Abstract
Objectives
Patients undergoing hemodialysis (HD) face high mortality, mainly from cardiovascular disease (CVD), infections, and dialysis withdrawal. Systemic inflammation contributes significantly to these outcomes. This study aimed to examine the associations between the systemic immune-inflammation index (SII) and systemic inflammation response index (SIRI) with all-cause, CVD, and infection-related mortality in HD patients, and to evaluate their prognostic value.
Methods
We conducted a retrospective analysis of 1190 HD patients recruited between January 2012 and December 2016. Kaplan-Meier survival curves, Cox regression analysis, and restricted cubic spline (RCS) models were employed to explore the associations between SII, SIRI, and all-cause, CVD, and infection-related mortality. In addition, receiver operating characteristic (ROC) curves was used to determine the predictive accuracy of SII and SIRI.
Results
Among the 1190 patients (median age 62.0 years; 64.0% male), the all-cause mortality rate was 38.2%, while the CVD mortality rate was 16.9%. Adjusted Cox regression analyses revealed that patients in the highest SIRI quartile (quartile 4) had significantly elevated risks of all-cause (hazard ratio [HR] 2.29, 95% CI 1.38–3.80, P = 0.001), CVD (HR 3.78, 95% CI 1.43–10.01, P = 0.007), and infection-related mortality (HR 2.42, 95% CI 1.70–3.01, P < 0.001) compared to those in the lowest SIRI quartile. Similar associations were found for SII (P < 0.001). Kaplan-Meier curves demonstrated comparable results. RCS analysis revealed nonlinear relationships between SII, SIRI, and mortality risk (P < 0.05). ROC analysis highlighted that Both SII and SIRI demonstrated moderate to strong prognostic value, with SIRI consistently offering the best risk stratification.
Conclusions
Elevated levels of SII and SIRI are linked to higher risks of all-cause, CVD, and infection-related mortality in HD patients. SIRI appears to be a more reliable predictor of mortality risk. Future studies should explore the underlying mechanisms and validate the predictive value of SII and SIRI for mortality risk among HD patients.
Clinical trail number
Not applicable.
Supplementary information
The online version contains supplementary material available at 10.1186/s12882-025-04223-y.
Keywords: Systemic immune-inflammation index, Systemic inflammation response index, Cardiovascular disease mortality, All-cause mortality, Hemodialysis
Introduction
End-stage kidney disease (ESKD) remains a significant global health burden, especially for patients undergoing maintenance hemodialysis (HD) [1–4]. Individuals undergoing hemodialysis face an alarmingly high mortality risk, with approximately 20% of patients dying within the first year of dialysis and a staggering 60% mortality rate over the course of five years. Despite advances in dialysis technologies and care protocols, mortality rates in this population remain persistently high [3, 4]. The primary causes of death among hemodialysis patients are cardiovascular disease (CVD) (53%), infections (18%), and withdrawal from dialysis (16%), with CVD remaining the leading cause of mortality [5–7]. This underscores the heightened vulnerability of these patients to cardiovascular events, which are further exacerbated by the physiological challenges of ESKD [3, 8, 9]. Between 2007 and 2017, CVD-related mortality in HD patients increased by 50.0%, highlighting the urgent need to identify novel risk factors to mitigate mortality [10].
Systemic inflammation has emerged as a key contributor to the high mortality rates in HD patients [11, 12]. Immune activation promotes the release of pro-inflammatory cytokines, such as interleukin-8 (IL-8), IL-6, and IL-1β, which have been associated with adverse outcomes in ESKD patients [13–15]. Inflammatory biomarkers may therefore serve as valuable tools for nephrologists to assess systemic inflammation and optimize therapeutic strategies. Against this backdrop, the systemic immune-inflammation index (SII) and the systemic inflammation response index (SIRI), as cost-effective and comprehensive indicators reflecting the overall immune-inflammatory state, have shown prognostic utility across various conditions, including malignancies and kidney disease [16, 17]. The SII and SIRI, can be readily obtained through routine blood tests, incorporating counts of lymphocytes, neutrophils, monocytes, and platelets in the peripheral blood [18, 19]. Not only is it simple to perform and cost-effective, but it also demonstrates good reproducibility. Clinically, SII and SIRI are important for predicting survival in chronic kidney disease (CKD) and peritoneal dialysis (PD) populations, and SII and SIRI may also be linked to poor outcomes in patients with CVD [20–22]. However, their predictive value in HD patients has not been fully investigated.
To address the identified research gaps, this study aims to investigate the relationship between various systemic inflammatory markers, specifically the SII and SIRI, and their association with all-cause, CVD, and infection-related mortality in patients undergoing HD. Additionally, we compare the predictive value of each inflammatory marker to determine their relative prognostic accuracy.
Materials and methods
Study design and population
This is a prospective cohort study with data retrospectively analyzed, carried out at the Department of hemodialysis center of the Traditional Chinese Medicine Hospital of Pujiang County. All incident HD patients between January.1 2012 and December 31, 2016, were screened. All patients were followed until death, transfer to renal transplantation, loss to follow-up or the end of study (December 30, 2022). The exclusion criteria were as follows: (1) younger than 18 years old; (2) diagnosis with malignancies, hematological diseases, autoimmune diseases, severe infection or liver failure; (3) receiving immunosuppression therapy within the past 6 months; (4) maintenance HD therapy less than 3 months; (5) missing data of survival information, laboratory parameters, or covariates. Follow-up was censored at the time of transfer to peritoneal dialysis (PD), renal transplantation, loss to follow-up, or study completion (December 30, 2022). Hence, 1190 HD patients with complete follow-up were included in the final analysis (Fig. 1). All patients provided informed consent at the commencement of HD. The study was in line with the principles of the Declaration of Helsinki and with approved by the Ethics Committee at the Traditional Chinese Medicine Hospital of Pujiang County, China (approval number: 2024C04).
Fig. 1.
Flowchart of patient inclusion and exclusion for this study. HD, hemodialysis
Data collection
Baseline characteristics, including demographic data, comorbidities, laboratory values, and medication use, were collected at the time of admission, prior to the initiation of hemodialysis. Demographic variables encompassed age, sex, body mass index (BMI), time on dialysis, primary kidney disease, education level, marital status, annual household income, and smoking and alcohol consumption status. Comorbidities assessed included hypertension, diabetes, hypotension, and cardiovascular disease. Laboratory assessments included a complete blood count (white blood cell count, hemoglobin, platelet count, neutrophils, lymphocytes, and monocytes), as well as measurements of serum creatinine, blood urea nitrogen (BUN), uric acid, calcium, phosphate, albumin, hemoglobin, C-reactive protein (CRP), ferritin, total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL). The BMI was calculated as weight (kg) divided by height squared (m²). All data were collected by trained, independent investigators, ensuring accuracy and consistency across the study cohort.
Definitions
Blood samples were collected using automated hematology analyzers prior to dialysis. Given the clinical relevance, accessibility, and established validity of systemic inflammatory markers in reflecting immune and inflammatory status in the HD population, five inflammation-related indices were selected for analysis: the SII, systemic inflammation response index (SIRI), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and lymphocyte-to-monocyte ratio (LMR). These indices have been validated in previous studies for their prognostic value and utility in guiding treatment strategies, making them suitable candidates for further investigation in this research setting.
SII=Platelet*Neutrophil/Lymphocyte
SIRI=Monocyte*Neutrophil/Lymphocyte
NLR=Neutrophil/Lymphocyte
PLR=Platelet/Lymphocyte
LMR=Lymphocyte/Monocyte
Outcomes
The primary outcome of this study was all-cause mortality, with CVD and infection-related mortality as the secondary outcome. CVD mortality was defined as death resulting from myocardial infarction, congestive heart failure, stroke (including cerebral hemorrhage and infarction), arrhythmia, peripheral arterial disease, or sudden cardiac death [23]. Infection-related mortality was classified as death from conditions such as pulmonary infection, sepsis, lower limb infection, cholecystitis, cholangitis, infections related to vascular access or catheters, enteritis, infectious endocarditis, urinary tract infection, cyst infection, cellulitis, or other infections [24]. Mortality data for patients who died during hospitalization were obtained from medical records, while for those who died at home, data were collected through telephone interviews with next-of-kin. Survival status was independently verified by trained physicians, blinded to the clinical and laboratory data of the patients.
Statistical analysis
Continuous variables were expressed as median with interquartile range (IQR), while categorical variables were reported as number (percentage, %). Demographic characteristics were compared using the rank sum test for continuous variables and the χ² test for categorical variables. To assess the association between various inflammatory markers and all-cause or cardiovascular mortality in HD patients, univariate and multivariable Cox regression analyses were performed. The unadjusted (crude) model included no covariates, whereas Model 1 adjusted for sex and age. Model 2 was fully adjusted, accounting for potential confounders, including age, sex, BMI, time on dialysis, primary kidney disease, marital status, education level, smoking and drinking status, annual family income, hypertension, diabetes mellitus, hypotension, cardiovascular disease, hemoglobin, albumin, calcium, phosphate, creatinine, CRP, ferritin, blood urea nitrogen, uric acid, estimated glomerular filtration rate (eGFR), FPG, TC, TG, HDL, and LDL. Kaplan-Meier survival curves were employed to illustrate differences in survival probabilities across groups with varying levels of systemic inflammatory markers. To explore potential nonlinear associations between systemic inflammatory markers and HD mortality, restricted cubic spline (RCS) regression with four knots was utilized, allowing for a more flexible model to capture variations in risk along the entire spectrum. Comparisons of area under the curve (AUCs) at different time points were conducted using DeLong’s test to ensure statistical rigor in assessing differences. Predictive accuracy for survival was evaluated using time-dependent receiver operating characteristic (ROC) curves. The AUC was calculated for each inflammatory marker at multiple time points to assess their discriminative power. All statistical analyses were performed using R software (version 4.3.2), with two-tailed P-values of < 0.05 considered statistically significant.
Results
Baseline characteristics
In this study, a total of 1190 HD patients were analyzed, with baseline characteristics stratified by SII and SIRI quartiles presented in Table 1 and Table 2. Among the participants, 762 (64.0%) were male, with a median age of 62.0 years (IQR: 54–72 years). Notable differences in admission characteristics, including age, duration of dialysis, primary kidney disease, educational level, and annual family income, were observed across SII quartiles. Laboratory indices such as hemoglobin, albumin, creatinine, CRP, urea nitrogen, uric acid, eGFR, FPG, TG, LDL, and WBC counts also exhibited significant variation. Furthermore, the prevalence of comorbidities, including hypertension, diabetes, hypotension, and cardiovascular disease, increased with higher SII quartiles (all P < 0.05). Similar pronounced differences in admission characteristics, laboratory indices, and medication use were noted across SIRI quartiles, reinforcing the relevance of these inflammatory markers in this population. Additional participant characteristics are detailed in Table 1 and Table 2.
Table 1.
Baseline characteristics stratified by systemic-immune inflammation index quartiles
| Clinical characteristics | Overall | Quartiles of systemic-immune inflammation index | P value | |||
|---|---|---|---|---|---|---|
| Quartile 1 (N = 316) | Quartile 2 (N = 278) | Quartile 3 (N = 298) | Quartile 4 (N = 298) | |||
| <426.78 | 426.78–662.30 | 662.30–998.13 | ≥998.13 | |||
| Age, years | 62 (54, 72) | 59 (54, 70) | 64 (55, 72) | 60 (52, 72) | 66 (55, 73) | 0.013 |
| Male sex | 762 (64.0) | 190 (60.1) | 172 (61.9) | 194 (65.1) | 206 (69.1) | 0.104 |
| BMI (kg/m2) | 21.2 (19.5, 24.0) | 21.6 (19.5, 23.6) | 21.4 (18.9, 23.9) | 21.6 (19.6, 24.2) | 21.3 (19.6, 24.0) | 0.344 |
| Time on dialysis (years) | 4 (3, 9) | 5 (3, 11) | 4 (2, 8) | 4 (3, 8) | 3 (2, 8) | <0.001 |
| Primary kidney disease, (n, %) | <0.001 | |||||
| Diabetic nephropathy | 426 (35.8) | 78 (24.7) | 96 (34.5) | 98 (32.9) | 264 (88.6) | |
| Glomerulonephritis | 698 (58.7) | 214 (67.7) | 164 (59.0) | 182 (61.1) | 22 (7.4) | |
| Others | 66 (5.5) | 24 (7.6) | 18 (6.5) | 18 (6.0) | 12 (4.0) | |
| Educational level, (n, %) | <0.001 | |||||
| Junior secondary or below | 1016 (85.4) | 238 (75.3) | 246 (88.5) | 268 (89.9) | 264 (88.6) | |
| Senior high school/technical secondary school | 144 (12.1) | 66 (20.9) | 30 (10.8) | 26 (8.7) | 22 (7.4) | |
| Junior college/university or above | 28 (2.4) | 12 (3.8) | 2 (0.7) | 4 (1.3) | 12 (4.0) | |
| Marital status, (n, %) | 0.146 | |||||
| Unmarried | 74 (6.2) | 24 (7.6) | 18 (6.5) | 14 (4.7) | 18 (6.0) | |
| Married | 936 (78.7) | 236 (74.7) | 220 (79.1) | 252 (84.6) | 228 (76.5) | |
| Divorced | 90 (7.6) | 32 (10.1) | 20 (7.2) | 16 (5.4) | 22 (7.4) | |
| Widowed | 90 (7.6) | 24 (7.6) | 20 (7.2) | 16 (5.4) | 30 (10.1) | |
| Annual family income (RMB thousand yuan), (n, %) | <0.001 | |||||
| <50 | 482 (40.5) | 120 (38.0) | 138 (49.6) | 130 (43.6) | 94 (31.5) | |
| 50 to < 100 | 196 (16.5) | 76 (24.1) | 34 (12.2) | 42 (14.1) | 44 (14.8) | |
| ≥ 100 | 512 (43.0) | 120 (38.0) | 106 (38.2) | 126 (42.3) | 160 (53.7) | |
| Current smoker, (n, %) | 692 (58.2) | 166 (52.5) | 164 (59.0) | 176 (59.1) | 186 (62.4) | 0.089 |
| Current drinker, (n, %) | 750 (63.0) | 190 (60.1) | 178 (64.0) | 180 (60.4) | 202 (67.8) | 0.189 |
| Hypertension, (n, %) | 360 (30.3) | 86 (27.2) | 78 (28.1) | 82 (27.5) | 114 (38.3) | 0.007 |
| Diabetes mellitus, (n, %) | 430 (36.1) | 78 (24.7) | 94 (33.8) | 98 (32.9) | 160 (53.7) | <0.001 |
| Hypotension, (n, %) | 892 (75.0) | 218 (69.0) | 194 (69.8) | 232 (77.9) | 248 (83.2) | <0.001 |
| Cardiovascular disease, (n, %) | 550 (46.2) | 120 (38.0) | 134 (48.2) | 128 (43.0) | 168 (56.4) | <0.001 |
| Anti-hypertension, (n, %) | 360 (30.3) | 100(31.5) | 80 (28.8) | 106 (35.6) | 74 (24.8) | 0.033 |
| Anti-diabetes, (n, %) | 366 (30.8) | 78 (24.7) | 86 (30.9) | 76 (25.5) | 126 (42.3) | <0.001 |
| Hypolipidemic agents, (n, %) | 546 (45.9) | 124 (39.2) | 130 (46.8) | 122 (40.9) | 170 (57.0) | <0.001 |
| Laboratory of parameters | ||||||
| Hemoglobin (g/L) | 108.0 (97.0, 117.0) | 108.5 (100.0, 117.0) | 110.0 (98.0, 118.0) | 108.0 (100.0, 117.0) | 103.0 (90.8, 117.0) | <0.001 |
| Albumin (g/L) | 36.8 (34.5, 38.8) | 36.9 (34.9, 38.7) | 36.7 (34.5, 38.9) | 36.9 (34.9, 39.4) | 36.3 (32.4, 38.7) | 0.042 |
| Calcium (mmol/L) | 2.2 (2.1, 2.4) | 2.3 (2.1, 2.4) | 2.2 (2.1, 2.4) | 2.2 (2.1, 2.4) | 2.2 (2.1, 2.4) | 0.128 |
| Phosphorus (mmol/L) | 1.9 (1.5, 2.3) | 1.9 (1.5, 2.3) | 1.8 (1.4, 2.3) | 1.9 (1.5, 2.2) | 1.9 (1.5, 2.3) | 0.559 |
| Creatinine (μmol/L) | 960.4 (737.0, 1173.9) | 969.1 (796.6, 1193.5) | 947.4 (756.0, 1154.0) | 1030.0 (841.0, 1197.7) | 876.0 (606.0, 1167.1) | <0.001 |
| C-reactive protein (mg/L) | 3.2 (1.1, 9.09) | 2.4 (0.9, 5.4) | 2.4 (1.0, 6.5) | 2.7 (0.8, 7.2) | 7.5 (2.6, 30.2) | 0.000 |
| Ferritin (mg/L) | 47.6 (25.5, 112.4) | 45.6 (23.9, 148.0) | 53.9 (29.5, 123.0) | 49.0 (24.7, 101.5) | 43.0 (24.8, 85.4) | 0.109 |
| Urea nitrogen (mmol/L) | 22.9 (18.2, 28.0) | 22.4 (17.6, 27.9) | 22.3 (17.9, 27.8) | 25.0 (19.9, 28.7) | 23.1 (17.9, 27.8) | <0.001 |
| Uric acid (μmol/L) | 450.0 (387.0, 519.0) | 464.0 (399.0, 539.0) | 435.0 (380.0, 482.0) | 453.0 (403.5, 532.0) | 444.0 (345.7, 519.0) | <0.001 |
| eGFR (mL/min/1.73 m2) | 4.0 (3.2, 5.5) | 3.8 (3.2, 5.1) | 3.9 (3.3, 5.3) | 3.9 (3.2, 4.8) | 4.7 (3.3, 7.0) | <0.001 |
| FPG (mg/dL) | 6.7 (5.5, 9.7) | 6.4 (5.2, 8.0) | 6.5 (5.3, 9.3) | 6.6 (5.1, 9.7) | 7.4 (6.0, 11.0) | <0.001 |
| TC (mmol/L) | 3.3 (2.7, 4.1) | 3.3 (2.6, 3.9) | 3.3 (2.7, 4.1) | 3.4 (2.8, 4.0) | 3.4 (2.9, 4.2) | 0.136 |
| TG (mmol/L) | 1.5 (1.1, 2.3) | 1.4 (1.0, 2.0) | 1.4 (1.1, 2.0) | 1.6 (1.2, 2.3) | 1.7 (1.1, 2.5) | <0.001 |
| HDL (mmol/L) | 0.9 (0.8, 1.1) | 0.9 (0.8, 1.1) | 0.9 (0.8, 1.2) | 0.9 (0.8, 1.1) | 0.9 (1.7, 2.9) | 0.423 |
| LDL (mmol/L) | 2.1 (1.7, 2.7) | 2.0 (1.6, 2.6) | 2.2 (1.7, 2.6) | 2.2 (1.7, 2.7) | 2.1 (1.7, 2.9) | 0.037 |
| NLR | 4.0 (3.0, 5.6) | 2.6 (2.0, 3.4) | 3.6 (3.0, 4.3) | 4.4 (3.7, 5.4) | 6.7 (5.4, 9.3) | 0.000 |
| PLR | 150.4 (110.9, 198.4) | 97.5 (77.1, 118.2) | 135.3 (112.3, 158.6) | 171.8 (148.8, 199.1) | 230.8 (187.2, 304.2) | 0.000 |
| MLR | 0.3 (0.3, 0.5) | 0.2 (0.2, 0.3) | 0.2 (0.2, 0.4) | 0.4 (0.3, 0.5) | 0.5 (0.3, 0.7) | 0.000 |
| SII | 662.3 (426.8, 998.1) | 314.1 (228.6, 372.1) | 551.8 (495.0, 608.9) | 786.6 (731.2, 894.1) | 1356.2 (1129.4, 1820.7) | 0.000 |
BMI, body mass index; eGFR, estimated glomerular filtration rate; Bun, blood urea nitrogen; CRP, C-reactive protein; TC, total cholesterol; TG, triglycerides, HDL, high-density lipoprotein cholesterol; LDL, low-density lipoprotein cholesterol; WBC, white blood cell; SII, systemic immune-inflammation index; SIRI, systemic inflammation response index; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; LMR, lymphocyte-to-monocyte ratio
Table 2.
Baseline characteristics stratified by system inflammation response index quartiles
| Clinical characteristics | Overall | Quartiles of system inflammation response index | P value | |||
|---|---|---|---|---|---|---|
| Quartile 1 (N = 296) | Quartile 2 (N = 294) | Quartile 3 (N = 302) | Quartile 4 (N = 298) | |||
| <0.98 | 0.98–1.53 | 1.53–2.31 | ≥2.31 | |||
| Age, years | 62 (54, 72) | 59 (51, 69) | 61 (53, 71) | 61 (53, 72) | 67 (57, 77) | <0.001 |
| Male sex | 762 (64.0) | 156 (52.7) | 176 (59.9) | 214 (70.9) | 216 (72.5) | <0.001 |
| BMI (kg/m2) | 21.2 (19.5, 24.0) | 20.8 (18.9, 22.8) | 21.6 (19.3, 24.2) | 21.4 (20.1, 24.2) | 21.2 (19.5, 24.2) | 0.012 |
| Time on dialysis (years) | 4 (3, 9) | 5 (3, 12) | 5 (3, 9) | 4 (3, 7) | 3 (2, 6) | <0.001 |
| Primary kidney disease, (n, %) | <0.001 | |||||
| Diabetic nephropathy | 426 (35.8) | 54 (18.2) | 98 (33.3) | 118 (39.1) | 156 (52.3) | |
| Glomerulonephritis | 698 (58.7) | 216 (73.0) | 170 (57.8) | 174 (57.6) | 138 (46.3) | |
| Others | 66 (5.5) | 26 (8.8) | 26 (8.8) | 10 (3.3) | 4 (1.3) | |
| Educational level, (n, %) | <0.001 | |||||
| Junior secondary or below | 1016 (85.4) | 236 (79.7) | 256 (87.1) | 272 (90.1) | 252 (84.6) | |
| Senior high school/technical secondary school | 144 (12.1) | 52 (17.6) | 34 (11.6) | 20 (9.9) | 28 (9.4) | |
| Junior college/university or above | 28 (2.4) | 8 (2.7) | 4 (1.4) | 0 (0) | 18 (6.0) | |
| Marital status, (n, %) | 0.025 | |||||
| Unmarried | 74 (6.2) | 22 (7.4) | 20 (6.8) | 14 (4.6) | 18 (6.0) | |
| Married | 936 (78.7) | 222 (75.0) | 242 (82.3) | 238 (78.8) | 234 (78.5) | |
| Divorced | 90 (7.6) | 32 (10.8) | 18 (6.1) | 26 (8.6) | 14 (4.7) | |
| Widowed | 90 (7.6) | 20 (6.8) | 14 (4.8) | 24 (7.9) | 32 (10.7) | |
| Annual family income (RMB thousand yuan), (n, %) | <0.001 | |||||
| <50 | 482 (40.5) | 126 (42.6) | 126 (42.9) | 132 (43.7) | 98 (32.9) | |
| 50 to < 100 | 196 (16.5) | 66 (22.3) | 54 (18.4) | 48 (15.9) | 28 (9.4) | |
| ≥ 100 | 512 (43.0) | 104 (35.1) | 114 (38.8) | 122 (40.4) | 172 (57.70) | |
| Current smoker, (n, %) | 692 (58.2) | 140 (47.3) | 158 (53.7) | 192 (63.6) | 202 (67.8) | <0.001 |
| Current drinker, (n, %) | 750 (63.0) | 166 (56.1) | 168 (57.1) | 196 (64.9) | 220 (73.8) | <0.001 |
| Comorbidities, (n, %) | ||||||
| Hypertension, (n, %) | 360 (30.3) | 72 (24.3) | 74 (25.2) | 90 (29.8) | 124 (41.6) | <0.001 |
| Diabetes mellitus, (n, %) | 430 (36.1) | 56 (18.9) | 102 (34.7) | 116 (38.4) | 156 (52.3) | <0.001 |
| Hypotension, (n, %) | 892 (75.0) | 206 (69.6) | 196 (66.7) | 238 (78.8) | 252 (84.6) | <0.001 |
| Cardiovascular disease, (n, %) | 550 (46.2) | 106 (35.8) | 118 (40.1) | 138 (45.7) | 188 (63.1) | <0.001 |
| Anti-hypertension, (n, %) | 360 (30.3) | 122 (41.2) | 86 (29.3) | 70 (23.2) | 82 (27.5) | <0.001 |
| Anti-diabetes, (n, %) | 366 (30.8) | 56 (18.9) | 88 (29.9) | 102 (33.8) | 120 (40.3) | <0.001 |
| Hypolipidemic agents, (n, %) | 546 (45.9) | 106 (35.8) | 122 (41.5) | 102 (33.8) | 174 (58.4) | <0.001 |
| Laboratory of parameters | ||||||
| Hemoglobin (g/L) | 108.0 (97.0, 117.0) | 109.0 (100.3, 118.0) | 107.0 (98.0, 116.0) | 109.0 (96.0, 117.0) | 106.0 (93.0, 117.0) | 0.027 |
| Albumin (g/L) | 36.8 (34.5, 38.8) | 37.3 (35.2, 39.3) | 36.9 (35.3, 38.9) | 36.8 (34.1, 38.9) | 36.2 (32.5, 38.0) | <0.001 |
| Calcium (mmol/L) | 2.2 (2.1, 2.4) | 2.3 (2.1, 2.4) | 2.2 (2.1, 2.4) | 2.2 (2.1, 2.4) | 2.2 (2.0, 2.4) | <0.001 |
| Phosphorus (mmol/L) | 1.9 (1.5, 2.3) | 2.0 (1.5, 2.3) | 1.8 (1.5, 2.3) | 2.0 (1.5, 2.4) | 1.8 (1.4, 2.3) | 0.050 |
| Creatinine (μmol/L) | 960.4 (737.0, 1173.9) | 957.2 (774.6, 1154.6) | 975.8 (766.0, 1215.4) | 1022.1 (820.3, 1197.0) | 844.7 (618.5, 1133.9) | <0.001 |
| C-reactive protein (mg/L) | 3.2 (1.1, 9.1) | 2.0 (0.7, 4.8) | 2.5 (1.0, 6.0) | 3.3 (1.1, 8.6) | 7.9 (2.9, 35.0) | 0.000 |
| Ferritin (mg/L) | 47.6 (25.5, 112.4) | 53.6 (29.5, 158.5) | 53.6 (25.8, 99.3) | 39.8 (21.0, 96.2) | 49.0 (25.7, 114.9) | <0.001 |
| Urea nitrogen (mmol/L) | 22.9 (18.2, 28.0) | 22.6 (18.0, 27.8) | 22.0 (18.5, 26.8) | 24.6 (18.0, 28.5) | 23.4 (17.5, 28.8) | 0.199 |
| Uric acid (μmol/L) | 450.0 (387.0, 519.0) | 464.0 (400.5, 517.0) | 443.0 (393.0, 522.0) | 459.0 (389.0, 534.0) | 430.0 (361.5, 502.2) | 0.009 |
| eGFR (mL/min/1.73 m2) | 4.0 (3.2, 5.5) | 3.9 (3.2, 5.3) | 3.8 (3.3, 5.2) | 3.9 (3.1, 4.9) | 4.5 (3.3, 6.8) | <0.001 |
| FPG (mg/dL) | 6.7 (5.5, 9.7) | 6.5 (5.3, 8.1) | 6.8 (5.4, 9.3) | 6.6 (5.6, 9.9) | 7.2 (5.6, 11.2) | 0.035 |
| TC (mmol/L) | 3.3 (2.7, 4.1) | 3.3 (2.8, 4.0) | 3.3 (2.6, 4.2) | 3.4 (2.7, 4.1) | 3.2 (2.7, 3.8) | 0.214 |
| TG (mmol/L) | 1.5 (1.1, 2.3) | 1.4 (1.1, 2.1) | 1.4 (1.1, 2.3) | 1.5 (1.1, 2.2) | 1.6 (1.1, 2.3) | 0.635 |
| HDL (mmol/L) | 0.9 (0.8, 1.1) | 1.0 (0.8, 1.1) | 0.9 (0.8, 1.2) | 0.9 (0.8, 1.1) | 0.9 (0.8, 1.1) | 0.154 |
| LDL (mmol/L) | 2.1 (1.7, 2.7) | 2.1 (1.7, 2.7) | 2.1 (1.6, 2.7) | 2.2 (1.7, 2.7) | 2.1 (1.6, 2.6) | 0.240 |
| NLR | 4.0 (3.0, 5.6) | 2.5 (2.0, 3.2) | 3.7 (3.1, 4.5) | 4.4 (3.5, 5.3) | 6.8 (5.3, 9.3) | 0.000 |
| PLR | 150.4 (110.9, 198.4) | 118.2 (91.0, 152.9) | 143.2 (106.4, 191.3) | 157.7 (122.1, 208.0) | 187.1 (140.1, 254.1) | 0.000 |
| MLR | 0.3 (0.3, 0.5) | 0.2 (0.2, 0.3) | 0.3 (0.3, 0.4) | 0.4 (0.3, 0.5) | 0.6 (0.5, 0.8) | 0.000 |
| SIRI | 1.5 (1.0, 2.3) | 0.7 (0.5, 0.8) | 1.2 (1.1, 1.4) | 1.8 (1.7, 2.1) | 3.6 (2.8, 5.4) | 0.000 |
BMI, body mass index; eGFR, estimated glomerular filtration rate; Bun, blood urea nitrogen; CRP, C-reactive protein; TC, total cholesterol; TG, triglycerides, HDL, high-density lipoprotein cholesterol; LDL, low-density lipoprotein cholesterol; WBC, white blood cell; SII, systemic immune-inflammation index; SIRI, systemic inflammation response index; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; LMR, lymphocyte-to-monocyte ratio
Impacts of SII and SIRI with mortality
In this cohort, the overall all-cause mortality rate was 38.2%, with CVD mortality representing 16.9% and infection-related mortality accounting for 5.2%. Mortality rates showed a significant increasing trend across higher quartiles of SII and SIRI (P < 0.001 for trend) (Table 3).
Table 3.
Mortality outcomes in patients with hemodialysis
| Outcome, no. of events | Overall | Quartiles of systemic-immune inflammation index | P value | Quartiles of system inflammation response index | P value | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Quartile 1 (N = 316) | Quartile 2 (N = 278) | Quartile 3 (N = 298) | Quartile 4 (N = 298) | Quartile 1 (N = 296) | Quartile 2 (N = 294) | Quartile 3 (N = 302) | Quartile 4 (N = 298) | ||||
| <426.78 | 426.78–662.30 | 662.30–998.13 | ≥998.13 | <0.98 | 0.98–1.53 | 1.53–2.31 | ≥2.31 | ||||
| All-cause mortality | 455 (38.2) | 99 (31.3) | 88 (31.7) | 125 (41.9) | 143 (48.0) | <0.001 | 83 (28.0) | 110 (37.4) | 121 (40.1) | 141 (47.3) | <0.001 |
| Cardiovascular mortality | 201 (16.9) | 26 (8.2) | 28 (10.1) | 52 (17.4) | 95 (31.9) | <0.001 | 23 (7.8) | 37 (12.6) | 50 (16.6) | 91 (30.5) | <0.001 |
| Infection-related mortality | 62 (5.2) | 9 (2.8) | 8 (2.9) | 19 (6.4) | 26 (8.7) | <0.001 | 7 (2.4) | 11 (3.7) | 16 (5.3) | 28 (9.4) | <0.001 |
We conducted Cox proportional hazards regression and restricted cubic spline (RCS) analyses to evaluate the prognostic value of SII, SIRI, NLR, PLR, MLR, and monocytes for all-cause, CVD, and infection-related mortality using a comprehensive multivariable-adjusted model (Table 4 and Fig. 2). This model accounted for age, sex, BMI, dialysis duration, primary kidney disease, marital status, education level, smoking and drinking status, annual family income, hypertension, diabetes mellitus, hypotension, cardiovascular disease, and laboratory parameters including hemoglobin, albumin, calcium, phosphate, creatinine, CRP, ferritin, blood urea nitrogen, uric acid, eGFR, FPG, TC, TG, HDL, and LDL. For all-cause mortality, Quartile 4 of SII and SIRI exhibited the highest hazard ratios (HRs of 2.26 and 2.29, respectively). SII demonstrated a nonlinear relationship (P for nonlinearity = 0.012), with an inflection point at 582, while SIRI showed a linear trend (P = 0.004). NLR and monocyte were also significantly associated with all-cause mortality, though their effect sizes were more modest. For CVD mortality, SII and SIRI remained significant predictors. Quartile 4 showed HRs of 3.33 and 3.78, respectively. SII exhibited a nonlinear relationship (P for nonlinearity < 0.001), with an inflection point at 464, whereas SIRI followed a linear trend (P = 0.005). NLR and monocytes were associated with an increased risk of CVD mortality, while PLR and MLR showed weaker or nonsignificant associations. Lastly, infection-related mortality was significantly associated with higher quartiles of both SII and SIRI. Quartile 4 of SII (HR = 1.70, P < 0.001) and SIRI (HR = 2.42, P < 0.001) showed an increased risk in both crude and adjusted model. RCS analysis revealed significant nonlinear associations for both markers, with SIRI exhibiting a sharp increase in risk (P for overall < 0.001, P for nonlinear = 0.001) and SII showing a similar trend (P for overall = 0.010, P for nonlinear = 0.060). Additionally, higher quartiles of NLR, PLR, MLR, and monocytes were significantly associated with increased infection-related mortality risk.
Table 4.
Association between quartiles of SII, SIRI, NLR, PLR, MLR, monocyte and death in patients with hemodialysis
| Outcomes | Quartile 1 (Ref) | Quartile 2 | Quartile 3 | Quartile 4 | Pvalue | |
|---|---|---|---|---|---|---|
| SII | All-cause mortality | |||||
| Crude model | 1.00 | 1.39 (0.92, 2.09) | 0.93 (0.61, 1.42) | 2.32 (1.62, 3.30)*** | <0.001 | |
| Model 1 | 1.00 | 1.13 (0.75, 1.71) | 0.96 (0.63, 1.46) | 2.05 (1.43, 2.94)**** | <0.001 | |
| Model 2 | 1.00 | 1.24 (0.76, 2.02) | 0.99 (0.60, 1.66) | 2.26 (1.43, 3.58)**** | <0.001 | |
| CVD mortality | ||||||
| Crude model | 1.00 | 1.94 (0.93, 4.04) | 2.30 (1.15, 4.62)*** | 3.08 (1.67, 5.65)**** | 0.002 | |
| Model 1 | 1.00 | 1.97 (0.94, 4.12) | 2.18 (1.08, 4.38)*** | 2.90 (1.57, 5.36)**** | 0.006 | |
| Model 2 | 1.00 | 2.00 (0.75, 5.32) | 2.20 (0.86, 5.67) | 3.33 (1.45, 7.69)*** | 0.035 | |
| Infection-related mortality | ||||||
| Crude model | 1.00 | 1.21 (0.85, 2.12) | 1.68 (1.18, 3.20)*** | 2.24 (1.88, 3.70)**** | <0.001 | |
| Model 1 | 1.00 | 1.14 (0.65, 2.05) | 1.52 (1.12, 2.99)*** | 2.12 (1.59, 2.84)**** | <0.001 | |
| Model 2 | 1.00 | 1.11 (0.72, 1.98) | 1.44 (1.04, 2.72)*** | 1.70 (1.25, 2.32)**** | <0.001 | |
| SIRI | All-cause mortality | |||||
| Crude model | 1.00 | 1.95 (1.25, 3.05)*** | 1.80 (1.15, 2.83)* | 3.36 (2.24, 5.05)*** | <0.001 | |
| Model 1 | 1.00 | 1.91 (1.21, 2.99)*** | 1.87 (1.18, 2.95)*** | 2.70 (1.77, 4.12)**** | <0.001 | |
| Model 2 | 1.00 | 1.66 (0.98, 2.82) | 1.26 (0.73, 2.16) | 2.29 (1.38, 3.80)*** | 0.003 | |
| CVD mortality | ||||||
| Crude model | 1.00 | 1.66 (0.77, 3.56) | 1.60 (0.74, 3.44) | 2.63 (1.32, 5.26)* | 0.008 | |
| Model 1 | 1.00 | 1.65 (0.77, 3.54) | 1.48 (0.69, 3.20) | 2.41 (1.20, 4.84)* | 0.024 | |
| Model 2 | 1.00 | 2.13 (0.82, 5.56) | 1.94 (0.72, 5.18) | 3.78 (1.43, 10.01)*** | 0.031 | |
| Infection-related mortality | ||||||
| Crude model | 1.00 | 1.52 (1.08, 2.22)*** | 2.26 (1.68, 3.22)**** | 2.55 (1.85, 3.32)**** | <0.001 | |
| Model 1 | 1.00 | 1.45 (1.02, 2.08)*** | 2.14 (1.62, 3.08)**** | 2.52 (1.74, 3.08)**** | <0.001 | |
| Model 2 | 1.00 | 1.32 (1.05, 2.11)*** | 2.03 (1.60, 2.98)**** | 2.42 (1.70, 3.01)**** | <0.001 | |
| NLR | All-cause mortality | |||||
| Crude model | 1.00 | 0.65 (0.42, 1.04) | 1.42 (1.13, 1.76)*** | 1.61 (1.14, 2.25)*** | <0.001 | |
| Model 1 | 1.00 | 0.81 (0.52, 1.24) | 1.21 (1.08, 1.51)*** | 1.06 (1.05, 1.07)**** | <0.001 | |
| Model 2 | 1.00 | 1.33 (0.80, 2.23) | 1.18 (1.11, 1.86)*** | 1.99 (1.24, 3.18)*** | 0.002 | |
| CVD mortality | ||||||
| Crude model | 1.00 | 2.30 (1.18, 4.51)* | 2.40 (1.30, 4.44)*** | 2.37 (1.35, 3.16)*** | 0.020 | |
| Model 1 | 1.00 | 2.22 (1.15, 4.28)* | 2.36 (1.34, 4.32)*** | 2.37 (1.35, 4.16)*** | 0.020 | |
| Model 2 | 1.00 | 2.07 (0.84, 5.12) | 2.34 (1.04, 5.24)* | 2.49 (1.10, 5.66)* | 0.046 | |
| Infection-related mortality | ||||||
| Crude model | 1.00 | 1.01 (0.92, 1.48) | 1.21 (1.08, 1.58)* | 1.45 (1.11, 1.98)*** | <0.001 | |
| Model 1 | 1.00 | 0.98 (0.72, 1.77) | 1.15 (1.04, 1.77)* | 1.32 (1.15, 1.78)*** | <0.001 | |
| Model 2 | 1.00 | 1.05 (0.94, 1.56) | 1.16 (1.01, 1.86)* | 1.38 (1.18, 1.92)*** | <0.001 | |
| PLR | All-cause mortality | |||||
| Crude model | 1.00 | 0.82 (0.57, 1.19) | 1.22 (1.11, 1.92)* | 1.32 (1.18, 1.70)* | 0.003 | |
| Model 1 | 1.00 | 0.94 (0.65, 1.36) | 0.72 (0.48, 1.07) | 1.17 (0.84, 1.63) | 0.099 | |
| Model 2 | 1.00 | 0.82 (0.54, 1.26) | 0.87 (0.55, 1.39) | 1.04 (0.68, 1.58) | 0.668 | |
| CVD mortality | ||||||
| Crude model | 1.00 | 1.89 (1.07, 3.35)* | 2.65 (1.52, 4.61)**** | 1.98 (1.22, 3.22)*** | 0.006 | |
| Model 1 | 1.00 | 1.94 (1.09, 3.44)* | 2.40 (1.36, 4.26)*** | 1.90 (1.16, 3.11)* | 0.020 | |
| Model 2 | 1.00 | 1.82 (0.94, 3.49) | 2.86 (1.41, 5.81)*** | 1.77 (0.91, 3.43) | 0.037 | |
| Infection-related mortality | ||||||
| Crude model | 1.00 | 1.25 (0.92, 1.88) | 1.56 (1.30, 1.92)*** | 1.88 (1.45, 2.12)**** | <0.001 | |
| Model 1 | 1.00 | 1.45 (1.02, 2.08)*** | 1.50 (1.32, 1.88)*** | 1.85 (1.38, 2.04)*** | <0.001 | |
| Model 2 | 1.00 | 1.32 (1.05, 2.11)* | 1.52 (1.28, 1.89)*** | 1.78 (1.32, 2.18)*** | 0.002 | |
| MLR | All-cause mortality | |||||
| Crude model | 1.00 | 1.34 (0.89, 2.01) | 1.42 (0.94, 2.15) | 2.19 (1.50, 3.18)**** | <0.001 | |
| Model 1 | 1.00 | 1.26 (0.84, 1.89) | 1.28 (0.84, 1.95) | 1.68 (1.13, 2.50)* | 0.071 | |
| Model 2 | 1.00 | 1.19 (0.74, 1.92) | 1.32 (0.81, 2.16) | 1.44 (0.91, 2.28) | 0.469 | |
| CVD mortality | ||||||
| Crude model | 1.00 | 1.51 (0.83, 2.73) | 1.19 (0.64, 2.22) | 1.86 (1.08, 3.20)* | 0.079 | |
| Model 1 | 1.00 | 1.50 (0.83, 2.73) | 1.19 (0.64, 2.22) | 1.74 (1.01, 3.01)* | 0.156 | |
| Model 2 | 1.00 | 0.75 (0.43, 2.33) | 1.94 (0.83, 4.53) | 3.00 (1.68, 4.54)* | 0.042 | |
| Infection-related mortality | ||||||
| Crude model | 1.00 | 1.45 (1.25, 1.99)*** | 1.66 (1.32, 2.18)**** | 2.12 (1.44, 2.62)**** | <0.001 | |
| Model 1 | 1.00 | 1.40 (1.21, 1.82)*** | 1.58 (1.28, 2.22)**** | 2.08 (1.55, 2.82)**** | <0.001 | |
| Model 2 | 1.00 | 1.28 (1.12, 1.48)* | 1.62 (1.26, 2.33)**** | 2.18 (1.52, 2.95)**** | <0.001 | |
| Mon | All-cause mortality | |||||
| Crude model | 1.00 | 1.62 (1.05, 2.49)* | 2.29 (1.55, 3.39)**** | 2.63 (1.79, 3.87)**** | <0.001 | |
| Model 1 | 1.00 | 1.51 (0.98, 2.32) | 1.73 (1.16, 2.58)*** | 1.70 (1.13, 2.56)* | 0.044 | |
| Model 2 | 1.00 | 2.11 (1.25, 3.56)*** | 2.15 (1.33, 3.45)*** | 1.74 (1.01, 2.99)* | 0.010 | |
| CVD mortality | ||||||
| Crude model | 1.00 | 0.76 (0.41, 1.42) | 1.19 (1.02, 2.00)* | 1.30 (1.21, 2.14)*** | 0.235 | |
| Model 1 | 1.00 | 0.79 (0.42, 1.47) | 1.12 (1.04, 1.81)* | 1.41 (1.15, 2.34)*** | 0.165 | |
| Model 2 | 1.00 | 1.27 (0.47, 3.48) | 2.48 (1.05, 5.88)* | 2.10 (1.96, 4.30)*** | 0.032 | |
| Infection-related mortality | ||||||
| Crude model | 1.00 | 1.08 (0.82, 1.45) | 1.12 (1.01, 1.22)*** | 1.32 (1.14, 2.52)* | <0.001 | |
| Model 1 | 1.00 | 1.02 (0.92, 1.82) | 1.08 (1.03, 1.12)* | 1.44 (1.12, 1.82)*** | 0.004 | |
| Model 2 | 1.00 | 1.08 (0.84, 1.77) | 1.02 (1.01, 1.44)* | 1.36 (1.18, 1.92)*** | 0.015 | |
Model 1 adjusted for sex and age
Model 2 was fully adjusted for age, sex, BMI, time on dialysis, primary kidney disease, marital status, education level, smoking and drinking status, annual family income, hypertension, diabetes mellitus, hypotension, cardiovascular disease, hemoglobin, albumin, calcium, phosphate, creatinine, CRP, ferritin, blood urea nitrogen, uric acid, eGFR, FPG, TC, TG, HDL, and LDL
SII, systemic immune-inflammation index; SIRI, systemic inflammation response index; NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio; MLR: monocyte-to-lymphocyte ratio; CVD, cardiovascular disease
Statistical significance is indicated for comparisons of Quartiles 2, 3, and 4 versus Quartile 1 (reference group), based on multivariable Cox regression analyses: *P < 0.05; **P < 0.01; **P < 0.001
Fig. 2.
Restricted cubic spline analysis between SIRI and the risk of all-cause mortality (A), CVD mortality (C), and infection-related mortality (E) in participants with hemodialysis. Restricted cubic spline analysis between SII and the risk of all-cause mortality (B), CVD mortality (D), and infection-related mortality (F) in participants with hemodialysis. Blue lines represent references for odds ratios, and blue areas represent 95% confidence intervals. The model was adjusted for age, sex, body mass index, time on dialysis, primary kidney disease, marital status, education level, smoking and drinking status, annual family income, hypertension, diabetes mellitus, hypotension, cardiovascular disease, hemoglobin, albumin, calcium, phosphate, creatinine, C-reactive protein, ferritin, blood urea nitrogen, uric acid, estimated glomerular filtration rate, fasting blood-glucose, total cholesterol, triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol
Moreover, SII and SIRI were analyzed as categorical variables. Kaplan-Meier curves demonstrated significantly different survival patterns across quartiles of systemic inflammatory indices (all P < 0.0001) (Fig. 3), indicating that higher quartiles of SII and SIRI corresponded to lower survival probabilities. Cumulative hazard curves for all-cause, CVD, and infection-related mortality across SII and SIRI quartiles are illustrated in Fig. 3. Patients in the highest quartile of SII (quartile 4) experienced significantly elevated rates of all-cause, CVD, and infection-related mortality compared to those in the lowest quartile (quartile 1) (all log-rank test P values < 0.05). These findings emphasize the strong association between systemic inflammation, reflected by SII and SIRI, and adverse mortality outcomes.
Fig. 3.
Kaplan-Meier analysis of all-cause mortality (A), CVD mortality (C), and infection-related mortality (E) based on SII groups among individuals with hemodialysis. Kaplan-Meier analysis of all-cause mortality (B), CVD mortality (D), and infection-related mortality (F) based on SIRI groups among individuals with hemodialysis
The predictive ability of SII and SIRI for mortality in HD patients
ROC curves were generated to assess the predictive accuracy of SII, SIRI, NLR, PLR, MLR, and monocytes for all-cause, CVD, and infection-related mortality (Fig. 4 and Fig. S1). Time-dependent ROC analysis and Delong’s test demonstrated that SIRI consistently yielded higher AUC values at 36, 60, and 96 months for all-cause mortality, with statistical significance reached at 36 months (AUC = 0.648, P = 0.042 vs. SII) and 60 months (AUC = 0.645, P = 0.016 vs. SII), though not at 96 months (AUC = 0.630, P = 0.143). In comparison, SII’s AUC ranged from 0.549 to 0.589. Other markers (NLR, PLR, MLR, and monocytes) showed lower and less stable predictive performance.
Fig. 4.
The time-dependent ROC curves for displaying the predictive abilities of different systemic inflammatory indicators (SII and SIRI). SII: systemic immune-inflammation index; SIRI: systemic inflammation response index; ROC: receiver operating characteristic
For CVD mortality, both SII and SIRI demonstrated higher predictive accuracy compared to other inflammatory markers at 36, 60, and 96 months (Fig. 4 and Fig. S1). SII showed AUC values ranging from 0.671 to 0.707, while SIRI exhibited comparable performance with AUC values between 0.667 and 0.737. SII performed slightly better at 96 months, whereas SIRI had a marginally higher AUC at 36 and 60 months (All P > 0.05). In contrast, NLR, PLR, MLR, and monocyte counts consistently showed lower AUC values across all time points.
Regarding infection-related mortality, SIRI demonstrated significantly better predictive power at all time points: AUC = 0.641 (P = 0.021 vs. SII) at 36 months, 0.658 (P = 0.035 vs. SII) at 60 months, and 0.640 (P = 0.044 vs. SII) at 96 months. In contrast, SII’s AUCs were notably lower (0.492, 0.525, and 0.578, respectively). Among other markers, NLR showed moderate performance, while PLR, MLR, and monocytes had weak and inconsistent associations. These findings further support the greater predictive utility of SIRI for this mortality outcome.
Overall, these results highlight that both SII and SIRI possess moderate predictive capabilities for mortality in hemodialysis patients. Specifically, SIRI demonstrated superior predictive accuracy for all-cause and infection-related mortality. Meanwhile, both indices showed similar predictive performance for CVD mortality.
Discussion
In this study, we evaluated the prognostic value of SII and SIRI in predicting all-cause, CVD, and infection-related mortality in patients undergoing maintenance hemodialysis. Higher levels of both SII and SIRI were significantly associated with an increased risk of mortality compared to lower levels, even after adjusting for confounders. Further analysis revealed a positive correlation between these markers and both all-cause and CVD mortality risk. Notably, SII and SIRI outperformed other inflammatory markers, including NLR, PLR, MLR, and monocytes, in predicting mortality. Among these, SIRI emerged as the most effective marker for predicting infection-related mortality. These findings underscore the importance of incorporating SII and SIRI into mortality risk assessments, suggesting that monitoring these indices could be a critical strategy in efforts to reduce mortality in HD patients.
Preceding studies have identified chronic inflammation as an important risk factor for several diseases such as cancer, diabetes, and atherosclerotic disease [25–27]. As simple, reliable, and minimally invasive biomarkers, SII and SIRI have been shown to serve as powerful prognostic indices in various inflammation-related diseases [28–30]. Chronic inflammation is well-recognized as a major contributor to the pathogenesis and progression of chronic kidney disease and is closely associated with the prognosis of patients undergoing dialysis [20–22]. Moreover, immune cells play a critical role in cardiac injury and repair, with numerous studies indicating that SII serves as an independent predictor of adverse outcomes in patients with ESKD, including mortality and vascular access survival in HD populations [20, 31–35]. Collectively, these findings, along with our own, suggest that SII and SIRI may significantly influence all-cause and CVD mortality in HD patients.
Monocytes are key players in the development of cardiovascular disease, with higher monocyte levels linked to CVD mortality, coronary plaque formation, and atherosclerosis [37–39]. Specifically, intermediate monocytes are associated with reduced left ventricular ejection fraction and a higher risk of cardiovascular events in both the general population and CKD patients [40–42]. These monocytes contribute to inflammation by releasing pro-inflammatory cytokines and expressing receptors involved in atherosclerosis [37]. Elevated monocyte levels are also predictive of poor outcomes in dialysis patients and are associated with worsening kidney function [42]. Our study further confirms that monocytes remain significantly associated with all-cause, cardiovascular, and infection-related mortality risk, supporting their role in adverse clinical outcomes.
In addition to monocytes, our findings demonstrate that NLR remains positively associated with both all-cause and cardiovascular mortality risk, even after adjusting for confounders, consistent with previous studies [40, 42, 43]. These studies have demonstrated that NLR exhibits the highest sensitivity and specificity for predicting 30-day mortality and is linked to longer hospital stays and increased dialysis frequency [43]. Furthermore, elevated values of NLR, MLR, and PLR correlate with the progression of chronic kidney disease, cardiovascular disease, and mortality, with higher NLR values particularly associated with an increased risk of CVD and death in dialysis patients [40, 42]. These results highlight the continued importance of traditional inflammatory markers such as monocytes and NLR in predicting mortality risk. However, our findings suggest that integrating markers like SIRI and SII, which account for a broader range of immune responses, may provide even more accurate risk stratification, especially in hemodialysis patients where immune dysregulation and chronic inflammation are prevalent.
In contrast to these traditional markers, SIRI and SII, as composite indices, integrate multiple immune cell types, offering a broader evaluation of the equilibrium between systemic inflammation and immune response [18, 19]. SIRI combines neutrophils, monocytes, and lymphocytes, reflecting both acute and chronic inflammatory responses as well as immune suppression [19, 44]. Neutrophils and monocytes suggest systemic inflammation, while lymphocytes indicate impaired immune competence [19, 44]. SII, on the other hand, incorporates platelets, which capture both inflammatory and coagulation processes [45]. However, platelet levels are influenced by various non-inflammatory factors, such as thrombocytopenia or uremic platelet dysfunction, common in dialysis patients [46, 47]. In contrast, monocytes offer greater specificity in reflecting chronic immune dysregulation [48, 49]. Thus, SIRI, by integrating markers of both immune activation and suppression, may provide a more accurate reflection of immune dysfunction, particularly in hemodialysis patients, where chronic low-grade inflammation and immune impairment are prevalent.
Mechanistically, chronic inflammation contributes to endothelial dysfunction, arterial stiffness, and vascular calcification, which are key determinants of CVD morbidity and mortality in this demographic [26, 50–54]. Furthermore, persistent immune activation, characterized by elevated levels of pro-inflammatory cytokines such as IL-6, IL-8, and IL-1β, exacerbates atherosclerosis and promotes arrhythmogenesis, thereby increasing the risk of sudden cardiac death [55–58]. These mechanisms may elucidate why patients with elevated SII and SIRI levels face a heightened risk for CVD mortality.
Given the significant association between SII, SIRI, and mortality risk, regular monitoring of these indices could be valuable for guiding clinical decision-making in HD patients. Identifying patients with elevated SII or SIRI levels may enable clinicians to initiate targeted interventions, such as improving dialysis quality, optimizing nutritional support, or introducing anti-inflammatory therapies, to reduce inflammation-related risks [59, 60]. Interventions aimed at lowering systemic inflammation, such as statins, omega-3 fatty acids, antioxidants, and other anti-inflammatory agents, have shown beneficial effects and may play a crucial role in improving clinical outcomes in HD patients [59–63]. As dialysis quality improves and patients’ lifestyles are better managed, controlling inflammation levels could lead to significant improvements in clinical outcomes.
The identification of SII and SIRI as significant prognostic markers in HD patients holds important clinical implications. There is an increasing need for robust, easily accessible biomarkers to guide risk stratification and treatment decisions in this population. Derived from routine hematological tests, SII and SIRI provide practical, cost-effective tools for clinical use. These composite markers offer a more comprehensive and accurate assessment of systemic inflammation compared to simpler markers like monocytes, improving risk prediction. Early identification of patients with elevated SII or SIRI could prompt more aggressive interventions to reduce systemic inflammation, such as tailored anti-inflammatory therapies, lifestyle changes, or adjustments to dialysis protocols. Furthermore, monitoring these indices may enable clinicians to implement targeted strategies to improve clinical outcomes. Given their strong prognostic value, future studies should assess whether interventions aimed at lowering SII or SIRI levels can enhance survival outcomes in HD patients and refine risk stratification.
Study limitations and future directions
Several limitations of this study should be acknowledged. First, as a single-center retrospective analysis, the generalizability of our findings to other populations, particularly those in different geographical regions or clinical settings, may be limited. The observational design also precludes establishing a causal relationship between systemic inflammation indices and mortality, and reverse causality cannot be ruled out. Additionally, unmeasured confounders, such as dialysis adequacy, medication use, or undiagnosed comorbidities, may have influenced our results, despite adjusting for a wide range of known covariates. The possibility of residual confounding remains. Moreover, while our study primarily focused on all-cause and cardiovascular mortality, we did not systematically collect data on non-fatal cardiovascular events, such as myocardial infarctions, strokes, or hospitalizations due to heart failure. Furthermore, we only assessed SII and SIRI at baseline and did not consider longitudinal changes in these markers, which could have provided a more comprehensive understanding of their dynamic relationship with mortality.
Future research should focus on validating our findings in larger, multicenter cohorts to improve generalizability and further explore the association between SII, SIRI, and infection-related mortality in HD patients. Additionally, evaluating whether serial measurements of these indices can enhance risk stratification and provide more dynamic insights into patient prognosis is warranted. Investigating the role of these indices in combination with other established inflammatory markers, such as CRP and albumin, may further improve the predictive accuracy of mortality risk models. Lastly, randomized controlled trials are needed to determine whether targeting systemic inflammation through specific therapeutic interventions can improve survival outcomes and reduce infection-related mortality in this high-risk population.
Conclusion
This study underscores the potential prognostic value of SII and SIRI in predicting all-cause, cardiovascular, and infection-related mortality in patients undergoing maintenance HD. These indices, which reflect systemic inflammation, may offer insights into risk stratification in this patient population. However, while SII and SIRI show promise as biomarkers, further research is required to validate these findings in larger, diverse cohorts. Additionally, the practical utility of these markers in guiding therapeutic interventions remains to be fully explored before they can be recommended for routine clinical use.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
We would like to thank all those who helped with this study.
Author contribution
Liang Dai contributed to the conception and design of the work, analyzed the data, wrote the manuscript, and critically revised the manuscript. Qianqian Zhu contributed to the acquisition of the data, analyzed the data, wrote the manuscript, and critically revised the manuscript. All authors have approved the final version of the manuscript for publication.
Funding
No funding.
Data availability
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Human ethics and consent to participate declarations
Ethics Committee of The Traditional Chinese Medicine Hospital of Pujiang County approved the study (2024C04), Written informed consent was obtained from all study subjects in accordance with the Declaration of Helsinki.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Anand SK, et al. Machine-learning based subgroups of AL amyloidosis and cumulative incidence of mortality and end stage kidney disease. Am J Hematol. 2024. [DOI] [PubMed]
- 2.Koyama AK, et al. Homelessness and risk of end-stage kidney disease and death in veterans with chronic kidney disease. JAMA Netw Open. 2024;7(9):e2431973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Al Awadhi S, et al. A Metabolomics approach to identify metabolites associated with mortality in patients receiving maintenance hemodialysis. Kidney Int Rep. 2024;9(9):2718–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pinter J, et al. Associations of abnormal fluid status, plasma sodium disorders, and low dialysate sodium with mortality in hemodialysis patients. Clin J Am Soc Nephrol. 2024. [DOI] [PMC free article] [PubMed]
- 5.Kosugi T, et al. Association between hemoglobin level and mortality in patients undergoing maintenance hemodialysis: A nationwide dialysis registry in Japan. Clin Exp Nephrol. 2025. [DOI] [PMC free article] [PubMed]
- 6.Men R, et al. Associations between serum potassium variability and mortality in patients undergoing maintenance hemodialysis: A retrospective study. Sci Rep. 2024;14(1):p.29998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Usui N, et al. Association of physiological reserve obtained from cardiopulmonary exercise testing and frailty with all-cause mortality in patients on hemodialysis. Clin J Am Soc Nephrol. 2024. [DOI] [PMC free article] [PubMed]
- 8.Ito H, Safety of apixaban and rivaroxaban in patients with atrial fibrillation on hemodialysis. Eur J Intern Med. 2024;124:154–55. [DOI] [PubMed] [Google Scholar]
- 9.Zheng SL, et al. Genome-wide association analysis provides insights into the molecular etiology of dilated cardiomyopathy. Nat Genet. 2024;56(12):2646–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sircuța AF, et al. The Relationship between circulating kidney injury molecule-1 and cardiovascular morbidity and mortality in hemodialysis patients. Biomedicines. 2024;12(8). [DOI] [PMC free article] [PubMed]
- 11.Ren S, et al. The predictive value of systemic immune-inflammation index for vascular access survival in chronic hemodialysis patients. Front Immunol. 2024;15:1382970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zununi Vahed S, et al. The impacts of dietary antioxidants on cardiovascular events in hemodialysis patients: An update on the cellular and molecular mechanisms. Nutr Rev. 2024. [DOI] [PubMed]
- 13.Arabi SM, et al. The effects of curcumin-containing supplements on inflammatory biomarkers in hemodialysis patients: A systematic review and meta-analysis. Phytother Res. 2022;36(12):4361–70. [DOI] [PubMed] [Google Scholar]
- 14.Dember LM, et al. A randomized controlled pilot trial of anakinra for hemodialysis inflammation. Kidney Int. 2022;102(5):1178–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zhao H, et al, Expanded Hemodialysis ameliorates uremia-induced impairment of vasculoprotective KLF2 and concomitant proinflammatory priming of endothelial cells through an ERK/AP1/cFOS-dependent mechanism. Front Immunol. 2023;14:1209464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Luo J, et al, Prognostic implications of systemic immune-inflammation index in myocardial infarction patients with and without diabetes: Insights from the NOAFCAMI-SH registry. Cardiovasc Diabetol. 2024;23(1):41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chen X, Li A, Ma Q, Neutrophil-lymphocyte ratio and systemic immune-inflammation index as predictors of cardiovascular risk and mortality in prediabetes and diabetes: A population-based study. Inflammopharmacology. 2024;32(5):3213–27. [DOI] [PubMed] [Google Scholar]
- 18.Fu C, et al, Association between complete blood cell count-derived inflammatory biomarkers and gallstones prevalence in American adults under 60 years of age. Front Immunol. 2024;15:1497068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Li B, et al, Peripheral blood immunoinflammatory biomarkers: Prospective predictors of postoperative long-term survival and chronic postsurgical pain in breast cancer. Front Immunol. 2025;16:1531639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chen Y, et al, Association between systemic inflammatory indicators with the survival of chronic kidney disease: A prospective study based on NHANES. Front Immunol. 2024;15:1365591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ren Z, et al, The advanced lung cancer inflammation index is associated with mortality in peritoneal dialysis patients. BMC Nephrol. 2024;25(1):208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lai W, et al, Elevated systemic immune inflammation level increases the risk of total and cause-specific mortality among patients with chronic kidney disease: a large multi-center longitudinal study. Inflamm Res. 2023;72(1):149–58. [DOI] [PubMed] [Google Scholar]
- 23.Zhang Y, et al, Dietary fatty acids intake and all-cause and cardiovascular mortality in patients on peritoneal dialysis. Clin Nutr. 2023;42(11):2188–97. [DOI] [PubMed] [Google Scholar]
- 24.Kitamura H, et al, Serum alkaline phosphatase and infection-related mortality in hemodialysis patients: Ten-year outcomes of the q-cohort study. Clin Exp Nephrol. 2022;26(11):1119–29. [DOI] [PubMed] [Google Scholar]
- 25.Ait-Oufella H, Libby P, Inflammation and atherosclerosis: Prospects for clinical trials. Arterioscler Thromb Vasc Biol. 2024;44(9):1899–905. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Miguel V, Shaw IW, Kramann R, Metabolism at the crossroads of inflammation and fibrosis in chronic kidney disease. Nat Rev Nephrol. 2024. [DOI] [PubMed]
- 27.Tezcan G, et al, Resolution of chronic inflammation and cancer. Periodontol 2000. 2024. [DOI] [PubMed]
- 28.Lu N, et al, Prognostic model for predicting recurrence in hepatocellular carcinoma patients with high systemic immune-inflammation index based on machine learning in a multicenter study. Front Immunol. 2024;15:1459740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Walther KA, et al, Inflammation indices in association with periodontitis and cancer. Periodontol 2000. 2024. [DOI] [PMC free article] [PubMed]
- 30.Xia Z, et al, Total bilirubin modified the association between diabetes and stroke: A cross-sectional study from NHANES 2011-2016. J Neurol Neurosurg Psychiatry. 2024. [DOI] [PubMed]
- 31.Bao Y, et al, Renal tubular epithelial cell quality control mechanisms as therapeutic targets in renal fibrosis. J Pharm Anal. 2024;14(8):100933. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Cheung CK, et al, The pathogenesis of IgA nephropathy and implications for treatment. Nat Rev Nephrol. 2024. [DOI] [PMC free article] [PubMed]
- 33.Shen P, et al, Demethylzeylasteral ameliorates podocyte damage in murine lupus by inhibiting inflammation and enhancing autophagy. Phytomedicine. 2024;134:155966. [DOI] [PubMed] [Google Scholar]
- 34.Fu S, et al, Inflammatory indexes and anemia in chronic kidney disease: correlation and survival analysis of the National Health and Nutrition Examination Survey 2005-2018. Ren Fail. 2024;46(2):2399314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Xue Y, et al, Review of clinical characteristics and mortality outcomes in patients on maintenance hemodialysis during the Omicron surge: a single center experience. BMC Public Health. 2024;24(1):1481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Dregoesc MI, et al, Intermediate monocytes are associated with the first major adverse cardiovascular event in patients with stable coronary artery disease. Int J Cardiol. 2024;400:131780. [DOI] [PubMed] [Google Scholar]
- 37.Khalaf K, et al, Cellular aspects of immunity involved in the development of atherosclerosis. Front Immunol. 2025;16:1461535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Yang X, et al, Causal effects of circulating immune cells on coronary atherosclerosis: Evidence from Mendelian randomization. Medicine (Baltimore). 2025;104(5):e41361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Yilmaz M, et al, Evaluation of systemic immune-inflammation index and systemic inflammation response index in the differentiation of acute Ischemic stroke and transient Ischemic attack. Malawi Med J. 2025;36(4):255–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Kim HK, Jun H, Ko SY, Neutrophil-to-lymphocyte ratio as a predictor for early mortality in older patients requiring hemodialysis; insights for hemodialysis access planning. BMC Nephrol. 2025;26(1):2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Yoshida Y, et al, Monocyte/macrophage-specific loss of ARNTL suppresses chronic kidney disease-associated cardiac impairment. Int J Mol Sci. 2024;25(23). [DOI] [PMC free article] [PubMed]
- 42.Zhou Q, et al, Association between monocyte-to-lymphocyte ratio and inflammation in chronic kidney disease: a cross-sectional study. Kidney Blood Press Res. 2024;49(1):1066–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Mureșan AV, et al, The predictive value of NLR, MLR, and PLR in the outcome of end-stage kidney disease patients. Biomedicines. 2022;10(6). [DOI] [PMC free article] [PubMed]
- 44.Qiu J, et al, Accelerated biological aging mediates the association between inflammatory markers with helicobacter pylori infection and mortality. J Transl Med. 2025;23(1):p.174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Zhang Y, et al, Associations of systemic inflammation and systemic immune inflammation with serum uric acid concentration and hyperuricemia risk: The mediating effect of body mass index. Front Endocrinol (Lausanne). 2024;15:1469637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Medina-González R, et al, Decrease in platelet count in patients with AKI and its association with major adverse kidney events. Ren Fail. 2024;46(1):2359643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Zhao X, et al, Platelet count has a u-shaped association with mortality in hemodialysis patients. Sci Rep. 2024;14(1):26572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Bacharaki D, et al, Immune responses of patients on maintenance hemodialysis after infection by SARS-CoV-2: A prospective observational cohort study. BMC Infect Dis. 2023;23(1):581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Losappio V, et al, Molecular mechanisms of premature aging in hemodialysis: The complex interplay between innate and adaptive immune dysfunction. Int J Mol Sci. 2020;21(10). [DOI] [PMC free article] [PubMed]
- 50.Lee SH, et al, STAT3 blockade ameliorates LPS-induced kidney injury through macrophage-driven inflammation. Cell Commun Signal. 2024;22(1):476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Ran S, et al, Association of metabolic signatures of air pollution with MASLD: Observational and mendelian randomization study. J Hepatol. 2024. [DOI] [PubMed]
- 52.Yang JD, et al, Imperatorin ameliorates ferroptotic cell death, inflammation, and renal fibrosis in a unilateral ureteral obstruction mouse model. Phytomedicine. 2024. 135:156066. [DOI] [PubMed] [Google Scholar]
- 53.Corken AL, et al, Platelets, inflammation, and purinergic receptors in chronic kidney disease. Kidney Int. 2024;106(3):392–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Hao XM, et al, Mechanisms of inflammation modulation by different immune cells in hypertensive nephropathy. Front Immunol. 2024;15:1333170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Eirin A, Chade AR, Lerman LO, Kidney intrinsic mechanisms as novel targets in renovascular hypertension. Hypertension. 2024;81(2):206–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Foresto-Neto O, et al, Immunology of kidney disease. Annu Rev Immunol. 2024;42(1):207–33. [DOI] [PubMed] [Google Scholar]
- 57.Kanbay M, et al, A novel risk factor for malignancy: Albuminuria. Eur J Intern Med. 2023;118:22–31. [DOI] [PubMed] [Google Scholar]
- 58.Roccatello D, et al, From inflammation to renal fibrosis: A one-way road in autoimmunity? Autoimmun Rev. 2024;23(4):103466. [DOI] [PubMed] [Google Scholar]
- 59.Bogacka A, Olszewska M, Ciechanowski K, Effects of diet and supplements on parameters of oxidative stress, Inflammation, and antioxidant mechanisms in patients with chronic renal failure undergoing hemodialysis. Int J Mol Sci. 2024;25(20). [DOI] [PMC free article] [PubMed]
- 60.Peris-Fernández M, et al, Metabolic pathways affected in patients undergoing hemodialysis and their relationship with inflammation. Int J Mol Sci. 2024;25(17). [DOI] [PMC free article] [PubMed]
- 61.Bavi Behbahani H, et al, The association malnutrition-inflammation score with chronic kidney disease-associated pruritus and quality of life in hemodialysis patients: A multicenter cross-sectional study. Sci Rep. 2024;14(1):31811. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Huang J, et al, Construction of a c-reactive protein-albumin-lymphocyte index-based prediction model for all-cause mortality in patients on maintenance hemodialysis. Ren Fail. 2025;47(1):2444396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Kalacun V, et al, Oxidative stress and inflammation in hemodialysis: a comparison of patients with or without advanced nonalcoholic fatty liver disease (NAFLD). Ren Fail. 2455523, 2025;47(1):2455523 [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
Data Availability Statement
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.