JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY VOL. 66, NO.

25, 2015

ª 2015 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER INC. http://dx.doi.org/10.1016/j.jacc.2015.10.047

REVIEW TOPIC OF THE WEEK

Obesity, Exercise, Obstructive Sleep

Apnea, and Modifiable Atherosclerotic
Cardiovascular Disease Risk Factors
in Atrial Fibrillation
Jared D. Miller, MD, Konstantinos N. Aronis, MD, Jonathan Chrispin, MD, Kaustubha D. Patil, MD,
Joseph E. Marine, MD, Seth S. Martin, MD, MHS, Michael J. Blaha, MD, MPH, Roger S. Blumenthal, MD,
Hugh Calkins, MD

ABSTRACT

Classically, the 3 pillars of atrial fibrillation (AF) management have included anticoagulation for prevention of throm-
boembolism, rhythm control, and rate control. In both prevention and management of AF, a growing body of evidence
supports an increased role for comprehensive cardiac risk factor modification (RFM), herein defined as management of
traditional modifiable cardiac risk factors, weight loss, and exercise. In this narrative review, we summarize the evidence
demonstrating the importance of each facet of RFM in AF prevention and therapy. Additionally, we review emerging data
on the importance of weight loss and cardiovascular exercise in prevention and management of AF. (J Am Coll Cardiol
2015;66:2899–906) © 2015 by the American College of Cardiology Foundation.

A trial fibrillation (AF) is the most common

sustained arrhythmia and represents
rapidly growing epidemic (1). It is projected
that the prevalence will rise from 5.2 million in
a
prevention efforts by focusing on risk modification”
as a key research focus (6). Of note, there are
numerous other risk factors for AF, such as valvular
heart disease, thyroid disease, or chronic obstruc-
2010 to 12.1 million in 2030, driven primarily by tive pulmonary disease, which are not thought of as
our aging population (2). AF is associated with classical risk factors for atherosclerotic heart dis-
increased morbidity, mortality, and rising health ease. Such conditions also have known causal
care–associated expenses (3,4). relationships with AF and also have implications for
Unsurprisingly, many modifiable risk factors management strategies. These other AF risk factors
associated with atherosclerotic cardiovascular dis- warrant investigation in the work-up of new di-
ease (ASCVD) have also been associated with AF agnoses of AF.
(Central Illustration). Although the most recent AF
practice guidelines acknowledge these risk factors, OBESITY AND WEIGHT LOSS
they provide no evidence-based recommendations
on diet, exercise, and lifestyle changes in AF pri- The rise in prevalence of obesity has been well

Listen to this manuscript’s

mary and secondary prevention (5). In 2013, the documented, particularly in the United States, where
audio summary by Heart Rhythm Society recognized “advancing AF an estimated one-third of adults are obese (7). During
JACC Editor-in-Chief
Dr. Valentin Fuster.

From the Division of Cardiology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland. Dr.
Calkins is a consultant for Boehringer Ingelheim, Medtronic, and Atricure. All other authors have reported that they have no
relationships relevant to the contents of this paper to disclose.

Manuscript received October 5, 2015; accepted October 8, 2015.

2900 Miller et al. JACC VOL. 66, NO. 25, 2015

Risk Factor Modification in AF DECEMBER 29, 2015:2899–906

ABBREVIATIONS the past 5 years, a number of studies have 13.5 years of follow-up, increased total body fat
AND ACRONYMS established that obesity is closely linked to mass assessed by bioelectrical impendence was
AF risk. Body mass index (BMI) is part of associated with higher incidence of AF (16).
AF = atrial fibrillation
prediction models for new-onset AF (8). The Several mechanisms underlie the association
ASCVD = atherosclerotic
connection between obesity and AF has been between obesity and AF. In a sheep model fed a high-
cardiovascular disease
shown to occur independently of the many caloric diet, obesity was associated with left atrial
BMI = body mass index
comorbidities associated with obesity (9,10). enlargement and fibrosis, atrial inflammatory, and
HTN = hypertension
In the ARIC (Atherosclerosis Risk In Com- lipid infiltration, as well as changes in atrial elec-
MET = metabolic equivalent
munities) study (N ¼ 14,598), 17% of AF risk trophysiological properties, ultimately leading to
OSA = obstructive sleep apnea
was attributed to obesity or overweight sta- increased rates of spontaneous and induced AF (17). In
PVI = pulmonary vein isolation tus (11). In the WHS (Women’s Health Study) an ovine model of obesity sustained for 8 months,
RFM = risk factor modification (N ¼ 34,309), for every 1 kg/m2 increase in obesity was associated with infiltration of the left atrial
BMI, the relative AF risk increased by 5% (10). In the posterior myocardium by epicardial fat and reduced
Women’s Health Initiative cohort of 93,676 females, endocardial voltage, representing a potential substrate
for every 1 kg/m 2 increase in BMI, AF relative risk for AF (18).
increased by 12%. (12). Interestingly, higher levels of In humans, increased BMI has been associated
physical activity attenuated the AF risk conferred by with increased left atrial size (19), which, in turn, is
obesity (13). BMI >35 kg/m 2 was also associated with associated with higher AF risk (20). Increased peri-
increased AF risk by a hazard ratio of 3.50 in young, cardial fat volume has also been described in obese
healthy women (14). individuals and is related to the presence, severity,
In a recent meta-analysis of 51 studies and 626,603 and post-ablation recurrence of AF, independent of
patients, for every 5 kg/m 2 increase in BMI, there was BMI, suggesting a local pathogenic effect of pericar-
a 10% to 29% higher relative risk for new-onset or dial fat (21). Obesity has also been associated with
post-operative AF (15). In a subanalysis of 16 studies increased epicardial fat thickness, which may lead to
and 5,864 patients undergoing AF ablation, the risk of altered atrial electrophysiology and sympathovagal
AF recurrence increased by 13% for every 5 kg/m2 imbalance of the atria (22–24). Clinically, epicardial
increase in BMI (15). Adiposity measures other than fat has been associated with AF (25). Lastly, obesity is
BMI have also been associated with increased AF a state of chronic, low-grade, systemic inflammation
risk. In a Danish registry of 55,273 participants and (26). Systemic inflammation has a key role in

C EN T RA L IL LUSTR AT I ON RFM in AF: The Associations Between Cardiometabolic RFs and AF

Risk Factor Modification in Atrial Fibrillation (AF)

Atrial electroanatomic
AF risk factor (RF) AF AF outcomes
remodeling

Modifiable: Recurrence of AF
Obesity AF-related symptoms
Hypertension
Impaired quality of life
Hyperlipidemia
Diabetes mellitus Stroke and
RF prevention systemic thromboembolism
Low cardiorespiratory fitness
and management Increased mortality
Obstructive sleep apnea
Coronary artery disease Weight loss through Increased morbidity
light to moderate exercise Rising health care-
Nonmodifiable: (not high-intensity exercise) associated expenses
Age and management of
Genetics modifiable RFs

Miller, J.D. et al. J Am Coll Cardiol. 2015; 66(25):2899–906.

Cardiometabolic risk factors contribute to the development and consequences of atrial fibrillation (AF) and can be modified by weight loss, exercise, and management of
comorbid cardiac risk factors. RF ¼ risk factor; RFM ¼ risk factor modification.
JACC VOL. 66, NO. 25, 2015 Miller et al. 2901
DECEMBER 29, 2015:2899–906 Risk Factor Modification in AF

initiation and perpetuation of AF and may contribute loss and AF therapy is likely multifactorial, because
to the association of obesity with AF (27). routine RFM and weight-loss counseling result in
Recent clinical trials demonstrate that weight improvement in other comorbid conditions that are
reduction, in the context of a comprehensive risk also independently associated with AF. However,
factor modification (RFM) plan, has an important role weight loss does result in atrial structural changes,
in AF management (Table 1). In a single-center ran- supporting a direct link between weight loss and
domized trial, 150 overweight patients with symp- improved AF outcomes (34). Currently, the body of
tomatic AF were randomized to weight management evidence is enough to strongly recommend weight
versus lifestyle advice, in addition to standard ther- loss for both prevention and management of AF.
apy for AF (28). The weight management consisted of
a prescribed exercise routine and a low-calorie diet, EXERCISE AND FITNESS
with improvement in BMI from 32.8 to 27.2 kg/m2 at
15 months versus no change for the control group. At The benefits of routine exercise in improving ASCVD
15 months, the weight management group had risk factors have been well established, and guide-
reduced frequency of AF episodes, reduced duration lines recommend regular brisk exercise for this
in AF, and lower symptom severity scores. purpose (35). Observational studies have shown
The LEGACY-AF (Long-Term Effect of Goal increased risk of AF in young athletes or in those
Directed Weight Management on Atrial Fibrillation performing endurance training (36,37). A common
Cohort: A 5 Year Follow-Up Study) study investigated theme in these studies is involvement in regular long-
the impact of RFM and weight loss in longer-term AF duration endurance training that is well beyond the
management (30). Patients with a BMI $27 kg/m 2 and scope of what would be practiced by the typical pa-
AF (N ¼ 825) were offered weight management and tient with AF and other comorbidities.
standard therapy for AF at the discretion of the In the Cardiovascular Health Study, the incidence
treating physician. At 5-year follow-up, a dose– of AF was lower in individuals performing light-
response relationship was seen, with greater weight to-moderate exercise compared with those per-
loss resulting in reduced AF burden and symptoms. forming no exercise (38). There was no difference in
Weight loss $10% was associated with >6-fold AF incidence between individuals performing high-
decrease in arrhythmia-free survival. Fluctuations of intensity exercise versus no exercise. Among those
weight >5% within the study time period offset some, who walked for exercise, walking greater distances
but not all, of the benefits of weight loss. or at a faster pace was associated with a greater
Weight loss has also been demonstrated to reduce reduction in incident AF. In the Women’s Health
AF recurrence after ablation procedures. In the Initiative study, increased physical activity was
ARREST-AF (Aggressive Risk Factor Reduction Study associated with less incident AF and appeared to
for Atrial Fibrillation and Implications for the Outcome mediate some of the relationship between obesity
of Ablation), patients with AF and obesity undergoing and AF (13). Two meta-analyses examining the
catheter ablation were offered the chance to partici- relationship of routine nonendurance exercise and
pate in a group in a nonrandomized fashion (29). At a AF reported no association between exercise and AF
mean follow-up of 42 months after catheter ablation, risk (39,40).
patients in the RFM group had significant reductions in In a cohort of 64,561 adults with a mean follow-up
weight, blood pressure, and lipid profile, and improved of 5.4 years, for each 1 metabolic equivalent (MET)
glycemic control. There was a significant decrease in achieved during treadmill stress testing, there was a
AF frequency, duration, symptoms, and arrhythmia- 7% decreased relative risk for AF development (41). In
free survival with the RFM group. Additionally, obese the recent CARDIO-FIT (Impact of CARDIOrespiratory
patients have been observed to receive at least a 2-fold FITness on Arrhythmia Recurrence in Obese In-
higher effective radiation dose compared with dividuals with Atrial Fibrillation) study, 308 obese
nonobese patients during an AF ablation (33). (BMI $27 kg/m 2) patients with AF were enrolled in a
The epidemiological link between obesity and AF is tailored exercise program. Patients that had high
clear, and there is increasing understanding of the cardiorespiratory fitness had greater arrhythmia-free
pathophysiology linking the 2 conditions. Recent data survival with and without rhythm-control strategies.
support the importance of weight loss in AF man- Patients who improved their fitness level by $2 METs
agement. Of note, weight loss did not occur in isola- had a 2-fold higher probability of AF-free survival, as
tion in the studies discussed earlier, but rather well as lower AF burden and symptom severity
included components of exercise, diet, and modifi- compared with those that improved their fitness
cation of other risk factors. The link between weight by <2 METs (32).
 2902
T A B L E 1 Interventional Studies Assessing the Effect of RFM on AF Outcomes

Risk Factor Modification in AF

Miller et al.
First Author Results (Intervention vs.
(Ref. #) Year Design Patient Population Sample Size Intervention Comparison Follow-up Control Group)

Obesity
Abed et al. (28) 2013 Single-center, Overweight and 150 Weight management General lifestyle 15 months  Greater weight loss (14.3 vs. 3.6 kg)
partially blinded, obese ambulatory (I ¼ 75; advice  Reduced AF burden and severity scores
RCT (weight-loss patients with C ¼ 75) (11.8 vs. 2.6 points and 8.4 and 1.7 points,
counselors were symptomatic AF respectively)
not blinded)  Less frequent AF episodes
 Reduced cumulative AF duration (692 min
reduction vs. 419 min increase)
 Reduced IVS and LAA
Pathak et al. (29) 2014 Single-center, Consecutive patients 149 RFM by a physician-led Information on 42 months  Greater reductions in weight and blood pressure,
nonrandomized, with BMI $27 kg/m2 and $1 RF (I ¼ 61; clinic (HTN control, RFM was as well as improved glycemic control and lipid
placebo- (HTN, glucose intolerance/DM, C ¼ 88) weight management, provided profile.
controlled HLD, OSA, smoking, or alcohol lipid management,  Decreased AF frequency, duration, symp-
trial excess) undergoing initial catheter glycemic control, OSA toms, and symptom severity.
ablation for symptomatic AF, management,  Improved arrhythmia-free survival after a
despite the use of antiarrhythmic smoking and alcohol single or multiple procedures (with or
medication counseling) without the use of anti-arrhythmic agents)
Pathak et al. (30) 2015 Single-center, Consecutive patients with 355 Weight management No control arm 46–48 months  Decreased AF burden and symptom severity
single-arm, symptomatic paroxysmal or in patients that lost $10% of their weight
study persistent AF and BMI $27 kg/m2 compared to those that lost <10%
 Weight loss $10% was associated with a
6-fold greater probability of arrhythmia-
free survival
 Fluctuations of weight >5% offset some,
but not all, benefits of weight loss
Diabetes
Fatemi et al. (31) 2014 Multicentric, Patients with CVD or aged 55 to 79 yrs 10,251 Intensive glucose Standard 4.7 yrs  There was no difference in AF incidence
RCT and had anatomic evidence of (I ¼ 5,040; control strategy between the 2 groups
significant ASCVD, albuminuria, C ¼ 5,042) (HbA1c <6.0%) (HbA1c:  Patients with DM and new-onset AF had a
LVH, or $2 additional RFs (HLD, 7.0%–7.9%) hazard ratio of 2.65 for all-cause mortality
HTN, current smoking status, or
obesity)
Exercise and cardiorespiratory fitness
Pathak et al. (32) 2015 Single-center, Consecutive patients with 308 Weight and risk No control arm 49 months  Greatest arrhythmia-free survival (with and
single-arm symptomatic paroxysmal or factor management without rhythm control strategies) was
study persistent AF and BMI $27 kg/m2 program, structured observed in patients with high
exercise program cardiorespiratory fitness compared with
adequate or low cardiorespiratory fitness
 Decreased AF burden and symptom severity
in patients with cardiorespiratory fitness
gain $2 METs, compared with <2 METs

DECEMBER 29, 2015:2899–906

 2-fold greater probability of arrhythmia-

JACC VOL. 66, NO. 25, 2015

free survival in patients with
cardiorespiratory fitness gain $2 METs,
compared with <2 METs

AF ¼ atrial fibrillation; ASCVD ¼ atherosclerotic cardiovascular disease; BMI ¼ body mass index; C ¼ sample size of the control group; CVD ¼ cardiovascular disease; DM ¼ diabetes mellitus; HLD ¼ hyperlipidemia; HTN ¼ hypertension; I ¼ sample size of the intervention
group; IVS ¼ intraventricular septum thickness; LAA ¼ left atrial area; LVH ¼ left ventricular hypertrophy; METs ¼ metabolic equivalents; OSA ¼ obstructive sleep apnea; RCT ¼ randomized controlled trial; RF ¼ risk factors; RFM ¼ risk factor management.
JACC VOL. 66, NO. 25, 2015 Miller et al. 2903
DECEMBER 29, 2015:2899–906 Risk Factor Modification in AF

Although there is a link between endurance exer- of 6 randomized controlled trials involving 3,557 pa-
cise and AF, this represents a small fraction of tients, statin therapy decreased the relative risk of
patients and a subgroup of people already at low risk incident or recurrent AF by 61% compared with stan-
of developing AF. There are no data to suggest that dard of care (52). In a subgroup analysis, the benefit
routine light or moderate exercise puts patients at from statin use was limited to those with a previous
increased risk of AF, and there is evidence that history of AF and those undergoing cardiac surgery or
light-to-moderate exercise may decrease incident AF. treatment for ACS.
In addition, increased fitness is associated with Prospective data on the role of fish oil in decreasing
reduced AF risk. Given the other cardiovascular the incidence of AF are limited. Two randomized,
benefits of routine exercise, it is logical to recom- double-blind, placebo-controlled trials of high-dose
mend regular, moderate exercise as part of routine AF omega-3 (4 to 8 g/day) failed to show a reduction
prevention and management. in AF recurrence among patients with a history
of paroxysmal or persistent AF (53,54). In meta-
HYPERTENSION
analyses, fish oil supplementation was not associ-
ated with reduction of AF risk (55).
Elevated blood pressure has consistently been one the
Statins are, however, associated with decreased risk
strongest predictors of the development of AF (42).
of post-operative AF. A meta-analysis of randomized
Even high-normal blood pressure has been linked with
controlled and observational studies in cardiac surgery
increased risk of AF (43). The increased afterload leads
including 17,643 patients demonstrated less post-
to both atrial and ventricular structural remodeling,
operative AF with pre-operative statin use (56).
resulting in diastolic dysfunction and left atrial
In total, the data linking abnormal lipid profiles
enlargement and fibrosis (44,45). These changes in
and incident AF has been mixed, as have been studies
turn lead to increased AF (45).
of statins and fish oils in prevention of AF. On the
AF risk reduction via blood pressure control has
basis of the evidence, recommendations for use of
not been consistently shown. Therefore, control of
lipid-lowering agents for the purpose of preventing or
hypertension (HTN) in those without cardiovascular
managing AF is limited to its role in cardiac surgery
disease is not currently recommended for prevention
patients.
of AF (5). However, HTN plays an important role in
the management of AF in regard to thromboembolic
DIABETES
risk. The presence of HTN is a risk factor for stroke in
patients with AF in both the CHADS2 and CHADS2 -
Diabetes is an independent risk factor for AF (57,58).
VASc risk stratification instruments (46). Control of
This has been corroborated with long-term prospec-
HTN with losartan has been associated with a 2-fold
tive population cohort studies. In the Framingham
reduction of stroke rates (47).
Heart Study, diabetes was associated with 40% higher
It is well-documented that HTN is associated with
risk of AF in men and 60% higher risk in women after
increased risk of AF and increased risk of complica-
38 years of follow-up (59).
tions of AF, particularly stroke. Although treatment of
Regarding mechanisms underlying the association
HTN has not been consistently shown to decrease AF
of diabetes and AF, cardiac autonomic neuropathy
risk, it is an important component of reducing throm-
has been implicated by leading to sympathetic over-
boembolic risk and of any AF management strategy.
activity and neural remodeling (60,61). Dysfunctional
CHOLESTEROL cardiac autonomic activity can trigger AF, especially
with changes in vagal tone (62).
Studies assessing associations between dyslipidemia There are limited data on diabetes management and
and incident AF report variable results. Multiple AF risk. The Action to Control Cardiovascular Risk in
studies have shown that low levels of high-density Diabetes study randomized 10,251 patients to intense
lipoprotein cholesterol (<35 mg/dl) are associated glycemic control (HbA 1c <6.0%) or a standard target
with increased risk of new-onset AF (48,49). How- (7.0% to 7.9%). There was no difference in new-onset
ever, other studies have found no association AF between the 2 arms (31). In a prospective study of
between high-density lipoprotein cholesterol and 263 consecutive patients undergoing first-time cath-
incident AF, and increased total cholesterol and low- eter ablation for AF, there was no difference in AF
density lipoprotein cholesterol have been associated recurrence between those patients with and without
with a lower incidence of AF (50,51). diabetes; however, there was a significant increase in
A number of studies have evaluated the roles of procedural complications including stroke, cardiac
statins and fish oil in AF prevention. In a meta-analysis tamponade, and hematomas among those with
2904 Miller et al. JACC VOL. 66, NO. 25, 2015

Risk Factor Modification in AF DECEMBER 29, 2015:2899–906

diabetes (63). Other studies have shown a greater atrial There is clear evidence demonstrating OSA to be a
tachyarrhythmia recurrence rate following catheter significant risk factor for AF. In addition, treatment of
ablation for AF in patients with diabetes (64,65). OSA is an important component of AF management,
In summary, there is currently insufficient evi- particularly when cardioversion or PVI is used. The
dence to suggest that any particular management strength of the evidence warrants consideration of
strategy for diabetes directly affects the risk of routine clinical screening for OSA prior to use of a
developing AF. However, it is reasonable to hypoth- rhythm control strategy.
esize that optimal management of diabetes and
prevention of cardiovascular complications may TOBACCO AND ALCOHOL
indirectly reduce risk of AF.
Studies yield conflicting results in regard to the
OBSTRUCTIVE SLEEP APNEA association of tobacco use and incident AF (71–73).
Alcohol consumption has been associated with
Obstructive sleep apnea (OSA) is highly prevalent in an increased risk of developing AF in a dose-
patients with AF. In a prospective analysis, approxi- dependent manner (74). Although tobacco and
mately 50% of AF patients had OSA, as compared with alcohol have been associated with increased risk of
32% of controls (66). After a multivariate analysis AF, the effect of tobacco cessation or reduction in
looking at traditional risk factors for OSA, AF had alcohol intake on management of AF is less clear.
a greater association with OSA than BMI, HTN, or They are, therefore, components of a comprehen-
diabetes. sive strategy to lower AF risk, but cannot be rec-
Mechanisms by which OSA contributes to AF ommended specifically for the purpose of improving
risk include intermittent nocturnal hypoxemia/ AF outcomes.
hypercapnia, surges in sympathetic tone and blood
pressure during apneic episodes, and increased CONCLUSIONS
inflammation (67,68). All of these factors may
contribute to left atrial remodeling and chamber dila- In addition to anticoagulation, rhythm control, and
tion, contributing to perpetuation of AF. rate control, a fourth pillar of AF management is
Rates of AF recurrence after electrical cardioversion emerging. There is growing evidence supporting
are higher in patients with OSA who are not treated with aggressive RFM, especially weight loss, in the
nocturnal positive-pressure therapy. In a prospective context of a comprehensive RFM plan. These bene-
study of patients with OSA referred for electrical car- fits are seen in AF prevention, success rates in AF
dioversion (N ¼ 39), the 12-month AF recurrence rate management, and in reducing complications of AF,
was 82% in patients who were not appropriately including stroke. Although fundamental in primary
receiving positive-pressure therapy versus 42% in pa- care and cardiology for the management of ASCVD,
tients using appropriate OSA treatment (69). RFM in AF deserves similar recognition. The current
In a prospective study of AF patients with OSA body of evidence supports a comprehensive strategy
referred for index pulmonary vein isolation (PVI) of weight loss, exercise, and fitness, screening for
procedure, arrhythmia-free survival at 1 year was OSA, and treatment of traditional modifiable ASCVD
higher in those who were compliant with positive- risk factors. Further research needs to be performed
pressure therapy (71.9%) compared with patients before making specific recommendations and
who were not (36.7%) (70). Arrhythmia-free survival guidelines on appropriate weight loss and fitness
after PVI in treated OSA patients was similar to targets.
arrhythmia-free survival in patients who did not
have OSA. The risk of AF recurrence after PVI in REPRINT REQUESTS AND CORRESPONDENCE: Dr.
untreated OSA patients was the same as the risk of Hugh Calkins, Division of Cardiology, Johns Hopkins
recurrence in OSA patients with AF who were medi- Hospital, 1800 Orleans Street, Zayed Tower 7125R, Bal-
cally managed without PVI. timore, Maryland 21287. E-mail: [email protected].

REFERENCES

1. Chugh SS, Havmoeller R, Narayanan K, et al. 2. Colilla S, Crow A, Petkun W, et al. Estimates of 3. Benjamin EJ, Wolf PA, D’Agostino RB, et al.
Worldwide epidemiology of atrial fibrillation: a current and future incidence and prevalence of Impact of atrial fibrillation on the risk of death: the
Global Burden of Disease 2010 Study. Circulation atrial fibrillation in the U.S. adult population. Am J Framingham Heart Study. Circulation 1998;98:
2014;129:837–47. Cardiol 2013;112:1142–7. 946–52.
JACC VOL. 66, NO. 25, 2015 Miller et al. 2905
DECEMBER 29, 2015:2899–906 Risk Factor Modification in AF

4. Kim MH, Johnston SS, Chu BC, et al. Estimation electrical remodeling: implications for atrial (from the Action to Control Cardiovascular Risk in
of total incremental health care costs in patients fibrillation. Heart Rhythm 2013;10:90–100. Diabetes Study). Am J Cardiol 2014;114:1217–22.
with atrial fibrillation in the United States. Circ
18. Mahajan R, Lau DH, Brooks AG, et al. Electro- 32. Pathak RK, Elliott A, Middeldorp ME, et al.
Cardiovasc Qual Outcomes 2011;4:313–20.
physiological, electroanatomical, and structural Impact of CARDIOrespiratory FITness on
5. January CT, Wann LS, Alpert JS, et al. 2014 remodeling of the atria as consequences of sus- Arrhythmia Recurrence in Obese Individuals with
AHA/ACC/HRS guideline for the management of tained obesity. J Am Coll Cardiol 2015;66:1–11. Atrial Fibrillation: the CARDIO-FIT study. J Am Coll
patients with atrial fibrillation: a report of the Cardiol 2015;66:985–96.
19. Stritzke J, Markus MRP, Duderstadt S, et al.
American College of Cardiology/American Heart 33. Ector J, Dragusin O, Adriaenssens B, et al.
The aging process of the heart: obesity is the main
Association Task Force on Practice Guidelines and Obesity is a major determinant of radiation dose in
risk factor for left atrial enlargement during
the Heart Rhythm Society. J Am Coll Cardiol 2014; patients undergoing pulmonary vein isolation for
aging: the MONICA/KORA (Monitoring of Trends
64:e1–76. atrial fibrillation. J Am Coll Cardiol 2007;50:
and Determinations in Cardiovascular Disease/
6. Van Wagoner DR, Piccini JP, Albert CM, et al. Cooperative Research in the Region of Augsburg) 234–42.
Progress toward the prevention and treatment of study. J Am Coll Cardiol 2009;54:1982–9. 34. Abed HS, Nelson AJ, Richardson JD, et al.
atrial fibrillation: a summary of the Heart Rhythm Impact of weight reduction on pericardial adipose
20. Marcus GM, Olgin JE, Whooley M, et al. Racial
Society Research Forum on the Treatment and
differences in atrial fibrillation prevalence and left tissue and cardiac structure in patients with atrial
Prevention of Atrial Fibrillation, Washington, DC, fibrillation. Am Heart J 2015;169:655–62.e2.
atrial size. Am J Med 2010;123:375.e1–7.
December 9-10, 2013. Heart Rhythm 2015;12:
21. Wong CX, Abed HS, Molaee P, et al. Pericardial 35. Thompson PD, Buchner D, Pina IL, et al.
e5–29.
fat is associated with atrial fibrillation severity and Exercise and physical activity in the prevention
7. Ogden CL, Carroll MD, Kit BK, et al. Prevalence ablation outcome. J Am Coll Cardiol 2011;57: and treatment of atherosclerotic cardiovascular
of childhood and adult obesity in the United 1745–51. disease: a statement from the Council on Clinical
States, 2011-2012. JAMA 2014;311:806–14. Cardiology (Subcommittee on Exercise, Rehabili-
 lu AS, Çiçek D, Akinci S, et al.
22. Balciog
8. Alonso A, Krijthe BP, Aspelund T, et al. Simple tation, and Prevention) and the Council on
Arrhythmogenic evidence for epicardial adipose
risk model predicts incidence of atrial fibrillation in Nutrition, Physical Activity, and Metabolism
tissue: heart rate variability and turbulence are
a racially and geographically diverse population: (Subcommittee on Physical Activity). Circulation
influenced by epicardial fat thickness. Pacing Clin
the CHARGE-AF consortium. J Am Heart Assoc 2003;107:3109–16.
Electrophysiol 2015;38:99–106.
2013;2:e000102. 36. Mont L, Sambola A, Brugada J, et al. Long-
23. Granér M, Seppälä-Lindroos A, Rissanen A,
9. Gami AS, Hodge DO, Herges RM, et al. lasting sport practice and lone atrial fibrillation.
et al. Epicardial fat, cardiac dimensions, and low-
Obstructive sleep apnea, obesity, and the risk of Eur Heart J 2002;23:477–82.
grade inflammation in young adult monozygotic
incident atrial fibrillation. J Am Coll Cardiol 2007; twins discordant for obesity. Am J Cardiol 2012; 37. Elosua R, Arquer A, Mont L, et al. Sport prac-
49:565–71. 109:1295–302. tice and the risk of lone atrial fibrillation: a case-
control study. Int J Cardiol 2006;108:332–7.
10. Tedrow UB, Conen D, Ridker PM, et al. The 24. Lin YK, Chen YC, Chang SL, et al. Heart failure
long- and short-term impact of elevated body epicardial fat increases atrial arrhythmogenesis. 38. Mozaffarian D, Furberg CD, Psaty BM, et al.
mass index on the risk of new atrial fibrillation: the Int J Cardiol 2013;167:1979–83. Physical activity and incidence of atrial fibrillation
WHS (Women’s Health Study). J Am Coll Cardiol in older adults: the Cardiovascular Health Study.
25. Stojanovska J, Kazerooni EA, Sinno M, et al.
2010;55:2319–27. Circulation 2008;118:800–7.
Increased epicardial fat is independently associ-
11. Huxley RR, Lopez FL, Folsom AR, et al. Abso- ated with the presence and chronicity of atrial 39. Ofman P, Khawaja O, Rahilly-Tierney CR, et al.
lute and attributable risks of atrial fibrillation in fibrillation and radiofrequency ablation outcome. Regular physical activity and risk of atrial fibrilla-
relation to optimal and borderline risk factors: the Eur Radiol 2015;25:2298–309. tion: a systematic review and meta-analysis. Circ
Atherosclerosis Risk in Communities (ARIC) study. Arrhythm Electrophysiol 2013;6:252–6.
26. Yang H, Youm YH, Vandanmagsar B, et al.
Circulation 2011;123:1501–8.
Obesity increases the production of proin- 40. Kwok CS, Anderson SG, Myint PK, et al.
12. Perez MV, Wang PJ, Larson JC, et al. Risk flammatory mediators from adipose tissue T cells Physical activity and incidence of atrial fibrillation:
factors for atrial fibrillation and their population and compromises TCR repertoire diversity: impli- a systematic review and meta-analysis. Int J
burden in postmenopausal women: the Women’s cations for systemic inflammation and insulin Cardiol 2014;177:467–76.
Health Initiative Observational Study. Heart 2013; resistance. J Immunol 2010;185:1836–45. 41. Qureshi WT, Alirhayim Z, Blaha MJ, et al.
99:1173–8. Cardiorespiratory fitness and risk of incident atrial
27. Guo Y, Lip GY, Apostolakis S. Inflammation in
13. Azarbal F, Stefanick ML, Salmoirago- atrial fibrillation. J Am Coll Cardiol 2012;60: fibrillation: results from the Henry Ford Exercise
Blotcher E, et al. Obesity, physical activity, and 2263–70. Testing (FIT) Project. Circulation 2015;131:
their interaction in incident atrial fibrillation in 1827–34.
28. Abed HS, Wittert GA, Leong DP, et al. Effect of
postmenopausal women. J Am Heart Assoc 2014; 42. Brunner KJ, Bunch TJ, Mullin CM, et al. Clinical
weight reduction and cardiometabolic risk factor
3:e001127. predictors of risk for atrial fibrillation: implications
management on symptom burden and severity in
14. Karasoy D, Bo Jensen T, Hansen ML, et al. patients with atrial fibrillation: a randomized for diagnosis and monitoring. Mayo Clin Proc
Obesity is a risk factor for atrial fibrillation among clinical trial. JAMA 2013;310:2050–60. 2014;89:1498–505.
fertile young women: a nationwide cohort study. 43. Grundvold I, Skretteberg PT, Liestøl K, et al.
29. Pathak RK, Middeldorp ME, Lau DH, et al.
Europace 2013;15:781–6. Upper normal blood pressures predict incident
Aggressive risk factor reduction study for atrial
15. Wong CX, Sullivan T, Sun MT, et al. Obesity fibrillation and implications for the outcome of atrial fibrillation in healthy middle-aged men: a
and the risk of incident, post-operative, and post- ablation: the ARREST-AF cohort study. J Am Coll 35-year follow-up study. Hypertension 2012;59:
ablation atrial fibrillation: a meta-analysis of 626, Cardiol 2014;64:2222–31. 198–204.
603 individuals in 51 studies. J Am Coll Cardiol EP 44. Vaziri SM, Larson MG, Lauer MS, et al. Influ-
30. Pathak RK, Middeldorp ME, Meredith M, et al.
2015;1:139–52. ence of blood pressure on left atrial size. The
Long-Term Effect of Goal-Directed Weight
16. Frost L, Benjamin EJ, Fenger-Grøn M, et al. Framingham Heart Study. Hypertension 1995;25:
Management in an Atrial Fibrillation Cohort: A
Body fat, body fat distribution, lean body 1155–60.
Long-Term Follow-Up Study (LEGACY). J Am Coll
mass and atrial fibrillation and flutter. A Danish Cardiol 2015;65:2159–69. 45. Kottkamp H. Human atrial fibrillation sub-
cohort study. Obesity (Silver Spring) 2014;22: strate: towards a specific fibrotic atrial cardiomy-
31. Fatemi O, Yuriditsky E, Tsioufis C, et al. Impact
1546–52. opathy. Eur Heart J 2013;34:2731–8.
of intensive glycemic control on the incidence of
17. Abed HS, Samuel CS, Lau DH, et al. Obesity atrial fibrillation and associated cardiovascular 46. Lip GY, Nieuwlaat R, Pisters R, et al. Refining
results in progressive atrial structural and outcomes in patients with type 2 diabetes mellitus clinical risk stratification for predicting stroke and
2906 Miller et al. JACC VOL. 66, NO. 25, 2015

Risk Factor Modification in AF DECEMBER 29, 2015:2899–906

thromboembolism in atrial fibrillation using a 55. Liu T, Korantzopoulos P, Shehata M, et al. 65. Lu ZH, Liu N, Bai R, et al. HbA1c levels as
novel risk factor-based approach: the Euro Heart Prevention of atrial fibrillation with omega-3 fatty predictors of ablation outcome in type 2 diabetes
Survey on Atrial Fibrillation. Chest 2010;137: acids: a meta-analysis of randomised clinical trials. mellitus and paroxysmal atrial fibrillation. Herz
263–72. Heart 2011;97:1034–40. 2015;40 Suppl 2:130–6.

47. Wachtell K, Lehto M, Gerdts E, et al. Angio- 56. Liakopoulos OJ, Choi YH, Kuhn EW, et al. 66. Gami AS, Pressman G, Caples SM, et al. As-
tensin II receptor blockade reduces new-onset Statins for prevention of atrial fibrillation after sociation of atrial fibrillation and obstructive sleep
atrial fibrillation and subsequent stroke cardiac surgery: a systematic literature review. apnea. Circulation 2004;110:364–7.
compared to atenolol: the Losartan Intervention J Thorac Cardiovasc Surg 2009;138:678–86.e1.
67. Somers VK, Dyken ME, Clary MP, et al. Sym-
For End Point Reduction in Hypertension (LIFE)
57. Dublin S, Glazer NL, Smith NL, et al. Diabetes pathetic neural mechanisms in obstructive sleep
study. J Am Coll Cardiol 2005;45:712–9. mellitus, glycemic control, and risk of atrial apnea. J Clin Invest 1995;96:1897–904.
48. Haywood LJ, Ford CE, Crow RS, et al. ALLHAT fibrillation. J Gen Intern Med 2010;25:853–8.
68. Otto ME, Belohlavek M, Romero-Corral A,
Collaborative Research Group. Atrial fibrillation at 58. Huxley RR, Filion KB, Konety S, et al. Meta- et al. Comparison of cardiac structural and func-
baseline and during follow-up in ALLHAT (Anti- analysis of cohort and case-control studies of type tional changes in obese otherwise healthy adults
hypertensive and Lipid-Lowering Treatment to 2 diabetes mellitus and risk of atrial fibrillation. with versus without obstructive sleep apnea. Am J
Prevent Heart Attack Trial). J Am Coll Cardiol Am J Cardiol 2011;108:56–62. Cardiol 2007;99:1298–302.
2009;54:2023–31.
59. Benjamin EJ, Levy D, Vaziri SM, et al. Inde- 69. Kanagala R, Murali NS, Friedman PA, et al.
49. Alonso A, Yin X, Roetker NS, et al. Blood lipids pendent risk factors for atrial fibrillation in a Obstructive sleep apnea and the recurrence of
and the incidence of atrial fibrillation: the Multi- population-based cohort. The Framingham Heart atrial fibrillation. Circulation 2003;107:2589–94.
Ethnic Study of Atherosclerosis and the Framing- Study. JAMA 1994;271:840–4.
70. Fein AS, Shvilkin A, Shah D, et al. Treatment of
ham Heart Study. J Am Heart Assoc 2014;3:
60. Tesfaye S, Chaturvedi N, Eaton SE, et al., obstructive sleep apnea reduces the risk of atrial
e001211.
EURODIAB Prospective Complications Study fibrillation recurrence after catheter ablation. J Am
50. Lopez FL, Agarwal SK, Maclehose RF, et al. Group. Vascular risk factors and diabetic neurop- Coll Cardiol 2013;62:300–5.
Blood lipid levels, lipid-lowering medications, and athy. N Engl J Med 2005;352:341–50.
71. Krahn AD, Manfreda J, Tate RB, et al. The
the incidence of atrial fibrillation: the atheroscle-
61. Kempler P, Tesfaye S, Chaturvedi N, et al., natural history of atrial fibrillation: incidence, risk
rosis risk in communities study. Circ Arrhythm
EURODIAB IDDM Complications Study Group. factors, and prognosis in the Manitoba Follow-Up
Electrophysiol 2012;5:155–62.
Autonomic neuropathy is associated with Study. Am J Med 1995;98:476–84.
51. Psaty BM, Manolio TA, Kuller LH, et al. Inci- increased cardiovascular risk factors: the EURO-
72. Stewart S, Hart CL, Hole DJ, et al. Population
dence of and risk factors for atrial fibrillation in DIAB IDDM Complications Study. Diabet Med
prevalence, incidence, and predictors of atrial
older adults. Circulation 1997;96:2455–61. 2002;19:900–9.
fibrillation in the Renfrew/Paisley study. Heart
52. Fauchier L, Pierre B, de Labriolle A, et al. 62. Dimmer C, Tavernier R, Gjorgov N, et al. Var- 2001;86:516–21.
Antiarrhythmic effect of statin therapy and atrial iations of autonomic tone preceding onset of atrial
73. Chamberlain AM, Agarwal SK, Folsom AR,
fibrillation a meta-analysis of randomized fibrillation after coronary artery bypass grafting.
et al. Smoking and incidence of atrial fibrillation:
controlled trials. J Am Coll Cardiol 2008;51: Am J Cardiol 1998;82:22–5.
results from the Atherosclerosis Risk in Com-
828–35.
63. Tang RB, Dong JZ, Liu XP, et al. Safety and munities (ARIC) study. Heart Rhythm 2011;8:
53. Nigam A, Talajic M, Roy D, et al., AFFORD efficacy of catheter ablation of atrial fibrillation in 1160–6.
Investigators. Fish oil for the reduction of atrial patients with diabetes mellitus–single center
74. Kodama S, Saito K, Tanaka S, et al.
fibrillation recurrence, inflammation, and oxidative experience. J Interv Card Electrophysiol 2006;17:
Alcohol consumption and risk of atrial fibrilla-
stress. J Am Coll Cardiol 2014;64:1441–8. 41–6.
tion: a meta-analysis. J Am Coll Cardiol 2011;57:
54. Kowey PR, Reiffel JA, Ellenbogen KA, et al. 64. Chao TF, Suenari K, Chang SL, et al. Atrial 427–36.
Efficacy and safety of prescription omega-3 fatty substrate properties and outcome of catheter
acids for the prevention of recurrent symptomatic ablation in patients with paroxysmal atrial fibrilla-
atrial fibrillation: a randomized controlled trial. tion associated with diabetes mellitus or impaired KEY WORDS diabetes, hyperlipidemia,
JAMA 2010;304:2363–72. fasting glucose. Am J Cardiol 2010;106:1615–20. hypertension