SURTAVI Trial: TAVR Non-inferior to SAVR in Patients With Intermediate Surgical Risk

Transcatheter aortic valve replacement (TAVR) was found to be non-inferior to surgical aortic valve replacement (SAVR) for the primary composite endpoint of all-cause mortality and disabling stroke at 24 months in patients with symptomatic, severe aortic stenosis who had an intermediate level of operative risk, according to the results of the SURTAVI trial presented March 17 at ACC.17 in Washington, DC, and simultaneously published in the New England Journal of Medicine.

In a modified intention-to-treat analysis that included patients in whom the procedure for their assigned group was attempted and using a novel Bayesian statistical model, Michael J. Reardon, MD, FACC, and colleagues demonstrated that the primary outcome was similar at an estimated 12.6 percent and 14.0 percent in the TAVR and SAVR arms, respectively. "TAVR was just as good as surgery, but it was not statistically superior to it," said Reardon.

SURTAVI, conducted at 87 centers in the U.S., Canada, and Europe, is the first study to evaluate outcomes with the self-expanding CoreValve and the new Evolut-R bioprosthesis valves, used in 84 percent and 16 percent of TAVR patients, respectively. To allow for TAVR to be evaluated against real-world surgery, the surgeons performing SAVR could select any biologic valve or whether to enlarge the annulus or base of the valve.

The Society of Thoracic Surgeons Predicted Risk of Operative Mortality (STS PROM) scores were 4.4±1.5 and 4.5±1.6 in the TAVR (n = 864) and SAVR (n=796) groups, respectively. The mean age of the patients was 79 years, about 43 percent were women, and about 60 percent had NYHA Class III/IV heart failure.

Overall, deaths from any cause were similar for TAVR and SAVR: occurring in 2.2 vs. 1.7 percent of patients at 30 days, 6.7 vs. 6.8 percent at one year, and in 11.4 vs. 11.6 percent at two years. The rate of major disabling stroke at two years, 2.6 percent for TAVR and 4.5 percent for SAVR, was not statistically significantly different. Although not a primary outcome of the study, researchers noted that the risk of any type of stroke at 30 days was statistically superior for TAVR, 3.4 percent compared with 5.6 percent for SAVR.

Reardon stated there is some indication that the TAVR valve worked better, with a statistically superior valve orifice and lower mean gradients than with surgery on echocardiography at all time points. Moderate-to-severe paravalvular leakage was more common with TAVR than SAVR, occurring in 5.4 vs. 0.4 percent of patients, respectively, and pacemaker implantation was more common with TAVR. In contrast, SAVR was associated with more acute kidney injury, atrial fibrillation, and transfusions. "We saw the best surgical outcomes we’ve seen yet and TAVR did just as well. This is now the second randomized trial that has met its non-inferiority endpoint and should lead to changes in clinical guidelines that will move the field forward and also benefit our patients," Reardon said.

"While this study further supports the safety and efficacy of TAVR in intermediate risk patients, demonstration of comparable long term durability of TAVR compared to SAVR out to 10 years and beyond is critical to decision making in younger individuals," said Kim A. Eagle, MD, MACC, editor-in-chief of

Keywords: ACC17, ACC Annual Scientific Session, Aortic Valve, Aortic Valve Stenosis, Bayes Theorem, Bioprosthesis, Heart Valve Prosthesis, Heart Valve Diseases, Standard of Care, Stroke, Transcatheter Aortic Valve Replacement, ACC Scientific Session Newspaper

< Back to Listings