Chest ultrasonography for the emergency diagnosis of traumatic pneumothorax and haemothorax: A systematic review and meta-analysis

Objective: To assess the accuracy of the chest ultrasonography for the emergency diagnosis of traumatic pneumothorax and haemothorax in adults. Study design: Systematic review and meta-analysis. Methods: PubMed, EMBASE, Scopus, Web of Science and LILACS (up to 2016) were systematically searched for prospective studies on the diagnostic accuracy of ultrasonography for pneumothorax and haemothorax in adult trauma patients. The references of other systematic reviews and the included studies were checked for further articles. The characteristics and results of the studies were extracted using a standardised form, and their methodological quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Primary analysis was performed considering each hemithorax as an independent unit, while secondary analysis considered each patient. The global diagnostic accuracy of the chest ultrasonography was estimated using the Rutter-Gatsonis hierarchical summary ROC method. Moreover, Reitsma's bivariate model was used to estimate the sensitivity, specificity, positive likelihood ratio (LR +) and negative likelihood ratio (LR-) of each sonographic sign. This review was previously registered (PROSPERO CRD42016048085). Results: Nineteen studies were included in the review, 17 assessing pneumothorax and 5 assessing haemothorax. The reference standard was always chest tomography, alone or in parallel with chest radiography and observation of the chest tube. The overall methodological quality of the studies was low. The diagnostic accuracy of chest ultrasonography had an area under the curve (AUC) of 0.979 for pneumothorax (Fig). The absence of lung sliding and comet-tail artefacts was the most reported sonographic sign of pneumothorax, with a sensitivity of 0.81 (95% confidence interval [95% CI], 0.71-0.88), specificity of 0.98 (95% CI, 0.97-0.99), LR+ of 67.9 (95% CI, 26.3-148) and LR- of 0.18 (95% CI, 0.11-0.29). An echo-poor or anechoic area in the pleural space was the only sonographic sign for haemothorax, with a sensitivity of 0.60 (95% CI, 0.31-0.86), specificity of 0.98 (95% CI, 0.94-0.99), LR+ of 37.5 (95% CI, 5.26-207.5), LR-of 0.40 (95% CI, 0.17-0.72) and AUC of 0.953. Conclusion: Notwithstanding the limitations of the included studies, this systematic review and meta-analysis suggested that chest ultrasonography is an accurate tool for the diagnostic assessment of traumatic pneumothorax and haemothorax in adults. (C) 2018 Elsevier Ltd. All rights reserved.