Background: Hemothorax is a common sequelae following thoracic trauma and is associated with significant morbidity and mortality. Current guidelines recommend all traumatic hemothoraces be considered for drainage with tube thoracostomy (TT), regardless of size. With increasing use of computed tomography, smaller hemothoraces not seen on x-ray (defined as an occult hemothorax) are frequently detected.
Methods: This systematic review was performed to gather data on patients with occult hemothorax managed with TT or without TT (termed expectant management [EM]). MEDLINE, EMBASE, and Cochrane databases from inception to October 2019 were searched for relevant articles. The primary outcome was rates of failure of expectant (conservative) management. Secondary outcomes of interest included predictors of TT insertion, predictors of failure of EM, and morbidity and mortality in patients with occult hemothorax.
Results: We screened 1,329 abstracts from which 6 articles reporting 1,405 patients with occult hemothorax were included. Of these patients, 601 (43.68%) were managed initially with TT, and 802 (56.32%) were managed expectantly. Of the 802 patients managed expectantly, 212 failed conservative management and underwent TT insertion (23.1% pooled failure rate estimate [95% confidence interval, 17.1-29.1%]). The presence of concomitant pneumothorax predicted upfront TT insertion. Of the patients who failed EM, the need for mechanical ventilation and the presence of a large hemothorax predicted failure. Mortality was similar in both groups.
Conclusions: Conservative treatment of occult hemothorax fails in 23.1% of patients. The presence of hemothorax greater than 300 mL and the need for mechanical ventilation predicted failure of conservative treatment and the need for TT. There was no difference in mortality between EM and TT cohorts. These data suggest that it may be possible to safely observe patients with occult hemothoraces less than 300 mL (1.5 cm pleural stripe) secondary to blunt trauma without upfront TT insertion.
Level of evidence: Systematic review and meta-analysis, level III.