Abstract
Background: General surgeons' recent familiarity with advanced laparoscopic techniques have rendered laparoscopy feasible safely in the trauma setting. Traditionally high rates of nontherapeutic laparotomies also contribute to this increased interest. This study was undertaken to determine the predictive value and accuracy of diagnostic laparoscopy (DL) in evaluation of penetrating thoracoabdominal trauma. Methods: Entry criteria included thoracoabdominal gunshot (GSW) or stab wounds (SW) in otherwise hemodynamically stable patients. A high index of suspicion for either hemoperitoneum, peritonitis, or diaphragmatic injury was required for inclusion. All patients underwent DL in the operating room followed by standard laparotomy. The findings of the two evaluations were compared. Results: Twenty-four patients were included in the study. Twenty males and 4 females with an average age of 34 years made up the group. Violation of the peritoneal cavity was present in 21 cases and absent in 3. No intraabdominal injuries were found during laparotomy in the latter three cases without peritoneal violation. The specificity and positive predictive value were 100% for lesions of the diaphragm, liver, spleen, pancreas, kidney, and hollow viscus. The sensitivity was highest for liver and spleen injuries (88%), followed by diaphragmatic injuries (83%), pancreas and kidney injuries (50%), and lowest for injuries of hollow viscus (25%). The negative predictive value was 95, 99, 91, and 57%, respectively, for these organs. Conclusions: DL could have avoided unnecessary laparotomy in 38% of cases in this study. There were no complications related to laparoscopy. The greatest value of DL in penetrating thoracoabdominal injuries is in the evaluation of peritoneal violation, diaphragmatic, and upper abdominal solid-organ injuries. It is not ideal for predicting hollow viscus injuries.