Study Affirms Cost-Effectiveness of Watchman LAA Closure Device for Stroke Reduction

As physicians and patients await news from the Centers for Medicare & Medicaid Services (CMS) regarding coverage for percutaneous left atrial appendage (LAA) closure, a new analysis suggests that it is cost-effective relative to warfarin for reducing risk of stroke in nonvalvular A-fib. 

“When compared over a lifetime, [LAA closure] proved to be the most cost-effective treatment,” write Vivek Y. Reddy, MD, of Mount Sinai Medical Center (New York, NY), and colleagues in a study published in the December 22, 2015, issue of the Journal of the American College of Cardiology.

The researchers examined a variety of outcomes, including 4-year clinical events, stroke rates, and quality of life (QoL) data, from the PROTECT AF trial and from meta-analyses that compared adjusted-dose warfarin and novel oral anticoagulants (NOACs).Cost was examined at different time periods over a 20-year lifetime in 2015 US dollars.

Cost-effective and Cost Saving

In the first year of treatment, LAA closure with the Watchman was more expensive than warfarin and resulted in fewer quality-adjusted life-years (QALYs). However, by 7 years, LAA closure was cost-effective relative to warfarin. Closure dominated the NOACs by year 5 and warfarin by year 10. In contrast, the NOACs were cost-effective relative to warfarin at 16 years. Additionally, LAA closure was associated with more QALYs at 10 years and beyond than warfarin or the NOACs.

Furthermore, sensitivity analyses found that even when procedure costs were doubled, LAA closure remained cost-effective relative to warfarin at 11 years ($41,470/QALY) and NOACs at 10 years ($21,964/QALY).

Using a willingness-to-pay threshold of $50,000/QALY, the overall probability of cost-effectiveness for LAA closure was 98%. At 20 years, there was a 95% probability that closure was cost-effective compared with the NOACs and a 75% probability that the NOACs were cost-effective vs warfarin.

In an email interview with TCTMD, Reddy said he believes the data are likely to be seen as supportive of a favorable final CMS decision regarding coverage for LAA closure procedures. “I would think that CMS would be particularly interested in the fact that the Watchman was not only cost-effective vs NOACs, but was actually cost saving,” he said.

Reddy added that the study is important because it looked at long-term Watchman data, which are only available via the PROTECT cohort, and the composite of all the NOACs as opposed to just one. Additionally, he said the study incorporated the level of disability associated with strokes into the economic model, which “is very important because a nondisabling stroke that, for example, simply causes some numbness in 1 hand, has a completely different financial impact on the patient and healthcare system than a large stroke that, for example, paralyzes half the body.... [M]ost cost-effectiveness analyses have not taken the stroke severity into account.”

Comparators, Data Sets at Issue

In an editorial accompanying the study, N.A. Mark Estes III, MD, of Tufts Medical Center (Boston, MA), cautions that clinicians should be mindful of the limitations of the cost-effectiveness model.

“Out of necessity, the analysis was performed using indirect comparison methodology, with warfarin as the common control,” Estes writes, adding that Watchman has not yet been compared directly to NOACs in prospective randomized trials. Furthermore, future head-to-head trials are needed to determine whether LAA closure provides a clear benefit when compared with factor Xa inhibitors or direct thrombin inhibitors.

Last month at the American Heart Association 2015 Scientific Sessions, an abstract presented by James Freeman, MD, MPH, of Yale University (New Haven, CT), created some buzz by showing that while Watchman was highly cost-effective compared with warfarin and dabigatran based on data from PROTECT AF, the device was dominated by both medications when PREVAIL data were used. Using a willingness-to-pay threshold of $100,000 per QALY, LAA closure with the Watchman was likely to be the most cost-effective option in 67% of cases based on PROTECT AF, but in only 40% based on PREVAIL.

Responding to that analysis, Reddy told TCTMD “it is very important to remember that PREVAIL was not a large study [nor was the follow-up very long], and was indeed not designed to be studied in isolation.” The most “intellectually honest way to look at the data,” he added, would be to combine the PROTECT AF and PREVAIL data sets. 

“My guess is that once longer follow-up is allowed, and the stroke severity information is incorporated—because most of the strokes in the PREVAIL Watchman group were actually minor strokes—the cost-effectiveness outcomes will look closer to what we published,” Reddy noted. He said he and his coauthors plan to redo their cost-effectiveness analysis with inclusion of the PREVAIL data.

But in his editorial, Estes points out that relying on data from trials at experienced centers to extrapolate cost-effectiveness may not provide the true picture of those costs in clinical practice.

Reddy acknowledged that while it is possible that the "real-world" economic data may look worse, they could also look better. 

“Remember that the Watchman complication rates in the FDA clinical trials have been decreasing,” he said. In fact, the recently presented EWOLUTION registry data, which included over a 1,000 European cases, demonstrated improved safety event rates, such as a cardiac tamponade rate of 0.3%, compared with the 5% rate in PROTECT AF.

Still, Estes concludes that any conclusions about costs are “highly dependent on assumptions regarding increases in costs of treatment and monitoring over time.” 

Sources: 
1. Reddy VY, Akehurst RL, Armstrong SO, et al. Time to cost-effectiveness following stroke reduction strategies in AF: warfarin versus NOACs versus LAA closure. J Am Coll Cardiol. 2015;66:2728-2739.
2. Estes NAM. Left atrial appendage closure for stroke prevention in AF: the quest for the Holy Grail [editorial]. J Am Coll Cardiol. 2015:66:2740-2742. 

Disclosures:

• Reddy reports serving as a consultant to Boston Scientific, as well as serving as a consultant to and receiving grant support from Coherex and St. Jude Medical.
• Estes reports consulting fees from Boston Scientific, Medtronic, and St. Jude Medical. 

Related Stories:

• Societies Issue Recommendations for Operators, Institutions Offering LAA Closure Procedures
• Meta-analysis Shows Watchman Still Noninferior to Warfarin, With Good Safety Outcomes
• LAA Occlusion More Cost-effective Than Both Warfarin and Dabigatran for A-fib Patients