CHICAGO – Two-year follow-up results on transcatheter aortic valve replacement identified for the first time the critical role that even mild paravalvular aortic regurgitation can play in patient survival and highlighted a new problem the percutaneous procedure needs to overcome in contending with aortic valve replacement by open surgery.
In the PARTNER (Placement of Aortic Transcatheter Valves) trial, which compared the performance of one system of transcatheter aortic valve replacement (TAVR) against aortic valve replacement by open surgery, patients with mild paravalvular aortic regurgitation (a leak around the outer edges of the inserted valve) had a 33% 2-year morality, compared with a 26% rate in TAVR patients with no leak or a trace leak, Dr. Susheel K. Kodali reported at the meeting. Patients with a moderate or severe leak had a 51% mortality rate at 2-year follow-up, a statistically significant difference, compared with patients with a trace or no leak.
The data Dr. Kodali reported also showed that the incidence of paravalvular leaks, which generally appeared during the first 30 days following intervention, occurred substantially more often in the TAVR patients, compared with those who underwent open valve replacement. In the TAVR group of the PARTNER trial, about 48% of patients had a trace or no leak at 2 years, compared with a roughly 90% rate in the surgery patients. By 2 years after treatment, 7% of the TAVR patients had a moderate or severe leak, compared with 1% of the open-surgery patients.
"The trend was to say that grade 1, grade 2 leaks didn’t matter. Now you show that aortic regurgitation, especially moderate or severe, is not acceptable," commented Dr. G. Alain Cribier, professor of cardiology at Charles Nicolle Hospital in Rouen, France. "The good part is that you can do something about them. We can do a lot more to assess the anatomy of the annulus and use more valve sizes. In my experience, the trend is to reduce the [number of] leaks," said Dr. Cribier, who performs TAVR.
Dr. Kodali agreed. "Paravalvular aortic regurgitation is an important target for us to improve. Factors that go into paravalvular aortic regurgitation include valve size and position. Two recent reports said that we have systematically undersized the valve 35%-40% of the time," when TAVR had been performed in the past. "We need to better understand valve anatomy," said Dr. Kodali, an interventional cardiologist and codirector of the transcatheter aortic valve program at Columbia University in New York.
Another step he and his associates have taken to cut leak incidence and severity has been valve dilatation following placement. "If we see a paravalvular leak that is mild or worse, we tend to reballoon the valve," he said. He cited a separate report he and his associates made at the meeting that documented the ability of balloon dilatation following valve placement to significantly reduce the volume of paravalvular regurgitation in 36 patients who began with moderate or severe regurgitation.
"But ideally, if we better size the valve up front we won’t need to do postimplant dilatation," he said.
"Leaks are significantly improved by improved techniques. There will probably be about a 50% improvement in leak" incidence and severity, predicted Dr. Robert A. Guyton, professor of surgery and chief of cardiothoracic surgery at Emory University in Atlanta and a collaborator on the PARTNER trial. "But I don’t think we’ll get rid of it." The incidence of mild or worse paravalvular leaks following TAVR may drop to perhaps 25% of patients, he predicted, "but that’s still a big number," compared with the rate following open surgery, he said in an interview.
The data Dr. Kodali reported showed the powerful impact that paravalvular leaks had on overall outcomes in the study. Although the overall 1- and 2-year outcome results from PARTNER showed essentially superimposable mortality rates for TAVR and open surgery, among those patients who emerged from TAVR with a trace or no paravalvular leak, 1-year mortality was 13% with TAVR, compared with 24% with open surgery, and 2-year mortality was 26% with TAVR, compared with 30% with open surgery, differences that approached statistically significance. In other words, patients who came through TAVR with a trace or mild leak had better outcomes than did the surgery patients, but because many TAVR patients had mild, moderate, or severe leaks, the poor prognosis of those patients counterbalanced the relatively good outcomes of the patients who had ideal TAVR outcomes.
And the problem, said Dr. Guyton, is that "going into the cath lab, you can’t predict which patients will have a leak and which won’t. But maybe [in the future] we can look at how much calcium there is on a valve and predict which patients may have leaks" and perform open surgery on those patients instead of TAVR.