ABSTRACT
PURPOSE: To evaluate the prognostic factors that influence the survival of patients with traumatic intractable oronasal bleeding treated by transarterial embolization (TAE). MATERIALS AND METHODS: Patients who received TAE for intractable oronasal bleeding in the National Taiwan University Hospital from 2002 through 2013 were included in the study. Retrospective reviews were undertaken to collect relevant clinical and neuroradiologic data that might be correlated with patients' survival. The Wilcoxon rank-sum test or Fisher exact test was adopted to analyze differences between the survival group and the mortality group. Odds ratios were estimated by univariate logistic regression. RESULTS: TAE successfully controlled the bleeding in 24 of 26 patients (92.3%) who had severe craniofacial injury in the 12-year period. Of the 24 patients with successful TAE, 13 patients were discharged alive from the hospital. The overall survival rate was 50% (13 of 26). Significantly higher initial Glasgow Coma Scale (GCS) score (P = .01) and lower Injury Severity Score (ISS; P < .01) were present in the survival group than in the mortality group by the Wilcoxon rank-sum test. Moreover, patients with an ISS of at least 30, a GCS score lower than 9, initial hemoglobin level lower than 10 g/dL, and computed tomographic (CT) findings of a brain midline shift had statistically higher odds ratios predicting mortality than their counterparts as estimated by univariate logistic regression. CONCLUSIONS: The results of this study showed that the combination of diagnostic angiography and therapeutic embolization is effective treatment for intractable oronasal bleeding in patients with severe craniofacial injury. The prognosis in patients who were rescued with successful TAE was statistically correlated with the severity of trauma and concomitant brain injury. An ISS of at least 30, a GCS score lower than 9, an initial hemoglobin level lower than 10, and CT findings of a brain midline shift were strong predictors for mortality.
