Effect of temporary shunting on extremity vascular injury: An outcome analysis from the Global War on Terror vascular injury initiative

Abstract

Extremity vascular injury during the current war has been defined by anecdotal description and case series. These reports focused on estimation of short-term limb viability and technical description of commonly used adjuncts. Temporary vascular shunting (TVS) has been advocated in current care structures, yet mostly due to war environments, broader statistical scrutiny is lacking. This study's purpose is to provide perspective on TVS's impact on limb salvage, and estimate longer-term freedom from amputation. Data from the Joint Theater Trauma Registry (JTTR), Balad Vascular Registry (BVR), Walter Reed Vascular Registry (WRVR), electronic medical records, and patient interviews were collected on American Troops sustaining extremity vascular injury from June 2003 through December 2007. Those in whom arterial TVS utilization was identified comprise the TVS group. These were compared with controls with similar injury date and anatomic location managed without TVS. Descriptive statistics were employed establishing overall univariate predictors of amputation and comparison between groups. Proportional-hazards modeling, with propensity score adjustment for systemic injury severity and Level 2 care, characterized risk factors of limb loss and effect of TVS. Freedom from amputation was estimated using Kaplan Meier log-rank methods. Cases and controls consisted of 64 and 61 extremity arterial injuries, respectively. Mean follow-up was 22 months (range: 1-54 months). The TVS group was more severely injured (mean injury severity score [ISS]: 18 [SD = 10] TVS vs. 15 [SD = 10] control, P = .05) and more likely to receive Level 2 care (TVS: 26%; control: 10%, P = .02). Overall, a total of 26 amputations occurred (21%). Penetrating blasts, compared with gunshot wounds, were associated with amputation (30% vs. 6%, P = .002). After propensity score adjustment, use of TVS suggested a reduced risk of amputation (relative risk [RR] = 0.47; 95% confidence interval [CI] [0.18-1.19]; P = .11). Venous repair was associated with limb salvage (RR = 0.2; 95% CI [0.04-0.99], P = .05). Associated fracture (RR = 5.0; 95% CI [1.45-17.28], P = .01), and elevated mangled extremity severity score (MESS) ([MESS 5-7] RR = 3.5, 95% CI [0.97-12.36], P = .06; [MESS 8-12] RR = 16.4; 95% CI (3.79-70.79), P < .001) predicted amputation. Amputation-free survival was 78% in the TVS group and 77% in the control group at three years (P = .5). Temporary vascular shunting used as a damage control adjunct in management of wartime extremity vascular injury does not lead to worse outcomes. Benefit from TVS is suggested, but not statistically significant. Injury specific variables of venous ligation, associated fracture, and penetrating blast mechanism are associated with amputation. Amputation-free survival after vascular injury in Operation Iraqi Freedom is 79% at three years. Further studies to statistically define any possible benefits of TVS are needed.