LAUSR

OBJECTIVE Blunt carotid artery injury (BCI) is present in approximately 1.0% to 2.7% of all blunt trauma admissions and can result in significant morbidity and mortality. Management ranges from antithrombotic therapy alone to surgery, where potential indications include pseudoaneurysm, failed or contraindication to medical therapy, and progression of neurologic symptoms. Still, optimal management, including approach and timing, continues to be an active area for debate. The goal of this study was to assess the epidemiologic characteristics of BCI, and, after controlling for presenting features intrinsic to the data, compare outcomes based on management, operative approach, and timing of intervention. METHODS A retrospective review was conducted of adult BCI patients identified within the National Trauma Data Bank from 2002 to 2016. The National Trauma Data Bank is the largest trauma database in the United States, collating data from each trauma admission for more than 900 trauma centers. Independent variables of interest included nonoperative versus operative management (OM); endovascular versus open intervention, and early (within 24 hours) versus delayed (after 24 hours) intervention. For each independent variable, groups were compared after propensity score matching to control for presenting factors and patterns of injury. RESULTS There were 9190 patients who met the inclusion criteria, 812 of whom underwent operative intervention (open, n = 288; endovascular, n = 481, both: n = 43). During the review, there was no difference in proportion of OM over time, although there was a statistically significant decrease in the proportion of open intervention (0.48% per year; P < .05). For outcomes, operative versus nonoperative management (nOM) resulted in no difference in mortality, but the operative group demonstrated an increased risk of stroke (11.8% vs 6.5%), longer hospital and intensive care length of stay, and more days on mechanical ventilation (P < .001 for each). With regard to timing: mortality was increased for early intervention (early, 16% vs delayed, 6.3%; P < .001), which was predominantly driven by the endovascular cohort (early, 19.2% vs delayed, 2.5%; P < .001). CONCLUSIONS In this study, there was no significant trend in the overall volume of operative or nOM; however, when considering approach to OM, there was a significant decrease in open procedures. Consistent with previous literature, injury to the neck, head, and chest was significant associated with BCI. Also outcomes demonstrated an increased prevalence of stroke after operative relative to nOM. Importantly, after critically assessing the timing to intervention, results strongly suggested that, if possible, intervention should be delayed for at least 24 hours.