Abstract Background/Introduction Non-ST-elevation myocardial infarction (NSTEMI) comprises 70% of acute coronary syndromes in contemporary practice.¹ Notwithstanding significant advancements in therapeutic approaches, in-hospital mortality rates in NSTEMI remain substantial, ranging from 3% to 7%.¹ Whilst the Global Registry of Acute Coronary Events (GRACE) score remains paramount for risk stratification, inflammatory indices may offer additional prognostic value.² The Systemic Inflammatory Response Index (SIRI) and Systemic Immune-Inflammation Index (SII) have demonstrated promising utility in cardiovascular risk assessment; however, their comparative predictive value remains insufficiently explored in NSTEMI.³ Purpose To compare SIRI and GRACE score for predicting in-hospital mortality in NSTEMI patients whilst evaluating whether inflammatory markers enhance prognostic accuracy. Methods We conducted a single-centre, retrospective cohort study examining 644 consecutive NSTEMI patients who underwent coronary angiography between July 2023 and October 2024. Inflammatory indices were calculated from admission blood samples obtained within 24 hours of presentation, whilst GRACE scores were determined utilising standardised algorithms. The primary endpoint comprised in-hospital all-cause mortality. Statistical analyses incorporated receiver operating characteristic curve analysis and Cox proportional hazards regression to identify independent predictors of mortality. Results Amongst the study cohort (n=644), 38 patients (5.9%) experienced in-hospital mortality. Non-survivors demonstrated significantly elevated SIRI values (6.54±9.22 vs 1.65±1.26×10³/µL, p0.001), SII measurements (3245.02±3416.11 vs 1028.83±1092.01×10³/µL, p0.001), and GRACE scores (145.1±27.8 vs 117.2±26.8, p0.001) compared with survivors (Table 1). ROC analysis revealed excellent discriminative capacity for mortality prediction: GRACE score (AUC=0.798, 95% CI: 0.719-0.878), SIRI (AUC=0.781, 95% CI: 0.704-0.858), and SII (AUC=0.771, 95% CI: 0.697-0.846) (Figure 1). Optimal threshold values were established at 121.50 for GRACE score (sensitivity 89.8%, specificity 52.2%) and 2.07 for SIRI (sensitivity 83.7%, specificity 42.6%) (Table 2). Multivariate Cox regression analysis identified both SIRI (HR: 1.612, 95% CI: 1.231-2.111, p=0.001) and GRACE score (HR: 1.029, 95% CI: 1.014-1.044, p0.001) as independent predictors of in-hospital mortality (Table 3). Conclusion This investigation demonstrates that both SIRI and GRACE score independently predict in-hospital mortality in NSTEMI patients, exhibiting comparable discriminative capabilities. The robust predictive value of SIRI, combined with its accessibility and cost-effectiveness, suggests its potential utility as a complementary risk stratification tool alongside the established GRACE score. These findings may facilitate the identification of high-risk patients who might derive benefit from more intensive therapeutic strategies. Figure 1, Table 2, and Table 3
Bedir et al. (Sat,) studied this question.