Evaluation of SI, MSI and DSI for very early (3-day) mortality in patients with septic shock

Abstract

Background Septic shock is associated with increased mortality. Predicting mortality, including early prediction for septic shock patients in intensive care units (ICUs), remains an important challenge. Method We searched the Medical Information Mart for Intensive Care IV database. Odds ratios (ORs) with 95% confidence intervals (CIs) of the relationships between shock index (SI), modified SI (MSI), and diastolic SI (DSI) of patients with septic shock requiring vasopressors and 3-day/in-hospital mortality were calculated using logistic regression models. The time-course changes of these parameters were compared between survivors and non-survivors. The performance of the different parameters was described by the area under the receiver operating characteristic (ROC) curve (AUC) and compared with DeLong analysis. Results A total of 1266 patients with septic shock requiring vasopressors were identified. The 3-day mortality rate and in-hospital mortality rate were 8.7% and 23.5%, respectively. Multivariable logistic regression analysis showed significant associations between pre-vasopressor SI/MSI/DSI and 3-day mortality in patients with septic shock requiring vasopressors in fully adjusted models (Ps for trend < 0.01). The AUCs of pre-vasopressor SI, MSI, and DSI were 0.746, 0.710, and 0.732 for 3-day mortality, respectively. There were significant differences in the time-course of SI, MSI, and DSI between survivors and non-survivors at 3-day/in-hospital mortality among patients with septic shock requiring vasopressors (repeated-measures ANOVA, inter-subjects difference P < 0.001). Conclusion Pre-vasopressor SI, MSI, and DSI values identified patients with septic shock requiring vasopressors who are at increased risk of early death. Of these easy-to-acquire values, SI and MSI show a comparatively better performance.