97. Re-Operation and Mortality Risk After Bioprosthetic Aortic Valve Replacement in Young Adults with Congenital Heart Disease: A Multi-Center Analysis
Stephanie Fuller1, Michele Borisuk2, Lynn A. Sleeper2, *Emile A. Bacha3, Luke Burchill4, *Kristine J. Guleserian5, *Michel N. Ilbawi6, Khanh Nguyen7, *Anees J. Razzouk8, Takeshi Shinkawa9, Minmin Lu2, *Christopher W. Baird2
1The Children's Hospital of Philadelphia, Philadelphia, PA;2Boston Children’s Hospital, Boston, MA;3Children's Hospital of New York/Columbia University, New York, NY;4University of Melbourne Royal Melbourne Hospital, Parkville, Australia;5Niklaus Children's Hospital, Miami, FL;6Advocate Children's Hospital, Oak Lawn, IL;7Maria Fareri Children's Hospital, Valhalla, NY;8Loma Linda University Hospital, Loma Linda, CA;9Tokyo Women's Hospital, Tokyo, Japan
Invited Discussant: *Craig J. Baker
Objective: Bioprosthetic aortic valve replacement (bAVR) in patients with congenital heart disease (CHD) is challenging due to patient age, size, complexity and the frequent need for re-operation. The objective of this multicenter collaboration was to assess rates of survival and re-operation and to identify predictors thereof in a contemporary cohort of bAVR patients.
Methods: Data were retrospectively collected for 314 patients undergoing bAVR at 8 centers from January 2000 to December 2014. Kaplan-Meier estimation of time to re-operation and Cox regression modeling was utilized.
Results: Average age at bAVR was 45.2 ± 27.0 years (IQR 17.8, 71.1) and 30% were <21 years of age. Indications for bAVR were aortic stenosis (150, 48%), aortic regurgitation (87, 28%) and mixed valvar disease (55, 18%). Twenty-eight patients (9%) had undergone prior AVR. Median valve size was 23 mm (IQR 21, 25). Concomitant procedures were performed in 55% of patients. Aortic annular enlargement was performed in 14%. Median follow-up was 2.9 (IQR 1.2, 5.7) years. Thirty-four patients (11%) required re-operation and 22 (65%) were less than 21 years old. Time to re-operation varied significantly among valve type (p=0.020). Crude 3-year rates of re-operation were 40% for one valve type and 0-14% for the others. Crude 3-year rates were 20%, 3%, and 1% in patients less than 21, 21 to 65, and greater than 65 years, respectively. Smaller valve size was associated with re-operation (p=0.002) even after being indexed to BSA. Variables independently associated with re-intervention by multivariable analysis were younger age at surgery (29% increase in hazard per 5-year decrease, p<0.001), valve type (HR=4 to 8 versus other valve types), smaller valve size (20% increase in hazard per 1 mm decrease, p=0.002) and concomitant ASD/PFO closure (HR=8.1, p<0.001). There was no covariate-adjusted difference in risk among the valve types. The overall 1, 3 and 5-year survival rates were 94%, 90% and 85%, respectively, without differences by valve type (p=0.19).
Conclusions: Re-operation after bAVR is common. Risk varies substantially by patient age and valve type. A concerning reduction in 5-year survival is shown for CHD patients undergoing bAVR. Further research is needed to guide optimal valve choice and improve survival.