Prospective evaluation of the sensitivity of physical examination in chest trauma

Abstract

Background: Chest radiographs are routine for patients presenting with blunt and penetrating chest trauma. The accuracy of physical examination in the diagnosis of hemopneumothorax in these patients is unclear. A prospective study was performed to define the utility of routine portable chest radiographs in 676 trauma patients.

Methods: Over 19 months (January 2000-July 2001), 676 patients who presented with penetrating or blunt chest trauma were interviewed and examined for signs and symptoms of hemopneumothorax. The incidence of chest pain or tenderness and tachypnea was noted and both lung fields were auscultated. A portable chest radiograph was then performed on all the patients.

Results: All the patients were hemodynamically stable. Five hundred twenty-three patients sustained blunt trauma, with seven hemopneumothoraces (1.3%). The negative predictive values of auscultation, pain or tenderness, and tachypnea were 99& to 100%. One hundred fifty-three patients sustained penetrating chest trauma. Of these injuries, 68 were gunshot wounds and 85 were stab wounds. Twenty-four (16%) of these patients had hemopneumothoraces. The sensitivities of auscultation, pain or tenderness, and tachypnea were 50%, 25%, and 32%, respectively. The negative predictive values of these tests were < 91%.

Conclusion: Blunt chest trauma patients who are hemodynamically stable with a normal physical examination do not require a routine chest radiograph. In contrast, all victims of penetrating trauma require chest radiographs because many will have hemopneumothorax in the absence of clinical findings.