BACKGROUND Information about predictors and safety of NDD after TAVR is limited.

METHODS The study reviewed 663 consecutive patients who underwent elective balloon-expandable TAVR (from July 2014 to July 2016) at our institution. We first determined predictors of NDD in patients who underwent minimalist transfemoral TAVR. After excluding cases with complications, we compared 30-day and 1-year outcomes between NDD patients and those with longer hospital stay using Cox regression adjusting for the Predicted Risk of Mortality provided by the Society of Thoracic Surgeons. The primary endpoint was the composite of mortality and readmission at 1 year.
RESULTS A total of 150 patients had NDD after TAVR and 210 patients had non-NDD. Mean age and the Society of Thoracic Surgeons Predicted Risk of Mortality were 80.7  8.8 years and 6.6  3.7%, respectively. Predictors of NDD were male sex (odds ratio [OR]: 2.02; 95% confidence interval [CI]: 1.28 to 3.18), absence of atrial fibrillation (OR: 1.62; 95% CI: 1.02 to 2.57), serum creatinine (OR: 0.71; 95% CI: 0.55 to 0.92), and age (OR: 0.95; 95% CI: 0.93 to 0.98). As expected, 84% of patients with complications had non-NDD. After excluding cases with complications, there was no difference in hazard rates of the 30-day composite outcome between NDD and non-NDD (hazard ratio: 0.62; 95% CI: 0.20 to 1.91), but the hazard of the composite outcome at 1 year was significantly lower in the NDD group (hazard ratio: 0.47; 95% CI: 0.27 to 0.81). This difference in the composite outcome can be explained by the lower hazard of noncardiovascular related readmission in the NDD group.
CONCLUSIONS Factors predicting NDD include male sex, absence of atrial fibrillation, lower serum creatinine, and younger age. When compared with patients without complications with a longer hospital stay, NDD appears to be safe, achieving similar 30-day and superior 1-year clinical outcomes. (J Am Coll Cardiol Intv 2018;11:107–15) © 2018 Published by Elsevier on behalf of the American College of Cardiology Foundation.