Predictive Value of IBI for In-Hospital Death in Elderly Patients with Non-ST-Segment Elevation Myocardial Infarction

Introduction

With the aging of the population, the treatment of elderly patients (≥75 years old) with non-ST-segment elevation myocardial infarction (NSTEMI) has become an important challenge in the field of cardiology [1]. Clinically, due to the decline in physiological reserve, complex comorbidities, and poor treatment tolerance, elderly NSTEMI patients exhibit a relatively high risk of in-hospital death [2]. Although primary percutaneous coronary intervention (PCI) can significantly reduce the risk of in-hospital death, real-world data show that only one-fourth of elderly NSTEMI patients receive PCI treatment [3,4]. In a clinical trial of frail elderly patients with NSTEMI, the routine invasive strategy provided no benefits [5]. Therefore, it is necessary to explore more risk factors for elderly NSTEMI patients who have not undergone PCI treatment to identify high-risk patients at an early stage.

Inflammation plays a crucial role in the pathogenesis, disease progression, and prognosis of elderly patients with NSTEMI [6]. In recent years, the inflammatory burden index (IBI), as an emerging inflammatory biomarker, has been widely applied in the field of risk stratification for various diseases [7–13]. Compared with single inflammatory markers, IBI, which is calculated as the product of C-reactive protein (CRP) and the neutrophil-to-lymphocyte ratio, can more comprehensively and stably assess the body’s inflammatory status, demonstrating higher accuracy in reflecting the degree of inflammation and predicting the prognosis [9–11]. Previous studies have confirmed that there is an independent association between IBI levels and the prevalence of cardiovascular diseases [11,12]. In another study, IBI was proven to be an independent predictor of all-cause mortality in patients with osteoarthritis [13]. However, in elderly patients with NSTEMI, the relationship between IBI and in-hospital death remains unclear. This study aims to evaluate the predictive efficacy of IBI for cardiogenic death during hospitalization in elderly NSTEMI patients. In line with the latest NSTEMI guideline [14], we set the age threshold at ≥75 years in this study to better capture a population with a higher burden of frailty, multimorbidity, and competing risks.

Material and Methods

STUDY POPULATION:

This single-center retrospective study consecutively enrolled patients diagnosed with NSTEMI at the First Affiliated Hospital of Nanjing Medical University between February 2021 and February 2025. NSTEMI was defined in accordance with the latest guidelines [14,15]. Briefly, NSTEMI was characterized by a rise and/or fall in cardiac injury biomarkers – primarily high-sensitivity cardiac troponin – with at least 1 value above the 99th percentile upper reference limit, accompanied by clear evidence of ischemia and no persistent ST-segment elevation on ECG. The study protocol was reviewed and approved by the Institutional Review Board (IRB) of the First Affiliated Hospital of Nanjing Medical University (2025-SR-272). This research was conducted in accordance with the Declaration of Helsinki. Given that the study posed no risks to the patients, the IRB waived the requirement for informed consent. The inclusion criteria were as follows: (1) aged ≥75 years; (2) not receiving PCI treatment; (3) having complete clinical data. The exclusion criteria were: (1) a history of myocardial infarction (MI); (2) comorbid inflammatory diseases or autoimmune diseases; (3) severe renal insufficiency (estimated glomerular filtration rate <30 mL/min/1.73 m2); (4) active bleeding in vital organs; (5) patients with malignant tumors or hematological diseases. We excluded patients with a history of MI to ensure a more homogeneous index-pathophysiology at presentation and to minimize confounding due to prior infarct-related remodeling, revascularization, and secondary prevention therapies. Patients were divided into 2 groups according to whether cardiogenic death occurred during hospitalization. Given the retrospective design, we provide a brief post hoc power/precision justification for the primary endpoint (in-hospital cardiogenic death). We assumed an event rate of 10%, set α=0.05, and targeted 80–90% power. The effect size for IBI was specified as an adjusted odds ratio per 1-SD (or per log-unit) increase. Based on these assumptions, we evaluated power for the available sample and confirmed that the final sample size is adequate to detect the prespecified effect size and to yield precise estimates of model performance (including sensitivity and specificity with 95% CI).

CLINICAL DATA COLLECTION:

The clinical data of all enrolled patients were collected based on their medical records, encompassing information such as age, sex, body mass index (BMI), previous medical history, and medications administered. The lymphocyte count, neutrophil count, and CRP levels measured for the first time after the patients’ admission were recorded. To more effectively reflect myocardial injury and cardiac function status, the recorded high-sensitivity troponin T (hs-TNT) and B-type natriuretic peptide (BNP) were peak values during hospitalization. The neutrophil-to-lymphocyte ratio (NLR) was defined as the ratio obtained by dividing the neutrophil count by the lymphocyte count. The IBI was defined as the product of CRP and NLR [9–11]. The medications recorded included aspirin, P2Y12 inhibitors, β-blockers, statins, sodium-glucose cotransporter 2 inhibitors, angiotensin-converting enzyme inhibitors, and angiotensin receptor blockers.

STATISTICAL ANALYSIS:

To assess the normality of continuous variables, the Kolmogorov-Smirnov test was employed. Normally distributed and non-normally distributed continuous data were statistically analyzed using the independent samples t-test and Mann-Whitney U test, respectively. The comparison of categorical variables between the 2 groups was performed using the chi-square test. Logistic regression analysis was used to explore risk factors associated with in-hospital death. Specifically, based on prior literature and pathophysiology, the following potentially relevant variables were incorporated into multivariate regression analysis via a stepwise forward method: IBI, age, sex, hypertension, diabetes mellitus, systolic blood pressure, heart rate, Killip class, baseline antithrombotic therapy, left ventricular ejection fraction (LVEF), hs-TNT, and BNP. The receiver operating characteristic curve (ROC) was used to evaluate the predictive efficacy of the IBI for in-hospital death, and the DeLong test was used to compare the differences between the areas under the curve (AUCs). The primary threshold selection rule was the Youden Index, with sensitivity/specificity-targeted alternative cut-offs considered in scenario analyses. We estimated sensitivity, specificity, and their 95% CIs at the selected thresholds [16]. Restricted cubic splines (RCS) were used to explore the dose-response relationship between IBI and in-hospital death. SPSS (Version 27.0, Chicago, USA) and R (Lucent Technologies, New Jersey, USA) were used for statistical analysis. Statistical significance was defined as a P value <0.05.

Results

BASELINE DATA COMPARISON BETWEEN GROUPS:

A total of 418 elderly patients with NSTEMI were enrolled in this study, and cardiogenic in-hospital deaths occurred in 43 (10.3%) (Figure 1). The mean age of the patients was 79.60±3.67 years, and males accounted for 66.03% of the total. The analysis of baseline data showed that, compared with the surviving elderly NSTEMI patients, the patients with cardiogenic death had significantly higher levels of the IBI, CRP, NLR, neutrophil count, and BNP, and a significantly higher proportion of patients with KILLIP >1. In contrast, the left ventricular ejection fraction (LVEF) and lymphocyte count were significantly lower (P<0.05) (Table 1).

ASSOCIATION BETWEEN THE IBI AND IN-HOSPITAL DEATH IN ELDERLY PATIENTS WITH NSTEMI:

Considering that the values were relatively large, IBI, hs-TNT, and BNP were logarithmically transformed and then subjected to logistic regression analysis. All variables were included in the univariate logistic regression analysis. The results demonstrated that the IBI, LVEF, KILLIP >1, lymphocyte count, CRP, and BNP were associated with cardiogenic death during hospitalization in elderly patients with NSTEMI (Table 2). To avoid the interference of collinearity, the relevant variables were incorporated into the multivariate logistic regression analysis using the stepwise forward method. The results showed that IBI (OR=2.22, 95% CI: 1.64–3.00), BNP, and LVEF were independent predictive risk factors for cardiogenic death during hospitalization in elderly patients with NSTEMI (Table 3). The results of RCS suggested a non-linear dose-response relationship between IBI and in-hospital death (p for nonlinearity <0.001, P for overall <0.001), showing that the higher the IBI, the greater the risk of in-hospital death (Figure 2).

ROC CURVES FOR IN-HOSPITAL DEATH IN ELDERLY PATIENTS WITH NSTEMI:

The results of ROC showed that the area under the curve (AUC) of IBI was 0.760 (cut-off value: 36.85, 95% CI: 0.689–0.832, P<0.001), the AUC of CRP was 0.676 (cut-off value: 2.41, 95% CI: 0.600–0.753, P<0.001), and the AUC of NLR was 0.710 (cut-off value: 9.12, 95% CI: 0.639–0.782, P<0.001). The DeLong test demonstrated that the discriminatory ability of IBI for cardiogenic death during hospitalization in elderly NSTEMI patients was significantly superior to that of CRP (z=3.967, P<0.001), while there was no statistically significant difference with NLR (z=1.398, P=0.162) (Figure 3, Table 4).

Discussion

LIMITATIONS:

First, as a retrospective study, its capacity for causal inference is limited and it is prone to selection bias, information bias, and unmeasured confounding. Second, our findings are primarily applicable to older NSTEMI patients aged ≥75 years, a group characterized by greater frailty and competing risks. Accordingly, caution is warranted when extrapolating these results to younger populations. Third, our study focused exclusively on in-hospital cardiogenic death. Whether IBI is associated with post-discharge cardiogenic mortality remains to be determined and will require future follow-up data for further clarification. Fourth, advanced age is often associated with frailty. Because this study was retrospective and frailty assessment is not routinely performed at our center, relevant data were unavailable. Consequently, the absence of a standardized measure of frailty limited our ability to control for confounding. Future prospective studies should incorporate validated frailty scales to enhance the adequacy of risk adjustment and the interpretability of results. Fifth, in our study, the IBI cut-off was set at 36.85, yielding a sensitivity of 74.4% and a specificity of 71.7%. Depending on the intended clinical application, higher sensitivity may be prioritized for rule-out and higher specificity for rule-in. It should be noted that these are in-sample estimates and may be subject to optimism; stronger evidence requires external validation in independent cohorts.

Conclusions

In elderly patients with NSTEMI, IBI demonstrated an independent association with in-hospital mortality, with modest discriminatory performance. There is a non-linear dose-response relationship between IBI and in-hospital death in elderly patients with NSTEMI.

Data Availability Statement

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.