Predictive ability of the ISS, NISS, and APACHE II score for SIRS and sepsis in polytrauma patients

Abstract

Systemic inflammatory response syndrome (SIRS) and sepsis as causes of multiple organ dysfunction syndrome (MODS) remain challenging to treat in polytrauma patients. In this study, the focus was set on widely used scoring systems to assess their diagnostic quality. A total of 512 patients (mean age: 39.2±16.2, range: 16-88years) who had an Injury Severity Score (ISS) ≥17 were included in this retrospective study. The patients were subdivided into four groups: no SIRS, slight SIRS, severe SIRS, and sepsis. The ISS, New Injury Severity Score (NISS), Acute Physiology and Chronic Health Evaluation II (APACHE II) scores, and prothrombin time were collected at admission. The Kruskal-Wallis test and χ(2)-test, multinomial regression analysis, and kernel density estimates were performed. Receiver operating characteristic (ROC) analysis is reported as the area under the curve (AUC). Data were considered as significant if p<0.05. All variables were significantly different in all groups (p<0.001). The odds ratio increased with increasing SIRS severity for NISS (slight vs. no SIRS, 1.06, p=0.07; severe vs. no SIRS, 1.07, p=0.04; and sepsis vs. no SIRS, 1.11, p=0.0028) and APACHE II score (slight vs. no SIRS, 0.97, p=0.44; severe vs. no SIRS, 1.08, p=0.02; and sepsis vs. no SIRS, 1.12, p=0.0028). ROC analysis revealed that the NISS (slight vs. no SIRS, AUC 0.61; severe vs. no SIRS, AUC 0.67; and sepsis vs. no SIRS, AUC 0.77) and APACHE II score (slight vs. no SIRS, AUC 0.60; severe vs. no SIRS, AUC 0.74; and sepsis vs. no SIRS, AUC 0.82) had the best predictive ability for SIRS and sepsis. Quick assessment with the NISS or APACHE II score could preselect possible candidates for sepsis following polytrauma and provide guidance in trauma surgeons' decision-making.