Quick Links

Recommendations Summary

CKD: Macronutrients: LC n-3 PUFA Nutritional Supplements and Hard Outcomes (2020)

Click here to see the explanation of recommendation ratings (Strong, Fair, Weak, Consensus, Insufficient Evidence) and labels (Imperative or Conditional). To see more detail on the evidence from which the following recommendations were drawn, use the hyperlinks in the Supporting Evidence Section below.


  • Recommendation(s)

    CKD: LC n-3 PUFA Nutritional Supplements for Mortality and Cardiovascular Disease, Maintenance Hemodialysis and Post-Transplant

    In adults with CKD 5D on MHD or posttransplantation, we suggest not routinely prescribing LC n-3 PUFA, including those derived from fish or flaxseed and other oils, to lower risk of mortality (2C) or cardiovascular events (2B).

    Rating: Fair
    Imperative

    CKD: LC n-3 PUFA Nutritional Supplements for Mortality and Cardiovascular Disease, Peritoneal Dialysis

    In adults with CKD 5D on PD, it is reasonable to not routinely prescribe LC n-3 PUFA, including those derived from fish or flaxseed and other oils, to lower risk of mortality or cardiovascular events (OPINION).

    Rating: Consensus
    Conditional

    • Risks/Harms of Implementing This Recommendation

      Adverse effects of capsule-based supplementation may lead to gastrointestinal side effects like stomach upset and eructation (though the latter can be masked by different formulations). Theoretical risks like bleeding have not been borne out in clinical trials.

    • Conditions of Application

      All treatment should be individualized according to a patient's preferences, health history and co-morbidities as well as clinical judgement. 

    • Potential Costs Associated with Application

      • LC n-3 PUFA supplementation considerations will differ depending on whether the intervention is diet-based or capsule-based.
      • For dietary interventions the goal of supplementation must be clearly defined. If it is to raise blood levels of a-linolenic acid then supplementation should focus on soybean, flaxseed, and other oils as well as meat and dairy products.  If it is to raise EPA or DHA blood/tissue levels then the primary dietary sources must be sardine, mackerel, salmon and other high-content marine-based foods. Potential limitations to dietary supplementation include their relatively high cost and difficulty in achieving high daily intake. In addition, the source and processing method will influence LC n-3 PUFA foodstuff content. For example, farmed fish typically (but not always) has lower LC n-3 PUFA compared to wild fish, while frying fish could alter the n-3/n-6 ratio which may be of clinical significance (Strobel, et al. 2012).
      • Capsule-based supplementation involves a set of different considerations. While dozens of commercial LC n-3 PUFA supplements are available, quality control is often lacking (Kleiner, et al. 2015).This makes precise dosing recommendations difficult. An alternative route is to have the patient obtain supplements via physician prescription (e.g. icosapent ethyl, omega-3 ethyl esters). For either option cost could be an issue. Achieving high dose supplementation will be easier with capsules then through dietary consumption. Adverse effects of capsule-based supplementation may lead to gastrointestinal side effects like stomach upset and eructation (though the latter can be masked by different formulations). Theoretical risks like bleeding have not been borne out in clinical trials.
      • LC n-3 PUFA content is listed on the website of the National Institutes of Health (NIH 2018) (https://ods.od.nih.gov/factsheets/Omega3FattyAcids-HealthProfessional/).

    • Recommendation Narrative

      Rationale/Background
      Long chain omega-3 polyunsaturated fatty acids (LC n-3 PUFA) include eicosapentaenoic (EPA), docosapentaenoic and docohexaenoic acids (DHA), all of which are obtained primarily from dietary sources like cold-water fish (i.e. fish oil), or linoleic acid, which is derived from flaxseed or certain other vegetable oils. In recent decades LC n-3 PUFA have demonstrated protean biologic effects that mediate eicosanoid production, cell membrane physiology, signal transduction, metabolism, apoptosis, oxidation, and inflammation. Accordingly, they have been tested in a variety of medical conditions. Of particular interest has been their putative effects on cardiac membrane stabilization leading to possible reduction of malignant arrhythmias and sudden cardiac death. Patients with CKD have been documented to have some of the lowest record blood levels of LC n-3 PUFA on record (Friedman, et al. 2012) thus making them potentially very suitable candidates for supplementation interventions. In fact, LC n-3 PUFA supplementation has been studied as possible therapy for a number of conditions commonly observed in patients with CKD including dyslipidemia, hemodialysis access failure, cardiovascular disease and death, as well as for their immunomodulatory effects in patients with kidney allografts.

      Detailed Justification

      All-Cause Mortality and Cardiovascular Events
      Despite the putative overall benefits of LC n-3 PUFA and the elevated risk of death in patients with CKD, three RCTs demonstrated no improvement in mortality with supplementation. However, the studies were heterogeneous in terms of study population (Svensson et al 2006 in patients with MHD, two in patients with CKD with kidney allografts- Maachi et al and Berthoux et al ), the dose of LC n-3 PUFA (Maachi et al 1.44g/d EPA + 0.96g/d DHA; Berthoux et al 1.62g/d EPA+1.08g/d DHA; Svensson et al 2006 0.77g/d EPA+ 0.64g/d DHA) and the study duration (1-2 years), with the combined study population being fairly modest in size (n=264).

      Two well-designed but modestly sized (combined N=351) RCTs in patients on MHD reported mixed results on the effect of LC n-3 PUFA supplementation on cardiovascular events (Svensson et al 2006,  Lok et al 2012). Lok et al. reported that 4 g/day fish oil (1.6 g/d EPA, 0.8 g/d DHA) for 12 months as compared to a corn oil-based placebo significantly lowered the cardiovascular event rate 0.41 (0.20 to 0.85) (p=0.02) and improved cardiovascular event-free survival 0.43 (0.19 to 0.96) (p=0.04) but did not influence the number of patients with one or more events 0.78 (0.55 to 1.09) (p=0.15). All were secondary outcomes in a trial designed to study MHD vascular access. In a secondary prevention trial Svennson et al. reported that 1.7 g/d fish oil (0.77g/d EPA and 0.64g/d DHA) for two years had no effect on the primary combined endpoint of cardiovascular events or death as compared to olive oil-based placebo (Svensson et al 2006) but did improve the secondary endpoints of myocardial infarctions (0.30 (0.10, 0.92) (p=0.036) and major coronary events (0.40 (0.17, 0.97) (p=0.043).

      Glomular Filtration Rate and CKD Progression
      Based on six RCTs with widely differing study design and populations (CKD stage 3, MHD, kidney allografts) (Maachi et al 1995, Berthoux et al 1992, Mori et al 2009, Bennett et ala 1995), fish oil supplementation was not found to influence estimated or measured GFR. In the study by Guebre Egziabher et al 2013, participants received 1.8g of n-3 fatty acids, but authors do not describe EPA or DHA amount. In the study by Bennett et al, participants received “9 or 18g/d EPA capsules”. In the remaining studies, EPA dose ranged from 0.46-1.62g/d and DHA ranged from 0.25-1.08g/d.

      Similarly, fish oil supplementation for 8-12 weeks did not influence serum creatinine levels in three studies of non-dialyzed patients who used placebo or non-placebo-based control groups. In the study by Guebre Egziabher et al 2013, participants received 1.8g/d of n-3 fatty acids, but authors do not describe EPA or DHA amount. In the remaining studies, EPA amount ranged from 0.69-1.44g/d and DHA amount ranged from 0.25-0.96g/d (Svensson et al 2004, Bouzidi et al 2010).

      Blood Pressure
      Five RCTs examined the effect of LC n-3 PUFA supplementation on BP, two in non-dialyzed patients (no stage reported) (Svensson et al 2004, Mori et al 2009), two in patients on MHD (Lok et al 2012, Khajehdehi et al 2000) and one in patients with CKD with kidney allografts Bennett et al 1995). The results were mixed.  In non-dialyzed patients Svensson et al reported that fish oil (0.96g/d from DHA and 1.44g/d EPA) for 8 weeks did not affect BP (Svensson et al 2004) while Mori et al found that fish oil (0.38g/d DHA and 0.46g/d EPA) for 8 weeks lowered both systolic (meaSEM, (-3.3 (±0.7)) and diastolic (-2.9 (± 0.5)), (p<0.0001 for each change) blood pressures. A pooled analysis of these two trials did not find an overall beneficial effect. In patients on MHD, Lok et al reported an improvement in systolic BP in patients on MHD with fish oil (0.8g/d DHA and 1.6g/d EPA) for one year [Mean Difference (95% CI): -8.10 (-15.4, -0.85), p=0.014) and a reduction in the number of BP medications but no effect on diastolic BP. In contrast, Khajehdehi et al found no effect at all on BP of 1.5 g/day fish oil (DHA and EPA content not reported) as compared to placebo for two months. Data from these two trials could not be pooled. Bennett et al randomized patients with CKD with kidney allografts and reported no benefit of “9 or 18 g EPA capsules” per day versus placebo for 26 weeks188 on systolic BP but did note a reduction in diastolic BP in both EPA arms (p<0.05 for each) only.

      Inflammatory Markers
      The putative anti-inflammatory effects of LC n-3 PUFA were tested on two established biomarkers of inflammation.

      C-Reactive Protein
      Fifteen RCTs studied the effect of fish oil supplementation on circulating CRP. In pre-dialysis patients throughout the stages of CKD, fish oil either compared to placebo (Mori et al 2009, Madsen et al 2007)  or at varying doses (Guebre-Egziabber et al 2005)  had no effect. The pattern in patients on MHD was similar. A pooled analysis of ten studies (Lemos et al 2012, Khalatbari et al 2013, Bowden et al 2009, Daud et al 2012, Kooshki et al 2011, Poulia et al 2011, Saifullah et al 2007, Gharekhani et al 2014, Hung et al 2015, Harving et al 2015) found no effect of LC n-3 PUFA supplementation (nine using fish oil containing 0.42-1.8 g/d EPA with 0.24-1.14 g/d DHA, one using 2g/d flaxseed oil) on circulating CRP as compared to placebo (MD -1.73 mg/L, 95% CI: -3.54, 0.09). Ewers et al found that fat supplementation (which also included fats other than LC-n-3 PUFA adn specific n-3 PUFAs were not described) was associated with a reduction in CRP (p=0.01) after 12 weeks as compared to non-supplemented patients.

      Interleukin-6 
      Six RCTs studied the effect of LC n-3 PUFA on circulating IL-6. Neither of the two studies in predialysis patients comparing fish oil supplementation to placebo or at varying doses found a significant effect on IL-6  (one study reported 1.8g/d n-3 PUFAs total and one reported 1.4g/d EPA with 1.0g DHA) (Guebre-Egziabher et al 2005, Deike et al 2012), nor did a pooled analysis of four studies in patients on MHD (MD 5.32 pg/ml, 95% CI: -5.637, 16.275) in which participants received 0-1.93 g/d EPA with 0.72-0.97 g/d DHA) (Kooshki et al 2011, Gharekhani et al 2014, Hung et al 2015, Himmelfarb et al 2007).

    • Recommendation Strength Rationale

      The evidence supporting the recommendation statements is based on Grade III/ Grade C evidence for MHD or post-transplant;  and for individuals on PD the supporting evidence is based on Consensus/expert opinion. The clinical impact of LC n-3 PUFA supplementation in patients with CKD was challenging to assess due to short study duration, modest sample size, and broad heterogeneity in in terms of the absolute and relative amounts of n-3 PUFA supplemented, the type of placebo used, and the study duration, making it more difficult to provide conclusive evidence for or against the use of LC n-3 PUFA as a treatment option. Furthermore, baseline LC n-3 PUFA levels (either in blood or tissues) were not typically used to target populations that would most benefit. This is an important but often overlooked point because the putative benefits of LC n-3 PUFA supplementation may be inversely related to baseline blood or tissue concentrations (Friedman, et al. 2012; Mozaffarian, et al. 2006).

    • Minority Opinions

      Consensus reached.