Comparison of Local Versus General Anesthesia Following Transfemoral Transcatheter Self-Expanding Aortic Valve Implantation (from the Transcatheter Valve Therapeutics Registry)

Abstract

Transfemoral transcatheter aortic valve replacement (TF-TAVR) is mostly performed under general anesthesia (GA) in most US centers. We examined in-hospital and 30-day outcomes in patients who underwent TF-TAVR with a self-expanding bioprosthesis using local anesthesia (LA) or GA. Patients from the Transcatheter Valve Therapeutics Registry who underwent TF-TAVR from January 2014 to June 2016 with LA or GA were evaluated. Propensity matching was performed and procedural and clinical outcomes compared up to 30 days. A total of 11,006 patients were included (GA: 8,239 [74.9%] and LA: 2,767 [25.1%]). After propensity matching (n = 1,988 matched sets), device success was similar (94.5% vs 94.6%, p = 0.905). No differences in in-hospital stroke (2.7% vs 2.3%, p = 0.413) or paravalvular regurgitation grade (p = 0.113) were noted. Fewer LA patients were converted to open heart surgery (0.2% vs 0.6%, p = 0.076) or experienced an in-hospital major vascular complication (0.7% vs 1.4%, p = 0.026). Intensive care unit time (40.1 ± 58.4 vs 50.9 ± 72.1 hours, p < 0.001) and postprocedure length of stay (4.1 ± 3.6 vs 5.0 ± 4.5 days, p < 0.001) were significantly shorter with LA. In-hospital and 30-day all-cause mortality were lower in the LA cohort compared to the GA cohort ([1.1% vs 2.7%, p < 0.001] and [2.1% vs 3.9%, p = 0.001]). In conclusion, in the largest series of self-expanding bioprostheses for TF-TAVR, these propensity-matched cohorts demonstrate that LA is an acceptable alternative to GA with comparable success, lower safety outcomes, complications rates, and in-hospital and 30-day all-cause mortality.