Background: As more pneumothoraxes (PTX) are being identified on chest computed tomography (CT), the empiric trigger for tube thoracostomy (TT) versus observation remains unclear. We hypothesized that PTX measuring 35 mm or less on chest CT can be safely observed in both penetrating and blunt trauma mechanisms.
Methods: A retrospective review was conducted of all patients diagnosed with PTX by chest CT between January 2011 and December 2016. Patients were excluded if they had an associated hemothorax, an immediate TT (TT placed before the initial chest CT), or if they were on mechanical ventilation. Size of PTX was quantified by measuring the radial distance between the parietal and visceral pleura/mediastinum in a line perpendicular to the chest wall on axial imaging of the largest air pocket. Based on previous work, a cutoff of 35 mm on the initial CT was used to dichotomize the groups. Failure of observation was defined as the need for a delayed TT during the first week. A univariate analysis was performed to identify predictors of failure in both groups, and multivariate analysis was constructed to assess the independent impact of PTX measurement on the failure of observation while controlling for demographics and chest injuries.
Results: Of the 1,767 chest trauma patients screened, 832 (47%) had PTX, and of those meeting inclusion criteria, 257 (89.0%) were successfully observed until discharge. Of those successfully observed, 247 (96%) patients had a measurement of 35 mm or less. The positive predictive value for 35 mm as a cutoff was 90.8% to predict successful observation. In the univariant analyses, rib fractures (p = 0.048), Glasgow Coma Scale (p = 0.012), and size of the PTX (≤35 mm or >35 mm) (P < 0.0001) were associated with failed observation. In multivariate analysis, PTX measuring 35 mm or less was an independent predictor of successful observation (odds ratio, 0.142; 95% confidence interval, 0.047-0.428)] for the combined blunt and penetrating trauma patients.
Conclusion: A 35-mm cutoff is safe as a general guide with only 9% of stable patients failing initial observation regardless of mechanism.
Level of evidence: Therapeutic, level III.