Effect on 30-Day Readmissions after Early Versus Delayed Discharge after Uncomplicated Transcatheter Aortic Valve Implantation (from the Nationwide Readmissions Database)

Abstract

Early discharge after transcatheter aortic valve implantation has been shown to be safe in single-center studies and trials, but outcomes in broader clinical practice are unknown. Using the National Readmission Databases (1/2014 to 9/2015), we compared 30-day readmission rates between early (<3 days) and late (≥3 days) discharges after uncomplicated endovascular TAVR in a propensity-matched cohort. We examined factors associated with failure of early discharge by testing for interactions of patient factors with discharge strategy. Among 4,955 hospitalizations for uncomplicated TAVR, 1,857 (37%) were discharged early with substantial site-level variability (range 0% to 87%; median odds ratio 3.69). In the propensity matched cohort (n = 3,346), there were similar rates of 30-day readmission by discharge strategy (early vs late: 10.3% vs 10.6%; stratified log-rank p = 0.555). There was a statistically significant interaction between discharge strategy and number of chronic conditions (p = 0.007), where readmission rates were lower in patients discharged early in those with 0 to 4 chronic conditions, but not in those with 5 to 10 or >10. In conclusion, in a large "real-world" cohort, early discharge after uncomplicated TAVR was not associated with a higher rate of 30-day rehospitalization, yet there was significant variability across US hospitals. No patient characteristics were associated with increased risk of readmission with early discharge.