Background: Elderly patients undergoing transcatheter aortic valve replacement (TAVR) are at risk of hospital readmission post-procedure. It is not known if the index hospital length of stay, and specifically early discharge after TAVR is associated with an increased risk of readmission. We hypothesized a non-linear relationship whereby both short and long lengths of stay were associated with increased readmission risk.
Methods: We performed a retrospective multi-center cohort analysis of patients undergoing elective transfemoral TAVR and surviving to discharge between Jan 2007 and March 2014. The exposure variable was hospital length of stay measured from the procedure date to the date of discharge and modeled as a continuous variable in a multivariable cause-specific Cox regression. Main outcome measures were 30-day and 1-year all-cause readmissions.
Results: The study population consisted of 709 patients with a median length of stay of 6 days (interquartile range: 4-8 days). At 30-days and 1-year, 13.5% (n=96) and 44.0% (n=312) of patients were readmitted, respectively. Although length of stay was not associated with 30-day all-cause readmissions (p=0.92), there existed a significant association with 1-year readmission (p=0.01) after adjustment for baseline clinical variables. The association between length of stay and 1-year readmission was linear (p=0.55 for non-linearity) with no evidence supporting an increased readmission risk for shorter length of stays.
Conclusions: Among elderly survivors of elective transfemoral TAVR, a short length of stay was not associated with an increased readmission risk within 30 days or 1 year. The 1-year readmission risk increased with longer length of stay.