In-hospital outcomes in autogenous vein versus synthetic graft interposition for traumatic arterial injury: A propensity-matched cohort from PROOVIT

BACKGROUND: The ideal conduit for traumatic arterial repair is controversial. Autologous vein was compared with synthetic interposition grafts in the acute setting. The primary outcome was in-hospital reoperation or endovascular intervention. METHODS: The American Association for the Surgery of Trauma PROspective Observational Vascular Injury Treatment registry from November 2013 to January 2019 was queried for arterial injuries requiring interposition vein or graft repair. Patients with no recorded Injury Severity Score were excluded, and multiple imputation was used for other missing data. Patients treated with synthetic grafts (SGs) were propensity matched to patients with vein grafts (VGs) to account for preoperative differences. RESULTS: Four hundred sixty from 19 institutions were identified, with 402 undergoing VG and 58 SG. In the SG group, 45 were PTFE grafts, 5 were Dacron, and 8 had other conduits. The SG group was more severely injured at admission with more gunshot wounds and higher mean Injury Severity Score, lactate, and first-24-hour transfusion requirement. In addition, the SG cohort had significantly lower admission systolic blood pressure, pH, and hemoglobin. After propensity matching, 51 patients with SG were matched with 87 patients with VG. There were no differences in demographics, clinical parameters, or diagnostic evaluation techniques postmatch. The need for reoperation or endovascular intervention between the matched groups was equivalent (18%; p = 0.8). Therewas no difference in any secondary outcome including thrombosis, stenosis, pseudoaneurysm, infection, or embolic event, and hospital and intensive care unit length of stay were the same. CONCLUSION: American Association for the Surgery of Trauma PROspective Observational Vascular Injury Treatment registry data demonstrate that SGs are used in more critically ill patients. After controlling for relevant clinical factors and propensitymatching, there is no inhospital difference in rate of reoperation or endovascular intervention, or any secondary outcome between VG and SG.