Type 2 diabetes (T2D) poses a significant global public health challenge, profoundly affecting human health and quality of life (1). Prediabetes and insulin resistance (IR) stand out as primary contributors to T2D development, characterized by aberrant glucose metabolism (2). IR is a pivotal factor in various metabolic disorders, signifying a state where insulin-responsive tissues exhibit reduced responsiveness to physiological insulin levels. This state results in hyperinsulinemia and elevated fasting blood glucose (FBG), diagnostic indicators of IR (3). Despite being the gold standard for assessing IR, the hyperinsulinemic-euglycemic clamp is a costly, intrusive, and time-consuming procedure that needs to be performed by skilled staff. Therefore, the homeostasis model assessment of insulin resistance (HOMA-IR) offers a more straightforward and useful option by measuring insulin and fasting blood glucose (4).

IR is not only a key pathogenic factor in T2D but is also associated with various pathological conditions such as cardiovascular diseases, certain types of cancer, infertility, polycystic ovary syndrome, non-alcoholic fatty liver disease, and metabolic syndrome (5). Given the substantial harm that IR and prediabetes inflicts on human health, the early detection and intervention of them are hot topics among scholars in the relevant field. Previous research indicates that systemic chronic inflammation plays a pivotal role in IR and prediabetes, with obesity frequently triggering this inflammatory state (6). Obesity can induce a chronic inflammatory state in various insulin-target tissues, including adipose tissue, liver, muscles, and the pancreas. This is often attributed to potential interactions between immune processes and metabolic defects (7). Metabolic tissues induce the occurrence of this chronic low-grade inflammation by recruiting, accumulating, and activating pro-inflammatory macrophages. Although macrophages play a central role, other immune cell types are involved in these inflammatory processes (8). Satoshi discovered that CD8+ T cells can activate macrophages within adipose tissue. This alteration of the immune microenvironment leads to the shift of adipose tissue from an anti-inflammatory state to a pro-inflammatory state (9). Hence, controlling inflammation seems to be a pivotal intervention for mitigating IR and prediabetes. Nevertheless, studies on the population-level association between inflammation and IR and prediabetes are still limited. A comprehensive exploration of this potential connection necessitates an urgently required objective assessment indicator that precisely mirrors the immune and inflammatory status of populations with IR and prediabetes.

The systemic immune-inflammation index (SII), developed by Hu et al., is a novel, comprehensive biomarker for immune inflammation based on blood cells (10). It integrates three types of inflammatory cells-platelets, neutrophils, and lymphocytes. It accurately reflects the local immune response and systemic inflammatory status of body (11). Multiple studies have substantiated the prognostic value of SII in assessing outcomes for various cancer patients, encompassing bladder cancer (12), cervical cancer (11), non-small cell lung cancer (13), colorectal cancer (14), and gastric cancer (15). In recent years, SII has emerged as an indicator for detecting chronic inflammatory disease beyond tumors. Yang et al. reported that in patients with coronary artery disease after coronary intervention, SII demonstrates superior predictive value for major cardiovascular events compared to traditional risk factors (16). Wang et al. suggested using SII as an indicator for detecting diabetes depression (17). Additionally, several studies utilizing the NHANES database reveal a robust association between SII and various metabolic diseases, such as diabetic nephropathy (18), hepatic steatosis (19), and osteoporosis (20). Moreover, numerous studies indicate that, compared to traditional immune-inflammatory indicators [including lymphocyte/monocyte ratio (LMR), platelet/lymphocyte ratio (NLR)], SII is a more accurate predictor of malignant tumors (21, 22).

In summary, SII is a non-invasive quantitative indicator with higher research value compared to traditional inflammatory markers. Given the relationship between inflammation and IR as well as prediabetes, we hypothesize that a high SII level may be positively associated with the risk of developing IR and prediabetes. However, research in this field is currently limited. Therefore, this study plans to employ the NHANES database for a more rigorous statistical analysis methods, controlling confounding variables to validate this hypothesis. Our goal is to identify individuals at high risk of IR through SII and explore the association between SII and markers of T2D risk. This will help investigate its potential to identify prediabetic patients.

To our knowledge, this is the initial investigation assessing the association between SII and prediabetes in the adults of the United States. A recent study have reported a positive linear associations between higher SII and increased risk of IR, which is consistent with our findings. However, their study did not adequately exclude patients with existing diabetes or those taking medications that could affect IR, and our research included a higher number of study populations (23). Therefore, our research results are more reliable and more complete.

In this cross-sectional study involving American adults, we observed a significant correlation between elevated SII levels and risk markers for T2D (FBG, FSI, and HOMA-IR). Furthermore, after further adjusting for potential confounding factors, the association between SII and increased risk of IR and prediabetes persisted. However, the positive correlation between SII and IR or prediabetes is more significant in females than in males. Additionally, the restricted cubic spline model indicates a linear dose-response relationship between SII and the odds of IR and prediabetes. These findings suggest that SII may serve as a monitoring indicator for IR and prediabetes.

Zhao et al. illustrated that continuous SII exhibit a skewed distribution, aligning with the structure of our original data. To approximate it to a normal distribution, they recommended logarithmic transformation of SII (24). From Table 1 , it can be seen that some traditional diabetes risk factors, such as older age, smoking, and higher BMI, are more likely to have higher SII values, while protective factors like MVPA are more likely to have lower SII values. Additionally, our study revealed a positive correlation between continuous SII and glycated hemoglobin. However, this correlation vanished after adjusting for confounding factors and categorizing SII into tertiles.

SII is a recognized indicator for predicting cancer treatment outcomes and prognosis. Apart from cancer, the predictive value of SII for other metabolic-related diseases, such as diabetes and cardiovascular diseases, is also gaining attention (25, 26). The study by Nie et al. revealed an association between increased SII and increased prevalence of diabetes (27). Bian et al. found that CAD patients undergoing PCI with worse prognostic outcomes tended to have higher SII values (28). In a cross-sectional study conducted in the United States with 12,402 participants, a significant correlation between SII and metabolic syndrome was found after inclusion (24). Metabolic syndrome is a condition characterized by an aggregation of various metabolic risk factors related to IR and impaired glucose regulation (29). Inspired by this discovery, we focus on whether SII holds equal value in identifying and predicting IR and prediabetes. There are few studies that have independently evaluated the association between SII and IR as well as prediabetes. Numerous studies have indicated a significant association between SII and complications of diabetes.

In the work by Elbeyli, the SII was identified as a potential diagnostic biomarker for diabetic macular edema, with positive implications for improving diabetic retinopathy (30). Özata et al. further explored the correlation between SII and diabetic macular edema, revealing that an elevated SII level might lead to an increased incidence of serous retinal detachment (31). The research of Safak et al. unveiled the potential of SII as a predictive indicator for diabetic foot osteomyelitis (32). Moreover, in a survey of Indonesian diabetic patients, Yohanes and Andy found a significant association between low SII levels and the regulation of psychological well-being in diabetic patients (33).

In clinical studies, both IR and prediabetes have been found to be associated with several traditional inflammatory markers. Jia et al. used the rate nephelometry method to measure serum IMA and hs-CRP concentrations in patients with diabetic retinopathy, finding a positive correlation between hs-CRP concentration and the incidence of diabetic retinopathy (34). Liu et al. suggested that serum hs-CRP concentration can predict the incidence of diabetes (35). However, some studies have found no difference in the presence or absence of hs-CRP with IR. Systemic levels of TNF-a, IL-1b, IL-6, and CRP are elevated in both type 1 and type 2 diabetes patients (36, 37), which is a result of the chronic activation of pro-inflammatory pathways within insulin-target cells (38). Consequently, these cytokines and inflammatory mediators, especially TNF-α, monocyte chemoattractant protein-1 (MCP-1), CRP, and interleukins, are considered potential contributors to IR or impaired B-cell function (39). Additionally, in the study by Shu et al., it was found that the dietary inflammatory index is positively correlated with FBG, FSI, and HOMA-IR, and a more pro-inflammatory diet is associated with increased odds of IR and prediabetes (40).

In recent years, an increasing number of studies have focused on the significance of common inflammatory markers in blood routine examinations in the diagnosis and treatment of metabolic diseases. Additionally, Christine Lee and colleagues investigated the relationship between various white blood cell subtypes and IR in high-risk individuals, finding a positive correlation with all white blood cell subtypes, including granulocytes, lymphocytes, and monocytes (41). Karakaya studied 96 obese patients and 40 healthy controls, discovering a positive correlation between IR and white blood cell count, with NLR higher in obese IR patients than non-IR obese patients (42). Rodríguez-Rodríguez et al. also found a similar phenomenon in children (43). Some studies suggest that participants resistant to insulin have significantly higher platelet counts than insulin-sensitive participants (44). Hwang et al. found that with an increase in platelet count, the incidence of diabetes also increased, indicating that platelets are a potential risk marker (11). However, in the study by Rodríguez-Rodríguez, it was found that high platelet values do not constitute a risk factor for the occurrence of IR in children, and no relationship was observed between IR and PLR (43). In related reports, it seems that NLR is better at predicting inflammation than PLR, as neutrophils play a dominant role in inflammation by releasing vasoactive and cytotoxic substances (such as reactive oxygen species and digestive enzymes) during inflammation, leading to increased endothelial permeability (45). Therefore, it is necessary to introduce a more accurate SII index that includes platelet count for further research.

In various studies on inflammatory markers, SII, with its advantage of integrating three key immune cells, provides a more comprehensive description of the body’s inflammatory state compared to traditional single inflammatory biomarkers. For instance, Berbudi et al. revealed that SII, in predicting the impact of T2D on the immune system, demonstrates more precise and effective predictive capability than NLR, PLR, and MLR, as significantly confirmed by ROC curve analysis (46). Furthermore, the study by Nicoară’s team confirms that, in differentiating whether obese children have metabolic syndrome, SII exhibits higher diagnostic efficacy compared to NLR, PLR, and SIRI. It also shows a positive correlation with the HOMA-IR (47). These findings offer a new perspective for clinically assessing inflammatory and metabolic abnormal states.

The primary strength of this study is the inclusion of a large number of samples. Based on NHANES database, we analyzed a total of 9250 samples from 2005 to 2018, including self-reports, laboratory tests, and physical examinations. These data were collected by professionals through a standardized procedure, significantly reducing errors caused by different methods. The enormous sample size and standardized data contribute to obtaining meaningful and highly reliable results even after multiple condition screenings. And this study not only screened diabetic patients based on fasting blood glucose level >7.0 mmol/L but also excluded potential diabetic patients based on medication use and HbA1c levels. Additionally, to explore the impact of confounding factors on the association between SII and IR and prediabetes, we conducted stratified analyses. The results revealed that factors such as age, gender, race, education level, and family income-poverty ratio had some impact on the incidence of IR and prediabetes. For example, older patients have higher SII values and are at higher risk for IR and prediabetes. And our results show that women are more likely to have higher SII values than men, suggesting that they are more likely to have IR as well as pre-diabetes. Furthermore, lifestyle factors such as smoking, more alcohol consumption, and lower physical activity level were also positively associated with the risk of developing IR and prediabetes. Finally, we used non-restrictive cubic spline plots to analyze the nonlinear relationship between SII and IR and prediabetes, providing a more comprehensive understanding of their development and risk factors.

However, this study has certain limitations. Firstly, the nature of observational research restricts our analysis of causal relationships, leaving room for multiple interpretations, including both causation and reverse causation. Therefore, prospective studies are urgently needed to clarify the precise connections among these factors. Additionally, despite adjusting for various covariates, potential confounding factors such as dietary patterns and a family history of T2D might still be overlooked. It’s worth noting that, although diet has a significant impact on circulating TG levels, fasting blood samples collected in this study may not fully capture this aspect. Future research should delve into exploring the specific influence of dietary factors on study results.

Despite some limitations and drawbacks, our study still holds significant clinical relevance. As a novel inflammatory biomarker, SII not only offers the advantage of non-invasiveness but also provides a more comprehensive approach to assessing immune and inflammatory responses. The results of this study confirm our previous hypothesis that SII can serve as crucial indicators for diagnosing IR and prediabetes. In an era where IR is prevalent, often elusive, and troubling to primary care communities, the SII scoring system, comprising three simple and efficient hematological indicators, provides a practical diagnostic tool for primary healthcare practitioners. In the future, we plan to initiate a multicenter prospective cohort study to explore the effectiveness of SII as an independent predictor of IR and prediabetes. The goal is to offer prospective guidance and intervention for high-risk individuals through routine blood cell count monitoring, aiming for prevention and protection.

In conclusion, this study indicates a positive correlation between SII and FBG, FSI, and HOMA-IR, so higher SII levels may increase the odds of IR and prediabetes. Therefore, SII is poised to be a direct and cost-effective method for identifying patients with IR and prediabetes.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.cdc.gov/nchs/nhanes/.

The studies involving humans were approved by National Center for Health Statistics. The studies were conducted in accordance with the local legislation and institutional requirements. The human samples used in this study were acquired from gifted from another research group. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

HG: Writing – original draft, Validation, Methodology, Data curation. CW: Writing – original draft, Methodology. JZ: Writing – original draft. XJ: Writing – review & editing, Project administration. SL: Writing – review & editing, Supervision, Conceptualization.