Outcomes of Upper Extremity Versus Transfemoral Access for Fenestrated-Branched Endovascular Aortic Repair

Abstract

Upper extremity (UE) access is frequently used for fenestrated-branched endovascular aortic repair (FB-EVAR). The advent of steerable sheaths has enabled FB-EVAR to be performed using a total transfemoral (TF) approach without UE access, potentially decreasing the risks of cerebral embolic events. The purpose of this study was to assess the outcomes of FB-EVAR using UE vs TF access. Prospectively collected data from nine physician-sponsored investigational device exemption studies at U.S. centers were analyzed using a standardized database. All patients had been treated for complex abdominal aortic aneurysms (CAAAs) and thoracoabdominal aortic aneurysms (TAAAs) using manufactured fenestrated and branched stent-grafts between 2005 and 2020. The outcomes were compared between patients undergoing UE vs TF access. The primary composite outcome was stroke and transient ischemia attack (TIA) during the perioperative period. The secondary outcomes included technical success, local access-related complications, and perioperative mortality. A total of 1681 patients (71% men; mean age, 73.43 ± 7.8 years) had undergone FB-EVAR for 502 CAAAs (30%), 535 extent IV TAAAs (32%), and 644 extent I to III TAAAs (38%). UE access was used in 1103 patients (67%). The right side was used in 395 patients (24%) and the left side in 705 patients (42%). UE access was preferentially used for TAAAs (74% vs 47%; P < .001), and TF access was used more frequently for CAAAs (53% vs 26%; P < .01). A total of 38 perioperative cerebrovascular events (2.5%), 32 strokes (1.9%), and 6 TIAs (0.4%) had occurred. Perioperative cerebrovascular events had occurred more frequently with UE access than with TF access (2.8% vs 1.2%; P = .036). An individual component analysis of the primary outcome revealed a trend for a greater incidence of stroke (2.3% vs 1.2%; P = .13) and TIA (0.54% vs 0%; P = .10) in the UE access group. Total TF access was associated with a 60% reduction in perioperative cerebrovascular events on multivariate analysis (odds ratio, 0.39; P = .029). No significant differences were observed between UE and TF access in technical success (96.5% vs 96.8%; P = .72), perioperative mortality (2.9% vs 2.6% P = .72), or local access-related complications (6.5% vs 5.5%; P = .43). A total transfemoral approach for F-BEVAR was associated with a lower rate of perioperative cerebrovascular events compared with UE access. Although the cerebrovascular event rate was low with UE access for appropriately selected patients, a total femoral approach offered a lower risk of stroke and TIA. UE access might be justified, however, for more complex repairs.