BACKGROUND: Aortobifemoral bypass (ABF) remains an important treatment modality in the revascularization of aortoiliac occlusive disease. Despite ABF being performed for decades, questions remain regarding the preferred technique for the proximal anastomosis, specifically whether an end-to-end (EE) or an end-to-side (ES) configuration is superior. The goal of this study was to compare the outcomes of ABF based on proximal configuration. METHODS: We queried the Vascular Quality Initiative registry for ABF procedures performed between 2009 and 2020. Univariate and multivariate logistic regression analyses were used to compare perioperative and 1-year outcomes between EE and ES configurations. RESULTS: Of the 6,782 patients (median [interquartile range] age, 60.0 [54-66 years]) who underwent ABF, 3,524 (52%) had an EE proximal anastomosis and 3,258 (48%) had an ES proximal anastomosis. Postoperatively, the ES cohort had a higher frequency of extubation in the operating room (80.3% vs. 77.4%; P < 0.01), lower change in renal function (8.8% vs. 11.5%; P < 0.01), and lower use of vasopressors (15.6% vs. 19.1%; P < 0.01), but higher rates of unanticipated return to the operating room (10.2% vs. 8.7%; P = 0.037) compared with the EE configuration. At 1-year follow-up, the ES cohort had a significantly lower primary graft patency rate (87.5% vs. 90.2%; P < 0.01) and higher rates of graft revision (4.8% vs. 3.1%; P < 0.01) and claudication symptoms (11.6% vs. 9.9%; P < 0.01). The ES configuration was significantly associated with a higher rate of 1-year major limb amputations in univariate (1.6% vs. 0.9%; P < 0.01) and multivariate (odds ratio, 1.95, confidence interval, 1.18-3.23, P=<0.01) analyses. CONCLUSIONS: While the ES cohort seemed to have less physiologic insult immediately postoperatively, the EE configuration appeared to have improved 1-year outcomes. To our knowledge, this study is one of the largest population-based studies comparing the outcomes of the proximal anastomotic configurations. Longer-term follow-up is needed to determine which configuration is optimal.