An inflammatory–hyperglycemic stress index predicts mortality in acute coronary syndrome: development and external validation in two large cohorts
K Zhang, K Yan, M Liu, F Xu, Y Yang, J Yuan, X ZhaoAbstract
Background
The stress hyperglycemia ratio (SHR) has emerged as a robust predictor of adverse outcomes in acute coronary syndrome (ACS). Though accumulating evidence suggests that its prognostic value may be modified by systemic inflammatory status, commonly reflected by high-sensitivity C-reactive protein (hsCRP), it remains unknown whether integrating these pathways into a unified index improves risk prediction.
Purpose
This study aimed to develop and validate an integrated inflammatory-hyperglycemic stress index (IHSI) for mortality prediction in ACS patients.
Methods
We analyzed 6,149 ACS patients from a derivation cohort and 5,218 from a multicenter external validation cohort. The novel index was developed using a LASSO-based covariate selection approach followed by multivariable Cox regression, integrating hsCRP and the absolute deviation of SHR from its optimal value (identified by restricted cubic spline analysis) as: IHSI = 0.0277×hsCRP+0.6289×|SHR - 0.76|.
Results
The mean age was 58.4±10.4 years in the derivation cohort and 60.4±11.6 years in the validation cohort; 76.8% and 77.9% were male, respectively. Time-dependent ROC analyses showed that SHR achieved higher AUC values among patients with elevated hsCRP compared with those with lower hsCRP levels (e.g., 1-year AUC: 0.67 vs. 0.42). In the derivation cohort, IHSI demonstrated superior discrimination compared with hsCRP and SHR (1-year AUC: 0.72 vs. 0.66 and 0.60), with consistent findings in the validation cohort (0.72 vs. 0.68 and 0.65). IHSI also provided significant net reclassification improvement (derivation: 26.9% vs. hsCRP; 32.5% vs. SHR). Furthermore, in adjusted Cox models, each standard deviation increase in IHSI was independently associated with higher mortality, with HRs of 4.33 (95% CI 2.14-8.76) for 5-year mortality in the derivation cohort and 2.46 (95% CI 1.96-3.09) for 2-year mortality in the validation cohort.
Conclusion
The IHSI was independently associated with mortality and improved risk discrimination beyond individual biomarkers. Its robust performance across independent cohorts supports its potential utility as a risk assessment tool derived from routinely available metrics.