The data used in this survey comes from Jiangxi-acute decompensated heart failure study II. This is a cohort study initiated by Jiangxi Provincial People's Hospital, consecutively enrolling 3,484 patients with ADHF admitted to the Jiangxi Provincial People's Hospital from January 2018 to January 2024. The primary objective of this project is to establish a high-quality cohort of ADHF patients, effectively utilize their clinical record data during hospitalization, and explore new methods for early risk stratification to improve the adverse prognosis of ADHF patients. In this study, the diagnosis of ADHF was based on the ESC and ACC/AHA/HFSA Heart Failure Guidelines (2, 3), incorporating clinical symptoms, physical signs, and laboratory findings. The diagnostic criteria were as follows: The presence of at least one sign of HF: (a) Elevated N-terminal pro-brain natriuretic peptide (NT-proBNP); (b) Pulmonary edema detected by physical examination or chest x-ray; (c) Abnormal cardiac structure and/or function as indicated by echocardiography. The presence of at least one symptom of worsening HF: (a) Systemic venous congestion; (b) Dyspnea; (c) Insufficient tissue perfusion.

In the current study, we established the following exclusion criteria based on the research objectives: (i) To account for the potential impact of additional fluid and sodium retention, we excluded patients with uremia or a history of hemodialysis (n = 231) and those with liver cirrhosis (n = 42); (ii) Considering the potential influence on life expectancy, we excluded patients with malignant tumors (n = 160); (iii) Due to the significant role of reperfusion therapy in short-term prognosis, participants who had undergone percutaneous coronary intervention (PCI) within the past 3 months were excluded (n = 102); (iv) Participants under the age of 18 (n = 22); (v) Pregnant individuals (n = 4); (vi) Individuals with pacemaker-controlled heart rhythms, as their heart rates were not expected to be regulated by autonomic nervous control (n = 121). Additionally, we excluded participants with missing IBI data (n = 1,561). Ultimately, 1,241 patients with ADHF were included in the analysis. Given the high rate of missing IBI data in this study, we conducted a systematic evaluation of baseline characteristic differences between the complete-case group and the missing-data group prior to formal analysis. As shown in Supplementary Table S1, no statistically significant differences were observed between the two groups across most baseline characteristics (P > 0.05), suggesting that the missing data mechanism aligns with the missing at-random assumption. This finding provides methodological assurance regarding data quality for subsequent analyses. A detailed flowchart of the study population screening process is shown in Figure 1.

This study adhered to the ethical principles outlined in the Declaration of Helsinki. The use of research data strictly complied with ethical review requirements, and authorization was obtained from patients and their families. The study protocol was approved by the Ethics Committee of Jiangxi Provincial People's Hospital (IRB: 2024-01). The study followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines to ensure transparency and scientific rigor of the findings (34).

The baseline data for this study were collected by two trained researchers from the hospital's electronic medical record system, with cross-verification to ensure accuracy. The specific details are as follows: (i) Demographic and clinical data: sex, age, drinking status, smoking status, comorbidities [including hypertension, diabetes, stroke, and CHD], cardiac function (New York Heart Association classification: NYHA), blood pressure data [measured using an Omron automatic blood pressure monitor (HBP-1300) in a quiet environment or at the bedside] and medication information during hospitalization [Includes the use of beta-blockers, diuretics, angiotensin-converting enzyme inhibitors (ACEI)/angiotensin receptor inhibitors (ARB)/angiotensin receptor neprilysin inhibitors (ARNI), and vasopressor medications]. (ii) Echocardiographic examination: Left ventricular ejection fraction (LVEF). (iii) Laboratory test data: The biochemical indicators measured included albumin, alanine aminotransferase, aspartate aminotransferase (AST), creatinine (Cr), uric acid (UA), total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and fasting plasma glucose (FPG). Additionally, other assessed parameters included white blood cell (WBC) count, red blood cell (RBC) count, platelet count, CRP, neutrophil count, lymphocyte count, and NT-proBNP. All blood samples were collected within 24 h of hospital admission, adhering strictly to the timing requirements for laboratory results. For liver enzymes, lipid profiles, and FPG, venous blood samples were collected either at admission under fasting conditions or on the morning of the second day after admission.

IBI = CRP × (neutrophil count/lymphocyte count) (17).

The primary endpoint of this study was all-cause mortality within 30 days after hospital admission in patients with ADHF. The 30-day survival status of all participants was tracked by trained medical staff through multiple methods, including text messages, phone calls, and face-to-face follow-ups during outpatient clinics or hospital admissions.

All statistical analyses in this study were performed using R software (version 4.2.1) and Empower® software (version 2.0). Statistical significance was defined as a two-sided p-value < 0.05.

First, we stratified ADHF patients into tertiles (low, moderate, and high) based on IBI, which were determined by calculating the 33.33% and 66.67% percentiles of IBI values. Baseline variables were described according to their type and distribution: categorical variables were expressed as counts (%), while continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range), as appropriate. Group differences were analyzed using chi-square tests, one-way ANOVA, or non-parametric tests, as appropriate.

To assess the association between IBI and 30-day all-cause mortality in ADHF patients, we performed Kaplan–Meier analysis to plot survival curves for the three IBI groups. The significance of differences in survival rates among the groups was assessed using the log-rank test. Subsequently, we developed three adjusted Cox proportional hazards regression models to evaluate the association between IBI and 30-day mortality. Model 1 adjusted for baseline information assessed at admission, including sex, age, hypertension, diabetes, stroke, and CHD. Model 2 added NYHA classification, drinking status, smoking status, and LVEF. Based on Model 2, Model 3 was further adjusted for monocyte count, RBC, platelet count, AST, Cr, UA, TC, TG, HDL-C, LDL-C, FPG, and NT-proBNP. The proportional hazards assumption was evaluated using Kaplan–Meier curves for IBI groups and Schoenfeld residual tests, revealing no evidence of violation of this assumption (Figure 2 and Supplementary Figure S1). Additionally, based on collinearity assessments, we confirmed the absence of multicollinearity among covariates in the multivariable regression models (Supplementary Table S2) (35).

We also performed subgroup analyses to examine whether the association between IBI and 30-day mortality in ADHF patients was consistent across different subgroups. The subgroup variables and detailed stratification were as follows: age (<65 years vs. ≥65 years), sex (male vs. female), LVEF (<50% vs. ≥50%), NYHA classification (class III vs. class IV), hypertension (yes vs. no), diabetes (yes vs. no), stroke (yes vs. no), and CHD (yes vs. no). The significance of interaction effects was assessed using likelihood ratio tests.

To evaluate the predictive ability of IBI, we performed receiver operating characteristic curve analysis to assess the predictive performance of IBI and common inflammatory markers (CRP, neutrophil count, lymphocyte count, monocyte count, and WBC count) for 30-day mortality. The area under the curve (AUC), optimal threshold, sensitivity, and specificity were calculated for each indicator. Differences in AUCs were evaluated using the DeLong test. Additionally, we investigated the incremental predictive value of adding IBI to the established clinical risk model [Acute Decompensated Heart Failure National Registry (ADHFRE)] (36, 37) and calculated the C-index to quantify the improvement in predictive performance.

To ensure the robustness of the study findings, we conducted several sensitivity analyses: (i) Considering the potential impact of acute inflammation, we excluded patients with pulmonary infections at admission and repeated the primary analysis; (ii) To reduce the influence of reverse causality, we excluded participants who died within three days after admission; (iii) Given that hypertension, diabetes, stroke, and CHD are strong risk factors for adverse prognosis in ADHF patients, we excluded patients with these comorbidities and repeated the analysis (38, 39); (iv) For partially missing data (Supplementary Table S3), we performed multiple imputation to estimate missing values and repeated the primary analysis. (v) Medical treatment serves as the cornerstone of ADHF interventions. In subsequent models, we adjusted for ADHF treatment factors including beta-blockers, diuretics, ARB/ACEI/ARNI, and vasopressor agents. (vi) Considering that CHD patients undergoing PCI are typically a susceptible population for ADHF and do not interfere with the prognostic evaluation of IBI, we re-included this patient subgroup in further sensitivity analyses. (vii) To assess the generalizability of our findings, we utilized data from the United States National Health and Nutrition Examination Survey (1998–2018) to examine the association between IBI and all-cause mortality among participants diagnosed with congestive HF.

Among the 1,241 ADHF patients who met the study criteria, 720 were male and 522 were female, with a mean age of 68 years. The baseline characteristics of ADHF patients stratified by IBI tertiles are summarized in Table 1. Compared to patients in the low IBI group, those in the high IBI group were more likely to be male, older, and have a higher prevalence of diabetes, CHD, and NYHA Class IV. Additionally, they exhibited higher levels of CRP, WBC count, neutrophil count, monocyte count, AST, Cr, UA, FPG, and NT-proBNP, as well as lower levels of lymphocyte count, RBC count, TC, HDL-C, and LDL-C (All p < 0.05). Additionally, regarding treatment, compared to patients with low IBI, high IBI patients demonstrated a significantly lower proportion of ACEI/ARB/ARNI use (p = 0.015) and a markedly higher utilization rate of vasopressors (p < 0.001), while no significant differences were observed in diuretic or beta-blocker administration (both p > 0.05).

During the 30-day follow-up, 108 deaths occurred among the 1,241 ADHF patients. The mortality rates in the low, moderate, and high IBI groups were 1.93%, 4.60%, and 19.57%, respectively (Figure 3): as IBI increased, the 30-day mortality rate among ADHF patients demonstrated a progressive increase. Kaplan–Meier analysis showed that higher IBI was associated with increased all-cause mortality: the high IBI group had a significantly higher 30-day mortality rate compared to the low and moderate IBI groups (Figure 2: log-rank p < 0.0001).

Table 2 shows the hazard ratios (HRs) for the association between IBI, analyzed as both a continuous and categorical variable, and all-cause mortality. From Model 1 to Model 3, the HRs for the association between IBI and 30-day mortality in ADHF patients were 1.37, 1.40, and 1.29, respectively. Despite the attenuation of HRs with increasing levels of model adjustment, the positive association between IBI and 30-day mortality persisted across all models in ADHF patients. In the final model (Model 3), each SD increase in IBI was associated with a 29% increased risk of 30-day mortality (HR: 1.29, 95% CI: 1.15–1.46). Additionally, compared to the low IBI group, the high IBI group had a 368% higher risk of 30-day mortality (HR: 4.68, 95% CI: 1.06–13.73). Across all models, IBI showed a significant positive trend with 30-day mortality in ADHF patients (all p-trend < 0.001). These findings suggest that elevated IBI serves as an independent risk factor for poor short-term prognosis in ADHF patients.

Table 3 presents the results of subgroup analyses stratified by age, sex, LVEF, NYHA classification, and comorbidities (hypertension, diabetes, stroke, and CHD). After further likelihood ratio tests, we found no significant interaction between IBI and 30-day mortality in ADHF patients across subgroups (LVEF, NYHA classification, and comorbidities; All p-interaction >0.05), except for sex. These findings indicate that the association between IBI and short-term mortality prognosis in ADHF patients demonstrates robust stability across the majority of patient populations. In the sex subgroup, females had a significantly higher risk of IBI-related all-cause mortality compared to males (HR: female1.52 vs. male1.33, p-interaction = 0.0431). By contrast, female patients with IBI-related ADHF demonstrated a 1.14-fold higher 30-day mortality risk compared to males.

To further explore the potential clinical explanations for gender-based differences in IBI-associated mortality risk among ADHF patients, we performed a gender-stratified analysis comparing baseline comorbidities and treatment factors between medium-to-high IBI subgroups (IBI ≥18.32): Our findings revealed that compared to male ADHF patients, female ADHF patients exhibited a higher prevalence of diabetes and stroke but lower rates of hypertension and CHD. Regarding treatment, women were less likely to receive ACEI/ARB/ARNI and diuretics but more likely to receive beta-blockers and vasopressors compared to men. However, despite these observed trends, no statistically significant differences were found between genders in comorbidities or medication use (Supplementary Table S4, all p > 0.05).

The results of the predictive value analysis for IBI and multiple common inflammatory markers for 30-day mortality in ADHF patients are shown in Table 4 and Figure 4. The study demonstrated that conventional inflammatory biomarkers— CRP, WBC, neutrophil count, lymphocyte count, and monocyte count—each exhibited predictive value for 30-day mortality in ADHF patients, with respective predictive accuracies of 74%, 66%, 70%, 68%, and 60%. Compared with these conventional inflammatory biomarkers, IBI demonstrated superior predictive performance for 30-day mortality in ADHF patients, achieving approximately 80% accuracy (all DeLong's test p < 0.0001). Additionally, the optimal threshold for IBI in predicting 30-day mortality in ADHF patients was calculated as 159.86, with a specificity of 0.76 and a sensitivity of 0.70. Collectively, as a novel inflammatory biomarker, IBI significantly enhances the predictive accuracy for short-term adverse outcomes in ADHF patients beyond conventional inflammatory indicators.

We further evaluated the incremental predictive value of adding IBI to the established clinical risk model (ADHFRE). The results demonstrated that incorporating IBI into the ADHFRE model for predicting 30-day mortality significantly improved its predictive performance: the C-index increased from 0.58 to 0.82 (P < 0.01). These findings highlight that the addition of IBI provides significant incremental value to the ADHFRE risk model for predicting short-term mortality.

In the sensitivity analysis, the association between IBI and 30-day mortality in ADHF patients remained significant after excluding those with pulmonary infection (Table 5: Sensitivity-1); Specifically, in ADHF patients without pulmonary infection, IBI remained positively associated with 30-day mortality, yielding a HR of 1.61 (95% CI: 1.08–2.41). Furthermore, the main results remained largely unchanged after further excluding patients with hypertension, diabetes, stroke, and CHD, or patients who died within three days of admission: IBI maintained a robust positive association with 30-day mortality across these subgroups (Table 5: Sensitivity Analyses 2 and 3). Repeating the primary analysis in the imputed complete dataset yielded robust results (Table 5: Sensitivity-4). After adjusting for treatment factors including β-blockers, diuretics, ARBs/ACEIs/ARNI, and vasopressors, the findings remained consistent with the primary analysis (Table 5, Sensitivity-5). Additionally, we repeated the analyses in the complete cohort including patients who underwent PCI within the last 3 months (n = 1,241 + 44; due to missing CRP data in 58 participants who received PCI in the past three months, 44 additional individuals were incorporated into the original cohort, yielding a total of 1,285 participants). The findings remained consistent with the primary results (Table 5, Sensitivity-6). Finally, analysis of the external United States cohort confirmed a positive association between IBI and mortality risk among individuals with congestive HF (Table 5, Sensitivity-7), which is consistent with the results reported in the present study.

The strengths of this study lie in its novel findings and study population: (1) IBI showed excellent predictive value for 30-day mortality in ADHF patients (AUC: 0.80), which is promising news for ADHF patients as IBI can be obtained conveniently and effectively. (2) To our knowledge, this is the first study to evaluate the association between IBI and short-term mortality in ADHF patients, with results validated through multiple sensitivity analyses.

Some potential limitations should also be mentioned: (1) The participants in this study were primarily from Jiangxi, a southern city in China, which may restrict the applicability of our findings to northern China or other ethnic populations. (2) As a non-interventional study, it could not assess the impact of anti-inflammatory treatments on outcomes in ADHF patients after hospital admission. (3) This study primarily evaluated the predictive capability of IBI at admission for subsequent adverse events. The impact of IBI changes during hospitalization on prognosis is still unclear and warrants further investigation. (4) As with other observational studies, residual confounding cannot be eliminated. (5) A substantial proportion of ADHF patients lacked baseline CRP measurements at admission, resulting in missing IBI data. Although these missing values met the criteria for missing at random, the relative reduction in sample size may have influenced the findings to some extent, necessitating external validation in larger cohorts. (6) While 30-day follow-up effectively captures acute-phase events, it does not evaluate the longitudinal prognostic impact of IBI on ADHF patients. Future studies with extended follow-up are required to characterize the temporal trajectory of IBI's effects across short-, medium-, and long-term outcomes.