A composite model incorporating systemic inflammatory burden and GRACE score for enhanced identification of elevated myocardial stress in acute myocardial infarction: an exploratory analysis
Acute myocardial infarction (AMI) remains a leading cause of mortality and heart failure worldwide. Early identification of patients with elevated myocardial stress, a key driver of adverse outcomes, is clinically important. However, the utility of a composite inflammatory burden score for this purpose remains unclear. This retrospective study enrolled 302 consecutive patients hospitalized for AMI at Shenzhen Qianhai Shekou Hospital between January 2023 and December 2024, with 50 healthy controls for descriptive purposes only. Patients were stratified into three groups based on their systemic inflammatory burden score (SIBS): low (score 0), intermediate (score 1), and high (score 2) inflammatory burden. Clinical characteristics, laboratory parameters, and in-hospital outcomes were compared across groups. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the discriminatory ability of SIBS, GRACE score, hs-CRP, albumin, C-reactive protein-to-albumin ratio (CAR), and a composite logistic regression model combining SIBS ≥ 1, GRACE score, and hs-CRP for identifying elevated myocardial stress, defined as NT-proBNP >75th percentile (647 pg/mL) as an exploratory surrogate endpoint. Areas under the curve (AUC) were compared using the DeLong test. Internal validation was performed using 10-fold cross-validation and bootstrap optimism correction on the final model. The SIBS was significantly correlated with heart rate, Killip class, GRACE score, left ventricular ejection fraction (LVEF), prevalence of multi-vessel coronary disease, and length of hospital stay in AMI patients ( P < 0.05). The high inflammatory burden group (SIBS = 2) exhibited more severe cardiac dysfunction and extensive coronary artery disease. For identifying elevated myocardial stress, the GRACE score yielded the highest AUC among individual markers (0.766), followed by albumin (0.707) and hs-CRP (0.671). The composite model incorporating SIBS ≥ 1, GRACE score, and hs-CRP significantly improved discriminatory capacity, achieving an apparent AUC of 0.815 (95% CI: 0.737–0.888), which was superior to any single marker alone ( P < 0.05). After bootstrap optimism correction, the corrected AUC was 0.798, and the Hosmer-Lemeshow test indicated good calibration ( P = 0.32). The model also showed a Brier score of 0.168, a calibration slope of 0.94 (95% CI: 0.82–1.06), and a calibration intercept of −0.12 (95% CI: −0.45 to 0.21). In this exploratory analysis, SIBS is associated with clinical severity and myocardial stress in AMI patients. A composite model combining SIBS, GRACE score, and hs-CRP improves identification of elevated myocardial stress compared with single biomarkers in this exploratory analysis. External validation is needed before clinical application.