Publication

• Title: Effect of Lower Tidal Volume Ventilation Facilitated by Extracorporeal Carbon Dioxide Removal vs Standard Care Ventilation on 90-Day Mortality in Patients With Acute Hypoxemic Respiratory Failure: The REST Randomized Clinical Trial
• Acronym: REST (pRotective vEntilation with veno-venouS lung assisT in respiratory failure)
• Year: 2021
• Journal published in: JAMA
• Citation: McNamee JJ, Gillies MA, Barrett NA, et al; REST Investigators. Effect of Lower Tidal Volume Ventilation Facilitated by Extracorporeal Carbon Dioxide Removal vs Standard Care Ventilation on 90-Day Mortality in Patients With Acute Hypoxemic Respiratory Failure: The REST Randomized Clinical Trial. JAMA. 2021;326(11):1013-1023.

Context & Rationale

• Background

  • Lung-protective ventilation (lower tidal volumes) improves outcomes in acute respiratory failure populations, but lung injury may persist when mechanical stress (plateau/driving pressure, mechanical power) remains high.¹
  • “Ultraprotective” ventilation (very low tidal volumes, e.g., ~3 mL/kg predicted body weight) is physiologically attractive but frequently induces severe hypercapnia and acidaemia.
  • Low-flow veno-venous extracorporeal carbon dioxide removal (VV-ECCO₂R) can remove CO₂, potentially enabling ultraprotective ventilation while maintaining acceptable pH, but introduces invasive cannulation and anticoagulation-related risks.
  • Pre-REST evidence for VV-ECCO₂R in ARDS/AHRF was dominated by feasibility studies and small trials, with persistent uncertainty regarding net benefit vs harm at patient-centred outcomes.²
• Research Question/Hypothesis

  • Does lowering tidal volume during invasive mechanical ventilation using VV-ECCO₂R (to facilitate ultraprotective ventilation) reduce 90-day mortality compared with standard lung-protective ventilation in acute hypoxaemic respiratory failure?
• Why This Matters

  • VV-ECCO₂R is complex, resource-intensive, and carries material bleeding/device risks; any clinically meaningful benefit must justify these harms and costs.
  • If VV-ECCO₂R-enabled ultraprotective ventilation is neutral or harmful, it should not diffuse into routine practice outside trials.

Design & Methods

• Research Question: In invasively ventilated adults with acute hypoxaemic respiratory failure, does ultraprotective ventilation facilitated by VV-ECCO₂R reduce 90-day all-cause mortality compared with standard lung-protective ventilation?
• Study Type: Multicentre, pragmatic, open-label, randomised, parallel-group clinical trial; 51 adult ICUs (United Kingdom); 1:1 allocation; stratified by site.
• Population:
  • Setting: Adult ICU; invasive mechanical ventilation.
  • Key inclusion features: Age ≥16 years; acute and potentially reversible cause of acute respiratory failure; invasive ventilation with PEEP ≥5 cmH₂O; within 48 hours of onset of qualifying hypoxaemia defined by PaO₂/FiO₂ ≤20 kPa (≤150 mmHg) on 2 arterial blood gases separated by ≥6 hours (48-hour consent clock commenced at the second qualifying blood gas); randomisation occurred 8–24 hours after the second qualifying blood gas.
  • Key exclusions (examples): Invasive ventilation ≥7 days prior to randomisation; ability to maintain tidal volume ≤3 mL/kg predicted body weight with pH ≥7.2 without VV-ECCO₂R; receipt of (or decision to commence) ECMO within 24 hours; use of HFOV or APRV; untreated pulmonary embolism/pleural effusion/pneumothorax as primary cause; acute respiratory failure fully explained by left ventricular failure or fluid overload; contraindication to limited systemic anticoagulation with heparin; inability to obtain central venous access (IJ/femoral) or IVC filter (if femoral access); treatment withdrawal imminent within 24 hours; not expected to survive 90 days due to premorbid illness; DNAR order in place; severe chronic respiratory disease requiring domiciliary ventilation (except sleep-disordered breathing); severe chronic liver disease (Child-Pugh score >11); platelet count <40,000/mm³ prior to catheter insertion; prior REST enrolment; prisoners.
• Intervention:
  • VV-ECCO₂R (Hemolung Respiratory Assist System) plus a protocolised ultraprotective ventilatory strategy.
  • Ventilation targets (key): Tidal volume target ≈3 mL/kg predicted body weight; plateau pressure target ≤25 cmH₂O; ventilator parameters adjusted in tandem with CO₂ removal to maintain physiologic acceptability (notably pH), while maintaining lung-protective oxygenation strategies.
  • Duration: VV-ECCO₂R intended as time-limited support (maximum 7 days; typical observed duration shorter).
• Comparison:
  • Standard-care ventilation without VV-ECCO₂R (lung-protective ventilation; typical tidal volume target ≈6 mL/kg predicted body weight, with standard plateau pressure constraints), with co-interventions and weaning managed according to usual ICU practice.
• Blinding: Unblinded (device-based intervention); primary endpoint (mortality) is objective, but ventilator-dependent secondary outcomes are susceptible to performance effects.
• Statistics: Planned sample size 1120 to detect a 9% absolute reduction in 90-day mortality (from 41% to 32%) with 90% power at the 5% significance level (two-sided); primary analysis intention-to-treat; effect estimates reported as relative risk for binary outcomes and mean difference for continuous outcomes (with 95% CIs as reported).
• Follow-Up Period: Primary endpoint at 90 days; key secondary outcomes to day 28 (ventilator-free days, ICU/hospital outcomes), with adverse event reporting during the index admission.

Key Results

This trial was stopped early. Recruitment was terminated on the recommendation of the data monitoring and ethics committee for futility and feasibility after 412 participants were randomised (405 included in the primary analysis).

• VV-ECCO₂R enabled lower delivered tidal volumes (day 2: 4.5 (SD 1.6) vs 6.5 (SD 0.8) mL/kg predicted body weight) and lower plateau pressure (day 2: 23.5 (SD 4.4) vs 25.2 (SD 4.7) cmH₂O), but did not improve 90-day survival.
• Secondary outcomes favoured standard care for ventilator-free days, and the VV-ECCO₂R strategy was associated with substantially more adverse and serious adverse events, including intracranial haemorrhage.
• Across prespecified subgroups (e.g., ARDS status, baseline PaO₂/FiO₂, plateau pressure, driving pressure, APACHE II), interaction tests were not significant and confidence intervals were wide (imprecision consistent with early stopping and limited sample size).

Internal Validity

• Randomisation and allocation: Central randomisation with stratification by site (mitigates selection bias); baseline balance was good across key prognostic factors (e.g., median PaO₂/FiO₂ 118.1 (IQR 96.0–134.3) vs 115.5 (IQR 93.8–132.8) mmHg; APACHE II 19 (IQR 15–23) vs 20 (IQR 16–23); SOFA 10 (IQR 7–12) in both arms).
• Dropout/exclusions: Primary analysis included 405/412 randomised (200 vs 205), indicating low missingness for the primary endpoint; differential attrition is unlikely to explain the neutral mortality result.
• Performance/detection bias: Unblinded delivery is unavoidable for VV-ECCO₂R; mortality is objective, but ventilator-free days and weaning-related endpoints are vulnerable to differences in sedation, ventilator mode, and clinician behaviour.
• Protocol adherence / treatment receipt: 186/202 (92.1%) allocated to VV-ECCO₂R received it; 1 patient in the control group received non-protocol VV-ECCO₂R for 2 days (minimal crossover).
• Separation of the variable of interest (ventilation targets): Clear but incomplete separation was achieved:
  • Tidal volume (mL/kg PBW): Day 2: 4.5 (SD 1.6) vs 6.5 (SD 0.8); Day 3: 4.4 (SD 1.5) vs 6.7 (SD 0.8).
  • Plateau pressure (cmH₂O): Day 2: 23.5 (SD 4.4) vs 25.2 (SD 4.7); Day 3: 22.9 (SD 4.6) vs 24.6 (SD 4.4).
  • Driving pressure (cmH₂O): Day 2: 13.5 (SD 4.1) vs 14.6 (SD 4.0); Day 3: 13.1 (SD 4.3) vs 14.3 (SD 3.8).
  • Minute ventilation (L/min): Day 2: 6.6 (SD 2.1) vs 9.0 (SD 2.2); Day 3: 6.5 (SD 2.0) vs 9.6 (SD 2.5).
  • PaCO₂ (mmHg): Day 2: 60.8 (SD 15.5) vs 56.0 (SD 12.1); Day 3: 59.1 (SD 15.2) vs 54.6 (SD 11.3).
  • pH: Day 2: 7.28 (SD 0.10) vs 7.31 (SD 0.09); Day 3: 7.29 (SD 0.10) vs 7.32 (SD 0.08).
  • PaO₂/FiO₂ (mmHg): Day 2: 124.0 (SD 41.5) vs 137.3 (SD 49.1); Day 3: 131.3 (SD 45.8) vs 150.4 (SD 53.5).
• Timing: Patients were enrolled early (within 48 hours of qualifying hypoxaemia), aligning with a biologically plausible window for VILI modification.
• Dose (VV-ECCO₂R exposure): Mean VV-ECCO₂R duration was 4 (SD 2) days, and delivered tidal volumes remained closer to ~4–4.5 mL/kg PBW than the intended ~3 mL/kg PBW in the overall intervention group, suggesting dose/target attainment limitations may have diluted any potential lung-protective effect.
• Weaning/sedation as a competing mechanism (critical to interpretation): Early ventilator management differed materially between groups (consistent with deeper sedation and reduced spontaneous breathing in the VV-ECCO₂R strategy):
  • Mandatory ventilation: Day 2: 97.5% vs 84.2% of patients; Day 3: 91.4% vs 77.5%.
  • Pressure support ventilation: Day 2: 2.5% vs 15.8%; Day 3: 8.6% vs 22.1%.
  • Neuromuscular blockade use: Day 1: 50.0% vs 34.8%; Day 2: 27.0% vs 14.7%.
  These systematic differences are a credible mechanism for fewer ventilator-free days in the VV-ECCO₂R group, independent of lung recovery (i.e., the intervention may intrinsically delay readiness for spontaneous breathing and extubation because ultralow tidal volumes are typically delivered under controlled ventilation).
• Adjunctive therapies: Early prone positioning differed (day 1: 5.5% vs 13.9%), with less proning in the VV-ECCO₂R group early; later use was similar. This introduces potential confounding for oxygenation trajectories (though mortality remained neutral).
• Outcome assessment: Mortality is robust; ventilator-free days incorporate both death and extubation timing and are therefore sensitive to sedation/ventilation practices and competing risk dynamics.
• Statistical rigour: The trial was underpowered for the originally specified mortality difference due to early stopping; confidence intervals are therefore the key interpretive output (rather than p values alone), particularly for mortality and ECMO progression.

Conclusion on Internal Validity: Moderate. Randomisation and baseline balance support a credible estimate for mortality, but early stopping and unblinded care with systematic differences in sedation/ventilation mode (a plausible direct mediator of ventilator-free days) limit interpretability of weaning-related secondary outcomes.

External Validity

• Population representativeness: Adults with moderate-to-severe acute hypoxaemic respiratory failure treated in general UK ICUs; baseline severity (median PaO₂/FiO₂ ~116–118 mmHg; APACHE II ~19–20; SOFA ~10) is typical of patients considered for escalated ventilatory strategies.
• Applicability to routine practice: VV-ECCO₂R requires specialised equipment, cannulation skills, anticoagulation management, and complication management; transferability is limited to centres with established extracorporeal capability and robust governance.
• Generalisation across settings: Findings are most directly applicable to high-income ICU systems; in resource-limited settings, the risk–resource trade-off is likely even less favourable.
• Case-mix considerations: The enrolled population included both ARDS and non-ARDS acute hypoxaemic respiratory failure (ARDS at enrolment: 59% vs 63%); any subgroup-specific signal remains uncertain due to imprecision.

Conclusion on External Validity: Moderate. The clinical phenotype is common, but the intervention is highly specialised and the observed harm burden makes routine extrapolation beyond controlled settings and experienced centres difficult to justify.

Strengths & Limitations

• Strengths:
  • Pragmatic, multicentre design across 51 ICUs enhances real-world relevance for the intended health system context.
  • Objective primary endpoint (90-day all-cause mortality) with low missingness reduces susceptibility to ascertainment bias.
  • Demonstrated physiologic separation in ventilation variables (lower tidal volume, plateau and driving pressure) verifies that the biological “lever” was at least partially pulled.
  • Detailed reporting of ventilation parameters and adverse events enables granular assessment of efficacy–harm trade-offs.
• Limitations:
  • Stopped early for futility/feasibility: substantial loss of statistical power for the prespecified mortality effect; mortality CI remains wide.
  • Open-label management and systematic differences in controlled ventilation/neuromuscular blockade likely confound ventilator-free days (and any extubation-related outcomes).
  • Ultraprotective “dose” may have been insufficient: delivered tidal volumes were ~4–4.5 mL/kg PBW in the overall intervention group rather than the ~3 mL/kg PBW target, and VV-ECCO₂R duration averaged 4 days.
  • Safety signal: substantial bleeding and device-related harms materially constrain benefit–risk acceptability.

Interpretation & Why It Matters

• Clinical implication

  In a broad UK ICU population with acute hypoxaemic respiratory failure, VV-ECCO₂R-facilitated ultraprotective ventilation did not reduce 90-day mortality and increased serious adverse events; routine use cannot be justified on these data.
• Mechanistic insight

  Lower tidal volumes and pressures were achieved, but CO₂ control was not improved (PaCO₂ remained higher and pH lower on days 2–3), highlighting that real-world ECCO₂R performance and achievable “ultraprotective dose” may be constrained.
• Weaning confounding

  Because ultralow tidal volumes tend to be delivered with controlled ventilation, the VV-ECCO₂R strategy plausibly delayed spontaneous modes and extubation readiness (day 2 pressure support 2.5% vs 15.8%), complicating interpretation of ventilator-free days as a surrogate for pulmonary recovery.

Controversies & Subsequent Evidence

• Early stopping and imprecision: The prespecified mortality effect size was ambitious; stopping at 412 randomised patients left wide uncertainty around clinically meaningful benefit or harm, and post hoc conditional power estimates underscored limited ability to detect modest effects.
• Interpretation of ventilator-free days: The VV-ECCO₂R group received more controlled ventilation and neuromuscular blockade early (day 2 mandatory ventilation 97.5% vs 84.2%; day 1 neuromuscular blockade 50.0% vs 34.8%), creating a credible pathway for delayed weaning/extubation that is not necessarily reflective of worse lung recovery (i.e., “deep sedation to deliver ultralow tidal volumes” competes with a weaning-first approach in standard care).
• Safety as the dominant signal: Serious adverse events were common in VV-ECCO₂R (30.7% vs 8.6%), including intracranial haemorrhage (5.0% vs 1.0%) and non-intracranial bleeding (8.4% vs 1.4%), raising the bar for any future efficacy signal and sharpening requirements for safer devices/anticoagulation strategies.
• Consistency with earlier evidence: Earlier randomised work (e.g., Xtravent) established feasibility of ECCO₂R-facilitated ultraprotective ventilation but did not establish patient-centred benefit; REST provides the clearest signal to date that feasibility does not translate into mortality benefit and that harms are non-trivial.³
• Synthesis of the broader literature: Systematic reviews prior to REST highlighted limited high-quality evidence and emphasised bleeding/device complications as a recurrent theme; REST’s adverse event profile is concordant with that concern.²
• Ongoing uncertainty about “who, when, and how”: Scholarly reviews after early feasibility work argued that ECCO₂R’s clinical promise (if any) hinges on identifying a patient subgroup with high VILI risk and modifiable hypercapnia burden, alongside device/anticoagulation improvements; REST suggests that broad application in AHRF is not beneficial and may be harmful.⁴

Summary

• REST randomised 412 ventilated adults with acute hypoxaemic respiratory failure to VV-ECCO₂R-facilitated ultraprotective ventilation vs standard lung-protective ventilation; the trial stopped early for futility/feasibility.
• 90-day mortality was similar: 41.5% vs 39.5% (RR 1.05; 95% CI 0.83 to 1.33; P=0.68).
• The VV-ECCO₂R strategy achieved lower tidal volumes and pressures (day 2 tidal volume 4.5 vs 6.5 mL/kg PBW; day 2 plateau 23.5 vs 25.2 cmH₂O) but did not improve gas exchange control (higher PaCO₂ and lower pH at days 2–3).
• Ventilator-free days were fewer with VV-ECCO₂R (mean 7.1 vs 9.2; mean difference −2.1 days; 95% CI −3.8 to −0.3; P=0.02), plausibly influenced by deeper sedation/controlled ventilation requirements.
• Harms were substantially greater with VV-ECCO₂R (serious adverse events 30.7% vs 8.6%; intracranial haemorrhage 5.0% vs 1.0%).

Further Reading

Other Trials

• Brower RG, Matthay MA, Morris A, et al; Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342(18):1301-1308.
• Bein T, Weber-Carstens S, Goldmann A, et al. Lower tidal volume strategy (≈3 mL/kg) combined with extracorporeal CO₂ removal versus “conventional” protective ventilation (6 mL/kg) in severe ARDS: the prospective randomised Xtravent-study. Intensive Care Med. 2013;39(5):847-856.
• Guérin C, Reignier J, Richard JC, et al; PROSEVA Study Group. Prone positioning in severe acute respiratory distress syndrome. N Engl J Med. 2013;368(23):2159-2168.
• Peek GJ, Mugford M, Tiruvoipati R, et al; CESAR Trial Collaboration. Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet. 2009;374(9698):1351-1363.
• Combes A, Hajage D, Capellier G, et al; EOLIA Trial Group, REVA, and ECMONet. Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome. N Engl J Med. 2018;378(21):1965-1975.

Systematic Review & Meta Analysis

• Fitzgerald M, Millar J, Blackwood B, et al. Extracorporeal carbon dioxide removal for patients with acute respiratory failure secondary to the acute respiratory distress syndrome: a systematic review. Crit Care. 2014;18(3):222.
• Serpa Neto A, Cardoso SO, Manetta JA, et al. Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. JAMA. 2012;308(16):1651-1659.
• Boyle AJ, Sklar MC, McNamee JJ, et al. Extracorporeal carbon dioxide removal for lowering the risk of mechanical ventilation: research questions and clinical potential for the future. Lancet Respir Med. 2018;6(11):874-884.
• Goligher EC, Damiano P, Talmor D, et al. Precision medicine for mechanical ventilation in acute respiratory distress syndrome: encouraging the conduct of clinical trials. Am J Respir Crit Care Med. 2017;196(5):558-568.

Observational Studies

• Bellani G, Laffey JG, Pham T, et al; LUNG SAFE Investigators; ESICM Trials Group. Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries. JAMA. 2016;315(8):788-800.
• Terragni PP, Rosboch G, Tealdi A, et al. Tidal hyperinflation during low tidal volume ventilation in acute respiratory distress syndrome. Am J Respir Crit Care Med. 2007;175(2):160-166.
• Needham DM, Colantuoni E, Mendez-Tellez PA, et al. Lung protective mechanical ventilation and two year survival in patients with acute lung injury: prospective cohort study. BMJ. 2012;344:e2124.
• Laffey JG, Madotto F, Bellani G, et al; LUNG SAFE Investigators; ESICM Trials Group. Geo-economic variations in epidemiology, patterns of care, and outcomes in patients with acute respiratory distress syndrome: insights from the LUNG SAFE prospective cohort study. Chest. 2020;158(5):1967-1982.

Guidelines

• Slutsky AS, Ranieri VM. Ventilator-induced lung injury. N Engl J Med. 2013;369(22):2126-2136.
• Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Med. 1996;22(7):707-710.
• Papazian L, Forel JM, Gacouin A, et al; ACURASYS Study Investigators. Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med. 2010;363(12):1107-1116.
• Guérin C, Reignier J, Richard JC, et al; PROSEVA Study Group. Prone positioning in severe acute respiratory distress syndrome. N Engl J Med. 2013;368(23):2159-2168.
• Brower RG, Matthay MA, Morris A, et al; Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342(18):1301-1308.

Notes

• The “Guidelines” list includes consensus-defining and practice-shaping evidence (definitions/frameworks and landmark strategy trials) because high-quality, post-REST guideline statements specifically endorsing VV-ECCO₂R for ARDS/AHRF are limited in the peer-reviewed literature.

Overall Takeaway

REST is a landmark negative and safety-informing trial: VV-ECCO₂R-facilitated ultraprotective ventilation achieved physiologic separation in tidal volume and pressures but did not reduce 90-day mortality and was associated with substantially more serious adverse events, particularly bleeding. Its secondary signal of fewer ventilator-free days is biologically plausible as a consequence of deeper sedation/controlled ventilation required to deliver ultralow tidal volumes, reinforcing that weaning-centred endpoints can be intrinsically coupled to the intervention itself.

Bibliography

• 1Serpa Neto A, Cardoso SO, Manetta JA, et al. Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. JAMA. 2012;308(16):1651-1659.
• 2Fitzgerald M, Millar J, Blackwood B, et al. Extracorporeal carbon dioxide removal for patients with acute respiratory failure secondary to the acute respiratory distress syndrome: a systematic review. Crit Care. 2014;18(3):222.
• 3Bein T, Weber-Carstens S, Goldmann A, et al. Lower tidal volume strategy (≈3 mL/kg) combined with extracorporeal CO₂ removal versus “conventional” protective ventilation (6 mL/kg) in severe ARDS: the prospective randomised Xtravent-study. Intensive Care Med. 2013;39(5):847-856.
• 4Boyle AJ, Sklar MC, McNamee JJ, et al. Extracorporeal carbon dioxide removal for lowering the risk of mechanical ventilation: research questions and clinical potential for the future. Lancet Respir Med. 2018;6(11):874-884.