TEE Guided Cardioversion of Atrial Fibrillation

Abstract & Commentary

Synopsis: The use of TEE to guide anticoagulant therapy for atrial fibrillation is feasible and safe and offers results comparable to that of a conventional oral anticoagulation strategy.

Source: Klein AL, et al. N Engl J Med. 2001;344: 1411-1420.

The Assessment of Cardioversion Using Transesophageal Echocardiography (ACUTE) study was designed to compare a conventional precardioversion anticoagulation strategy with one that used transesophageal echocardiography (TEE) and short-term anticoagulation therapy prior to cardioversion. Patients with atrial fibrillation of greater than 48 hours duration were eligible for the study if they were at least 18 years of age and were hemodynamically stable. Patients were randomly assigned to 1 of 2 groups. In the conventional treatment group, anticoagulation was begun with warfarin with a target INR of 2-3. Electrical cardioversion with or without preceding antiarrhythmic therapy was performed in these patients after 3 weeks of anticoagulation and then warfarin was continued for 4 weeks post cardioversion. In the TEE group, therapeutic anticoagulation was established, with unfractionated heparin for hospitalized patients or with warfarin for outpatients. A TEE was then performed. If no thrombus was detected, the patients were cardioverted and continued on warfarin for 4 weeks postcardioversion. If a thrombus was detected, there was no direct current cardioversion. These patients were treated with warfarin for 3 weeks and underwent repeat transesophageal echo. If the thrombus was no longer seen, they underwent DC cardioversion at that time.

The primary outcome of the study was a composite end point of cerebrovascular accident, transient ischemic attack, and peripheral embolism. Secondary outcomes were major and minor hemorrhagic events, death from any cause, restoration, and maintenance at 8 weeks of sinus rhythm and functional status using the Duke Activity Status Index. The costs of the 2 treatment strategies were also analyzed. During the course of the study, an interim analysis by an independent data safety monitoring board indicated that the rates of the primary events were too low to achieve sufficient statistical power to detect differences between the groups. Enrollment was therefore terminated after a total of 1222 out of a planned 3000 patients had been enrolled. This paper presents data on these patients. The 2 groups were well matched. The mean age was approximately 64 years, and two-thirds of the patients were male. A little more than 50% of the patients had a history of hypertension and more than 25% of the patients had a history of congestive heart failure. The median estimated duration of atrial fibrillation was 13 days.

There were 603 patients in the conventional treatment group. Three hundred thirty-three of these patients underwent direct current cardioversion more than 3 weeks after randomization. Restoration of sinus rhythm was observed in 266 patients (79.9%). Two hundred seventy patients in the conventional treatment group did not undergo direct current cardioversion. One hundred twenty five of these (20.7%) had converted spontaneously, and 2 underwent chemical cardioversion during the 3 weeks of warfarin. Twenty-six patients declined the cardioversion attempt. Anticoagulation was not therapeutic in 46 patients. Bleeding complications were observed in 8 patients. Thirty-four patients either had difficulties during the cardioversion or were lost to follow-up. Among the 619 patients randomized to TEE guided therapy, 70 patients did not undergo TEE. These patients either spontaneously converted (33 patients), declined (4 patients), had other medical complications (3 patients), could not have a TEE probe passed successfully (7 patients), had medical complications precluding TEE (3 patients), or were lost to follow-up (23 patients) before the TEE was performed. Among the remaining 549 patients, 425 patients underwent a direct current cardioversion attempt within 24 hours after the TEE. Cardioversion was successful in 344 (80.9%) of these patients. In the 124 patients who completed the TEE but did not go on cardioversion, thrombi were detected in 76, 29 spontaneously converted, 8 had pharmacologic conversion, 7 were lost to follow-up, 3 declined, and 1 underwent another surgical procedure. Eventually, 20 of these 124 patients were successfully cardioverted. Cardioversion was accomplished within 3 ± 5 days after enrollment using the TEE guided approach vs. 30.6 ± 10.6 days with the conventional approach.

In the study, the total number of embolic events was low, and there was obviously no statistically significant differences between the groups. There were 5 (0.8%) embolic events in the TEE group vs. 3 (0.5%) events in the conventional treatment group.

A total of 76 patients had right or left heart thrombi detected during TEE. A left atrial thrombus was detected in 67 of the 76 patients (88.2%). None of the patients with a thrombus had an embolic event during the 8-week study. Four of these patients died from various causes, and 2 had hemorrhagic complications of anticoagulation.

Hemorrhagic events were more common in the conventional treatment group. There were 5 major and 14 minor hemorrhagic events among 18 patients in the TEE group and 9 major and 24 minor hemorrhagic complications among 33 patients in the conventional strategy group (P = .03). There were 15 deaths in the TEE group vs. 6 deaths in the conventional strategy group (P = .06). Only 1 death was from an embolic event. Most of the patients who died had significant coexisting conditions.

Although a slightly higher proportion of patients in the TEE arm had restoration of sinus rhythm at some point during the study (71.1% vs 65.2%; P = .03). At 8 weeks, there was no significant difference between the 2 groups in the proportion of patients in whom sinus rhythm had been maintained (52.7% vs 50.4%; P = .43). There was no significant difference between the treatment groups in functional status either at baseline or at 8 weeks.

Klein and colleagues concluded that the use of TEE to guide anticoagulant therapy for atrial fibrillation is feasible and safe and offers results comparable to that of a conventional oral anticoagulation strategy.

Comment by John P. DiMarco, MD, PhD

The ACUTE study provides careful data about the embolic risks and benefits of anticoagulation around the time of cardioversion in patients with atrial fibrillation. It has been recognized for more than 40 years that embolic events were the most frequent serious complications of electrical cardioversion in patients with atrial fibrillation. Although guidelines have been published concerning anticoagulation in this setting, these guidelines were based largely on expert consensus opinion and retrospective, uncontrolled studies. They had not been evaluated in any systematic fashion and their efficacy was uncertain. The ACUTE study investigators hoped that use of transesophageal echo would yield a lower rate of embolic events associated with cardioversion of atrial fibrillation. However, the study failed to show an advantage for the transesophageal echo group in large part due to the extremely low event rate seen in the conventional treatment group. An additional hope that led to the ACUTE study design was that earlier cardioversion would enhance the ability to restore and maintain sinus rhythm. This hope was based on the principle of "electrical remodeling," which means that atrial fibrillation itself produces duration-dependent changes in atrial electrophysiology that predisposes to more resistant atrial fibrillation. However, this hypothesis was not proven in the trial. At 8 weeks, there was no significant difference in the proportion of patients who were in sinus rhythm.

The conclusions that we should take from ACUTE are that either the conventional approach using 3 weeks of preconversion anticoagulation and a TEE guided approach can be selected based on patient circumstances. Patients who are highly symptomatic in atrial fibrillation, despite appropriate therapy to control rate should be referred for a TEE guided approach to lessen the duration of symptoms. The TEE guided approach would also apply to patients who have a high risk for bleeding with anticoagulation in whom any possible shortening of the anticoagulation period might be beneficial. Patients who have only minor symptoms while in atrial fibrillation however, can still be managed effectively by the conventional approach of outpatient anticoagulation and cardioversion.

Hopefully, within the next few years, more effective agents for preventing embolic events during atrial fibrillation will be developed and brought to market. These hopefully will be safer than the current strategies, which involve warfarin yet remain highly effective. For now, the choice of either a TEE or a conventional strategy can be based on individual patient considerations.