The long climb of left atrial appendage closure into approved and covered medical practice is not yet over, but three medical societies have started on the physician end of the program with an overview document. The societies offer some perspectives on a number of issues associated with this procedure, but state nonetheless that the data for patient selection are still largely absent, with the exception of data generated for the Watchman device by Boston Scientific (Marlborough, Mass.), a situation that leaves the patient selection question still in play.
FDA gave the Watchman the nod in April after the device appeared at three FDA advisory committees, and the Centers for Medicare & Medicaid Services (CMS) announced a national coverage analysis in which CMS indicated an interest in a registry requirement (Medical Device Daily, May 27, 2015). The registry discussion was already at play among cardiologists in 2013, however (MDD, Dec. 13, 2013), giving interested parties at least some lead time to consider how a registry might be assembled.
The article in the Journal of the American College of Cardiology argued that the data behind the Watchman "are limited" in terms of which patients should be selected for device use. The authors noted that of the currently available devices for LAA closure, only the Watchman has gone through randomized, controlled trials (RCTs) to evaluate the device against the current standard of care. They list the St. Jude Medical (St. Paul, Minn.) Amplatzer and the SentreHeart (Redwood City, Calif.) Lariat as other devices that are less well characterized in randomized trials, although they point out that the Lariat is a 510(k) device that was not cleared for this application (the Lariat is in perhaps a less-than-ideal situation in this area thanks to an FDA advisory triggered by a series of adverse events and six reported deaths (Medical Device Daily, July 14, 2015).
The authors also said that the paucity of professional guidelines can be traced back to the low volume of RCTs, although the volume of procedures in non-U.S. markets is substantial. The article predicted that use of LAA closure devices is certain to expand, and describe this as "the first of a series ... to address issues critical to the integration of new technologies into patient care."
The document noted that this version of the article is still a pre-publication manuscript and is still up for further editing before it is available in its final form. Upon completion, the article will appear not only in JACC, but also at the websites for the Heart Rhythm Society and the Society for Cardiovascular Angiography and Interventions (SCAI; both Washington). By some accounts, the global market for LAA closure was in excess of $7.5 billion two years ago, and the projected compound annual growth rate (CAGR) for this procedure of 13.2% between 2014 and 2020 should push global volume to more than $16 billion by 2020.
Among the authors of the JACC article is Fred Masoudi, a practicing cardiologist at the University of Colorado Hospital (Aurora, Colo.), who also chairs the writing committee at the American College of Cardiology (ACC; Washington). Masoudi acknowledged that there is a substantial body of outside-U.S. (OUS) experience behind the Watchman and other LAA devices. However, he said there is nonetheless a dearth of solid data regarding patient selection.
"It really is too scattered," Masoudi said of the OUS data. "There's not a lot of great systematic data on this area" outside of Watchman trials, which he said demonstrates "there is clearly a need for more data to help us make informed decisions" on using such devices. He also noted that FDA has approved an investigational device exemption for another device (perhaps a reference to the June FDA approval of an investigational device exemption for the Lariat), a trial that may or may not wrap up before any registry data emerge for the Watchman.
"My expectation is that [patient selection] will be addressed in a future version" of the afib guidelines, Masoudi said, but he added the uncertainty of timelines for the possible sources of data make it "hard for me to know what the timing of that will be."
Masoudi said the development of a more exhaustive set of guidelines "calls for . . . multiplicity of sources of data, not just RCTs." He reiterated the view that registries and RCTs "are complementary types of data," but he said that the question of whether the upcoming trial for the Lariat will clarify matters is difficult to forecast. The Lariat IDE, he said, "is more focused on the use of the Lariat as an adjunct" to ablation rather than a primary treatment, and Masoudi said clinicians will have to see the trial design before much will be known about the study's potential for adding to the body of knowledge.
There are limits to observational data, even from the best of registries, Masoudi indicated, although he declined to speak for HRS and SCAI on this point. He said the current guidelines for atrial fibrillation "definitely implement observational data," but he advised that observational data alone might not suffice to award any recommendations a 1A designation. However, registry data "ha[ve] been influential in guidelines in the past. It's my expectation – depending on what those data are – that they could influence" future guidelines as well. "It's also possible that a registry will refine our understanding as to who is experiencing complications," he said.
The article in JACC did not take up any requirements for operator or facility volume, and Masoudi said the societies are going to address this later, although he declined to predict when they would publish their thinking. CMS indicated in the Watchman national coverage analysis that the deadline for submitting comments was June 20.
The question of how to finance a registry hangs over the discussion, but Masoudi declined to speculate as to what sort of registration and per-patient fees an LAA closure registry might have to charge participating hospitals. He said, however, "there's a great deal of experience and understanding of registry logistics," adding that the TVT registry for transcatheter aortic valves is "a good place to start in terms of baseline expertise."
The article makes the case for limiting utilization to centers of excellence, but Masoudi said he is not sure how many hospitals could be counted as centers of excellence, although sites that participated in the Watchman trials might be counted toward a total. As for operator volume requirements, he said, "there does appear to be a fairly perceptible learning curve," adding that the Watchman trials depicted "declining rates of complications," including tamponade. The operator experience question is "a perfect question for a registry to follow," Masoudi said, but he observed that "there probably is no good hard and fast number" of procedures for determining when an operator has flattened the learning curve.