 159 (89.33) | 85 (95.51) | 74 (83.15) |
BUN, blood urea nitrogen; CRP, C-reactive protein; M, Median; NLR, neutrophil-to-lymphocyte ratio; PTH, parathyroid hormone; Q1, 1st quartile; Q₃, 3rd quartile; SD, standard deviation; WBC, white blood cell count; t, t-test, Z, Mann–Whitney test, χ2, Chi-square test, −, Fisher exact; SD, standard deviation, M, Median, Q1, 1st Quartile, Q3, 3st Quartile.
3.2. Association between neutrophil-to-lymphocyte ratio and peritoneal Dialysis-associated peritonitis
The association between NLR and PDAP risk was first evaluated using conditional logistic regression accounting for the matched pairs. As shown in Table 2, when analyzed as a continuous variable (per 1-unit increase in ln(NLR)), NLR showed a trend toward increased peritonitis risk in the unadjusted model (OR 1.83, 95% CI 0.97–3.46, p = 0.062). After adjustment for the matching variables (age and sex) in Model I, the association became statistically significant (OR 2.13, 95% CI 1.17–3.89, p = 0.014). This positive association remained robust with further adjustment for additional demographic and clinical covariates (Model II: OR 2.32, 95% CI 1.29–4.15, p = 0.005) and in the fully adjusted model incorporating key laboratory parameters (Model III: OR 2.25, 95% CI 1.26–4.02, p = 0.006).
Table 2.
Multivariable logistic regression analysis of NLR and peritoneal dialysis-related peritonitis.
| Variable | n.total | n.event_% | Crude model | Model I | Model II | Model III | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| OR (95%CI) | p value | OR (95%CI) | p value | OR (95%CI) | p value | OR (95%CI) | p value | |||
| NLR (continuous) | 178 | 89 (50) | 1.83 (0.97–3.46) | 0.062 | 2.13 (1.17–3.89) | 0.014 | 2.32 (1.29–4.15) | 0.005 | 2.25 (1.26–4.02) | 0.006 |
| NLR (categorical) | ||||||||||
| Q1 (<0.52) | 59 | 22 (37.3) | 1(Ref) | 1(Ref) | 1(Ref) | 1(Ref) | ||||
| Q2 (0.52–1.24) | 59 | 21 (35.6) | 0.93 (0.44–1.97) | 0.848 | 0.85 (0.39–1.83) | 0.672 | 0.79 (0.34–1.82) | 0.577 | 0.99 (0.38–2.54) | 0.981 |
| Q3 (≧1.24) | 60 | 46 (76.7) | 5.53 (2.49–12.27) | <0.001 | 7.41 (2.93–18.76) | <0.001 | 9.13 (3.37–24.69) | <0.001 | 11 (3.06–39.52) | <0.001 |
| P for trend | <0.001 | <0.001 | <0.001 | <0.001 | ||||||
OR, odds ratio; CI, confidence interval; NLR, neutrophil-to-lymphocyte ratio.
Crude model: unadjusted.
Model I: Adjusted for sex and age.
Model II: Adjusted for sex, age, education level, smoking, diabetes, hypertension, and dialysis duration.
Model III: Fully adjusted for sex, age, education level, smoking, diabetes, hypertension, dialysis duration, PTH, hemoglobin, albumin, calcium, globulin, CRP, and WBC.
When NLR was categorized into tertiles, patients in the highest tertile (NLR ≥ 1.24) had markedly increased odds of PDAP compared to those in the lowest tertile (NLR < 0.52). In the fully adjusted model (Model III), the odds ratio for the highest tertile was 11.00 (95% CI 3.06–39.52, p < 0.001). A significant dose–response relationship was evident across increasing tertiles (P for trend < 0.001).
To visualize the dose–response relationship, a restricted cubic spline analysis was performed. As illustrated in Figure 2, a significant linear association was observed between ln(NLR) and the log odds of peritonitis (P for overall association = 0.004; P for non-linearity = 0.511), confirming a monotonically increasing risk with higher NLR levels.
Figure 2.
Restricted cubic spline analysis showing the relationship between neutrophil-to-lymphocyte ratio (NLR) and the risk of peritoneal dialysis-associated peritonitis (PDAP). Models were adjusted for sex, age, education level, smoking, diabetes, hypertension, dialysis duration, parathyroid hormone (PTH), hemoglobin, albumin, calcium, globulin, C-reactive protein (CRP), and white blood cell count (WBC).
3.3. Subgroup analysis
Stratified analyses demonstrated a consistent positive association between elevated NLR (≥ median) and PDAP risk across most clinical subgroups. In the overall cohort, an elevated NLR was associated with significantly increased odds of peritonitis (OR: 5.76; 95% CI: 2.87–11.57) (Figure 3). A statistically significant interaction was observed for gender (P for interaction = 0.038). Consistently positive associations were also observed in several high-risk subgroups, including patients without hypertension (OR: 47.28; 95% CI: 3.27–684.00), those with albumin <35 g/L (OR: 16.30; 95% CI: 3.72–71.42), and those with globulin ≥30 g/L (OR: 12.12; 95% CI: 3.82–38.40). However, the wide confidence intervals in these subgroups reflect limited sample sizes and should be interpreted with caution. No significant interactions were detected for hypertension, hemoglobin, albumin, globulin, or age (all P for interaction >0.05), suggesting that the association between NLR and PDAP risk remains broadly consistent across patient profiles. Given the exploratory nature of these analyses and the modest sample size in certain strata, these findings warrant confirmation in larger, adequately powered studies.
Figure 3.
Subgroup analysis of the association between NLR and the risk of peritonitis. NLR was dichotomized at the median value. Multivariable logistic regression was performed to assess the relationship between NLR categories and peritonitis incidence.
3.4. Sensitivity analysis
To assess the robustness of our primary finding, we performed complementary sensitivity analyses. First, we employed an alternative modeling strategy using unconditional multivariable logistic regression and entering all variables with p < 0.20 from bivariate analysis into a single model. In this analysis, NLR remained an independent predictor of PDAP (adjusted OR 3.01, 95% CI 1.23–7.33, p = 0.015) (Supplementary Table S1).
Second, we evaluated the discriminative ability of NLR alone for predicting peritonitis using receiver operating characteristic (ROC) curve analysis. The area under the ROC curve (AUC) was 0.78 (95% CI 0.71–0.85) (Supplementary Figure S1), indicating moderate discriminatory power. At the optimal cutoff value of 0.459, sensitivity was 0.85 (95% CI 0.78–0.93) and specificity was 0.65 (95% CI 0.55–0.75) (Supplementary Table S2).
To mitigate the risk of overfitting, we applied LASSO regression for variable selection (Supplementary Figure S2 and S3). The core variables identified-ln(NLR), dialysis vintage, albumin, and CRP-were used to construct a simplified model. The effect estimate for ln(NLR) remained consistent in direction with that of the fully adjusted model (simplified model OR = 7.31, 95% CI: 3.05–17.52, p < 0.001; Supplementary Table S3), supporting the robustness of the primary finding. Collectively, these sensitivity analyses reinforce the conclusion that an elevated pre-existing NLR is a robust risk factor for subsequent PDAP in this patient population.
4. Discussion
The principal finding of this retrospective case–control study is that a higher neutrophil-to-lymphocyte ratio (NLR), measured up to 3 months prior to clinical presentation, is independently associated with an increased risk of developing peritoneal dialysis-associated peritonitis (PDAP). This association demonstrated robustness across multiple analytical approaches: it remained significant in multivariable logistic regression models with progressive adjustment for potential confounders, exhibited a significant positive linear dose–response relationship in curve fitting analysis, and was confirmed in complementary sensitivity analyses.
The NLR is an emerging, readily accessible biomarker of systemic inflammation. Its prognostic utility extends beyond dialysis populations, with established links to conditions such as contrast-induced nephropathy following vascular procedures (7). In the context of end-stage renal disease, elevated NLR has been consistently associated with heightened inflammatory burden, increased mortality, and adverse cardiovascular outcomes in both hemodialysis (HD) and peritoneal dialysis (PD) patients (3, 7, 8–15). Furthermore, an elevated NLR at the time of peritonitis diagnosis has been linked to treatment failure (4). Complementary evidence from other HD cohorts further supports the prognostic value of NLR as an independent predictor of mortality, with its significance persisting across different dialysis-vintage strata (16). Among patients undergoing PD and younger than 60 years, NLR was identified as an independent risk factor for poor cardiovascular outcomes (17). Additionally, another study demonstrated that an elevated NLR was a significant risk factor for developing pneumonia in patients undergoing PD, with multivariate Cox proportional hazards models confirming NLR as an independent predictor (18). Our study extends this evidence by demonstrating, for the first time, that a pre-morbid elevation in NLR-assessed during a clinically stable period preceding the infection-serves as a significant risk factor for the subsequent development of PDAP itself. This suggests that a subclinical pro-inflammatory state, reflected by an elevated NLR, may predispose patients to infectious complications. Within this context, our study is the first to demonstrate that the NLR measured 3 months before the onset of peritonitis is a risk factor for peritonitis in patients undergoing PD.
Current guidelines and the existing literature have identified a broad spectrum of risk factors for PDAP, encompassing patient demographics (e.g., age, diabetes), procedural aspects (e.g., training duration, aseptic technique), and dialysis-related factors (e.g., modality, solution storage) (19–21). Our findings introduce pre-infection NLR as a novel, biologically plausible candidate within this risk profile. The observed association remained robust after adjustment for several of these established factors, including dialysis vintage and nutritional markers such as albumin and globulin, suggesting its potential as an independent contributor to peritonitis risk.
Our findings corroborate and extend the growing literature on the prognostic value of the neutrophil-to-lymphocyte ratio (NLR) in the peritoneal dialysis (PD) population. Previous work showed that NLR measured after PD catheterization independently predicts subsequent PD-associated peritonitis (PDAP) with strong discrimination (22). In contrast, we demonstrate for the first time that NLR elevation can precede clinical infection, detectable during a stable, infection-free period up to 3 months before PDAP onset, supporting its role as a predisposing risk factor rather than solely an acute-phase marker.
Beyond NLR, inflammatory-nutritional markers have been linked to PDAP risk in other work. Nomograms incorporating lymphocyte count, albumin, and hemoglobin have achieved high predictive performance (AUC ≈ 0.93, Wang et al. 23). Lymphopenia, a key component of elevated NLR, and related nutritional risks have repeatedly correlated with infection risk, with large retrospective cohorts (You et al. 24) confirming that low albumin and low hemoglobin are PDAP risk factors alongside socioeconomic and lifestyle factors. These findings across designs and populations strengthen the biological plausibility that systemic inflammation and nutritional status influence PDAP susceptibility. Differences in timing of NLR measurement, population characteristics, cutoff choices, and analytic approaches likely explain heterogeneity in effect sizes. Prospective validation in diverse cohorts and standardization of measurement protocols will be essential before clinical implementation.
Several limitations of this retrospective case–control study should be acknowledged. First, although we used pre-peritonitis laboratory values to establish a temporal relationship between NLR and subsequent PDAP risk, reverse causation cannot be entirely excluded, as subclinical or prodromal infection may elevate NLR before diagnosis, despite excluding tests after symptom onset and performing sensitivity analyses restricting measurements to >30 days before the event. Prospective cohorts with baseline NLR at dialysis initiation are needed to better establish temporal precedence. Second, despite rigorous matching, blinded data extraction, and adjustment for known confounders, residual confounding from unmeasured factors-such as detailed comorbidity scores, medication use, socioeconomic status, and treatment adherence-may influence the observed associations. Third, NLR was measured at a single time point, which may not capture longitudinal fluctuations in inflammatory status; serial measurements could provide a more complete inflammatory trajectory and potentially improve PDAP risk prediction. Fourth, modest sample sizes in some subgroup analyses yielded wide confidence intervals and unstable estimates (for example, an OR of 47.28 in patients without hypertension). These exploratory findings require validation in larger, adequately powered studies. Fifth, the ROC analysis for NLR alone was exploratory; we did not compare NLR to a baseline clinical model, assess incremental predictive value, evaluate calibration, or perform internal validation. Consequently, while NLR showed moderate discrimination (AUC 0.78), conclusions about its clinical utility for risk stratification remain preliminary, and external validation in large, prospective multicenter cohorts is necessary before clinical implementation. Finally, selection bias may exist because patients with incomplete medical records were excluded, and the single-center design with a relatively modest sample size may limit generalizability and precision of effect estimates; multicenter studies are needed to validate results and explore potential effect modification across diverse populations and settings.
In conclusion, our study provides evidence that an elevated neutrophil-to-lymphocyte ratio measured during a stable clinical period is associated with a significantly increased risk of subsequent peritonitis in patients undergoing peritoneal dialysis. If validated in larger, prospective multicenter cohorts, the NLR could serve as a simple and inexpensive tool for risk stratification, potentially identifying patients who may benefit from intensified monitoring or targeted preventive strategies. Future research should aim to confirm this association, elucidate the underlying mechanisms linking systemic inflammation to peritoneal infection risk, and explore the integrated predictive value of NLR alongside traditional and novel risk factors.
Acknowledgments
We thank Jie Liu (Beijing 301 Hospital) for his constructive suggestions on the statistical analysis and for his helpful comments that improved the manuscript.
Funding Statement
The author(s) declared that financial support was received for this work and/or its publication. This project is supported by the Medical and Health Project of Zhejiang Province (2024KY1846), the Zhejiang Provincial Traditional Chinese Medicine Clinical Research Technology Program (2026ZL1063), and the Lishui City Science and Technology Plan Projects (2023SJZC024, 2025RKX55).
Edited by: Luca Molinari, Azienda Ospedaliero Universitaria Maggiore della Carità, Italy
Reviewed by: Arriel Makembi Bunkete, Centre Hospitalier de l'Ouest Guyanais, French Guiana
Marilaeta Cindryani, Ngoerah General Hospital, Indonesia
Abbreviations: PD, Peritoneal dialysis; PDAP, Peritoneal dialysis-associated peritonitis; NLR, neutrophil-to-lymphocyte ratio; OR, odds ratio; CI, confidence interval; HD, hemodialysis; ISPD, International Society for Peritoneal Dialysis; WBC, white blood cell; BUN, blood urea nitrogen; Kt/V, urea clearance index; ANC, absolute neutrophil count; ALC, absolute lymphocyte count; aOR, adjusted odds ratio; CRP, C-reactive protein.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by The Ethics Committee of the Fifth Affiliated Hospital of Wenzhou Medical University. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin because the retrospective nature of the study.
Author contributions
WL: Formal analysis, Writing – original draft, Methodology, Conceptualization, Investigation. H-pL: Methodology, Investigation, Writing – review & editing, Formal analysis. XS: Writing – review & editing, Methodology, Investigation. JX: Software, Investigation, Conceptualization, Writing – review & editing, Resources.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmed.2026.1787005/full#supplementary-material
Receiver operating characteristic (ROC) curve evaluating the predictive performance of NLR for peritonitis.
LASSO Logistic Regression for Feature Selection. Selection of the optimal penalization parameter (lambda) in the LASSO model using five-fold cross-validation based on the minimum deviance criterion.
LASSO Logistic Regression for Feature Selection. LASSO coefficient paths for the 14 candidate features. Coefficient profiles are plotted as a function of the log(lambda) sequence. The vertical dashed line indicates the optimal lambda value selected by five-fold cross-validation, at which four features retained non-zero coefficients. LASSO, Least Absolute Shrinkage and Selection Operator.
Risk factors for peritoneal dialysis-related peritonitis.
ROC analysis for NLR in predicting peritoneal dialysis-associated peritonitis.
Simplified multivariable logistic regression model for the association between NLR and peritoneal dialysis-associated peritonitis.
References
- 1.Cho Y, Chow KM, Kam-Tao Li P, Runnegar N, Johnson DW. Peritoneal dialysis-related infections. Clin J Am Soc Nephrol: CJASN. (2024) 19:641–9. doi: 10.2215/CJN.0000000000000280, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Fung WW, Chow KM, Ng JK, Chan GC, Li PK, Szeto CC. The clinical utility of the neutrophil-to-lymphocyte ratio as a discriminatory test among bacterial, mycobacterium tuberculosis, and nontuberculous mycobacterium peritoneal dialysis-related peritonitis. Kidney360. (2022) 3:1031–8. doi: 10.34067/KID.0000842022, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Yang N, Yang K, Pan S, He Q, Jin J. Progress in the application of the neutrophil-to-lymphocyte ratio in dialysis-related complications. Ren Fail. (2023) 45:2259996. doi: 10.1080/0886022X.2023.2259996, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.He P, He LJ, Huang C, Hu JP, Sun SR. Neutrophil-to-lymphocyte ratio and treatment failure in peritoneal Dialysis-associated peritonitis. Front Med. (2021) 8:699502. doi: 10.3389/fmed.2021.699502, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. (2007) 147:573–7. doi: 10.7326/0003-4819-147-8-200710160-00010, [DOI] [PubMed] [Google Scholar]
- 6.Mickey RM, Greenland S. The impact of confounder selection criteria on effect estimation. Am J Epidemiol. (1989) 129:125–37. doi: 10.1093/oxfordjournals.aje.a115101, [DOI] [PubMed] [Google Scholar]
- 7.He T, Mohammadpour B, Willman M, Yaghoobpoor S, Willman J, Lucke-Wold B, et al. Prognostic role of neutrophil to lymphocyte ratio in contrast-induced nephropathy: a systematic review and Meta-analysis. Angiology. (2025) 76:819–32. doi: 10.1177/00033197241238512, [DOI] [PubMed] [Google Scholar]
- 8.Ke J, Qiu F, Fan W, Wei S. Associations of complete blood cell count-derived inflammatory biomarkers with asthma and mortality in adults: a population-based study. Front Immunol. (2023) 14:1205687. doi: 10.3389/fimmu.2023.1205687, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhang J, Fan X, Xu Y, Wang K, Xu T, Han T, et al. Association between inflammatory biomarkers and mortality in individuals with type 2 diabetes: NHANES 2005-2018. Diabetes Res Clin Pract. (2024) 209:111575. doi: 10.1016/j.diabres.2024.111575, [DOI] [PubMed] [Google Scholar]
- 10.Kara AV, Soylu YE. The relationship between vitamin D and inflammatory markers in maintenance hemodialysis patients. Int Urol Nephrol. (2019) 51:1659–65. doi: 10.1007/s11255-019-02250-x, [DOI] [PubMed] [Google Scholar]
- 11.Li H, Lu X, Xiong R, Wang S. High neutrophil-to-lymphocyte ratio predicts cardiovascular mortality in chronic hemodialysis patients. Mediat Inflamm. (2017) 2017:9327136. doi: 10.1155/2017/9327136, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Willman J, Lucke-Wold B. Commentary: "inflammation and the role of infection: complications and treatment options following neurotrauma". J Clin Neurosci. (2023) 113:147–8. doi: 10.1016/j.jocn.2023.01.005, [DOI] [PubMed] [Google Scholar]
- 13.van Tellingen A, Grooteman MP, Schoorl M, Bartels PC, Schoorl M, van der Ploeg T, et al. Intercurrent clinical events are predictive of plasma C-reactive protein levels in hemodialysis patients. Kidney Int. (2002) 62:632–8. doi: 10.1046/j.1523-1755.2002.00470.x, [DOI] [PubMed] [Google Scholar]
- 14.Zhang J, Lu X, Wang S, Li H. Neutrophil-to-lymphocyte ratio and erythropoietin resistance among maintenance hemodialysis patients. Blood Purif. (2022) 51:708–13. doi: 10.1159/000519644, [DOI] [PubMed] [Google Scholar]
- 15.Yu Y, Zhong Z, Yang W, Yu J, Li J, Guo X, et al. Neutrophil percentage-to-albumin ratio and risk of mortality in patients on peritoneal Dialysis. J Inflamm Res. (2023) 16:6271–81. doi: 10.2147/JIR.S437256, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhang J, Lu X, Wang S, Li H. High neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio are associated with poor survival in patients with hemodialysis. Biomed Res Int. (2021) 2021:9958081. doi: 10.1155/2021/9958081, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhang Y, Lai X, Chen Q, Zhan X, Long H, Peng F, et al. The relationship between neutrophil-to-lymphocyte ratio and the first occurrence of pneumonia in peritoneal dialysis patients. Clin Exp Nephrol. (2020) 24:770–8. doi: 10.1007/s10157-020-01894-9, [DOI] [PubMed] [Google Scholar]
- 18.Zeng Y, Chen Z, Chen Q, Zhan X, Long H, Peng F, et al. Neutrophil to lymphocyte ratio predicts adverse cardiovascular outcome in peritoneal dialysis patients younger than 60 years old. Mediat Inflamm. (2020) 2020:4634736. doi: 10.1155/2020/4634736 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Li PT, Chow KM, Cho Y. Corrigendum to 2022 ISPD peritonitis guidelines: 2022 update on prevention and treatment. Peritoneal Dialysis Int. (2023) 43:279. doi: 10.1177/08968608231166870, [DOI] [PubMed] [Google Scholar]
- 20.Perl J, Fuller DS, Bieber BA, Boudville N, Kanjanabuch T, Ito Y, et al. Peritoneal dialysis-related infection rates and outcomes: results from the peritoneal Dialysis outcomes and practice patterns study (PDOPPS). Am J Kidney Diseases. (2020) 76:42–53. doi: 10.1053/j.ajkd.2019.09.016, [DOI] [PubMed] [Google Scholar]
- 21.Pan L, Wang M, Yu Q, Gao X, Xia Y, Huang X. Risk factors for peritoneal Dialysis-associated peritonitis due to home-based operation management: a retrospective cohort study. J Clin Nurs. (2025) 34:3848–56. doi: 10.1111/jocn.17600, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Dai R, Peng C, Sang T, Cheng M, Wang Y, Zhang L. Construction and validation of a predictive model for the risk of peritoneal dialysis-associated peritonitis after peritoneal dialysis catheterization. Front Med. (2023) 10:1193754. doi: 10.3389/fmed.2023.1193754, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wang Y, Wu Z, Huang L, Suo D, Zhang M, Dai M, et al. A nomogram for predicting the risk of peritoneal dialysis-associated peritonitis in patients with end-stage renal disease undergoing peritoneal dialysis: model development and validation study. BMC Nephrol. (2025) 26:248. doi: 10.1186/s12882-025-04165-5, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.You L, Zhang B, Zhang F, Wang J. Pathogenic spectrum and risk factors of peritoneal dialysis-associated peritonitis: a single-center retrospective study. BMC Infect Dis. (2024) 24:440. doi: 10.1186/s12879-024-09334-9, [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Receiver operating characteristic (ROC) curve evaluating the predictive performance of NLR for peritonitis.
LASSO Logistic Regression for Feature Selection. Selection of the optimal penalization parameter (lambda) in the LASSO model using five-fold cross-validation based on the minimum deviance criterion.
LASSO Logistic Regression for Feature Selection. LASSO coefficient paths for the 14 candidate features. Coefficient profiles are plotted as a function of the log(lambda) sequence. The vertical dashed line indicates the optimal lambda value selected by five-fold cross-validation, at which four features retained non-zero coefficients. LASSO, Least Absolute Shrinkage and Selection Operator.
Risk factors for peritoneal dialysis-related peritonitis.
ROC analysis for NLR in predicting peritoneal dialysis-associated peritonitis.
Simplified multivariable logistic regression model for the association between NLR and peritoneal dialysis-associated peritonitis.
Data Availability Statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.