Lung ultrasonography underdiagnoses clinically significant pneumothorax

Abstract

BackgroundUltrasonography for trauma is an integral part of the Advanced Trauma Life Support algorithm and supported extensively in the literature. The reliability of chest ultrasonography as a screening examination for pneumothorax during initial trauma evaluation is unclear. We performed a prospective study where we hypothesized that chest ultrasonography would have low sensitivity for detecting clinically significant pneumothorax. MethodsA prospective observational analysis of patients with blunt chest trauma at a level 1 trauma center was performed. Patients included had supine chest radiography and chest ultrasonography performed prior to intervention as well as confirmatory computed tomographic imaging. All chest ultrasonography was performed in the trauma bay by a registered sonographer. All imaging was evaluated by an attending trauma surgeon and radiologist in real time. ResultsOf 2,185 patients screened with a diagnosis of blunt thoracic trauma, 1,489 patients had chest radiography, chest ultrasonography, and confirmatory computed tomography and were included for analysis. Patients were 71% male, with median age of 42 years, and mean Injury Severity Score of 6. The sensitivity of chest ultrasonography to detect pneumothorax was low. Chest ultrasonography had a false negative rate of 72% (n = 58), with 22% (n = 13) undergoing tube thoracostomy. Patients with false negative examinations had lower initial O2 saturation and systolic blood pressure and were more likely to have rib fractures compared with true negative chest ultrasonography examinations. ConclusionChest ultrasonography performed on initial trauma evaluation has low sensitivity with a high rate of false negative examinations. Because many of these false negative results are clinically significant requiring thoracostomy, using chest ultrasonography alone to screen for pneumothorax should be done with caution.