Abstract
Early identification of high-risk emergency surgical patients is crucial. The shock index (SI) and modified shock index (MSI) are simple vital sign-based parameters with proven prognostic value in trauma, yet their role in non-trauma emergency surgery remains unclear. This study aimed to evaluate the prognostic value of SI, MSI, and their dynamic changes (ΔSI and ΔMSI) between emergency department (ED) admission and the preoperative period in predicting perioperative mortality. This single-center retrospective cohort included 796 adults undergoing emergency surgery within 48 hours of ED admission in 2024. SI and MSI were calculated at ED admission and preoperatively, and ΔSI and ΔMSI were defined as the differences between these time points. The primary outcome was in-hospital mortality; secondary outcomes were postoperative complications. Logistic regression and receiver operating characteristic analyses were used to assess predictive performance. In-hospital mortality was 7.3% (58/796). Non-survivors were older, had higher American Society of Anesthesiologists scores, underwent major surgery more often, and showed higher preoperative heart rate, SI, and MSI with lower blood pressures (P < .05). ΔSI predicted mortality with an AUC of 0.73, outperforming ΔMSI (AUC 0.67). A ΔSI ≥ 0.1 showed 87% specificity and was associated with higher mortality (23% vs 4%). In multivariate analysis, ΔSI ≥ 0.1, and both ED and preoperative heart rates remained independent predictors. Dynamic changes in SI between ED admission and preoperative evaluation provide strong prognostic information. A ΔSI threshold ≥ 0.1 is highly specific for mortality risk, supporting its utility as a practical, easily calculated tool for perioperative risk stratification. Prospective multicenter validation is warranted.