Omega-3 and Risk of Atrial Fibrillation: Vagally-Mediated Double-Edged Sword

Abstract

Objective:

Studies regarding effects of omega-3 fatty acids, specifically eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), on risk of atrial fibrillation (AF) have reported discordant results. The aim of this review is to clarify effects of marine omega-3 intake on risk of AF.

Patients and Methods:

A PubMed search was performed using terms: atrial fibrillation, omega-3, EPA, DHA, vagal tone. We summarized findings from randomized clinical trials (RCTs), epidemiology studies, and meta-analyses evaluating effects/associations of DHA+EPA on risk of AF. Also, vagal tone was explored as a mediator between omega-3 and risk of AF.

Results:

Meta-analyses of 8 RCTs and 17 prospective cohort studies comprised of 83,112 and 54,799 individuals, respectively, investigated the link between omega-3 intake and incident AF. The RCTs reported that treatment with DHA and/or EPA was associated with a 24% increased relative risk of AF (absolute risk 4.0% vs 3.3%; relative risk [RR] 1.24, 95% confidence interval [CI] 1.11–1.38, p=0.0002). This was dose-dependent; DHA+EPA doses of ~1,000 mg/d increased AF risk ~12%, whereas 1,800 to 4,000 mg/d increased AF risk by ~50%. In contrast, observational studies focused on DHA+EPA blood levels or dietary intake have generally reported that higher omega-3 levels/consumption are associated with lower AF risk. Maximal AF risk reduction (12%) occurred at ~650 mg/d of dietary DHA+EPA. Other studies have indicated that omega-3 fatty acids can dose-dependently increase vagal tone, which could explain the biphasic relationship between DHA+EPA and AF risk. Experimental studies show that low-level vagal stimulation decreases risk of AF, whereas high-level vagal stimulation increases risk of AF.

Conclusion:

Higher consumption of dietary omega-3 is associated with decreased AF risk. In contrast, pharmaceutical dosing of omega-3 increases AF in a dose-dependent manner, which may be mediated by vagal tone.

Introduction

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, with 3 million new cases worldwide annually, and a global prevalence of >37 million people.¹ AF puts patients at increased risk for hospitalization, heart failure, thromboembolic events/stroke, and premature death.¹ Prior studies assessing the association of omega-3 fatty acids (FA), specifically eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), and the risk of AF have been inconsistent—some studies have reported that omega-3 increases AF risk,^{2, 3} while others showed a reduced risk or no effect.⁴

Oral intake of DHA and EPA stimulates a dose-dependent increase in vagal tone, with effects apparent starting at relatively low doses (~500 mg/d).^5–7 Low-level vagal nerve stimulation is associated with a reduced risk of AF,⁸ whereas high-level vagal nerve stimulation is a reliable way to induce AF in animal models.⁹ This suggests a mechanism whereby high-dose versus low-dose omega-3 might be associated with divergent effects on risk of AF. The aim of this review is to clarify the effects of marine omega-3 intake on the long-term risk of AF and explore the role of vagal tone as a mediator of this relationship.

Methods:

We performed two literature searches in PubMed, from inception through September 2023, to query the relationships between omega-3, atrial fibrillation, and vagal tone. The first search utilized the terms “omega-3” and “atrial fibrillation.” The titles, abstracts, full texts, and references were screened to identify either meta-analysis, or randomized controlled trials (RCT) with at least 500 participants, or prospective observational studies employing omega-3 biomarker data. The second literature search utilized the terms “omega-3,” “eicosapentaenoic acid,” “EPA,” “docosahexaenoic acid,” “DHA,” “vagal,” “vagus,” “autonomic” and “parasympathetic.” The titles, abstracts, full texts, and references were screened to identify research examining the relationship between omega-3 and autonomic tone. This query was not restricted by language and the data are publicly available.

Findings

Recent meta-analyses focused on incident AF risk associated with marine omega-3 treatment (DHA and/or EPA) among 83,112 individuals from 8 large, cardiovascular (CV) RCTs (Table 1). ^{2, 3} Overall, pharmacologic omega-3 treatment was associated with a 24% increased relative risk (RR) of AF (p=0.0002) (Figure 1). ^{2, 3} This relationship, however, demonstrated a dose-dependent response. The risk for AF was mildly increased ~12% in the 5 RCTs (VITAL, RP, GISSI-HF, ASCEND, DO-HEALTH) testing ~1,000 mg/d of DHA+EPA, trials, OMEMI,¹⁵ REDUCE-IT¹⁶ and STRENGTH.¹⁷ Both STRENGTH and REDUCE-IT used dosages of ~4,000 mg/d of DHA and/or EPA, and OMEMI used 1,800 mg/d. Among the omega-3 FAs, EPA at high doses may be particularly prone to increasing risk of AF.¹⁸

Table 1 -.

RCTs in Meta-analysis of Omega-3: Baseline Characteristics ³

Year	Trial Name	n-3 FA dose (mg/d)	EPA/DHA	Study Size	Follow-up (mo)	Mean Age (yr)	Secondary Prevention (%)	AF in Treatment Arm (%)	AF in control arm (%)
2013	Risk and prevention	1	473/600	Treatment n=6239 Control n=6266	60	64	30%	1.8%	1.5%
2013	GISSI-HF	1	386/401	Treatment n=2921 Control n=2914	46.8	66	40%	15.2%	14.0 %
2018	ASCEND	0.84	460/380	Treatment n=7740 Control n=7740	88.8	63	0%	2.1%	1.7%
2019	REDUCE-IT	4	4000/0	Treatment n=4089 Control n=4090	58.8	64	71%	5.3%	3.9%
2020	STRENGTH	4	2200/800	Treatment n=6539 Control n=6539	42	62.5	55%	2.2%	1.3%
2020	OMEMI	1.8	930/660	Treatment n=372 Control n=387	24	75	100%	7.5%	3.9%
2020	DO-HEALTH	1	330/660	Treatment n=1073 Control n=1084	35.9	74.9	0%	7.3%	6.2%
2020	VITAL Rhythm	0.84	460/380	Treatment n=12,542 Control n=12,577	63.6	67	0%	3.7%	3.4%

Figure 1 -.

Meta-analysis of association between omega-3 fatty acid treatment with atrial fibrillation. Forest plot shows overall pooled data as well as by subgroups of lower dose (≤ 1 g/d) and higher dose (> 1 g/d). Trials included in the analysis were ASCEND; DO-Health; GISSI-HF; Risk and Prevention; VITAL Rhythm; OMEMI; REDUCE-IT; STRENGTH.³

Several RCTs have tested the effects of short-term, moderate-dose omega-3 supplementation on the risk of AF among patients undergoing elective cardiac surgery or with paroxysmal AF. A meta-analysis of these trials reported that a relatively brief course of omega-3 FA compared to placebo was not associated with either an increased or decreased risk of post-operative or recurrent AF, although heterogeneity was present.¹⁹

Many large and methodologically robust observational studies have focused on levels of EPA and DHA circulating in the blood, which reflects chronic intake of marine omega-3 FA. These investigations generally reported that higher blood levels of DHA+EPA were associated with lower risk of AF. These studies typically had > 10 years of follow-up, with a only a small percentage of study subjects taking fish oil supplements or pharmaceutical omega-3.⁴

Recently, FORCE (the Fatty Acids and Outcomes Research Consortium) published a pooled analysis of de novo-generated data from 17 prospective cohort studies from around the world using baseline data on blood levels of omega-3 FA and risk for AF. Among the 55,214 participants there were 7,720 incident cases of AF during a median follow-up of 13.3 years (Table 2).⁴ In this harmonized, multivariable meta-analysis, there was a 12% decrease in risk of AF for patients in the top quintile of DHA+EPA blood levels compared to the bottom quintile (Figure 2).⁴ Similarly, a study based on DHA+EPA levels in adipose tissue reported inverse associations between omega-3 and risk of AF.²⁰

Table 2 -.

Baseline Characteristics of Studies Assessing Omega-3 Fatty Acid Biomarker Levels and Incident AF ⁴

Study	Country	Study Design	Baseline Years	AF Cases, n/N	Median Follow-up, y	Mean Age, y	Women %	Mean BMI kg/m	Baseline ASCV D, %b	Baseline HF, %
60YO	Sweden	Prospective Cohort	1997-1999	3987/581	16.5	60.0	51.7	26.8	8.2	1.2
ARIC	USA	Prospective Cohort	1987	3821/896	29.1	54.0	52.0	27.1	5.3	3.5
CHS	USA	Prospective Cohort	1992-1993	3526/1459	9.5	74.9	60.0	26.8	24.3	4.9
DCH	Denmark	Prospective Cohort	1993-1997	3187/183	13.5	56.7	46.1	26.2	3.0	0.2
EPIC- Norfolk	UK	Prospective Cohort	1993-1998	7383/1070	14.3	63.3	19.2	26.6	5.0	-
FHS	USA	Prospective Cohort	2005-2008	2488/329	11.3	65.9	55.5	28.2	13.1	1.1
Hisayama	Japan	Prospective Cohort	2002-2003	3126/153	9.0	62.0	57.5	23.0	5.6	-
HPFS	USA	Prospective Cohort	1994	1529/64	11.4	64.6	0.0	25.8	0.0	0.0
KIHD	Finland	Prospective Cohort	1998-2001	1774/435	18.1	62.8	53.0	27.8	30.1	6.7
MERLIN TIMI-36	17 countries	Prospective case-cohort	2004	1769/161	0.9	63.2	37.4	29.1	100.0	21.0
MESA	USA	Prospective Cohort	2000	5203/816	12.9	62.0	52.7	28.2	0.0	0.0
PIVUS	Sweden	Prospective Cohort	2001-2004	950/205	15.0	70.2	51.1	27.0	10.0	4.7
PRE-DETERMINE	USA and Canada	Prospective Cohort	2007-2013	4732/505	7.9	63.0	23.9	30.2	100.0	23.3
RS	Netherlands	Prospective Cohort	2002-2005	2361/299	9.9	74.9	58.8	27.4	14.7	4.9
RUTI-HF	Spain	Prospective Cohort	2006-2020	700/84	2.9	64.8	30.4	27.4	32.4	100.0
ULSAM	Sweden	Prospective Cohort	1971-1974	2006/406	33.3	49.7	0.0	25.0	1.2	2.7
WHMS	USA	Prospective Cohort	1995	5257/74	6.0	70.0	100.0	28.3	18.1	0.8
Total				54,799/7720	13.3	63.4	46.7	27.3	19.6	6.6

Figure 2 –

Relative risk reductions in AF for highest quintile versus lowest quintile of various omega-3 blood levels.¹⁴

Dietary intake studies based on participant recall of food consumption have reported inconsistent findings regarding the link between omega-3 and risk of AF.^21–27 However, because these studies were based on food frequency questionnaires, their findings may be unreliable. The recently published Million Veteran Program is the largest ever conducted prospective dietary observational study focused on intake of DHA+EPA + docosapentaenoic acid.²⁸ In that study, the median marine omega-3 FA intake was only 219 mg/d. The researchers reported a nonlinear inverse relationship with incident AF, showing a RR reduction of 11% at 750 mg/d of DHA+EPA intake, followed by a plateau in risk out to quintile 5, where the median intake was 1038 mg/d (Figure 3).

Figure 3 –

Million Veteran Program – 301,294 people; self-reported omega-3 intake via food frequency questionnaires.²⁸

A Danish cohort study of 55,246 individuals reported a U-shaped relationship between the intake of DHA+EPA and incidence of AF over 13.6 years.²⁹ The lowest risk of AF was seen at 630 mg/d of omega-3, which was associated with a 13% RR reduction.²⁹ This association with lower risk of AF was lost at higher DHA+EPA intakes—median consumption of omega-3 was 1250 mg/d in fifth quintile where no AF reduction was observed.²⁹

A recently published observational study of over-the-counter omega-3 supplements and AF risk analyzed 427,678 participants in the UK Biobank, with 25,748 AF events during a median of 11 years of follow-up.³⁰ Habitual users of fish oil supplements (n=133,438, 31% of participants) had a statistically significant, albeit modestly, higher risk of incident AF compared with nonusers, with a hazard ratio (HR) of 1.10, and a 95% confidence interval (CI) of 1.07 to 1.13.³⁰ Yet, these fish oil users in the UK Biobank study also had lower death rates; multivariable adjusted HR for habitual fish oil users versus non-users were 0.87 (CI 0.83 to 0.90) for all-cause mortality, and 0.84 (CI 0.78 to 0.91) for CV mortality.³¹ Information about the doses of DHA+EPA consumed as fish oil in the UK Biobank participants was not available.

The Omega-3—Vagus Nerve—AF Connection

Marine omega-3 fatty acids, which are abundant in the cell membranes of cardiomyocytes and neurons, play an important role in modulation of the autonomic nervous system. Omega-3 enhances vagal tone, as evidenced by its effects in reducing resting heart rate, augmenting heart rate variability (HRV), and improving post-exercise heart rate recovery.^32–36 HRV is a useful non-invasive predictor of sudden cardiac death and increased mortality risk during follow-up.³⁷ Omega-3 reliably increases HRV in healthy individuals and in those with ischemic heart disease, diabetes mellitus, or chronic kidney disease.^{6, 33, 34}

A RCT that used 840 mg/d of DHA+EPA versus placebo showed a significant decrease in resting heart rate of 4 beats/minute, and improvements in HRV and post-exercise heart rate recovery after exercise (another marker of vagal tone).⁶ Similarly, RBC omega-3 content was strongly correlated with better post-exercise heart rate recovery immediately following cessation of treadmill exercise in a cohort of ~14,000 patients.³⁴ The Vital Rhythm study—a 2-year RCT that tested marine omega-3 versus placebo—reported that one daily capsule supplying 840 mg of DHA+EPA increased HRV, lengthened PR-interval, decreased P-wave amplitude and increased P-wave duration—which is associated with an increased AF risk.⁷ These findings suggest that ingested DHA+EPA heightens vagal tone and slows intra-atrial and AV nodal conduction.

These vagal-augmenting effects of omega-3 are dose dependent. Pregnant women randomized to higher dose of DHA (800 mg/d) versus lower dose (200 mg/d) had slower resting heart rate, reduced sympathetic tone, higher vagally-mediated HRV indices, and enhanced HRV complexity, (p < 0.05 for all comparisons).⁵ These metrics demonstrate improved sympathovagal balance with higher dose omega-3.

Low-level vagus nerve stimulation (VNS) was recently shown to prevent AF,⁸ which seems paradoxical, given that VNS has been used for almost a century to induce AF in experimental models.⁹ Yet, recent evidence suggests that the ability of VNS to induce AF is proportional to the degree of heart rate slowing, with no increase in AF inducibility until VNS slows the heart rate by at least 40%.³⁸ On the other hand, it has been consistently shown that VNS at levels significantly below the bradycardia threshold confers strong antiarrhythmic activity.³⁹ Taken together, these observations suggest that the ultimate outcome of VNS depends on the level of stimulation, with antiarrhythmic effects prevailing at low levels that do not substantially slow the sinus rate,⁴⁰ and proarrhythmic events including AF/atrial flutter predominating at higher levels of VNS, especially in the setting of sinus bradycardia.

Discussion

Emerging evidence suggests that the vagal effects of DHA+EPA may be the underlying driver of the biphasic omega-3/AF relationship. At low doses the vagally-mediated antiarrhythmic effects reduce risk of AF, but at progressively higher doses of DHA+EPA the heightened vagal tone can induce sinus bradycardia and increase risk of AF.

The omega-3 dose-dependent amplification of vagal tone might be particularly problematic for people at risk for bradycardia-dependent AF, such as individuals who engage in large amounts of strenuous endurance exercise, have untreated sleep apnea, or have sick sinus syndrome with chronically slow heart rates.

VNS can alter other electrical properties that render atrial cardiomyocytes more susceptible to AF. Vagally-mediated, bradycardia-dependent AF may in part be due to parasympathetic-induced inhomogeneous changes in atrial effective refractory period.⁸ Bradycardia can allow for differential recovery times of the myocardium throughout the atria. This results in electrical heterogeneity, whereby some regions of myocardium might still be refractory, while others are still capable of being depolarized by an electrical wavefront.⁹ This dispersion of refractoriness can create the substrate for a re-entry circuit, such as atrial flutter. Slow HRs also can prolong the action potential durations in atrial cardiomyocytes, which increases the likelihood of abnormal electrical activity and re-entry circuits in the atria.³⁸ To our knowledge, no published studies have reported the effects of omega-3 on atrial flutter versus AF.

Despite the tendency for high-dose DHA+EPA treatment to increase risk of AF, and considering that stroke is the major complication of AF, it is surprising that higher omega-3 blood levels are associated with reduced risk of stroke.^{16, 41} In a large meta-analysis (n >180,000 subjects) with harmonized statistical methodology, being in the highest quintile of blood levels of marine omega-3 fatty acids was associated with a 18% reduction in risk of ischemic stroke, and no association with hemorrhagic stroke (Figure 4).⁴¹ Likewise, high-dose EPA in the REDUCE-IT trial, despite increasing relative risk of AF by 35%, reduced relative risk of stroke by 28%.^{42, 43} And even the cohort who developed AF during the REDUCE-IT study experienced lower risks of stroke and major adverse CV events compared to the placebo group. Moreover, the REDUCE-IT study found that in the group randomized to 4,000 mg/d of EPA, the 1.0% absolute risk increase for AF hospitalization was outweighed by 4.8% absolute risk reduction in the primary composite endpoint of major adverse CV events (CV death, myocardial infarction, stroke, coronary revascularization, or unstable angina requiring hospitalization).⁴³

Figure 4 -.

Associations of circulating DHA levels by quintile with total stroke, ischemic stroke, and hemorrhagic stroke (HR 95% CI).⁴¹

These studies on the effects of marine omega-3 FAs on atrial rhythms suggest that a daily intake of about 600 to 700 mg/d of DHA+EPA may be ideal for minimizing risk of AF. Consumption levels less than this are associated with modestly higher risk of AF, whereas omega-3 intakes >1,000 mg/d are also linked with progressively higher risks of AF. The typical American adult eats <1 serving per week of fish/seafood; so the average intake of DHA+EPA in the United States (US) is only about 100 mg/d.^{44, 45} Accordingly, the mean DHA+EPA level in red blood cells (RBC) (i.e., the omega-3 index) is 5.4% in the US.⁴⁶ The target omega-3 index that is ideal for reducing risk of major adverse CV events, stroke and all-cause mortality is ≥8%; to attain this would require approximately 1,000 mg/d of DHA+EPA.⁴⁴ In Japan and Scandinavia, adults typically consume about 900 to 1000 mg/d of DHA+EPA, and their omega index is ~8%; which could possibly be playing a role in the favorable life expectancies in those countries.⁴⁷ However, for individuals with a history of AF or those who have AF risk factors, a lower target of ~600 to 700 mg/d of DHA+EPA, preferably from fish/seafood, may be a safer level of intake.⁴²

Limitations

The contradictory effects on AF of dietary omega-3 versus high-dose DHA+EPA via supplements/prescription omega-3 could in part be due to confounding whereby other components in fish/seafood might be having salutary effects on risk of AF. Higher blood levels of omega-3 may be a marker for a “healthier” diet with more fish/seafood and less red meat and processed meat, which could be another source of confounding. This is a hypothesis-generating review based on meta-analyses and other diverse sources of data. Heterogeneity in study design and duration, doses of omega-3 used, and study populations make it difficult to draw firm conclusions, and causality cannot be established. Randomized placebo-controlled trials evaluating omega-3’s effects on atrial fibrillation and vagal tone would be helpful in clarifying this hypothesis.

Conclusion

Consumption of marine sources of omega-3 such as fish and seafood has been consistently associated with lower risks for developing AF, where an ideal daily dose of DHA+EPA appears to be ~600 to 700 mg/d. In contrast, higher dose omega-3 interventions (>1,000 mg/d) appear to increase the risk of AF in a dose-dependent fashion, though the absolute risk is small (~1%). Hypothetically, the biphasic relationship between risk of AF and omega-3 dose could be mediated, at least in part, by vagal tone.

Individuals with a history of AF and/or risk factors for vagally-mediated AF should consider avoiding pharmaceutical treatment with omega-3, and instead consume at least 2 servings/week of non-fried fish/seafood, as recommended by the American Heart Association guidelines.⁴⁸

Abbreviations:

AF

Atrial Fibrillation

CI

Confidence Interval

CV

Cardiovascular

DHA

Docosahexaenoic Acid

EPA

Eicosapentaenoic Acid

FA

Fatty Acids

FORCE

Fatty Acids and Outcomes Research Consortium

HR

Hazard Ratio

HRV

Heart Rate Variability

RCT

Randomized Clinical Trials

RBC

Red blood Cells

RR

Relative Risk

SCD

Sudden Cardiac Death

US

United States