Multicenter retrospective study of noncompressible torso hemorrhage: Anatomic locations of bleeding and comparison of endovascular versus open approach

BACKGROUND Rational development of technology for rapid control of noncompressible torso hemorrhage (NCTH) requires detailed understanding of what is bleeding. Our objectives were to describe the anatomic location of truncal bleeding in patients presenting with NCTH and compare endovascular (ENDO) management versus open (OPEN) management. METHODS This is a retrospective study of adult trauma patients with NCTH admitted to four urban Level I trauma centers in the Houston and San Antonio metropolitan areas in 2008 to 2012. Inclusion criteria include named axial torso vessel disruption, Abbreviated Injury Scale chest or abdomen score of 3 or higher with shock (base excess, <-4) or truncal operation in 90 minutes or less, or pelvic fracture with ring disruption. Exclusion criteria include isolated hip fractures, falls from standing, or prehospital cardiopulmonary resuscitation. After dichotomizing into OPEN, ENDO, and resuscitative thoracotomy (RT) groups based on the initial approach to control NCTH, a mixed-effects Poisson regression with robust error variance (controlling for age, mechanism, Injury Severity Score, shock, hypotension, and severe head injury as fixed effects and site as a random effect) was used to test the hypothesis that ENDO was associated with reduced in-hospital mortality in NCTH patients. RESULTS Five hundred forty-three patients with NCTH underwent ENDO (n = 166, 31%), OPEN (n = 309, 57%), or RT (n = 68, 12%). Anatomic bleeding locations were 25% chest, 41% abdomen, and 31% pelvis. ENDO was used to treat relatively few types of vascular injuries, whereas OPEN and RT injuries were more diverse. ENDO patients had more blunt trauma (95% vs. 34% vs. 32%); severe injuries (median Injury Severity Score, 34 vs. 27 vs. 21), and increased time to intervention (median, 298 vs. 92 vs. 51 minutes) compared with OPEN and RT. Mortality was 15% versus 20% versus 79%. ENDO was associated with decreased mortality compared to OPEN (relative risk, 0.58; 95% confidence interval, 0.46-0.73). CONCLUSION Although ENDO may reduce mortality in NCTH patients, significant group differences limit the generalizability of this finding. LEVEL OF EVIDENCE Therapeutic, level V.