More than 95 percent of implantations with the Watchman device for left atrial appendage occlusion in patients with non-valvular atrial fibrillation were successful, according to an analysis of procedural data.
Although new operators performed half of the procedures, there were few complications in this real-world cohort. The procedural complication rates were 1.02 percent for pericardial tamponades, 0.078 percent for procedure-related strokes, 0.25 percent for device embolizations and 0.078 percent for procedure-related deaths.
Lead researcher Vivek Y. Reddy, MD, presented the findings Nov. 2 in a late-breaking clinical trials session at the Transcatheter Cardiovascular Therapeutics scientific symposium in Washington, D.C.
The results were simultaneously published online in the Journal of the American College of Cardiology. Boston Scientific, which manufactures the Watchman device, funded the study. Reddy is a consultant to Boston Scientific.
“The data shows that the U.S. clinical experience with [left atrial appendage occlusion] using the Watchman device has been favorable,” Reddy said in a news release. “Since the device was approved by the FDA, the technology has been used in increasing numbers of patients and procedural success rates remain high. In addition, despite a large percentage of previously inexperienced operators performing the implantation procedures, complication rates continue to improve compared to earlier clinical trials.”
The FDA approved the Watchman in March 2015 to reduce the risk of stroke in patients with non-valvular atrial fibrillation who seek an alternative to warfarin or novel anticoagulants. It is the only left atrial appendage occlusion device approved in the U.S.
In February, the Centers for Medicare & Medicaid Services (CMS) released a national coverage determination for percutaneous left atrial appendage closure for non-valvular atrial fibrillation. The decision made the Watchman available to some Medicare and Medicaid patients and required all patients undergoing left atrial appendage closure to be included into a national prospective registry. CMS approved the first registry on Aug. 17, in which hospitals submit data each quarter to the American College of Cardiology Foundation’s National Cardiovascular Data Registry.
For this study, the researchers obtained data on 3,822 consecutive implants with the Watchman device between March 2015 and May 2016 at 169 U.S. Centers. They mentioned that Boston Scientific required a trained Watchman clinical specialist to be present during all implants. The company also recorded procedure details and complications in a standardized, de-identified patient form.
Of the 382 operating physicians, 71 percent had not performed any Watchman procedures before the FDA approval. Those new physicians performed half of the procedures examined in this analysis.
The researchers noted that the procedure was successfully performed in 95.5 percent of patients. They added that a partial recapture of the device was required in 23 percent of the cases.
“Even though 71 percent of the operators were novice, these are probably operators that were chosen because of their skill in doing transfemoral procedures and left atrial procedures or other structural heart procedures,” Reddy said during a news conference. “The question of how this will roll out to novice operators is a continuing question.”
During the study, there were three strokes attributed to the procedure, including two patients who presented with symptoms consistent with ischemic stroke and one patient who had a stroke the day after the procedure. There were also three deaths attributed to the procedures, including one patient who died during pericardiocentesis before surgery could be performed, one patient who died during surgery and one patient who died two days after undergoing surgical repair. Another patient died within seven days of the procedure due to a pulmonary embolism that was not related to the Watchman.
The researchers mentioned that the analysis had a few limitations, including that there was no source verification, no core laboratory for procedural testing and assessment and no informed consent for collecting patient characteristics or follow-up outcomes. They also noted that operators did not have to report procedure-related adverse events that occurred outside the procedural period. In addition, they did not collect patient demographics as part of the standardized implant form.
“This is not a prospective registry with monitoring of each individual site,” Reddy said during a news conference. “There may have been events that were missed in this particular analysis, but we thought it was still important to report because this is the first elaboration of this technology beyond the clinical trial experience in the United States.”