FRANCE-TAVI: AF, renal failure, sex strong predictors of post-TAVR mortality

MUNICH — Patient characteristics, including atrial fibrillation, renal failure and sex, not antithrombotic therapy, had the greatest impact on long-term mortality after successful transcatheter aortic valve replacement, according to data from the FRANCE-TAVI registry.

Oral anticoagulation, which was administered in nearly 71% of patients for AF, was a correlate of mortality, independent of AF. Moreover, oral anticoagulation after TAVR decreased risk for bioprosthetic valve dysfunction, as opposed to antiplatelet therapy, Jean-Philippe Collet, MD, PhD, with the ACTION Study Group, Unité de Recherche Clinique, Hôpital Lariboisière, APHP, Paris, said at the European Society of Cardiology Congress.

The prospective, multicenter FRANCE-TAVI registry included 12,804 patients who underwent TAVR at 48 centers in France from 2013 to 2015.

“The objectives of this study were to investigate whether antithrombotic treatment influences long-term mortality at 3 years follow-up and early bioprosthetic valve dysfunction based on transthoracic echocardiographic finding during early follow-up,” Collet said during a presentation of the data.

Correlates of mortality

Of the nearly 13,000 patients in the registry, the final analysis with respect to mortality could be performed in 11,469 patients. Mean patient age was 83 years, logistic EuroSCORE I was 17.8 and about half were men. Median follow-up was 428 days.

The survival rate was 90.4% at 1 year, 80.1% at 2 years and 69.3% at 3 years.

About one-third of patients were discharged on oral anticoagulants, which was attributable to a history of AF in 70.8% of patients, according to Collet.

The researchers identified several independent correlates of long-term mortality, including NYHA class III or IV HF, EuroSCORE I, prior CABG, prior non-CABG cardiac surgery, diabetes and use of auto-expandable vs. balloon-expandable valves or smaller valves. The strongest independent correlates were male sex (adjusted HR = 1.63; 95% CI, 1.44-1.84), history of AF (aHR = 1.41; 95% CI, 1.23-1.62) and chronic renal failure (aHR = 1.37; 95% CI, 1.23-1.53), according to the findings.

Other independent factors associated with mortality in this patient population included anticoagulation at discharge (aHR = 1.18; 95% CI, 1.04-1.35), TAVR performed via non-femoral access (aHR = 1.18; 95% CI, 1.04-1.35) and moderate to severe prosthetic regurgitation (aHR = 1.28; 95% CI, 1.11-1.4) — the latter two of which are related to the procedure itself, he noted.

“Much of these data are already known, but I want to drive attention toward the role of anticoagulation, which is associated with increased mortality, and the independent association between AF and increased mortality,” Collet said. “Both factors are present independently, although they do vary co-linearly, as the vast majority of patients who were discharged on oral anticoagulants received oral anticoagulants because they had AF.”

Kaplan-Meier curves also showed an unadjusted HR of 1.5 (95% CI, 1.35-1.66), with a clear trend toward associating anticoagulation with mortality. However, these results are preliminary, Collet noted.

Neither aspirin nor clopidogrel were associated with mortality, according to the data.

Bioprosthetic valve dysfunction

Of all patients, 2,555 were included in the final analysis of early bioprosthetic valve dysfunction, which was defined as increased prosthetic gradient of 10 mm Hg or greater or new gradient of 20 mm Hg or greater. About one-third of patients were discharged on oral anticoagulants, the median follow-up time to echocardiography was 12 months and 5.5% were diagnosed with bioprosthetic valve dysfunction.

Oral anticoagulation at discharge was protective against bioprosthetic valve dysfunction (aHR = 0.54; 95% CI, 0.35-0.82), according to Collet. Additionally, correlates of increased risk for bioprosthetic valve dysfunction were prior TAVR (aHR = 2.96; 95% CI, 1.15-7.64) and a prosthesis of 23 mm or smaller (aHR = 3.43; 95% CI, 2.41-4.89).

Future research necessary

In this study, post-TAVR antithrombotic treatment appears to be mainly driven by patient characteristics, according to Collet.

“It’s not the device that is driving the use of antithrombotic therapy, but patient characteristics,” he said. “Oral anticoagulation is also associated with less bioprosthetic valve dysfunction, but the potential clinical benefit is missing and I think that’s due to gaps in knowledge.”

He also noted that oral anticoagulant use is co-linear with an indication for AF, but it is not yet known at this time whether oral anticoagulant use reflects the impact of AF on late outcome or whether it is clearly independent.

Collet highlighted several study limitations, including lack of randomization, potential treatment crossover and the identification of bioprosthetic valve dysfunction using only echocardiography.

Major ongoing trials, such as GALILEO, ATLANTIS, ENVISAGE, POPULAR, AVATAR and AUREA, have different designs, but are looking at the same endpoints and will likely provide more information, he noted.

“In this study, we’ve seen that the interrelation of anticoagulation and outcomes are very complex. We know the independent correlates of long-term mortality and probably anticoagulation at discharge may play a role, but we need to wait for ongoing studies,” Collet said. “Things are clearer for the use of oral anticoagulants in bioprosthetic valve dysfunction, but as we have seen, there are procedural characteristics that may be key players on this outcome. Many of the ongoing trials will have more in-depth data and investigation.”– by Melissa Foster

References:

Collet J-P, et al. Late-Breaking Science 72. Presented at: European Society of Cardiology Congress; Aug. 25-29, 2018; Munich.

Overtchouk P, et al. J Am Coll Cardiol. 2018;doi:10.1016/j.jacc.2018.08.1045.

Disclosures: The FRANCE-TAVI registry was partly funded by Edwards Lifesciences and Medtronic and supported by the ACTION Study Group. Collet reports he has received research grants to his institution or honorarium from AstraZeneca, Bayer, Bristol-Myers Squibb, Daiichi-Sankyo, Eli Lilly, Fédération Française de Cardiologie, Lead-Up, Medtronic, MSD, Sanofi-Aventis and WebMD.