Skip to main content

Foundational Trials

RE-ENERGIZE

Image: Shutterstock / luchschenF

Publication

  • Title: A randomized trial of enteral glutamine for treatment of burn injuries
  • Acronym: RE-ENERGIZE
  • Year: 2022
  • Journal published in: New England Journal of Medicine
  • Citation: Heyland DK, Wibbenmeyer L, Pollack JA, et al. A randomized trial of enteral glutamine for treatment of burn injuries. N Engl J Med. 2022;387(11):1001-10.

Context & Rationale

  • Background
    • Major thermal injury produces profound inflammation, hypermetabolism, and catabolism, with high risks of infection, organ dysfunction, prolonged hospitalisation, and death.
    • Glutamine is a key nitrogen carrier and precursor for nucleotides and glutathione, and is biologically plausible as a “conditionally essential” amino acid during severe stress (gut barrier, immune cell function, redox balance).
    • Early, small burn trials suggested glutamine supplementation could reduce infectious morbidity (including Gram-negative bacteraemia) and possibly mortality.
    • In broader critically ill populations, subsequent large trials of glutamine-containing strategies raised concern for harm in some settings, creating genuine uncertainty about benefit–risk in burns.
    • Practice variation persisted internationally, and burn clinicians required a definitive, adequately powered, blinded trial in contemporary burn care.
  • Research Question/Hypothesis
    • Does early enteral L-glutamine (0.5 g/kg/day) compared with placebo reduce time to live hospital discharge (up to 90 days) in adults with severe burn injury?
    • Key mechanistic/clinical hypotheses included improved survival and reduced acquired Gram-negative bacteraemia, with better longer-term function and health-related quality of life among survivors.
  • Why This Matters
    • Enteral glutamine had become embedded in some burn nutrition pathways despite uncertain net benefit and evolving safety signals from non-burn ICU trials.
    • A large, double-blind, international RCT could resolve whether routine supplementation improves patient-centred outcomes or should be abandoned.
    • The findings were expected to influence nutrition guidelines, resource allocation, and standard burn centre protocols globally.

Design & Methods

  • Research Question: In adults with severe thermal burns, does enteral L-glutamine (0.5 g/kg/day) versus placebo reduce time to discharge alive from hospital (up to 90 days)?
  • Study Type:
    • Randomised, parallel-group, placebo-controlled, double-blind, investigator-initiated international trial.
    • Multicentre: 54 burn centres in 14 countries.
    • Recruitment: March 2011 to June 2020.
    • Setting: specialised burn acute care units (including ICU-level care) with established enteral feeding pathways.
  • Population:
    • Adults (≥18 years) with deep second-degree and/or third-degree burns expected to require grafting, enrolled within 72 hours after hospital admission.
    • Total body-surface area (TBSA) eligibility thresholds by age: 18–39 years (TBSA ≥20% or TBSA ≥15% with inhalation injury); 40–59 years (TBSA ≥15%); ≥60 years (TBSA ≥10%).
    • Key exclusions included baseline renal dysfunction (pre-specified creatinine/urine-output thresholds), injuries from high-voltage electrical contact, moribund patients not expected to survive 72 hours, liver cirrhosis, contraindication to enteral nutrition, pregnancy/lactation, extreme BMI, and prior glutamine exposure before randomisation (per protocol exclusions).
  • Intervention:
    • L-glutamine powder administered enterally at a total dose of 0.5 g/kg/day (obesity-adjusted body weight when BMI ≥35).
    • Delivered via nasogastric/enteral tube (or orally if appropriate), started soon after randomisation and continued until 7 days after the last skin grafting procedure, discharge from the acute care unit, or 3 months after hospital admission (whichever occurred first).
    • Study product supplied as blinded sachets, mixed with liquid/food for administration.
  • Comparison:
    • Matching placebo (maltodextrin) administered enterally on the same schedule and in the same manner as the intervention.
    • All other burn care (including nutrition support targets and co-interventions) delivered according to local practice within participating centres.
  • Blinding:
    • Double-blind (participants, clinicians, investigators, and outcome assessors), with pharmacy-controlled allocation and indistinguishable study products.
    • Blinding mitigated differential co-intervention and discharge decision-making between groups.
  • Statistics:
    • Power calculation: A sample size of 1,200 patients was planned to provide 80% power (two-sided α=0.05) to detect a clinically important difference in time to discharge alive, based on simulation assumptions including a 20% relative reduction in 90-day mortality and a 20% relative increase in the daily discharge rate among 90-day survivors.
    • Analysis type: Modified intention-to-treat (patients receiving study product) for primary analyses; time-to-event methods accounting for competing risk of death (subdistribution hazard ratio), with pre-specified sensitivity analyses.
    • Per reporting plan, p-values were emphasised for the primary outcome and serious adverse events; secondary outcomes were generally reported with confidence intervals.
  • Follow-Up Period:
    • Primary endpoint: up to 90 days after randomisation (time to discharge alive, with death as competing risk).
    • Mortality and patient-centred outcomes assessed to 6 months after randomisation.

Key Results

This trial was not stopped early. Recruitment proceeded to the revised target sample size (an interim safety analysis was planned at approximately 600 participants).

Outcome Glutamine Placebo Effect p value / 95% CI Notes
Time to discharge alive from hospital (up to 90 days; competing risk of death) 40 days (IQR 24–87) 38 days (IQR 22–75) Subdistribution HR 0.91 95% CI 0.80 to 1.04; P=0.17 Primary outcome; values <1 favour placebo (slower discharge with glutamine)
Mortality at 6 months 17.2% 16.2% HR 1.06 95% CI 0.80 to 1.41; P=Not reported Time-to-event analysis over 6 months
In-hospital mortality 91/596 (15.3%) 84/604 (13.9%) Adjusted RR 0.96 95% CI 0.79 to 1.16; P=Not reported Adjusted model per trial report
Acquired Gram-negative bacteraemia (within 90 days) 113/596 (19.0%) 109/604 (18.0%) Adjusted RR 0.97 95% CI 0.80 to 1.16; P=Not reported Pre-specified tertiary endpoint linked to gut-barrier hypothesis
Days in hospital (excluding withdrawals; truncated at 90 days) 32 (IQR 20–53) 30 (IQR 18–53) Adjusted mean diff 0.92 days 95% CI −1.57 to 3.41; P=Not reported Length-of-stay summary reported with unadjusted and adjusted mean differences
Time to live discharge from acute care unit (up to 90 days) 36 days (IQR 21–81.5) 35 days (IQR 19–67) Adjusted HR 0.91 95% CI 0.79 to 1.03; P=Not reported Values <1 favour placebo
Time to live extubation (baseline mechanically ventilated subgroup) 17 days (IQR 7–29) 15 days (IQR 6–28) Adjusted HR 0.91 95% CI 0.77 to 1.08; P=Not reported Analysed among those ventilated at baseline; values <1 favour placebo
SF-36 Physical Component Scale at 6 months (survivors assessed) 42.3 ± 11.2 (n=274) 42.4 ± 11.9 (n=290) Mean diff 0.10 95% CI −1.71 to 1.91 No clinically important between-group difference
Activities of daily living (Katz) at 6 months (survivors assessed) 5.3 ± 1.5 (n=293) 5.3 ± 1.5 (n=301) Mean diff −0.01 95% CI −0.24 to 0.23 Scores range 0–6; higher scores indicate greater independence
Patients with any serious adverse event (SAE) 26/596 (4.4%) 37/604 (6.1%) Not reported P=0.20 Fisher’s exact test; SAEs largely deemed not/unlikely related; one acute kidney injury possibly related in glutamine arm
Average urea levels (mmol/L; median [Q1–Q3]) 8.7 [5.9–15.6] 6.8 [4.8–12.2] Not reported Not reported Higher urea consistent with increased nitrogen load
  • Enteral glutamine did not shorten time to discharge alive (median 40 vs 38 days; subdistribution HR 0.91; 95% CI 0.80 to 1.04; P=0.17).
  • Mortality and the hypothesised infection endpoint (acquired Gram-negative bacteraemia) were not improved, with effect estimates close to null.
  • Serious adverse events were not increased (26 vs 37 patients with any SAE; P=0.20), although biochemical safety monitoring showed higher urea in the glutamine arm.

Internal Validity

  • Randomisation and allocation:
    • Central randomisation with allocation concealment; study product preparation by pharmacy enabled robust blinding.
    • Balance achieved across key baseline variables (e.g., age 49.2 ± 17.1 vs 48.7 ± 17.6 years; TBSA 33.1 ± 17.2% vs 31.9 ± 17.1%; mechanical ventilation 40.6% vs 39.6%).
  • Drop out or exclusions:
    • 1,209 patients consented and randomised; 9 withdrew consent early and did not receive study product; 1,200 received study product and were included in the modified intention-to-treat analyses (596 vs 604).
    • Primary endpoint ascertainment was high: known discharge time or death in 587/596 (98.5%) vs 597/604 (98.8%).
    • Six-month patient-centred follow-up was incomplete (e.g., SF-36 available in 274 vs 290; ADL/IADL assessed in 293 vs 301), introducing potential attrition bias for long-term functional outcomes.
  • Performance/Detection Bias:
    • Double blinding limited differential co-interventions and differential outcome assessment.
    • Primary outcome (time to live discharge) is operational and routinely recorded, reducing subjective measurement bias.
  • Protocol Adherence:
    • Early initiation: time from randomisation to first dose 5.9 ± 19.6 hours (glutamine) vs 5.1 ± 9.5 hours (placebo).
    • Exposure intensity: proportion of prescribed doses received 91.3 ± 13.6% vs 91.0 ± 13.5%.
    • Duration: 27.0 ± 22.4 days vs 25.6 ± 22.7 days (range included 0 days in both groups, consistent with early death/withdrawal in some participants).
  • Baseline Characteristics:
    • Groups were clinically comparable at baseline (sex ~73% male; APACHE II 14.0 ± 7.0 vs 13.8 ± 6.8; inhalation injury 33.4% vs 34.3%).
    • Severity profile suggests a cohort at meaningful risk of complications and death (mean TBSA ~32–33%; ~40% ventilated at baseline).
  • Heterogeneity:
    • Wide geographic participation (54 sites, 14 countries) increases clinical heterogeneity but reduces centre-specific effects.
    • Analytic strategy incorporated competing-risk methods and pre-specified sensitivity approaches; subgroup modelling did not suggest effect modification by age or TBSA (interaction tests largely non-significant).
  • Timing:
    • Time from acute care unit admission to randomisation: 43.7 ± 21.6 hours vs 42.8 ± 21.9 hours.
    • Time from burn injury to initiation of study agent: 60.8 ± 25.1 hours vs 59.5 ± 25.6 hours.
  • Dose:
    • Enteral L-glutamine 0.5 g/kg/day (obesity-adjusted weight when BMI ≥35), continued for a prolonged clinically relevant period (mean ~26 days) and aligned to grafting/acute care milestones.
  • Separation of the Variable of Interest:
    • High separation in study-product exposure (91% dose receipt in both groups) with matched placebo administration.
    • Delivered nutrition among mechanically ventilated patients was broadly similar: energy adequacy 74.2 ± 25.4% vs 71.9 ± 25.6%; protein adequacy 75.3 ± 22.8% vs 72.3 ± 22.3%.
  • Key Delivery Aspects:
    • Pragmatic delivery across burn centres with early dosing and sustained administration, maximising the chance of detecting a clinically meaningful effect if present.
    • Co-interventions were monitored (nutrition delivery metrics, renal replacement therapy exposure, and selected concomitant medications per protocol), supporting interpretability of group comparisons.
  • Outcome Assessment:
    • Primary outcome used competing-risk methods to account for death precluding discharge, a statistically appropriate approach for burn recovery trajectories.
    • Secondary outcomes included both in-hospital endpoints and 6-month patient-centred measures (SF-36 and ADL/IADL), though longer-term outcomes had substantial missingness.
  • Statistical Rigor:
    • Primary analysis aligned with the pre-specified approach: competing-risk time-to-event methods with subdistribution hazard ratios and a formal primary p-value.
    • Confidence intervals for key endpoints were narrow enough to exclude large clinically important benefit for the primary endpoint (subdistribution HR 0.91; 95% CI 0.80 to 1.04).

Conclusion on Internal Validity: Overall, internal validity appears strong, supported by robust blinding, central randomisation, high adherence, near-complete ascertainment of the primary endpoint, and appropriate competing-risk modelling; the main internal threat is incomplete follow-up for 6-month patient-reported outcomes.

External Validity

  • Population Representativeness:
    • Adults with substantial burn severity (mean TBSA ~32–33%) treated in specialised burn centres, including a high proportion requiring mechanical ventilation (~40%).
    • Large, international recruitment (14 countries) supports broad representativeness for high-resource and mixed-resource burn systems.
    • Important exclusions (e.g., baseline renal dysfunction, high-voltage electrical injuries, moribund patients, liver cirrhosis, extremes of BMI) limit applicability to these subgroups.
  • Applicability:
    • Findings apply to adult burn patients in centres where early enteral nutrition is feasible and where glutamine would be delivered enterally at 0.5 g/kg/day.
    • Generalisation to paediatric burns, to patients with established renal failure/dialysis at baseline, or to parenteral glutamine strategies is limited.
    • Long recruitment period (2011–2020) spans evolving burn practices, but multicentre participation enhances applicability to contemporary real-world care variation.

Conclusion on External Validity: External validity is moderate-to-strong for adult severe thermal burns managed in specialised burn centres with early enteral feeding, but is limited for patients excluded by design (notably baseline renal dysfunction) and for settings without reliable enteral nutrition delivery.

Strengths & Limitations

  • Strengths:
    • Largest double-blind, placebo-controlled RCT evaluating glutamine supplementation in adult burn patients (n=1,200).
    • International, multicentre design (54 sites; 14 countries) improves generalisability and reduces single-centre bias.
    • Pragmatic administration strategy with early initiation, long duration, and high adherence (≈91% of prescribed doses received).
    • Patient-centred primary endpoint incorporating survival and hospital recovery trajectory, analysed with appropriate competing-risk methods.
    • Extended follow-up to 6 months with health-related quality of life and functional measures among survivors.
  • Limitations:
    • Incomplete 6-month follow-up for patient-reported and functional outcomes (SF-36 available in ~46–48% of dosed participants), which may bias long-term comparisons.
    • Primary endpoint depends on discharge processes that vary by system and centre; competing-risk methods address death as a competing event but cannot eliminate practice-variation effects.
    • Baseline glutamine status (e.g., plasma glutamine concentration) was not used to select/enrich participants, so targeted benefit in a deficient subgroup could not be directly evaluated.
    • Exclusion of baseline renal dysfunction improves safety but reduces applicability to an important high-risk subgroup.

Interpretation & Why It Matters

  • Clinical practice
    • Routine enteral glutamine supplementation (0.5 g/kg/day) in adult severe burns did not improve time to discharge alive, survival, or key infection outcomes, and should not be adopted as standard care on efficacy grounds.
    • Nutrition priorities should remain early enteral feeding, adequate protein/energy delivery, and multidisciplinary burn rehabilitation rather than single-nutrient supplementation without demonstrated benefit.
  • Trial methodology
    • RE-ENERGIZE demonstrates the importance of large, blinded, pragmatic trials to validate effects suggested by small studies in highly specialised populations.
    • Competing-risk approaches for discharge outcomes are essential in burns because death precludes discharge and length-of-stay distributions are strongly non-normal.
  • Biological interpretation
    • The absence of clinical benefit despite biologic plausibility implies that either glutamine is not rate-limiting in recovery for most severe burn patients, or that any benefit is confined to an unselected subgroup not captured by this broad pragmatic design.

Controversies & Subsequent Evidence

  • Earlier burn RCT evidence suggested clinically important reductions in infection and mortality with enteral glutamine; RE-ENERGIZE, as the largest and most rigorous burn trial, substantially down-weights the plausibility of large benefit in unselected patients.1
  • Safety concerns about glutamine in heterogeneous ICU populations (particularly at high doses and in patients with baseline renal dysfunction) informed burn-trial safety design and interpretation, including renal-dysfunction exclusions and biochemical monitoring.23
  • Updated burn-specific systematic reviews/meta-analyses incorporating RE-ENERGIZE report diminished or uncertain mortality benefit and highlight persistent heterogeneity and small-study effects in the pre-RE-ENERGIZE literature.45
  • Guideline trajectory shifted away from routine glutamine supplementation in critical illness; pre-trial burn recommendations that considered glutamine now sit in tension with post-trial evidence and should be interpreted cautiously in light of RE-ENERGIZE.67

Summary

  • In 1,200 adults with severe burns, enteral L-glutamine 0.5 g/kg/day did not reduce time to discharge alive (median 40 vs 38 days; subdistribution HR 0.91; 95% CI 0.80 to 1.04; P=0.17).
  • Six-month mortality was similar (17.2% vs 16.2%; HR 1.06; 95% CI 0.80 to 1.41).
  • No improvement was seen in acquired Gram-negative bacteraemia (19.0% vs 18.0%; adjusted RR 0.97; 95% CI 0.80 to 1.16) or in-hospital mortality (15.3% vs 13.9%; adjusted RR 0.96; 95% CI 0.79 to 1.16).
  • Health-related quality of life and functional outcomes at 6 months were similar among assessed survivors (e.g., SF-36 Physical Component Scale mean difference 0.10; 95% CI −1.71 to 1.91).
  • Serious adverse events were not increased (26 vs 37 patients with any SAE; P=0.20), though urea levels were higher in the glutamine arm (median 8.7 vs 6.8 mmol/L).

Overall Takeaway

RE-ENERGIZE is a landmark burn nutrition trial because it definitively tested a widely adopted, biologically plausible supplementation strategy in a large, double-blind, international cohort and found no meaningful clinical benefit. The results strongly argue against routine enteral glutamine supplementation in unselected adults with severe thermal burns, refocusing practice on proven elements of burn critical care and high-quality nutrition delivery rather than single-nutrient adjuncts.

Overall Summary

  • Enteral glutamine (0.5 g/kg/day) did not improve time to discharge alive, mortality, or Gram-negative bacteraemia in adults with severe burns.
  • High protocol adherence and rigorous blinding make a large hidden benefit unlikely.
  • These findings should reset burn nutrition practice away from routine glutamine supplementation.

Bibliography