Fish and Omega-3 Fatty Acids: Sex and Racial Differences in Cardiovascular Outcomes and Cognitive Function

Abstract

Both cardiovascular disease (CVD) and cognitive decline are common features of aging. One in five deaths are cardiac for both men and women in the US and an estimated 50 million are currently living with dementia worldwide. In this review, we summarize sex and racial differences in the role of fish and its very long chain omega-3 polyunsaturated fatty acids (n-3 PUFA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), in preventing CVD events and cognitive decline. In prospective studies, women with higher nonfried and fatty fish intake and women and Blacks with higher plasma levels of EPA and DHA had lower risk of CVD. In randomized controlled trials (RCTs) of EPA and DHA supplementation in primary CVD prevention, Black subjects benefited in a secondary outcome. In secondary CVD prevention, both men and women benefited, and Asians benefited as a prespecified subgroup. Fish and n-3 PUFA are associated with prevention of cognitive decline in prospective studies. In RCTs of EPA and DHA supplementation, women have cognitive benefit. DHA appears more beneficial than EPA, and supplementation is more beneficial when started before cognitive decline.

Although studies in women and racial groups are limited, life-long intake of nonfried and fatty fish lowers risk of CVD and cognitive decline, and RCTs also show benefit with EPA and DHA supplementation. These findings should be factored into recommendations for future research and clinical recommendations as dietary modalities could be cost-effective for disease prevention.

Introduction

Both cardiovascular disease (CVD) and cognitive decline are common features of aging. CVD is the leading cause of death globally and in 2020, accounted for approximately 19.05 million deaths.¹ About 697,000 people including 314,186 women in the United States died from heart disease in 2020—that’s 1 in every 5 deaths – and rates increased in both men and women between 2019 to 2020.^2,3 For these reasons, identifying and aggressively reducing risk factors for CVD in both men and women not only will reduce health care costs but will also reduce pain and suffering and premature death. Dementia is a major increasing global health challenge. An estimated 50 million people worldwide are currently living with dementia and this number is expected to increase to 131.5 million by 2050.⁴ Worldwide annual dementia costs are $1 trillion; therefore, modalities to prevent cognitive decline are particularly important. Approximately 15-20% of those age 65 years and older have mild cognitive impairment (MCI), an intermediate stage in the continuum from normal aging to dementia.⁵ Alzheimer’s disease is a progressive, neurodegenerative condition, with senile plaques, neurofibrillary tangles, neuronal dysfunction and neural failure. No cure or drug is available to change the progression to Alzheimer’s disease; therefore, dietary modalities become important because they can be cost-effective, have few side effects and are easy to institute over the lifespan on a population basis. In this review, we summarize the role of fish and its very long chain omega-3 polyunsaturated fatty acids (n-3 PUFA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in preventing CVD and cognitive decline with a focus on sex and racial differences. Potential beneficial mechanisms of EPA and DHA are discussed including lowering of triglyceride levels and increases in levels of specialized proresolving lipid mediators, which are the downstream products of EPA and DHA.

Fish Intake and CVD in Prospective Cohort Studies

Prospective cohort studies have reported benefit on coronary heart disease (CHD) events with life-long fish intake (Table 1).^6–10 In the Nurses’ Health Study, two or more servings of fish per week were associated with a 30% lower risk of CHD in women.⁶ At 40-year follow-up of 852 Dutch men free of CHD in 1960, those consuming on average 22 g per day or 1-2 fish meals per week had a 27% lower CHD death risk (P=0.16) and 54% lower sudden coronary death risk (hazard ratio [HR] 0.46; 95% CI:0.27-0.78) compared to those eating no fish.⁷

Table 1.

Prospective Cohort Studies of Fish Intake in Primary Prevention of Cardiovascular Outcomes by Sex and Race

Study Year	# subjects Female/male	Mean age Follow-up	Race ethnicity	Characteristics (exposure) n-3 type and dosage	Endpoint	Outcome of the exposure or intervention
Nurses’ Health Study6 2002	84,668 All female	16 years Age 30-55 in 1976	98.% White	≥2 fish servings/week	Risk of CHD	30% lower
Zutphen Study7 2008	1373 Males only	40 years	Dutch	Ingesting 30 g/day or 2 fish meals/week compared to none	CHD mortality Sudden cardiac death	27% lower 54% lower: HR: 0.46; 95% CI: 0.27-0.78
Shanghai 8 2013	73,159 women 61,137 men Women and men reported separately	Age 40-70 at entry between 1997 to 2000 12 years	Chinese	Total fish intake – self reported Data shown for women Total death is each quintile compared to the lowest quintile CVD death and stroke are top quintile compared to lowest	Total death: Quintile 5 Quintile 4 Quintile 3 CVD death Hemorrhagic stroke	HR: 0.82; 95% CI: 0.73-0.93 HR: 0.86; 95% CI: 0.77-0.96 HR: 0.88; 95% CI: 0.79-0.97 P trend 0.002 HR: 0.78; 95% CI 0.62-0.98; P trend 0.04 HR: 0.62; 95% CI: 0.40-0.96
Danish9 2016	57,053 28,991 female 25,913 male	Female: 56.2 years Male: 55.9 years Median 17 years	Danish	Fatty fish intake- self reported	MI Women Men	aHR: 0.78; 95% CI: 0.63-0.96 aHR: 0·88; 95% CI: 0·77 - 1·00
NIH AARP10 2018	180,580 women 240,729 men	Median age 62 yrs for both men and women 16 years	92% Caucasian	Nonfried Fish Top quintile omega-3 intake compared to bottom quintile	Total mortality Women Men Total mortality Women Men CVD mortality Women Men Alzheimer’s disease death Women Men	aHR: 0.92; 95% CI: 0.88-0.95 aHR: 0.91; 95% CI: 0.89-0.94 10% lower: aHR: 0.90;95% CI: 0.86-0.94 11% lower: aHR: 0.89; 95% CI: 0.86–0.92 18% lower; aHR: 0.82; 95% CI: 0.75-0.90 15% lower; aHR: 0.85; 95% CI: 0.80-0.90 41% lower; aHR: 0.59; 95% CI: 0.43-0.80 30% lower; aHR: 0.70; 95% CI: 0.54-0.89

Abbreviations: aHR indicates adjusted hazard ratio; CHD, coronary heart disease; CI, confidence interval; CVD, cardiovascular disease; HR, hazard ratio; MI, myocardial infarction; n-3, omega-3; NIH AARP, National Institute of Health American Association of Retired Persons.

In the Shanghai Women’s Health study, women in the top three quintiles of total fish intake had significantly lower rates of total death compared to the lowest quintile (P trend 0.002), but there was no difference in men.⁸ Women in the top quintile (median 105.2 g/d) had a 22% lower CVD death (HR:0.78, 95% CI:0.62-0.98, P trend 0.04) and 38% lower hemorrhagic stroke death (HR:0.62: 95% CI:0.40-0.96, P trend 0.04) compared to the lowest quintile (median 10.2 g/d).⁸ No difference was noted in men, illustrating a sex difference in outcomes.

A systematic review of 24 studies in 2021 including a total of 714,526 individuals from several countries reported that consumption of nonfried fish is associated with a reduced risk of CVD events and myocardial infarction (MI).¹¹ Of the 24 studies, only 4 reported women separately. Of these, Danish women eating >23 g/d of fatty fish, which are high in EPA and DHA, had a 22% lower rate of MI (HR:0.78; 95% CI:0.63-0.96) compared to a 12% lower rate in men (HR:0.88;95% CI:0.77-1.00) after adjustment at 17-year follow-up.⁹ The other 3 found no difference in women; however, total fish was reported rather than fatty fish. Fatty fish include salmon, Atlantic mackerel, Atlantic herring, sardines, anchovies and Bluefin and albacore tuna. In the National Institute of Health American Association of Retired Persons (NIH-AARP) prospective study of 180,580 women, those in the top quintile of n-3 PUFA intake estimated at > 23 g/d from dietary records, compared to the bottom quintile (0-6 g/d), had 10% lower all-cause mortality (HR:0.90; 95% CI:0.86-0.94), 18% lower CVD mortality (HR:0.82; 95% CI:0.75-0.90) and 41% lower rates of death from Alzheimer’s disease (HR:0.59; 95% CI:0.43-0.80).¹⁰

Prospective Cohorts Reporting CVD Outcomes by EPA and DHA Levels

The beneficial effect of fish appears due to EPA and DHA, which are found in fatty fish (Table 2). In the prospective Cardiovascular Health Study of 2,692 US adults free of coronary artery disease (CAD) at baseline, higher plasma levels of EPA and DHA were associated with a 27% lower rate of total mortality at 16-year follow-up due to fewer CVD deaths (HR:0.73; 95% CI:0.61-0.86, P trend 0.008).¹² Sex was adjusted for rather than presented separately. In the prospective Multi-Ethnic Study of Atherosclerosis (MESA) at 10 year follow-up of 2,837 subjects, the highest quartiles of plasma EPA levels were associated with HRs of 0.39 (0.17-0.88) in women and 0.58 (0.30-1.11) in men, and highest quartiles of DHA were associated with HRs of 0.32 (0.12-0.87) in women and 0.41 (0.20-0.86) in men for incident CVD compared with the lowest quartile.¹³ At 15.6-year follow-up in 1,799 Black participants in MESA, significant inverse associations were noted between DHA levels and a composite of incident MI, cardiac arrest, stroke, CHD or stroke death (adjusted HR:0.56; 95% CI:0.34-0.90) compared to other racial groups.¹⁴ These differential benefits may be attributed to genetic variations relevant to n-3 PUFA metabolism¹⁵ in addition to the potential anti-inflammatory actions of n-3 PUFA benefits in Black subjects who have increased inflammatory biomarkers compared to other racial groups.¹⁶

Table 2.

Prospective Cohort Studies by EPA and DHA Levels in Primary Prevention of CVD by Sex and Race

Study Year	# subjects Female/male	Mean age Follow-up	Race ethnicity	Characteristics (exposure) n-3 type and dosage	Endpoint	Outcome of the exposure or intervention
Cardiovascular Health Study12 2013	2,692 63.7% women Adjusted for sex except length of life reported separately for women and similar to men	74 years 16 years	687 Blacks (11.7%) Adjusted for white race	Highest Quintile EPA+DHA compared to lowest Highest Quintile EPA compared to lowest Highest Quintile DHA Compared to lowest	Total mortality	27%↓ HR:0.73, 95% CI, 0.61-0.86, P trend 0.008 Highest quintile lived 2.2 years longer than lowest 17% ↓ HR: 0.83; 95% CI: 0.71-0.98; P trend < 0.005 20% ↓HR: 0.80; 95% CI: 0.67-0.94; P trend 0.006
MESA13 2013	2,837 52% female Men and women reported separately	61.5 years 10 years 2000-2010	White n=724 African American n=697 Chinese American n=711 Hispanic n=705	Top quartile EPA+DHA compared to lowest Top quartile EPA by sex Top quartile DHA by sex	Incident CVD Total group Women Men Women Men	HR 0.49 and 0.39 for EPA and DHA, respectively, for incident CVD; P trend <0.001 HR 0.39; 95% CI: 0.17-0.88 HR 58; 95% CI: 0.30-1.11 HR 0.32; 95% CI: 0.12- 0.87 HR 0.41; 95% CI: 0.20-0.86
MESA14 2018	6,568 2,837 women 52.9% female	Mean age 62.1 years 15.6 years	Black: n=1,799 27.4%	Inverse association between DHA levels and CVD events in Black subjects	Incident MI, cardiac arrest, stroke, CHD or stroke death	aHR 0.57; 95% CI: 0.34-0.90

Abbreviations: aHR indicates adjusted hazard ratio; CHD, coronary heart disease; CI, confidence interval; CVD, cardiovascular disease; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; HR, hazard ratio; MESA, Multiethnic Study of Atherosclerosis; MI, myocardial infarction; n-3, omega-3.

Randomized Trials of Omega-3 Fatty Acids in Primary Prevention of CVD

The VITamin D and OmegA-3 Trial (VITAL) was the first large-scale, randomized controlled trial (RCT) investigating the effect of n–3 PUFA in primary CVD prevention and cancer among 25,871 men and women 50 and 55 years of age or older, respectively, in the US randomized to 860 mg EPA+DHA or placebo (Table 3).¹⁷ At 5.3-year follow-up, both women and men had nonsignificant 9 and 7% reductions, respectively, for major CVD events: men, HR:0.91 [0.76-1.10]; women, HR:0.93 [0.76-1.15]. However, a significant 28% reduction in the risk of MI, a secondary outcome, was noted for men, HR:0.72 [0.55-0.95] but not women, HR:0.73 [0.52-1.03].¹⁷ African Americans (n=5,106) experienced a significant 77% reduction in MI (HR:0.23; 95% CI: 0.11-0.47) whereas other racial/ethnic groups had no benefit (non-Hispanic Whites, HR:0.93; 95% CI:0.73-1.18; others: HR:0.54; 95% CI:0.23-1.26; P interaction by race/ethnicity=0.001), indicating a difference in outcome by race.¹⁸ Of note, African Americans derived this treatment benefit irrespective of their baseline dietary fish intake, whereas non-Hispanic Whites benefitted only when their fish intake was low. Supplemental n-3 PUFA led to a 19% reduction in major CVD events (HR:0.81; 95% CI:0.67-0.98) and a 40% reduction in MI (HR:0.60; 95% CI 0.45-0.81) among individuals with low fish intake (below the cohort median) but not among those with higher fish intake (HR:1.08; 95% CI:0.88-1.32; MI: HR:0.94; 95% CI:0.67-1.31).¹⁸

Table 3.

Randomized Controlled Trials of EPA and DHA Supplementation in Primary Prevention of CVD by Sex and Race

Study Year	# subjects Female/male	Mean age Follow-up	Race ethnicity	Characteristics (exposure) n-3 type and dosage Baseline triglyceride and % reduction	Endpoint	Outcome of the exposure or intervention
Vital17 2019 202018	25,871 Female=13,085 (50.6%)	Median age 67.1±7.1 5.3 years	5,106 Black (20.2%)	860 mg EPA+DHA vs placebo No triglyceride data	Major CVD events Total group Men Women MI, secondary outcome Total group Men Women Black Non-Hispanic White Others	Nonfatal stroke, nonfatal MI & CVD death HR: 0.92; 95% CI: 0.80-1.06. HR: 0.91; 95% CI: 0.76-1.10 Women: HR: 0.93; 95% CI: 0.76-1.15 28% ↓; HR: 0.72; 95% CI: 0.59-0.99 HR: 0.72; 95% CI: 0.55-0.95 HR: 0.73; 95% CI: 0.52-1.03 HR: 0.23; 95% CI: 0.11-0.47 HR: 0.93; 95 % CI: 0.73-1.18 HR: 0.54; 95% CI: 0.23-1.26
Ascend19 2018 All with DM	15,480 37.4% female- didn’t report separately or adjust for sex	63.3±9.2 7.4 years	96.5% white	460 mg EPA 380 DHA vs 1 g olive oil No triglyceride data	Nonfatal MI, nonfatal stroke, TIA and vascular death	No difference

Abbreviations: ASCEND indicates Study of Cardiovascular Events in Diabetes; CI, confidence interval; CVD, cardiovascular disease; DHA, docosahexaenoic acid; DM, diabetes mellitus; EPA, eicosapentaenoic acid; g, gram; HR, hazard ratio; MI, myocardial infarction; n-3, omega-3; TIA, transient ischemic attack; VITAL, VITamin D and OmegA-3 Trial.

In The Study of Cardiovascular Events in Diabetes (ASCEND) of 15,480 subjects with diabetes (96.5% white) without clinical CVD, no reduction of CVD risk was observed in those randomized to 460 mg of EPA and 380 mg of DHA compared to an olive oil placebo (Table 3).¹⁹ The trial did not report women separately or adjust for sex.

Randomized Trial of Fish in Secondary Prevention of CVD

In the Diet and Reinfarction Trial, 2,033 Welsh men who were post-MI and randomized to 2-3 portions of fish per week or taking 3,500 mg fish oil capsules daily had a 29% reduction in total mortality (9.3% vs. 12.8%; P<0.05) and 32% reduction in CHD death (7.7% vs. 11.4%; P<0.01) at 2-year follow-up compared to those randomized to either reduction in fat intake to 30% of total energy with an increased polyunsaturated/saturated fat ratio or increased cereal fiber of 18g/dL (Table 4).²⁰ This study strongly suggested that fish or fish oil intake decreased CHD mortality; however, a limitation is that it examined only men.

Table 4.

Randomized Controlled Trials of Fish Intake and Omega-3 Ethyl Ester (EPA and DHA) Supplementation in Primary and/or Secondary Prevention of CVD by Sex and Race

Study Year	# subjects Female/male	Mean age Follow-up	Race ethnicity	Characteristics (exposure) n-3 type and dosage Baseline TG and % reduction	Endpoint	Outcome of the exposure or intervention
Diet and Reinfarction Trial20 1989 Post-MI Mean 41 days after MI	2,033 Male only	56.7 years 2 years	Welsh men	3 arms: fat intake 30% with ↑P/S fat ratio; cereal fiber 18g/d or 2-3 fish portions/week or 3,500 mg fish oil capsules daily	Total mortality CHD death	Those eating fish had 29% ↓ in total mortality (9.3% vs 12.8%, P<0.05 compared to other 2 groups Those in fish group had 32% ↓ in CHD death 7.7% vs 1.4%, P<0.01.
GISSI 21 1999 Post-MI within 3 months	5,664 15.3% female Women not reported separately; sex not adjusted for	3 months 3.5 years	Italian	850-882 mg EPA+DHA daily or none (open-label) TG: n-3: −4.6% vs 0.4% control; net: −4%	3 months 3.5 years	15%↓total death, nonfatal MI and nonfatal stroke RR: 0.85, 0.74-0.98, (P=0.02) due to 45% decrease in sudden death 0.55 (0.40–0.76) 20% ↓ in CVD death, nonfatal MI and stroke RR: 0.80, 95%CI: 0.68-0.95, P=0.008
GISSI HF22 2008 NYHA class II-IV heart failure	6,975 22% female Female not reported separately; sex not adjusted for	Mean (SD) age 67 (11) 3.9 yrs IQR: 3-4.5	Italian	850-882 mg EPA+DHA TG: net −6% in n-3 group	Total mortality	HR: 0·91; 95·5% CI: 0·833–0·998, P=0·041
Omega23 Post-MI 2010	3,851 25.6% female Did not report sex separately or adjust for sex	Mean age 64 1 year	104 German centers Race data not collected	460 mg EPA and 380 mg DHA vs 1 g olive oil placebo TG: n-3: −7% vs.- 3% placebo; net −4%	Sudden Cardiac death Total mortality Major adverse cerebrovascular and cardiovascular events Revascularization	1.5% vs 1.5%, P=0.84 4.6% vs 3.7%, P=0.18 10.4% and 8.8%, P=0.10 27.6% vs 29.1%, P=0.34
Alpha Omega24 2010 Post-MI on average 4 years	4,837 32% women Statin use 85%	mean age 69 years 40 months	32 hospitals in Holland	4 margarines: 226 mg of EPA+150 mg of DHA or ALA 1.9 g or EPA+DHA+ALA or Placebo TG: n-3: −5% vs. −3% for placebo; net: −2%	Major CVD events EPA+DHA Male EPA+DHA Women ALA Women	Major CVD events include fatal and nonfatal CVD, PCI and CABG HR:1.06; 95% CI: 0.89 to 1.25; P=0.51 HR: 0.82, 95% CI: 0.58-1.16, P=0.27 HR, 0.73; 95% CI: 0.51 to 1.03; P=0.07.
Origin25 2012 Primary and secondary prevention	12,536 all with DM or impaired fasting glucose 59% with CVD 34% female; didn’t report by sex or adjust for sex	63.5±7.8 years Median 6.2 years	40 countries; no race data	465 mg of EPA and 375 mg of DHA Placebo: 1 g olive oil TG: n-3: −17% vs. −6% for placebo; net: −11%	CVD death	n-3 vs placebo: 9.1% vs. 9.3%; HR: 0.98; 95% CI: 0.87-1.10; P=0.72
Risk and Prevention 26 2013 Primary and secondary prevention DM + at least 1 additional risk factor: 47.9% CVD: 29.5% 4 CVD risk factors but no DM: 20.8%	12,513 total 38.5% women 40% on statin	Mean age: 64 years Median follow-up: 5 years	860 general practitioners in Italy	Omega-3 ethyl ester 1 g daily or placebo (olive oil). Omega-3 ranged in a ratio of EPA:DHA 0.9:1 to 1.5:1). Baseline TG 150 mg/dL (range 107 to 201) TG decreased 28.2±1.3 mg/dL in omega-3 and 20.1±1.3 mg/dL in placebo, P<0.001.	Primary Endpoint Total group Men Women	Time to death from cardiovascular causes or hospital admission for cardiovascular causes Omega 3: 11.7% vs 11.9% placebo (aHR: 0.97; 95% CI: 0.88 to 1.08; P=0.58). 1.04 (0.92–1.17), P=NS 0.82 (0.67–0.99), P=0.04

Abbreviations: aHR indicates adjusted hazard ratio; ALA, alpha linolenic acid; CABG, coronary artery bypass graft; CHD, coronary heart disease; CI, confidence interval; CVD, cardiovascular disease; docosahexaenoic acid; DM, diabetes mellitus; EPA, eicosapentaenoic acid; g, gram; HF, heart failure; HR, hazard ratio; MI, myocardial infarction; n-3, omega-3; PCI, percutaneous coronary intervention; P/S, polyunsaturated/saturated; TG, triglyceride.

Randomized Trials of Omega-3 Fatty Acids and CVD Outcomes in Primary and/or Secondary Prevention of CVD

Low-Dose Omega-3 Fatty Acids

Six RCTs have reported on the use of low-dose omega-3 fatty acids in prevention of CVD (Table 4).^21–26 The Gruppo Italiano per lo Studio della Sopravvivenza nell-Infarto Miocardico (GISSI) prevention study²¹ and GISSI Heart Failure study both showed benefit with 840 mg of ethyl esters EPA/DHA in a 1000 mg capsule administered once daily (former Omacor) in an open-label trial. ²² Two subsequent placebo-controlled studies (Omega and Origin) with the same dose showed no benefit (Table 4).^23,25 A fifth RCT, Alpha Omega, used margarines containing very low-dose EPA and DHA or alpha-linolenic acid (ALA) and also reported no benefit.²⁴ None of these studies reported on women or race separately or adjusted for sex or race. The Risk and Prevention trial reported a benefit in 4,818 women randomized to 1 g daily of omega-3 ethyl ester or an olive oil placebo with 40% on statin treatment. At a median follow-up of 5 years, no difference was observed in the 2 arms for men and women combined; however, women randomized to omega-3 had a significant 18% reduction (HR: 0.82; 95% CI 0.67–0.99), P=0.04, indicating a difference in outcome by sex.²⁶ Several reasons could account for the differences in these trials. The GISSI trials were open-label, a study design not as definitive as a placebo-controlled trial, and the survival rates were lower in the GISSI-Prevenzione trial than in Alpha Omega, Omega, Origin and Risk and Prevention. The introduction of statins represented a major change in the treatment of CVD patients in the early 1990s. Only 5% of patients in the GISSI-Prevenzione trial received statins at baseline (1993–1995), whereas 46% were receiving them after 42 months of intervention. In Alpha Omega, 85% were receiving statins; 81% in Omega, 53% in Origin and 40% in Risk and Prevention.^23–26

Randomized Trials of Pure-EPA

Several trials have examined pure EPA-containing products (Table 5). The Japan Eicosapentaenoic Lipid Intervention Study (JELIS) was a mixed trial of primary and secondary prevention with addition of 1800 mg/day of open-label EPA to either 10-20 mg pravastatin or 5-10 mg simvastatin therapy. For the combined group, a significant 19% reduction (95% CI:0.69–0.95; P=0.011) in major coronary events compared to statins alone was observed at 4.6-year follow-up, suggesting that high-dose EPA added to low-intensity statin therapy can further reduce residual risk ²⁷ The primary prevention group had no benefit in major coronary events (HR:0.82; 95% CI:0·63–1.06; P=0.132) whereas the secondary prevention group had a significant 19% reduction (HR:0.81; 95% CI:0.657–0.998, P=0.048). Men (n=5,859) had a significant 24% reduction (HR:0.76; 95% CI:0.62–0.94) whereas women (n=12,786) had a nonsignificant 13% reduction (HR:0.87; 95% CI:0.68–1.13).²⁷ This lack of significance in women could have been due to a 2-3 fold lower event rate compared to men. Major coronary events included sudden cardiac death, fatal and nonfatal MI, unstable angina and coronary revascularization. Stroke was not included. RESPECT-EPA was open-label and randomized 2,506 Japanese subjects with CAD and a low EPA/arachidonic acid ratio to a similar dose of pure EPA as JELIS - 1800 mg EPA or none. At 6-year follow-up, the primary endpoint of sudden cardiac death, MI, unstable angina, nonfatal stroke and coronary revascularization approached significance whereas the secondary endpoint, which did not include stroke, was significant, 8% vs 11.3%, HR: 0.734, P=0.0306.^{28, 29}

Table 5.

Randomized Controlled Trials of Pure EPA Supplementation in Primary and/or Secondary Prevention of CVD by Sex and Race

Study Year	# subjects Female/male	Mean age Follow-up	Race ethnicity	Characteristics (exposure) n-3 type and dosage Baseline TG and % reduction	Endpoint	Outcome of the exposure or intervention
JELIS 27 2007 Primary and secondary prevention	18,645 total 12,786 women (69%) 5,859 men	61±8 years 4.6 years	Japanese	1800 mg icosapent ethyl+ statin or statin only; open-label On Prava 10 or 20 mg or simva 5 or 10 mg Baseline TG: 153.2 mg/dL TG: n-3: −9% vs. −4% (control); net: −5% Mean LDL-C: 180 mg/dL	Major coronary events Primary prevention group Total group CAD group Women Men	Coronary events include sudden cardiac death, fatal and nonfatal MI, unstable angina, PCI and CABG 18% ↓; HR:0.82; 95% CI: 0·63–1.06, P=0.132 19% ↓; HR: 0·81; 95% CI: 0·69–0·95; P=0.011 19% ↓; HR:0.81; 95% CI: 0.657–0.998, P=0.048 13% ↓; HR: 0.87; 95% CI: 0.68–1.13 24% ↓; HR: 0.76; 95% CI: 0.62–0.94
RESPECT-EPA 28,29 DM: 45% All with CAD	2,506 17.3% women n-3: 1,249 control: 1,257 EPA/AA ratio < 0.4 All on statin	Median [IQR] age: 68 (61.0, 73.0) Follow-up: 6 years	95 centers in Japan	Purified EPA; 1800 mg daily vs none (open label) Baseline median TG: EPA: 120 mg/dL; control: 117 mg/dL TG reduction not given Median LDL-C: 80 mg/dL Median EPA: 45 μg/ml	Primary endpoint Secondary endpoint	Cardiovascular death, non-fatal MI, non-fatal ischemic stroke, unstable angina pectoris requiring hospitalization, and clinically indicated coronary revascularization 10.9% versus 14.9%, HR 0.785, P=0.0547). Sudden cardiac death, MI, unstable angina, and coronary revascularization 8.0% versus 11.3%, HR 0.734, P=0.0306)
REDUCE-IT 201930 202131 CAD 70.7% T2DM 29.3% with at least 1 additional risk factor All with TG 135–499 mg/dL;	5,822 men 2,357 women	Median [IQR] 64 [57-69] Median follow-up 4.9 years	90.3% white median TG of 216.5 mg/dL TG reduction: EPA: −18.3% vs. +2.2% in placebo; net −20.5% for EPA	4 g icosapent ethyl vs mineral oil placebo Primary Prevention Secondary Prevention Central adjudication30 Investigator adjudication31 Central adjudication30 Investigator adjudication31 Central adjudication30 Central adjudication30	Primary endpoint Total group30 Men Women White race Other race	CVD death, nonfatal MI, nonfatal stroke, PCI or CABG, unstable angina HR: 0.88; 95% CI: 0.70–1.10 HR: 0.73; 95%CI: 0.65–0.81 HR: 0.73; 95% CI: 0.65-0.82 HR: 0.74; 95% CI: 0.67-0.82 HR: 0.82; 95% CI: 0.66-0.1.01 HR: 0.74; 95% CI: 0.61-0.89 HR: 0.77; 95% CI: 0.69–0.85 HR: 0.60; 95% CI: 0.43–0.83

Abbreviations: CABG indicates coronary artery bypass graft; CAD, coronary artery disease; CI, confidence interval; CVD, cardiovascular disease; DM, diabetes mellitus; EPA, eicosapentaenoic acid; g, gram; HR, hazard ratio; MI, myocardial infarction; n-3, omega-3; PCI, percutaneous coronary intervention; REDUCE-IT, Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial; TG, triglyceride.

In the Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial (REDUCE-IT) with a triglyceride level of 135 to 499 (median 216.5 mg/dL) and either cardiovascular disease or diabetes with one other risk factor, the total group had a 26% reduction (HR:0.74, 95% CI:0.65–0.83, P<0.001).³⁰ Both women and men had similar 26% significant reductions (HR:0.74, 95% CI:0.61-0.89 vs 0.74, 0.67-0.82, respectively) for investigator-adjudicated events of a composite of cardiovascular death, nonfatal MI and nonfatal stroke in statin-treated subjects compared to a mineral oil placebo at a median follow-up of 4.9 years ³¹ Those in the mineral oil placebo had significant 10% and 30% increases from baseline in levels of low-density lipoprotein cholesterol (LDL-C) and high-sensitivity C-reactive protein (hs-CRP), respectively, at 12 months.³² Increases in other inflammatory markers included lipoprotein(a), homocysteine, oxidized low-density lipoprotein, interleukin-6, Lp-PLA2 activity and interleukin-1β. The effect of these increases in the mineral oil placebo arm on the overall risk reductions in clinical events is uncertain, and some have called for a new trial with a neutral placebo.

Randomized Trials of Free Omega-3 Carboxylic Acid Containing EPA and DHA

Two trials examined the effect of omega-3 carboxylic acid containing EPA and DHA on cardiovascular outcomes (Table 6). The Omega-3 Fatty acids in Elderly with Myocardial Infarction (OMEMI) trial examined 1800 mg daily of free omega-3 carboxylic acid vs a corn oil placebo and found no benefit on CVD events. The primary outcome did not differ by sex.³³ However, one child spoon full of cod liver oil was allowed in both arms, which could have reduced risk in the placebo arm. The Long-Term Outcomes Study to Assess Statin Residual Risk with Epanova in High Cardiovascular Risk Patients with Hypertriglyceridemia (STRENGTH) trial tested a higher dose of free omega-3 carboxylic acid (Epanova), 4 g daily (2,220 mg EPA and 800 mg DHA), in 13,078 patients with CVD or DM and at least 1 additional risk factor.³⁴ The primary endpoint, which included cardiovascular death, nonfatal MI, revascularization, hospitalized unstable angina and nonfatal stroke, was not significant for the total group with no difference by sex. However, Asian subjects (a prespecified subgroup, n=1,355) had a significant 28% reduction in the primary endpoint (HR:0.72; 95% CI:0.54-0.96), indicating a difference in outcome by race.³⁴ Moreover, a significant 15% reduction (HR: 0.85; 95% CI: 0.75-0.97, P=0 02) was noted for those with established CVD in the secondary endpoint of coronary events, which was similar to the primary endpoint but did not include fatal and nonfatal stroke. Of note, JELIS did not include fatal and nonfatal stroke in the primary endpoint.²⁷ Stroke is more strongly influenced by blood pressure than a triglyceride-lowering agent; therefore, the composite of cardiac events without stroke may be more reflective of the true effect of EPA+DHA supplementation. These significant findings in Asians and the secondary endpoint of cardiac events without stroke warrant consideration for investigation in future studies.

Table 6.

Randomized Controlled Trials of free Omega-3 Carboxylic Acid (EPA and DHA) Supplementation in Primary and/or Secondary Prevention of CVD by Sex and Race

Study Year	# subjects Female/male	Mean age Follow-up	Race ethnicity	Characteristics (exposure) n-3 type and dosage Baseline TG and % reduction	Endpoint	Outcome of the exposure or intervention
OMEMI 33 2021 All post-MI for 2-8 weeks	1,027 patients 29% female	Mean±SD age: 75±3.6 years (range 70-82 years) 2 year follow-up	4 Centers in Norway White: 99.8%	EPA: 930 mg DHA: 660 mg vs corn oil placebo Mean triglyceride: 111.4±61.9 mg/dL.	Primary Endpoint Secondary endpoint:	Composite of nonfatal MI, unscheduled revascularization, stroke, all-cause death, heart failure hospitalization 21.4% on n-3 PUFA vs 20.0% on placebo (HR, 1.08 [95% CI, 0.82–1.41]; P=0.60). Afib: 7.2% (28) on n-3 PUFA vs 4.0% (15) on placebo (1.84 [0.98–3.45]; P=0.06).
STRENGTH34 2020 55.9% with CVD; n= 7,316 70% with DM and either CVD or at least 1 additional risk factor	13,078 35% women CVD group: n-3 = 3,638 Placebo: 3,678	Mean age 62.5±9 years Median follow-up: 42 months	22 countries Black=345 *Other= 654 TG reduction: −19.0% vs −0.9%; net −18.1% for omega-3; Non-HDL: −6.1% vs −1.1%; net 5% for omega-3	Omega-3 carboxylic acid 4 g- EPA: 2,200 mg; DHA: 800 mg daily vs corn oil placebo -Primary outcome: cardiovascular death, nonfatal MI, nonfatal stroke, coronary revascularization, or unstable angina requiring hospitalization Primary prevention Secondary prevention -Secondary outcome of coronary events: cardiac death, MI, revasc and hospitalized unstable angina	Total group Men n=8,510 Women n=4,563 Asian n=1355 n = 5,762 n=7,316 Secondary endpoint: Total group	HR: 0.99; 95% CI, 0.90-1.09; P=0.84 HR: 1.01; 95% CI: 0.90-1.13 HR: 0.94; 95% CI: 0.78-1.14 HR: 0.72;95% CI: 0.54-0.96 HR: 1.16; 95% CI: 0.95-1.41 HR: 0.94; 95% CI: 0.84-1.05 HR: 0.85; 95% CI: 0.75-0.97, P=0.02

^*Other: American Indian or Alaskan Native, Native Hawaiian or Pacific Islander, multiple race or unknown

Abbreviations: Afib indicates atrial fibrillation; CI, confidence interval; CVD, cardiovascular disease; DHA, docosahexaenoic acid; DM, diabetes mellitus; EPA, eicosapentaenoic acid; g, gram; HR, hazard ratio; MI, myocardial infarction; n-3, omega-3; OMEMI, Omega-3 Fatty acids in Elderly with Myocardial Infarction; PUFA, polyunsaturated fatty acid; RR, relative risk; STRENGTH, Long-Term Outcomes Study to Assess Statin Residual Risk with Epanova in High Cardiovascular Risk Patients with Hypertriglyceridemia; TG, triglyceride.

Effect of Omega-3 Ethyl Ester on Coronary Plaque and Cardiovascular Events

The Slowing HEART diSease with lifestyle and omega-3 fatty acids (HEARTS) randomized trial examined the effect of supplementation with 3.36 g omega-3 ethyl ester (EPA and DHA) daily for 30 months on coronary plaque as measured by coronary computed tomographic angiography and cardiovascular events. In 240 statin-treated subjects, mean (SD) age 63.3 (7.6) years with stable CAD and normal triglyceride levels (median 122 mg/dL), those with an omega-3 fatty acid index ≥ 4% had prevention of progression of noncalcified and regional calcified coronary plaque compared to those with an omega-3 fatty acid index <3.4% who had progression.³⁵ Those with an omega-3 fatty acid index of 8.4% had regression of noncalcified plaque.³⁶ Those with regression of fatty plaque had a 14.9% reduction in triglyceride that correlated with fatty plaque regression (r=0.135; P=0.036) and also had 4-fold fewer cardiac events than those with progression (5% versus 22.3%, respectively, P<0.001).³⁷ Normotensive subjects on EPA+DHA had regression of noncalcified coronary plaque that correlated with triglyceride reduction (r=0.35; P=0.034) and a significant decrease in neutrophil/lymphocyte ratio.³⁷ In contrast, hypertensive subjects had no change in noncalcified coronary plaque or neutrophil/lymphocyte ratio. These results suggest that triglyceride reduction was associated with coronary plaque regression and lower rates of cardiac events.

Several meta-analyses have been reported with the 2 most recent reporting a dose-response benefit (Table 7). In a 2019 meta-analysis of 127,477 subjects in 13 RCTs without including REDUCE-IT, supplementation with n-3 PUFA was associated with a significantly lower risk of MI (rate ratio [RR]:0.92; 95% CI:0.86-0.99, P=0.020), CHD death (RR:0.92; 95% CI:0.86-0.98, P=0.014), total CHD (RR:0.95; 95% CI:0.91-0.99, P=0.008), CVD death (RR:0.93; 95% CI:0.88-0.99, P=0.013), and total CVD (RR:0.97; 95% CI:0.94-0.99, P=0.015).³⁸ However, no benefit was observed for stroke (RR:1.05; 95% CI: 0.98-1.14). Inverse associations for all outcomes were strengthened after including REDUCE-IT. Risk reductions were linearly related to marine omega-3 fatty acid dose.³⁸ Moreover, a 2021 meta-analysis of 40 trials also reported a dose-dependent effect of EPA+DHA with significant reductions in MI, CHD, fatal MI and mortality but not CVD events (Table 7).³⁹ It was estimated that increasing intake by 1 g/day EPA+DHA corresponded to a 9% and 7% lower risk of MI and total CHD, respectively, and to a 5.8% lower risk of CVD events (P<0.01).

Table 7.

Meta-analyses of Omega-3 Fatty Acid in Prevention of CVD Events

Study Year	# Studies	Mean age Mean follow-up	Number subjects	Characteristics (exposure) n-3 type and dosage Baseline TG and % reduction	Endpoint	Outcome of the exposure or intervention
Hu et al38 2019	19 RCTs	64.3 years 5 years	127,477 59.7% male	Omega-3 dose ranged from 376 to 4000 mg/day	Dose-dependent effect	Lower risk for MI, CHD, total CHD, CVD death and total CVD
Bernasconi et al 39 2021	40 studies Protective effect against fatal MI and CHD death achieved with <800-1200 mg daily of EPA+DHA	Not given	135,267 Race/sex distribution not given	EPA+DHA dose ranged from 400 mg/day to 5500 mg/day; Weighted average dosage received was 1221 mg/day of EPA + DHA	Dose dependent for CVD events and MI	MI ↓ 13%, RR:0.87; 95% CI: 0.80-0.96 CHD ↓ 10%, RR: 0.90; 95% CI: 0.84-0.97 Fatal MI ↓ 35%, RR:0.65; 95% CI: 0.46-0.91 CHD Mortality ↓ 9%, RR: 0.91; 95% CI: 0.85-0.98 CVD events:↓ 5%, RR:0.95; 95% CI: 0.90-1.00

Abbreviations: CHD indicates coronary heart disease; CI, confidence interval; CVD, cardiovascular disease; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; MI, myocardial infarction; n-3, omega-3; RR, relative risk; TG, triglyceride.

Risk of Atrial Fibrillation (Afib) with Omega-3 Fatty Acid Supplementation

When recommending omega-3 fatty acid supplementation, one must take into consideration the potential increased risk of Afib, which can be associated with an increase in the risk of stroke.

Low-Dose Omega-3 Fatty Acid and Afib

Trials of low-dose omega-3 fatty acid supplementation showed no increased risk of Afib. In the ASCEND trial of 840 mg EPA and DHA, in which no benefit was observed on CV events as described earlier, no difference was observed in the incidence of AFib in the omega-3 arm, 2.1% vs 1.7% for placebo (HR, 1.23; 95% CI, 0.98-1.54).¹⁹ In the neutral CV trial, VITAL with 860 mg EPA+DHA, Afib occurred at a similar rate in the active arm and placebo arms (3.7% vs 3.4%, respectively; HR, 1.09; 95% CI, 0.96-1.24; P=0.19).¹⁷

Moderate-Dose Omega-3 Fatty Acid and Afib

In the neutral OMEMI trial of 930 mg of EPA and 660 mg of DHA for 2 years, 43 patients developed new-onset Afib and 27 patients developed paroxysmal Afib. In the omega-3 group, 11.9% developed Afib compared to 6.5% in the placebo group; HR 1.90 (95%CI 1.16–3.11), P=0.011.⁴⁰ Changes in serum EPA (but not DHA) mediated the effect from n-3 PUFA on Afib risk, explaining 65% of the association.

High-Dose Omega-3 Fatty Acid and Afib

In the STRENGTH trial of 4g daily of omega-3 carboxylic acid, risk of new-onset Afib was significantly higher in the omega-3 group: 2.2% vs 1.3% for corn oil (HR, 1.69; 95% CI, 1.29-2.21; P <0.001).³⁴ In the REDUCE-IT trial of 4 g daily of icosapent ethyl after median follow-up of 4.9 years, hospitalization for Afib or flutter occurred in 3.1% of the active arm vs 2.1% of the mineral oil group (P=0.004).³⁰

Meta-Analysis of Omega-3 Fatty Acid Trials and Afib

A meta-analysis of these five trials plus 2 more (GISSI-HF and Risk and Prevention Study) included more than 81,000 patients followed for nearly 5 years.⁴¹ Omega-3 fatty acid supplements were associated with a 25% increase in the risk for Afib (HR, 1.25; 95% CI, 1.07-1.46, P=0.013). A dose-dependent relationship was observed with most of the increased risk occurring in trials that tested > 1 g/d.

Potential Mechanisms of Benefit for CVD Prevention with Omega-3 Fatty Acids

Potential protective effects of omega-3 fatty acids against CVD are possibly related to reduced levels of plasma triglycerides. Apolipoprotein (apo) B-100 is synthesized by the liver and secreted in the form of very low density lipoprotein (VLDL), a triglyceride-rich-lipoprotein which is the major triglyceride carrier in plasma.⁴² ApoB-48 is the amino terminal 48% of apoB-100 and accepts dietary triglyceride in intestinal enterocytes to form chylomicrons.⁴³ In the plasma, the triglyceride of both chylomicrons, triglyceride-rich-lipoprotein and VLDL particles is hydrolyzed to free fatty acids and glycerol leading to the formation of VLDL remnants, intermediate density lipoprotein and chylomicron remnants.^44,45 Oxidation of fatty acids of surface phospholipids of these apoB-containing lipoprotein particles, results in modification of lysine residues of apoB.⁴⁶ These “modified” VLDL remnants, intermediate density lipoprotein particles and chylomicron remnants are small enough to cross the endothelial barrier in blood vessels where they are taken up by the scavenger receptor on macrophages, a process leading to the formation of foam cells and fatty streaks, the precursor of more advanced atherosclerotic plaque. Thus, elevated triglyceride levels can contribute to the formation of atherosclerotic plaque. Omega-3 fatty acids reduce plasma triglyceride levels through reduced hepatic lipogenesis and VLDL production⁴⁷ and increased triglyceride-rich-lipoprotein lipolysis.⁴⁸ Moreover, biliary C22:6 omega-3 fatty acid–derived N-acyl taurines prevent intestinal triglyceride absorption, providing an additional mechanism for how DHA lowers plasma and liver triglyceride levels.⁴⁹

A post-hoc analysis of the REDUCE-IT trial reported a strong relationship between plasma EPA concentration and reduction of CVD events in the range of 140–200 ug/mL.⁵⁰ Moreover, in a post-hoc analysis of JELIS, a plasma EPA level ≥150 ug/dL was associated with a significant reduction in CHD event risk.⁵¹

Specialized Pro-Resolving Lipid Mediators as Potential Mechanisms of Benefit

Other potential cardiovascular protective mechanisms for EPA and DHA include increased levels of the anti-inflammatory and specialized proresolving lipid mediators (SPMs), the downstream products of EPA and DHA.⁵² EPA and DHA are precursors of the SPMs, resolvins and maresins, which actively resolve chronic inflammation.⁵² The HEARTS trial has taken a close look at this potential mechanism. Higher plasma levels of EPA+DHA were associated with significantly increased levels of two SPMs - resolvin E1 and maresin 1- and 18-hydroxy-eicosapentaenoic acid (HEPE), the precursor of resolvin E1.³⁶ Those with low plasma EPA+DHA levels had a low (18-HEPE+resolvin E1)/leukotriene B₄ (LTB₄) ratio and significant plaque progression. LTB₄ is a proinflammatory lipid mediator. Those with high plasma EPA+DHA levels had either low (18-HEPE+resolvin E1)/LTB₄ ratios with significant plaque progression or high (18-HEPE+resolvin E1)/LTB₄ ratios with significant plaque regression (Figure 1).³⁶ These findings suggest that an imbalance between pro-resolving and proinflammatory lipid mediators is associated with plaque progression and potentially mediates the beneficial effects of EPA and DHA. Moreover, the level of SPMs, the downstream products of omega-3 fatty acids, are more important in determining plaque regression and progression than the omega-3 fatty acid level. The role of SPMs in plaque progression and regression should be examined further as the ability to form the downstream products of omega-3 fatty acids could be a crucial determining factor in the effect on plaque regression and reduction of cardiovascular events.

Figure 1. Relationship between Omega-3 Fatty Acid Index, ratio of 18 HEPE+RvE1/LTB4 and Regression and Progression of Coronary Plaque.

A low omega-3 fatty acid index is associated with coronary plaque progression and a low ratio of 18 HEPE+RvE1/LTB4. Those with a high omega-3 fatty acid index fell into 2 groups: those with a high 18- HEPE+RvE1/LTB4 ratio had regression of coronary plaque whereas those with a low 18- HEPE+RvE1/LTB4 ratio had progression of coronary plaque. 18-HEPE indicates, 18-hydroxy-eicosapentaenoic acid; RvE1, resolvin E1; LTB₄, leukotriene B₄.

Omega-3 Fatty Acids and Cognitive Function

DHA is a major fatty acid in membrane phospholipids in the grey matter of the brain and makes up approximately 25% of total fatty acids in the human cerebral cortex and 50% of all PUFA in the central nervous system.^53–55 DHA is transported into the brain in the form of lysophosphatidylcholine by the major facilitator superfamily domain-containing protein 2a (Mfsd2a).⁵⁶ Mfsd2a knockout mice had 50% lower levels of DHA in brains with consequent cognitive deficits and neuronal cell loss in hippocampus and cerebellum, providing strong support for the role of DHA in cognitive function.⁵⁶

DHA and Neural Development

An increased risk of poor neural development in infants has been observed with low levels of maternal DHA in epidemiological studies.⁵⁷ A deficiency of dietary n-3 PUFA during fetal development in utero and postnatal has been associated with a loss of discriminative learning ability.⁵⁸ Infants born to mothers whose diet was supplemented with DHA had higher mental processing scores and higher degrees of stereopsis and stereoacuity at 4 years of age.^59,60 Breast-fed infants of mothers who were supplemented with DHA during lactation had significantly better psychomotor development, hand-eye coordination and visual acuity at 2.5 years compared to breast-fed infants of mothers who received a placebo.⁶¹ At least 300 mg of DHA daily is recommended for pregnant and nursing women to meet the needs of the fetus and breast-feeding infant.⁶²

Prospective Studies of Cognitively Healthy Subjects

In 2021, Wood et al.⁶³ reported a reduction of development of Alzheimer’s disease with fish intake in 5 prospective cohort studies in those > age 55. At 5.4-year follow-up in one of the cohorts, the Cardiovascular Health Cognition Study (n= 2,233 individuals age ≥ 65 years), consumption of more than 2 fatty fish meals per week showed a 41% decreased risk of Alzheimer’s disease in those without the apoE4 allele (95% CI: 0.36-0.95) compared to those eating less than once/month.⁶⁴ Outcome was not reported by sex or race.

Differences by sex were shown to be important in a 2022 cross-sectional report of 386 healthy adults (mean age 77.4±3.8 years).⁶⁵ Higher n-3 PUFA concentrations were associated with better non-verbal memory and processing speed in fully adjusted models (β’s of 0.21 and 0.19, respectively). Men with higher n-3 PUFA levels had better executive functioning and processing speed whereas women had better verbal and non-verbal episodic memory.⁶⁵ The difference by sex suggests that failure to account for sex could account for variability in studies and needs to be considered in future studies.

Prospective Studies in Mild Cognitive Decline

In 2022, Kosti et al.⁶⁶ reported a dose-response meta-analysis of 11 prospective, cohort studies of 46,488 individuals some of whom had mild cognitive impairment but were without dementia or Alzheimer’s disease at baseline. Fish intake of up to 2 portions (250 g) per week was associated with a 30% reduction in risk of Alzheimer’s disease (95% CI:0.54-0.89, 3 studies). Sex or race were not reported separately.

Randomized Controlled Trials of Omega-3 Fatty Acids in Mild Cognitive Impairment or Alzheimer’s Dementia

In 9 RCTs of which 5 included mild cognitive impairment or Alzheimer’s disease reviewed by Kosti et al.⁶⁶ in 2022, a dose-response relationship was found between the omega-3 fatty acid index achieved and improvement in executive function but not in general cognition. Of these 9, age range was > 50 in 7 trials but > 18 in 2 trials which may have minimized benefit. A wide range in EPA and/or DHA supplementation (from 2 to 36 months) and doses (112.5 mg to 1680 mg EPA and 297 mg to 1680 mg of DHA) may also have affected outcome.

In a 2022 review by Wood et al.⁶³ of 13 RCTs of adults ≥50 years with mild cognitive impairment, supplemental n-3 PUFA had a protective effect against cognitive decline in 9 studies but no effect in 4 studies with higher amounts showing most benefit. Importantly, significant improvement occurred prior to cognitive impairment or with early and mild cognitive impairment whereas those with Alzheimer’s disease had no benefit, suggesting that for the best outcome, n-3 PUFA should be started before cognitive decline or with mild cognitive decline.

Randomized Trials of Omega-3 Fatty Acid Supplementation in Cognitively Healthy Subjects

Table 8 reports the results of 15 RCTs with EPA+DHA in cognitively healthy subjects; 7 reported benefit on cognitive function measured by neuropsychological testing whereas 8 reported no benefit.^67–82 Differences in outcomes in the trials could be due to differences in dose of EPA+DHA, trial duration, the type of placebo, number of subjects, the types of cognitive function tests employed and apoE genotype. The VITAL RCT reported above for CVD outcomes had an ancillary study of cognitive function. No benefit was observed by sex or race in the 4,218 subjects randomized to 860 mg EPA+DHA compared to placebo.⁷⁹ In a small RCT of 57 women, mean age 72±7 years, in a retirement center in urban South Africa in 2022, those randomized to 2.2 g n-3 PUFA daily (canned pilchards and fish spread) had a significantly higher cognitive abilities screening instrument score at 12 weeks compared to canned meat balls and texturized soya.⁸⁰ Otherwise, none of the trials reported outcomes by sex or race.

Table 8.

Summary of randomized, controlled trials of omega-3 fatty acids in cognitively healthy subjects

First Author Year, Ref #	Number subjects Age range Age, mean±SD, years Study length % Male	Intervention	COWAT	DSST	TMT A	TMT B	Rey Auditory Verbal Learning Test
							RAVLT1 Immediate Recall	RAVLT2 Short delay	RAVLT3 Delayed recall	RAVLT 4 Delayed recognition
			P values
Malik 67 2021 CAD subjects	N=250 30 months Mean age, yrs Placebo: 63.9 ±7.4 years EPA/DHA: 62.9±7.6 years 83.2% male	DHA 1.5 g EPA 1.86 g Placebo: none, open label	0.011	0.12	.015	0.043	0.38	0.44	0.33	0.18
Kulzow 68 2016 *	N=44 50-75 years Mean age: years EPA/DHA: 63±6 Placebo: 61±6 26 weeks Male: 45%	EPA: 1,320 mg DHA: 880 mg Both with Vit E: 15 mg Placebo: sunflower oil 406 mg	Words starting with S: 0.25 Fruits: 0.62			0.55	0.33		0.48	0.87
Howe 69 2018	N=38 40-85 years Mean age: years EPA:63.2±1.6 Placebo:64.1±2.3 20 weeks Male: 68% Mildly Hypertensive	EPA: 400mg DHA: 1,600 mg Placebo: corn oil			P=0.5 for the sum of TMT A and TMT B
van de Rest 70 2008	N=302 Age ≥ 65 years Mean age: years EPA/DHA: 1800 mg: 69.9±3.4 EPA/DHA: 400 mg: 69.5±3.2 Placebo:70.1±4 26 weeks Male: 55%	High dose: EPA: 1,093 mg DHA 847 mg Low dose: EPA:226 mg DHA: 174 mg Placebo: high – oleic sunflower oil	NS		NS	NS	NS		NS	NS
Dangour 71 2010 †	N=867 70-79 years Mean age: years Fish oil: 74.7±2.5 Placebo:74.6±3 24 months Male: 55%	EPA: 200 mg DHA: 500 mg Placebo: olive oil	NS	NS			0.14		0.46
Chew 72 2015 ‡	N=3501 50-85 years Mean age: years 72.7±7.7 5 years Male: 42.5%	EPA 650 mg DHA 350 mg vs AREDS formulation Vit C, E, beta carotene and zinc	Animals: 0.34 Letters starting with F, A and S: 0.78
Witte 73 2014 §	N=65 ages 50-75 years Mean age: years 63.9 ± 6.6 26 weeks Male: 53.9%	EPA 1,320 mg DHA 880 mg placebo: sunflower oil	0.009
Mazereeuw 74 2016 CAD subjects	N=92 Mean age: years 61.7 ± 8.7 12 weeks 76% male	DHA,1200 mg/d EPA,600 mg/d EPA:DHAratio,1:2 Other: LC-PUFAs,100 mg/d Placebo: soybean/corn oil	NS			NS	No benefit on CVLT which is similar to RAVLT
Geleijnse 75 2012 Post-MI patients	N=2911 40 months Mean age, yrs Men: 68.7+5.5 years Women: 70.3+5.6 years	1. EPA:DHA 400mg/d EPA:DHAratio,3:2 2.EPA/DHA,400 mg/d ALA, 2g/d EPA:DHAratio,3:2 Placebo: margarine	No change in MMSE
Konagai 76 2013	N=45 3 months Mean age, yrs 67.1+3.4 years 100% male	1.Krill oil: DHA, 92mg/d EPA, 193mg/d DHA:EPAratio,1:2 2.Sardine oil: DHA, 251mg/d EPA, 491mg/d Placebo: medium chain triglyceride	Both krill oil & sardine oil had better P 300 ‖ compared to placebo
Maltais 77 2022	N=193 6 months 20-80 yrs	EPA, 1.7 g DHA, 0.8 g Ethyl ester; Placebo: high oleic acid soybean/corn oil	No difference in cognitive performance but those with low episodic memory score improved scores P = 0.043 ApoE genotype was not effect modifier
Ichinose 78 2021	N=53 12 months Mean age, yrs 69.1±5.3 55% female	EPA, 137 mg DHA, 297 mg in 200 ml milk beverage Placebo: 200 ml milk beverage	MMSE and HSD-R scores higher P < 0.05 in DHA compared to placebo at 12 months but only mean difference in MMSE was higher P < 0.05
Kang 79 2022	N=4,218 2-3 years Mean age, yrs 70.9±5.8 58.1% female	EPA: 460 mg DHA: 380 mg	No difference in annual rate of cognitive change using a global score for 8 neuropsychological tests.
Kuhn 80 2022	N=57 12 weeks Mean age, yrs 72±7 74% female	Canned pilchards and fish spread – 2.2 g omega-3 daily Control: canned meatballs and texturized soya	Significantly higher cognitive abilities screening instrument (CASI) score in intervention (P < 0.05).
Danthiir 81 2018	N=390 18 months Mean age, yrs 73.1±5.4 53% female	DHA: 1720 mg EPA: 600 mg Placebo: 990 mg oleic acid	No difference except negative effect on psychomotor speed in men. Treatment interaction with sex. ApoE4 carriers improved on reasoning (P=0.02)

^*Kulzow et al.⁶⁸ employed LOCATO which examines visual spatial object location of buildings and the RAVLT. After adjustment for changes in diet, those on omega-3 fatty acids did better on the LOCATO (p=0.004). As noted above, the RAVLT was not significant.

^†California Verbal Learning Test (CVLT) was performed which is similar to Rey Auditory Verbal Learning Test (RAVLT).

^‡This is an ancillary study of the Age-Related Eye Disease Study 2 (AREDS2) randomized clinical trial. A composite score was computed to obtain an overall score for a battery of 8 cognitive function tests and was not significant. The composite score included the scores of the following: Telephone Interview Cognitive Status-Modified –TISC-M word list for immediate recall, Verbal fluency with animal category, Verbal fluency with letter category, Verbal fluency alternating between animal and letter categories, Wechsler logical memory I and Wechsler logical memory II which measured immediate and delayed recall of 2 stories, Digits backward and TICS-M delayed recall. All test results were converted into z scores which were then added.

^§Witte et al.⁷³ calculated a composite executive score which was significant at P = 0.023. The composite executive score was defined as the sum of phonemic fluency and semantic fluency minus TMT (part B – part A)/part A and minus STROOP (part 3 – (part 1 + part 2))/2]/4. Preservation of both the integrity of the microstructure of white mass and the volume of gray mass was observed with a dose of 1320 mg of EPA plus 880 mg of DHA per day for 26 weeks.

^‖P 300 measures working memory by measuring event–related potentials using an electroencephalograph.⁷⁵ P300 depicts central nervous system activity involved in the processing of new information when attention is engaged to update memory representation. P300 has two components: latency, which reflects cognitive processing speed, and amplitude, which measures attentional resources engaged in task completion. Latency is prolonged and amplitude is decreased with aging, a finding suggesting that P300 testing can evaluate cognitive function in healthy elderly adults.

Abbreviations: ARA indicates arachidonic acid; COWAT, Controlled Oral Word Association Test; DSST, Digit-Symbol Substitution tests; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; MMSE, Mini-Mental Status examination; RAVLT, Rey Auditory Verbal Learning Test; RAVLT 1, immediate recall as a sum of 4 trials; RAVLT 2, short delay recall; RAVLT 3, delayed recall; RAVLT 4, delayed recognition; TMT, Trail Making Test.

In the HEARTS trial, cognitive function was assessed at baseline and at 12- and 30-month follow-up with neuropsychological testing as a prespecified secondary outcome. Over the 30-month period, subjects randomized to EPA and DHA had significant improvement at 12 and 30 months for verbal fluency, language and memory compared to control.⁶⁷ Post-hoc analyses indicated no difference by age, sex or diabetes status. Since DHA comprises approximately 25% of total fatty acids in the human cerebral cortex and 50% of all PUFA in the central nervous system,⁵³ outcome by plasma levels of EPA and DHA was examined. After adjustment for age, sex, diabetes, hypertension and statin, DHA levels above the median (85 ug/mL) were associated with improvement in verbal fluency, language and memory (controlled oral word association test, COWAT 1), and visual-motor coordination (trail making test, TMT A), but EPA levels were not (Figure 2, panel A).⁸³ The lack of benefit with EPA levels alone is in line with the findings from the Chicago Memory and Aging project which reported that total intake of n-3 PUFA and DHA were associated with reduced risk of Alzheimer’s disease but EPA alone was not.⁸⁴ Moreover, total intake of n-3 PUFA was protective among women (P interaction=0.02). In the HEARTS trial, an omega-3 index ≥ 4% had the strongest association for COWAT 1 and was also associated with improvement in the digit-symbol substitution test (DSST). Therefore, DHA appears more important than EPA for improved cognition in cognitively healthy adults with CAD; however, combined EPA and DHA led to the best improvement that could be due to conversion of EPA to DHA. Figure 2, panel B, shows that those with an omega-3 fatty acid index ≥4% had significant improvement at 12 months that was maintained at 30 months whereas those with an omega-3 fatty acid index <4% had no change, thus, cognitive aging was delayed by at least 30 months. These findings support combined EPA and DHA for improved cognitive function and may be especially important for patients with CAD, a risk factor for cognitive decline.

Figure 2. Cognitive Function Scores at 30 Months Grouped by EPA, DHA and Omega-3 Fatty Acid Index.

A. Median difference in cognitive function score for the three groups: DHA, EPA and Omega-3 Fatty Acid Index of 4%. Data shown as median [interquartile range]. Median for DHA is 85 ug/ml; median for EPA is 52.75 ug/ml. *Adjusted for age, sex, body mass index, hypertension status, diabetes status and statin use. A lower number is better for TMT; for all others, a higher number is better. Continuous variables were reported as the median (interquartile range [IQR]) and compared using the Wilcoxon-Mann-Whitney U test for unpaired analyses and Wilcoxon signed-rank test for paired analyses. COWAT indicates Controlled Oral Word Association Test; DHA, Docosahexaenoic Acid; DSST, Digit-Symbol Substitution test; EPA, Eicosapentaenoic Acid; FA, Fatty Acid; n-3, Omega-3; TMT, Trail Making Test.

B. Change in Digit-Symbol Substitution Test Score by Omega-3 Fatty Acid Index Cut Point of 4%. Mean changes from baseline are shown. Compared to baseline, a significant increase is seen in the DSST score for an omega-3 fatty acid index ≥4% at both 12 and 30 months (paired t-tests; P<0.001 for both) whereas an omega-3 fatty acid index <4% had no difference (P=0.48 and 0.31 for 12 and 30 months, respectively). These results support that those in the omega-3 fatty acid index ≥4% group had delay in cognitive aging of at least 30 months.

Summary

Differences in outcome by sex and race are summarized as follows:

CVD Outcomes:

Prospective cohort studies have reported a lower rate of MI in Danish women with fatty fish intake and lower rate of CHD with 2 fish servings per week in the Nurse’s Health Study. Women in Shanghai had lower rates of total death, CVD death and hemorrhagic stroke whereas men did not. Benefit appears related to blood levels of EPA and DHA – women in the top quintile of n-3 PUFA intake had lower rates of total mortality, CVD mortality and death from Alzheimer’s disease (NIH-AARP). In MESA, both women and men in the highest quartiles of EPA and DHA had lower rates of incident CVD. Moreover, black subjects in MESA had lower rates of MI, cardiac arrest, stroke, CHD or stroke death compared to other racial groups based on DHA levels, indicating a racial difference.

Supplementation with omega-3 fatty acid in RCTs has shown a race difference with reduction of MI in primary prevention of CVD for African Americans, a pre-specified subgroup, in Vital. For secondary prevention of CVD, women benefited in the Risk and Prevention Trial, Asians as a pre-specified subgroup in the Strength Trial and both men and women in the REDUCE-IT trial. A review has summarized the medical societies that have endorsed the use of icosapent ethyl for established CVD or type 2 DM for patients on a statin and with elevated triglyceride levels generally in the range of 200-499 mg/dL.⁸⁵ A meta-analysis reported that those receiving EPA+DHA had benefit on cardiac events (but not stroke) that was comparable to those receiving EPA alone.³⁹ It is also noteworthy that in STRENGTH, CVD events (which include stroke) were not significantly reduced whereas cardiac events that did not include stroke were reduced in those with established CVD. Taken together, these findings suggest that a new trial should be considered in Asians with omega-3 carboxylic acid and in Black subjects with low dose EPA+DHA in primary prevention and with cardiac events (without stroke) as the primary endpoint. All future trials should be required to enroll enough women and racial groups in order to assess benefit. The HEARTS trial reported benefit with high-dose EPA+DHA on cardiac events and plaque regression in patients with normal triglyceride levels and provided the first evidence that the downstream products of EPA and DHA, the resolvins, are more important in determining benefit than the omega-3 fatty acid level. Moreover, normotensive subjects had greater benefit than hypertensive subjects potentially due to lower levels of inflammation. Future studies should examine the role of blood pressure and inflammation, including resolvins, in response to omega-3 fatty acids as differential responses could account for differences in outcomes of RCTs.

Cognitive Outcomes:

Prospective studies suggest that fish intake of 2 portions/week was associated with a 30% reduction in Alzheimer’s disease with a dose-response relationship noted between the omega-3 fatty acid index achieved; however, for the best outcome, n-3 PUFA should be started before cognitive decline. DHA may be more important than EPA due to the high content of DHA in the brain.

The Office of Dietary Supplements has provided Global Recommendations for fish and EPA and DHA Intake at the following link: https://www.issfal.org/assets/globalrecommendationssummary19nov2014landscape_-3-.pdf. These recommendations are for all life stages including infants, pregnant women, healthy adults, including for both primary and secondary prevention of cardiovascular disease, heart failure and multiple other conditions and diseases.

Conclusions

These findings suggest that intake of fatty and nonfried fish and higher plasma levels of DHA and EPA are important in primary prevention of CVD and cognitive decline. Higher doses of EPA and DHA, especially icosapent ethyl, may be required in secondary CVD prevention in the setting of high-intensity statin use. The secondary endpoint of cardiac events, which excluded fatal and nonfatal stroke, was significant in the STRENGTH trial and MI was significant in Black men in VITAL These findings suggest that EPA+DHA may be beneficial in reducing cardiac events but have less benefit on stroke, the latter affected more by hypertensive status. These findings should be examined in future trials of EPA+DHA supplementation. Given the prevalence of CVD and cognitive decline, clinicians and patients should be aware of the potential benefits of fatty fish and EPA and DHA supplementation outlined in this review to make informed decisions. This becomes especially important on a population basis considering the huge financial burden of CVD and cognitive decline and the low cost and side effects of fish intake and omega-3 fatty acid supplementation.

Highlights.

• Prospective cohort studies of fatty fish reported lower rates of MI in Danish women, lower rates of MI, total death, CVD death, hemorrhagic stroke and Alzheimer’s death with total fish intake in Asian women and lower total mortality, CVD mortality and Alzheimer’s death in women in NIH-AARP.

• Primary Prevention of CVD: The VITAL RCT of 860 mg EPA+DHA daily was neutral for CVD for the total group although African-American subjects had a 77% reduction in MI as a secondary endpoint.

• Secondary Prevention of CVD RCTs: Risk and Prevention: 18% reduction in women with low dose EPA+DHA; REDUCE-IT: 26% reduction in cardiac events and stroke with pure EPA in both men and women; Strength: no benefit with EPA+DHA for total group although Asians had a significant 28% reduction in the primary endpoint of major CVD events; 15% reduction occurred in cardiac events (without stroke) as a secondary endpoint.

• Resolvins, the downstream products of EPA+DHA, predicted regression of coronary plaque better than omega- fatty acid levels in the HEARTS RCT.

• In a RCT of 3.36 g EPA+DHA, an omega-3 fatty acid index ≥4% was associated with significant improvement of cognitive function at 12 and 30 months compared to <4% with DHA more important than EPA.

Non-standard Abbreviations and Acronyms

AFIB

atrial fibrillation

APO

apolipoprotein

ASCEND

Study of Cardiovascular Events in Diabetes

CAD

coronary artery disease

CHD

coronary heart disease

COWAT

controlled oral word association test

CVD

cardiovascular disease

DHA

docosahexaenoic acid

EPA

eicosapentaenoic acid

GISSI

Gruppo Italiano per lo Studio della Sopravvivenza nell-Infarto Miocardico

HEARTS

Slowing HEART disease with lifestyle and omega-3 fatty acids

18-HEPE

18-hydroxy-eicosapentaenoic acid

HR

hazard ratio

JELIS

Japan Eicosapentaenoic Lipid Intervention Study

LTB₄

leukotriene B₄

MCI

mild cognitive impairment

MESA

Multi-Ethnic Study of Atherosclerosis

Mfsd2a

major facilitator superfamily domain-containing protein 2a

MI

myocardial infarction

n-3 PUFA

omega-3 polyunsaturated fatty acids

NIH-AARP

National Institute of Health American Association of Retired Persons

OMEMI

Omega-3 Fatty acids in Elderly with Myocardial Infarction

OR

odds ratio

REDUCE-IT

Reduction of Cardiovascular Events with Icosapent Ethyl-Intervention Trial

RCT

randomized controlled trial

STRENGTH

Long-Term Outcomes Study to Assess Statin Residual Risk with Epanova in High Cardiovascular Risk Patients with Hypertriglyceridemia

SPMs

specialized proresolving lipid mediators

VITAL

VITamin D and OmegA-3 Trial

VLDL

very low density lipoprotein