Mechanical Ventilation Does Not Predict Pneumothorax Observation Failure in the Severely Injured

Abstract

Background Observative management of small traumatic pneumothoraces (PTX) has been shown to decrease chest tube utilization in non-mechanically ventilated patients without compromising outcomes. This approach could be used in mechanically ventilated (MV) patients, though many feel these patients are at increased risk of observation failure. Methods A single center retrospective study of all adults undergoing observation of a computed tomography (CT) diagnosed PTX from 2015-2019. Patients with chest tube placement within 4-hours of arrival, concurrent hemothorax, or death within 24-hours were excluded. Observation failure was defined as chest tube placement. Results Of 340 patients, 64 were on MV. The groups were of similar age, BMI, underlying pulmonary comorbidities, and PTX size (10.1 mm vs 8.8 mm, P = .20). The MV group was more severely injured (ISS [25+] [60.9% vs 11.2%, P < .001]). There was no difference in observation failure rates by MV status overall (6.3% vs 5.1%, P = .75) or by PTX size (<15 mm [5% vs 2.2%, P = .37], <20 mm [4.8% vs 3.1%, P = .45], <25 mm [4.8% vs 4.1%, P = .73], <30 mm [4.8% vs 4.1%, P = .73], <35 mm [4.8% vs 4.7%, P = 1.00]). MV was not an independent predictor of observation failure on multivariable analysis (OR .64, 95% CI .18-2.20), though PTX size was (OR 1.11, 95% CI 1.05-1.17). When comparing those who failed vs those who did not, the only difference was PTX size (9.34 mm vs 19.41 mm, P < .001). Conclusion MV is not an independent predictor of PTX observation failure. While PTX size appears to play a role, further studies are needed to outline safe parameters for observation in those undergoing MV.