EPA Fish Oil

High-dose EPA (the omega-3 fat eicosapentaenoic acid) is one of the most evidence-backed supplements for cardiovascular risk - but the benefit is specific to statin-treated patients with elevated triglycerides, not the general population. So if your doctor mentioned high triglycerides, or a headline tied fish oil to heart attacks, that distinction is the place to start. In REDUCE-IT (n=8,179), 4g/day of icosapent ethyl (the prescription pure-EPA drug sold as Vascepa) cut the primary composite endpoint by 25% (HR 0.75, p<0.001), with an NNT of 21 over 4.9 years - that is how many such patients had to take it, for that long, to prevent one cardiovascular event. The effect appears specific to pure EPA: the STRENGTH trial of EPA+DHA at the same dose was terminated for futility (HR 0.99). If you are high-risk and on a statin, this is a candidate conversation with your doctor, not a routine OTC purchase.

The split between EPA alone and EPA-plus-DHA is the whole story here, so it is worth slowing down on. The larger Japanese JELIS trial (n=18,645) confirmed that 1.8g/day EPA added to statin therapy significantly reduces coronary events. A 2021 meta-analysis (Khan et al., PMID: 34505026) put numbers on the EPA-specific advantage: pure EPA reduced CV mortality by 18% (RR 0.82, 95% CI 0.68-0.99), nonfatal MI by 28% (RR 0.72, 95% CI 0.62-0.84), and total coronary events by 27%. The EPA+DHA combination, by contrast, barely moved CV mortality (RR 0.94) or MI (RR 0.92). That gap is the reason a high-EPA product is not interchangeable with a generic "fish oil" capsule that bundles in DHA.

The clearest single piece of evidence for that gap is STRENGTH (n=13,078), which tested 4g/day of EPA+DHA carboxylic acids (Epanova) against corn oil. The result: HR 0.99 (95% CI 0.90-1.09, p=0.84), terminated early for futility. Every individual component was null. Even patients in the highest tertile of EPA levels showed no benefit. This is the strongest evidence that EPA alone drives the cardiovascular benefit, not omega-3s in general. The leading biochemical explanation is that DHA competes with EPA at the cellular level - EPA's anti-inflammatory and membrane-stabilizing effects appear to require high selectivity, and DHA's competing metabolic pathways may blunt EPA's cardiovascular benefit when the two are taken together at high doses. That would explain why JELIS (pure EPA) and REDUCE-IT (pure EPA) succeeded while STRENGTH (EPA+DHA) failed completely.

There is also a dose-in-the-blood angle that may explain much of the gap between trials. REDUCE-IT pushed median EPA plasma levels to 144.0 μg/mL at 1 year (up ~400% from baseline). STRENGTH reached an average peak EPA of only 89.6 μg/mL despite using similar doses - likely because the formulation was absorbed differently. Even the STRENGTH patients in the highest EPA tertile (>116 μg/mL) showed no cardiovascular benefit, which hints that the level needed for plaque stabilization and endothelial protection may sit above what EPA+DHA combinations reach. In REDUCE-IT, higher EPA plasma levels tracked with fewer ischemic events.

One honest caveat keeps the REDUCE-IT number from being taken at face value: the mineral oil placebo was not biologically inert. hs-CRP increased >30%, LDL-C rose slightly, and ApoB increased significantly in the control arm. Modeling suggests the "true" HR may be closer to 0.88 than 0.75. Even that adjusted estimate is still a clinically meaningful 13%+ relative risk reduction from pure EPA beyond lipid changes - so the benefit holds up, it is just probably smaller than the headline.

For triglycerides specifically, the AHA confirms that 2-4g daily of EPA lowers triglycerides by 20-30% in moderate hypertriglyceridemia (200-499 mg/dL). Dose matters, and not gently: 0.85g/day failed to move TG at all, while 3.4g/day achieved a 27% reduction (p=0.002). Below the threshold, you are paying for very little.

Form matters too, because it decides how much of the oil you actually absorb. Triglyceride (TG) and re-esterified triglyceride (rTG) forms absorb substantially better than ethyl ester (EE), especially if you take them without a fatty meal. The major cardiovascular trials happened to use the ethyl ester form, but for everyday supplementation rTG is the one to reach for. High-dose EPA for cardiovascular risk should be guided by a physician.

The safety picture at high doses is the part most labels skip. The Khan meta-analysis found EPA monotherapy uniquely increases bleeding risk (RR 1.49, 95% CI 1.20-1.84) - a signal not seen with EPA+DHA combinations. Atrial fibrillation (an irregular, often racing heartbeat) risk is elevated across all omega-3 formulations (RR 1.26, 95% CI 1.08-1.48), with EPA alone showing the highest signal (RR 1.35, 95% CI 1.10-1.66). On the other side of the ledger, pure EPA has one clear advantage over EPA+DHA: it does not raise LDL-C, while EPA+DHA combinations increase LDL by 5-20% in patients with severe hypertriglyceridemia.