Abstract
OBJECTIVE: We examined the effect of frailty on in-hospital mortality, readmission rates, and hospitalization costs after transcatheter and surgical aortic valve replacement in a population-level cohort. METHODS: The Nationwide Readmissions Database was queried for patients who underwent transcatheter or surgical aortic valve replacement during 2016-2018. Multivariate logistic regression was used to discern independent effects of frailty on outcomes. Kaplan-Meier time-to-event analysis was used to evaluate the effect of frailty on freedom from readmission. RESULTS: A total of 243,619 patients underwent aortic valve replacement: 142,786 (58.6%) transcatheter aortic valve replacements and 100,833 (41.4%) surgical aortic valve replacements. Frail patients constituted 16,388 (11.5%) and 7251 (7.2%) in the transcatheter aortic valve replacement and surgical aortic valve replacement cohorts, respectively. Compared with nonfrail patients, frail patients had greater in-hospital mortality (transcatheter aortic valve replacement: 3.2% vs 1.1%; surgical aortic valve replacement: 6.1% vs 2.0%; both P < .001), longer length of stay (transcatheter aortic valve replacement: 4 vs 2 days; surgical aortic valve replacement: 13 vs 6 days; P < .001), and greater cost (transcatheter aortic valve replacement: $51,654 vs $44,401; surgical aortic valve replacement: $60,782 vs $40,544; P < .001). Time-to-event analysis showed that frail patients had higher rates of readmission over the calendar year in both transcatheter aortic valve replacement (P < .001) and surgical aortic valve replacement (P < .001) cohorts. This association persisted on adjusted multivariate regression for mortality (transcatheter aortic valve replacement odds ratio [95% CI] 1.98 [1.65-2.37], surgical aortic valve replacement 1.96 [1.60-2.41]), 30-day readmission (transcatheter aortic valve replacement 1.38 [1.27-1.49], surgical aortic valve replacement 1.47 [1.30-1.65]), and 90-day readmission (transcatheter aortic valve replacement 1.41 [1.31-1.52], surgical aortic valve replacement 1.60 [1.43-1.79]) (P < .001 for all). CONCLUSIONS: For patients undergoing transcatheter or surgical aortic valve replacement, frailty is associated with in-hospital mortality, readmission, and higher costs. Further efforts to optimize outcomes for frail patients are warranted.