Final 5-year data from PREVAIL, pooled with PROTECT AF
DENVER -- Left atrial appendage (LAA) closure with the Watchman device is comparable to warfarin (Coumadin) long-term for stroke prevention in atrial fibrillation (Afib), according to pooled data from the PREVAIL and PROTECT AF trials.
Final results of the PREVAIL trial at 5 years again showed that it met one of its two co-primary efficacy endpoints, Saibal Kar, MD, of Cedars-Sinai Medical Center in Los Angeles, told the audience at the Transcatheter Cardiovascular Therapeutics meeting here.
The device failed to meet the threshold for noninferiority to warfarin for the efficacy endpoint of combined stroke, systemic embolism, or cardiovascular/unexplained death (0.066% versus 0.051% over 18 months, posterior probability for noninferiority=88.4%).
But it did meet noninferiority for the mechanistic endpoint -- combined occurrence of post-procedure ischemic stroke and systemic embolism more than a week after randomization -- with a rate of 0.0255% versus 0.0135% with warfarin over 18 months (posterior probability for noninferiority=97.5%).
The PREVAIL data and meta-analysis with 5-year data from PROTECT AF (combined n=1,114) were simultaneously published online in the Journal of the American College of Cardiology.
The missed efficacy endpoint in PREVAIL was likely because the warfarin arm did better than expected for stroke, with a surprisingly low 0.73% incidence over 5 years, Kar's group argued.
It appears this could be attributed to chance because of the endpoint's wide confidence intervals. If warfarin had done so well in previous trials, "NOACs [non–vitamin K oral anticoagulants] would not have been approved," Kar said at a press conference.
Also of note in that 5-year data was the 55% reduced risk of disabling/fatal stroke when patients got LAA closure instead of long-term warfarin (HR 0.45; P=0.03), driven by an 80% relative reduction in hemorrhagic stroke (HR 0.20, 95% CI 0.07-0.56).
Nonetheless, "this is not a first-line therapy. This is for those who cannot, will not, and shall not take anticoagulants," Kar emphasized.
The meta-analysis showed Watchman and warfarin groups sharing similar rates of the composite endpoint (3.4% device versus 2.8%, HR 0.82, 95% CI 0.58-1.17) and all-stroke/systemic embolism over 5 years (HR 0.96, 95% CI 0.60-1.54).
The full dataset showed that LAA closure reduced cardiovascular mortality by 41% and all-cause mortality by 27% over 5 years.
Kar and colleagues warned that "these results do not necessarily apply to patients with true contraindications to oral anticoagulation," because patients had to take warfarin for 6 weeks post-procedure and dual antiplatelet therapy for 6 months.
Another caveat of the study was the question of how Watchman would stack up against NOACs. That question is of importance because while LAA closure addresses just one source of potential stroke, medical therapy is supposed to provide more systemic prevention.
The next generation of studies will come in two forms: large registries and comparisons of LAA closure with NOACs, said lead author Vivek Reddy, MD, of New York's Icahn School of Medicine at Mount Sinai, in an interview with MedPage Today (a Boston Scientific representative monitored the telephone call).
When it comes to the latter, "my guess is the large trials will start in 2 years and it will take 3 to 5 years to see some data, and it will be somewhere around 5 to 7 years before we get definitive data," Reddy predicted.
Panelists at a press conference pointed to the COMPASS trial, which showed that a very low dose of the NOAC rivaroxaban (Xarelto) in combination with aspirin could improve cardiovascular outcomes -- although it was unclear how many patients presented with Afib in that investigation.
Rivaroxaban may thus be the "sweet spot" for some patients in balancing ischemic and bleeding risks, Robert Yeh, MD, of Boston's Beth Israel Deaconess Medical Center, suggested in an interview.
Journal of the American College of Cardiology
Source Reference: Reddy VY, et al "5-year outcomes after left atrial appendage closure: from the PREVAIL and PROTECT AF trials" J Am Coll Cardiol 2017; DOI: 10.1016/j.jacc.2017.10.021.