麻豆传媒

Long-term outcomes with left atrial appendage (LAA) closure have been holding steady as compared with anticoagulation, the PRAGUE-17 trial affirmed, although without answering questions about late device-related thrombus and leaks.

As in the initial trial results at 20 months of follow-up, LAA closure with an Amulet or Watchman device remained noninferior (subdistribution HR 0.81, 95% CI 0.56-1.18, P=0.006 for noninferiority) to non-vitamin K oral anticoagulants (NOACs) for the primary composite of safety and efficacy results at 3.5 years, reported Pavel Osmancik, MD, PhD, of University Hospital Kralovske Vinohrady in Prague.

The findings were presented at the Transcatheter Cardiovascular Therapeutics (TCT) meeting, held virtually and in-person in Orlando, and appeared simultaneously in the .

Overall bleeding events were similar between groups, but the relative contribution of periprocedural events shrank over time. By 3.5 years, non-procedural clinically-relevant bleeding was significantly more common in the NOAC group (40 vs 23 cases; HR 0.55, 95% CI 0.31-0.97).

The 402-patient study was not powered for individual components of the composite endpoint, which included stroke, transient ischemic attack, systemic embolism, cardiovascular death, major or non-major clinically relevant bleeding, and procedure- or device-related complications.

"But given the observed trend of significantly increased non-procedural clinically-relevant bleeding in the NOAC arm, it is likely that late bleeding events will increasingly favor LAAC [LAA closure] over time," noted Faisal M. Merchant, MD, of Emory University in Atlanta, in an .

At the same time, 8.5% of the LAA closure group ended up on anticoagulants at some point later in follow-up, perhaps due to treatment of venous thromboembolism or for prevention of other sources of emboli, Merchant wrote. "If significant numbers of patients treated with LAAC end up on anticoagulation in the long run, the benefits of LAAC are likely to be attenuated."

For that reason, it's important when selecting patients for LAA closure to ensure that there is no reason for non-LAA thrombus, such as a degenerative bioprosthetic valve or mitral stenosis, noted TCT press conference panelist Ruby Satpathy, MD, of Baptist Heart Specialists in Jacksonville, Florida.

One of the big questions with LAA closure devices has been durability of outcomes for patients.

"Unfortunately, with regard to understanding late device-associated risks, PRAGUE-17 is a real missed opportunity," wrote Merchant.

Planned transesophageal echocardiograms (TEE) at 6 to 18 months after LAA closure had to be canceled due to the COVID-19 pandemic for two-thirds of patients. Osmancik said his group hoped to finish those eventually, but that the clinical outcomes provided no reason to expect an excess of device-related thrombus.

LAA closure had hazard ratios of 0.68 (95% CI 0.39-1.20) for cardiovascular death and 1.14 (95% CI 0.56-2.30) for all-stroke or transient ischemic attack.

Merchant agreed that there was no obvious signal of late ischemic events. "Admittedly the number of ischemic events in PRAGUE-17, even during longer follow-up, is small, but these data are consistent with large, real-world registries which demonstrate similar long-term rates of [stroke/systemic embolism] following LAAC compared to propensity score matched patients treated with NOAC."

He called for more detailed assessment of late device-related thrombus and peri-device leak. But, he added, "until such data are available, the longer-term PRAGUE-17 results provide some reassurance that the originally-reported results appear stable over time."