Does pre-injury anticoagulation make chest tubes any less safe? A nationwide retrospective analysis

Abstract

BackgroundOral anticoagulation is becoming more common with the aging population, which raises concern for the risk of invasive procedures that can cause bleeding, such as chest tube placement (thoracostomy). With the increase in CT imaging, more pneumothoraces and hemothoraces are being identified. The relative risk of thoracostomy in the presence of anticoagulation is not well-established. The objective of this study was to determine whether pre-injury anticoagulation affects the relative risk of tube thoracostomy following significant chest trauma. MethodsThis retrospective cohort study used data from the 2019 American College of Surgeons-Trauma Quality Program (ACS-TQP) database using R version 4.2.2. Data from the database was filtered based on inclusion and exclusion criteria. Outcomes were then assessed with the population of interest. Demographics, vitals, comorbidities, and injury parameters were also collected for each patient.This study included all adult patients (≥ 18 years) presenting with traumatic hemothorax, pneumothorax, or hemopneumothorax. Patients with missing data in demographics, vitals, comorbidities, injury parameters, or outcomes, as well as those with no signs of life upon arrival, were excluded from the study. Patients were stratified into groups based on whether they had pre-injury anticoagulation and whether they had a chest tube placed in the hospital. The primary outcome was mortality, and the secondary outcome was hospital length of stay (LOS). Logistic and standard regressions were used by a statistician to control for age, sex, and Injury Severity Score (ISS). ResultsOur study population included 72,385 patients (4250 with pre-injury anticoagulation and 68,135 without pre-injury anticoagulation). Pre-injury anticoagulation and thoracostomy were each independently associated with increased mortality and LOS. However, there was a non-significant interaction term between pre-injury anticoagulation and thoracostomy for both outcomes, indicating that their combined effects on mortality and LOS did not differ significantly from the sum of their individual effects. ConclusionThis study suggests that both pre-injury anticoagulation and thoracostomy are risk factors for mortality and increased LOS in adult patients presenting with hemothorax, pneumothorax, or hemopneumothorax, but they do not interact with each other. We recommend further study of this phenomenon to potentially improve clinical guidelines. Level of evidenceTherapeutic, Level III.