CHICAGO – Frailty and three common chronic health conditions place patients at greater risk of late mortality following transcatheter aortic valve replacement, according to extended follow-up of 339 patients in the Multicenter Canadian Experience study.
The overall survival rate was 57% after a median follow-up of 3 years (interquartile range 26-44 months) following transcatheter aortic valve implantation (TAVI).
Importantly, two-thirds of deaths (67%) were of noncardiac origin, 26% were due to cardiac causes, and 7% were of unknown causes, principal investigator Dr. Josep Rodés-Cabau said at the i2 Summit, a conference sponsored by the American College of Cardiology. Nearly half of noncardiac deaths were from respiratory failure (49%).
In multivariate analysis, a history of chronic obstructive pulmonary disease (hazard ratio, 1.99), chronic kidney disease (HR, 1.62), chronic atrial fibrillation (HR 1.82), and frailty (HR, 1.76) emerged as significant independent predictors of all-cause mortality more than 30 days after TAVI.
Surprisingly, there was no difference in survival between the transfemoral and transapical approaches in the multicenter TAVI program using a balloon-expandable valve, said Dr. Rodés-Cabau of the Quebec Heart and Lung Institute, Laval University, Quebec City, Canada. At 48 months follow-up, 45% of patients were free from death and stroke, as were 47% of those treated with the transfemoral approach and 44% treated with the transapical approach.
Causes of cumulative cardiac death included cardiac failure (65.5%), sudden death (13.8%), myocardial infarction (13.8%), endocarditis (3.4%), and postmitral replacement (3.4%).
Two patients required valve explantation because of endocarditis at 7 months and 13 months following TAVI, but no cases of structural valve failure occurred during follow-up, Dr. Rodés-Cabau stressed.
At 48 months, 76% of patients were free from cardiac death, as were 78% of transfemoral patients and 75% of transapical patients.
In multivariate analysis, pulmonary hypertension (HR, 1.92) and chronic kidney disease (HR, 2.30) were significant predictors of late cardiac death.
Although the data are compelling, the patient numbers are not sufficient to use to guide patient selection for TAVI, but instead represent a first step toward TAVI risk stratification, Dr. Rodés-Cabau said. After the introduction of TAVI, the focus was on technique, and many very-sick patients underwent the procedure only to die within the first year. Now, when a patient arrives at his center with these comorbidities, they are evaluated very carefully, including involvement of specialists such as respiratory or dietary specialists.
"I have to say that, based on the results of this study and also in our own daily experience, we have been refusing more patients now than we were in the past," he told reporters at a press briefing.
When asked during the formal study presentation why late mortality was similar with the two TAVI approaches, Dr. Rodés-Cabau said there could be several factors, but that the transapical approach is very well developed in Canada.
"These data were concentrated in a few number of centers with a huge experience in transapical approach, and this could explain the absence of difference between the two approaches," he said.
The investigators also evaluated regurgitation but found no association with late mortality, Dr. Rodés-Cabau said in an interview. The rate of moderate or severe regurgitation was relatively low at 8% at discharge. Studies have found a suggestion that mild regurgitation may be related to late mortality, but the Canadian data clearly show no difference in left ventricular diameter or LV ejection fraction changes between patients with and without mild regurgitation at 3 years, he said.
Overall, valve function remained stable up to 4 years of follow-up, with only mild, nonsignificant changes in transvalvular gradient and valve area.
"Mild aortic regurgitation, mostly paravalvular, was frequent after TAVI but had no impact on left ventricular diameters and function," Dr. Rodés-Cabau said at the meeting, also sponsored by the Cardiovascular Research Foundation.
Thirty-day mortality results from the Canadian experience were previously published in 2010 (J. Am. Coll. Cardiol. 2010;55:1080-90). Overall mortality was 10.4%, with pulmonary hypertension associated with a significantly higher rate of periprocedural death. Notably, patients with either porcelain aorta or frailty had acute outcomes similar to the rest of the study population.
Dr. Rodés-Cabau stressed that data in the study was analyzed at the Echo Core Lab and that only those patients with serial echocardiographic exams were included in the analysis. In all, 158 echocardiographic exams were analyzed at 1 year follow-up, 86 at 2 years, 34 at 3 years, and 11 exams at 4 years.
The median age of the patients was 81 years, 34% had prior cardiac bypass grafting, and 91% had a New York Heart Association functional class III-IV. The majority of patients received the Edwards Sapien valve (275 patients), with 57 implanted with the Cribier-Edwards valve, and 7 with the Sapien XT valve. The transfemoral approach was performed in 162 and transapical in 177 patients.