Diagnostic accuracy of computed tomography findings for hollow viscus injuries following thoracoabdominal gunshot wounds.

Abstract

Selective nonoperative management (SNOM) of abdominal gunshot wounds (GSWs) is increasingly used as computed tomography (CT) has become a diagnostic adjunct for the evaluation of intraabdominal injuries including hollow viscus injuries (HVIs). Currently, there is scarce data on the diagnostic accuracy of CT for identifying HVI. The purpose of this study was to determine the diagnostic accuracy of different CT findings in the diagnosis of HVI following abdominal GSW. This retrospective single-center cohort study was performed from January 2015 to April 2019. We included consecutive patients (≥18 years) with abdominal GSW for whom SNOM was attempted and an abdominal CT was obtained as a part of SNOM. Computed tomography findings including abdominal free fluid, diffuse abdominal free air, focal gastrointestinal wall thickness, wall irregularity, abnormal wall enhancement, fat stranding, and mural defect were used as our index tests. Outcomes were determined by the presence of HVI during laparotomy and test performance characteristics were analyzed. Among the 212 patients included for final analysis (median age: 28 years), 43 patients (20.3%) underwent a laparotomy with HVI confirmed intraoperatively whereas 169 patients (79.7%) were characterized as not having HVI. The sensitivity of abdominal free fluid was 100% (95% confidence interval [CI]: 92-100). The finding of a mural defect had a high specificity (99%, 95% CI: 97-100). Other findings with high specificity were abnormal wall enhancement (97%, 95% CI: 93-99) and wall irregularity (96%, 95% CI: 92-99). While there was no singular CT finding that confirmed the diagnosis of HVI following abdominal GSW, the absence of intraabdominal free fluid could be used to rule out HVI. In addition, the presence of a mural defect, abnormal wall enhancement, or wall irregularity is considered as a strong predictor of HVI. Diagnostic Test or Criteria; Level II.