Objective: Arterial involvement in Behçet disease (BD) is rare, and its surgical management is a major concern because of its high recurrence rate. This study evaluated the influence of the surgical technique, device, and immunosuppressive treatment used on the postoperative recurrence in patients with non-pulmonary arterial BD. Methods: A single-center, retrospective study was conducted of 23 patients meeting the international criteria for BD who underwent surgery for arterial involvement between May 1996 and September 2015. Recurrence was defined as the occurrence of arterial aneurysm or thrombosis during follow-up. Perioperative medical treatment and surgical technique used were reported. Results: There were 47 surgical procedures performed in 23 patients. Mean follow-up was 8.4 ± 7.5 years. Initial arterial lesions were aneurysms and thrombosis in 85% and 15% of cases, respectively. Arterial lesions were aortic and peripheral in 48% and 52% of cases. Recurrence rate was 51%. Recurrences developed within <1 year in 24% of cases and at the same anatomic site in 92% of cases. Among the 24 recurrences, 17 were false aneurysms, 6 were thrombosis, and 1 was a true aneurysm in a different arterial site. To treat the arterial lesion, direct anastomosis was performed in 6 cases; bypass using the saphenous vein, graft, or allograft was performed in 6, 27, and 5 cases, respectively; and stent graft was used in 3 cases. Vascular lesions involved the aorta in 19 cases and a peripheral artery in 28 cases. Preoperative medical treatments, including colchicine, steroids, and immunosuppressants, significantly decreased recurrence rate: 28% (7/25) vs 75% (15/20) in untreated patients (P = .002). The recurrence rate was 42.5% (17/40) in patients treated postoperatively vs 80% (4/5) in untreated patients. The nature of the device used (vein, prosthetic graft, allograft, stent graft, or direct anastomosis) did not change the risk of recurrence. When anastomoses were protected using the prosthetic sleeving technique, the recurrence rate was three times lower (P = .08). Conclusions: Relapse is a main concern after surgical repair of arterial BD. This study suggests the need for targeted perioperative medical management to reduce the risk of arterial recurrence in BD patients. To this end, a multidisciplinary approach is mandatory. The use of sleeve anastomosis is associated with a numerically lower risk of recurrence. However, further studies are needed to confirm this efficacy.