Diagnostic Accuracy of the eFAST Protocol in Trauma Patients Stratified by Severity according to the Injury Severity Score: A Multicenter Retrospective Study

Background: Traumatic injuries represent a leading cause of death and disability. Although contrast-enhanced Computed Tomography (CT) is the gold standard for trauma diagnosis, the Extended Focused Assessment with Sonography in Trauma (eFAST) protocol remains an integral component of the primary evaluation. The diagnostic utility of this technique has been well described in the literature, especially for its high specificity, but current evidence on its accuracy stratified by the severity of trauma remains limited. Objectives: The primary aim of this study was to evaluate whether eFAST diagnostic accuracy varies according to the severity of the trauma, stratified using the Injury Severity Score (ISS). Secondarily, the impact of the eFAST protocol in the evaluation of hemodynamically unstable patients and the correlation between eFAST-CT concordance and in-hospital mortality were investigated. Methods: A multicenter, retrospective study was conducted, drawing on data from the Baggiovara, Ferrara, and Cesena trauma centers, from January 2023 to December 2024. Adult patients presenting to the Emergency Department with a Code Red and a triage diagnosis of polytrauma or major trauma were included, provided both eFAST and CT scan were performed. The accuracy of the eFAST, stratifying patients into Mild, Moderate, and Severe groups based on the ISS, was evaluated by comparing its results with the contrast-enhanced CT scans as the gold standard, specifically assessing the ability of ultrasound to detect hemopericardium, hemoperitoneum, hemothorax, and pneumothorax. Hemodynamic instability, pertaining to the first secondary outcome, was defined as a Shock Index ≥ 1. Results: Of the 1,508 patients evaluated, 876 met the inclusion criteria. Results demonstrated an overall excellent specificity (95.0-100%) of the eFAST protocol, while sensitivity in the general population was moderate, ranging from 16.7% for hemothorax to 48.4% for perihepatic hemoperitoneum. Moreover, the study showed that as severity increased, eFAST sensitivity did not differ significantly among severity groups in the thorax and upper abdomen (p > 0.05), with the exception of the pelvis (p = 0.0007). Furthermore, the eFAST protocol demonstrated enhanced sensitivity in hemodynamically unstable patients within the Morison’s pouch and the pelvis, while pneumothorax detection sensitivity decreased in this cohort. No statistically significant correlation was found between in-hospital mortality and eFAST-CT concordance. Conclusions: This study confirmed the rule-in power of eFAST, demonstrating high specificity and therefore representing a useful method to rapidly guide clinical interventions in the event of a positive result. The moderate sensitivity and the number of false negatives, however, preclude its use as a rule-out tool, confirming the importance of CT imaging integration. The stratification by severity demonstrated no statistically significant effect on sensitivity, apart from the pelvic window. Specifically, the clinical severity and technical limitations of critically ill patients impaired ultrasound performance, thereby counterbalancing the greater injury burden across most anatomical regions. Future prospective studies with certified operators are needed to refine accuracy estimates and further characterize the relationship between injury severity and eFAST diagnostic performance.