Exercise Testing and Exercise Rehabilitation for Patients With Atrial Fibrillation : Journal of Cardiopulmonary Rehabilitation and Prevention

Secondary Logo

Journal Logo

Invited Review

Exercise Testing and Exercise Rehabilitation for Patients With Atrial Fibrillation

Keteyian, Steven J. PhD; Ehrman, Jonathan K. PhD; Fuller, Brittany MD; Pack, Quinn R. MD

Author Information
Journal of Cardiopulmonary Rehabilitation and Prevention 39(2):p 65-72, March 2019. | DOI: 10.1097/HCR.0000000000000423
  • Free

Atrial fibrillation (AF) is a common cardiac arrhythmia associated with an increasing prevalence with advancing age. This review provides a concise discussion of exercise testing and training/rehabilitation considerations for patients with persistent or permanent AF.


Atrial fibrillation is observed in >3 million Americans, <2% in the population that is <65 yr of age, and approximately 10% in the population aged ≥65 yr.1 Among people aged 75 to 84 yr, the prevalence of AF is 12%, and one-third of patients with AF are ≥80 yr of age.1,2 The incidence of AF appears to be greater in people of European decent, with an approximate lifetime risk of 26% for men and 23% for women >40 yr of age. Comparatively, the incidence of AF in African Americans appears to be lower, despite African Americans having a higher prevalence of the more common risk factors (eg, hypertension, diabetes).1 There is a higher prevalence of AF in men than in women, and women typically present with the condition at approximately 5 yr older than men.1–4

Interestingly, despite fewer risk factors, competitive athletes also are at an increased risk for developing AF. A meta-analysis of >9000 patients showed that the risk for having or developing AF was 64% higher among athletes than the general population (OR = 1.64, 95% CI, 1.10-2.43).5 This increase in risk among athletes was observed in younger individuals (eg, <54 yr, OR = 1.96, 95% CI, 1.06-3.65) but not people >54 yr (OR = 1.41, 95% CI = 0.81-2.44). Higher volume, endurance-type activities (ie, long distance running) may also be associated with the increased risk for incident AF, whereas weekly physical activity levels at or slightly above recommended guidelines (ie, 150 min of moderate-intensity activity per week) appear to have protective properties.6


The mechanism of AF is complex and involves ≥1 microreentrant foci of cardiac tissue, usually in the left atrium that results in multiple reentrant electrical circuits leading to fibrillation-like activity.7 Patients with advanced age, hypertension, heart failure, diabetes, ischemic heart disease, valvular heart disease, left ventricular hypertrophy, left atrial enlargement, and hyperthyroidism are at an increased risk for developing AF.1,8 Among athletes, the precise mechanism(s) responsible for the increased risk for AF is unclear and may include alterations in the autonomic nervous system and/or abnormalities in atrial structure or function due to the chronic pressure and volume overload that results from a sustained increase in cardiac output during exercise.

Atrial fibrillation is a complex disorder to manage and associated with exercise intolerance, reduced quality of life, frequent hospitalizations (∼500 000 annually), and thromboembolic events. It is associated with a 5-fold increased risk for stroke, a 3-fold increased risk for heart failure, and a 2-fold increased risk for mortality.1,2,9


Atrial fibrillation is an easily diagnosed condition that is common in patients participating in a rehabilitation program. This section describes several important clinical elements pertaining to AF.


The diagnosis of AF should be suspected when patients have an irregular pulse on physical examination or irregular heart rhythm on electrocardiogram (ECG). Although palpitations, tachycardia, dyspnea, and/or fatigue with exertion often accompany a new diagnosis of AF, they are not always present and are not required to make the diagnosis. When AF is suspected, a 12-lead ECG should be obtained to confirm the diagnosis. This will show no repetitive pattern in the RR interval (ie, an irregularly irregular pattern) and P waves will not be distinct. Atrial fibrillation should not be confused with a pattern of sinus rhythm and frequent premature atrial contractions, which can often have an irregular heart rhythm but, unlike AF, will have distinct P waves for the sinus node-initiated beats.

In the rehabilitation setting, ECG telemetry can be the first indication of a potentially new diagnosis of AF. When a rehabilitation professional believes that new-onset AF is likely, consultation with a patient's physician is required for further evaluation. In an asymptomatic patient, an episode of AF should generally last at least 6 min before being considered a formal diagnosis,9 which then usually involves an echocardiogram and laboratory testing to determine the cause. Identifying the duration/persistence of AF separates patients into 3 general patterns of disease (Table 1). This is an important step because each type has unique implications for treatment. In patients with intermittent palpitations and suspected AF, ambulatory outpatient ECG monitoring for several days or longer may be necessary to document bouts of AF.

Table 1 - Classification of Atrial Fibrillation by Duration and Persistence
Type or Class of Atrial Fibrillation Diagnostic Criteria
Paroxysmal An episode of AF that spontaneously converts back to normal sinus rhythm within 7 d.
Persistent Patient requires electrical or chemical/drug cardioversion to restore normal sinus rhythm.
Permanent Present for >6 mo or the patient and the physician decide to no longer attempt to restore normal sinus rhythm.
Abbreviation: AF, atrial fibrillation.


The initial goal in the treatment of AF is to reduce symptoms (eg, palpitations) and a rapid resting heart rate (HR). A patient's resting ventricular rate on initial presentation with AF is often between 120 and 150 beats/min and is typically treated to reduce the rate to between 60 and 100 beats/min. First-line agents used to achieve rate control include β-adrenergic blocking agents, calcium channel entry blockers, and digoxin (Table 2). However, older adults and those already receiving drugs that slow conduction through the atrioventricular node (eg, β-blockers or calcium channel blockers) may have resting HRs within the normal range.

Table 2 - Common Medications Used in the Treatment of Atrial Fibrillation
Usual Dose Treatment Notes
Rhythm control strategy
Amiodarone Initial load is 10 g
Maintenance dose is 200-400 mg daily
Effective for multiple types of arrhythmias; very slowly metabolized and excreted; multiple side effects that include pulmonary, hepatic, and thyroid fibrosis
Sotalol 80-160 mg twice daily Prolongs QT interval and requires monitoring; contraindicated with reduced ejection fraction or left ventricular hypertrophy; useful in patients with coronary artery disease
Rate control strategy
β-adrenergic blocking agents Comment: Can worsen asthma, hypoglycemia unawareness, erectile dysfunction
Metoprolol 25-100 mg twice daily Most commonly used β-blocker, available in an extended release formula
Carvedilol 3.125-25 mg twice daily Effective antihypertensive; also indicated for heart failure with reduced ejection fraction
Calcium channel blockers (nondihydropyridine) Comment: Should be avoided in heart failure with reduced ejection fraction
Diltiazem 120-420 mg daily (controlled/extended release) Also useful for hypertension and angina; immediate release formula requires dosing every 6-8 hr
Digoxin 0.0625-0.25 mg daily Most useful in combination with β-blockers; requires monitoring with blood-level testing; toxicity possible, particularly in hypokalemia; must reduce dose in patients with kidney disease
Aspirin 81-325 mg daily Mostly ineffective for stroke prevention; use only for patients at very high risk of bleeding
Warfarin (Coumadin) 0.5-10 mg daily Very narrow therapeutic window; requires obtaining frequent blood samples and involves dose adjustments; multiple drug and diet interactions
Direct oral anticoagulants Comment: Not to be used in prosthetic heart valves or mitral stenosis; does not need monitoring; generics not available; minimal drug interactions; reduce dose in patients with kidney disease; drugs are more equivalent than different
Apixaban (Eliquis) 2.5-5 mg twice daily Lowest risk of bleeding; increases diltiazem blood levels by 40%
Dabigatran (Pradaxa) 75-150 mg twice daily Lowest risk of stroke; occasionally causes dyspepsia; reversal agent available
Rivaroxaban (Xarelto) 15-20 mg daily Should be taken with food

If patients remain symptomatic with complaints such as palpitations or fatigue, or prefer restoration of sinus rhythm, additional treatment options are available. This usually involves agents such as amiodarone or sotalol (Table 2), both of which are used in conjunction with direct current electrical cardioversion. These steps often restore normal sinus rhythm and relieve patient symptoms; however, AF often recurs. For patients with recurrent AF and symptoms, invasive procedures such as catheter-based ablation (using heat or extreme cold) may be considered. Catheter-based ablation procedures are curative in 53% to 83% of cases; in selected patients with AF and heart failure, ablation can also reduce heart failure mortality and hospitalizations.10,11

The second major treatment goal of AF is to reduce the risk of a stroke. Atrial fibrillation is a risk factor for stroke because lack of organized atrial electrical activity often leads to slower atrial blood flow (so-called, “blood pooling”), which can increase the risk for intracardiac blood clots; therefore, the potential exists for embolization of a blood clot to the brain or the periphery. Such embolization, however, can be dramatically reduced with adequate anticoagulation therapy or catheter-based closure of the atrial appendage.12,13 For decades, warfarin was used almost exclusively for anticoagulation, but within the last decade several new direct oral anticoagulants have been approved in the United States (Table 2). All of these medications have substantial advantages over warfarin but should not be used in AF that is associated with cardiac valvular disease. Ideally, all treatment decisions should weigh the estimated annual risk of stroke against the treatment-related risk of bleeding. Although the presence of AF has been consistently associated with higher mortality rates and incident heart failure,14 it is unclear whether AF is just a marker for greater underlying disease severity or whether AF independently contributes to poorer outcomes.


Atrial fibrillation can influence physiological responses or adjustments during exercise testing, as well as adaptations to a physical rehabilitation program. These and other important issues are discussed below.


By itself, AF is not an indication for performing an exercise test. However, a graded exercise test can be useful in patients with AF to help assess myocardial ischemia or evaluate chronotropic response to exercise, particularly in those treated for rate control.1,15 For patients with AF enrolled in a rehabilitation program, an exercise test could help better guide exercise training intensity or quantify cardiorespiratory fitness at baseline or in response to physical training. Exercise testing can also help when titrating a medication to control ventricular rate response or to assess for QRS widening among patients starting a class 1C antiarrhythmic drug (eg, flecainide).15,16

Given that there are no procedures specifically recommended for exercise testing in patients with AF, such testing should follow the principles set forth by the American Heart Association15 and the American College of Sports Medicine.16 In the absence of any of the standard indications for stopping an exercise test, patients with AF can be safely exercised up to symptom-limited maximum effort, instead of simply 85% of age-predicted maximum HR. Concerning the latter, it is not uncommon that patients with AF achieve a peak HR that exceeds 100% of age-predicted maximum.17,18 The use of expired gas exchange analysis can be added as a useful tool to more accurately quantify functional capacity. Before testing, those with AF should continue to take medications prescribed for anticoagulation or control of ventricular rate.


No studies have specifically addressed the safety of exercise testing in patients with AF. In 1968, Hornsten and Bruce19 performed exercise treadmill tests on 25 male and 40 female patients with AF and compared their responses to an equal number of patients in normal sinus rhythm. They observed that HR at rest, during submaximal exercise, and at peak exercise was higher in both men and women with AF than in the people in sinus rhythm. No adverse events were reported. Two decades later, Atwood et al20 reported no adverse complications in 29 men with AF who performed a maximal exercise test. Based on these reports and decades of anecdotal experience, and also giving consideration to the test-related procedures summarized in Table 3, a maximal effort exercise test under adequate supervision is likely associated with a low risk for test-related complications. For those people in sinus rhythm, exercise testing–induced AF occurs in <1% of those who perform a test.20

Table 3 - Important Exercise Testing Considerations and Procedures in Patients With Atrial Fibrillation
Acute physiologic adjustments
Because of loss of a normal atrial contraction and an associated reduction in cardiac output, exercise capacity in patients with AF is 15%-20% below that of people in sinus rhythm.
Heart rate at rest, during submaximal exercise, and at peak exercise is often higher in patients with AF than in people in sinus rhythm.
Peak ventricular rate often exceeds age-predicted maximum.
Testing considerations and procedures
Adopt testing principles used for patients without AF.
Patients should undergo testing while taking effective rate-controlling and anticoagulation medications.
Exercise testing can help with titrating medications that target ventricular control during exertion.
Sign- and symptom-limited testing is appropriate.
An exercise test should not be performed in an AF patient with an uncontrolled, tachycardic ventricular rate.
There are relatively limited data addressing the safety of exercise testing in patients with AF; current data suggest that maximal effort under proper supervision is appropriate.
Abbreviation: AF, atrial fibrillation.

Measuring HR at Rest and During Exercise

Heart rate in patients with AF is best measured by ECG recording or cardiac auscultation. With AF, fluctuations in HR occur constantly both at rest and during exercise, thus making the accurate determination of HR a challenge. Atrial rate in patients with AF is between 400 and 600 beats/min and prior to diagnosis or treatment, the ventricular rate is typically high (ie, rapid ventricular rate) at >100 beats/min. Once adequately treated, usually with a β-adrenergic blocking agent, the ventricular rate at rest is reduced to between 60 and 80 beats/min, while the atrial rate remains unaffected. Because of interpatient variability in ventricular response, when determining HR by ECG assessment the technique of extrapolating data from a shorter time (eg, 6 sec or a single RR interval) may provide an erroneously slow or fast HR. Instead, count the number of QRS complexes over 10 sec and multiply by 6 to more accurately determine rate at rest and during exercise. In exercise rehabilitation, ECG telemetry systems that provide a visual display of HR usually measure the RR intervals over a relatively short time period to provide a ventricular rate. For that reason, the displayed HR on the ECG monitor typically fluctuates in patients with AF. The amount of change or fluctuation in HR can vary from 1 patient to the next and is also influenced by the rate controlling medication(s) they are using.

Self-monitoring of HR in patients with AF is challenging. Over a 60-sec period of time, Sneed and Hollerbach21 compared resting HR in patients with AF using (1) standard ECG measurement; (2) palpation of the radial pulse; and (3) auscultation of the chest. When compared with HR measured by ECG, they reported that the apical/chest method was more accurate than radial pulse palpation, with the latter associated with a slower apparent HR (ie, pulse deficit). Specifically, heart beats with shortened R-to-R intervals are associated with less ventricular filling and a resultant lower stroke volume. This makes detecting a radial pulse more difficult due to less blood being ejected into the arterial system. Self-monitoring of HR using radial pulse palpation is not generally recommended because of the inaccuracy associated with the usually short duration of assessment, occurrence of a pulse deficit, and the inability of many patients to palpate their own pulse and count a rapidly changing rate. Although many commercially available technology devices are being investigated for the remote detection and reporting of AF, the accuracy of these devices for measuring HR at rest and during exercise has not been adequately described; therefore, additional research is warranted.22–24

Cardiorespiratory Responses During Exercise

Patients with new-onset AF often experience an excessive ventricular rate response during low-level activities of daily living. This is also observed during the early, low-level stages of an exercise test, even when the resting HR is controlled.15 In AF, the net effect of the loss of a normal atrial contraction (aka, atrial kick) is a reduction in cardiac output due to reduced ventricular filling. To help maintain a normal cardiac output response during exercise, the higher ventricular rate represents a compensatory response to the reduced ventricular filling and the associated lower stroke volume. Atwood et al20 observed that the peak exercise HR of a group of 21 patients with AF and no other known heart disease was significantly higher than that of a group of 29 subjects in sinus rhythm (189 ± 32 vs 166 ± 24 beats/min, P < .01).

To assess the effects of AF on exercise capacity, Atwood and colleagues25 studied 655 patients who underwent cardioversion. Comparisons were made between those who maintained sinus rhythm following cardioversion and those who remained in AF, with multiple tests completed over 52 wk. By week 8, there was a significantly greater (∼15%) exercise capacity (ie, exercise time) among those who converted to sinus rhythm. This difference was sustained over the 52 wk. In addition, patients who converted to sinus rhythm had a significantly lower HR at rest (∼25 beats/min) and at peak exercise (∼40 beats/min) versus the AF group.

Finally, Atwood et al20 noted a significantly lower peak oxygen uptake (o2) in AF patients with known heart disease compared with patients with AF but without heart disease (19.1 ± 5.0 vs 22.7 ± 5 mL/kg/min, respectively, P < .05). Among patients with heart failure due to systolic dysfunction, peak o2 was approximately 20% lower in patients with AF (13.8 ± 3.6 mL/kg/min) than in those in sinus rhythm (17.1 ± 5.6 mL/kg/min).26 In the aforementioned 2 studies, there were no differences in either age or peak respiratory exchange ratio between the study groups, and 1 study included males only and in the other study, 99% were males.20,26 This suggests that individual effort, sex, and age did not play an important role in the observed differences in peak o2.


Two of the hallmark features of AF are exercise intolerance and diminished quality of life, especially when coexisting with other conditions (eg, heart failure, diabetes, or valvular heart disease). To that end and given that regular exercise is known to improve functional capacity and lessen fatigue, it is often helpful to include exercise training in the care of patients with AF if appropriate control of HR has been achieved. With the above in mind, we first point out that there are only a few small prospective randomized controlled trials that describe the effects of exercise rehabilitation in patients with AF. For patients with AF, this section addresses (1) responses to regular exercise training; (2) the use of exercise rehabilitation in the management of the disease; and (3) the essential elements for prescribing exercise.

Responses or Adaptations to Exercise Training

Across a variety of exercise assessments, such as peak o2, 6-min walk test, power output, and measures of activities of daily living, most studies27–32 and meta-analyses33–35 show that aerobic exercise training is consistently associated with modest (10%-15%) improvements in exercise capacity in patients with AF (Table 4). A secondary analysis of patients with heart failure and AF enrolled in the HF-ACTION trial showed small but significant improvements in both peak o2 and 6-min walk distance after 3 mo for patients undergoing exercise training versus controls (o2in mL/kg/min: 5% vs 0%, respectively; 6-min walk distance in m: 6% vs 1%, respectively).29 Using ratings of perceived exertion and HR to guide intensity, Malmo and colleagues30 were the first group to apply higher-intensity interval training in a cohort of patients with AF and a higher level of fitness at baseline (33 mL/kg/min) and showed a 3.2 mL/kg/min (10%) increase in peak o2.

Table 4 - Physiologic and Clinical Responses to Aerobic Exercise Training in Patients With Atrial Fibrillation
Parameter or Outcome Response or Adaptation
Exercise capacity (eg, 6-min walk distance, peak oxygen uptake) Modestly improved, 10%-15%
Skeletal muscle strength, endurance, or histochemistry Insufficient data
Clinical responses: AF burden (time in AF, ventricular rate at rest, and during exercise) Generally favorable responses, with less time in AF and lower ventricular rate at rest and during exercise
Health status or quality of life Generally improved across most submeasures
Clinical outcomes: mortality, rehospitalization Insufficient data
Abbreviation: AF, atrial fibrillation.

Similar to exercise capacity, overall and various subcomponent measures of quality of life and health status are typically improved with exercise training27–35 (Table 4). This includes reductions in the common AF-related symptoms of feeling of tiredness (−26%) and shortness of breath (−33%), as well as symptom frequency and severity.30,32 Using the Kansas City Cardiomyopathy questionnaire, the HF-ACTION trial showed small improvements (+3.1 points) in health status among patients with AF randomized to exercise training versus usual care.29 It is important to point out, however, that the sample sizes and the quality of the evidence in most of the studies reported to date have been rated as low.33 Therefore, additional multisite trial-level evidence involving larger samples is needed to better describe the impact of exercise training on exercise capacity, health status, and quality of life in patients with AF.

A large randomized clinical trial designed to evaluate the safety of exercise training or its effect on clinical outcomes in patients with AF has not been undertaken to date. Younis et al.36 showed that among patients with AF who underwent an exercise stress test before and after cardiac rehabilitation, improvement in peak exercise capacity was associated with a significantly lower adjusted risk for the combined end-point of all-cause mortality or cardiac-related hospitalization. A secondary analysis of the HF-ACTION trial showed that among patients with chronic heart failure, there was no interaction between AF status and exercise training on the clinical outcomes of combined all-cause mortality or hospitalization, all-cause mortality alone, or combined cardiovascular mortality or heart failure hospitalization.29 In a systematic review by Giacomantonio et al,35 there were no major adverse events during exercise training among those with AF; however, several studies included in that analysis did not report safety information.

Exercise Training as a Therapy in the Management of AF

Using different measures to assess AF burden, several studies have investigated the use of exercise training as a therapy to help manage the condition. In a study of 51 patients with paroxysmal or persistent AF, Malmo and colleagues30 used an implanted loop recorder to measure percentage of time in AF during a 4-wk period to assess the effect of 12 wk of higher-intensity interval training versus usual care. They observed that time in AF decreased from 8.1% to 4.8% with exercise training versus an increase in controls from 10.4% to 14.6% (P = .001 for the mean difference in change between groups). Using Holter ECG monitoring in an uncontrolled trial of 10 patients with permanent AF who underwent 4 mo of exercise training, Plisiene et al37 reported a 12% decrease in 24-hr mean ventricular rate. They also reported an approximate 8% decrease in mean ventricular rate during all levels of submaximal exercise.

An uncontrolled, prospective observational study by Pathak et al38 involved 308 obese patients with paroxysmal or persistent AF who were entered into a risk factor management program that included structured exercise; they showed that those who improved cardiorespiratory fitness by ≥2 metabolic equivalents of task over ≥24 mo experienced greater weight loss, blood pressure control, and improved cardiac structure than patients achieving a <2 metabolic equivalents of task gain in fitness. Among patients with a greater gain in fitness, there were also a greater reduction in AF burden (ie, frequency and duration of AF; symptoms) and higher arrhythmia-free survival rates. In this study, the recommendation for exercise volume was 200 min/wk of moderate intensity (ie, up to 85% of age-predicted maximum). Although much more research is needed, in general, regular aerobic exercise seems to have a modest, favorable effect on AF burden.

Prescribing Exercise in Patients With AF

Since a primary goal for exercise training in patients with AF is to reverse exercise intolerance as measured by peak o2, the principle of specificity of training dictates that large-muscle, whole-body activities (eg, walking, cycling) that stimulate the cardiorespiratory system be employed (Tables 5 and 6). In addition, since disorders of skeletal muscle strength, function, and endurance are likely common as well in these patients, consideration for the incorporation of strength training into the overall exercise plan for selected patients may be justified. However, there are no prospective, controlled resistance training-specific trials that specifically involve patients with AF.

Table 5 - Exercise Prescriptions Used in Randomized Controlled Trials Involving Exercise-Based Cardiac Rehabilitation in Patients With Atrial Fibrillation
Study Sample Size; Mean Age; Primary Disorder Exercise Prescription
Frequency (d/wk) Duration Intensity Modalities Comments
Min/wk Number of Weeks
Hegbom et al (2007),32 n = 30; 64 yr; chronic AF 3 225 8 70%-90% maximum HR Aerobic Strengthening exercises for back, thighs, and abdomen
Osbak et al (2011),31 n = 49; 70 yr; permanent AF 3 ≥90 12 70% of maximal exercise capacity; 14-16 on Borg scale C, W, R, IT Subjects encouraged to also do 30 min of light exercise daily
Malmo et al (2016),30 n = 51; 59 yr; nonpermanent AF 3 75 12 4-min work by 4-min recovery interval training, up to 95% peak HR; 11-14 on Borg scale W, R First full, formal evaluation of higher-intensity interval training
Luo et al (2017),29 n = 1984; 59 yr; HF with AF or SR 3 90 for first 3 mo, then 120 for duration of the trial ≥52 11-14 on Borg scale C, W Secondary analysis from an exercise training trial evaluating clinical outcomes in patients with HF
Abbreviations: AF, atrial fibrillation; C, cycling; HF, heart failure; HR, heart rate; IT, interval training; R, running; SR, sinus rhythm; W, walking.

Table 6 - Exercise Prescription for Patients With Atrial Fibrillation
Training Parameter Recommendation
Aerobic or cardiorespiratory
Type of activity Aerobic, dynamic such as walking or cycling
Intensity Rhythm = AF
RPE of 11-14 on Borg 6-20 scale
Talk test; fastest pace, while still being able to comfortably carry on a conversation
Rhythm = Sinus
55%-80% of HRRa
Frequency of sessions per weekb 3-5; most, if not all, days of the week
Duration/sessionb 30-60 min
Muscular strength and endurancec
Type of activity Higher repetition, lower resistance muscle-specific exercises using fixed weight machines and handheld weights for 4-6 primary muscle groups; avoid free weights
Intensity 50%-70% of 1 repetition maximum for lifts involving the hips and lower body; 40%-70% of 1 repetition maximum for lifts involving the upper body
Frequency 1-2 d/wk; 1-2 sets per day for each muscle group or lift; 10-15 repetitions per set
Duration Total time = 12-20 min; contraction speed = 6 sec (3-sec concentric and 3-sec eccentric); rest to work ratio ≥2 to 1
Abbreviations: AF, atrial fibrillation; HRR, heart rate reserve; RPE, ratings of perceived exertion.
aHRR computed as (measured peak heart rate − seated resting heart rate) × percent training level (expressed as a decimal) + seated resting heart rate.
bTarget amount = 150 min/wk.
cNo definitive, disease-specific guidelines currently exist for prescribing resistance training to patients with atrial fibrillation. In the absence of such, training methods advanced for healthy individuals or patients with cardiovascular diseases could be adapted and considered.16,39,40

In addition to the type of exercise, there are 3 other training factors to consider in order to impose the necessary training stimulus or overload39; these factors are intensity, duration, and frequency of training (Table 6). In all patients with AF, the clinical exercise physiologist or other exercise professional responsible for writing the exercise prescription and overseeing patient progress needs to ensure that the volume of exercise performed each week is slowly but consistently adjusted over time.39 For most patients, progressing up to the initially targeted volume of exercise should require between 2 and 3 wk.

Duration and frequency of effort should both be progressively up-titrated, before intensity is increased, to a minimum target amount of 150 min/wk (eg, 30 min, 5 times/wk). With respect to prescribing exercise training intensity, the common approach involving the HR reserve method likely does not apply for 2 reasons. First, such an approach requires a measurement of peak HR from a maximal exercise test, which is not routinely completed for patients enrolled in cardiac rehabilitation. Second, the variable HR response to exercise in AF previously discussed makes titrating training workloads to a prescribed target HR range a challenge in these patients. The latter might be overcome by calculating an HR from an ECG telemetry printout that was recorded over a longer time interval (ie, number of QRS complexes in a 30-sec strip), but this approach may prove impractical during an actual rehabilitation session. Interestingly, a recent controlled trial by Malmo et al30 that involved higher-intensity interval training in patients with nonpermanent AF used HR (set at 85%-95% of measured peak HR) during periods of sinus rhythm. During periods of AF, workloads were simply matched to those trained at during periods of sinus rhythm (regardless of the HR while in AF), aided with the Borg ratings of perceived exertion 6 to 20 scale.

Although prescribing exercise intensity can be a challenge in patients with AF, some valid and helpful tools are available. For patients with heart failure and AF, the American College of Sports Medicine recommends guiding exercise intensity at 11 to 14 on the ratings of perceived exertion scale.41 Shown to be a feasible and valid method to guide exercise intensity across a variety of patient populations, the ratings of perceived exertion scale requires that patients be correctly taught and demonstrate an understanding of how to properly express their level of exertion. Another strategy for guiding exercise intensity in patients with AF is the Talk Test, which has experienced widespread use and validation in other clinical populations and in free-living conditions (Table 6).42–44 This method is based on the premise that it is difficult for patients to carry on a conversation when exercising at or above ventilatory threshold, thus representing the basis for the instruction that patients exercise at the highest intensity at which they can still comfortably carry on a conversation.44–46 For most patients, this represents training at an intensity that approximates 60% to 75% of peak o2.

Regarding resistance training, although 1 controlled trial included muscle strengthening exercises as part of a comprehensive exercise intervention in patients with AF,32 no prospective controlled trials have reported on the responses to resistance training only in these patients. Following treatment with catheter ablation for AF, Risom et al47 randomized 210 patients at 1 mo after the procedure to cardiac rehabilitation, which included some strength exercises, versus usual care. Compared with usual care, peak o2, but not self-rated mental health, was improved with cardiac rehabilitation. Nonserious events, mainly musculoskeletal complaints, were more frequent in the cardiac rehabilitation group, but no data were provided regarding any effects on skeletal muscle function. For clinicians who wish to incorporate resistance training into the care of their patients with AF, it is likely prudent to consider methods advanced for healthy individuals and patients with other cardiovascular diseases (Table 6).16,39,40


Atrial fibrillation is a disorder that is easily diagnosed and associated with well-defined treatment guidelines. However, despite AF representing a common comorbidity among patients enrolled in rehabilitation programs, there is a paucity of controlled trials that use standard laboratory methods and endpoints to evaluate the utility and effectiveness of exercise testing and training in these patients. To that end, the questions surrounding the impact of aerobic and resistance training in patients with AF are many, including safety; quality of life clinical outcomes; and central, autonomic, and peripheral (ie, skeletal muscle, vascular) adaptations. Both larger observational and multicenter trials involving standard and novel evaluation and training methods are needed to better address these issues, with a special emphasis placed on the elderly, women, and vulnerable populations.


Dr Pack was supported by a grant (#1K23HL135440) from the National Institutes of Health.


1. January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2014;64:e1–e76.
2. Piccini JP, Hammill BG, Sinner MF, et al. Incidence and prevalence of atrial fibrillation and associated mortality among Medicare beneficiaries, 1993-2007. Circ Cardiovasc Qual Outcomes. 2012;5:85–93.
3. Potpara TS, Marinkovic JM, Polovina MM, et al. Gender-related differences in presentation, treatment and long-term outcome in patients with first-diagnosed atrial fibrillation and structurally normal heart: the Belgrade atrial fibrillation study. Int J Cardiol. 2012;161:39–44.
4. Andrade JG, Deyell MW, Lee AYK, Macle L. Sex differences in atrial fibrillation. Can J Cardiol. 2018;34:429–436.
5. Ayinde H, Schweizer ML, Crabb V, Ayinde A, Abugroun A, Hopson J. Age modifies the risk of atrial fibrillation among athletes: a systematic literature review and meta-analysis. Int J Cardiol Heart Vasc. 2018;18:25–29.
6. Elliott AD, Maatman B, Emery MS, Sanders P. The role of exercise in atrial fibrillation prevention and promotion: finding optimal ranges for health. Heart Rhythm. 2017;14:1713–1720.
7. Mann D, Zipes D, Libby P, Bonow R. Atrial fibrillation: Clinical features, mechanisms and management. In, Braunwald's Heart Disease: A Textbook of Cardiovascular Medicine. 10th ed. New York, NY: Elsevier; 2015:800–801.
8. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest. 2010;137:263–272.
9. Healey JS, Connolly SJ, Gold MR, et al. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med. 2012;366:120–129.
10. Marrouche NF, Brachmann J, Andresen D, et al. Catheter ablation for atrial fibrillation with heart failure. N Engl J Med. 2018;378:417–427.
11. Leef GC, Perino AC, Cluckey A, et al. Geographic and racial representation and reported success rates of studies of catheter ablation for atrial fibrillation: findings from the SMASH-AF meta-analysis study cohort. J Cardiovasc Electrophysiol. 2018;29:747–755.
12. Connolly SJ. Prevention of vascular events in patients with atrial fibrillation: evidence, guidelines, and practice. J Cardiovasc Electrophysiol. 2003;14(suppl):S52–S55.
13. Reddy VY, Doshi SK, Kar S, et al. 5-year outcomes after left atrial appendage closure: from the PREVAIL and PROTECT AF Trials. J Am Coll Cardiol. 2017;70:2964–2975.
14. Odutayo A, Wong CX, Hsiao AJ, Hopewell S, Altman DG, Emdin CA. Atrial fibrillation and risks of cardiovascular disease, renal disease, and death: systematic review and meta-analysis. BMJ. 2016;354:i4482.
15. Fletcher GF, Ades PA, Kligfield P, et al. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation. 2013;128:873–934.
16. Atwood JE, Myers JN. Atrial fibrillation. In: Durstine JL, Moore GE, Painter PL, Roberts SO, eds. ACSM's Exercise Management for Persons With Chronic Diseases and Disabilities. 4th ed. Champaign, IL: Human Kinetics; 2009:73–78.
17. Pinkstaff S, Peberdy MA, Kontos MC, Finucane S, Arena R. Quantifying exertion level during exercise stress testing using percentage of age-predicted maximal HR, rate pressure product, and perceived exertion. Mayo Clin Proc. 2010;85:1095–1100.
18. Arena R, Myers J, Kaminsky LA. Revisiting age-predicted maximal heart rate: can it be used as a valid measure of effort? Am Heart J. 2016;173:49–56.
19. Hornsten TR, Bruce RA. Effects of atrial fibrillation on exercise performance in patients with cardiac disease. Circulation. 1968;37:543–548.
20. Atwood JE, Myers J, Sullivan M, et al. Maximal exercise testing and gas exchange in patients with chronic atrial fibrillation. J Am Coll Cardiol. 1988;11:508–513.
21. Sneed NV, Hollerbach AD. Accuracy of heart rate assessment in atrial fibrillation. Heart Lung. 1992;21:427–433.
22. Tison GH, Sanchez JM, Ballinger B, et al. Passive detection of atrial fibrillation using a commercially available Smartwatch. JAMA Cardiol. 2018;3:409–416.
23. Koshy AN, Sajeev JK, Nerlekar N, et al. Smart watches for heart rate assessment in atrial arrhythmias. Int J Cardiol. 2018;266:124–127.
24. Bumgarner JM, Lambert CT, Hussein AA, et al. Smartwatch algorithm for automated detection of atrial fibrillation. J Am Coll Cardiol. 2018;71:2381–2388.
25. Atwood JE, Myers JN, Tang XC, Reda DJ, Singh SN, Singh BN. Exercise capacity in atrial fibrillation: a substudy of the Sotalol-Amiodarone Atrial Fibrillation Efficacy Trial (SAFE-T). Am Heart J. 2007;153:566–572.
26. Pardaens K, Van Cleemput J, Vanhaecke J, Fagard RH. Atrial fibrillation is associated with a lower exercise capacity in male chronic heart failure patients. Heart. 1997;78(6):564–568.
27. Mertens DJ, Kavanagh T. Exercise training for patients with chronic atrial fibrillation. J Cardiopulm Rehabil. 1996;16:193–196.
28. Vanhees L, Schepers D, Defoor J, Brusselle S, Tchursh N, Fagard R. Exercise performance and training in cardiac patients with atrial fibrillation. J Cardiopulm Rehabil. 2000;20:346–352.
29. Luo N, Merrill P, Parikh KS, et al. Exercise training in patients with chronic heart failure and atrial fibrillation. J Am Coll Cardiol. 2017;69:1683–1691.
30. Malmo V, Nes BM, Amundsen BH, et al. Aerobic interval training reduces the burden of atrial fibrillation in the short term: a randomized trial. Circulation. 2016;133:466–473.
31. Osbak PS, Mourier M, Kjaer A, Henriksen JH, Kofoed KF, Jensen GB. A randomized study of the effects of exercise training on patients with atrial fibrillation. Am Heart J. 2011;162:1080–1087.
32. Hegbom F, Stavem K, Sire S, Heldal M, Orning OM, Gjesdal K. Effects of short-term exercise training on symptoms and quality of life in patients with chronic atrial fibrillation. Int J Cardiol. 2007;116:86–92.
33. Risom SS, Zwisler AD, Johansen PP, et al. Exercise-based cardiac rehabilitation for adults with atrial fibrillation. Cochrane Database Syst Rev. 2017;2:CD011197.
34. Reed JL, Mark AE, Reid RD, Pipe AL. The effects of chronic exercise training in individuals with permanent atrial fibrillation: a systematic review. Can J Cardiol. 2013;29:1721–1728.
35. Giacomantonio NB, Bredin SS, Foulds HJ, Warburton DE. A systematic review of the health benefits of exercise rehabilitation in persons living with atrial fibrillation. Can J Cardiol. 2013;29:483–491.
36. Younis A, Shaviv E, Nof E, et al. The role and outcome of cardiac rehabilitation in patients with atrial fibrillation. Clin Cardiol. 2018;41:1170–1176.
37. Plisiene J, Blumberg A, Haager G, et al. Moderate physical exercise: a simplified approach for ventricular rate control in older patients with atrial fibrillation. Clin Res Cardiol. 2008;97:820–826.
38. Pathak RK, Elliott A, Middeldorp ME, et al. Impact of CARDIOrespiratory FITness on arrhythmia recurrence in obese individuals with atrial fibrillation: the CARDIO-FIT Study. J Am Coll Cardiol. 2015;66:985–996.
39. Squires RW, Kaminsky LA, Porcari JP, Ruff JE, Savage PD, Williams MA. Progression of exercise training in early outpatient cardiac rehabilitation: an official Statement from the American Association of Cardiovascular and Pulmonary Rehabilitation. J Cardiopulm Rehabil Prev. 2018;38:139–146.
40. Williams MA, Haskell WL, Ades PA, et al. Resistance exercise in individuals with and without cardiovascular disease: 2007 update. Circulation. 2007;116:572–584.
41. Riebe D, Ehrman JK, Liguori G, Magal M. Exercise prescription for patients with cardiovascular and cerebrovascular disease. ACSM's Guidelines for Exercise Testing and Prescription. 10th ed. Philadelphia, PA: Wolters Kluwer; 2018:239.
42. Reed JL, Birnie DH, Pipe AL. Five things to know about exercise training in patients with paroxysmal, persistent or permanent atrial fibrillation. Can Med Assoc J. 2014;186:E558.
43. Reed JL, Pipe AL. The talk test: a useful tool for prescribing and monitoring exercise intensity. Curr Opin Cardiol. 2014;29:475–480.
44. Zanettini R, Centeleghe P, Franzelli C, et al. Validity of the talk test for exercise prescription after myocardial revascularization. Eur J Prev Cardiol. 2013;20:376–382.
45. Brawner CA, Vanzant MA, Ehrman JK, et al. Guiding exercise using the talk test among patients with coronary artery disease. J Cardiopulm Rehabil. 2006;26:72–75.
46. Foster C, Porcari JP, Anderson J. The talk test as a marker of exercise training intensity. J Cardiopulm Rehabil Prev. 2008;28:24–30.
47. Risom SS, Zwisler AD, Rasmussen TB, et al. Cardiac rehabilitation versus usual care for patients treated with catheter ablation for atrial fibrillation: results of the randomized CopenHeartRFA trial. Am Heart J. 2016;181:120–129.

atrial fibrillation; exercise; rehabilitation

© 2019 Wolters Kluwer Health, Inc. All rights reserved.