August 3rd, 2012
To Repeat — or Not to Repeat — the TEE
Eric Joseph Oligino, M.D. and James Fang, MD
Mrs. K is a physically active 84-year-old woman with a history of hypertension and hyperlipidemia who presented to the ED reporting 1 week of intermittent palpitations, accompanied by dyspnea and atypical chest pain. Physical exam findings were a pulse of 140 bpm, blood pressure 130/80 mm Hg, respirations 10/minute, oxygen saturation 100% on room air, non-distended neck veins, and clear lungs. Cardiac exam revealed irregular/irregular tachycardia with no murmurs or gallops. Mrs. K’s extremities were warm, well-perfused, and without edema. An initial electrocardiogram showed atrial fibrillation with rapid ventricular response.
Mrs. K was admitted for rate control and sent for a transesophageal echocardiogram (TEE) to guide electrical cardioversion. The TEE revealed a small thrombus in the left-atrial appendage but was otherwise unremarkable—normal ventricular function, no significant valvular disease. The cardioversion was canceled, and Mrs. K was discharged home on therapeutic anticoagulation and rate control. At follow-up, she reported diminished exercise capacity and mild dyspnea on exertion since her hospital discharge. Therefore, the plan was to pursue a rhythm control strategy for symptomatic atrial fibrillation.
To determine whether — before electrical cardioversion — to perform a repeat TEE in a patient with atrial fibrillation, a previously identified thrombus in the left-atrial appendage, and 3 weeks of therapeutic anticoagulation, I turned to the ACUTE (Assessment of Cardioversion Using Transesophageal Echocardiography) trial. Here are the details about this trial:
Investigators randomized 1222 patients with AF of at least 2 days’ duration to undergo either a conventional, non–TEE-guided anticoagulation strategy (lasting 3 weeks prior to cardioversion) or an early TEE-guided strategy. Patients in the TEE-guided group who had a thrombus detected were given 3 weeks of warfarin, followed by a repeat TEE. After the repeat TEE, cardioversion was performed only if no thrombus was detected (thrombi were detected in 13.8% of the TEE-guided group).
During a mean follow-up of 8 weeks, incidence of the composite endpoint — cerebrovascular accident, transient ischemic attack, or peripheral embolism — was similar in the two groups (conventional, 0.5%; TEE-guided, 0.8%). The incidence of any bleeding was significantly higher in the conventional group (5.5% vs. 2.9% in the TEE-guided group), but the two groups were similar in major bleeding, death, and functional status. Normal sinus rhythm was maintained in about half of each group at 8 weeks.
Question:
Given that the conventionally treated and TEE-guided groups in the ACUTE trial had a similar rate of embolism at follow-up and that no trial participants had precisely Mrs. K’s profile — thrombus detected on initial TEE, 3 weeks of anticoagulation, and diminished exercise capacity with dyspnea on follow-up — is cardioversion without a repeat TEE an appropriate management strategy?
Response:
August 16, 2012
For this woman, cardioversion would not be appropriate without a repeat TEE, given that the original TEE documented a thrombus (before prolonged anticoagulation had been initiated). The best randomized-trial evidence we have to guide management of this patient comes from the ACUTE trial, despite some limitations.
The ACUTE protocol mandated a repeat TEE when a thrombus was detected on initial TEE. This strategy, which prompted prolonged anticoagulation in 76 of the 619 patients in the TEE-guided group, kept the thromboembolic rate at a very low <1%. Although the rate in the conventional-therapy arm was similarly low, all patients were treated as if they had a thrombus (i.e., assuming the worst-case scenario) and therefore underwent prolonged anticoagulation (3 weeks before plus 4 weeks after cardioversion), rather than the abbreviated anticoagulation strategy used in the TEE-guided arm. The downside was more bleeding, albeit manageable, and decreased immediate restoration of sinus rhythm.
Alternatively, if symptoms and heart rate could be reasonably managed, the patient could have undergone 3 weeks of anticoagulation and then elective cardioversion followed by anticoagulation without TEE (i.e., conventional management). Relevant to this point is the substantial number of patients who spontaneously converted to sinus rhythm (125/603 in the non-TEE arm; 62/619 in the TEE arm) without direct-current cardioversion.
Thus, the ACUTE trial showed that an accelerated strategy of TEE-guided search for thrombus (to limit the duration of anticoagulation and, therefore, its complications) was safe and effective for achieving sinus rhythm in the short term. However, the trial was not an endorsement of TEE cardioversion over conventional management. In fact, at 8-week follow-up, the two groups were similar in their rates of sinus rhythm, functional capacity, and death.
Finally, keep in mind that the ACUTE trial participants were highly selected—clinical equipoise had to exist for the clinician and the investigator. In fact, the trial was stopped early, given low rates of enrollment and events at the time of the interim analysis.
Follow-Up:
August 22, 2012
Mrs. K. had documented therapeutic anticoagulation lasting more than 3 weeks during her post-discharge follow-up, before returning for electrical cardioversion. However, given the small left-atrial thrombus detected on her previous TEE, a repeat TEE was performed. A small thrombus was again identified in the left-atrial appendage, and the electrical cardioversion was canceled. Mrs. K was discharged home on an increased dose of warfarin with a target INR of 2.5 to 3.5. During follow-up, she continued to be symptomatic but did not have any bleeding complications related to the increased INR goal. Mrs. K. is scheduled to return for electrical cardioversion without a repeat TEE in the upcoming weeks.
11 Responses to “To Repeat — or Not to Repeat — the TEE”
NEJM Journal Watch — Recent Cardiology Articles- Persistent Symptoms After Valve Replacement for Severe Aortic Stenosis
- Is Diltiazem Safe in Patients with AF Taking Apixaban or Rivaroxaban?
- Does Lowering High Childhood Lipid Levels Prevent Adverse Cardiovascular Events in Adulthood?
- Self-Expanding Supraannular Valve vs. Balloon-Expandable Valve in TAVR for Small Aortic Annulus
- A New TAVR Prosthesis for Aortic Regurgitation Shows Promise
I prefer to repeat the TEE …. bec i believe in one have to do his best if possible . esp when u are dealing with a case belonging to paracardiology (Heart facts without evidence !)
Since 3 weeks of anticoagulation without the initial TEE results in low risk – what are we really accomplishing with the second TEE? Are there clinicians who would just go straight to cardioversion after 3 weeks of anticoagulation and not a repeat the TEE? If you do repeat, and see a thrombus – do you stay with a rate control strategy or continue the cycle?
I agree that 3-6 weeks of anticoagulation results in low risk of thromboembolism. Repeat TEE might well show persistent thrombus, but would not be clinically helpful, as there is no way to tell if it is “organized” (which was the pre-TEE era clinical rationale for the use of anticoagulation prior to cardioversion: allowing the thrombus to organize to avoid embolism). Thus, repeat TEE would be an additional procedure without demonstrated benefit to the patient, in my opinion.
I would proceed with cardioversion at low risk without a follow-up TEE, remembering (and advising the patient) that there is no “zero risk” scenario.
What are the chances that an 84-year old with AF and no reversible precipitating cause could be realistically kept in sinus rhythm for a reasonable length of time? My perception would be ‘low’, and given some risks of a rhythm control policy, I would suggest that the patient pursue long-term rate control and anticoagulation more thoroughly.
Amen,Dr. Mann. There is no proven role for cardioversion here if one measures outcomes. Some countries have adverse payment-for -proceedure incentives–let’s be clear and honest.
Do we really want to subject this 84 y.o. to amio for life? I’d take rate control/A/C Rx. any day.
New data on rhythm control in elderly looked better, but only after a year. Needs replication.
No. Re-evaluation is important . Cardioversion with persistent thrombosis is ridiculous. But so is not doing anything to decrease the thrombotic risk. Early thrombolysis would have been most reasonable, but not having done that, catheter-directed thrombolysis or clot removal would be most efficacious and safe. Would not cardiovert without clot disposal.
There are two possibile choices:
1)not CVE but only rate control by drugs(digitalis and betablocker if necessary) and OACT,for daylife
2) CVE – expecially if she remains symptomatic as effort-syndrome – only if a repeated TEE shows the trhombus-free left auricula.
I prefer the first choice, for an “even physically active 84-year-old woman”: she has to do the OACT neverthless.. (CHDS score 2 and more) Please, no other surgical or device tool for such a patient: it’s like to cath a fly by bazooka!!
First of all, instead of discussing the need of a second TEE before cardioversion, someone is questioning if cardioversion is even appropriate in this patient. In my opinion, in a patient with a first episode af AF, without a structural heart disease and a dilated left atrium and especially given the fact that it is symptomatic even after 3 weeks of rate control therapy, it is worth to try to obtain the rhythm control (at least make a first try). The only problem is that I’d like to know how much the rate control therapy is being effective (what’s the heart rate after 3 weeks?).
Regarding the question of the need for a second TEE, I think that in front of the evidence of the ACUTE trial, a second TEE is not necessary. The symptoms of diminished exercise capacity with dyspnea are not suggestive of persistence of thrombus in LA but rather show a poorly tolerated AF (if heart rate is well controlled, this strentgthens the indication for cardioversion). Of course, even if not obligatory, in real clinical practice the second TEE could be performed (especially if easily provided by the structure), in order to perform the cardioversion with more serenity. Of course this choiche is not supported by any evidence.
Finally, I have some doubt about the ACUTE trial: it’s not clear how many of the patients in the TEE group who had a thrombus on the first TEE (76 out of 549, 13.8%), had a persistent thrombus on the second TEE performed after 3 weeks of anticoagulation (futhermore, I can’t understand why only 27 of these 76 underwent cardioversion 3 weeks later). This would be very intersting to know in order to assess the indication for a second TEE in the patient of this clinical case.
I would suggest that a second TEE be performed, regardless of the current evidence, just for the sake of common sense and also for forensic reasons…
1/Suppose that you perform an ECV without TEE and the patient gets a stroke! What will you do then? Sue the investigators of the ACUTE trial?! And what about the poor lady?
2/ I agree with the previous colleague from italy: we dont know how good the rate control is, I mean it makes a huge difference if the patient has an average heart rate of 100 or 60 bpm. Based on her current symtomes, she has apparently developped tachycardia induced cardiomyopathy. SO now you have two options left: either improve Rate control (then I agree, no need for an 2nd TEE) or switch to rhythm control (then you will need a TEE because of the reasons mentioned above).
But remember this: rhythm control calls for prophylaxis: flecainide or even worse: amiodarone! whereas rate control can be obtained thith beta blocker +/- cardiac glycoside.
In last resort you still have the kamikaze operation: destruction of the av-node and pace-maker implantation (but I would strongly advise against such a method…).In both cases she will have to receive a lifelong oral anticoagulation due to her CHA2DS2VASC-score above 2.
I can agree with all (except thrombolysis). Data is so limited. Indeed, we dont even lnow that the presence of thrombus merits abortion of cardioversion provided that anticoagulation is immediately administered. We go by the ACUTE strategy because it appears safe. The thrombus is a high risk marker. Postconversion stunning ( either electrical or chemical) requires anticoagulation irrespective of the presence of thrombus. The cardioversion itself does not “eject” the thrombus. Perhaps post conversion stunning and stasis is more likely to promote embolism in the setting of a preexisting fresh thrombus even with ongoing anticoagulation…but this is not clear clinically.
So in real life…would continue rate control agents, attempting to avoid cardioversion. If rhythm control still desirable, would repeat TEE 6 to 8 weeks after first, as thrombus usually will not disappear in 3 weeks and thrombus disappearance will make me feel safer. I might also rationalize the decision, by claiming that an enlarging thrombus would alter therapy (who knows..add low dose ASA?).
I guess I don’t know why anyone is still using warfarin in patients with this issue. Despite all the noise about bleeding issues with dabigatran, when properly used, it is a very useful drug — and rivaroxaban is also an extremely useful drug. Repeating a TEE prior to cardioversion in a patient who has had a demonstrable atrial thrombus is, I think, a no brainer. Why would one have done the initial TEE if not to detect a thrombus, and if one has been identified and the patient has been given 3 weeks of warfarin (good God, don’t most cardiologists recognize that warfarin makes things worse before it makes things better?), could one simply intuit that the thrombus had resolved? I confess that I don’t know if the accumulated data on cardioversion without repeat TEE suggests that 3 weeks of warfarin makes it safe to perform cardioversion in patients with previously demonstrable atrial thrombi, but I would pursue a path that employed an agent with immediate antithrombotic action — not warfarin –and obtain verification that the thrombus has resolved before cardioversion.