Association between Red Blood Cell Transfusion and 30-day Readmission Following Transcatheter Aortic Valve Replacement

Abstract

ABSTRACT Background: We sought to evaluate the association between post-procedural packed red blood cell (PRBC) transfusion following transcatheter aortic valve replacement (TAVR) and 30-day all-cause readmission. We assessed incidence, causes and predictors of 30-day readmission. Methods: We retrospectively analyzed 417 patients who underwent TAVR and survived the index hospitalization. A propensity-score adjusted multivariable logistic regression model was utilized to relate PRBC transfusion to 30-day readmission and to identify predictors of 30-day readmission. Results: The overall 30-day readmission rate was 19.4% and was for non-cardiac causes in 54.3% of patients. Of patients who received PRBC transfusion and those who were not transfused, 30.9% and 21.7% were readmitted within 30 days, respectively (p = 0.08). After propensity adjustment, the odds of readmission were not different among transfused and non-transfused patients (1.33 [95% CI 0.74, 2.40, p = 0.34]). However, among non-anemic patients, transfusion was associated with a greater likelihood of readmission (50% vs. 11.8%, OR 4.92 [95% CI 1.74, 13.91], p = 0.003), in contrast to anemic patients in whom it was not (OR 0.96, [95% CI 0.53, 1.73], p = 0.89; interaction p = 0.002). Independent predictors of 30-day readmission included history of atrial fibrillation (OR: 2.06; CI: 1.23, 3.46, p = 0.006), urgent TAVR procedure (OR: 2.29; CI: 1.20, 4.38, p = 0.020), discharge to nursing home or rehabilitation facility (OR: 1.95; CI: 1.11, 3.44, p = 0.014) and any post-operative complication (OR: 2.08; CI: 1.19, 3.63, p = 0.007). Conclusions: Pre-procedure atrial fibrillation and urgent procedures are novel predictors of early readmission following TAVR. PRBC transfusion did not independently predict 30-day readmission following TAVR.