Fish oil halved serious cardiovascular events in a dialysis trial. The size is the puzzle

People on maintenance haemodialysis die of cardiovascular disease more than of anything else, and very few preventive therapies have been shown to help them [s1]. The trial investigators put it flatly: cardiovascular disease is the leading cause of death in patients receiving haemodialysis, yet effective preventive therapies remain limited [s1]. The protocol authors made the same point, noting that people with kidney failure on replacement therapy suffer premature cardiovascular mortality and events with few proven pharmacological interventions [s2]. So a positive result is news on its own terms.

PISCES, published in the New England Journal of Medicine on 7 November, reports that daily fish-oil supplementation lowered the rate of serious cardiovascular events in haemodialysis patients compared with placebo, with a hazard ratio of 0.57 [s1]. The journal ran an accompanying editorial whose title — "Red Herring or Great Catch?" — signals that the size of the effect is doing the arguing [s3].

The trial

PISCES was a double-blind, randomised, placebo-controlled trial at 26 sites in Canada and Australia [s1]. Adults receiving maintenance haemodialysis were assigned to daily supplementation with fish oil — 4 g of n-3 polyunsaturated fatty acids, comprising 1.6 g of eicosapentaenoic acid and 0.8 g of docosahexaenoic acid — or to a corn-oil placebo [s1].

The primary endpoint was a composite of all serious cardiovascular events: sudden and non-sudden cardiac death, fatal and non-fatal myocardial infarction, peripheral vascular disease leading to amputation, and fatal and non-fatal stroke [s1].

The word doing the work in that sentence is all. The published protocol is explicit that the primary outcome is the rate of all serious cardiovascular events, "not just the first" [s2]. That is a deliberate design choice, and a defensible one in a population that accumulates repeated events, but it is not the analysis most cardiovascular trials report, and it is not directly comparable to them.

Recruitment ran from 28 November 2013 to 22 July 2019, and 1,228 participants were randomised — 610 to fish oil and 618 to placebo [s1]. The protocol had targeted 1,100 patients across 26 dialysis units, with an expected intervention period of at least 3.5 years [s2]; follow-up in the published report was 3.5 years [s1].

The results

The rate of serious cardiovascular events was 0.31 per 1,000 patient-days in the fish-oil group and 0.61 per 1,000 patient-days in the placebo group (hazard ratio 0.57; 95% CI 0.47 to 0.70; P<0.001) [s1].

The individual components moved in the same direction. The hazard ratio was 0.55 (95% CI 0.40 to 0.75) for cardiac death, 0.56 (0.40 to 0.80) for fatal and non-fatal myocardial infarction, 0.57 (0.38 to 0.86) for peripheral vascular disease leading to amputation, and 0.37 (0.18 to 0.76) for fatal and non-fatal stroke [s1].

Two secondary analyses are the ones to read carefully. When the primary endpoint was extended to include non-cardiac causes of death, the hazard ratio rose to 0.77 (95% CI 0.65 to 0.90) [s1]. And when the outcome was recast as a first cardiovascular event or death from any cause — the conventional time-to-first-event framing — the hazard ratio was 0.73 (95% CI 0.61 to 0.87) [s1].

Both of those are still favourable, and both are meaningfully smaller than 0.57. The gap is not a contradiction; it is what happens when a trial counts recurrent events in a population where the sickest patients accrue several. Readers comparing PISCES against other cardiovascular prevention trials should compare against 0.73, not 0.57.

Adherence to the trial regimen and the incidence of adverse events did not differ meaningfully between the groups [s1].

Why the size invites scrutiny

A hazard ratio of 0.57 for serious cardiovascular events, produced by a fish-oil supplement, is a large effect by the standards of preventive cardiology [s1]. The authors frame the question the trial was built to answer as genuinely open: supplementation with n-3 polyunsaturated fatty acids, especially EPA and DHA, may have cardiovascular benefits in the general population, but efficacy among patients receiving haemodialysis was uncertain [s1]. A result at that magnitude, in a population where little has worked, is the kind of finding that gets checked hard before it changes anything.

Several features of the trial constrain how far the result travels. It was conducted at 26 sites in two countries with broadly similar dialysis systems [s1][s2]. Randomisation ended in mid-2019, so the standard of background care reflects that era [s1]. The comparator was a corn-oil placebo rather than an inert capsule, which is a design detail worth noting when interpreting the size of a between-group difference [s1]. And the trial reports a rate-based primary endpoint that will not line up cleanly with the endpoints readers know from other trials [s1][s2].

None of that makes the finding wrong. It makes it a single trial with an unusually large effect, which is precisely the situation in which replication matters most.

What this is not

This is not a recommendation to take fish oil, and PISCES does not speak to people who are not on maintenance haemodialysis. The dose studied — 4 g of n-3 polyunsaturated fatty acids daily, of a specified EPA and DHA composition — is a trial regimen in a supervised population with a specific disease, not a general-purpose supplement schedule [s1]. Over-the-counter fish-oil products vary widely in content and are not interchangeable with what was tested here.

The trial was supported by the Heart and Stroke Foundation of Canada and others, and was registered under ISRCTN00691795 [s1].

What to watch

Whether the effect survives independent replication in another dialysis population, and whether the recurrent-event analysis is reproduced alongside a conventional time-to-first-event one, are the two things that will determine how PISCES is read in a few years. Until then, the honest summary is a large single-trial benefit in a population that has had almost nothing else work.

Sources

• [s1] Fish-Oil Supplementation and Cardiovascular Events in Patients Receiving Hemodialysis, The New England Journal of Medicine, 7 November 2025. https://doi.org/10.1056/NEJMoa2513032
• [s2] Protection against Incidences of Serious Cardiovascular Events Study with daily fish oil supplementation in dialysis patients (PISCES): protocol for a randomised controlled trial, BMJ Open, 10 January 2024. https://doi.org/10.1136/bmjopen-2023-072239
• [s3] Fish Oil for Patients Receiving Hemodialysis — Red Herring or Great Catch?, The New England Journal of Medicine, 7 November 2025. https://doi.org/10.1056/NEJMe2515057