Publication

• Title: Effect of Lung Recruitment and Titrated Positive End-Expiratory Pressure (PEEP) vs Low PEEP on Mortality in Patients With Acute Respiratory Distress Syndrome: A Randomized Clinical Trial
• Acronym: ART
• Year: 2017
• Journal published in: JAMA
• Citation: Cavalcanti AB, Suzumura ÉA, Laranjeira LN, et al; Writing Group for the Alveolar Recruitment for Acute Respiratory Distress Syndrome Trial (ART) Investigators. Effect of lung recruitment and titrated positive end-expiratory pressure (PEEP) vs low PEEP on mortality in patients with acute respiratory distress syndrome: a randomized clinical trial. JAMA. 2017;318(14):1335-1345.

Context & Rationale

• Background

  • In ARDS, collapsed and heterogeneously aerated lung units promote atelectrauma and stress concentration during ventilation.
  • Higher PEEP and recruitment manoeuvres can improve oxygenation and may reduce cyclic opening/closing, but risk haemodynamic compromise and overdistension-related lung injury.
  • Prior randomised comparisons of higher vs lower PEEP and “open lung” approaches produced mixed results, leaving uncertainty about patient selection and how best to titrate PEEP.
  • Physiology-based strategies proposed individualising PEEP after recruitment using respiratory-system compliance (as a surrogate for “best balance” between recruitment and overdistension).
• Research Question/Hypothesis

  • Does an “open lung” strategy (recruitment manoeuvre + decremental PEEP titration to best compliance) reduce 28-day mortality compared with an established low-PEEP strategy in adults with moderate-to-severe ARDS?
  • Hypothesis: lung recruitment with compliance-guided higher PEEP would reduce mortality by reducing ventilator-induced lung injury (via improved mechanics and reduced driving pressure).
• Why This Matters

  • “Recruit and titrate” approaches were widely used in practice and studied in smaller trials, yet lacked definitive patient-centred outcome data.
  • A large, pragmatic, international ICU trial could clarify whether aggressive recruitment plus compliance-guided PEEP improves survival or causes harm.
  • Results would directly inform ventilator management strategies in moderate-to-severe ARDS and influence guideline recommendations.

Design & Methods

• Research Question: In adults with moderate-to-severe ARDS, does a strategy of recruitment manoeuvre and compliance-titrated PEEP reduce 28-day mortality compared with a low-PEEP strategy?
• Study Type: Randomised, multicentre, investigator-initiated, international, open-label, parallel-group trial conducted in ICUs (120 ICUs across 9 countries).
• Population:
  • Setting: ICU; mechanically ventilated adults with ARDS within 72 hours of onset.
  • Key inclusion: Berlin-definition ARDS; after 3 hours of standardised ventilation (VT 6 mL/kg predicted body weight, RR 35/min, FiO₂ 1.0, PEEP 10 cmH₂O) and a further 30 minutes at FiO₂ 1.0 and PEEP 10 cmH₂O, required PaO₂:FiO₂ ≤200 mm Hg.
  • Key exclusion: More than 72 hours since ARDS onset; contraindications to recruitment/higher airway pressures (e.g., untreated pneumothorax, bronchopleural fistula); severe haemodynamic instability; intracranial hypertension; severe chronic lung disease or other specific protocol-defined exclusions.
• Intervention:
  • Core strategy: Maximum recruitment manoeuvre followed by decremental PEEP titration to best respiratory-system compliance (then set PEEP at best compliance +2 cmH₂O) and subsequent reassessment.
  • Recruitment manoeuvre (initial protocol): Pressure-controlled ventilation with ΔP 15 cmH₂O; PEEP 25 cmH₂O for 1 min, 35 for 1 min, 45 for 2 min (target maximum airway pressure 60 cmH₂O).
  • PEEP titration (initial protocol): Volume-control VT 5 mL/kg; decremental PEEP from 23 cmH₂O in steps of 3 cmH₂O down to 11, each step maintained 4 min, selecting best compliance.
  • Second recruitment (initial protocol): Repeated maximum recruitment at PEEP 45 cmH₂O for 2 min, then commence ventilation at selected PEEP.
  • Protocol modification: After the 556th patient, recruitment manoeuvre pressures/duration were reduced (PEEP 25 for 1 min, 30 for 1 min, 35 for 2 min; maximum airway pressure 50 cmH₂O; titration step duration reduced to 3 min; final recruitment performed at PEEP 35 for 2 min).
  • Co-interventions: Lung-protective ventilation targets in both groups; rescue therapies permitted for refractory hypoxaemia (e.g., prone positioning, inhaled nitric oxide, extracorporeal support) at clinician discretion.
• Comparison:
  • Ventilation strategy: Established low-PEEP strategy using a PEEP–FiO₂ table (ARDSNet-style) with lung-protective targets.
  • Recruitment manoeuvres: Not part of protocol in the control arm.
  • Rescue strategies: Permitted as clinically indicated.
• Blinding: Unblinded (open-label) due to the nature of ventilator strategies; primary outcome was objective (mortality), mitigating detection bias for the primary endpoint.
• Statistics: Event-driven design; planned to continue until 520 deaths by 28 days accrued to provide 90% power to detect HR 0.75 with type I error 5%; primary endpoint final α=0.042 due to interim monitoring; intention-to-treat analysis with complete-case approach (except cases lost to follow-up).
• Follow-Up Period: Mortality to 28 days (primary) and to 6 months (key secondary); additional in-hospital and ICU outcomes assessed within index admission timeframes.

Key Results

This trial was not stopped early. Two interim analyses were performed (after one-third and two-thirds of planned enrolment), and recruitment continued to completion.

Outcome	Lung recruitment manoeuvre + titrated PEEP (n=501)	Low-PEEP strategy (n=509)	Effect	p value / 95% CI	Notes
Death ≤28 days (primary)	277/501 (55.3%)	251/509 (49.3%)	HR 1.20	95% CI 1.01 to 1.42; P=0.041	Final significance threshold for primary outcome α=0.042
Death ≤6 months	327/501 (65.3%)	303/509 (59.5%)	HR 1.18	95% CI 1.01 to 1.38; P=0.04	Survival follow-up to 6 months
ICU mortality	272/501 (54.3%)	251/509 (49.3%)	RD 4.8%	95% CI 1.0 to 8.7; P=0.03	Absolute risk difference reported
Hospital mortality	295/501 (58.9%)	277/509 (54.4%)	RD 4.5%	95% CI 0.6 to 8.4; P=0.04	Index hospitalisation
Ventilator-free days ≤28 days	Mean 5.3 (SD 8.4)	Mean 6.4 (SD 9.5)	MD −1.1 days	95% CI −2.2 to 0.1; P=0.07	Distribution influenced by higher early mortality
Barotrauma ≤7 days	28/501 (5.6%)	8/509 (1.6%)	RD 4.0%	95% CI 2.1 to 5.9; P<0.001	Includes pneumothorax, subcutaneous emphysema, pneumomediastinum, etc (trial definition)
Pneumothorax requiring drainage ≤7 days	16/501 (3.2%)	6/509 (1.2%)	RD 2.0%	95% CI 0.4 to 3.7; P=0.03	Clinically relevant barotrauma subset
Death ≤7 days	160/501 (31.9%)	130/509 (25.5%)	RD 6.4%	95% CI 2.7 to 10.1; P=0.001	Signal suggests early harm
Commencement/increase vasopressors or hypotension ≤1 hour	174/501 (34.8%)	144/509 (28.3%)	RD 6.5%	95% CI 1.0 to 12.0; P=0.02	Exploratory mechanistic outcome
ICU length of stay	Mean 18.2 (SD 18.2)	Mean 19.2 (SD 20.1)	MD −1.0 days	95% CI −3.4 to 1.4; P=0.42	Secondary outcome

• Mortality increased with recruitment + compliance-titrated PEEP at 28 days (55.3% vs 49.3%) and 6 months (65.3% vs 59.5%), with a prominent early mortality signal (≤7 days: 31.9% vs 25.5%).
• Physiology improved but harms increased: early haemodynamic instability and barotrauma were more frequent in the intervention group (vasopressor initiation/increase or hypotension ≤1 hour: 34.8% vs 28.3%; barotrauma ≤7 days: 5.6% vs 1.6%).
• Subgroups: No statistically significant effect modification; harm direction was broadly consistent (e.g., PaO₂:FiO₂ >100 mm Hg: HR 1.30; 95% CI 1.00 to 1.69; interaction P=0.33; pulmonary ARDS: HR 1.32; 95% CI 1.03 to 1.69; interaction P=0.15).

Internal Validity

• Randomisation and allocation: Central web-based randomisation with permuted blocks (size 4), stratified by ICU, age, and PaO₂:FiO₂ (≤100 vs >100); allocation concealment before assignment was credible.
• Drop out or exclusions: 1013 randomised; 3 patients withdrew consent and were excluded from analyses; primary analysis sample 1010 (501 vs 509).
• Performance/detection bias: Open-label ventilation strategies; primary outcome (mortality) objective and centrally defined; co-interventions (rescue therapies) allowed and tracked.
• Protocol adherence: 480/501 (95.8%) received the recruitment manoeuvre; 21/501 (4.2%) did not undergo manoeuvre (per supplementary adherence table).
• Baseline characteristics: Well balanced (e.g., age 51.4 vs 50.8 years; SAPS 3 score 63.5 vs 62.7; baseline PaO₂:FiO₂ at eligibility 119.5 vs 117.2 mm Hg; shock at enrolment 63.5% vs 61.7%).
• Heterogeneity: Broad multicentre enrolment with prespecified subgroup analyses; no strong evidence of treatment-effect heterogeneity across key strata in the primary analysis.
• Timing: Early ARDS enrolment (median duration of ARDS at randomisation 15.1 vs 16.3 hours), aligning with a biologically plausible window for recruitment/PEEP strategies.
• Dose: Recruitment manoeuvre and titration were protocolised and high-intensity; a mid-trial protocol modification reduced maximum pressures and duration after safety concerns, potentially attenuating “dose” thereafter.
• Separation of the Variable of Interest: Clear early between-group separation in ventilator settings and mechanics (examples from the first 7 days, reported values):
  • At 1 hour: PEEP 16.4 (95% CI 16.0 to 16.7) vs 13.0 (12.7 to 13.3) cmH₂O; driving pressure 11.5 (11.1 to 11.8) vs 13.0 (12.6 to 13.3) cmH₂O; plateau pressure 27.9 (27.5 to 28.3) vs 25.9 (25.5 to 26.3) cmH₂O.
  • Day 1: PEEP 16.2 (15.9 to 16.6) vs 12.0 (11.7 to 12.3) cmH₂O; driving pressure 11.7 (11.3 to 12.1) vs 13.5 (13.1 to 13.8) cmH₂O.
  • Day 3: PEEP 14.2 (13.8 to 14.6) vs 10.5 (10.2 to 10.9) cmH₂O; driving pressure 12.1 (11.7 to 12.5) vs 13.5 (13.1 to 13.9) cmH₂O.
• Key delivery aspects: The intervention reliably increased PEEP and improved oxygenation but also raised plateau pressures early; haemodynamic optimisation and neuromuscular blockade were frequently used as part of delivery.
• Crossover: Not reported in the primary manuscript tables as a formal crossover metric; protocol deviations and additional manoeuvres were described in supplementary materials.
• Adjunctive therapy use: Rescue therapies within 7 days were similar (15.6% vs 17.5%); prone positioning use was 15.2% vs 16.1% during the first 7 days (supplementary co-interventions table).
• Outcome assessment: Mortality outcomes were clear; safety outcomes included prespecified barotrauma definitions and clinically relevant pneumothorax requiring drainage.
• Statistical rigor: Prespecified interim monitoring and adjusted primary alpha; primary time-to-event analysis with Cox model; secondary outcomes interpreted as exploratory per trial report.

Conclusion on Internal Validity: Overall, internal validity appears moderate-to-strong: randomisation and baseline balance were robust, separation of ventilator strategy was clear, and the primary endpoint objective, though open-label delivery and mid-trial protocol modification introduce complexity in interpreting mechanism and implementation fidelity.

External Validity

• Population representativeness: Adults with moderate-to-severe ARDS early in the syndrome, enrolled across diverse international ICUs; pneumonia was the most common ARDS risk factor, with substantial septic shock prevalence.
• Applicability: Findings apply most directly to invasively ventilated, moderate-to-severe ARDS where an aggressive recruitment manoeuvre and compliance-guided PEEP titration are being considered early.
• Important exclusions: Patients with contraindications to high airway pressures, severe haemodynamic instability, or other protocol-defined safety exclusions may limit application to the most unstable phenotypes.
• Resource considerations: Protocolised recruitment and decremental titration require experienced staff and close haemodynamic monitoring; feasibility may be limited in resource-constrained settings.

Conclusion on External Validity: External validity is good for early moderate-to-severe ARDS managed in ICUs with capability for protocolised ventilator manoeuvres; generalisability is more limited for very unstable patients or settings without capacity for intensive haemodynamic monitoring during recruitment.

Strengths & Limitations

• Strengths:
  • Large, international, multicentre ARDS ventilation trial with patient-centred primary outcome.
  • Early ARDS enrolment with standardised eligibility confirmation to reduce misclassification.
  • Protocolised and clearly separated ventilator strategies with objective primary endpoint.
  • Preplanned interim monitoring and time-to-event primary analysis.
• Limitations:
  • Open-label intervention with potential for performance bias (mitigated for mortality but relevant for process outcomes).
  • Mid-trial protocol modification after safety concerns, potentially introducing treatment heterogeneity over time.
  • Complex intervention bundle (recruitment + titration + repeated manoeuvres) limits attribution of harm/benefit to individual components.
  • Control strategy PEEP levels may be higher than “low PEEP” in some historic comparators, affecting interpretation of the counterfactual.

Interpretation & Why It Matters

• Clinical practice

  • Routine use of aggressive recruitment manoeuvres combined with compliance-guided PEEP titration in early moderate-to-severe ARDS increased mortality and should not be adopted as standard care on the basis of ART.
  • When oxygenation is poor, prioritise evidence-supported adjuncts (e.g., prone positioning) and lung-protective limits, while treating recruitment manoeuvres as high-risk interventions requiring careful patient selection and monitoring.
• Mechanistic insight

  • Improved oxygenation and modest reductions in driving pressure did not translate into benefit, and were accompanied by increased haemodynamic instability and barotrauma.
  • ART supports the view that “open lung” benefits are highly contingent on recruitability, haemodynamic tolerance, and the specific recruitment “dose” (pressure/time), rather than being universally beneficial.
• Trial design implications

  • Large mechanistically motivated ventilation trials must explicitly balance physiologic endpoints against early safety signals (haemodynamics, barotrauma) and consider pragmatic deliverability across sites.

Controversies & Subsequent Evidence

• Editorial interpretation (benefit–harm balance): The accompanying editorial emphasised that the intensity of recruitment/PEEP titration and the observed early haemodynamic and barotrauma signals provide plausible mechanisms for increased mortality, tempering enthusiasm for “open lung” strategies applied broadly.¹
• PEEP titration target (compliance as surrogate): Correspondence argued that selecting PEEP by best respiratory-system compliance after recruitment may not reliably identify a safe or optimal PEEP for the individual patient, potentially permitting overdistension despite “improved” compliance metrics.²
• Intervention delivery and protocol complexity: Correspondence highlighted implementation complexity (including strict prerequisites, monitoring, and interruption criteria) and questioned whether elements of manoeuvre timing/intensity could have contributed to harm in real-world ICU delivery.³
• Hyperoxia and stretch interaction: Correspondence emphasised that eligibility confirmation and manoeuvre phases used FiO₂ 1.0, raising concern that hyperoxia could amplify ventilator-induced lung injury during high-stress manoeuvres, complicating mechanistic attribution of harm.⁴
• Trialists’ reply (mechanisms and interpretation): The investigators responded to these concerns, addressing protocol design, feasibility/training, and alternative mechanistic explanations (including breath stacking and the unexpectedly lung-protective features of the control strategy).⁵
• Outcome set and reporting governance: The published statistical analysis plan clarified outcome handling (including ICU length of stay and several exploratory outcomes that were not in the earliest protocol versions), underscoring the importance of prospectively specified analysis frameworks in complex ventilation trials.⁶
• Subsequent trials: PHARLAP (a phase II maximal recruitment “open lung” strategy) halted enrolment after ART publication and reported no improvement in ventilator-free days or mortality, with increased cardiovascular adverse events in the intervention group, aligning with a safety-limited benefit–harm profile for high-intensity recruitment strategies.⁷
• Evidence synthesis after ART: A Bayesian network meta-analysis comparing PEEP selection strategies in moderate-to-severe ARDS suggested higher PEEP without lung recruitment manoeuvres had the highest probability of mortality benefit versus lower PEEP, while higher PEEP with prolonged recruitment manoeuvres had a high probability of harm versus higher PEEP without recruitment, concordant with ART’s direction of effect for prolonged/intense manoeuvre strategies.⁸
• Guideline incorporation: ESICM guidelines recommend against using prolonged high-pressure recruitment manoeuvres in ARDS and advise against routine use of brief recruitment manoeuvres, reflecting ART-informed safety concerns and the uncertain net benefit of recruitment strategies applied indiscriminately.⁹
• Updated ATS guidance: The ATS guideline update suggests higher PEEP without lung recruitment manoeuvres rather than lower PEEP in moderate-to-severe ARDS and recommends against prolonged recruitment manoeuvres (strong recommendation), explicitly reflecting the post-ART evidence base.¹⁰

Summary

• In 1010 analysed patients with early moderate-to-severe ARDS, recruitment manoeuvre plus compliance-titrated PEEP increased 28-day mortality vs a low-PEEP strategy (55.3% vs 49.3%; HR 1.20; 95% CI 1.01 to 1.42).
• Mortality harm persisted to 6 months (65.3% vs 59.5%; HR 1.18; 95% CI 1.01 to 1.38) and appeared early (≤7 days: 31.9% vs 25.5%).
• The intervention improved oxygenation and reduced driving pressure modestly, but increased haemodynamic instability (vasopressor initiation/increase or hypotension ≤1 hour: 34.8% vs 28.3%) and barotrauma (5.6% vs 1.6%).
• No convincing subgroup demonstrated benefit; the direction of effect was broadly consistent across prespecified strata.
• ART materially shifted practice and guideline recommendations away from routine recruitment manoeuvres and towards higher PEEP strategies without recruitment (when used) plus prioritisation of proven adjuncts.

Overall Takeaway

ART is a landmark trial because it overturned a physiologically appealing “open lung” strategy by demonstrating increased mortality and clear safety signals when aggressive recruitment manoeuvres were paired with compliance-titrated higher PEEP in early moderate-to-severe ARDS. Subsequent trials, evidence syntheses, and major guidelines have aligned away from prolonged recruitment manoeuvres and towards cautious use of higher PEEP without recruitment when indicated, prioritising proven ARDS interventions.

Bibliography

• 1Sahetya SK, Brower RG. Lung Recruitment and Titrated PEEP in Moderate to Severe ARDS: Is the Door Closing on the Open Lung? JAMA. 2017;318(14):1327-1329.
• 2McKown AC, Semler MW, Rice TW. Lung Recruitment and Positive End-Expiratory Pressure Titration in Patients With Acute Respiratory Distress Syndrome. JAMA. 2018;319(9):932-933.
• 3Barbas CSV, Nemer SN. Lung Recruitment and Positive End-Expiratory Pressure Titration in Patients With Acute Respiratory Distress Syndrome. JAMA. 2018;319(9):933.
• 4Morris IS, Lane AS, Seppelt I. Lung Recruitment and Positive End-Expiratory Pressure Titration in Patients With Acute Respiratory Distress Syndrome. JAMA. 2018;319(9):933-934.
• 5Cavalcanti AB, Amato MBP, Ribeiro de Carvalho CR. Lung Recruitment and Positive End-Expiratory Pressure Titration in Patients With Acute Respiratory Distress Syndrome—Reply. JAMA. 2018;319(9):934-935.
• 6Damiani LP, Berwanger O, Paisani D, et al. Statistical analysis plan for the Alveolar Recruitment for Acute Respiratory Distress Syndrome Trial (ART). A randomized controlled trial. Rev Bras Ter Intensiva. 2017;29(2):142-153.
• 7Hodgson CL, Cooper DJ, Arabi Y, et al. Maximal Recruitment Open Lung Ventilation in Acute Respiratory Distress Syndrome (PHARLAP): A Phase II, Multicenter Randomized Controlled Clinical Trial. Am J Respir Crit Care Med. 2019;200(11):1363-1372.
• 8Dianti J, Tisminetzky M, Ferreyro BL, et al. Association of Positive End-Expiratory Pressure and Lung Recruitment Selection Strategies with Mortality in Acute Respiratory Distress Syndrome: A Systematic Review and Network Meta-analysis. Am J Respir Crit Care Med. 2022;205(11):1300-1310.
• 9Grasselli G, Calfee CS, Camporota L, et al; European Society of Intensive Care Medicine Taskforce on ARDS. ESICM guidelines on acute respiratory distress syndrome: definition, phenotyping and respiratory support strategies. Intensive Care Med. 2023;49(7):727-759.
• 10Qadir N, Sahetya S, Munshi L, et al. An Update on Management of Adult Patients with Acute Respiratory Distress Syndrome: An Official American Thoracic Society Clinical Practice Guideline. Am J Respir Crit Care Med. 2024;209(1):24-36.