PARIS – Patients with severe aortic stenosis at intermediate operative risk have a significantly lower 30-day risk of stroke and other neurologic complications with transcatheter aortic valve replacement than with surgical replacement, according to new results from the landmark SURTAVI trial.
“This is the first time the stroke rate has been shown to be lower with TAVR than with surgery,” A. Pieter Kappetein, MD, noted in presenting the results at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.
It’s a finding that adds to the momentum for studying TAVR in low-surgical-risk patients, he added.
“As we move toward lower-risk patients it will become even more important to see whether there is a difference in stroke. Suppose the stroke rate in SURTAVI had been a little higher with TAVR than SAVR? It would really make us more cautious about moving toward lower-risk patients. Now that we see that in intermediate-risk patients the stroke rate is actually a little bit lower than with surgery, I think we’ll feel more comfortable moving toward lower-risk patients,” according to Dr. Kappetein, professor of cardiothoracic surgery at Erasmus University Medical Center in Rotterdam.
SURTAVI (Surgical Replacement and Transcatheter Aortic Valve Implantation) involved randomization of 1,660 patients with severe symptomatic aortic stenosis to TAVR or SAVR. All participants were deemed to be at intermediate operative risk based upon a predicted surgical mortality of 3%-15%. The primary outcome -- a composite of all-cause mortality and disabling stroke at 2 years -- was presented at the 2017 meeting of the American College of Cardiology and simultaneously published (N Engl J Med. 2017 Apr 6;376(14):1321-1331). The rate was 12.6% with TAVR using the self-expanding CoreValve or Evolut R bioprosthesis and noninferior at 14% with SAVR.
Dr. Kappetein presented a prespecified secondary analysis of the 30-day rate of all neurologic complications, including nondisabling strokes and encephalopathy. He and the other SURTAVI organizers felt this was an important outcome because these early neurologic events have a major impact upon quality of life, including whether a patient will be discharged home or to a rehabilitation clinic or skilled nursing facility following aortic valve replacement.
The 30-day incidence of all stroke was 3.3% in the TAVR patients, significantly lower than the 5.4% rate in the SAVR group. By the 2-year mark, however, the difference was no longer statistically significant, with a rate of 6.3% in the TAVR group compared with 8.0% with SAVR.
Ninety-five percent of the early strokes were ischemic.
The 30-day incidence of disabling stroke was 1.2% with TAVR and 2.4% with SAVR, a difference that was not significant (P=0.057). The 2-year rate was 2.4% in the TAVR arm and 4.5% with SAVR, again not significantly different.
Half of the strokes in the TAVR group had a modified Rankin score of 0-1 at 30 days, meaning no or only minimal signs of stroke. In contrast, most of the strokes in the SAVR group were disabling, with a modified Rankin score of 2-6.
Only 36% of patients who had an early stroke were discharged home, compared with 87% of patients without a stroke. Not surprisingly, quality of life as assessed using the SF-36 physical summary was significantly worse in the early-stroke group. However, with or without stroke, TAVR patients recovered quality of life faster than SAVR patients.
He noted that the timing of the early strokes differed between the two groups. The great majority of both disabling and nondisabling strokes in the TAVR patients were periprocedural, occurring on the day of TAVR or the next day. Strokes in the SAVR group occurred then as well, but also on days 2-6.
One reason why SURTAVI is the first study to show a lower stroke risk with TAVR is that it was the first TAVR-versus-SAVR study to feature comprehensive neurologic testing pre- and post-procedure, along with evaluation of all suspected events by a neurologist or stroke specialist, according to Dr. Kappetein.
“As surgeons we all have said the stroke rate after SAVR is 1%-1.5%, but only when the patient wasn’t waving to us the next morning would we say, ‘Oh, that patient may have a stroke.’ Then we would call a neurologist. So there were many more subtle strokes that we never actually detected. If you do a proper examination of the patient before and after the procedure you’ll find many more strokes,” he said.
He and his coinvestigators systematically searched in vain for predictors of increased stroke risk among the TAVR and SAVR patients.
“Actually, the stroke risk is present for every patient we treat with TAVR or SAVR,” the surgeon continued.
However, discussant Adnan Kastrati, MD, chief physician at the German Heart Center in Munich, thought he spied in the SURTAVI data a potential opportunity to reduce early strokes in SAVR patients. He noted that new-onset atrial fibrillation is consistently more common in SAVR than TAVR patients, and that many of the strokes in the SAVR group occurred on days 2-6. When do heart surgeons typically start oral anticoagulation in their patients with postsurgical atrial fibrillation? he asked.
Not until after 48 hours, Dr. Kappetein replied.
“Those strokes on days 4, 5, and 6 might have to do with atrial fibrillation, and we may need to be more aggressive as surgeons in anticoagulating patients with atrial fibrillation after surgery,” he said.
Dr. Kappetein reported receiving research grant support from Medtronic, sponsor of SURTAVI.
This article was updated July 28, 2107.