There is not enough evidence to give percutaneous or surgical left atrial appendage (LAA) closure an edge over long-term oral anticoagulation, a meta-analysis found. In the little data available for percutaneous procedures, there was no difference in stroke or death after Watchman implantation. However, there was moderate-level evidence that harm occurred in one in 15 patients (6.5%) within 7 days of percutaneous interventions. “There is limited evidence that the Watchman device may be noninferior to long-term oral anticoagulation in selected patients. Data on effectiveness of LAA exclusion devices is lacking in patients ineligible for long-term oral anticoagulation. Percutaneous LAA devices are associated with high rates of procedure-related harms,” according to North Noelck, MD, MPH, of VA Portland Health Care System, and colleagues in their investigation published online in Circulation: Cardiovascular Quality and Outcomes. “Although surgical LAA exclusion during heart surgery does not seem to add incremental harm, there is insufficient evidence of benefit,” they added. Surgical LAA interventions failed to show a benefit in stroke or mortality prevention in the limited evidence base, whereas serious safety events occurred in 6.9% to 32.0% of patients. “LAA exclusion might be especially attractive for patients who are unable to take oral anticoagulants,” Noelck’s group noted. “However, the trial data most closely apply to patients who do not have contraindications to long-term oral anticoagulation therapy. In these trials, warfarin [Coumadin] was used typically for 3 to 6 months until device endothelialization, and LAA closure was achieved.” Calling the meta-analysis “comprehensive and methodologically sound,” Richard Lange, MD, MBA, president of Texas Tech University Health Sciences Center El Paso, told MedPage Today that he agreed that “there are no studies that support the use of LAA exclusion in patients who are ineligible for warfarin therapy.” Noelck and colleagues themselves added that for percutaneous procedures like Watchman placement, “there is no randomized controlled trial data available to compare deployment success, LAA exclusion achieved, or health outcome benefits among the different devices.” “Over the long term in observational studies, patients had low rates of stroke and bleeding, and there were no reported technical device failures. However, data on longer-term safety from the observational studies are limited, in part, by high rates of attrition, lack of information about the loss to follow-up, wide variation in follow-up duration (6 months to 5 years), and the lack of a consistent standard for adverse events reporting,” they noted. Likewise, the evidence base for surgery was far from ideal: “The randomized controlled trials, although at low risk of bias, were underpowered for determining the clinical effectiveness of this procedure. Both the observational studies demonstrated no significant difference in stroke-free survival during the course of their follow-up, but important data such as information about anticoagulation use among the groups was lacking.” The pooled analysis included 20 studies involving patients who received LAA closure or removal. The two randomized trials of percutaneous closure both involved Watchman; another 11 observational studies of percutaneous closure were included. Three randomized trials and two observational studies on surgical exclusion were included. Noelck’s group suggested that LAA interventions may be a choice of convenience for some patients. “Some patients may simply prefer the placement of an LAA exclusion device over the inconvenience of long-term oral anticoagulation,” they wrote. “Policy makers will need to consider whether routine availability of periprocedural LAA exclusion for preference-sensitive indications is warranted.” “There are a variety of devices being used for LAA exclusion, but there is inadequate evidence that the efficacy and safety of each of these devices are similar enough to comfortably extrapolate data from [the Watchman] and apply to the use of a different device. Although the techniques used for many of the devices are similar, there are still important differences perhaps most notably for the Lariat device, which takes an epicardial approach to snaring and externally excluding the LAA.” “For the time being, the evidence for device efficacy applies most closely to the Watchman device, and there is insufficient evidence to determine the efficacy of other devices,” they emphasized. Disclosures Noelck and Lange disclosed no relevant conflicts of interest.