Registry Data Suggest Protection Up to 2 Years With LAA Closure

Patients with nonvalvular A-fib who are contraindicated for oral anticoagulants have better outcomes for 2 years after left atrial appendage (LAA) closure with the Amplatzer cardiac plug than would be expected based on their risk profile, according to registry data published online January 8, 2015, ahead of print in Heart. The reduction in thromboembolic and bleeding events is most apparent during the second year postimplantation.

The findings, initially presented last spring at EuroPCR in Paris, come amid anticipation that the competing Watchman device (Boston Scientific) will finally receive a verdict from the FDA in 2015.Take Home: Registry Data Suggest Protection Up to 2 Years With LAA Closure

Using data from the Iberian registry, researchers led by José Ramón López Mínguez, MD, PhD, of Infanta Cristina University Hospital (Badajoz, Spain), looked at 167 patients with nonvalvular A-fib and contraindications to oral anticoagulation who underwent LAA closure with the Amplatzer cardiac plug (St. Jude Medical) at 12 centers in Spain and Portugal. The most common indications for LAA closure were GI hemorrhage (30.5%) and cranial hemorrhage (22.8%).

Procedural complications, which occurred in 5.38% of patients, included 2 transient ischemic attacks (TIAs), 4 vascular complications, 2 cases of cardiac tamponade, and 1 instance of device migration (percutaneously snared).

In the 158 patients (94.6%) with procedural success, cumulative 24-month adverse event rates were:

  • Death: 10.8%
  • Major bleeding: 5.7%
  • Minor relevant bleeding: 4.4%
  • Any bleeding: 10.1%
  • Stroke/TIA: 4.4%
  • Any event: 19.0%

Device thrombosis occurred in 8.2%, atrial septal defect in 11.4%, and leak (< 3 mm) in 8.2%.

Observed rates of bleeding and stroke/TIA per year were lower than those predicted by the HAS-BLED, CHADS2, and CHA2DS2-VASc risk scores had patients not undergone intervention (table 1). The annual rate of death was 5.8%, and most major bleeding events were GI hemorrhages.

Table 1. Annual Event Ratesa Based on 2-Year Follow-up  

Using data only from 12-month follow-up, further calculations showed that the annual rates of major bleeding and stroke/TIA were still lower than expected at 5.2% and 3.9%, respectively. The reduction in thromboembolic risk was seen “despite periimplantation events and possible thrombus formation on the devices in the first few months,” the researchers note, pointing out that in the second year, the disparity between observed and expected outcomes was even more dramatic.

According to Dr. López Mínguez and colleagues, the rate of thrombus formation on the device ranged from 5% to 17% in earlier studies, with its likelihood affected by the stringency with which it is defined and monitored. The risk rises when dual antiplatelet therapy (DAPT) is stopped in the first 2 months, they say, and it falls when an enoxaparin course of 2 to 3 weeks is given. Choice of antithrombotic therapy in trials has varied.

Overall, the findings of the current study “show that LAA closure with an occlusion device in patients contraindicated for [oral anticoagulation] is a reasonably safe therapeutic option…,” the authors write.

“New studies with longer follow-up periods and early discontinuation of DAPT would be of great interest to verify whether bleeding events during the first year are reduced without losing the efficacy of lowering thromboembolic events,” they conclude. “This could broaden current indications.”

Comparator a “Straw Man”?

In an email to TCTMD, Zoltan G. Turi, MD, of Robert Wood Johnson University Hospital (New Brunswick, NJ), described the study as “a nice effort, but [the] bottom line is that it is yet another ‘device versus historical predicted event rate’ study.”

Predicted event rates based on CHADS scores, he noted, “have been uniformly too easy a ‘straw man’ in the LAA closure literature. In the only 2 randomized trials in this arena, PROTECT AF and PREVAIL, the control group event rate has fallen well under predicted; thus, the wildly optimistic results from both the PLAATO and WATCHMAN pilot studies, which used the same CHADS score [calculations] as the current study to predict event rates, purported to reduce events by two-thirds and by 85% respectively…, [well] in excess of what was seen in the subsequent randomized trials.”

Moreover, Dr. Turi said, the current study fails to mitigate concern regarding device thrombus. “It is hard to know how to interpret the data given that there were different follow-up TEE timetables at the main site versus the rest of the hospitals, and information is not supplied on exact definitions,” he said. “Lack of a core laboratory compounds uncertainty, and the treatment with intravenous [low molecular weight heparin], while reasonable for addressing the presence of thrombus on the device, is concerning, in part because these patients were enrolled with contraindications to anticoagulation. The small number of patients and events makes correlating device thrombus formation with embolic events impossible.”

Still, the study adds to the “anemic literature” on Amplatzer, a device that “has been widely marketed for use in this population—patients who have contraindication for [oral anticoagulants]—despite the fact that the evidence base is extremely limited,” he said. It has been approved for sale in Europe since 2008.

“The crux of the question is whether an LAA closure device plus antiplatelets provides sufficient benefit to warrant implantation and its associated hazards,” he explained. “There is a major need for randomized trials of this population, until which this question simply cannot be answered.”

 


Source: 
López Mínguez JR, Nogales Asensio JM, Gragera JE, et al. Two-year clinical outcome from the Iberian registry patients after left atrial appendage closure. Heart. 2015;Epub ahead of print.

Related Stories:

Disclosures
  • Dr. López Mínguez reports receiving minimal economic compensation for serving as a proctor for LAA closure with the Amplatzer cardiac plug.
  • Dr. Turi reports no relevant conflicts of interest.

Comments