Predictors of Outcome in Infrainguinal Bypass for Trauma

Abstract

Trauma remains a leading cause of morbidity and mortality worldwide, with vascular injuries present in 1% to 2% and the majority of injuries occurring to the extremities. This study aimed to determine predictors of poor outcome in infrainguinal bypasses performed for traumatic arterial injury. Patients admitted between September 1999 and July 2015 who underwent infrainguinal arterial bypass for trauma at a single level one trauma center were included for analysis. Poor outcome was defined as a composite of thrombosis leading to graft abandonment, reoperation, major amputation, or death. Data were analyzed by stratified Kaplan-Meier and multiple Cox regression analysis using SAS version 9.4 (SAS Institute, Cary, NC). During the study period, 108 patients had infrainguinal arterial bypass for traumatic arterial injury, 45 end-to-side and 63 interposition bypasses. Mean age was 35 ± 17 years, and 17 of 108 (16%) were female. The median Injury Severity Score was 12 (interquartile range [IQR], 9-18); admission glomerular filtration rate was 77.5 (IQR, 59-92); median Mangled Extremity Severity Score (MESS) was 6 (IQR, 5-7); median injury to surgery time was 5.1 hours (IQR, 2-24); 39 of 108 (36%) had blunt injury; 30 of 108 (28%) had crush injury; 39 of 108 (36%) had motor vehicle collision; 10 of 108 (10%) had diabetes mellitus; and 76 of 108 (70%) had an infrageniculate target for bypass. Univariate risk factors for poor outcome included age >40 years (48% vs 21%; P < .01), MESS >7 (51% vs 14%; P < .01), blunt mechanism (39% vs 15%; P < .01), diabetes (70% vs 27%; P < .01), and infrageniculate target vessel (37% vs 16%; P < .03). Popliteal artery injury (38% vs 21%; P < .06) and concomitant orthopedic injuries (35% vs 17%; P < .06) had worse outcomes yet were not statistically significant. Baseline renal function, injury to surgery time, surgeon's specialty, and associated venous injuries were not significantly predictive of poor outcome. MESS was strongly predictive of poor outcome, with probability rising as high as 95% when MESS reached 12 (Fig). A MESS score ≥7 was 73% sensitive and 70% specific for prediction of poor outcomes. Age (odds ratio, 1.03/year; P < .04) and MESS >7 (odds ratio, 3.8; P < .02) were persistent predictors of poor outcome in multivariable analysis. In stratified analysis, interposition grafts fared worse only in penetrating injuries, with composite poor outcome in 0 of 13 (0%) in end-to-side bypass vs 6 of 26 (23%) in interposition bypass (relative risk, 1.3; 95% confidence interval, 1.1-1.6; P < .03). Poor outcome in infrainguinal bypass for trauma is predicted by age >40 years, MESS >7, blunt mechanism of injury, infrageniculate target, and history of diabetes mellitus. Interestingly, interposition bypasses had worse outcomes compared with end-to-side bypasses only in penetrating injures.