Cardiovascular disease (CVD) is the leading cause of non-communicable diseases (NCDs) and is the largest single cause of death worldwide (1, 2). Approximately 18 million people die from CVD each year, accounting for half of all NCD deaths (3). According to statistics, the annual cost of CVD in the United States is US$378 billion (4). As a result, the United Nations has set a number of Sustainable Development Goals, including reducing premature deaths from NCDs by one-third by 2030 (5). To reduce the burden and damage caused by CVD, it is important to explore the correlation between various indicators and CVD risk to help identify people at high risk of CVD who need early intervention.

Stroke and ischemic heart disease are two major diseases that contribute to the burden of CVD and cause of disability-adjusted life years (6), and their major cause is atherosclerosis (7). Atherosclerosis is a systemic chronic inflammatory vascular disease, and thrombosis, oxidative stress, and endothelial damage may be the potential pathogenic mechanisms (8, 9). Studies have shown a profound link between immune and inflammatory responses in the development of atherosclerosis (8–10). The white blood cells that make up the immune system, including lymphocytes, neutrophils, monocytes, and macrophages, play different roles in atherosclerosis. For example, neutrophils can accelerate all stages of atherosclerosis by activating macrophages, promoting monocyte recruitment, and cytotoxicity, while lymphocytes regulate the inflammatory response and therefore have an anti-atherosclerotic effect (11, 12). There is also evidence that platelets adhere to the vessel wall to enhance leukocyte aggregation and initiate atherosclerotic progression before leukocytes invade the atherosclerotic plaque (13, 14). In recent years, a growing number of studies have explored some low-cost, validated, reproducible indicators of inflammation and immunity, such as the neutrophil-lymphocyte ratio (NLR) and platelet-lymphocyte ratio (PLR), with results suggesting that these combined markers are better predictors of CVD prognosis than single-cell counts (15–19). A new index, the systemic immune-inflammation index (SII), calculated as (neutrophil × platelet)/lymphocyte has recently been introduced for CVD, which can provide a relatively comprehensive reflection of the balance between host’s inflammation and immunity status (20); SII can be easily obtained by the routine blood test for complete blood count, the most commonly performed test in clinical practice. This marker was first proposed by Hu et al. (20) and has shown a higher predictive value for prognosis in the cancer field than other inflammatory factors, such as NLR and PLR (21, 22). Interestingly, recent studies have suggested that SII may be associated with prognosis and mortality in CVD (23).

Considering the importance of CVD risk identification and early intervention, some studies have explored the association between SII and risk of CVD, but no consistent conclusions have been reached. Therefore, we conducted a meta-analysis and systematic review to comprehensively examine the value of SII as an inexpensive and accessible indicator for predicting the risk of different CVD.

This study was reported based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (24). The protocol of the study is not registered.

To the best of our knowledge, this is the first study to focus on examining the association between SII and CVD risk. SII is a new inflammatory index based on platelet, neutrophil, and lymphocyte counts that reflects the host’s inflammatory and immune status. We found a significantly higher risk of CVD in people with high SII levels; this risk can be observed in almost every type of CVD. SII values were also significantly increased at the onset of various CVD. These findings suggest that SII may serve as an indicator to identify people at high risk of CVD and contribute to early diagnosis. However, due to the nature of observational studies, the quality of the current evidence is rated as low or very low.

Cardiovascular disease encompasses a wide range of diseases with diverse and non-specific clinical presentations, making it easy to delay diagnosis and misdiagnosis, which leads to deterioration of the patient’s condition. Therefore, exploring a simple and widely available index to predict CVD is crucial. A large body of evidence suggests that chronic inflammation plays a key role in the pathogenesis of CVD (41–43). Specifically, neutrophils secrete inflammatory mediators that can lead to endothelial dysfunction and vessel wall degeneration (44). Platelets may release some chemokines, proinflammatory cytokines, and platelet-derived growth factors that promote endothelial cell depletion (45, 46). In addition, neutrophils promote atherosclerosis and thrombosis through interactions with platelets, proteolytic cleavage of coagulation factors, release of prothrombotic molecules, and promotion of monocyte infiltration, ultimately leading to cardiovascular events (47). Conversely, lymphocytes can regulate inflammatory responses and have anti-atherosclerotic effects. However, due to the diversity of T cell types, their roles are also diverse (48). For example, CD8+, CD4+, γδ T cells exacerbate stroke injury by secreting interferon-γ and IL-17, while T regulatory cells (CD4+, CD25+, Foxp3+, and T regulatory cell) release the anti-inflammatory cytokine IL-10, which is beneficial to the condition (49–51). Therefore, the detection of lymphocyte subpopulations may further enhance the strength of the association.

The SII has unique advantages over other biological indicators, such as NLR, PLR, and C-reactive protein (CRP). Liu et al. (32) found that SII had better predictive power for the occurrence of coronary artery disease compared with NLR, PLR, and C-reactive protein (CRP). Xu et al. (29) observed a significant linear dose-response association between SII and risk of total stroke. SII appears to be more stable than blood counts alone, and these counts are susceptible to various factors, such as dehydration and fluid overload (52, 53). Of note, among these diseases, SII was correlated most significantly with cerebral venous thrombosis, and the study by Zhang et al. (28) found that SII could be used as a strong independent predictor in multivariable regression analysis (SII degree, OR = 13.136, 95%CI 5.675–30.407, P < 0.001). This means that in patients with unexplained headache and a normal plain CT, SII can help clinicians to quickly infer cerebral venous thrombosis and determine its severity and the necessity for immediate MRI/MRV confirmation, which can help to shorten the time between symptom onset and diagnosis and reduce misdiagnosis (28). Furthermore, Zhang et al. (40) found that preoperative SII can help predict the risk of perioperative ischemic stroke in non-cardiac surgery patients, suggesting that physicians can formulate more rational perioperative prevention and treatment plans based on SII levels. In the chronic disease population, Xu et al. (29) found a stronger positive association between SII and total CVD in patients with hypertension or diabetes. This may be related to endothelial dysfunction, inflammatory infiltration, and vascular remodeling caused by long-term inflammation in patients with diabetes or hypertension (54).

In addition to the correlation with CVD risk, SII has been found to predict the severity, complications, and mortality of CVD. For example, Zhang et al. (28) found that SII was positively correlated with the baseline National Institutes of Health Stroke Score (NIHSS); Weng et al. (34) demonstrated that SII was associated with stroke severity and that SII level was an independent prognostic indicator of poor prognosis at 3 months. Interestingly, Liu et al. (32) found by correlation analysis that SII was a better predictor of coronary artery disease severity than NLR, PLR, and CRP levels.

The SII reflects a significant association of the host’s inflammatory and immune status with the risk of CVD. People with elevated SII on physical examinations may be recommended to adopt a lifestyle that can reduce inflammation, such as smoking cessation, weight control, reasonable emotional regulation, reduced sedentary behavior, regular exercise and rest, and reduction of saturated fat and refined carbohydrate content in the diet (55–57). Moreover, reminding the general population with high SII levels to increase the frequency of medical screening as appropriate may help patients benefit from secondary prevention treatment (58).

The main limitation of this study is the small number of studies in some CVD subtypes resulting in conservative 95%CI. For example, for venous thrombosis, although a strong correlation was found, it was not significant, which may be due to insufficient statistical power. Second, populations outside of China are relatively underrepresented, and more studies from other geodemographic settings are necessary to examine possible regional specificities. Third, although Egger’s and Begg’s tests did not show significance, small study effects may still be present; studies with smaller sample sizes seemed to tend to report stronger correlations. Fourth, the limited data did not support a more detailed stratified analysis to identify populations that would benefit more from SII indicators. Furthermore, current studies on the association between SII and CVD risk used different SII cutoff values, which need to be standardized before SII can be widely used.

In conclusion, our study indicated that SII may be a potential predictor of CVD. Healthcare providers need to be concerned about CVD risk in people with high SII and appropriate interventions may be needed. Due to the nature of observational studies, the quality of the evidence was graded from low to very low, which means that further research on the subject is necessary to draw solid conclusions. In addition, future studies should further determine the optimal cutoff value, characterize the benefited population, and investigate whether SII can be used in conjunction with other known CVD risk factors to build an accurate CVD risk assessment system.

The original contributions presented in this study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.

ZS and ML: conception, design, and critical revision of the manuscript. ZY, XL, JW, and RX: database search, literature review, and data extraction. XL and TH: statistical analysis. ZY, TH, XL, JW, and RX: drafting of the manuscript. All authors contributed to the article and approved the submitted version.