Observation-first versus angioembolization-first approach in stable patients with blunt liver trauma: A WTA multicenter study

Abstract

BACKGROUND Prior studies evaluating observation versus angioembolization (AE) for blunt liver injuries (BLT) with contrast extravasation (CE) on computed tomography imaging have yielded inconsistent conclusions, primarily due to limitations in single-center and/or retrospective study design. Therefore, this multicenter study aims to compare an observation versus AE-first approach for BLT, hypothesizing decreased liver-related complications (LRCs) with observation. METHODS We conducted a post hoc analysis of a multicenter, prospective observational study (2019–2021) across 23 centers. Adult patients with BLT + CE undergoing observation or AE within 8 hours of arrival were included. The primary outcome was LRCs, defined as perihepatic fluid collection, bile leak/biloma, pseudoaneurysm, hepatic necrosis, and/or hepatic abscess. A multivariable logistic regression analysis was used to evaluate risk factors associated with LRCs. RESULTS From 128 patients presenting with BLT + CE on imaging, 71 (55.5%) underwent observation-first and 57 (45.5%) AE-first management. Both groups were comparable in age, vitals, mechanism of injury, and shock index (all p > 0.05), however the AE group had increased frequency of American Association for the Surgery of Trauma Grade IV injuries (51.0% vs. 22.0%, p = 0.002). The AE cohort demonstrated increased rates of in-hospital LRCs (36.8% vs. 12.7%, p = 0.038), emergency department representation (25.0% vs. 10.0%, p = 0.025), and hospital readmission for LRCs (12.3% vs. 1.4%, p = 0.012). However, the two cohorts had similar mortality rates (5.7% vs. 5.3%, p = 0.912). After adjusting for age, ISS, and grade of liver injury, an AE-first approach had a similar associated risk of LRCs compared with observation-first management (odds ratio, 1.949; 95% confidence interval, 0.673–5.643; p = 0.219). CONCLUSION Patients with blunt liver injury and CE undergoing an observation-first approach were associated with a similar adjusted risk of LRCs and rate of mortality compared with AE-first approach. Overall, this calls for reevaluation of the role of routine AE in blunt liver trauma patients with CE. Future prospective randomized trials are needed to confirm these findings. LEVEL OF EVIDENCE Therapeutic/Care Management, Level IV.