Comparison of open and endovascular left subclavian artery revascularization for zone 2 thoracic endovascular aortic repair

Objective In patients undergoing elective thoracic endovascular aortic repair (TEVAR) and left subclavian artery (LSA) coverage, routine preoperative LSA revascularization is recommended. However, in the current endovascular era, the optimal surgical approach is debated. We compared baseline characteristics, procedural details, and perioperative outcomes of patients undergoing open or endovascular LSA revascularization in the setting of TEVAR. Methods Adult patients undergoing TEVAR with zone 2 proximal landing and LSA revascularization between 2013-2023 were identified in the Vascular Quality Initiative. We excluded patients with traumatic aortic injury, aortic thrombus, or ruptured presentations, and stratified based on revascularization type (open vs. any endovascular). Open LSA revascularization included surgical bypass or transposition. Endovascular LSA revascularization included single-branch, fenestration, or parallel stent grafting. Primary outcomes were stroke, spinal cord ischemia, and perioperative mortality (Pearson's χ2-test). Multivariable logistic regression was used to evaluate associations between revascularization type and primary outcomes. Secondarily, we studied other in-hospital complications and 5-year mortality (Kaplan-Meier, multivariable Cox-regression). Sensitivity analysis was performed in patients undergoing concomitant LSA revascularization to TEVAR. Results Of 2,489 patients, 1,842 (74%) underwent open and 647 (26%) received endovascular LSA revascularization. Demographics and comorbidities were similar between open and endovascular cohorts. Compared with open, endovascular revascularization had shorter procedure times (median 135 vs. 174min, p<.001), longer fluoroscopy time (median 23 vs. 16min, p<.001), lower estimated blood loss (median 100 vs. 123ml, p<.001), and less preoperative spinal drain use (40% vs. 49%, p<.001). Patients undergoing endovascular revascularization were more likely to present urgently (24% vs. 19%) or emergently (7.4% vs. 3.4%) (p<.001). Compared with open, endovascular patients experienced lower stroke rates (2.6% vs. 4.8%, p=.026; aOR 0.50[95%C.I., 0.25-0.90]), but had comparable spinal cord ischemia (2.9% vs. 3.5%, p=.60; 0.64[0.31-1.22]) and perioperative mortality (3.1% vs. 3.3%, p=.94; 0.71[0.34-1.37]). Compared with open, endovascular LSA revascularization had lower rates of overall composite in-hospital complications (20% vs. 27%, p<.001; 0.64[0.49-0.83]) and shorter overall hospital stay (7 vs. 8 days, p<.001). After adjustment, 5-year mortality was similar among groups (aHR 0.85[0.64-1.13]). Sensitivity analysis supported the primary analysis with similar outcomes. Conclusions In patients undergoing TEVAR starting in zone 2, endovascular LSA revascularization had lower rates of postoperative stroke and overall composite in-hospital complications, but similar spinal cord ischemia, perioperative and 5-year mortality rates compared with open LSA revascularization. Future comparative studies are needed to evaluate the mid- to long-term safety of endovascular LSA revascularization and assess differences between specific endovascular techniques.