Self-Expanding Transcatheter Aortic Valve Replacement in Patients With Low-Gradient Aortic Stenosis

Objectives: The authors sought to compare clinical and hemodynamic outcomes in patients receiving transcatheter aortic valve replacement (TAVR) for low-gradient (LG) aortic stenosis in the CoreValve EUS (Expanded Use Study) versus those with high-gradient (HG) aortic stenosis from the CoreValve U.S. Pivotal Extreme Risk Trial and CAS (Continued Access Study). Background: The EUS examined the impact of TAVR in patients unsuitable for surgical aortic valve replacement who were excluded from the U.S. Pivotal Extreme Risk Trial due to LG aortic stenosis. Methods: EUS patients were stratified by left ventricular ejection fraction: normal (≥50%, LG–normal ejection fraction), and low (<50%, did not respond to dobutamine by generating a mean gradient >40 mm Hg and/or velocity >4.0 m/s, “nonresponders”), and compared with extreme-risk patients from U.S. Pivotal and CAS that had either low resting gradient and responded to dobutamine (“responders”), or a high resting gradient (HG) or velocity. The primary endpoint was all-cause mortality or major stroke at 1 year. Hemodynamics and quality of life are reported at 30 days and 1 year. Results: At 30 days, patients with LG/low left ventricular ejection fraction (nonresponders and responders) had significantly higher rates of all-cause mortality or major stroke, all-cause mortality, and cardiovascular mortality than both HG and LG–normal ejection fraction patients. At 1 year, only the responders had higher rates of these outcomes in comparison to the other 3 groups. Mean gradient and effective orifice area improved significantly in all patients and were maintained through 1 year. New York Heart Association functional classification and Kansas City Cardiomyopathy Questionnaire overall summary scores improved (p < 0.05) in all cohorts through 1 year. When all 4 subgroups were pooled, both decreasing mean gradient and stroke volume index were associated with increased mortality. Pre-procedural mean gradient was the only hemodynamic independent predictor of 1-year mortality by multivariate analysis.