Size of traumatic pneumothorax on initial chest x-ray is independently associated with failed observation in children

Traumatic pneumothorax (tPTX) in children is typically treated with tube thoracostomy (TT). We aimed to determine if the size of tPTX on initial chest x-ray (CXR) was associated with a need for TT and failure of observation without TT.

We performed a single-institution retrospective review of children (18 years or younger) presenting to a Level 1 pediatric trauma center between 2010 and 2023 with tPTX, excluding children without CXR and those transferred after TT. Observation was defined as progression in care with known tPTX but without TT; failed observation was defined as TT after initial observation. The volumetric Collins method was used to estimate size of tPTX on CXR. Area under the receiver operating characteristic curve (AUROC), Youden's index, and multivariable regression analyses were performed.

There were 313 children with 358 instances of tPTX; of those undergoing observation, 45 (13.5%) failed. Those who failed observation had larger tPTX (14.1% vs. 6.5%, p < 0.001) and more frequently were hypotensive on arrival (26.7% vs. 10.2%, p = 0.006), had hemopneumothorax (22.2% vs. 3.5%, p < 0.001), received supplemental oxygenation (nasal cannula, 33.3% vs. 11.1%; nonrebreather, 37.8% vs. 58.5%; ventilated, 15.6% vs. 10.7%; overall p < 0.001), and were admitted to the pediatric intensive care unit (47.4% vs. 30.6%, p = 0.008). Optimal size predicting need for TT placement on CXR was 12.5% (AUROC 0.715). On multivariable regression controlling for tPTX size ≥12.5%, mechanical ventilation, hemopneumothorax, and hypotension, only size ≥12.5% was associated with failure of observation.

In children observed with tPTX, Collins size of ≥12.5% on CXR was independently associated with failure. In children with smaller tPTX, prophylactic TT placement may not be necessary, and observation should be considered.

Therapeutic/Care Management; Level III.