Publication

• Title: Enteral Omega-3 Fatty Acid, γ-Linolenic Acid, and Antioxidant Supplementation in Acute Lung Injury
• Acronym: OMEGA
• Year: 2011
• Journal published in: JAMA
• Citation: Rice TW, Wheeler AP, Thompson BT, deBoisblanc BP, Steingrub J, Rock P, et al; National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network of Investigators. Enteral omega-3 fatty acid, γ-linolenic acid, and antioxidant supplementation in acute lung injury. JAMA. 2011;306(14):1574-1581.

Context & Rationale

• Background

  • Acute lung injury (ALI)/ARDS is characterised by neutrophilic lung inflammation, permeability oedema, and prothrombotic signalling, with substantial mortality and prolonged mechanical ventilation.
  • Omega-3 (n-3) fatty acids (EPA/DHA) can shift membrane phospholipid substrates towards less inflammatory eicosanoid mediators, while γ-linolenic acid (GLA) and antioxidants have plausibly anti-inflammatory and anti-oxidant effects.
  • Prior small randomised trials of an enteral formula enriched with n-3 fatty acids, GLA, and antioxidants reported improved oxygenation and other outcomes, but interpretation was limited by small sample sizes and “as-treated” analyses restricted to patients who tolerated full continuous feeding.
  • Observational data suggested n-3 fatty acid levels are reduced in patients at risk of, and with, established ALI, supporting a biologically plausible “replacement/therapeutic” strategy.
• Research Question/Hypothesis

  • In adults with early ALI receiving mechanical ventilation and enteral nutrition, does twice-daily enteral bolus supplementation with EPA/DHA, GLA, and antioxidants increase ventilator-free days (VFDs) to day 28 compared with an isocaloric control supplement?
  • Mechanistic hypothesis: supplementation would increase the n-3:n-6 fatty acid balance, reduce inflammatory mediators, and translate into improved clinical outcomes.
• Why This Matters

  • “Immunomodulatory” lipid/antioxidant-enriched enteral products were already available and used in some ICUs, but high-certainty efficacy and safety evidence in ALI/ARDS was lacking.
  • A robust multicentre, blinded, intention-to-treat evaluation could confirm benefit, demonstrate harm, or resolve uncertainty created by earlier small trials.
  • Because nutritional interventions can change macronutrient delivery (not only “pharmaconutrients”), a rigorous trial could clarify whether any signal reflected biology or confounding by feeding tolerance and comparator composition.

Design & Methods

• Research Question: In early ALI, does twice-daily bolus enteral supplementation with EPA/DHA, GLA, and antioxidants (vs an isocaloric control bolus) improve ventilator-free days to day 28?
• Study Type: Randomised, double-blind, placebo-controlled, multicentre trial in 44 hospitals within the NHLBI ARDS Network (2 Jan 2008 to 21 Feb 2009); conducted alongside EDEN (trophic vs full-calorie feeding) in a factorial design.
• Population:
  • Setting: Mechanically ventilated adults with early ALI in ICUs at 44 ARDSNet hospitals.
  • Key inclusion criteria: Within 48 hours of meeting ALI criteria; PaO₂/FiO₂ <300 (altitude-adjusted if >1000 m); bilateral infiltrates consistent with pulmonary oedema; positive-pressure ventilation via endotracheal tube; no clinical evidence of left-sided cardiac failure; clinical intent to provide enteral nutrition.
  • Frequent exclusions (counts reflect multiple possible reasons per patient): severe chronic lung disease (n=500); inability to obtain consent (n=417); likely fatal underlying disease (n=306); recent intracranial haemorrhage (n=250); severe liver disease (n=194); moribund (n=167); refractory shock (n=113); coagulopathy (n=108); time window exceeded (ALI >48 h or intubated >72 h; n=445).
  • Randomised: 272/2778 screened (143 intervention; 129 control); all had complete follow-up to hospital discharge or day 60.
• Intervention:
  • Delivery: Twice-daily enteral boluses of 120 mL (total 240 mL/day), started within 6 hours of randomisation; continued until the earliest of 21 days, 48 hours of unassisted breathing, or extubation.
  • Composition (per day; 240 mL): 480 kcal; fat 44.6 g; carbohydrate 4.2 g; protein 3.8 g; EPA 6.84 g; DHA 3.40 g; GLA 5.92 g; vitamin C 1000 mg; vitamin E 440 IU; beta-carotene 4.8 mg; zinc 24.2 mg; selenium 85.2 µg; L-carnitine 180 mg; taurine 350 mg.
  • Background nutrition: Standard continuous non–n-3-enriched enteral formula given separately under a protocolised algorithm for gastrointestinal intolerance; study bolus given even if continuous feeds were interrupted provided enteral medications were tolerated.
  • Co-interventions: Simplified ARDSNet lung-protective ventilation and fluid-conservative haemodynamic protocols; insulin protocols targeting blood glucose 80–150 mg/dL; semirecumbent positioning.
• Comparison:
  • Delivery: Twice-daily isocaloric/isovolaemic control boluses of 120 mL (total 240 mL/day), identical in appearance and smell to the intervention.
  • Composition (per day; 240 mL): 474 kcal; fat 22 g; carbohydrate 51.8 g; protein 20 g; EPA/DHA/GLA 0 g; vitamin C 76 mg; vitamin E 12 IU; zinc 5.6 mg; selenium 18 µg; L-carnitine 38 mg; taurine 138 mg.
  • Background nutrition and co-interventions: Same protocolised enteral feeding approach and ARDSNet-style co-interventions as the intervention group.
• Blinding: Double-blind (participants, clinicians, investigators, and outcome assessors); randomisation via a central web-based system with stratification by hospital and baseline shock.
• Statistics: Maximum planned sample size 1000 with 4 interim analyses; 90.7% power to detect a 2.25-day increase in VFDs (assumed mean 14 days; SD 10.5) using a group sequential design with asymmetric efficacy/futility boundaries (two-sided alpha 0.05; beta 0.093); primary analysis was intention-to-treat.
• Follow-Up Period: Outcomes followed to day 60 or hospital discharge; primary endpoint assessed to day 28.

Key Results

This trial was stopped early. Stopped for futility at the first interim analysis after 272 patients were randomised (143 n-3 supplement; 129 control) because ventilator-free days and mortality crossed the prespecified futility boundary.

• Despite biochemical uptake (plasma EPA increased almost 8-fold), the intervention did not improve oxygenation and was associated with fewer VFDs (14.0 vs 17.2; difference −3.2 days; 95% CI −5.8 to −0.7; P=0.02).
• The direction of effect was consistently unfavourable for ICU-free days (14.0 vs 16.7; difference −2.7 days; 95% CI −5.1 to −0.3; P=0.04) and nonpulmonary organ failure–free days (12.3 vs 15.5; difference −3.2 days; 95% CI −5.9 to −0.5; P=0.02).
• Unadjusted 60-day mortality was higher with n-3 supplementation (26.6% vs 16.3%; difference 10.3%; 95% CI 0.7 to 19.9; P=0.054), but uncertainty remains given early stopping and lack of multiplicity correction.

Internal Validity

• Randomisation and allocation concealment: Central web-based randomisation with stratification by hospital and baseline shock; double-blind design reduces selection and ascertainment bias.
• Dropout/exclusions after randomisation: Complete follow-up to day 60/hospital discharge in all 272 participants; 1/143 patients randomised to n-3 received no n-3 doses.
• Performance/detection bias: Intervention and control boluses were identical in appearance/smell; outcomes largely objective (VFDs, ICU-free days, mortality), reducing detection bias.
• Protocol adherence: Mean receipt of 85% of planned twice-daily bolus doses in both groups; background ventilation and fluid management were protocolised.
• Baseline characteristics: Groups were broadly similar (APACHE III scores similar; pneumonia and sepsis were common aetiologies), but there were small imbalances (minute ventilation 11.4 [SD 3.1] vs 10.6 [SD 2.9] L/min; P=0.04; trend to higher prestudy fluid intake 5085 vs 4387 mL; P=0.09) that may have favoured control.
• Heterogeneity: Multicentre (44 hospitals) increases practice variability, but co-interventions were standardised; there was no evidence of interaction with EDEN feeding allocation (interaction P=0.47).
• Timing: Enrolment occurred within 48 hours of ALI onset; bolus supplementation began within 6 hours of randomisation, aligning with an “early intervention” biological premise.
• Dose: Delivered a high-fat bolus (44.6 g fat/day) rather than continuous infusion; the intervention increased diarrhoea (29% vs 21%; P=0.001), raising a plausible pathway for reduced tolerance/absorption or indirect harms.
• Separation of the variable of interest: Clear biochemical separation (plasma EPA increased almost 8-fold; plasma EPA:arachidonic acid ratio increased); percent EPA rose from ~0.5% at baseline to ~3.5% by day 3 in the intervention subset, while remaining ~0.5% in controls.
• Key delivery aspects: The intervention was dissociated from baseline enteral feeding (bolus dosing even when continuous feeds were interrupted), supporting intention-to-treat evaluation despite feeding intolerance.
• Outcome assessment: VFDs and ICU-free days are clinically relevant and prespecified; however, P values were not corrected for multiple comparisons or early stopping, and early termination reduces estimate stability.
• Statistical rigor: Prespecified group sequential monitoring with asymmetric stopping boundaries; trial stopped at first interim analysis for futility, increasing uncertainty around the magnitude (and even direction) of some effects, particularly mortality.

Conclusion on Internal Validity: Moderate to strong: randomisation, concealment, blinding, protocolised co-interventions, objective outcomes, and excellent follow-up support internal validity, but early stopping and the non-inert nutritional composition differences between intervention and control boluses complicate attribution and precision of effect estimates.

External Validity

• Population representativeness: Typical early ALI/ARDS population (mechanically ventilated within 48 hours of onset) in a large multicentre network; common aetiologies included pneumonia (52%) and sepsis (23%).
• Key exclusions: Many screened patients were excluded for severe chronic lung disease, delayed presentation (ALI >48 h or intubated >72 h), consent barriers, and severe comorbidity, which may limit generalisability to frailer or late-presenting cohorts.
• Intervention generalisability: Findings apply most directly to twice-daily bolus supplementation dissociated from baseline feeds; extrapolation to continuous “immune-modulating” enteral formulas (different delivery kinetics) or to parenteral omega-3 lipid emulsions is uncertain.
• Health-system applicability: Protocolised ARDSNet practice may differ from lower-resource settings, but the intervention itself is pragmatic and widely deliverable where enteral feeding is available.

Conclusion on External Validity: Moderate: results are broadly applicable to early, mechanically ventilated ALI/ARDS patients receiving enteral nutrition in well-resourced ICUs, but may not fully generalise to different formulations, continuous delivery strategies, or later-phase ARDS.

Strengths & Limitations

• Strengths:
  • Randomised, double-blind, multicentre ARDSNet trial with concealed allocation and excellent follow-up.
  • Protocolised co-interventions (ventilation, haemodynamic/fluid strategy, glucose targets), reducing confounding from practice variation.
  • Demonstrated pharmacodynamic separation (plasma EPA increased almost 8-fold), supporting that “lack of delivery” was not the primary explanation for negative results.
  • Novel bolus strategy allowed intention-to-treat assessment even when continuous feeds were interrupted.
• Limitations:
  • Stopped early for futility after 272/1000 planned participants, reducing precision and stability of treatment effect estimates (especially for mortality).
  • Intervention and control boluses were isocaloric but not macronutrient-matched (control provided up to 20 g/day additional protein and substantially more carbohydrate), complicating attribution to n-3/GLA/antioxidants vs broader nutritional effects.
  • Bolus delivery of a high-fat supplement may increase gastrointestinal intolerance (diarrhoea 29% vs 21%) and may not reflect continuous immunomodulatory formula delivery used in earlier trials.
  • P values were not adjusted for multiple comparisons or early stopping.

Interpretation & Why It Matters

• Clinical practice

  Twice-daily bolus supplementation with EPA/DHA, GLA, and antioxidants should not be used routinely in early ALI/ARDS given fewer ventilator-free days (14.0 vs 17.2; P=0.02) and a mortality signal favouring control (26.6% vs 16.3%; P=0.054).
• Mechanistic inference

  Biochemical incorporation (EPA increased almost 8-fold) did not translate into improved oxygenation or clinical outcomes, highlighting limits of surrogate-driven nutrition “pharmaconutrition” in heterogeneous ARDS biology.
• Trial-methods lesson

  Large multicentre, blinded RCTs can overturn strong signals from small single-centre nutrition trials; comparator composition and delivery method are integral to interpreting nutrition interventions.

Controversies & Other Evidence

• Early termination for futility reduces precision and can yield unstable effect estimates; the editorial emphasised that the nominally significant morbidity endpoints occurred in the context of many tested endpoints and early stopping, warranting caution in interpreting “harm” magnitude.¹
• Discordant findings versus prior “immune-modulating” enteral formula trials raised debate about whether differences in delivery strategy (bolus dissociated from feeds vs continuous complete formula), feeding tolerance, and control formula choice could explain divergent results.¹
• Correspondence highlighted that the OMEGA findings contrasted with earlier sepsis/ARDS literature supporting n-3 interventions, underscoring potential context-dependence (population, timing, delivery, comparator) and the need for careful interpretation across heterogeneous nutrition trials.²³
• Subsequent systematic reviews/meta-analyses focusing on ARDS/ALI have reported heterogeneous results; pooled estimates have variably suggested improved oxygenation and shorter ventilation/ICU stay in some analyses, but mortality effects are inconsistent and sensitive to trial selection, delivery mode, and small-study effects.⁴⁵⁶
• Guideline synthesis has incorporated OMEGA as a key negative multicentre trial and has not supported routine use of enteral formulas/supplements enriched with fish oils/GLA/antioxidants for ARDS outside research contexts, highlighting uncertainty, heterogeneity, and potential confounding by macronutrient delivery differences.⁷

Summary

• OMEGA randomised 272 adults with early ALI at 44 hospitals to a twice-daily bolus n-3/GLA/antioxidant supplement vs an isocaloric control bolus, alongside protocolised enteral feeding.
• The trial stopped early for futility at the first interim analysis after 272 participants.
• The intervention resulted in fewer ventilator-free days (14.0 vs 17.2; difference −3.2 days; 95% CI −5.8 to −0.7; P=0.02) and fewer ICU-free and organ failure–free days.
• There was a mortality signal favouring control (26.6% vs 16.3%; P=0.054), but estimates are imprecise due to early stopping and multiplicity considerations.
• Biochemical separation was confirmed (plasma EPA increased almost 8-fold), and diarrhoea was more frequent with the n-3 supplement (29% vs 21%; P=0.001).

Overall Takeaway

The OMEGA trial is a landmark ARDS nutrition study because it rigorously tested a biologically compelling “pharmaconutrition” intervention in a large multicentre, blinded, intention-to-treat design and found no benefit—indeed, worse morbidity outcomes—with a signal towards harm. It shifted practice and guideline thinking away from routine enteral n-3/GLA/antioxidant supplementation in early ALI/ARDS, and it refocused the field on the importance of delivery strategy, comparator composition, and high-quality evidence before widespread adoption.

Bibliography

• 1.Cook DJ, Heyland DK. Pharmaconutrition in acute lung injury. JAMA. 2011;306(14):1599-1600.
• 2.König K, Rosenberger P, Mirakaj V. Supplementation in acute lung injury. JAMA. 2012;307(2):144-146.
• 3.Rice TW, Wheeler AP, Thompson BT. Supplementation in acute lung injury—Reply. JAMA. 2012;307(2):144-146.
• 4.Zhu D, Zhang Y, Li S, Gan L, Feng H, Nie W. Enteral omega-3 fatty acid supplementation in adult patients with acute respiratory distress syndrome: a systematic review of randomised controlled trials with meta-analysis and trial sequential analysis. Intensive Care Med. 2014;40(4):504-512.
• 5.Li C, Bo L, Liu W, Lu X, Jin F. Enteral immunomodulatory diet (omega-3 fatty acid, γ-linolenic acid and antioxidant supplementation) for acute lung injury and acute respiratory distress syndrome: an updated systematic review and meta-analysis. Nutrients. 2015;7(7):5572-5585.
• 6.Langlois PL, D'Aragon F, Hardy G, Manzanares W. Omega-3 polyunsaturated fatty acids in critically ill patients with acute respiratory distress syndrome: a systematic review and meta-analysis. Nutrition. 2019;61:84-92.
• 7.Singer P, Blaser AR, Berger MM, Alhazzani W, Calder PC, Casaer MP, et al. ESPEN guideline on clinical nutrition in the intensive care unit. Clin Nutr. 2019;38(1):48-79.