Lifestyle and Risk Factor Modification for Reduction of Atrial Fibrillation: A Scientific Statement From the American Heart Association

Animal models of obesity support epidemiological observations. Conduction slowing and larger atrial size and pressure contribute to AF vulnerability in humans. Obesity causes similar findings in animal models. In a sheep overfeeding model, increasing weight correlated with increased biatrial volumes, inflammatory infiltrates, TGFβ1 (transforming growth factor-β1), PDGF (platelet-derived growth factor), and fibrosis.²⁴ Overfed, obese, or higher-weight sheep had heterogeneity of activation and conduction velocity slowing, rate-dependent conduction slowing, spontaneous AF episodes, easier AF induction, and longer AF episodes.²⁵ Chronic AF models showed marked increases in pericardial fat volume associated with fat cell infiltration into adjacent myocardium, forming a unique substrate for AF in obesity. In a pig AF model, a high-fat diet resulted in shortening of pulmonary vein effective refractory period and more sustained AF compared with control.²⁶

Several mechanistic studies suggest that pericardial and epicardial fat exerts arrhythmogenicity via stearic acid effects on cardiac ionic currents to shorten the action potential,²⁷ and a direct effect of epicardial fat–associated inflammatory markers can promote AF stability.^28,29 Obesity-related hypertension and diastolic dysfunction can also activate stretch-activated left atrial (LA) channels, increasing AF susceptibility. Moreover, obesity contributes to the development of obstructive sleep apnea (OSA), hypertension, and DM, which independently increase AF risk.³⁰

Weight change is a reliable quantitative measure for tracking the success of weight loss strategies. Several observational studies noted an impressive impact on AF burden by weight loss.³¹ This inspired prospective studies directed toward weight and other risk factor reduction (Table 3 and Figure 2). In the LEGACY trial (Long-Term Effect of Goal-Directed Weight Management in an Atrial Fibrillation Cohort), patients who underwent goal-directed weight loss and risk factor management over 4 years were analyzed on the basis of percent body weight lost.¹³ Patients who lost and maintained the loss of ≥10% of their body weight enjoyed a 6-fold arrhythmia-free likelihood compared with those who lost <3% or gained weight. Weight fluctuation of >5%, but not stable weight, was independently associated with AF recurrence and offset the weight loss effect to some degree. In the REVERSE-AF trial (Prevention and Regressive Effect of Weight-Loss and Risk Factor Modification on Atrial Fibrillation), a subanalysis of LEGACY, patients who had the least percent of body weight lost had the highest progression to persistent AF (48%).¹⁵ In the subcohort of patients with ≥10% weight reduction who started with persistent AF, 88% of patients became paroxysmal or had no AF. Although it is difficult to discern the effect of weight reduction alone, several studies approached AF risk by intervening on multiple risk factors, including obesity. A 15-month prospective randomized controlled trial (RCT) used a risk factor modification scheme, including weight loss (along with OSA, hypertension, tobacco, alcohol, and glycemic control), and measured AF burden, AF symptoms score, and echocardiographic features.³² The overall trend was toward decreased AF symptoms and burden in the intervention group. Similarly, the RACE 3 study (Routine Versus Aggressive Upstream Rhythm Control for Prevention of Early AF in Heart Failure) randomized patients with early persistent AF and HF to conventional therapy or risk factor–driven upstream therapy and cardiac rehabilitation (physical activity, dietary restrictions, counseling).¹⁴ Although this study did not specifically focus on weight loss (therefore, the effect attributable to weight loss alone cannot be defined), sinus rhythm at 1 year after cardioversion by 7-day Holter monitoring occurred in 75% of the upstream and 63% of the conventional group. The ARREST-AF Cohort Study (Aggressive Risk Factor Reduction Study for Atrial Fibrillation) followed up risk factor reduction in patients with a BMI ≥27 kg/m² who had AF ablation.¹² The 41% who elected to undergo risk factor reduction had higher arrhythmia-free survival compared with those who declined this strategy.

Whether intensive lifestyle intervention is effective in reducing incident AF remains to be established. In a secondary analysis of the randomized Look AHEAD trial (Action for Health in Diabetes) of 5067 overweight or obese patients with DM,³³ AF was ascertained from ECGs or discharge summaries; in this study, weight loss and increased physical activity did not affect incident AF. However, bariatric surgery may produce benefits in reduction of AF risk. In the SOS matched cohort study (Swedish Obese Subjects) of 2000 obese subjects who underwent bariatric surgery matched to 2021 obese control subjects, risk of new AF, ascertained from the Swedish National Patient Registry, was 29% lower in the surgery group with more risk reduction in younger subjects and those with higher blood pressure (BP).³⁴ Moreover, a retrospective study of 239 morbidly obese patients undergoing AF ablation reported reduced AF recurrence in patients who had bariatric surgery before ablation.³⁵

Conversely, there appears to be a paradoxical effect with an increased risk of AF in individuals with very low body weight.¹⁶ More specifically, the risk of AF as determined in a large cohort analysis appears to be driven by low lean body mass rather than BMI alone or anthropomorphic obesity patterns such as waist or hip circumference.³⁶

Several gaps in knowledge remain for AF in obese populations. The increase in AF risk by pericardial/epicardial fat accumulation³⁷ requires further mechanistic investigation and expansion to additional cohorts. Defining the mechanism of risk may help tailor recommendations for AF prevention. However, although the most effective weight loss techniques for longitudinal success remain undefined, optimal weight management is considered an important component of AF management. The Australian LEGACY¹³ and ARREST-AF¹² studies suggest that AF burden reduction can be achieved with initial targets of at least a 10% reduction in weight and a BMI <27 kg/m² in conjunction with the management of concurrent risk factors. Although BMI reduction has been targeted, BMI has limitations, as is apparent from sex-based studies and the potential for lean body mass (adjusted by BMI and other measurements) and different adipose depots to contribute to AF. Finally, the relationship, either direct or indirect, between weight loss and AF reduction is an area of active investigation.

The 2018 Physical Activity Guidelines Advisory Committee Report recommends 150 min/wk of moderate-intensity or 75 min/wk of vigorous-intensity aerobic exercise (Table 4) for all adults because this exercise volume improves cardiovascular health.⁴³ Observational and intervention studies indicate that regular aerobic exercise at the levels recommended by the Physical Activity Guidelines Advisory Committee may also reduce the risk of new-onset AF (Table 5). Moreover, physical activity may partially offset the elevated risk of AF associated with obesity.^42,44 In addition, better cardiorespiratory fitness—as a consequence of aerobic exercise—has a graded and inverse relationship to AF burden in both middle-aged and elderly people.⁴⁶ Data are more limited for the reduction of AF burden. CARDIO-FIT (Cardiorespiratory Fitness on Arrhythmia Recurrence in Obese Individuals With Atrial Fibrillation), a cohort study, assessed the impact of cardiorespiratory fitness gain on AF burden in overweight and obese individuals with nonpermanent AF. Higher cardiorespiratory fitness and a gain in cardiorespiratory fitness over time were associated with a greater reduction in AF burden.¹¹ Every metabolic equivalent (MET) gained from baseline to follow-up was associated with a 9% decline in arrhythmia recurrence even after adjustment for weight loss and baseline cardiac risk factors. Improvements in exercise capacity as small as ≥2 METs in overweight individuals on top of weight loss were associated with a 2-fold greater freedom from AF.¹¹

Regular aerobic exercise may also improve AF-related symptoms,^11,47 quality of life,^47,48 and exercise capacity.^47,48 An RCT of 28 patients with persistent AF found that compared with a nonexercise control, 2 months of moderate-intensity continuous training reduced AF-related symptoms and improved quality of life and exercise capacity.⁴⁷ Similarly, another RCT of 49 patients with permanent AF reported that 12 weeks of moderate-intensity continuous training was superior to nonexercise control in increasing exercise capacity and improving quality of life.⁴⁸ A meta-analysis of 5 RCTs reported that exercise training significantly improved exercise capacity, left ventricular ejection fraction, and scores in the General Health and Vitality sections of the Short Form-36.⁴⁹

In summary, regular aerobic exercise is beneficial for preventing and treating AF, although the reach of physical activity and exercise interventions is currently limited in that <10% of adults meet current Physical Activity Guidelines, primarily because of time constraints.⁵⁰

Participation in extreme endurance exercise races has grown in popularity. In parallel, there is growing concern that excessive exercise (such as that undertaken by endurance athletes) may be deleterious to cardiovascular health, including paradoxically increasing AF risk. Although observational studies may be limited by sex bias and are uncontrolled for factors such as height and ethnicity, a meta-analysis of 6 studies showed that athletes had a 5-fold increased risk of AF compared with age-matched controls (odds ratio, 5.3 [95% CI, 3.6–7.9]).⁵¹ A study of 52 755 long-distance cross-country skiers⁵² reported hazard ratios of 1.20 to 1.29 comparing the fastest with slower finishers and the most number of races with 1 race completed, possibly suggesting a gradient of AF risk among extreme exercisers. Most important, this higher risk of AF has been observed only with exercise doses that far exceed recommendations of the Physical Activity Guidelines Advisory Committee Report (Tables 5 and 6) and are therefore achieved by only <1% of Americans.

There has been recent interest in the potential health benefits of high-intensity interval training (HIIT). Compared with moderate-intensity continuous training, potential advantages of HIIT include greater improvements in physiological parameters such as cardiorespiratory fitness,⁵⁶ left ventricular ejection fraction,⁵⁷ diastolic function,⁵⁸ and endothelial/vascular function^58,59 for the same level of energy expenditure; shorter duration; and better time-efficiency. Because lack of time is the most commonly cited reason for poor adherence to regular exercise,⁶⁰ the time-efficiency of HIIT may be advantageous.

A randomized study of 51 patients with nonpermanent AF suggests that HIIT may reduce AF burden in the short term.⁶¹ The time commitment per session and per week was 40 and 120 minutes, respectively. Compared with a control group of patients with AF who continued their previous exercise habits, HIIT was associated with a greater reduction in AF burden after 12 weeks of exercise.⁶¹ However, another randomized study of moderate-intensity continuous training versus HIIT guided by the Borg maximal perceived exertion score over 12 weeks in 76 subjects with nonpermanent AF failed to show a significant difference in AF burden,⁶² yet in another study, patients with coexistent hypertension and kidney disease had more AF occurrence with HIIT compared with moderate-intensity exercise.⁶³

The impact of HIIT on structural and electric remodeling of the LA is worth noting. In a study of 61 sedentary adults randomized to HIIT versus yoga for 10 months,⁶⁴ there was no significant change in atrial electric activity by P-wave signal-averaged ECG in the HIIT group, although changes in echocardiographic LA structure, LA mechanical function, and left ventricular remodeling were noted. Of note, although there was adverse LA and left ventricular remodeling in the HIIT group compared with the yoga group, the LA and left ventricular volume increases resulting from vigorous exercise remained lower than those observed in competitive endurance athletes. In summary, further research is needed to determine whether a longer duration of HIIT remains beneficial in reducing AF burden or risk or whether it may paradoxically induce the electric changes thought to cause AF in endurance athletes.

Multicomponent exercise refers to physical activity that includes >1 type of exercise domain such as aerobic, muscle strengthening, and balance training. Mind-body exercise is a form of exercise that combines body movement, mental focus, and controlled breathing to improve strength, balance, flexibility, and overall health. Examples of multicomponent and mind-body exercises include yoga, tai chi (taiji), and qigong. Given the role of the autonomic nervous system in AF genesis⁶⁵ and the salutary effect of mind-body exercises on cardiac autonomic function,^66–69 yoga and tai chi may enhance the management of AF. Indeed, the YOGA My Heart Study demonstrated that 3 months of yoga training reduced AF burden and symptoms and improved several domains of quality of life.⁷⁰ In a randomized study of HIIT versus yoga, although HIIT adversely affected atrial remodeling, there was no significant change in the yoga group, which was neutral on atrial changes that might improve AF risk.⁶⁴

Although associations between SDB and AF may be driven in part by shared risk factors such as aging, male sex, obesity, hypertension, and HF, with postulated connections including hemodynamic, autonomic, and inflammatory mechanisms, a recent meta-analysis demonstrated an adjusted 2-fold risk of AF among patients with SDB.⁷³ Studies consistently demonstrate a high SDB prevalence in AF populations, ranging from 21% to 87%.^74–80

The bulk of evidence suggests a “dose-response” relationship between SDB severity and AF incidence, burden, and response to treatment. Patients with severe SDB are less likely to respond to antiarrhythmic drug therapy than those with milder forms of SDB.⁸¹ In a study of patients attending a sleep clinic, SDB diagnosis and severity were independently associated with incident AF.⁸²

It is important to screen patients with AF for concomitant SDB. Patients with AF rarely report OSA-related symptoms, and lack of symptoms or excessive daytime sleepiness should not preclude patients from being investigated for the potential presence of concomitant OSA.⁸³ Initial screening could include taking a sleep symptom–oriented clinical history and physical examination or questionnaires,⁸⁴ with referral for further testing as indicated. Although polysomnography is the gold standard for the diagnosis of SDB,⁸⁵ home sleep apnea testing is an alternative method to diagnose SDB.⁸⁶

Although there are few randomized trials, current evidence suggests that treatment of SDB in patients with AF leads to better outcomes. Among patients with SDB and AF who underwent cardioversion, those who received continuous positive airway pressure (CPAP) therapy had a lower recurrence of AF than those who did not.⁸⁷ Similarly, an analysis of the ORBIT-AF (Outcomes Registry for Better Informed Treatment of AF) cohort demonstrated that those with SDB who were using CPAP were less likely to have progression of AF disease to more permanent AF.⁸⁸ A recent meta-analysis demonstrated the consistently lower risk of AF recurrence in patients with SDB who used CPAP versus those who did not,⁸⁹ especially in those who underwent AF ablation. Furthermore, some studies have demonstrated similar outcomes after ablation in CPAP-treated patients and in those without SDB.^89–91

An important weakness of the aforementioned data is that they were derived almost exclusively from observational studies and are therefore prone to potential bias. Raising some concern, the SAVE study (Sleep Apnea Cardiovascular Endpoints), although underpowered for AF, was a large multicenter randomized study (n=2717) that compared CPAP and usual care alone and did not show a reduction in cardiovascular events, including incident new-onset AF, in patients with moderate to severe OSA and established CVD.⁹² Thus, more randomized trials are needed to support stronger recommendations for SDB diagnosis and treatment in patients with AF. In this population, weight loss also plays an important role in reducing SDB severity.⁹³

Data on the mechanistic pathways by which DM can cause AF are scarce. One proposed pathway is the effect of long-term DM on atrial structure and function. The Strong Heart Study showed that subjects with DM had greater echocardiographic left ventricular mass and wall thickness and higher pulse pressure/stroke volume (a surrogate for arterial stiffness) compared with patients without DM.¹⁰¹ Such chronic changes may eventually contribute to atrial dilatation and remodeling. Furthermore, increased production of advanced glycation end products can contribute to atrial fibrosis, a major contributor to AF.^102,103

Besides atrial structural changes, DM may contribute to electric, electromechanical, and autonomic remodeling. In patients with AF undergoing radiofrequency ablation, those with DM had significantly lower right atrial and LA voltages than patients without DM.¹⁰⁴ Patients with DM also have lower atrial reservoir and conduit function, as assessed by strain, and higher interatrial and intra-atrial electromechanical delays.^105,106 As for autonomic differences, the atrial tissue in patients with DM has a greater ability to release acetylcholine and a higher susceptibility to AF after sympathetic simulation.^107,108

Although studies have been limited by their retrospective nature, aggressive glycemic control has been associated with reduced risk of AF development and recurrence. In 12 605 patients with DM, thiazolidinedione treatment over a 5-year duration was associated with reduced AF occurrence risk by ≈30% after adjustment for age and comorbidities.¹⁰⁹ Similarly, a large population-based study of 645 710 Taiwanese subjects showed that metformin use was associated with a 19% lower risk of AF over 13 years compared with control, and metformin reduced cellular myolysis and oxidative stress in paced HL-1 atrial cells.¹¹⁰ Glycemic control may also reduce the risk of repeat ablation, as was shown in subjects with DM taking pioglitazone.¹¹¹ Another study associated higher hemoglobin A_1c levels and worsening 12-month trend in hemoglobin A_1c with AF recurrence after ablation.¹¹² Thus, blood sugar control may be an important strategy for reducing recurrent AF burden. However, the degree of control should follow current guidelines because there is no separate target for AF, and no randomized trials focusing on glycemic control as a sole intervention are available. Further research is needed to assess whether aggressive glycemic control may help reduce AF incidence and recurrence.

Epidemiological and clinical trial substudy data have linked AF risk with hypertension; AF and hypertension share common risk factors.^30,113–115 Joint National Committee guidelines¹¹⁶ recommend BP targets <140/90 and <150/90 mm Hg for those <60 and >60 years of age, respectively, except for patients with comorbid DM or chronic kidney disease. Because of the high prevalence of hypertension, this risk factor provides the highest attributable risk for the development of AF. The population-attributable fraction for AF in the ARIC study (Atherosclerosis Risk in Communities) was 21.6% for elevated BP, 12.7% for BMI, 7.45% for smoking, 8.77% for DM, and 5.35% for history of cardiac disease.¹¹⁵

Poorly controlled BP has been associated with elevated AF risk.^30,115 Clinical trial data suggest linear relationships between BP management and adverse cardiovascular outcomes (ie, “the lower the better”).¹¹⁷ Secondary analyses of randomized anticoagulant trials show variable associations of BP control with mortality in patients with AF. In general, however, studies support BP control as a strategy to lower stroke risk in patients with AF,^118–120 and limited data suggest that sustained exposure to prehypertension (BP, 120–139/80–89 mm Hg) is associated with an increased risk of AF.^121–124 To date, available evidence supports targeting BP at guideline-recommended levels for general cardiovascular health for primary AF prevention.

Elevated circulating levels of angiotensin II and aldosterone lead to inflammation, fibrosis, and anisotropic conduction and result in a medium that fosters AF.¹²⁵ The use of angiotensin-converting enzyme inhibitors and angiotensin receptor blockers as antihypertension therapy has yielded inconsistent results with regard to the primary prevention of AF.^126,127 Mineralocorticoid receptor antagonist treatment was associated with reduced AF risk and recurrence in a meta-analysis of clinical trials and observational studies of mineralocorticoid receptor antagonist treatment (Table 7).¹²⁸

The SMAC AF study¹²⁹ (Substrate Modification With Aggressive Blood Pressure Control) was a randomized, open-label trial of tight BP control compared with standard care in patients undergoing AF ablation. When moderate hypertension was managed as an isolated factor, no difference in arrhythmia control was observed. In contrast, in a small study of 76 patients with severe resistant hypertension and symptomatic AF randomized to AF ablation with or without renal sympathetic denervation, renal denervation was associated with a significant reduction in BP and AF burden at 12 months,¹³⁰ and tight BP control was practiced in LEGACY.¹³ These observations suggest that, in the presence of severe hypertension, targeting this factor may aid arrhythmia control. The ongoing ASAF trial (Ablation of Sympathetic Atrial Fibrillation; NCT02115100) is randomizing 138 hypertensive patients with AF with signs of sympathetic overdrive to pulmonary vein isolation versus pulmonary vein isolation with renal denervation.¹³¹

Hypertension management for AF follows the current guidelines for general cardiovascular health and should include management of contributory lifestyle factors (obesity, physical inactivity, and diet), along with pharmacotherapy.

Initial goals such as 10% weight loss and 2-MET increases in physical fitness to reduce AF burden in overweight and obese patients appear reasonable, as demonstrated by LEGACY, CARDIO-FIT,¹¹ and REVERSE-AF.^13,15 These studies recommended progressive goals to achieve a target BMI of ≤25 kg/m² and initial low-intensity exercise for 20 minutes 3 times weekly, increasing to at least 200 min/wk of moderate-intensity activity. Screening and management of other risk factors follow American Heart Association/American College of Cardiology guidelines. Guideline goals include at least 150 min/wk of moderate-intensity or 75 min/wk of vigorous-intensity physical activity.^187,188 Specific targets for other risk factors are a highly evolving area of continuing research.

For patients who are able, the use of a pedometer or smartphone/watch that can provide activity feedback may be an effective strategy to reduce sedentary activity and increase daily physical activity.¹⁸⁹ Setting a step goal (eg, 5000 steps per day initially and then working up to >10 000 steps per day) can help to sustain effort over time.

Similarly, dietary apps can help patients track their weight and caloric intake and can be useful in promoting weight loss.¹⁹⁰ However, significant and sustained weight loss is rarely accomplished without at least 6 months of sustained effort, so it is essential that patients are advised about, referred to, or enrolled in a comprehensive weight management program.¹⁹¹ For patients who are eligible and interested, bariatric surgery is highly effective at producing sustained weight loss and risk factor reduction¹⁹² and may reduce new AF incidence³⁴ or recurrence after ablation.³⁵

CPAP adherence can be difficult, and among patients with sleep apnea, clinicians should inquire about barriers to CPAP adherence.¹⁹³ Referral to a specialty sleep clinic and several sequential trials of different delivery devices may increase hours of effective CPAP therapy.¹⁹³

Alcohol and tobacco use should be managed as chronic diseases that necessitate multiple repeated interventions.^194,195 Prescription medications such as nicotine replacement therapy for smoking cessation and naltrexone for alcohol use disorder can be considered, and referral to addiction centers is helpful.

The number of factors related to AF management increases its complexity and may result in suboptimal care if only some of the risk factors are addressed adequately. The Euro Heart Survey showed that adherence to established guidelines is poor, which leads to increased morbidity and mortality,¹⁹⁶ highlighting the need for a comprehensive approach to service delivery for patients with AF. Use of or referral to multidisciplinary teams may facilitate intensive and comprehensive lifestyle counseling. If these programs are not available locally, health systems interested in reducing the burden of AF could invest in such treatment teams (Figures 3 and 4).

An RCT demonstrated that a multidisciplinary, integrated, protocol-driven, guideline-based clinic resulted in a 35% relative risk reduction of the composite end point of cardiovascular hospitalization and mortality.¹⁹⁷ The Before and After Study recruited patients with AF from the emergency department to a dedicated AF clinic within 7 days.¹⁹⁸ Management was systematically undertaken to assess causative factors, symptoms, quality of life, and stroke risk. Patients were provided a treatment plan based on guidelines. Compared with a propensity-matched control cohort, there was a significant reduction in the composite end point of all-cause death, cardiovascular hospitalizations, and emergency department visits for AF (odds ratio, 0.71 [95% CI, 0.50–1.0]; P=0.049).

When available, providers may find it helpful to refer patients to structured programs that focus on risk factor management and prevention.^199,200 Studies show that a structured program for appropriate intensive behavioral counseling with accompanying physician encouragement is effective in achieving change.^201–203

Resources for additional information, clinical organizations, and guidelines in lifestyle medicine are given in Table 9. Key references by section may be found in the online data supplement.