In recent decades, the surge in diabetes mellitus (DM) cases has propelled it into a critical global health concern, imposing substantial economic burdens worldwide (1). Among its complications, diabetic kidney disease (DKD) looms large, affecting individuals with both type 1 and type 2 DM. Patients with DKD, especially those receiving dialysis, endure a substantial symptom burden and frequent hospital admissions stemming from prevalent comorbidities such as hypertension, coronary artery disease, congestive heart failure, and cerebrovascular disease (2–6). These comorbidities frequently contribute to psychological issues, disabilities, and substantial healthcare costs, markedly impairing patients’ quality of life. Despite current clinical management strategies, which include renin-angiotensin system blockade and meticulous control of hypertension, hyperglycemia, and dyslipidemia, DKD remains a primary contributor to end-stage renal disease (ESRD) necessitating renal replacement therapy (1). The ongoing therapeutic hurdles highlight the pressing necessity for a more profound understanding of DKD’s pathophysiological intricacies, from its onset to advanced renal failure, to identify potential risk factors for screening and intervention. Bridging this knowledge gap is essential for developing novel and effective strategies to prevent and manage the progression of DKD in clinical practice.

Recent investigations have implicated various factors, including metabolic disruptions and hemodynamic irregularities triggered by hyperglycemia and insulin resistance (IR), in the pathogenesis of DKD (7–9). Moreover, IR is closely associated with chronic low-grade inflammation marked by heightened levels of mediators like interleukin-1, interleukin-6, and tumor necrosis factor-α (10, 11). The evolving understanding of DKD portrays it as a disorder driven by metabolic and immunological interplay. Both systemic and localized renal inflammation are recognized as pivotal in DKD progression (12), wherein numerous novel pro-inflammatory signaling pathways have been implicated, including the NOD-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome activation (13), the nuclear factor kappa B (NF-κB) signaling pathway (14), toll-like receptor 4 (TLR4) signaling pathway (15), adenosine 5′-monophosphate-activated protein kinase signaling pathway (16), and the hypoxia-inducible factor-1 signaling pathway (17).

The diabetic microenvironment, marked by hyperglycemia, fluctuating glucose levels, and IR, triggers both systemic and localized inflammatory responses through the TLR4/NF-κB/NLRP3 pathway. These cascades activate platelets, an atypical first-line inflammatory biomarker that may attach to leukocytes and endothelial cells, modifying their pro-inflammatory activities (18–20). For instance, platelet activating factor (PAF), a pro-inflammatory mediator significantly increased by activated platelets, exaggerates leukocyte chemotaxis, complement activation, reactive oxygen species and eicosanoids production in DKD (21–23). PAF also stimulates lymphocytes to produce immunoglobulins and elevates circulating pro-inflammatory cytokines such as interleukin-1, interleukin-6, and tumor necrosis factor-α (24). These cytokines not only amplify platelet activation but also co-stimulate lymphocytes, further exacerbating renal inflammation (25–27). Platelet-derived platelet factor 4 (PF4), a potent chemoattractant produced by platelet that promotes neutrophil adhesion to endothelial cells and lymphocyte chemotaxis (21, 28), has been found to be markedly elevated in DKD patients with macroalbuminuria (23). Neutrophils, which comprise the majority of white blood cells and are critical in initiating and regulating inflammatory processes, release neutrophil elastase-a key player in chronic inflammation and potential contributor to renal damage in DKD (29). Lymphocytes are inflammatory mediators that do have regulatory or protective functions for preventing the progression of chronic kidney disease (30). More importantly, DKD patients exhibit significantly higher neutrophils and platelet counts alongside notably lower lymphocyte counts compared to healthy populations (29–33), indicating heightened inflammation and an imbalance in immune regulation. Taken together, the diabetic milieu orchestrates a broad variety of inflammatory responses, including secretion of pro-inflammatory cytokines, platelet-lymphocyte interaction, and platelet-neutrophil interaction, all of which synergistically contribute to the deterioration of renal function. These findings collectively underscore the critical role of inflammation and immune cells interaction in driving DKD progression.

The systemic immune-inflammation index (SII), an innovative inflammatory marker derived from platelet count × neutrophil count/lymphocyte count, has emerged as a comprehensive measure of inflammation. Initially utilized to assess prognosis in hepatocellular carcinoma patients (34), SII has shown prognostic utility in various cancers and is recognized for its precision in gauging inflammatory status. Recent studies have further linked higher SII with increased incidence of metabolic syndrome (35), cardiovascular disease (36), nonalcoholic fatty liver disease (37), DM (38), urinary albumin excretion (39), diabetic retinopathy (40), and other DM-related complications (41, 42). Moreover, prospective cohort studies have associated SII with elevated risks of cardiovascular, cardio-cerebrovascular, and all-cause mortality in DM individuals (43) and the general population (44). However, the relationship between SII and mortality outcomes in individuals with DKD remains unexplored. Examining the relationship between SII and long-term mortality risk in individuals with DKD is crucial as it provides a deeper understanding of the impact of inflammation and immune status on the health outcomes of DKD patients. Furthermore, such research provides valuable insights for improving clinical management and intervention approaches for DKD patients. To bridge this gap, the present study investigates the association between SII, a novel index reflecting systemic inflammatory state, and all-cause mortality and cardio-cerebrovascular disease mortality in individuals with DKD, using data from the National Health and Nutrition Examination Survey (NHANES) database.

In this study, we for the first time conducted a novel investigation into the relationship between the SII and both all-cause mortality and cardio-cerebrovascular disease mortality in individuals diagnosed with DKD using data from the NHANES database. Our analysis unveiled notable findings: a U-shaped correlation between SII levels and the risk of all-cause mortality in DKD individuals was observed, and the inflection point of lnSII with the lowest HR was 5.82, indicating that both excessively low and high concentrations were associated with an increased risk. This relationship remained consistent across various stratified analyses. Additionally, we found a linear correlation between SII levels and the risk of cardio-cerebrovascular disease mortality. Taken together, these findings suggest that SII could serve as a predictive marker for mortality risk in DKD patients and may represent a potential target for interventions aimed at improving health outcomes.

Emerging evidence suggests that chronic immune cell overactivation and subsequent low-grade inflammation may underlie the pathogenesis of DKD and its dire prognosis (15, 54–56). A comprehensive analysis of multiple genome-wide association studies revealed that a significant portion of single nucleotide polymorphisms associated with DKD are directly or indirectly linked to inflammation and immunity (57, 58). Numerous clinical and epidemiological investigations have consistently shown elevated levels of plasma inflammatory markers (59), such as C-reactive protein (60), high-sensitivity C-reactive protein (61), and interleukin-6 (62) in DKD patients. Although large-scale clinical trials specifically targeting therapies for DKD are lacking, the Finerenone in Reducing Kidney Failure and Disease Progression in Diabetic Kidney Disease (FIDELIO-DKD) trial has indicated that DKD patients derive greater benefit from anti-inflammatory treatment with Finerenone, a first novel, highly potent, selective mineralocorticoid receptor antagonist (63). This previous evidence suggests a potential synergistic relationship between inflammation levels and progression of DKD.

The SII is derived from the counts of three circulating immune cell types: neutrophils, lymphocytes, and platelets. It provides a comprehensive assessment of immune and inflammatory status, offering more clinical insights than single or dual peripheral blood parameters. Elevated SII levels often coincide with thrombocytosis, neutrophilia, or lymphopenia, reflecting heightened inflammatory responses and serving as a valuable diagnostic biomarker for systemic inflammatory activity (44). Particularly, numerous studies have underscored SII’s predictive capacity, linking higher SII levels to increased risks of various renal disease subtypes, including contrast-induced nephropathy (64), renal cell carcinoma (65), and peritoneal dialysis-treated chronic kidney disease patients (66). Moreover, several previous studies have shown the close relationship between SII and the incidence and severity of DKD, suggesting SII as a widely available, non-invasive, cost-effective, and straightforward approach to detecting and monitoring DKD (54, 55, 67). Accordingly, the therapeutic potential of anti-inflammation-based regimen is promising, emphasizing the importance of addressing systemic inflammation for better mortality risk prevention and prediction of DKD. Nevertheless, the relation between SII and clinical outcomes in DKD individuals remains largely undetermined.

In our current study, we for the first time identified a positive association between SII and increased risk of all-cause mortality, coupled with a linear association between SII and increased risk of cardio-cerebrovascular disease mortality, which underscores the value of SII in identifying high-risk individuals in DKD populations, thereby enabling early intervention. It is noteworthy that we observed a U-shaped relation between SII and all-cause mortality. Specifically, below the threshold value of 5.82 for lnSII, higher SII was significantly associated with lower all-cause mortality, while above the thresholds, SII was positively associated with all-cause mortality. Consistent with our present results, a previous study by Chen et al. revealed a U-shaped correlation between SII and all-cause mortality in populations with DM (43). Besides, Yan et al. observed that higher SII was closely associated with an increased risk of the presence and severity of DKD in Chinese population (67). These pieces of evidence, in conjunction with our findings, support the notion that SII holds promise as a potential biomarker for DKD. Importantly, both low and high SII levels might elevate mortality risk in individuals with DKD or DM, which aligns with the understanding that low platelet levels are typically associated with a heightened risk of bleeding that could contribute to higher all-cause mortality. While our study contributes to understanding the relationship between SII and mortality, the precise mechanisms underlying this association remain enigmatic and warrant further investigation.

Our study revealed an intriguing interaction between the SII and BMI in our subgroup analysis, highlighting the intricate relationship between inflammation and obesity in the development of DKD (35). This finding supports previous research indicating a complex interplay between SII and BMI, with BMI serving as a critical mediator in the association between SII and the risk of DM (68). Additionally, Kong et al.’s study identified SII as an independent risk factor for both all-cause and CVD-specific mortality in obese populations (53). Furthermore, numerous preclinical studies have underscored the pivotal role of dysregulated inflammatory responses in obesity-related pathogenesis (69). Intriguingly, the paradoxical association was observed when we seek to investigate the independent impact of BMI on the prognosis among the DKD cohorts. As illustrated in Supplementary Figure 1A , the association between BMI and all-cause mortality in DKD cohorts exhibits a U-shaped pattern. Specifically, the relationship between BMI levels on a continuous scale and the risk of all-cause mortality in the DKD cohort is U-shaped (P for nonlinearity = 0.003); both low and high BMI levels were linked to an increased risk of all-cause mortality ( Supplementary Figure 1A ). The BMI level associated with the lowest risk of incident all-cause mortality was 35.04 in the fully adjusted analyses. In contrast, as shown in Supplementary Figure 1B , BMI appears to have no correlation with cardio-cerebrovascular mortality. The obesity paradox observed in our study among DKD subjects aligns with numerous previous studies that demonstrate a significant association between BMI and all-cause mortality risk in diabetes cohorts (70, 71), chronic kidney disease cohorts (72–74), and the general population (75). However, elucidating the precise mechanisms underlying this interplay between inflammation and obesity in exacerbating DKD pathogenesis warrants further investigation.

Our study presents several strengths. Firstly, we analyzed a substantial sample size of 3195 individuals, ensuring robust representation of the population. Secondly, meticulous attention was paid to controlling for confounding variables, enhancing the reliability of our findings. Thirdly, our investigation is pioneering in its exploration of both linear and nonlinear relationships between the SII and mortality in DKD populations, evolving methodologically over time. Lastly, SII serves as an easily accessible, cost-effective measure with potential therapeutic implications or as an early warning indicator.

However, several limitations warrant consideration. Firstly, the observational nature of our study precludes establishing causal relationships. Secondly, despite efforts to adjust for various confounding factors, the influence of unmeasured variables such as diabetes duration, dietary habits, and treatment modalities remains a concern. Thirdly, while SII offers ease of measurement, factors affecting neutrophil, lymphocyte, and platelet counts could introduce selection bias. Lastly, our reliance on data from a single blood test may not fully capture temporal fluctuations in SII levels due to the short lifespan of blood cells. Continuous monitoring could provide more robust evidence than a one-time assessment.

In conclusion, our study revealed a U-shaped relationship between the SII and all-cause mortality, with threshold values of 5.82 for lnSII. Additionally, higher concentrations of SII exhibited a linear association with increased risk of cardio-cerebrovascular disease mortality. These findings underscore the independent prognostic significance of SII for patients with DKD. However, further extensive prospective investigations are warranted to validate and consolidate our findings.