Left subclavian artery coverage during thoracic endovascular aortic repair and risk of perioperative stroke or death.

INTRODUCTION: Left subclavian artery (LSA) coverage during thoracic endovascular aortic repair (TEVAR) is often necessary due to anatomic factors and is performed in to up to 40% of procedures. Despite the frequency of LSA coverage during TEVAR, reported associations with risk of periprocedural stroke or death are inconsistent in reported literature. We examined the 2005-2008 American College of Surgeons National Surgical Quality Improvement Program Participant Use Data file to determine associations between LSA coverage during TEVAR and risk of perioperative stroke or death.

METHODS: Current procedural terminology (CPT) codes were used to identify patients undergoing TEVAR, LSA coverage, and subclavian revascularization. Patients undergoing coronary bypass, ascending aortic repair, abdominal aortic aneurysm repair, or nonvascular intra-abdominal procedures during the same operation were excluded. Perioperative stroke and mortality associations with LSA coverage were examined using logistic regression models for each outcome. Significance was assessed at α = 0.05, with univariable P < .05 required for multivariable model entry.

RESULTS: Eight hundred forty-five TEVAR procedures were identified, of which 52 patients were excluded due to additional major procedures performed with TEVAR. Seven hundred thirty-three of the remaining 793 procedures included CPT codes indicating primary placement of an initial thoracic endograft and form the basis of this analysis. LSA coverage occurred in 279 procedures (38%). Thirty-day stroke and mortality rates were 5.7% and 7.0%, respectively. LSA coverage was associated with increased 30-day risk of stroke in multivariable modeling (odds ratio [OR], 2.17 95% confidence interval [CI], 1.13-4.14; P = .019). Other significant multivariable risk factors for stroke included proximal aortic cuff placement during TEVAR (OR, 2.58; 95% CI, 1.30-5.16; P = .007) and emergency procedure status (OR, 3.60; 95% CI, 1.87-6.94; P < .001). No significant association between LSA coverage and perioperative mortality was identified (univariable OR, 1.70; 95% CI, 0.98-2.93; P = .0578).

CONCLUSION: LSA coverage during thoracic endovascular repair is associated with increased risk of perioperative stroke following TEVAR. Further evidence is needed to determine whether procedural modifications, including LSA revascularization, reduce the incidence of stroke associated with TEVAR.