Is the Watchman device better than no therapy?
Is FDA approval and use of the Watchman device misguided?
These are two questions that cardiologist John Mandrola, MD addressed in writing for Medscape.com in November, 2017. The Watchman device, which was recently approved by the FDA, was developed to block off the left atrial appendage – or “LAA” – (a sack-like structure that is off to the side of the left atrium chamber of the heart). The idea behind the Watchman device is that in patients with atrial fibrillation (AF), the risk of stroke is based on the “fact” that most clots develop in the LAA before they are pumped out to the brain. Adequate anticoagulation is usually very effective at preventing such clots, but some patients are not good candidates for anticoagulant therapy because of the associated bleeding risk. The Watchman device is designed for such patients to close off the LAA and, thereby, prevent clots from forming in the LAA. Such an approach may prevent clot-induced strokes while avoiding the need for anticoagulation. The FDA initially declined to approve the device until the results of a 2nd study became available. In a Medscape posting (www.medscape.com/viewarticle/888355), Dr. Mandrola outlined seven reasons why he remains skeptical about the value of this new device. The following comments are meant to summarize his assessments, but the reader is encouraged to review the entire article at the medscape link above.
Point 1: The additional data did not change the direction of trends in that, at 5 years, ischemic (clot-based) stroke rates were still higher with the device and the apparent lower hemorrhagic (bleeding-based) stroke rate with the device may be due, in part, to an unusually high hemorrhagic stroke rate with warfarin. The hemorrhagic stroke rate with warfarin was approximately twice as high as what had been reported in several earlier AF studies with warfarin. Dr. Mandrola also questioned inclusion of cardiovascular and unexplained death as a valid endpoint in the trials.
Point 2: The new data do not change the net benefit calculation in that all the available data indicate a 6% major procedural complication rate. This means that LAAC patients start off with a 6% chance of being harmed by the procedure and that risk has to be balanced against the possible benefit of therapy that may be “non-inferior” to warfarin. By far, the most common procedural complication was serious pericardial effusion, but major bleed, ischemic stroke, and device embolization also occurred as procedure-related events.
Point 3: The LAA is not the only source of AF-related clots from the heart, especially in those who are not good candidates for anticogulation. Dr Mandrola points out that the idea that about 90% of AF-related clots come from the LAA has not been proven and that more recent data indicate that the sickest AF patients have higher rates of clots forming outside of the LAA. It is precisely these “sicker” AF patients who are more likely to be candidates for a LAAC device due to being less likely to tolerate anticoagulation.
Point 4: Use of antiplatelet agents and anticoagulants remains a concern. Because the LAAC device is a foreign body inside the heart, usual therapy includes a period of warfarin therapy followed by antiplatelet therapy. As Dr. Mandrola points out, the bleeding risk with aspirin in AF patients has been reported to be as high as that seen with warfarin, and in one study, almost 60% of LAAC device patients remained on anticoagulation or dual antiplatelet therapy at about one year after the procedure.
Point 5: “Stroke is a systemic disease”. AF produces structural changes in the heart and blood components that lead to increased clotting risk. Insertion of a LAAC device does nothing to reverse or correct these factors that increase the risk of clots.
Points 6 & 7: I chose to combine these last two issues because they both are related to statistical methods. The first point Dr. Mandrola makes is that the authors used Bayesian statistical methods which he indicates are based on “evidence or beliefs about something”. My “take” on his discussion is that the analysis following the 2nd study may have been biased by the authors’ beliefs or views of the data from the first study. The second point, is that the non-inferiority limits were set so wide that it is possible that outcomes with the Watchman device could have been almost 5 fold worse than the warfarin outcomes and still the analysis would have met the non-inferiority limits.
So, what’s the “bottom line”? Dr. Mandrola concludes that “… we have no RCT-level evidence that LAAC is better than no therapy.” (and) “… to stretch the available evidence from Watchman vs. warfarin and conclude that it is worth exposing anticoagulation-ineligible patients to an invasive, unproven preventive procedure is misguided.”