Are Left Atrial Thrombi That Defy Preprocedure Anticoagulation Predictable?

June 09, 2021

Three or more weeks of oral anticoagulation (OAC) sometimes isn't up to the job of clearing any potentially embolic left atrial (LA) thrombi before procedures like cardioversion or catheter ablation in patients with atrial fibrillation (AF). Such OAC-defiant LA thrombi aren't common, but nor are they rare enough to ignore, suggests a new meta-analysis that might also have identified features that predispose to them.

Such predictors of LA clots that persist despite OAC could potentially guide selective use of transesophageal echocardiography (TEE) instead of more routine policies to either use or not use TEE for thrombus rule-out before rhythm-control procedures, researchers propose.

Their prevalence was about 2.7% among the study's more than 14,000 patients who received at least 3 weeks of OAC with either vitamin K antagonists (VKA) or direct oral anticoagulants (DOAC) before undergoing TEE.

But OAC-resistant LA thrombi were two to four times as common in patients with than without certain features, including AF other than paroxysmal and higher CHADS2 and CHA2DS2-VASc stroke risk-stratification scores.

"TEE imaging in select patients at an elevated risk of LA thrombus, despite anticoagulation status, may be a reasonable approach to minimize the risk of thromboembolic complications following cardioversion or catheter ablation," propose the study's authors, led by Antony Lurie, BMSC, Population Health Research Institute, Hamilton, Ontario, Canada. Their report was published in the June 15 issue of the Journal of the American College of Cardiology.

Guidelines don't encourage TEE before cardioversion in patients who have been on OAC for at least 3 weeks, the group notes, and policies on TEE use before AF ablation vary widely regardless of anticoagulation status.

The current study suggests that 3 weeks of OAC isn't enough for a substantial number of patients, who might be put at thromboembolic risk if TEE were to be skipped before rhythm-control procedures.

Conversely, many patients unlikely to have LA thrombi get preprocedure TEE anyway. That can happen "irrespective of how long they've been anticoagulated, their pattern of atrial fibrillation, or their stroke risk," senior author Jorge A. Wong, MD, MPH, Population Health Research Institute and McMaster University, told theheart.org | Medscape Cardiology.

But "TEE is an invasive imaging modality, so it is associated with small element of risk." The current study, Wong said, points to potential risk-stratification tools clinicians might use to guide more selective TEE screening.

"At sites where TEEs are done all the time for patients undergoing ablation, one could use several of these risk markers to perhaps tailor use of TEE in individuals," Wong said. "For example, in people with paroxysmal atrial fibrillation, we found that the risk of left atrial appendage clot was approximately 1% or less." Screening by TEE might reasonably be avoided in such patients.

"Fortunately, continued oral anticoagulation already yields low peri-procedural stroke rates," observes an accompanying editorial from Paulus Kirchhof, MD, and Christoph Sinning, MD, from the University Heart & Vascular Center and German Centre of Cardiovascular Research, Hamburg.

"Based on this new analysis of existing data, a risk-based use of TEE imaging in anticoagulated patients could enable further improvement in the safe delivery of rhythm control interventions in patients with AF," the editorialists agree.

The meta-analysis covered 10 prospective and 25 retrospective studies with a total of 14,653 patients that reported whether LA thrombus was present in patients with AF or atrial flutter (AFL) who underwent TEE after at least 3 weeks of VKA or DOAC therapy. Reports for 30 of the studies identified patients by rhythm-control procedure, and the remaining five didn't specify TEE indications.

The weighted mean prevalence of LA thrombus at TEE was 2.73% (95% CI 1.95% - 3.80%). The finding was not significantly changed in separate sensitivity analyses, the report says, including one limited to studies with low risk of bias and others excluding patients with valvular AF, interrupted OAC, heparin bridging, or subtherapeutic anticoagulation, respectively.

Patients treated with VKA and DOACs showed similar prevalences of LA thrombi, with means of 2.80% and 3.12%, respectively (P = .674). The prevalence was significantly higher in patients:

  • with nonparoxysmal than with paroxysmal AF/AFL (4.81% vs 1.03%; P < .001)

  • undergoing cardioversion than ablation (5.55% vs 1.65; P < .001)

  • with CHA2DS2-VASc scores of at least 3 than with scores of 2 or less (6.31% vs 1.06%; P < .001).

A limitation of the study, observe Kirchhof and Sinning, "is that all patients had a clinical indication for a TEE, which might be a selection bias. When a thrombus was found on TEE, clinical judgment led to postponing of the procedure," thereby avoiding potential thromboembolism.

"Thus, the paper cannot demonstrate that presence of a thrombus on TEE is related to peri-procedural ischemic stroke," they write.

The literature puts the risk for stroke or systemic embolism at well under 1% for patients anticoagulated with either VKA or DOACs for at least 3 weeks prior to cardioversion, in contrast to the nearly 3% prevalence of LA appendage thrombus by TEE in the current analysis, Wong observed.

"So we're seeing a lot more left atrial appendage thrombus than we would see stroke," but there wasn't a way to determine whether that increases the stroke risk, he agreed.


Wong, Lurie, and the other authors report no relevant conflicts. Kirchhof discloses receiving partial support "from several drug and device companies active in atrial fibrillation," and to being listed as inventor on two AF-related patents held by the University of Birmingham. Sinning reports no relevant relationships. 

J Am Coll Cardiol. 2021;77: 2875-2886, 2887-2889. Abstract, Editorial

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....