Simplified acute physiology score ii to predict hospital mortality among ST-elevation myocardial infarction patients in Switzerland

Abstract Background ST-elevation myocardial infarction (STEMI) is a prevalent diagnosis in Swiss intensive care units (ICU). Nevertheless, a validated scoring system to predict early mortality prediction among these patients is lacking. SAPS II (Simplified Acute Physiology Score II) is collected in each ICU patient on a mandatory base. However, this score was developed to predict mortality in critically ill patients, explicitly excluding cardiac patients in its original conception. Evidence about its predictive validity for hospital mortality in patients with STEMI is scarce. Purpose This study aims to assess the performance of the above scoring system in predicting in-hospital mortality of intensive care patients with STEMI using contemporary data from a long-term period in Switzerland. Material and Methods ICU patients with STEMI enrolled in the Acute Myocardial Infarction in Switzerland (AMIS) Plus registry between 2005 and 2024 were included. In-hospital mortality was analysed using a multivariate logistic regression model. Results Out of 6,124 patients included (mean age 64.4 ±12.6; men n=4,695, 77%), 13% (n=818) had a return of spontaneous circulation after cardiac arrest and 19% (n=1,164) a Charlson Comorbidity Index 1. Complications during hospitalisation included cardiogenic shock (n=362, 5.9%), sepsis (n=175, 2.9%) and acute kidney injury (n=191, 3.1%). The overall in-hospital mortality was 8.1% (n=498), while the mean SAPS II score recorded was 28.3. Patients who died had markedly higher SAPS II scores (mean 61.2 (±23); median 63.0, IQR 43, 80) compared to survivors (mean 25.4 (±14); median 22, IQR 18, 29) (p0.001). For SAPS II between 30 and 60 in-hospital mortality was higher than predicted by the score, whereas for scores exceeding 60, predictions overestimated observed mortality (Figure 1). After adjusting for age and sex, each one-point increase in SAPS II was associated with an 8% higher chance of mortality (OR 1.08; 95%CI 1.08-1.09; p0.001). Conclusion The SAPS II is associated with hospital mortality in STEMI patients hospitalised in Swiss ICUs. The higher observed mortality compared to the predicted mortality in STEMI patients with lower SAPS II suggests the existence of mortality-related confounders not captured by the SAPS II.