Atrial flutter: Risk of thromboembolism and role of anticoagulation

INTRODUCTION — 

Anticoagulation to reduce the risk of embolic stroke and other thromboembolic events is a cornerstone of management for patients with atrial flutter (AFL). Our approach to anticoagulation for patients with AFL is described here. (See "Electrocardiographic and electrophysiologic features of atrial flutter".)

Other aspects of the diagnosis and management of atrial flutter are discussed separately:

●(See "Atrial flutter: Overview of diagnosis and management" and "Electrocardiographic and electrophysiologic features of atrial flutter".)

●(See "Control of ventricular rate in atrial flutter".)

●(See "Atrial flutter: Restoration of sinus rhythm".)

●(See "Atrial flutter: Maintenance of sinus rhythm".)

●(See "Atrial fibrillation and flutter after cardiac surgery".)

THROMBOEMBOLIC RISK

Magnitude of risk

Risk of thromboembolic events — The risk of thromboembolic events in individuals with AFL is greater than in the general population without atrial arrhythmias and similar to the risk in individuals with atrial fibrillation (AF) [1-4]. Many individuals with AFL also have episodes of AF. These risks are illustrated by the following studies:

●In a prospective study of 112 participants in the Framingham Heart study with AFL, AFL (compared with lack of AFL and AF) was associated with increased 10-year risk of AF (101 versus 21 per 1000 person-years; adjusted hazard ratio [HR] 5.01, 95% CI 3.14-7.99), stroke (28 versus 15 per 1000 person-years; adjusted HR 2.17, 95% CI 1.13-4.17), myocardial infarction (adjusted HR 3.05, 95% CI 1.42-6.59), heart failure (adjusted HR 4.14, 95% CI 1.90-8.99), and mortality (adjusted HR 2.00, 95% CI 1.44-2.79) [5]. The risk of stroke in participants with AFL was similar to the risk of stroke among those with AF (28 versus 32 per 1000 person-years; adjusted HR 0.9, 95% CI 0.5-1.7).

●In a population-based retrospective cohort study, stroke occurred during three-year follow-up in 4.1 percent of 9339 patients with typical AFL and in 1.2 percent of 7248 adults in a matched general population cohort (HR 3.6, 95% CI 2.8-4.7) [1]. During three-year follow-up, AF occurred in 40.4 percent of patients with AFL compared with 3.3 percent of the general population (HR 21.8, 95% CI 18.3-25.0 percent).

Limitations of risk estimates — One problem in interpreting data on thromboembolic events in patients with AFL is that many patients with AFL also have episodes of AF, as described above. Patients with isolated AFL may have lower thromboembolic risk than patients with a combination of AFL and AF. (See 'Risk of thromboembolic events' above.)

This issue was illustrated by a Medicare database study which found an increased risk of stroke in patients with AFL (relative risk 1.41 compared with a control group [2]). In these patients, the relative risk of stroke was 1.56 in patients who subsequently had an episode of AF (similar to the risk with AF alone), while those with isolated AFL had a stroke risk similar to that for the control population (relative risk 1.11) (figure 1).

Another issue in identifying the risk of thromboembolism associated with AFL is that most patients with AFL have concomitant cardiovascular disease (eg, hypertension and atherosclerotic vascular disease), which may impact the risk of stroke. AFL without concurrent cardiovascular or pulmonary disease (previously called lone AFL) is uncommon (eg, 1.7 percent of a population of adults with AFL [6]). Risk factors for stroke are discussed separately. (See "Stroke: Etiology, classification, and epidemiology".)

Prevalence of thrombus — The prevalence of atrial thrombus and associated thromboembolic risk in patients with AFL are related not only to the thrombus formation during AFL but also thrombus formation during AF, which is common in patients with AFL. Many patients with AFL have alternating periods of AF, making it difficult to know the exact risk of thrombus formation (and subsequent embolization) specifically attributable to AFL [2,7].

Atrial mechanical function is not normal with AFL, though it is not as impaired as with AF. As demonstrated by transmitral and left atrial appendage Doppler recordings, organized atrial and atrial appendage mechanical function is present during sustained AFL and absent during AF [7]. Shorter AFL cycle length and larger atrial size are associated with lower left atrial appendage emptying velocity and greater risk of left atrial spontaneous echogenic contrast and/or left atrial thrombus identified by transesophageal echocardiography (TEE) [8].

TEE evidence of atrial thrombi has been documented in a number of reports of patients with AFL not receiving chronic anticoagulation [7,9-14]. As with AF, nearly all atrial thrombi in patients with AFL are located within the left atrial appendage. The frequency with which thrombus occurs may vary with the duration of the arrhythmia and other risk factors, as illustrated by the following observations:

●Two series evaluated patients with AFL for a mean duration of 33 to 36 days who did not have a history of AF, rheumatic heart disease, or a prosthetic heart valve [9,10]. Left atrial appendage thrombus was found in 1 to 1.6 percent, right atrial appendage thrombus in 1 percent, and left atrial appendage spontaneous echocardiographic contrast in 11 to 13 percent [9,10]. In one of these reports, there was a close correlation between a history of thromboembolism and periods of AF in patients with AFL [10].

●Left atrial thrombus was present in 5 of 47 consecutive patients (11 percent) with AFL for a mean duration of 28 days who did not have a history of AF or mitral stenosis [12].

●Atrial thrombi and spontaneous echocardiographic contrast may be more common in patients with AFL of longer duration. In a TEE study of 30 patients with persistent AFL (duration 6.4 months), two patients (7 percent) had thrombus in the left atrial appendage, and 25 percent had spontaneous echocardiographic contrast prior to cardioversion [11].

Thromboembolic risk factors — Although evidence is more limited in patients with AFL compared with patients with AF, risk factors for stroke appear to be the same in patients with AFL as for patients with AF.

In the above-cited retrospective cohort study, among patient with isolated AFL (ie, with no documented AF), clinical characteristics associated with risk of stroke included older age, female sex, hypertension, diabetes mellitus, coronary artery disease, and chronic kidney disease [1]. Two scores used to stratify thromboembolic risk in patients with AF (CHADS₂ and CHA₂DS₂-VASc scores) were higher among patients with isolated AFL with stroke compared with those without stroke. Among patients with isolated AFL, a CHA₂DS₂-VASc score ≥2 was associated with a three-year stroke risk of 5.3 percent (95% CI 4.6-6.0 percent) compared with 2.5 percent (95% CI 1.9-3.0 percent) in those with a CHA₂DS₂-VASc score of 0 or 1. A higher CHA₂DS₂-VASc score was also associated with a higher risk of development of AF (42.3 percent, 95% CI 40.8-43.8 versus 37.9 percent, 95% CI 36.2-39.6).

These findings support managing thromboembolic risk in patients with AFL in a manner similar to that for AF, as discussed below. (See 'Indications for anticoagulation' below and 'Alternatives to anticoagulation' below.)

Cardioversion — The risk of thromboembolism following cardioversion of AFL to sinus rhythm is likely increased compared with the risk with ongoing AFL. A systematic review of studies of patients with AFL undergoing cardioversion found a short-term rate of thromboembolic events ranging from 0 to 7.3 percent during follow-up time ranging from 2 to 42 days [3]. The thromboembolic risk of cardioversion for pure AFL is uncertain given heterogeneity in study populations and interventions [3,4,9,15-19]. Many patients included in reports of AFL cardioversion-related events also had episodes of AF (but happened to be in AFL at the time of cardioversion). Some studies included patients with a prior history of thromboembolism and were thus more likely to report high event rates. On the other hand, studies in which at least some patients were anticoagulated or underwent precardioversion TEE to assess for thrombus were more likely to report low event rates.

Embolization may be related to a transient reduction in left atrial appendage mechanical function after cardioversion, which may lead to thrombus formation, referred to as atrial "stunning," and is present after successful cardioversion of AFL [11,12,20-22]. In one report, left atrial appendage peak ejection velocity fell by 26 percent within 15 minutes of cardioversion and almost 50 percent of subjects had new or more pronounced spontaneous echocardiographic contrast [11]. These changes predispose to de novo thrombus formation [15]. Similar observations have been made in patients with AF. (See "Hemodynamic consequences of atrial fibrillation and cardioversion to sinus rhythm", section on 'Atrial stunning'.)

The severity of atrial stunning appears to be somewhat less pronounced in AFL than in AF, which could explain the lower embolic risk after cardioversion in AFL. In a report that compared 19 patients with AFL with 44 patients with AF, the left atrial appendage peak ejection velocity was significantly higher in the patients with AFL at baseline (42 versus 28 cm/s in AF) and after cardioversion (27 versus 15 cm/s) [20]. In addition, new or more pronounced spontaneous echocardiographic contrast was less likely in those with AFL (21 versus 50 percent for AF). Nearly all atrial thrombi among patients with either AFL or AF were located in the left atrial appendage.

Catheter ablation — Catheter ablation is a key intervention for maintenance of sinus rhythm in patients with AFL. However, there is significant short- and long-term risk of thromboembolism after catheter ablation.

Atrial thrombus formation after catheter ablation for AFL is likely related to atrial stunning [22,23]. The likelihood of developing atrial stunning and its duration were assessed in a review of 15 patients with persistent AFL (mean duration 17 months) and seven with paroxysmal AFL who underwent radiofrequency catheter ablation [22]. Significant left atrial appendage stunning and spontaneous echocardiographic contrast on TEE were observed after ablation in 80 percent of those with persistent AFL but in none with paroxysmal AFL, suggesting that, like AF, left atrial stunning in AFL is related to the duration of the arrhythmia and not the mode of reversion. These changes resolved after three weeks of sustained sinus rhythm. (See "Atrial flutter: Maintenance of sinus rhythm", section on 'RF catheter ablation'.)

Risk of AF and stroke — While catheter ablation for AFL offers high success rates for preventing recurrent AFL (see "Atrial flutter: Maintenance of sinus rhythm", section on 'RF catheter ablation'), there are risks of development of stroke and of AF after ablation. In a retrospective study following 218 patients for three years after ablation for AFL, AF occurred in 21.6 percent, and the incidence of stroke was between 0 and 2.3 percent [1].

INDICATIONS FOR ANTICOAGULATION

General approach — The approach to anticoagulation for patients with AFL is similar to that for AF [24,25]. (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation" and "Atrial fibrillation in adults: Use of oral anticoagulants".)

●With AF – In patients with AFL and prior or intercurrent AF, the approach to anticoagulation for AF applies. (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation" and "Atrial fibrillation in adults: Use of oral anticoagulants".)

●For isolated AFL – Although there is limited evidence on management of thromboembolic risk in patients with AFL without any documented AF, we and others manage thromboembolic risk in patients with isolated AFL similarly to those with AF [24,25]. (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation" and "Atrial fibrillation in adults: Use of oral anticoagulants".)

●Assessment of potential risk and benefit – For each patient, the estimated absolute risk reduction for thromboembolic events is weighed against the estimated increase in absolute risk of intracranial hemorrhage and other life-threatening or severely debilitating bleeding complications.

●CHA₂DS₂-VASc risk score – Our approach to deciding whether to prescribe anticoagulant therapy for patients with AFL (excluding those with a separate indication for anticoagulation, such as rheumatic mitral stenosis that is severe or clinically significant [mitral valve area ≤1.5 cm²], a bioprosthetic valve [surgical or transcatheter] within the first three to six months after implantation, or a mechanical heart valve) is as follows (see "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'Approach to deciding whether to anticoagulate'):

•For a CHA₂DS₂-VASc score ≥2 in males or ≥3 in females (calculator 1) (table 1), we recommend long-term oral anticoagulant (OAC).

•For a CHA₂DS₂-VASc score of 1 in males and 2 in females (calculator 1) (table 1):

-For patients with CHA₂DS₂-VASc score of 1 in males and 2 in females based on age 65 to 74 years or heart failure, we recommend long-term OAC. Age 65 to 74 years and heart failure are stronger risk factors than the other factors conferring one CHA₂DS₂-VASc score point.

-For patients with other risk factors, the decision to anticoagulate is based upon the specific nonsex risk factor and the burden of AFL. For patients with very low burden of isolated AFL (eg, AFL that is well documented as limited to an isolated episode that may have been due to a reversible cause such as recent surgery, heavy alcohol ingestion, or sleep deprivation), it may be reasonable to forgo long-term OAC and institute close surveillance for recurrent AFL or AF, although it may not be possible to reliably estimate AFL and AF burden from surveying symptoms or infrequent monitoring.

•For patients with a CHA₂DS₂-VASc of 0 in males or 1 in females (calculator 1) (table 1), we suggest against OAC. Patient values and preferences may impact the decision. For example, a patient who is particularly stroke averse and is not at increased risk for bleeding (see "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'Bleeding risk') may reasonably choose OAC, particularly if the patient is a candidate for treatment with a direct oral anticoagulant (DOAC).

●Bleeding risk – For all potential candidates for OAC, bleeding risk and related possible contraindications to OAC should be reviewed (table 2 and table 3). (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'Bleeding risk'.)

The bleeding risk assessment should identify any modifiable bleeding risk factors that can be mitigated and to flag high-bleeding-risk patients for early review and follow-up. (See "Risks and prevention of bleeding with oral anticoagulants", section on 'Bleeding risk scores' and "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'Mitigating bleeding risk'.)

●Choice of anticoagulant – Although there are limited data on OAC therapy in patients with AFL, anticoagulant selection for patients with AFL is the same as for patients with AF. Of the DOACs evaluated in large clinical trials for stroke prevention in AF, only apixaban enrolled patients with AF or AFL [26]. (See "Atrial fibrillation in adults: Use of oral anticoagulants".)

At the time of cardioversion — Anticoagulation leading to, at the time of, and after cardioversion of AFL is managed in a manner similar to that for AF. (See "Prevention of embolization prior to and after restoration of sinus rhythm in atrial fibrillation".)

After radiofrequency catheter ablation — Anticoagulation following catheter ablation of AF is discussed separately (algorithm 1). (See "Atrial flutter: Maintenance of sinus rhythm", section on 'Anticoagulation after RF catheter ablation'.)

ALTERNATIVES TO ANTICOAGULATION

Left atrial appendage occlusion — Though unproven in an exclusively AFL cohort, left atrial appendage exclusion is an option for patients with AFL who are poor oral anticoagulation (OAC) candidates. As discussed separately, left atrial appendage occlusion is the primary alternative to anticoagulation for patients with AFL or AF who have an indication for anticoagulation but have a contraindication for long-term OAC (excluding those with a concurrent indication for anticoagulation such as severe or clinically significant rheumatic stenosis, a bioprosthetic valve [surgical or bioprosthetic] within the first three to six months after implantation, or a mechanical valve). Nearly all the evidence on the efficacy and safety of left atrial appendage occlusion has been developed in patients with AF. (See "Atrial fibrillation: Left atrial appendage occlusion".)

Lack of pharmacologic alternative — For patients with AFL and/or AF who have an indication for OAC to reduce thromboembolic risk, no other antithrombotic regimen is an effective and safe alternative to standard therapeutic OAC. In the setting of AFL and/or AF, nonanticoagulant antithrombotic regimens (eg, antiplatelet agents) are less effective in lowering thromboembolic risk than standard therapeutic OAC, and some of these investigated alternative antithrombotic regimens entail a bleeding risk similar to or greater than standard therapeutic OAC. (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'Pharmacologic agents'.)

RECOMMENDATIONS OF OTHERS — 

The 2020 European Society of Cardiology guidelines for the management of AF and the 2023 American College of Cardiology/American Heart Association/American College of Clinical Pharmacy/Heart Rhythm Society guideline for the diagnosis and management of AF recommend managing anticoagulation in patients with AFL in a manner similar to that for patients with AF [24,27], recognizing that no report has been sufficiently large to accurately define both the risk of embolization and benefit of antithrombotic therapy in a pure AFL population.

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Atrial fibrillation" and "Society guideline links: Arrhythmias in adults".)

SUMMARY AND RECOMMENDATIONS

●Thromboembolic risk with atrial flutter – The risk of thromboembolic events in individuals with atrial flutter (AFL) is greater than in the general population without atrial arrhythmias and similar to the risk in individuals with atrial fibrillation (AF) [1-4]. Of note, episodes of AF occur in many individuals with AFL; thromboembolic risk in these patients may be largely related to episodes of AF.

●Anticoagulation – The approach to anticoagulation for patients with AFL is similar to that for AF.

•With AF – In patients with AFL and prior or intercurrent AF, the approach to anticoagulation for AF applies. (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation" and "Atrial fibrillation in adults: Use of oral anticoagulants".)

•For isolated AFL – Although there is limited evidence on management of thromboembolic risk in patients with AFL without any documented AF, we and others manage thromboembolic risk in patients with isolated AFL similarly to those with AF [24,25]. (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation" and "Atrial fibrillation in adults: Use of oral anticoagulants".)

●Assessment of potential risk and benefit – For each patient, the estimated absolute risk reduction for thromboembolic events is weighed against the estimated increase in absolute risk of intracranial hemorrhage and other life-threatening or severely debilitating bleeding complications. For all potential candidates for oral anticoagulation (OAC), bleeding risk and related possible contraindications to OAC should be reviewed (table 2 and table 3). (See "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'Bleeding risk'.)

The bleeding risk assessment should identify any modifiable bleeding risk factors that can be mitigated and flag high-bleeding-risk patients for early review and follow-up. (See "Risks and prevention of bleeding with oral anticoagulants", section on 'Bleeding risk scores' and "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'Mitigating bleeding risk'.)

●CHA₂DS₂-VASc risk score – Our approach to deciding whether to prescribe anticoagulant therapy for patients with AFL (excluding those with a concurrent indication for anticoagulation) is similar to the approach for patients with AF, although supporting data are more limited for patients with AFL (see "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'Approach to deciding whether to anticoagulate'):

•For a CHA₂DS₂-VASc score ≥2 in males or ≥3 in females (calculator 1) (table 1), we recommend long-term OAC (Grade 1B).

•For a CHA₂DS₂-VASc score of 1 in males and 2 in females (calculator 1) (table 1):

-For patients with this score based on age 65 to 74 years or heart failure, we recommend long-term OAC (Grade 1B). Age 65 to 74 years and heart failure are stronger risk factors than the other factors conferring one CHA₂DS₂-VASc score point.

-For patients with other risk factors, the decision to anticoagulate is based upon the specific nonsex risk factor and the burden of AFL. For patients with very low burden of AFL (eg, AFL that is well documented as limited to an isolated episode that may have been due to a reversible cause such as recent surgery, heavy alcohol ingestion, or sleep deprivation), it may be reasonable to forgo long-term OAC and institute close surveillance for recurrent AFL or AF, although it may not be possible to reliably estimate AFL burden from surveying symptoms or infrequent monitoring.

•For patients with a CHA₂DS₂-VASc of 0 in males or 1 in females (calculator 1) (table 1), we suggest against OAC (Grade 2C). Patient values and preferences may impact the decision. For example, a patient who is particularly stroke averse and is not at increased risk for bleeding (see "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'Bleeding risk') may reasonably choose anticoagulation, particularly if the patient is a candidate for treatment with a direct oral anticoagulant (DOAC).

●Choice of anticoagulant – Although data are limited in patients with AFL, anticoagulant selection for patients with AFL is the same as for patients with AF. (See "Atrial fibrillation in adults: Use of oral anticoagulants".)

●Anticoagulation after catheter ablation – Management of anticoagulation following catheter ablation is discussed separately. (See "Atrial flutter: Maintenance of sinus rhythm", section on 'Anticoagulation'.)