Clots Found in the Left Atrial Appendage: A Big Deal for TAVR?
The TAVR team at James Cook University Hospital, in Middlesbrough, England, was starting a procedure when the echocardiographer reported a finding that made them stop. Transesophageal echo (TEE) showed a clot in the left atrial appendage (LAA). After a quick discussion, the team elected to proceed with balloon valvuloplasty. By the time this step was complete, the clot had vanished from the LAA, halting the procedure again. Further imaging revealed that the thrombus had embolized and become entangled on a wire in the left ventricle. After multiple attempts, operators managed to aspirate the clot and successfully complete the procedure.
“This individual case suggested to us that this was a real clinical phenomenon and certainly could happen in cases. It’s not just a theoretical concern,” Paul Williams, BMBCh, who participated in the procedure, told TCTMD. “And I think if we hadn’t managed to extract that thrombus and it embolized up to the brain, it would have caused a very significant stroke.”
The case highlights a clinical area without many definitive answers, as it’s not clear what the next steps should be when clinicians find a thrombus in the LAA during a pre-TAVR evaluation. There’s little in the way of information on how common the finding is, whether it is a major contributor to stroke after the procedure, or how patient management should be modified to mitigate any risks.
But despite the uncertainty, “there’s enough of a question mark about the importance of this that we should be looking for it [routinely],” Williams told TCTMD. “I think that it does potentially change your management, even if we don’t know for a fact that it increases the risk of stroke.”
Operators must make decisions, for example, about implementing or optimizing anticoagulation, selecting the most appropriate patients to receive embolic protection devices, delaying TAVR, or consenting patients for a procedure with a potentially higher risk for stroke, Williams said.
Since TAVR was introduced, clinicians and researchers have focused on reducing the periprocedural risk of stroke, and evidence suggests that the efforts are working, with rates going from 5% to 6% in the earliest trials to 1.7% to 3.4% in more recent registries.
“So it’s an important clinical problem, but there’s really been no data on whether embolization of left atrial appendage thrombus is an important cause of some of these strokes,” Williams said.
To begin exploring that question, Williams along with Sonny Palmer, MBBS (St. Vincent’s Hospital, Melbourne, Australia), performed a more systematic investigation of 198 patients who received cardiac CT scans for TAVR evaluation at James Cook University Hospital between July 2013 and October 2015. As Williams reported in a presentation at EuroPCR in Paris, France, last month, definite LAA thrombus was found in 11.1%, a “very high” rate. All but two patients with thrombus had a diagnosis of A-fib; definite thrombus was seen in 32% of patients with A-fib and only 1.6% of those without.
Among those who ultimately underwent TAVR, the data also hinted at a higher stroke risk in patients with LAA thrombus; the stroke rate was 20% in those with thrombus and 3.8% in those without. Williams cautioned, however, that that was based on only six strokes total. Two of those patients had LAA thrombus, and both had A-fib that was being treated with warfarin. The remaining four did not have A-fib, were not on oral anticoagulation, and did not have LAA thrombus identified on CT.
Despite the limitations, “it does suggest there may be an increased stroke rate in patients who have atrial appendage thrombus,” Williams said. He said there are two main ways LAA thrombus could be causing strokes after TAVR: the rapid ventricular pacing required for many procedures could be causing the thrombus to detach and embolize or wires could be directly snagging the clot. Williams said that even though there is no equipment placed into the left atrium under normal circumstances, sometimes when positioning the wire in the left ventricle it can go back through the mitral valve into the left atrium and theoretically make its way into the appendage.
Samir Kapadia, MD (Cleveland Clinic, Cleveland, OH), was skeptical of both possibilities. He said rapid pacing could potentially cause embolization, although there are no data to support or refute that. He added, however, that “there is no rhyme or reason that 30-second pacing will do something bad to the appendage.”
As for wire entanglement, Kapadia said the only time there is a risk of wires going near the appendage is when TAVR is performed via transapical access, which is infrequent in current practice. “It’s not an issue,” he said.
Asked whether thrombus in the LAA should be considered a major issue, Lars Søndergaard, MD (Rigshospitalet, Copenhagen, Denmark), pointed to the fact that only two of the six strokes identified in Williams’ study occurred in patients with thrombus, indicating that looking for thrombus routinely would not identify most patients at risk for an event.
It is more important, he said, to ensure that patients with A-fib are on oral anticoagulation at a relevant level before TAVR to minimize the risk of thrombus. If thrombus is seen on CT, however, it could provide support for the use of embolic protection devices, he added. Such devices are currently under investigation and have not yet been approved in the United States.
Søndergaard pointed out that there has been a lot of discussion about reducing not only strokes but also silent brain lesions, which have been shown to occur in 70% to 80% of patients after TAVR. That raises the question of which patients should have a protection device used during the procedure, he said, adding that patients who have A-fib, previous stroke, or very calcified aortic valves and those undergoing valve-in-vale procedures would probably make the list.
Thus, even though the findings of Williams’ study are important, they may not lead to a major change in practice because patients with A-fib—who made up nearly all of the patients with LAA thrombus—would receive a protection device anyway, Søndergaard concluded.
CT Versus TEE
Traditionally, LAA thrombus has been diagnosed using TEE, typically in patients with A-fib being evaluated before cardioversion, catheter ablation, or implantation of an appendage closure device. There is accumulating evidence, however, that CT performs as well as echo for diagnosing LAA thrombus, or at least in terms of ruling it out.
In the TAVR population, cardiac CT has become standard for preprocedural planning, according to Jonathon Leipsic, MD (St. Paul’s Hospital, Vancouver, Canada), president of the Society of Cardiovascular Computed Tomography.
A standard CT scan will already include a look at the LAA and will be sensitive enough to exclude the presence of a thrombus, Leipsic said. To increase specificity, however, a delayed-phase study might be required. That entails doing a second pass 30 to 40 seconds later—after the contrast has had more time to spread—to determine whether an apparent abnormality seen on the first pass is an actual thrombus or was simply incomplete filling of the appendage.
But Williams said he doesn’t think any other tests are needed if the CT image has high enough quality, noting that TEE might not be feasible if conscious sedation is planned. He said that at his center, CT is performed in every patient before TAVR—even those with renal impairment—because the imaging is considered important enough to be worth exposing patients to a small risk of nephropathy.
Chandan Devireddy, MD (Emory University, Atlanta, GA), told TCTMD that pre-TAVR screening mandates cardiac CT angiography in all patients, unless there’s a concern about critical chronic kidney disease. Those patients would undergo 3D TEE.
“In our opinion, every patient that’s going through a TAVR evaluation should be getting some type of diagnostic evaluation that can assess for the presence of left atrial appendage thrombus,” he said. “The important thing is to look. If you’re getting the data and you’re not necessarily paying attention to the other secondary information that’s present, then you’re doing patients a disservice.
“I think what’s more challenging is that once you see thrombus, especially in a CT scan, what do you do with it?” he said.
Delay and Anticoagulate
Practice guidelines don’t provide advice on what to do if LAA thrombus is identified in a patient being evaluated prior to TAVR, creating a “clinical dilemma in decision making,” Devireddy said, adding that the verdict on how to proceed takes into account a patient’s overall presentation and level of risk and input from discussions with the patient and family.
If thrombus is found in a patient who does not have a known diagnosis of A-fib, especially if he or she is not on oral anticoagulation, there is a need to perform some type of heart rhythm monitoring while initiating anticoagulation, he said. After waiting 4 to 6 weeks, the patient would potentially be reevaluated to see whether the thrombus dissolved, following by a risk-benefit discussion, Devireddy said, noting that the stroke risk associated with LAA thrombus in a TAVR population remains unclear.
Other physicians contacted by TCTMD agreed that patients with LAA thrombus who are not already taking anticoagulation should be treated with an oral anticoagulant for several weeks before TAVR is performed.
“That is the current strategy, but there is no very good data for that,” Kapadia said.
As for patients who are already on oral anticoagulation when the thrombus is found, they can either be switched to a different agent—Williams said some patients in his study were switched from warfarin to one of the newer non-vitamin K antagonists—or have their dose adjusted to improve the level of anticoagulation.
Doing TAVR With Thrombus in Place
But in some cases, when patients are very sick, waiting to perform TAVR is not an option. “If [the patient’s procedure is postponed in order] to anticoagulate the clot further, then they take the risk of further decompensation, sudden death, syncope, worsening exacerbation of congestive heart failure, and it’s that balance that we have to find in making these decisions,” Devireddy said.
Williams said that if the decision is made to move forward with TAVR despite the presence of thrombus, consideration will be given to using embolic protection devices. If a patient is not a candidate for one of those devices, however, he or she will need to be informed of the risks and give consent for a procedure with a potentially higher risk for stroke.
Risk of the thrombus embolizing during the procedure could be mitigated by choosing a valve type that does not require rapid ventricular pacing or by avoiding balloon valvuloplasty, which would reduce the amount of rapid ventricular pacing required, Williams said. Also, he said, being meticulous about wire placement will ensure that the thrombus will not be disturbed.
“I would gather that any qualified TAVR operator should not have any wires or equipment in the left atrium,” Devireddy commented. “If that’s happening, especially in a case where you’re aware of the risk of thrombus, then that operator needs to really think about what they’re doing.”
Kapadia said that TAVR can mostly be performed as usual in the presence of LAA thrombus, but pointed out that he would not cardiovert such patients and also would not administer protamine to reverse the effects of heparin because it promotes clotting.
Ripe Area for Research
Considering the questions that remain regarding the impact of LAA thrombus in a TAVR population, it would be worthwhile to explore the issue with larger data sets, Devireddy said.
Investigators from the pivotal TAVR trials could go back and review their data on LAA thrombus if it’s available, he suggested. If not, it’s an issue that could be tackled by multicenter collaborations or by the Society of Thoracic Surgeons/American College of Cardiology TVT Registry, Devireddy said. “Because these patients are being followed so closely, we’re getting very interesting information about the incidence of atrial fibrillation afterwards, the incidence of stroke afterwards, and potentially this is part of that equation.”
According to Kapadia, the most important line of research would be centered on management once LAA thrombus is identified because there is a lot of variation in practice currently.
Leipsic did not agree that LAA thrombus should be routinely ruled out in patients undergoing TAVR—“Right now, I don’t necessarily think the data’s that compelling,” he explained—but said that it “absolutely” is an area worth studying further.
“I think it’s another example of trying to understand where CT could add value and help patients, so I think it’s definitely interesting in scope,” he said.
Williams P, Palmer S. Left atrial appendage thrombus in TAVI: prevalence, clinical impact, and the role of cardiac computed tomography. Presented at: EuroPCR 2016. May 19, 2016. Paris, France.
- Williams reports receiving an educational grant from Edwards Lifesciences and conference support from St. Jude Medical.
- Devireddy reports serving on the scientific advisory board for Medtronic and participating in TAVR trials for various device companies.
- Leipsic reports that his center serves as the CT core lab for Edwards Lifesciences and Medtronic.
- Kapadia and Søndergaard report no relevant conflicts of interest.