Oil supplementation and cardiovascular mortality in the dialysis population

For many years and despite the best efforts of the community in optimizing dialysis adequacy, fluid management, blood pressure, anemia, and bone mineral disorder control, the mortality of patients receiving maintenance hemodialysis remains unacceptably high: ∼15% per year in the USA in 2022 [1]. In the European population, the expected remaining lifetime is 67% shorter in patients undergoing maintenance dialysis [2]. In more detail, almost half of these deaths are from cardiovascular causes [3], yet they resist the standard preventive arsenal. Many intervention trials (4D, AURORA, CHOIR, CREATE, TREAT, IDEAL, FACT, CONTRAST, ESHOL, ALCHEMYST) have been negative in the context of kidney failure whereas comparable interventions have been successful in other diseases (Fig. 1). Fortunately, the results of the Protection against Incidences of Serious Cardiovascular Events Study (PISCES) reverses this trend with a lot of power and clinical interest [4]. An encouraging and emerging signal for reducing cardiovascular mortality across the petals of clinical trials in hemodialysis patients. CVD, cardiovascular disease; HDF, hemodiafiltration; CI, confidence interval; MHD, maintenance hemodialysis. *Hazard ratio (HR) for all-cause mortality. PISCES is not merely another nutritional study; it is a challenge to our current understanding of dialysis cardiomyopathy. By randomizing 1228 patients to a medium range-dose regimen of 4 g of n-3 polyunsaturated fatty acids (n-3 PUFAs), specifically containing 1.6 g EPA (eicosapentaenoic acid) and 0.8 g docosahexaenoic acid (DHA) daily versus a corn-oil placebo, the investigators achieved a risk reduction that has been uncommonly met in recent nephrology trials. The primary endpoint—a composite of serious cardiovascular events—was reduced by 43% [4]. Even more compelling is the consistency across secondary endpoints. Authors reported a 45% reduction in cardiac death, a 44% reduction in myocardial infarction, and a staggering 63% reduction in stroke. In a population where “negative trials” have been the norm, these hazard ratios, with confidence intervals well below 1 call for an immediate and rigorous attention. Moving from unsuccessful standard cardiovascular risks treatment to specific uremic toxicity targets may explain why is this study successful. The answer may lie in the specific pathology of the patient undergoing maintenance dialysis. We are not dealing primarily with unstable atherosclerotic plaques (the target of statins), but rather with an “uremic milieu” characterized by circulating toxins, organ fibrosis, left ventricular hypertrophy, and electrolyte fluxes that may predispose patients to arrhythmias. The dose of 4 g of n-3 PUFAs used in PISCES is remarkable. It stands between usual western dietary intakes (actual daily intake of 250 vs 500 mg recommended), and high pharmacological doses up to 15 g daily, which are not considered anymore due to important side-effects. Early trials in the nondialysis population using low dose omega-3 supplementation yielded mixed results. The GISSI Heart Failure trial (n = 11 324; 0.8 g EPA + DHA, EPA: DHA ratio 1:2) suggested modest benefits in a population with limited background cardiovascular therapy, whereas later trials in statin-treated populations failed to confirm these effects, as exemplified by ASCEND (n = 15 480; 0.4 g EPA + 0.3 g DHA; 75% taking statins)[5]. The PISCES n-3 PUFAs supplement aligns with the successful REDUCE-IT trial using a slightly higher dose (n = 8 179, 4 g per day of purified EPA (icosapent ethyl), 97% on statins) in a nondialysis population [6]. At this dosage, EPA and DHA incorporate into the phospholipid bilayer of cardiomyocytes, modulating ion channels (specifically the Na and Ca channels) [7]. This stabilizes the electrical potential of the cell, raising the threshold for ventricular fibrillation—essentially providing an “internal chemical defibrillator” [7]. This aligns with the observed reduction in sudden cardiac death, a mechanism distinct from lipid-lowering alone. It also suggests not just a delayed anti-atherosclerotic effect, but an immediate anti-arrhythmic or hemodynamic benefit, crucial for the dialysis population. Although a risk of increased atrial fibrillation (AF) was suggested in two previous trials [8, 9] a recent large report in >260 000 adults from the UK Biobank showed a reduced risk of AF with higher plasma omega-3 levels and no association between fish-oil supplements and AF [10]. Thus there might be a possible discrepancy between the dietary intake of fish oil and the actual plasma concentration of n-3 PUFAs, which may deserve further research into the absorption and metabolism of n-3 PUFAs. This might not have happened in the PISCES since a substudy of plasma fatty acids showed that patients in the omega-3 group increased their n-3 PUFAs by 2.4% compared with the corn-oil group (Table S6b4). N-3 PUFAs are potent precursors of resolvins and protectins that actively improve inflammatory resolution and endothelial function, and experimental studies consistently demonstrate beneficial metabolic effects, partly mediated through a free fatty acid receptor 4, also known as GPR120, which may explain benefits observed in stroke and peripheral vascular disease [11–13]. Interestingly, n-3 PUFAs may have a Janus face in inflammation biology. A pilot randomized dose–response trial reported the effects of two doses of n-3 PUFAs (1.8 and 3.6 g/day) in nondialysis CKD patients [14]. Serum adiponectin, leptin, C-reactive protein, and tumor necrosis factor-α levels were not modified by either dose of n-3 PUFAs, whereas interleukin-6 was only reduced by the 1.8 g/d n-3 PUFAs dose, suggesting an improvement in the inflammatory status and the subsequent cardiovascular risk reported in patients on maintenance hemodialysis. With the higher n-3 PUFAs dose of 3.6 g/d, improvement in lipid profile and adipocyte metabolism was observed, whereas there was no change in inflammatory markers [14]. This suggests a dose-effect of n-3 PUFAs in CKD patients that could not been captured in PISCES who studied a single n-3 PUFAs dose of 4 g [4]. The EPA to DHA balance should also be explored, as experimental data indicate distinct biological effects, with EPA showing more favorable actions [5]. Bleeding is an issue during high-dose n-3 PUFAs administration. PISCES found no significant increase in serious bleeding events (4.8% in fish oil vs. 7.6% in placebo). This reassures us that 4 g n-3 PUFAs/day is safe, even in patients on anticoagulation for fistula patency. It has been hypothesized that the corn-oil supplement in the control arm might have had some biological effects per se. The daily 4 g corn oil might have brought 1 g of omega-9 oleic acid and some vitamin E, both protective compounds and 2.4 g omega-6 linoleic acid. Daily omega-9 and omega-6 fatty acids requirements are 20–30 and 8–9 g, respectively, thus the corn-oil supplement does not seem excessive compared to a usual dietary intake. Also, experimental data do not support a clinically meaningful pro atherogenic effect of corn oil [15]. The modest amount of omega-6 might have been counterbalanced by the protective properties of vitamin E and oleic acid from the corn oil as well. A 43% reduction is a rare and outlier result in clinical research, and often regresses to the mean in subsequent studies. Therefore, additional confirmatory data are warranted: a well-known condition in clinical research. Furthermore, we must scrutinize the safety signal. There seemed to be more infections and respiratory symptoms in the treatment group. In addition, and not fully documented, in the n-3 PUFAs group there were 35% more deaths (112 vs 82 in the corn-oil group) from noncardiovascular causes as the overall death rate was not different between groups. In a population where infection is the second leading cause of death, any immunomodulatory agent—even a nutritional one—must be checked for immunosuppressive risks [16]. While the total serious adverse events were balanced, this signal warrants careful monitoring in clinical practice. Finally, more information would have been welcome on the routine dialysis treatment and adequacy to ensure that this group of patients was not different from other maintenance dialysis populations worldwide and that a fish-oil treatment may be proposed elsewhere. We now stand at a crossroads. Regulatory bodies are unlikely to grant a labeling indication based on a single trial, and a confirmatory study is scientifically warranted. Yet, we face the ethical dilemma of the clinician: should we withhold a safe, low-cost intervention with a potential 43% mortality benefit while waiting another 5 years for a replication study? Pending confirmatory data, the use of high-dose n-3 PUFAs (specifically 4 g, as lower doses are likely ineffective) should be strongly considered for high-risk patients, particularly those with a history of AF or ischemic heart disease. Also, we must remain alert and monitor our patients for infection and bleeding, ensuring that this major outcome does not come with hidden concerns. Before widespread adoption, parallel lines of investigation are needed. The mode of n-3 PUFAs administration warrants scrutiny, as manufacturing may degrade bioactive compounds, introduce contaminants, or require additives that affect product quality [17]. In addition, the dietary context is critical, since high omega-6 and ultra-processed food intake antagonizes omega-3 metabolism. This raises the question of whether the K-DOQI nutrition guidelines [18] should evolve to more strongly promote omega-3 rich dietary patterns, such as Nordic or Mediterranean diets, which combine higher omega-3 with lower omega-6 intakes and have shown benefits for CKD progression and complications [19]. Finally, the PISCES has broken the curse of failure in dialysis cardiovascular trials and brings nutrition back into focus as a potentially effective therapeutic strategy. Whether this is a singular anomaly or the dawn of a new metabolic standard of care remains to be seen, but we have here a genuine reason for optimism. The authors declare no conflict of interest or any funding for writing this paper.