乌鸦传媒

One-year data from cohort A of the PARTNER trial confirm a higher risk of stroke in those undergoing transcatheter aortic valve implantation (TAVI) using the transapical approach, according to a substudy of the trial.

In the substudy, researchers examined 47 patients with 49 neurological events and found the incidence of stroke to be 6.1% in the TAVI arm compared with 4.5% in the surgical group although the difference was not significant, Michael J. Mack, MD, medical director of cardiovascular surgery for the Baylor Health Care System in Plano, Texas, said during a conference call Monday morning.

Mack noted that overall the stroke risk was low in both arms at 30 days -- 3.8% versus 2.3% favoring surgery.

The early procedural stroke risk factors included receiving a valve percutaneously, and a small aortic valve area index (for those in the TAVI arm). "In other words, the tighter the valve, the more calcium on the valve, the greater the chance of stroke in the early procedural phase," Mack said.

Prior to the conference call, hosted by Wells Fargo Securities, study results were presented by Craig Miller, MD, from Stanford University, at the American Association of Thoracic Surgery meeting in Philadelphia.

"What was surprising to us was that we thought there was a neurological protective benefit early on with the transapical approach compared to the transfemoral approach, which we did not find," Mack said, adding that the trial was not powered to directly compare these two approaches.

PARTNER B results, presented at the 2010 Transcatheter Cardiovascular Therapeutics meeting, compared TAVI to optimal medical therapy. The results from PARTNER cohort A , presented in March 2011 at the American College of Cardiology meeting, compared TAVI to surgical aortic valve replacement.

The overall neurological event rate of cohort A at 30 days, presented at the ACC meeting in March, was 5.6% in the TAVI arm and 2.6% in the surgical arm. The risk of stroke was also two times greater in the TAVI arm at one year: 8.3% versus 4.3%, Mack recounted.

The overall incidence of major stroke at 30 days was 3.8% in the TAVI arm versus 2.3% in the surgical arm. At one year, those figures were 4.8% versus 2.6% in the TAVI and AVR arms, respectively.

Substudy results raises important questions

In the TAVI arm of the new substudy, researchers found a 58% rate of major stroke, 26% for TIA, and 16% for minor stroke. There was not much difference in the surgical arm, Mack said: 69%, 25%, and 6%, respectfully.

Risk factors for late incidence of stroke, after 30 days, included a higher NYHA classification, history of previous stroke, and undergoing the procedure transapically.

Researchers found that those in the transapical TAVI group had a higher incidence of stroke compared with those in the transfemoral arm. Those in the transapical arm tended to be women and had higher incidences of peripheral artery disease, carotid surgery, and CABG surgery compared with the transfemoral arm.

"The higher incidence of stroke in the transapical arm seems to be related to the underlying disease substrate rather than procedurally related," Mack said during the conference call.

"This really elicits more questions than answers. First, would a cerebral protection device offer any benefit? About half of the strokes occurred in the periprocedural phase and seemed to be related to heavier calcium of the valves. You could make a reasonable hypothesis that there may be a role for cerebral protection in those patients and in managing about half of the strokes," he said.

Regarding later strokes, Mack said it is unclear whether they are device related, and if they are, perhaps there is a role for anticoagulation or dual-antiplatelet therapy. "The protocol of the trial recommended dual-antiplatelet therapy, but it is impossible to discern the compliance rate," he commented.

The definition of major stroke used in the PARTNER trial was a Rankin score of 2 or greater. A clinical events committee adjudicated all events. "This was the first time such a committee was used in a major trial," Mack said.

Miller and Mack reviewed all case report forms and source documents again, over and above the clinical events committee. "Any question as to whether a stroke was major or minor, we adjudicated it as a major stroke. We took a very conservative approach and erred on the side of recording the worst possible outcome," Mack said.

He also noted that after the early hazard of stroke with the transfemoral approach, the late hazard lines become parallel.

How will the stroke data impact adoption of TAVI? "Patients clearly indicate in informed consent forms that they do not want to experience a stroke. They'd rather die than have a stroke. This is an issue that needs to be addressed. The good news is that it can be addressed," Mack said.

First, the newer generation of devices is smaller and techniques have improved, which should help to reduce stroke risk, he said. Second, the use of cerebral protection devices will help reduce the risk, and future trials should be designed to evaluate optimal anticoagulation and antiplatelet therapy.

"The concern now is that if we treat a lower risk population, is that stroke rate acceptable," Mack said. "However, it is not known whether a lower risk population would have the same stroke risk, given the potential for a lower atherosclerotic burden."

He added that going forward, trials should be designed with a composite endpoint that includes at least death and stroke.

Also, whether results from the PARTNER trial can be generalized to operators outside the tightly controlled trial is debatable, he said. He noted that 30-day results in Europe where the device is approved, and where they are using third-generation devices, are not better than in the PARTNER trial.

Comment