Triple-contrast helical CT in penetrating torso trauma: a prospective study to determine peritoneal violation and the need for laparotomy.

Abstract

A prospective study was performed to determine the usefulness of triple-contrast helical CT in predicting peritoneal violation and the need for laparotomy in the treatment of penetrating torso trauma. Triple-contrast helical CT scans were obtained in 104 hemodynamically stable patients with penetrating injuries to the torso (thoracoabdominal region including tangential wounds to the anterior abdomen, flank, back, and pelvis) over a 17-month period. The study group included 54 patients with gunshot wounds and 50 with stab wounds. No patient had a radiographic or clinical indication for immediate laparotomy. A positive finding on CT was defined as evidence of peritoneal violation or injury to the retroperitoneal colon, major vessel, or urinary tract. Patients with a positive CT, except patients with isolated liver injury or free fluid, underwent laparotomy. Patients with a negative finding on CT were initially observed. CT studies were positive in 35 (34%) of 104 patients and negative in 69 (66%) of 104 of patients. Laparotomy was performed in 21 (60%) of 35 patients with positive CT; 19 (86%) of 22 were therapeutic, two (9%) were nontherapeutic, and one (5%) was negative (no injury was found). Nine patients with isolated hepatic injuries were successfully treated without laparotomy. Among patients with a negative CT, 67 (97%) of 69 were treated nonoperatively with success. CT had 100% (19/19) sensitivity, 96% (69/72) specificity, 100% (69/69) negative predictive value, and 97% (101/104) accuracy in predicting the need for laparotomy. Triple-contrast helical CT can accurately predict the need for laparotomy and exclude peritoneal violation in penetrating torso trauma including tangential abdominal wounds.