EPA+DHA · lipids & cardiovascular risk
Overview
SummaryIntroductionAim & questionMethods
Results
Study selection (PRISMA)Included studies Lipids — meta-analysisSensitivity Cardiovascular outcomesWomen-specific (HR)NetworkRanking (SUCRA)
Conclusions
DiscussionLimitationsConclusionsReferences
RUEN
Draft

Omega-3 (EPA+DHA) in peri- and postmenopausal women: effects on lipids and cardiovascular risk — a systematic review and meta-analysis

Quantitative synthesis of the effect of EPA+DHA on the lipid profile (primary outcome — triglycerides) and cardiovascular events in peri- and postmenopausal women.
4088 PubMed recordsPrimary population — pure postmenopausePrimary outcome — triglycerides
Search date: 28 May 2026Analysis date: 10 June 2026Published: 10 June 2026
Key findings

Summary

In peri- and postmenopausal women, EPA+DHA produces a significant reduction in triglycerides with a neutral effect on LDL-C and HDL-C. This is a robust quantitative result obtained in dedicated women / postmenopausal cohorts. Evidence on hard cardiovascular outcomes in women is limited and treated as supportive.

−8.7 mg/dL
Triglycerides (TG) · mean difference
95% CI [−12.8, −4.6] · k=8
+0.11 mmol/L
LDL-C · mean difference
95% CI [0.07, 0.15] · k=8
+0.05 mmol/L
HDL-C · mean difference
95% CI [−0.04, 0.13] · k=7
Cardiovascular outcomes
HR 0.87
Major CVD — women · direct data
95% CI [0.75, 1.00] · p=0.055
OR 0.94
All-cause mortality · whole pop.
95% CI [0.90, 0.98] · k=43
OR 0.91
Myocardial infarction · whole pop.
95% CI [0.85, 0.98] · k=23
OR 1.10
Stroke · whole pop.
95% CI [1.01, 1.20] · k=17

Cardiovascular estimates in women come from published HRs (direct data); whole-population ORs are shown as a supportive, indirect tier — mainly from male-predominant trials. See the «Cardiovascular outcomes» section for details.

Context

Introduction

After menopause, a woman's lipid profile predictably worsens — triglycerides and non-HDL cholesterol rise — contributing to increased cardiovascular risk. The long-chain omega-3 fatty acids eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) lower triglycerides and are widely used in cardiometabolic prevention.

Despite a wealth of omega-3 research, quantitative data specific to postmenopausal women remain fragmented: the large cardiovascular trials enrolled mostly men, and women's subgroups were reported inconsistently. We focused precisely on this population and on the most reproducible, quantifiable outcome — the lipid response, primarily triglycerides.

Objective

Aim and review question

To quantitatively assess the effect of EPA+DHA on the lipid profile in peri- and postmenopausal women, and to characterise the available evidence on cardiovascular outcomes in this population.

PICO. P — peri-/postmenopausal women · I — EPA+DHA · C — placebo or standard care/diet · O — primarily triglycerides (TG); secondarily LDL-C, HDL-C and cardiovascular events (MACE, mortality, MI, stroke). Primary population — pure postmenopause; broader women-percentage definitions are sensitivity analyses.
Methodology

Methods

Search

A systematic search of PubMed (omega-3 / EPA / DHA) and the ClinicalTrials.gov registry, supplemented by targeted citation searching to capture landmark cardiovascular trials. Reproducible query (simplified form):

("fatty acids, omega-3"[Mesh] OR omega-3[tiab] OR "n-3 PUFA"[tiab] OR eicosapentaenoic[tiab] OR docosahexaenoic[tiab] OR EPA[tiab] OR DHA[tiab] OR "fish oil"[tiab]) AND (postmenopaus*[tiab] OR menopaus*[tiab] OR women[tiab] OR female[tiab]) AND (triglycerid*[tiab] OR lipid*[tiab] OR cardiovascular[tiab]) AND randomized controlled trial[pt]

The base search returned 4088 PubMed records and 414 registered RCTs; an expanded iteration and citation searching added further records. Open the query in PubMed →

Inclusion criteria

Randomised controlled trials; intervention — EPA+DHA (and EPA- and DHA-mono preparations for the network comparison); comparator — placebo, standard care or diet; a population of peri-/postmenopausal women; reported lipid and/or cardiovascular outcomes. The primary population is pure postmenopause (postmenopause fraction 100% or explicitly postmenopausal status); less strict women-percentage definitions are used in sensitivity analyses.

Extraction and analysis

Data were extracted independently over two iterations with disagreement resolution; values feeding the pooled estimates were checked against the source (including reconstructing the standard error of the change from reported SE / 95% CI per Cochrane methods). The primary analysis pooled the mean difference of change, EPA+DHA minus control, under a random-effects model (DerSimonian–Laird) with inverse-variance weighting. Triglycerides were harmonised to mg/dL; LDL-C and HDL-C to mmol/L. Binary cardiovascular outcomes were pooled as odds ratios (OR, log scale) in a two-tier scheme: Tier 1 — women population, Tier 2 — the whole study population (supportive, indirect). Additionally, a women-subgroup analysis based on published hazard ratios (HR) and a network meta-analysis of omega-3 formulations were performed.

Risk-of-bias assessment. For the included randomised trials, risk of bias was assessed with the Cochrane RoB 2 tool across the domains of randomisation, deviations from the intended intervention, missing data, outcome measurement and selective reporting.
Selection

Study selection flow (PRISMA 2020)

Identification: PubMed 4088 + ClinicalTrials.gov 414 RCTs + expanded & citation search
↓
Screening (title/abstract): selected for full text ~1082
Excluded at screening: irrelevant population/intervention, no outcomes, non-RCT, non-human
↓
Corpus assembled and processed: 616 articles; full text retrieved ≈ 96.6%
Excluded at full text: insufficient female %, no relevant outcome, duplicates, protocol-only
↓
Included in the quantitative synthesis: lipid pool (postmenopause) k=8 for triglycerides; cardiovascular outcomes — dozens of trials in the supportive (whole-population) tier

A detailed interactive selection flow with per-branch counts is maintained in the project workbench.

Evidence base

Characteristics of included studies

The quantitative lipid pool is formed by trials conducted directly in postmenopausal / women cohorts with measured lipid change. The largest contribution to the triglyceride estimate comes from the large Gaengler 2024 trial (n ≈ 535).

StudyYearN (T/C)InterventionPostmenopause %Women %
Stark 2000200018/17EPA+DHA_high_>=3g100100
Ciubotaru 2003200310/10EPA+DHA_high_>=3g100100
Kabir 2007200712/14EPA+DHA_standard_1-2g100100
Nelson 2011201120/19EPA+DHA_standard_1-2g100100
Crochemore 2012201214/13EPA+DHA_standard_1-2g100100
Murphy 2021202131/27EPA+DHA_standard_1-2g100100
Félix-Soriano 2021202115/20EPA+DHA_standard_1-2g100100
Lee 2023202310/10EPA+DHA_standard_1-2g100100
Gaengler 20242024266/269EPA+DHA_standard_1-2g100100
Primary outcome

Lipids — meta-analysis (EPA+DHA vs control)

Each study is shown as a point estimate of the mean difference in change (a square whose area is proportional to the study weight) and its 95% confidence interval; the diamond is the pooled random-effects estimate. Triglyceride values are in mg/dL; LDL-C and HDL-C in mmol/L. Clicking a row opens the publication in PubMed.

Robustness

Sensitivity analysis by population strictness

The primary triglyceride estimate is robust to broadening the population from pure postmenopause to women-predominant cohorts — the direction and significance of the effect are preserved (TG in mg/dL, LDL-C/HDL-C in mmol/L):

Population definitionOutcomekMD95% CI
Pure postmenopause (primary)TG8−8.7[−12.8, −4.6]
Pure postmenopause (primary)LDL-C80.11[0.07, 0.15]
Pure postmenopause (primary)HDL-C70.05[−0.04, 0.13]
Women-predominant cohorts (≥50%)TG10−7.5[−11.5, −3.6]
Women-predominant cohorts (≥50%)LDL-C110.10[0.06, 0.15]
Women-predominant cohorts (≥50%)HDL-C100.03[−0.05, 0.11]
Secondary outcomes

Cardiovascular outcomes — a two-tier analysis

Tier 1 (women) — trials in a women population; few and less stable. Tier 2 (whole population) — all omega-3 trials, including the large landmark, predominantly-male studies; treated as a supportive, indirect evidence tier for women, not the primary estimate. Odds ratios (OR) are shown; OR < 1 favours EPA+DHA.
Tier 1 — women population
OR 0.86 [0.78, 0.95]; k=5; p = 0.004.
Study OR Lok 2012 (9/99 vs 17/97) Lok 2012 (9/99 vs 17/97) · weight 1.4% 0.47 [0.20, 1.11] Mozaffarian 2012 (13/758 vs 20/758) Mozaffarian 2012 (13/758 vs 20/758) · weight 2.1% 0.64 [0.32, 1.30] Macchia 2013 (16/289 vs 20/297) Macchia 2013 (16/289 vs 20/297) · weight 2.3% 0.81 [0.41, 1.60] Manson 2019 (386/12933 vs 419/12938) Manson 2019 (386/12933 vs 419/12938) · weight 53.1% 0.92 [0.80, 1.06] Hamaya 2025 (281/12933 vs 342/12938) Hamaya 2025 (281/12933 vs 342/12938) · weight 41.1% 0.82 [0.70, 0.96] Pooled OR (k=5) 0.86 [0.78, 0.95] · p=0.004 ← EPA+DHA control →
Tier 2 — whole population (supportive)
OR 0.95 [0.91, 0.99]; k=21; p = 0.006.
Study OR GISSI-Prevenzione Investigators 1999 ( GISSI-Prevenzione Investigators 1999 (715/5666 vs 785/5668) · weight 11.6% 0.90 [0.81, 1.00] Nilsen 2001 (42/150 vs 36/150) Nilsen 2001 (42/150 vs 36/150) · weight 0.5% 1.23 [0.73, 2.07] Marchioli 2001 (358/2835 vs 419/2828) Marchioli 2001 (358/2835 vs 419/2828) · weight 5.9% 0.83 [0.71, 0.97] Grundt 2004 (52/150 vs 49/150) Grundt 2004 (52/150 vs 49/150) · weight 0.6% 1.09 [0.68, 1.77] Brouwer 2006 (81/273 vs 90/273) Brouwer 2006 (81/273 vs 90/273) · weight 1.0% 0.86 [0.60, 1.23] GISSI-HF investigators 2008 (1981/3494 GISSI-HF investigators 2008 (1981/3494 vs 2053/3481) · weight 15.1% 0.91 [0.83, 1.00] Kromhout 2010 (336/2404 vs 335/2433) Kromhout 2010 (336/2404 vs 335/2433) · weight 5.1% 1.02 [0.86, 1.20] Rauch 2010 (182/1752 vs 149/1701) Rauch 2010 (182/1752 vs 149/1701) · weight 2.6% 1.21 [0.96, 1.52] Galan 2010 (202/1259 vs 200/1263) Galan 2010 (202/1259 vs 200/1263) · weight 3.0% 1.02 [0.82, 1.26] Lok 2012 (9/99 vs 17/97) Lok 2012 (9/99 vs 17/97) · weight 0.2% 0.47 [0.20, 1.11] Bosch 2012 (1034/6281 vs 1017/6255) Bosch 2012 (1034/6281 vs 1017/6255) · weight 15.3% 1.01 [0.92, 1.12] Mozaffarian 2012 (13/758 vs 20/758) Mozaffarian 2012 (13/758 vs 20/758) · weight 0.3% 0.64 [0.32, 1.30] Macchia 2013 (16/289 vs 20/297) Macchia 2013 (16/289 vs 20/297) · weight 0.3% 0.81 [0.41, 1.60] Bosch 2013 (733/6239 vs 745/6266) Bosch 2013 (733/6239 vs 745/6266) · weight 11.6% 0.99 [0.89, 1.10] ASCEND Study Collaborative Group 2018 ASCEND Study Collaborative Group 2018 (689/7740 vs 712/7740) · weight 11.3% 0.96 [0.86, 1.08] Foroughinia 2018 (4/37 vs 10/43) Foroughinia 2018 (4/37 vs 10/43) · weight 0.1% 0.40 [0.11, 1.40] Manson 2019 (386/12933 vs 419/12938) Manson 2019 (386/12933 vs 419/12938) · weight 6.9% 0.92 [0.80, 1.06] Kalstad 2021 (108/505 vs 102/509) Kalstad 2021 (108/505 vs 102/509) · weight 1.5% 1.09 [0.80, 1.47] Myhre 2022 (68/438 vs 60/443) Myhre 2022 (68/438 vs 60/443) · weight 1.0% 1.17 [0.81, 1.71] Bernhard 2024 (65/180 vs 62/178) Bernhard 2024 (65/180 vs 62/178) · weight 0.7% 1.06 [0.69, 1.63] Hamaya 2025 (281/12933 vs 342/12938) Hamaya 2025 (281/12933 vs 342/12938) · weight 5.4% 0.82 [0.70, 0.96] Pooled OR (k=21) 0.95 [0.91, 0.99] · p=0.006 ← EPA+DHA control →
Tier 1 — women population
OR 0.96 [0.89, 1.05]; k=25; p = 0.412.
Study OR Eritsland 1996 (8/317 vs 6/293) Eritsland 1996 (8/317 vs 6/293) · weight 0.6% 1.24 [0.42, 3.61] Ness 1999 (94/1015 vs 130/1018) Ness 1999 (94/1015 vs 130/1018) · weight 9.3% 0.70 [0.53, 0.92] Burr 2003 (283/1571 vs 242/1543) Burr 2003 (283/1571 vs 242/1543) · weight 20.7% 1.18 [0.98, 1.43] Calò 2005 (1/79 vs 2/81) Calò 2005 (1/79 vs 2/81) · weight 0.1% 0.51 [0.04, 5.70] Leaf 2005 (13/200 vs 12/202) Leaf 2005 (13/200 vs 12/202) · weight 1.1% 1.10 [0.49, 2.47] Friesecke 2008 (18/83 vs 22/82) Friesecke 2008 (18/83 vs 22/82) · weight 1.4% 0.76 [0.37, 1.54] Holman 2009 (1/401 vs 0/399) Holman 2009 (1/401 vs 0/399) · weight 0.1% 2.99 [0.12, 73.68] Saravanan 2010 (0/54 vs 2/54) Saravanan 2010 (0/54 vs 2/54) · weight 0.1% 0.19 [0.01, 4.11] Sorice 2011 (0/96 vs 0/105) Sorice 2011 (0/96 vs 0/105) · weight 0.0% 1.09 [0.02, 55.64] Han 2012 (0/18 vs 0/12) Han 2012 (0/18 vs 0/12) · weight 0.0% 0.68 [0.01, 36.35] Mozaffarian 2012 (8/758 vs 15/758) Mozaffarian 2012 (8/758 vs 15/758) · weight 1.0% 0.53 [0.22, 1.25] Macchia 2013 (4/289 vs 5/297) Macchia 2013 (4/289 vs 5/297) · weight 0.4% 0.82 [0.22, 3.08] Kumar 2013 (1/39 vs 1/39) Kumar 2013 (1/39 vs 1/39) · weight 0.1% 1.00 [0.06, 16.58] Hoogeveen 2014 (166/2179 vs 67/739) Hoogeveen 2014 (166/2179 vs 67/739) · weight 8.3% 0.83 [0.61, 1.11] Shinto 2014 (1/13 vs 1/13) Shinto 2014 (1/13 vs 1/13) · weight 0.1% 1.00 [0.06, 17.90] Maki 2014 (0/37 vs 0/36) Maki 2014 (0/37 vs 0/36) · weight 0.0% 0.97 [0.02, 50.37] Nigam 2014 (0/153 vs 1/163) Nigam 2014 (0/153 vs 1/163) · weight 0.1% 0.35 [0.01, 8.73] Hoogeveen 2015 (166/2179 vs 67/739) Hoogeveen 2015 (166/2179 vs 67/739) · weight 8.3% 0.83 [0.61, 1.11] Vanderbilt 2015 (0/126 vs 0/64) Vanderbilt 2015 (0/126 vs 0/64) · weight 0.0% 0.51 [0.01, 25.99] Omrani 2015 (1/30 vs 0/30) Omrani 2015 (1/30 vs 0/30) · weight 0.1% 3.10 [0.12, 79.23] de Borst 2017 (46/171 vs 39/140) de Borst 2017 (46/171 vs 39/140) · weight 2.9% 0.95 [0.58, 1.57] Stroes 2018 (0/81 vs 0/81) Stroes 2018 (0/81 vs 0/81) · weight 0.0% 1.00 [0.02, 51.01] Manson 2019 (493/12933 vs 485/12938) Manson 2019 (493/12933 vs 485/12938) · weight 44.9% 1.02 [0.90, 1.16] Yokote 2020 (1/306 vs 0/77) Yokote 2020 (1/306 vs 0/77) · weight 0.1% 0.76 [0.03, 18.86] Мареев 2020 (1/30 vs 1/10) Мареев 2020 (1/30 vs 1/10) · weight 0.1% 0.31 [0.02, 5.48] Pooled OR (k=25) 0.96 [0.89, 1.05] · p=0.412 ← EPA+DHA control →
Tier 2 — whole population (supportive)
OR 0.94 [0.90, 0.98]; k=43; p = 0.008.
Study OR Sacks 1995 (0/41 vs 1/39) Sacks 1995 (0/41 vs 1/39) · weight 0.0% 0.31 [0.01, 7.82] Eritsland 1996 (8/317 vs 6/293) Eritsland 1996 (8/317 vs 6/293) · weight 0.2% 1.24 [0.42, 3.61] GISSI-Prevenzione Investigators 1999 ( GISSI-Prevenzione Investigators 1999 (472/5666 vs 545/5668) · weight 10.8% 0.85 [0.75, 0.97] Ness 1999 (94/1015 vs 130/1018) Ness 1999 (94/1015 vs 130/1018) · weight 2.3% 0.70 [0.53, 0.92] Nilsen 2001 (11/150 vs 11/150) Nilsen 2001 (11/150 vs 11/150) · weight 0.2% 1.00 [0.42, 2.38] Marchioli 2001 (239/2835 vs 299/2828) Marchioli 2001 (239/2835 vs 299/2828) · weight 5.7% 0.78 [0.65, 0.93] Burr 2003 (283/1571 vs 242/1543) Burr 2003 (283/1571 vs 242/1543) · weight 5.1% 1.18 [0.98, 1.43] Calò 2005 (1/79 vs 2/81) Calò 2005 (1/79 vs 2/81) · weight 0.0% 0.51 [0.04, 5.70] Leaf 2005 (4/100 vs 10/100) Leaf 2005 (4/100 vs 10/100) · weight 0.1% 0.38 [0.11, 1.24] Leaf 2005 (13/200 vs 12/202) Leaf 2005 (13/200 vs 12/202) · weight 0.3% 1.10 [0.49, 2.47] Brouwer 2006 (8/273 vs 14/273) Brouwer 2006 (8/273 vs 14/273) · weight 0.2% 0.56 [0.23, 1.35] Friesecke 2008 (18/83 vs 22/82) Friesecke 2008 (18/83 vs 22/82) · weight 0.4% 0.76 [0.37, 1.54] GISSI-HF investigators 2008 (955/3494 GISSI-HF investigators 2008 (955/3494 vs 1014/3481) · weight 16.6% 0.92 [0.82, 1.02] Holman 2009 (1/401 vs 0/399) Holman 2009 (1/401 vs 0/399) · weight 0.0% 2.99 [0.12, 73.68] Saravanan 2010 (0/54 vs 2/54) Saravanan 2010 (0/54 vs 2/54) · weight 0.0% 0.19 [0.01, 4.11] Kromhout 2010 (186/2404 vs 184/2433) Kromhout 2010 (186/2404 vs 184/2433) · weight 4.0% 1.02 [0.83, 1.27] Rauch 2010 (88/1919 vs 70/1885) Rauch 2010 (88/1919 vs 70/1885) · weight 1.8% 1.25 [0.90, 1.72] Galan 2010 (67/1259 vs 60/1263) Galan 2010 (67/1259 vs 60/1263) · weight 1.4% 1.13 [0.79, 1.61] Nodari 2011 (0/67 vs 0/66) Nodari 2011 (0/67 vs 0/66) · weight 0.0% 0.99 [0.02, 50.38] Farquharson 2011 (2/97 vs 1/97) Farquharson 2011 (2/97 vs 1/97) · weight 0.0% 2.02 [0.18, 22.66] Sorice 2011 (0/96 vs 0/105) Sorice 2011 (0/96 vs 0/105) · weight 0.0% 1.09 [0.02, 55.64] Han 2012 (0/18 vs 0/12) Han 2012 (0/18 vs 0/12) · weight 0.0% 0.68 [0.01, 36.35] Bosch 2012 (951/6281 vs 964/6255) Bosch 2012 (951/6281 vs 964/6255) · weight 19.1% 0.98 [0.89, 1.08] Mozaffarian 2012 (8/758 vs 15/758) Mozaffarian 2012 (8/758 vs 15/758) · weight 0.2% 0.53 [0.22, 1.25] Sandesara 2012 (0/120 vs 0/123) Sandesara 2012 (0/120 vs 0/123) · weight 0.0% 1.02 [0.02, 52.07] Macchia 2013 (4/289 vs 5/297) Macchia 2013 (4/289 vs 5/297) · weight 0.1% 0.82 [0.22, 3.08] Kumar 2013 (1/39 vs 1/39) Kumar 2013 (1/39 vs 1/39) · weight 0.0% 1.00 [0.06, 16.58] Hoogeveen 2014 (166/2179 vs 67/739) Hoogeveen 2014 (166/2179 vs 67/739) · weight 2.1% 0.83 [0.61, 1.11] Shinto 2014 (1/13 vs 1/13) Shinto 2014 (1/13 vs 1/13) · weight 0.0% 1.00 [0.06, 17.90] Wilbring 2014 (1/99 vs 2/99) Wilbring 2014 (1/99 vs 2/99) · weight 0.0% 0.49 [0.04, 5.55] Maki 2014 (0/37 vs 0/36) Maki 2014 (0/37 vs 0/36) · weight 0.0% 0.97 [0.02, 50.37] Nigam 2014 (0/153 vs 1/163) Nigam 2014 (0/153 vs 1/163) · weight 0.0% 0.35 [0.01, 8.73] Vanderbilt 2015 (0/126 vs 0/64) Vanderbilt 2015 (0/126 vs 0/64) · weight 0.0% 0.51 [0.01, 25.99] Omrani 2015 (1/30 vs 0/30) Omrani 2015 (1/30 vs 0/30) · weight 0.0% 3.10 [0.12, 79.23] Feguri 2017 (1/29 vs 1/28) Feguri 2017 (1/29 vs 1/28) · weight 0.0% 0.96 [0.06, 16.21] de Borst 2017 (46/171 vs 39/140) de Borst 2017 (46/171 vs 39/140) · weight 0.7% 0.95 [0.58, 1.57] Stroes 2018 (0/81 vs 0/81) Stroes 2018 (0/81 vs 0/81) · weight 0.0% 1.00 [0.02, 51.01] ASCEND Study Collaborative Group 2018 ASCEND Study Collaborative Group 2018 (752/7740 vs 788/7740) · weight 16.3% 0.95 [0.85, 1.05] Manson 2019 (493/12933 vs 485/12938) Manson 2019 (493/12933 vs 485/12938) · weight 11.1% 1.02 [0.90, 1.16] Yokote 2020 (1/306 vs 0/77) Yokote 2020 (1/306 vs 0/77) · weight 0.0% 0.76 [0.03, 18.86] Мареев 2020 (1/30 vs 1/10) Мареев 2020 (1/30 vs 1/10) · weight 0.0% 0.31 [0.02, 5.48] Kalstad 2021 (28/505 vs 28/509) Kalstad 2021 (28/505 vs 28/509) · weight 0.6% 1.01 [0.59, 1.73] Bernhard 2024 (20/180 vs 12/178) Bernhard 2024 (20/180 vs 12/178) · weight 0.3% 1.73 [0.82, 3.65] Pooled OR (k=43) 0.94 [0.90, 0.98] · p=0.008 ← EPA+DHA control →
Tier 1 — women population
OR 0.95 [0.76, 1.19]; k=7; p = 0.654.
Study OR Leaf 2005 (9/200 vs 9/202) Leaf 2005 (9/200 vs 9/202) · weight 5.5% 1.01 [0.39, 2.60] Mozaffarian 2012 (0/758 vs 3/758) Mozaffarian 2012 (0/758 vs 3/758) · weight 0.6% 0.14 [0.01, 2.76] Kumar 2013 (1/38 vs 1/38) Kumar 2013 (1/38 vs 1/38) · weight 0.6% 1.00 [0.06, 16.59] Maki 2014 (0/49 vs 0/49) Maki 2014 (0/49 vs 0/49) · weight 0.3% 1.00 [0.02, 51.41] Vanderbilt 2015 (0/81 vs 0/81) Vanderbilt 2015 (0/81 vs 0/81) · weight 0.3% 1.00 [0.02, 51.01] Manson 2019 (142/12933 vs 148/12938) Manson 2019 (142/12933 vs 148/12938) · weight 92.3% 0.96 [0.76, 1.21] Yokote 2020 (0/152 vs 0/77) Yokote 2020 (0/152 vs 0/77) · weight 0.3% 0.51 [0.01, 25.86] Pooled OR (k=7) 0.95 [0.76, 1.19] · p=0.654 ← EPA+DHA control →
Tier 2 — whole population (supportive)
OR 0.90 [0.84, 0.95]; k=22; p < 0.001.
Study OR Sacks 1995 (0/4 vs 1/4) Sacks 1995 (0/4 vs 1/4) · weight 0.0% 0.26 [0.01, 8.52] GISSI-Prevenzione Investigators 1999 ( GISSI-Prevenzione Investigators 1999 (291/5666 vs 348/5668) · weight 13.5% 0.83 [0.71, 0.97] Nilsen 2001 (8/150 vs 8/150) Nilsen 2001 (8/150 vs 8/150) · weight 0.3% 1.00 [0.37, 2.74] Marchioli 2001 (144/2835 vs 204/2828) Marchioli 2001 (144/2835 vs 204/2828) · weight 7.2% 0.69 [0.55, 0.86] Grundt 2004 (8/150 vs 16/150) Grundt 2004 (8/150 vs 16/150) · weight 0.4% 0.47 [0.20, 1.14] Leaf 2005 (2/28 vs 5/28) Leaf 2005 (2/28 vs 5/28) · weight 0.1% 0.35 [0.06, 2.00] Leaf 2005 (9/200 vs 9/202) Leaf 2005 (9/200 vs 9/202) · weight 0.4% 1.01 [0.39, 2.60] Brouwer 2006 (6/273 vs 13/273) Brouwer 2006 (6/273 vs 13/273) · weight 0.4% 0.45 [0.17, 1.20] GISSI-HF investigators 2008 (712/3494 GISSI-HF investigators 2008 (712/3494 vs 765/3481) · weight 26.3% 0.91 [0.81, 1.02] Kromhout 2010 (80/2404 vs 82/2433) Kromhout 2010 (80/2404 vs 82/2433) · weight 3.5% 0.99 [0.72, 1.35] Galan 2010 (41/1259 vs 38/1263) Galan 2010 (41/1259 vs 38/1263) · weight 1.7% 1.09 [0.69, 1.70] Nodari 2011 (0/13 vs 0/13) Nodari 2011 (0/13 vs 0/13) · weight 0.0% 1.00 [0.02, 54.16] Bosch 2012 (574/6281 vs 581/6255) Bosch 2012 (574/6281 vs 581/6255) · weight 23.7% 0.98 [0.87, 1.11] Mozaffarian 2012 (0/758 vs 3/758) Mozaffarian 2012 (0/758 vs 3/758) · weight 0.0% 0.14 [0.01, 2.76] Bosch 2013 (142/6239 vs 137/6266) Bosch 2013 (142/6239 vs 137/6266) · weight 6.2% 1.04 [0.82, 1.32] Kumar 2013 (1/38 vs 1/38) Kumar 2013 (1/38 vs 1/38) · weight 0.0% 1.00 [0.06, 16.59] Maki 2014 (0/49 vs 0/49) Maki 2014 (0/49 vs 0/49) · weight 0.0% 1.00 [0.02, 51.41] Vanderbilt 2015 (0/81 vs 0/81) Vanderbilt 2015 (0/81 vs 0/81) · weight 0.0% 1.00 [0.02, 51.01] ASCEND Study Collaborative Group 2018 ASCEND Study Collaborative Group 2018 (196/7740 vs 240/7740) · weight 9.5% 0.81 [0.67, 0.98] Foroughinia 2018 (0/37 vs 0/43) Foroughinia 2018 (0/37 vs 0/43) · weight 0.0% 1.16 [0.02, 59.90] Manson 2019 (142/12933 vs 148/12938) Manson 2019 (142/12933 vs 148/12938) · weight 6.5% 0.96 [0.76, 1.21] Yokote 2020 (0/152 vs 0/77) Yokote 2020 (0/152 vs 0/77) · weight 0.0% 0.51 [0.01, 25.86] Pooled OR (k=22) 0.90 [0.84, 0.95] · p<0.001 ← EPA+DHA control →
Tier 1 — women population
OR 0.74 [0.60, 0.91]; k=4; p = 0.004.
Study OR Mozaffarian 2012 (10/758 vs 10/758) Mozaffarian 2012 (10/758 vs 10/758) · weight 5.6% 1.00 [0.41, 2.42] Macchia 2013 (1/289 vs 1/297) Macchia 2013 (1/289 vs 1/297) · weight 0.6% 1.03 [0.06, 16.51] Maki 2014 (0/49 vs 0/49) Maki 2014 (0/49 vs 0/49) · weight 0.3% 1.00 [0.02, 51.41] Manson 2019 (145/12933 vs 200/12938) Manson 2019 (145/12933 vs 200/12938) · weight 93.6% 0.72 [0.58, 0.90] Pooled OR (k=4) 0.74 [0.60, 0.91] · p=0.004 ← EPA+DHA control →
Tier 2 — whole population (supportive)
OR 0.91 [0.85, 0.98]; k=23; p = 0.010.
Study OR Sacks 1995 (1/4 vs 2/4) Sacks 1995 (1/4 vs 2/4) · weight 0.1% 0.33 [0.02, 6.65] GISSI-Prevenzione Investigators 1999 ( GISSI-Prevenzione Investigators 1999 (424/5666 vs 485/5668) · weight 27.8% 0.86 [0.75, 0.99] Nilsen 2001 (21/150 vs 15/150) Nilsen 2001 (21/150 vs 15/150) · weight 1.0% 1.47 [0.72, 2.97] Marchioli 2001 (104/2835 vs 113/2828) Marchioli 2001 (104/2835 vs 113/2828) · weight 7.0% 0.91 [0.70, 1.20] Grundt 2004 (21/150 vs 34/150) Grundt 2004 (21/150 vs 34/150) · weight 1.4% 0.56 [0.31, 1.01] Leaf 2005 (1/28 vs 3/28) Leaf 2005 (1/28 vs 3/28) · weight 0.1% 0.31 [0.03, 3.16] Brouwer 2006 (1/273 vs 3/273) Brouwer 2006 (1/273 vs 3/273) · weight 0.1% 0.33 [0.03, 3.20] Calabresi 2006 (1/30 vs 0/27) Calabresi 2006 (1/30 vs 0/27) · weight 0.0% 2.80 [0.11, 71.59] GISSI-HF investigators 2008 (107/3494 GISSI-HF investigators 2008 (107/3494 vs 129/3481) · weight 7.6% 0.82 [0.63, 1.07] Galan 2010 (113/1259 vs 112/1263) Galan 2010 (113/1259 vs 112/1263) · weight 6.8% 1.01 [0.77, 1.33] Farquharson 2011 (0/52 vs 3/52) Farquharson 2011 (0/52 vs 3/52) · weight 0.1% 0.13 [0.01, 2.67] Bosch 2012 (344/6281 vs 316/6255) Bosch 2012 (344/6281 vs 316/6255) · weight 20.8% 1.09 [0.93, 1.27] Mozaffarian 2012 (10/758 vs 10/758) Mozaffarian 2012 (10/758 vs 10/758) · weight 0.7% 1.00 [0.41, 2.42] Sandesara 2012 (1/37 vs 1/37) Sandesara 2012 (1/37 vs 1/37) · weight 0.1% 1.00 [0.06, 16.61] Macchia 2013 (1/289 vs 1/297) Macchia 2013 (1/289 vs 1/297) · weight 0.1% 1.03 [0.06, 16.51] Wilbring 2014 (3/29 vs 2/29) Wilbring 2014 (3/29 vs 2/29) · weight 0.1% 1.56 [0.24, 10.09] Maki 2014 (0/49 vs 0/49) Maki 2014 (0/49 vs 0/49) · weight 0.0% 1.00 [0.02, 51.41] Feguri 2017 (1/19 vs 0/19) Feguri 2017 (1/19 vs 0/19) · weight 0.0% 3.16 [0.12, 82.64] ASCEND Study Collaborative Group 2018 ASCEND Study Collaborative Group 2018 (186/7740 vs 200/7740) · weight 12.6% 0.93 [0.76, 1.14] Foroughinia 2018 (2/37 vs 6/43) Foroughinia 2018 (2/37 vs 6/43) · weight 0.2% 0.35 [0.07, 1.86] Manson 2019 (145/12933 vs 200/12938) Manson 2019 (145/12933 vs 200/12938) · weight 11.1% 0.72 [0.58, 0.90] Kalstad 2021 (39/505 vs 35/509) Kalstad 2021 (39/505 vs 35/509) · weight 2.3% 1.13 [0.71, 1.82] Woo 2021 (0/38 vs 1/38) Woo 2021 (0/38 vs 1/38) · weight 0.0% 0.32 [0.01, 8.22] Pooled OR (k=23) 0.91 [0.85, 0.98] · p=0.010 ← EPA+DHA control →
Tier 1 — women population
OR 1.02 [0.82, 1.28]; k=5; p = 0.850.
Study OR Mozaffarian 2012 (4/758 vs 8/758) Mozaffarian 2012 (4/758 vs 8/758) · weight 3.5% 0.50 [0.15, 1.66] Macchia 2013 (3/289 vs 3/297) Macchia 2013 (3/289 vs 3/297) · weight 1.9% 1.03 [0.21, 5.14] Maki 2014 (0/49 vs 0/49) Maki 2014 (0/49 vs 0/49) · weight 0.3% 1.00 [0.02, 51.41] Nigam 2014 (1/153 vs 0/163) Nigam 2014 (1/153 vs 0/163) · weight 0.5% 3.22 [0.13, 79.56] Manson 2019 (148/12933 vs 142/12938) Manson 2019 (148/12933 vs 142/12938) · weight 93.8% 1.04 [0.83, 1.31] Pooled OR (k=5) 1.02 [0.82, 1.28] · p=0.850 ← EPA+DHA control →
Tier 2 — whole population (supportive)
OR 1.10 [1.01, 1.20]; k=17; p = 0.022.
Study OR Sacks 1995 (1/4 vs 0/4) Sacks 1995 (1/4 vs 0/4) · weight 0.1% 3.86 [0.12, 126.74] GISSI-Prevenzione Investigators 1999 ( GISSI-Prevenzione Investigators 1999 (98/5666 vs 80/5668) · weight 7.8% 1.23 [0.91, 1.66] Stone 2000 (178/5666 vs 98/5658) Stone 2000 (178/5666 vs 98/5658) · weight 11.2% 1.84 [1.43, 2.36] Marchioli 2001 (50/2835 vs 40/2828) Marchioli 2001 (50/2835 vs 40/2828) · weight 3.9% 1.25 [0.82, 1.90] GISSI-HF investigators 2008 (122/3494 GISSI-HF investigators 2008 (122/3494 vs 103/3481) · weight 9.8% 1.19 [0.91, 1.55] Galan 2010 (59/1259 vs 60/1263) Galan 2010 (59/1259 vs 60/1263) · weight 5.1% 0.99 [0.68, 1.42] Farquharson 2011 (1/52 vs 2/52) Farquharson 2011 (1/52 vs 2/52) · weight 0.1% 0.49 [0.04, 5.58] Bosch 2012 (314/6281 vs 336/6255) Bosch 2012 (314/6281 vs 336/6255) · weight 27.8% 0.93 [0.79, 1.09] Mozaffarian 2012 (4/758 vs 8/758) Mozaffarian 2012 (4/758 vs 8/758) · weight 0.5% 0.50 [0.15, 1.66] Sandesara 2012 (3/37 vs 3/37) Sandesara 2012 (3/37 vs 3/37) · weight 0.2% 1.00 [0.19, 5.31] Macchia 2013 (3/289 vs 3/297) Macchia 2013 (3/289 vs 3/297) · weight 0.3% 1.03 [0.21, 5.14] Maki 2014 (0/49 vs 0/49) Maki 2014 (0/49 vs 0/49) · weight 0.0% 1.00 [0.02, 51.41] Nigam 2014 (1/153 vs 0/163) Nigam 2014 (1/153 vs 0/163) · weight 0.1% 3.22 [0.13, 79.56] Feguri 2017 (0/19 vs 1/19) Feguri 2017 (0/19 vs 1/19) · weight 0.1% 0.32 [0.01, 8.26] ASCEND Study Collaborative Group 2018 ASCEND Study Collaborative Group 2018 (217/7740 vs 214/7740) · weight 18.9% 1.01 [0.84, 1.23] Manson 2019 (148/12933 vs 142/12938) Manson 2019 (148/12933 vs 142/12938) · weight 12.9% 1.04 [0.83, 1.31] Kalstad 2021 (17/505 vs 12/509) Kalstad 2021 (17/505 vs 12/509) · weight 1.2% 1.44 [0.68, 3.05] Pooled OR (k=17) 1.10 [1.01, 1.20] · p=0.022 ← EPA+DHA control →
Tier 1 — women population
OR 0.95 [0.81, 1.11]; k=17; p = 0.503.
Study OR Goodfellow 2000 (0/13 vs 0/15) Goodfellow 2000 (0/13 vs 0/15) · weight 0.1% 1.15 [0.02, 61.90] Heidt 2009 (9/32 vs 15/32) Heidt 2009 (9/32 vs 15/32) · weight 2.2% 0.44 [0.16, 1.25] Saravanan 2010 (29/52 vs 22/51) Saravanan 2010 (29/52 vs 22/51) · weight 3.9% 1.66 [0.76, 3.62] Phelan 2011 (0/22 vs 0/22) Phelan 2011 (0/22 vs 0/22) · weight 0.2% 1.00 [0.02, 52.63] Kumar 2011 (6/18 vs 13/18) Kumar 2011 (6/18 vs 13/18) · weight 1.2% 0.19 [0.05, 0.80] Nodari 2011 (37/66 vs 56/66) Nodari 2011 (37/66 vs 56/66) · weight 3.4% 0.23 [0.10, 0.52] Sorice 2011 (11/37 vs 24/37) Sorice 2011 (11/37 vs 24/37) · weight 2.5% 0.23 [0.09, 0.61] Mozaffarian 2012 (227/758 vs 233/758) Mozaffarian 2012 (227/758 vs 233/758) · weight 49.1% 0.96 [0.77, 1.20] Macchia 2013 (69/289 vs 56/297) Macchia 2013 (69/289 vs 56/297) · weight 15.0% 1.35 [0.91, 2.01] Lomivorotov 2014 (6/17 vs 5/18) Lomivorotov 2014 (6/17 vs 5/18) · weight 1.1% 1.42 [0.34, 5.94] Nigam 2014 (98/153 vs 103/163) Nigam 2014 (98/153 vs 103/163) · weight 11.2% 1.04 [0.66, 1.64] Vanderbilt 2015 (74/81 vs 30/81) Vanderbilt 2015 (74/81 vs 30/81) · weight 2.9% 17.97 [7.33, 44.06] Kristensen 2016 (0/73 vs 1/72) Kristensen 2016 (0/73 vs 1/72) · weight 0.2% 0.32 [0.01, 8.09] Ip 2017 (3/5 vs 5/5) Ip 2017 (3/5 vs 5/5) · weight 0.2% 0.13 [0.00, 3.52] Vasheghani Farahani 2017 (17/142 vs 29 Vasheghani Farahani 2017 (17/142 vs 29/142) · weight 5.6% 0.53 [0.28, 1.02] Javid 2018 (0/27 vs 0/27) Javid 2018 (0/27 vs 0/27) · weight 0.2% 1.00 [0.02, 52.22] Feguri 2019 (3/21 vs 14/21) Feguri 2019 (3/21 vs 14/21) · weight 1.0% 0.08 [0.02, 0.38] Pooled OR (k=17) 0.95 [0.81, 1.11] · p=0.503 ← EPA+DHA control →
Tier 2 — whole population (supportive)
OR 0.94 [0.81, 1.09]; k=20; p = 0.408.
Study OR Goodfellow 2000 (0/13 vs 0/15) Goodfellow 2000 (0/13 vs 0/15) · weight 0.1% 1.15 [0.02, 61.90] Heidt 2009 (9/32 vs 15/32) Heidt 2009 (9/32 vs 15/32) · weight 2.0% 0.44 [0.16, 1.25] Saravanan 2010 (29/52 vs 22/51) Saravanan 2010 (29/52 vs 22/51) · weight 3.6% 1.66 [0.76, 3.62] Phelan 2011 (0/22 vs 0/22) Phelan 2011 (0/22 vs 0/22) · weight 0.1% 1.00 [0.02, 52.63] Kumar 2011 (6/18 vs 13/18) Kumar 2011 (6/18 vs 13/18) · weight 1.1% 0.19 [0.05, 0.80] Farquharson 2011 (36/52 vs 47/52) Farquharson 2011 (36/52 vs 47/52) · weight 1.8% 0.24 [0.08, 0.71] Nodari 2011 (37/66 vs 56/66) Nodari 2011 (37/66 vs 56/66) · weight 3.1% 0.23 [0.10, 0.52] Sorice 2011 (11/37 vs 24/37) Sorice 2011 (11/37 vs 24/37) · weight 2.3% 0.23 [0.09, 0.61] Mozaffarian 2012 (227/758 vs 233/758) Mozaffarian 2012 (227/758 vs 233/758) · weight 45.2% 0.96 [0.77, 1.20] Macchia 2013 (69/289 vs 56/297) Macchia 2013 (69/289 vs 56/297) · weight 13.8% 1.35 [0.91, 2.01] Lomivorotov 2014 (6/17 vs 5/18) Lomivorotov 2014 (6/17 vs 5/18) · weight 1.1% 1.42 [0.34, 5.94] Nigam 2014 (98/153 vs 103/163) Nigam 2014 (98/153 vs 103/163) · weight 10.3% 1.04 [0.66, 1.64] Vanderbilt 2015 (74/81 vs 30/81) Vanderbilt 2015 (74/81 vs 30/81) · weight 2.7% 17.97 [7.33, 44.06] Kristensen 2016 (0/73 vs 1/72) Kristensen 2016 (0/73 vs 1/72) · weight 0.2% 0.32 [0.01, 8.09] Ip 2017 (3/5 vs 5/5) Ip 2017 (3/5 vs 5/5) · weight 0.2% 0.13 [0.00, 3.52] Vasheghani Farahani 2017 (17/142 vs 29 Vasheghani Farahani 2017 (17/142 vs 29/142) · weight 5.1% 0.53 [0.28, 1.02] Feguri 2017 (3/19 vs 14/19) Feguri 2017 (3/19 vs 14/19) · weight 0.8% 0.07 [0.01, 0.33] Javid 2018 (0/27 vs 0/27) Javid 2018 (0/27 vs 0/27) · weight 0.1% 1.00 [0.02, 52.22] Feguri 2019 (3/21 vs 14/21) Feguri 2019 (3/21 vs 14/21) · weight 0.9% 0.08 [0.02, 0.38] Kalstad 2021 (28/505 vs 15/509) Kalstad 2021 (28/505 vs 15/509) · weight 5.3% 1.93 [1.02, 3.66] Pooled OR (k=20) 0.94 [0.81, 1.09] · p=0.408 ← EPA+DHA control →
Direct women data

Cardiovascular risk in women — pooling published HRs

Because event counts in women are not published for the large landmark trials, a direct women estimate can only be obtained by pooling published women-subgroup hazard ratios (HR) using generic inverse-variance. For the EPA+DHA family this yields a borderline benefit signal on the primary composite outcome.

EPA+DHA, major CVD composite in women: HR 0.87 [0.75, 1.00]; k=2; p = 0.055.
Study HR VITAL VITAL · weight 47.0% 0.93 [0.76, 1.15] Risk & Prevention Risk & Prevention · weight 53.0% 0.82 [0.67, 0.99] Pooled HR (k=2) 0.87 [0.75, 1.00] · p=0.055 ← EPA+DHA control →
Network

The analyzed comparison network

Shown below is exactly the network used in the network meta-analysis: omega-3 formulations are compared with one another through a common control node (placebo/standard care/diet), because there are no head-to-head trials between formulations in the evidence base. This is a star topology. Node size and edge width are proportional to the number of studies; edge labels give the number of studies for each comparison.

Network for triglycerides (4 formulations)
16 2 1 Control EPA+DHA k=16 EPA-mono k=2 DHA-mono k=1 Node size and edge width ∝ number of studies (k). All comparisons anchored to a common control (star).

For cardiovascular outcomes the network has three nodes (control / EPA+DHA / EPA-mono) — there are no DHA-mono trials with cardiovascular events. The per-outcome networks with their k counts are shown below, in the ranking section.

Network meta-analysis

Indirect comparisons and treatment ranking (P-score / SUCRA)

A network meta-analysis (frequentist random-effects model, netmeta) compares omega-3 formulations with one another even without head-to-head trials, via a common comparator node. Treatment ranking is expressed as the P-score — the frequentist analogue of SUCRA (surface under the cumulative ranking curve): a value closer to 100% means the treatment is more likely to be among the best for that outcome.

How to read this. The network has a star-shaped topology — there are no head-to-head EPA+DHA vs EPA-mono/DHA-mono trials, so cross-formulation comparisons are entirely indirect and inconsistency cannot be assessed; the result rests on the transitivity assumption. The cardiovascular networks are built on the whole study population (predominantly male) — a supportive, indirect tier, not a direct women estimate. DHA-mono is absent for cardiovascular outcomes.
k = 36 studies · 3 nodes · I² = 6%
21 15 Control EPA+DHA k=21 EPA-mono k=15 Node size and edge width ∝ number of studies (k). All comparisons anchored to a common control (star).

Ranking (P-score)

EPA-mono
100.0%
EPA+DHA
49.8%
Control
0.2%
A higher P-score is more favourable for this outcome (fewer events / lower triglycerides).

Network estimates (including indirect)

The diamond is the network estimate of a comparison with its 95% confidence interval; the dashed vertical line is no difference. Cross-formulation (indirect) comparisons are highlighted on a purple background.

Comparison OR [95% CI] EPA+DHA vs Control 0.95 [0.91, 0.99] · p=0.010 EPA-mono vs Control 0.71 [0.68, 0.74] · p<0.001 EPA+DHA vs EPA-mono 1.33 [1.26, 1.42] · p<0.001 ← favours first favours second →
ComparisonOR95% CIp
EPA+DHA vs Control0.95[0.91, 0.99]0.010
EPA-mono vs Control0.71[0.68, 0.74]< 0.001
EPA+DHA vs EPA-mono1.33[1.26, 1.42]< 0.001
k = 56 studies · 3 nodes · I² = 0%
43 13 Control EPA+DHA k=43 EPA-mono k=13 Node size and edge width ∝ number of studies (k). All comparisons anchored to a common control (star).

Ranking (P-score)

EPA-mono
96.0%
EPA+DHA
53.8%
Control
0.2%
A higher P-score is more favourable for this outcome (fewer events / lower triglycerides).

Network estimates (including indirect)

The diamond is the network estimate of a comparison with its 95% confidence interval; the dashed vertical line is no difference. Cross-formulation (indirect) comparisons are highlighted on a purple background.

Comparison OR [95% CI] EPA+DHA vs Control 0.94 [0.90, 0.98] · p=0.008 EPA-mono vs Control 0.89 [0.82, 0.96] · p=0.002 EPA+DHA vs EPA-mono 1.07 [0.98, 1.16] · p=0.159 ← favours first favours second →
ComparisonOR95% CIp
EPA+DHA vs Control0.94[0.90, 0.98]0.008
EPA-mono vs Control0.89[0.82, 0.96]0.002
EPA+DHA vs EPA-mono1.07[0.98, 1.16]0.159
k = 28 studies · 3 nodes · I² = 78%
20 8 Control EPA+DHA k=20 EPA-mono k=8 Node size and edge width ∝ number of studies (k). All comparisons anchored to a common control (star).

Ranking (P-score)

EPA+DHA
97.5%
Control
48.7%
EPA-mono
3.8%
A higher P-score is more favourable for this outcome (fewer events / lower triglycerides).

Network estimates (including indirect)

The diamond is the network estimate of a comparison with its 95% confidence interval; the dashed vertical line is no difference. Cross-formulation (indirect) comparisons are highlighted on a purple background.

Comparison OR [95% CI] EPA+DHA vs Control 0.73 [0.52, 1.03] · p=0.076 EPA-mono vs Control 1.42 [0.90, 2.23] · p=0.128 EPA+DHA vs EPA-mono 0.51 [0.29, 0.91] · p=0.022 ← favours first favours second →
ComparisonOR95% CIp
EPA+DHA vs Control0.73[0.52, 1.03]0.076
EPA-mono vs Control1.42[0.90, 2.23]0.128
EPA+DHA vs EPA-mono0.51[0.29, 0.91]0.022
k = 19 studies · 4 nodes · I² = 66% · exploratory, whole population, low certainty (mono arms k=1–2)
16 2 1 Control EPA+DHA k=16 EPA-mono k=2 DHA-mono k=1 Node size and edge width ∝ number of studies (k). All comparisons anchored to a common control (star).

Ranking (P-score)

DHA-mono
75.4%
EPA+DHA
72.8%
EPA-mono
39.2%
Control
12.6%
A higher P-score is more favourable for this outcome (fewer events / lower triglycerides).

Network estimates (including indirect)

The diamond is the network estimate of a comparison with its 95% confidence interval; the dashed vertical line is no difference. Cross-formulation (indirect) comparisons are highlighted on a purple background.

Comparison MD [95% CI] EPA+DHA vs Control −17.8 [−26.0, −9.7] · p<0.001 EPA-mono vs Control −7.5 [−28.7, 13.7] · p=0.490 DHA-mono vs Control −25.7 [−70.8, 19.4] · p=0.265 EPA+DHA vs EPA-mono −10.4 [−33.1, 12.4] · p=0.371 EPA+DHA vs DHA-mono 7.9 [−38.0, 53.7] · p=0.737 EPA-mono vs DHA-mono 18.2 [−31.6, 68.1] · p=0.474 ← lower in first lower in second →
ComparisonMD95% CIp
EPA+DHA vs Control−17.8[−26.0, −9.7]< 0.001
EPA-mono vs Control−7.5[−28.7, 13.7]0.490
DHA-mono vs Control−25.7[−70.8, 19.4]0.265
EPA+DHA vs EPA-mono−10.4[−33.1, 12.4]0.371
EPA+DHA vs DHA-mono7.9[−38.0, 53.7]0.737
EPA-mono vs DHA-mono18.2[−31.6, 68.1]0.474
Interpretation

Discussion

In a postmenopausal women population, EPA+DHA consistently lower triglycerides — a pooled mean difference of about 8.7 mg/dL — with a neutral effect on LDL-C and HDL-C. This is consistent with the classic lipid profile of long-chain omega-3 (marked TG reduction, minimal LDL effect) and confirms its reproducibility specifically in the female postmenopausal population, not only in mixed samples.

For cardiovascular events, direct women data are limited. At the supportive whole-population tier, EPA+DHA are associated with a modest reduction in all-cause and cardiovascular mortality and in myocardial infarction, but a signal of increased stroke and atrial fibrillation is also seen. These estimates come mainly from male-predominant trials and cannot be transferred directly to women; they are presented as context, not as a primary result.

Network ranking. A network meta-analysis with P-score ranking (a SUCRA analogue) revealed an outcome-dependent picture. For MACE and all-cause mortality, EPA-mono ranks best, but this result is driven entirely by high-dose trials in high-risk groups (chiefly REDUCE-IT and JELIS) and is burdened by their known concerns (mineral-oil placebo, a particular population) — it reflects the specific context of those trials rather than the superiority of the formulation as such. Conversely, for atrial fibrillation EPA+DHA ranks as the most favourable formulation and EPA-mono as the least (EPA+DHA vs EPA-mono — a significant reduction in odds), consistent with the accumulated evidence of increased arrhythmia risk on high-dose EPA-mono. This is a clinically important difference in the safety profiles of the formulations.

Completeness of the evidence base. The set of included studies was cross-checked against prior omega-3 reviews and meta-analyses: Mora 2024 · AbuMweis 2018 · Hartweg 2009 · Balk 2016 · Hu 2019 · Bernasconi 2021. Landmark cardiovascular trials (VITAL, ASCEND, GISSI, ORIGIN, Risk & Prevention) were identified and accounted for.
Caveats

Limitations

  • Scarcity of women-only cardiovascular data. Hard cardiovascular outcomes were measured mainly in large, male-predominant trials; the whole-population tier is only indirect support for women.
  • Dependence of the TG estimate on individual large trials. The large Gaengler 2024 trial contributes substantially; robustness holds when the population is broadened, but the weight of individual studies is high.
  • Heterogeneity of doses and formulations of omega-3 (EPA+DHA, EPA-mono, DHA-mono; from <1 g to >2 g/day) and of lipid units, harmonised to a common scale.
  • Reconstruction of change standard errors from reported SE / 95% CI for some studies per Cochrane methods.
  • Stroke and atrial fibrillation signals at the whole-population tier require cautious interpretation and do not apply directly to the women population.
Bottom line

Conclusions

In peri- and postmenopausal women, EPA+DHA produce a significant reduction in triglycerides (pooled mean difference of about 8.7 mg/dL; k=8) with a neutral effect on LDL-C and HDL-C — a robust quantitative result in dedicated postmenopausal cohorts. Evidence on hard cardiovascular outcomes in women is limited: at the supportive whole-population tier a modest benefit on mortality and myocardial infarction is seen, alongside a cautionary signal for stroke and atrial fibrillation. EPA+DHA are justified for correcting hypertriglyceridaemia in this population; extrapolating a cardiovascular benefit to women requires further data.

Sources

References

Studies included in the quantitative analyses (lipid, cardiovascular and network) — 106 in total:

  1. ? 2024 (2024). Cost-Effectiveness of Icosapent Ethyl in REDUCE-IT USA: Results From Patients Randomized in the United States. J Am Heart Assoc. PubMed
  2. ASCEND Study Collaborative Group 2018 (2018). Effects of n-3 Fatty Acid Supplements in Diabetes Mellitus. N Engl J Med. PubMed
  3. Asztalos 2016 (2016). Effects of eicosapentaenoic acid and docosahexaenoic acid on cardiovascular disease risk factors: a randomized clinical trial. Metabolism. PubMed
  4. Ballantyne 2012 (2012). Efficacy and safety of eicosapentaenoic acid ethyl ester (AMR101) therapy in statin-treated patients with persistent high triglycerides (from the ANCHOR study). Am J Cardiol. PubMed
  5. Bays 2011 (2011). Eicosapentaenoic acid ethyl ester (AMR101) therapy in patients with very high triglyceride levels (from the Multi-center, plAcebo-controlled, Randomized, double-blINd, 12-week study with an open-label Extension [MARINE] trial). Am J Cardiol. PubMed
  6. Bernhard 2024 (2024). Effect of six month's treatment with omega-3 acid ethyl esters on long-term outcomes after acute myocardial infarction: The OMEGA-REMODEL randomized clinical trial. Int J Cardiol. PubMed
  7. Bhatt 2019 (2019). Effects of Icosapent Ethyl on Total Ischemic Events: From REDUCE-IT. J Am Coll Cardiol. PubMed
  8. Bhatt 2020 (2020). REDUCE-IT USA: Results From the 3146 Patients Randomized in the United States. Circulation. PubMed
  9. Bhatt 2022 (2022). Treatment With Icosapent Ethyl to Reduce Ischemic Events in Patients With Prior Percutaneous Coronary Intervention: Insights From REDUCE-IT PCI. J Am Heart Assoc. PubMed
  10. Bhatt 2022 (2022). Prevention of Cardiovascular Events and Mortality With Icosapent Ethyl in Patients With Prior Myocardial Infarction. J Am Coll Cardiol. PubMed
  11. Bhatt 2025 (2025). Cardiovascular Outcomes With Icosapent Ethyl by Baseline Low-Density Lipoprotein Cholesterol: A Secondary Analysis of the REDUCE-IT Randomized Trial. J Am Heart Assoc. PubMed
  12. Bosch 2012 (2012). n-3 fatty acids and cardiovascular outcomes in patients with dysglycemia. N Engl J Med. PubMed
  13. Bosch 2013 (2013). n-3 fatty acids in patients with multiple cardiovascular risk factors. N Engl J Med. PubMed
  14. Brouwer 2006 (2006). Effect of fish oil on ventricular tachyarrhythmia and death in patients with implantable cardioverter defibrillators: the Study on Omega-3 Fatty Acids and Ventricular Arrhythmia (SOFA) randomized trial. JAMA. PubMed
  15. Burger 2024 (2024). Effects of icosapent ethyl according to baseline residual risk in patients with atherosclerotic cardiovascular disease: results from REDUCE-IT. Eur Heart J Cardiovasc Pharmacother. PubMed
  16. Burr 2003 (2003). Lack of benefit of dietary advice to men with angina: results of a controlled trial. Eur J Clin Nutr. PubMed
  17. Calabresi 2006 (2006). Effect of n-3 fatty acids on carotid atherosclerosis and haemostasis in patients with combined hyperlipoproteinemia: a double-blind pilot study in primary prevention. Ann Med. PubMed
  18. Calò 2005 (2005). N-3 Fatty acids for the prevention of atrial fibrillation after coronary artery bypass surgery: a randomized, controlled trial. J Am Coll Cardiol. PubMed
  19. Ciubotaru 2003 (2003). Dietary fish oil decreases C-reactive protein, interleukin-6, and triacylglycerol to HDL-cholesterol ratio in postmenopausal women on HRT. J Nutr Biochem. PubMed
  20. Crochemore 2012 (2012). ω-3 polyunsaturated fatty acid supplementation does not influence body composition, insulin resistance, and lipemia in women with type 2 diabetes and obesity. Nutr Clin Pract. PubMed
  21. de Borst 2017 (2017). Effect of Omega-3 Fatty Acid Supplementation on Plasma Fibroblast Growth Factor 23 Levels in Post-Myocardial Infarction Patients with Chronic Kidney Disease: The Alpha Omega Trial. Nutrients. PubMed
  22. Doenyas-Barak. N-3 fatty acid supplementation to routine statin treatment inhibits platelet function, decreases patients' daytime blood pressure, and improves inflammatory status. PubMed
  23. Eritsland 1996 (1996). Effect of dietary supplementation with n-3 fatty acids on coronary artery bypass graft patency. Am J Cardiol. PubMed
  24. Farquharson 2011 (2011). Effect of dietary fish oil on atrial fibrillation after cardiac surgery. Am J Cardiol. PubMed
  25. Feguri 2017 (2017). Preoperative carbohydrate load and intraoperatively infused omega-3 polyunsaturated fatty acids positively impact nosocomial morbidity after coronary artery bypass grafting: a double-blind controlled randomized trial. Nutr J. PubMed
  26. Feguri 2019 (2019). Benefits of Fasting Abbreviation with Carbohydrates and Omega-3 Infusion During CABG: a Double-Blind Controlled Randomized Trial. Braz J Cardiovasc Surg. PubMed
  27. Foroughinia 2018 (2018). Impact of Omega-3 Supplementation on High Sensitive C-Reactive Protein Level and 30-Day Major Adverse Cardiac Events After the Implementation of Coronary Stent in Patients with Chronic Kidney Disease: A Randomized Clinical Study. Adv Pharm Bull. PubMed
  28. Friesecke 2008 (2008). Fish oil supplementation in the parenteral nutrition of critically ill medical patients: a randomised controlled trial. Intensive Care Med. PubMed
  29. Félix-Soriano 2021 (2021). Effects of DHA-Rich n-3 Fatty Acid Supplementation and/or Resistance Training on Body Composition and Cardiometabolic Biomarkers in Overweight and Obese Post-Menopausal Women. Nutrients. PubMed
  30. Gaengler 2024 (2024). Effects of vitamin D, omega-3 and a simple strength exercise programme in cardiovascular disease prevention: The DO-HEALTH randomized controlled trial. J Nutr Health Aging. PubMed
  31. Galan 2010 (2010). Effects of B vitamins and omega 3 fatty acids on cardiovascular diseases: a randomised placebo controlled trial. BMJ. PubMed
  32. GISSI-HF investigators 2008 (2008). Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial. Lancet. PubMed
  33. GISSI-Prevenzione Investigators 1999 (1999). Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione trial. Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico. Lancet. PubMed
  34. Goodfellow 2000 (2000). Dietary supplementation with marine omega-3 fatty acids improve systemic large artery endothelial function in subjects with hypercholesterolemia. J Am Coll Cardiol. PubMed
  35. Grundt 2004 (2004). Clinical outcome and atherothrombogenic risk profile after prolonged wash-out following long-term treatment with high doses of n-3 PUFAs in patients with an acute myocardial infarction. Clin Nutr. PubMed
  36. Gurevitz 2025 (2025). Benefit of Icosapent Ethyl Across Types and Sizes of Myocardial Infarction in REDUCE-IT. Eur J Prev Cardiol. PubMed
  37. Hamaya 2025 (2025). Identification of individuals who benefit from omega-3 fatty acid supplementation to prevent coronary heart disease: a machine-learning analysis of the VITAL. Eur J Epidemiol. PubMed
  38. Han 2012 (2012). Effects of fish oil on inflammatory modulation in surgical intensive care unit patients. Nutr Clin Pract. PubMed
  39. Heidt 2009 (2009). Beneficial effects of intravenously administered N-3 fatty acids for the prevention of atrial fibrillation after coronary artery bypass surgery: a prospective randomized study. Thorac Cardiovasc Surg. PubMed
  40. Holman 2009 (2009). Atorvastatin in Factorial with Omega-3 EE90 Risk Reduction in Diabetes (AFORRD): a randomised controlled trial. Diabetologia. PubMed
  41. Hoogeveen 2014 (2014). No effect of n-3 fatty acids on high-sensitivity C-reactive protein after myocardial infarction: the Alpha Omega Trial. Eur J Prev Cardiol. PubMed
  42. Hoogeveen 2015 (2015). No effect of n-3 fatty acids supplementation on NT-proBNP after myocardial infarction: the Alpha Omega Trial. Eur J Prev Cardiol. PubMed
  43. Ip 2017 (2017). A Small Cohort Omega-3 PUFA Supplement Study: Implications of Stratifying According to Lipid Membrane Incorporation in Cardiac Surgical Patients. Heart Lung Circ. PubMed
  44. Ishikawa 2010 (2010). Preventive effects of eicosapentaenoic acid on coronary artery disease in patients with peripheral artery disease. Circ J. PubMed
  45. Isley. Pilot study of combined therapy with ω-3 fatty acids and niacin in atherogenic dyslipidemia. PubMed
  46. Javid 2018 (2018). Impact of Cranberry Juice Enriched with Omega-3 Fatty Acids Adjunct with Nonsurgical Periodontal Treatment on Metabolic Control and Periodontal Status in Type 2 Patients with Diabetes with Periodontal Disease. J Am Coll Nutr. PubMed
  47. Kabir 2007 (2007). Treatment for 2 mo with n 3 polyunsaturated fatty acids reduces adiposity and some atherogenic factors but does not improve insulin sensitivity in women with type 2 diabetes: a randomized controlled study. Am J Clin Nutr. PubMed
  48. Kalstad 2021 (2021). Effects of n-3 Fatty Acid Supplements in Elderly Patients After Myocardial Infarction: A Randomized, Controlled Trial. Circulation. PubMed
  49. Kita 2020 (2020). Effects of Fatty Acid Therapy in Addition to Strong Statin on Coronary Plaques in Acute Coronary Syndrome: An Optical Coherence Tomography Study. J Am Heart Assoc. PubMed
  50. Kristensen 2016 (2016). The effect of marine n-3 polyunsaturated fatty acids on cardiac autonomic and hemodynamic function in patients with psoriatic arthritis: a randomised, double-blind, placebo-controlled trial. Lipids Health Dis. PubMed
  51. Kromhout 2010 (2010). n-3 fatty acids and cardiovascular events after myocardial infarction. N Engl J Med. PubMed
  52. Kumar 2011 (2011). Effects of chronic omega-3 polyunsaturated fatty acid supplementation on human pulmonary vein and left atrial electrophysiology in paroxysmal atrial fibrillation. Am J Cardiol. PubMed
  53. Kumar 2013 (2013). Effects of long-term ω-3 polyunsaturated fatty acid supplementation on paroxysmal atrial tachyarrhythmia burden in patients with implanted pacemakers: results from a prospective randomised study. Int J Cardiol. PubMed
  54. Leaf 2005 (2005). Prevention of fatal arrhythmias in high-risk subjects by fish oil n-3 fatty acid intake. Circulation. PubMed
  55. Leaf 2005 (2005). Fish oil supplementation and risk of ventricular tachycardia and ventricular fibrillation in patients with implantable defibrillators: a randomized controlled trial. JAMA. PubMed
  56. Lee 2019 (2019). Docosahexaenoic acid reduces resting blood pressure but increases muscle sympathetic outflow compared with eicosapentaenoic acid in healthy men and women. Am J Physiol Heart Circ Physiol. PubMed
  57. Lee 2023 (2023). Fish Oil Supplementation with Resistance Exercise Training Enhances Physical Function and Cardiometabolic Health in Postmenopausal Women. Nutrients. PubMed
  58. Lok 2012 (2012). Effect of fish oil supplementation on graft patency and cardiovascular events among patients with new synthetic arteriovenous hemodialysis grafts: a randomized controlled trial. JAMA. PubMed
  59. Lomivorotov 2014 (2014). Randomized trial of fish oil infusion to prevent atrial fibrillation after cardiac surgery: data from an implantable continuous cardiac monitor. J Cardiothorac Vasc Anesth. PubMed
  60. Macchia 2013 (2013). Omega-3 fatty acids for the prevention of recurrent symptomatic atrial fibrillation: results of the FORWARD (Randomized Trial to Assess Efficacy of PUFA for the Maintenance of Sinus Rhythm in Persistent Atrial Fibrillation) trial. J Am Coll Cardiol. PubMed
  61. Maki 2014 (2014). A new, microalgal DHA- and EPA-containing oil lowers triacylglycerols in adults with mild-to-moderate hypertriglyceridemia. Prostaglandins Leukot Essent Fatty Acids. PubMed
  62. Manson 2019 (2019). Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer. N Engl J Med. PubMed
  63. Marchioli 2001 (2001). Efficacy of n-3 polyunsaturated fatty acids after myocardial infarction: results of GISSI-Prevenzione trial. Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico. Lipids. PubMed
  64. Matsuzaki 2009 (2009). Incremental effects of eicosapentaenoic acid on cardiovascular events in statin-treated patients with coronary artery disease. Circ J. PubMed
  65. Miller 2023 (2023). Effectiveness of icosapent ethyl on first and total cardiovascular events in patients with metabolic syndrome, but without diabetes: REDUCE-IT MetSyn. Eur Heart J Open. PubMed
  66. Miller 2026 (2026). Icosapent ethyl reduces CVD risk in cardiovascular-kidney-metabolic syndrome: REDUCE-IT CKM. Am J Prev Cardiol. PubMed
  67. Mozaffarian 2012 (2012). Fish oil and postoperative atrial fibrillation: the Omega-3 Fatty Acids for Prevention of Post-operative Atrial Fibrillation (OPERA) randomized trial. JAMA. PubMed
  68. Murphy 2021 (2021). Does supplementation with leucine-enriched protein alone and in combination with fish-oil-derived n-3 PUFA affect muscle mass, strength, physical performance, and muscle protein synthesis in well-nourished older adults? A randomized, double-blind, placebo-controlled trial. Am J Clin Nutr. PubMed
  69. Myhre 2022 (2022). Changes in eicosapentaenoic acid and docosahexaenoic acid and risk of cardiovascular events and atrial fibrillation: A secondary analysis of the OMEMI trial. J Intern Med. PubMed
  70. Nelson 2011 (2011). Omega-3 fatty acids and lipoprotein associated phospholipase A(2) in healthy older adult males and females. Eur J Nutr. PubMed
  71. Ness 1999 (1999). The long-term effect of advice to eat more fish on blood pressure in men with coronary disease: results from the diet and reinfarction trial. J Hum Hypertens. PubMed
  72. Nigam 2014 (2014). Fish oil for the reduction of atrial fibrillation recurrence, inflammation, and oxidative stress. J Am Coll Cardiol. PubMed
  73. Nilsen 2001 (2001). Effects of a high-dose concentrate of n-3 fatty acids or corn oil introduced early after an acute myocardial infarction on serum triacylglycerol and HDL cholesterol. Am J Clin Nutr. PubMed
  74. Nodari 2011 (2011). n-3 polyunsaturated fatty acids in the prevention of atrial fibrillation recurrences after electrical cardioversion: a prospective, randomized study. Circulation. PubMed
  75. Nodari 2011 (2011). Effects of n-3 polyunsaturated fatty acids on left ventricular function and functional capacity in patients with dilated cardiomyopathy. J Am Coll Cardiol. PubMed
  76. Omrani 2015 (2015). The effect of omega-3 on serum lipid profile in hemodialysis patients. J Renal Inj Prev. PubMed
  77. Origasa 2010 (2010). Clinical importance of adherence to treatment with eicosapentaenoic acid by patients with hypercholesterolemia. Circ J. PubMed
  78. Phelan 2011 (2011). Hormonal and metabolic effects of polyunsaturated fatty acids in young women with polycystic ovary syndrome: results from a cross-sectional analysis and a randomized, placebo-controlled, crossover trial. Am J Clin Nutr. PubMed
  79. Rauch 2010 (2010). OMEGA, a randomized, placebo-controlled trial to test the effect of highly purified omega-3 fatty acids on top of modern guideline-adjusted therapy after myocardial infarction. Circulation. PubMed
  80. REDUCE-IT (2019). Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia. N Engl J Med. PubMed
  81. Rizza. Fish oil supplementation improves endothelial function in normoglycemic offspring of patients with type 2 diabetes. PubMed
  82. Sacks 1995 (1995). Controlled trial of fish oil for regression of human coronary atherosclerosis. HARP Research Group. J Am Coll Cardiol. PubMed
  83. Sandesara 2012 (2012). A Randomized, Placebo-Controlled Trial of Omega-3 Fatty Acids for Inhibition of Supraventricular Arrhythmias After Cardiac Surgery: The FISH Trial. J Am Heart Assoc. PubMed
  84. Saravanan 2010 (2010). Omega-3 fatty acid supplementation does not reduce risk of atrial fibrillation after coronary artery bypass surgery: a randomized, double-blind, placebo-controlled clinical trial. Circ Arrhythm Electrophysiol. PubMed
  85. Sasaki 2012 (2012). Relationship between coronary artery disease and non-HDL-C, and effect of highly purified EPA on the risk of coronary artery disease in hypercholesterolemic patients treated with statins: sub-analysis of the Japan EPA Lipid Intervention Study (JELIS). J Atheroscler Thromb. PubMed
  86. Sayah 2024 (2024). Icosapent ethyl following acute coronary syndrome: the REDUCE-IT trial. Eur Heart J. PubMed
  87. Shinto 2014 (2014). A randomized placebo-controlled pilot trial of omega-3 fatty acids and alpha lipoic acid in Alzheimer's disease. J Alzheimers Dis. PubMed
  88. Skulas-Ray 2015 (2015). Dose-response effects of marine omega-3 fatty acids on apolipoproteins, apolipoprotein-defined lipoprotein subclasses, and Lp-PLA2 in individuals with moderate hypertriglyceridemia. J Clin Lipidol. PubMed
  89. So 2022 (2022). Ethyl EPA and ethyl DHA cause similar and differential changes in plasma lipid concentrations and lipid metabolism in subjects with low-grade chronic inflammation. J Clin Lipidol. PubMed
  90. Sorice 2011 (2011). N-3 polyunsaturated fatty acids reduces post-operative atrial fibrillation incidence in patients undergoing "on-pump" coronary artery bypass graft surgery. Monaldi Arch Chest Dis. PubMed
  91. Stark 2000 (2000). Effect of a fish-oil concentrate on serum lipids in postmenopausal women receiving and not receiving hormone replacement therapy in a placebo-controlled, double-blind trial. Am J Clin Nutr. PubMed
  92. Stone 2000 (2000). The Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardio (GISSI)-Prevenzione Trial on fish oil and vitamin E supplementation in myocardial infarction survivors. Curr Cardiol Rep. PubMed
  93. Stroes 2018 (2018). Omega-3 carboxylic acids in patients with severe hypertriglyceridemia: EVOLVE II, a randomized, placebo-controlled trial. J Clin Lipidol. PubMed
  94. Szarek 2024 (2024). Lipoprotein(a) Blood Levels and Cardiovascular Risk Reduction With Icosapent Ethyl. J Am Coll Cardiol. PubMed
  95. Taheri. The effect of omega-3 fatty acid supplementation on oxidative stress in continuous ambulatory peritoneal dialysis patients. PubMed
  96. Vanderbilt 2015 (2015). Effect of omega-three polyunsaturated fatty acids on inflammation, oxidative stress, and recurrence of atrial fibrillation. Am J Cardiol. PubMed
  97. Vasheghani Farahani 2017 (2017). Fish oil supplementation for primary prevention of atrial fibrillation after coronary artery bypass graft surgery: A randomized clinical trial. Int J Surg. PubMed
  98. Verma 2021 (2021). Icosapent Ethyl Reduces Ischemic Events in Patients With a History of Previous Coronary Artery Bypass Grafting: REDUCE-IT CABG. Circulation. PubMed
  99. Wang 2024 (2024). A 12-week randomized double-blind clinical trial of eicosapentaenoic acid intervention in episodic migraine. Brain Behav Immun. PubMed
  100. Wilbring 2014 (2014). Omega-3 polyunsaturated Fatty acids reduce the incidence of postoperative atrial fibrillation in patients with history of prior myocardial infarction undergoing isolated coronary artery bypass grafting. Thorac Cardiovasc Surg. PubMed
  101. Woo 2021 (2021). Comparison of the Efficacy and Safety of Atorvastatin 40 mg/ω-3 Fatty Acids 4 g Fixed-dose Combination and Atorvastatin 40 mg Monotherapy in Hypertriglyceridemic Patients who Poorly Respond to Atorvastatin 40 mg Monotherapy: An 8-week, Multicenter, Randomized, Double-blind Phase III Study. Clin Ther. PubMed
  102. Wu 2006 (2006). Effects of docosahexaenoic acid supplementation on blood lipids, estrogen metabolism, and in vivo oxidative stress in postmenopausal vegetarian women. Eur J Clin Nutr. PubMed
  103. Yokote 2020 (2020). Short-Term Efficacy (at 12 Weeks) and Long-Term Safety (up to 52 Weeks) of Omega-3 Free Fatty Acids (AZD0585) for the Treatment of Japanese Patients With Dyslipidemia　- A Randomized, Double-Blind, Placebo-Controlled, Phase III Study. Circ J. PubMed
  104. Yokoyama 2007 (2007). Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis. Lancet. PubMed
  105. Yuan. Twelve-week combined arginine and fish oil supplementation is associated with reduced sarcopenia severity: a randomized, double-blind, placebo-controlled study. PubMed
  106. Мареев 2020 (2020). [Influence of Omega-3 PUFA on Non-invasive factors determining the risk of arrhYthmias eXcess and sudden cardiac death in patients with HFpEF with ischemic etiology (ONYX)]. Kardiologiia. PubMed