Our study followed the guidelines in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) statement (12). The research protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO: CRD42024582003). A comprehensive search strategy was meticulously devised by two investigators, YJJ and ZT, who independently used specific terms and keywords to retrieve relevant literature from PubMed, Embase, Web of Science, and the Cochrane Library from their inception until July 9, 2024. The terms included, but were not limited to, “HF”, “systemic immune-inflammation index”, and “SII”. The search strategy is detailed in Supplementary Table S1.

Eligible studies met the following criteria: (1) patients diagnosed with HF; (2) SII directly or indirectly obtained and used as a primary analytical indicator; (3) investigation of the association between SII levels and HF incidence or prognosis; (4) RR or standardized mean difference (SMD) with 95% CI provided or calculable from available data; (5) fully published studies.

The exclusion criteria were: (1) Reviews, commentaries, conference abstracts, case reports, and letters; (2) Studies lacking sufficient data to compute RR or SMD with 95% CI; (3) Abstracts or full texts that were inaccessible; (4) Non-English language publications; (5) Studies not reporting data on the relation of SII to HF diagnosis or survival outcomes; (6) Studies containing duplicate or overlapping datasets.

Two researchers (YJJ and ZT) independently screened titles and abstracts, followed by full-text assessments to determine study eligibility. Discrepancies were resolved by consensus.

Data were extracted independently by two researchers, YJJ and ZT. Disagreements were settled via consensus among co-authors. Extracted information encompassed the first author, publication year, study period, region, study design, population, diagnostic criteria, number of patients, gender, SII cut-off, mean/median age, mean/median body mass index (BMI), mean/median EF%, and RR or SMD (95% CIs) for the incidence and the mortality of HF.

The Newcastle-Ottawa Scale (NOS) (13) was employed to assess eligible studies in selection, comparability, and outcomes. The maximum score was nine (10), and studies scoring from 7 to 9 were deemed high-quality ones.

RR and 95% CI were utilized to pool data for categorical variables, while SMD and 95% CI were employed for continuous variables, in order to evaluate the incidence and prognostic significance of SII in HF cohorts. Cochran's Q test and Higgins’ I² statistic were used to assess heterogeneity (14). A random-effects model was subsequently applied for data analysis. To ensure the robustness of the findings regarding the incidence and mortality of HF, subgroup and sensitivity analyses were performed. Funnel plots and Egger's tests were employed to detect potential publication bias for outcomes with more than 10 studies. A p-value less than 0.05 was considered statistically significant. In results showing publication bias, the Trim-and-Fill method was applied for adjustment. All statistical analyses were conducted using STATA version 15.0 and Review Manager version 5.4. Additionally, based on the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach, the quality of evidence for each outcome was systematically evaluated and categorized as “high”, “moderate”, “low”, or “very low” (15).

328 articles were initially retrieved. Among the initially identified records, 34 were excluded due to duplicate publications. Following a screening of titles and abstracts, an additional 277 studies were removed. The full texts of the remaining 17 studies were then reviewed in detail. Of these, two were excluded due to insufficient data for survival analysis. Ultimately, fifteen studies (7–11, 16–25) comprising a total of 77,917 patients were included in the present meta-analysis (Figure 1).

Of the fifteen eligible articles (7–11, 16–25), five (19, 20, 22, 23, 25) were conducted in Turkey, six in the United States, and the remaining four in China. Notably, among the included publications, two articles encompassed comparison groups each, two encompassed four comparison groups each, and one encompassed three comparison groups, resulting in a total of 35 comparison groups. Of these, fourteen were case-control studies and nineteen were cohort studies. All studies were retrospective in design, published in English, and appeared between 2020 and 2024. The characteristics of the included studies are summarized in Supplementary Table S2. Two studies reported the association between SII and the incidence of HF using continuous variables. The mean SII value among patients with HF was 646.10, in comparison to 537.98 in those without HF. Additionally, four studies reported the association between SII and mortality in HF patients using continuous variables. The mean SII value in deceased HF patients was 2390.89, whereas it was 1,668.40 in survivors.

The fifteen studies received scores of 7 and 8, which signify high quality. The most common point of deduction was in the category “comparability based on the most important and additional risk factors”. Among all studies, one was rated as 7, while seven had a score of 8 (Supplementary Tables S3, S4).

The robustness of the findings regarding the clinical significance of SII was evaluated by conducting a sensitivity analysis. The results indicated that the effect sizes remained stable within the original confidence range following the sequential exclusion of individual comparison groups. This consistency suggests that no single study exerted a disproportionate influence on the pooled estimates for the incidence of HF (Figure 3A), dichotomous mortality (Figure 3B), or continuous mortality (Figure 3C), thereby supporting the reliability and stability of the overall results.

Publication bias was assessed via funnel plots and Egger's test. In contrast, Egger's test revealed significant publication bias in the meta-analysis of dichotomous mortality outcomes (p = 0.0001) (Figure 4). Due to the limited number of included comparison groups (fewer than ten), assessment of publication bias could not be performed for the remaining outcomes. Publication bias for the dichotomous mortality outcome was further evaluated via the Trim-and-Fill method. The analysis indicated that statistical significance remained unchanged after adjustment (OR = 1.205; 95% CI: 1.062–1.368) (Figure 5), suggesting the association between SII and mortality is robust and reliable.

When graded using the GRADE framework, outcomes including mortality (dichotomous and continuous) and incidence rate (dichotomous and continuous) were all rated as very low quality of evidence (Table 2).

This meta-analysis, encompassing 77,917 participants, evaluated the diagnostic and prognostic utility of SII in HF. The evidence indicates that SII does not predict the incidence of HF; however, it possesses significant prognostic value for survival. Higher SII values were positively correlated with mortality, signifying that SII may identify patients at elevated risk of death.

In our study, no clear association between SII values and the incidence of HF as a dichotomous variable was detected. However, a significant correlation was observed between SII and the continuous incidence of HF. Among the studies reviewed, five reported a relationship between SII and HF incidence, with one indicating no significant association (8), which aligns with our findings. However, findings from four studies and continuous incidence data suggest that SII is associated with the onset of HF and may serve as an effective biomarker for its prediction. Nevertheless, given that HF is influenced by genetic, environmental, and lifestyle factors, and considering that the included studies were clinical with relatively small sample sizes (<1,000) and numerous confounding variables, the SII values reported reflect only the inflammatory and immune status at a specific time point (e.g., baseline), rather than capturing the dynamic progression of HF. Moreover, the onset of HF is often affected by multiple comorbid conditions. Our integrated analysis of multiple studies did not reveal a statistically significant association between SII and HF onset. Therefore, SII cannot currently be considered a predictive marker for the occurrence of HF. Future studies should incorporate serial measurements of SII across different stages, establish standardized cutoff values, and further explore their associations with HF as well as the modifying effects of comorbidities, in order to verify the correlation between SII and HF. Regarding prognosis, our findings support previous conclusions that SII is a reliable predictor of mortality risk in HF. In addition, elevated SII levels in patients with CAD have been associated with an increased risk of subsequent cardiac death, as well as a higher incidence of non-fatal stroke and HF (7). A meta-analysis involving 11,117 participants across eight studies demonstrated that SII could be a potential predictor of MACE in post-percutaneous coronary intervention patients (30). SII also exhibits moderate diagnostic accuracy for the identification of postoperative atrial fibrillation, thereby serving as a potential adjunctive marker for identifying high-risk patients and predicting adverse outcomes (31). As a novel biomarker reflecting the immune-inflammatory status of the body, SII holds promising potential for application in the field of cardiovascular diseases.Subgroup evaluations based on sample size, age, country, study type, SII cut off and cardiac function were conducted to further unravel the relationship between SII and incidence and death rates. Regarding mortality, despite our efforts, sources of heterogeneity could not be identified. The heterogeneity may be related to factors such as the type of HF, gender distribution, and BMI (1). Due to the limited number of available studies, subgroup analyses for these factors were not feasible. A reduction in heterogeneity to 0% was observed for incidence (categorized) within the United States, suggesting that incidence rates are regionally dependent and may be influenced by socioeconomic status, age distribution, and variations in lifespan. The incidence of HF appeared higher in studies from China and Turkey relative to those conducted in America. However, these observations were derived from single-center studies with limited sample sizes. Therefore, further research is warranted to determine whether geographic or regional factors modulate the relationship between SII and HF incidence (1). Moreover, subgroup analyses stratified by SII cutoff values suggested no clear dose-response relationship between SII levels and mortality risk. While low-to-moderate cutoff values (<2,000) were linked to increased mortality risk, the inverse association observed at thresholds ≥2,000 may be influenced by outliers, non-linear trends, or random error in small samples. Future studies utilizing multicenter, large-sample designs are encouraged to refine SII stratification, account for potential confounders, and validate the robustness of any dose-response relationships, thereby improving the generalizability and reliability of the findings. Moreover, mechanistic studies are needed to explore the role of SII across various stages and subtypes of HF, which may better support its integration into clinical decision-making. Furthermore, publication bias tests demonstrated that there exists certain publication bias in mortality (categorized), which could affect the quality of the evidence.Other immune-inflammatory markers, such as the neutrophil-to-lymphocyte ratio (NLR) and the platelet-to-lymphocyte ratio (PLR), have also been shown to be closely associated with HF. A meta-analysis involving over 18,000 patients with HF reported that NLR may serve as a useful tool for risk stratification, aiding in the identification of high-risk individuals. Elevated NLR levels were significantly associated with increased mortality risk in HF patients (HR: 1.12, 95% CI: 1.02–1.23, P = 0.013), suggesting its potential clinical utility (32). Another study (33) explored the relationship between PLR and clinical outcomes in HF patients, including mortality, rehospitalization, and worsening HF. Subgroup analyses (e.g., among patients with HFrEF) were conducted, and the study concluded that PLR had limited predictive value for prognosis in HF. Some conflicting findings may be attributable to differences in HF subtypes, underscoring the need for further research to validate these associations. In comparison to NLR and PLR, SII, also an immune-inflammatory marker, offers a more comprehensive reflection of the imbalance between systemic inflammation and immunity. In light of the present findings, it is suggested that elevated SII levels are associated with poor prognosis in HF and may, to a certain extent, indicate an increased risk of mortality.

SII is calculated by multiplying the neutrophil count by the platelet count and subsequently dividing by the lymphocyte count. An elevated SII value reflects a relative increase in platelet and neutrophil counts or a relative decrease in lymphocyte count, which may result from inflammatory responses and platelet activation. Neutrophils are integral to inflammatory processes, encompassing phagocytosis and degradation of bacteria and viruses, as well as the secretion of enzymes and reactive oxygen species to eradicate pathogens (34). In HF sufferers, it can be activated to release a significant amount of pro-inflammatory cytokines and oxidative stress substances (35–37), thereby promoting the progression of HF. Lymphocytes are involved in regulating the body's immune response, and their number and function are related to the onset and progression of HF (38). Lymphocytes are fewer in the HF population compared to the general population, and lymphopenia can independently predict poor survival in these patients (38–40). The reason is that lymphocytes modulate the inflammatory response by suppressing the release of inflammatory cytokines and activation of neutrophils and monocytes, thereby protecting the heart from damage (38). Moreover, platelets also contribute to the onset and progression of HF. Derived from the monocyte-macrophage system, platelets mediate between inflammation and thrombosis (38, 41, 42), with their activation and aggregation related to a higher incidence of cardiovascular events. Activated platelets promote inflammation by releasing pro-inflammatory cytokines and interacting with other sorts of cells, such as neutrophils, monocytes, and endothelial cells. Simultaneously, platelets, which can influence neutrophil function, damage surviving myocardial cells and exacerbate cardiac injury (33, 38, 43, 44). A study by Cătălina Liliana Andrei et al. demonstrated that increased mean platelet volume (MPV) is associated with a heightened mortality risk in patients with decompensated chronic HF, with elevated MPV serving as an independent predictor of hospital readmission and one-year mortality (45). Elevated platelet counts have been correlated with atherosclerosis, CAD, cerebrovascular disease (CD), and systemic inflammatory disorders (46–48), all of which significantly contribute to the pathophysiology of HF (2, 49). In summary, neutrophils, lymphocytes, and platelets exert distinct influences on the progression of myocardial fibrosis, perpetuation of systemic inflammation, endothelial injury, and impairment of cardiac function, collectively exacerbating HF pathology. Accordingly, SII, as a composite marker reflecting systemic inflammatory and immune status, may serve as both a diagnostic and prognostic biomarker in HF.

Our meta-analysis represents the first study unraveling the significance of SII in diagnosing and predicting HF. The results present considerable implications for assessing mortality risk in HF patients and refining clinical management strategies. Specifically, elevated SII levels are associated with increased mortality risk in patients with HF, highlighting the urgent need for early detection and comprehensive intervention in this population. Early detection strategies may include regular blood testing to dynamically monitor SII levels, natriuretic peptides, inflammatory markers, and complete blood counts; close clinical surveillance of symptoms and signs (e.g., through standardized questionnaires, monitoring of weight, edema, and pulmonary rales); and timely assessment of cardiac function via echocardiography and electrocardiography. In addition, proactive screening and management of key comorbidities such as diabetes, renal insufficiency, and atrial fibrillation are imperative. Comprehensive intervention should be multifaceted and intensified, with core components including strict adherence to guideline-directed medical therapy (e.g., ARNI/ACEI/ARB, beta-blockers, MRAs, SGLT2 inhibitors), individualized optimization of volume status through tailored diuretic strategies, intensive lifestyle modification (e.g., low-sodium diet, structured exercise rehabilitation, smoking cessation, and alcohol restriction), aggressive treatment of all comorbid conditions and precipitating factors, and precise utilization of device-based or surgical therapies (e.g., CRT, ICD, or valvular interventions). Furthermore, implementation of a structured, multidisciplinary HF management program and patient-centered self-management education system is essential to achieve comprehensive control of inflammation, improve cardiac function, slow disease progression, reduce mortality risk, and enhance patients’ quality of life.

Nevertheless, this meta-analysis possesses inherent limitations. First, all included studies were conducted solely in Asia and the Americas, specifically in China and Turkey. Therefore, the generalizability of these findings to patient populations in Europe, Africa, and other regions remains uncertain and should be interpreted with caution due to the geographic constraints of the available data. Further investigations are warranted to validate the prognostic significance of SII in HF populations outside Asia and the Americas.

Second, most included studies were retrospective, so potential confounding factors may affect the reliability of the results. HF has a heterogeneous etiology, including myocardial infarction, hypertension, valvular heart disease, and cardiomyopathies, all of which may influence the predictive value of inflammatory indices. However, due to limited reporting of specific etiologies in the original dataset, we were unable to conduct detailed subgroup analyses based on the underlying cause of HF. Therefore, the results may have been affected by the variation in HF etiology. Future research should investigate whether the predictive value of SII differs across distinct etiological subgroups to further elucidate its prognostic utility in varied pathological contexts. Substantial heterogeneity was observed across studies. Subgroup analyses indicated that heterogeneity diminished when stratified by country, study design, and ejection fraction category, suggesting these factors may contribute to the observed variability. However, due to incomplete data in the original publications, meta-regression analyses could not be conducted. Future studies should incorporate random-effects meta-regression to investigate these sources of heterogeneity in greater detail, thereby providing more precise guidance for clinical applications.

Additionally, the cut-off values of SII reported in these studies ranged from 590.4 to 3,986.19. Due to inconsistencies in the data and the limited number of studies, further analysis cannot be made on these values. To enhance reliability and comparability in future research, it is crucial to define a standard cut-off value for SII.