ARDS – Evidence-Based Management & Landmark Trials

ARDS 2026: Evidence-Based Management & Landmark Trials

Current standard of care, organized by treatment area, with the trials behind each recommendation.

Reviewed by the DosePilot Medical Team · Last reviewed: Jun 2026 · Independent — no industry funding
How to read this page. Each treatment area opens with the current approach, then the trials behind it (design, key results, limitations), and a short practical note (In practice). All doses and figures are taken from the cited primary sources (DOI + PMID). In practice notes are general practical guidance, not a substitute for clinical judgment.
How we verify evidence & editorial policy
Written and maintained by the DosePilot Medical Team. Recommendations are based on randomized controlled trials and current society guidelines (GRADE). Every citation is verified directly against PubMed (PMID) and the publisher DOI — no memory-based or AI-generated references. Each topic is peer-reviewed by the clinical team before publication and re-reviewed at least every 12 months or when major new evidence appears.

Funding & conflicts of interest. Produced independently; no industry funding was received for this content and contributors report no relevant financial conflicts of interest.
On this page
  1. Evidence timeline
  2. Definition & diagnosis
  3. Treat the cause & supportive care
  4. Lung-protective ventilation
  5. PEEP & recruitment
  6. Prone positioning
  7. Neuromuscular blockade
  8. Corticosteroids
  9. Fluid management
  10. Oxygenation targets
  11. Refractory hypoxemia & ECMO
  12. Oxygenation & NIV (pre-intubation)
  13. Inhaled vasodilators
  14. HFOV & ECCO₂R
  15. Where it’s heading
  16. Bottom line
  17. FAQ
  18. Related DosePilot tools
  19. Guidelines & evidence basis
At a glance — current standard of care
AreaCurrent approachTrial
Definition / diagnosisBerlin + 2023 Global Definition (incl. HFNO, SpO₂/FiO₂)Berlin; Global
Ventilation (tidal volume)6 mL/kg PBW, Pplat ≤30, minimize driving pressureARMA; Amato
PEEP / recruitmentAdequate PEEP; avoid aggressive recruitment maneuversALVEOLI/LOVS/EXPRESS; ART
Prone positioning≥16 h/day if PaO₂/FiO₂ <150PROSEVA
Neuromuscular blockadeNot routine; short cisatracurium in selected severeACURASYS → ROSE
CorticosteroidsEarly dexamethasone in moderate–severe; avoid lateDEXA-ARDS; RECOVERY
FluidsConservative strategy after resuscitationFACTT
AreaCurrent approachTrial
Refractory hypoxemia / ECMOEarly referral; VV-ECMO in selected very severeEOLIA; CESAR
Pre-intubation O₂ supportHigh-flow nasal oxygen; awake proning (COVID)FLORALI; Ehrmann
Treat the causeSource control + antibiotics + supportive bundle—
Oxygenation targetSpO₂ ~92–96%; avoid hyperoxia & ultra-conservativeICU-ROX; LOCO2; PILOT
Inhaled vasodilatorsRescue only; no mortality benefitiNO (Cochrane)
HFOV / ECCO₂RNot recommended routinelyOSCILLATE/OSCAR; REST

1Evidence timeline (2000–2026)

2000ARMA/ARDSNet — low tidal volume reduces mortality
2006FACTT — conservative fluids; LaSRS — late steroids don’t help
2009CESAR — ECMO referral benefit
2010ACURASYS — early cisatracurium benefit (severe)
2012Berlin Definition
2013PROSEVA — prone ≥16 h/day reduces mortality; OSCAR/OSCILLATE — HFOV not beneficial
2015FLORALI — high-flow nasal oxygen; Amato — driving pressure
2016Cochrane — inhaled nitric oxide: no mortality benefit
2017ART — aggressive recruitment harmful; ATS/ESICM/SCCM guideline
2018EOLIA — VV-ECMO in very severe ARDS
2019ROSE — routine paralysis no benefit
2020DEXA-ARDS — dexamethasone benefit; CoDEX (COVID-19); ICU-ROX/LOCO2 — oxygen targets
2021RECOVERY — dexamethasone in COVID-19; awake proning; REST — ECCO₂R no benefit
2022PILOT — SpO₂ target makes no difference
2023Global Definition of ARDS; ESICM ARDS guidelines
2024ATS ARDS guideline update (steroids, VV-ECMO, NMBA, PEEP); SCCM corticosteroid focused update
2025–26Subphenotype- & biomarker-guided (“predictive enrichment”) trials; 2023 Global Definition validation — no new mortality-changing RCT

2Definition & diagnosis

Current approach (2026)
Consensus definition
Diagnose ARDS by the Berlin criteria as updated by the 2023 Global Definition: acute hypoxemic respiratory failure within 1 week of a known insult, bilateral opacities (X-ray, CT, or lung ultrasound) not fully explained by cardiac failure/fluid overload, with hypoxemia graded by PaO₂/FiO₂ or SpO₂/FiO₂ (≤315 when SpO₂ ≤97%). ARDS is now recognized on HFNO ≥30 L/min and in resource-limited settings.
  • Mild: 200 < P/F ≤ 300 · Moderate: 100 < P/F ≤ 200 · Severe: P/F ≤ 100 (PEEP ≥5 if intubated). By SpO₂/FiO₂ (when SpO₂ ≤97%): mild ≤315 · moderate ≤235 · severe ≤148.
Evidence — old → new
2012
Berlin Definition. ESICM/ATS/SCCM consensus replacing the 1994 AECC definition; three severity tiers (PEEP ≥5) with stepwise mortality of 27% / 32% / 45%. Retired “acute lung injury” and four ancillary variables; modestly better mortality discrimination than AECC (AUROC 0.577 vs 0.536).
2023
Global Definition. 32-expert consensus expanding Berlin: include HFNO ≥30 L/min; accept SpO₂/FiO₂ ≤315 when no ABG; add ultrasound to imaging; drop PEEP/device requirements in resource-limited settings. Creates non-intubated, intubated, and resource-limited categories.

What changed. Earlier recognition (catches ARDS on HFNO before intubation), works without an ABG or advanced imaging, and is applicable globally.

Significance. A broader, earlier net — relevant for starting lung-protective care sooner and for cleaner trial enrolment. The non-intubated category formalizes the COVID-19 HFNO experience.

In practice

The Global Definition mainly enables earlier recognition — SpO₂/FiO₂ and ultrasound allow a bedside ARDS call (including on HFNO) without waiting for an ABG or formal imaging. It changes the timing of diagnosis, not the core therapy.

Sources (2)
  • ARDS Definition Task Force (Ranieri VM, et al). The Berlin Definition. JAMA. 2012;307(23):2526–2533. DOI 10.1001/jama.2012.5669. PMID 22797452.
  • Matthay MA, et al. A New Global Definition of ARDS. Am J Respir Crit Care Med. 2024;209(1):37–47. DOI 10.1164/rccm.202303-0558WS. PMID 37487152.

3Treat the cause & supportive care

Current approach (2026)
Foundational
The single most important step is treating the precipitating cause — source control and prompt, appropriate antibiotics for sepsis/pneumonia, plus management of the underlying insult. Combine with the supportive ICU bundle: lung-protective ventilation, conservative fluids after resuscitation, light sedation with daily interruption and early mobility, VTE and stress-ulcer prophylaxis, nutrition, and glucose control. Apply daily sedation interruption, spontaneous-breathing trials, and early mobility with standard safety screens (hold during active paralysis, unstable hemodynamics, or raised ICP).

Why it matters. No ARDS-specific therapy substitutes for reversing the cause; most ARDS is driven by pneumonia, sepsis, aspiration, or trauma. Supportive measures (the ABCDEF bundle, sedation strategy) reduce ventilator days and delirium and are the backbone of recovery.

In practice

Anchor every ARDS plan on the cause and the bundle: source control + antibiotics, lung-protective settings, light sedation with spontaneous-breathing trials and mobility, and conservative fluids once shock resolves.

Related DosePilot tools

4Lung-protective ventilation

Current approach (2026)
Guideline: strong (ATS/ESICM/SCCM 2017)RCT: mortality benefit
Ventilate with low tidal volume — 6 mL/kg predicted body weight (titrate 4–8) and plateau pressure ≤30 cmH₂O, and minimize driving pressure (ΔP = Pplat − PEEP) — keep it low (observational target ~≤15 cmH₂O). This is the foundational, mortality-reducing intervention in ARDS.
Evidence — old → new
2000
ARMA / ARDSNet (the foundation). RCT, 861 patients (stopped early). 6 mL/kg PBW (Pplat ≤30) vs traditional 12 mL/kg. Hospital mortality 31.0% vs 39.8% (P=0.007), more ventilator-free days (12 vs 10; P=0.007).
2015
Driving pressure (Amato). Mediation analysis of 3,562 patients across 9 RCTs. ΔP was most strongly associated with survival: each ~7 cmH₂O higher ΔP raised relative risk of death to 1.41 (95% CI 1.31–1.51), even within “protective” Vt/Pplat. Vt/PEEP changes helped only when they lowered ΔP.

What changed. From “limit tidal volume and plateau pressure” to “limit the pressure actually distending the lung — driving pressure,” normalizing Vt to each patient’s aerated lung size.

Significance. Low Vt remains the only ventilation strategy with RCT-proven mortality benefit; ΔP adds a bedside target. (ΔP evidence is observational/mediation, not yet a randomized target.)

In practice

Keep 6 mL/kg PBW and Pplat ≤30 as core targets. Use a rising driving pressure (ΔP) as a prompt to recheck PEEP, Vt, and recruitability rather than a number to chase.

Sources (2)
  • ARDS Network (Brower RG, et al). Ventilation with lower tidal volumes… N Engl J Med. 2000;342(18):1301–1308. DOI 10.1056/NEJM200005043421801. PMID 10793162.
  • Amato MBP, et al. Driving pressure and survival in ARDS. N Engl J Med. 2015;372(8):747–755. DOI 10.1056/NEJMsa1410639. PMID 25693014.

5PEEP & recruitment

Current approach (2026)
Guideline: conditional — higher PEEP (ATS 2017)RCT: harm — aggressive RM (ART)
Use adequate PEEP titrated to oxygenation/compliance (a higher-PEEP/FiO₂ table is reasonable in moderate–severe ARDS). Avoid aggressive sustained recruitment maneuvers with decremental PEEP titration — they increase mortality.
Evidence — old → new
2004–08
Higher vs lower PEEP (ALVEOLI, LOVS, EXPRESS). No mortality benefit from higher PEEP per se. ALVEOLI (n=549): 27.5% vs 24.9% (P=0.48). LOVS (n=983): 36.4% vs 40.4% (RR 0.90, 0.77–1.05; P=0.19), fewer refractory-hypoxemia episodes. EXPRESS (n=767): 31.2% vs 27.8% (RR 1.12, 0.90–1.40; P=0.31), more ventilator-free days. (Later meta-analyses suggested benefit limited to moderate–severe ARDS.)
2017
ART (the cautionary trial). RCT, 1,010 patients; recruitment + PEEP titration to best compliance vs low PEEP. 28-day mortality was higher with aggressive recruitment: 55.3% vs 49.3% (HR 1.20, 95% CI 1.01–1.42; P=0.041), with more pneumothorax (3.2% vs 1.2%) and barotrauma (5.6% vs 1.6%).
TrialPEEP comparisonMortality
ALVEOLI (2004)higher vs lower PEEP27.5% vs 24.9% (ns)
LOVS (2008)open-lung (RM + high PEEP)36.4% vs 40.4% (ns)
EXPRESS (2008)recruitment vs minimal distension31.2% vs 27.8% (ns)
ART (2017)RM + PEEP titration vs low PEEP55.3% vs 49.3% (worse, P=.04)

What changed. From “open the lung with high PEEP and recruitment” to “enough PEEP to oxygenate, but don’t force it.” ART reversed enthusiasm for routine aggressive recruitment.

Significance. PEEP should be individualized; aggressive recruitment is a documented harm, not a default.

In practice

Set PEEP to a PEEP/FiO₂ target and judge by oxygenation, compliance, and hemodynamic tolerance. After ART, avoid sustained high-pressure recruitment with decremental titration unless there is a specific rescue rationale.

Sources (4)
  • Brower RG, et al (ALVEOLI). N Engl J Med. 2004;351(4):327–336. DOI 10.1056/NEJMoa032193. PMID 15269312.
  • Meade MO, et al (LOVS). JAMA. 2008;299(6):637–645. DOI 10.1001/jama.299.6.637. PMID 18270352.
  • Mercat A, et al (EXPRESS). JAMA. 2008;299(6):646–655. DOI 10.1001/jama.299.6.646. PMID 18270353.
  • Cavalcanti AB, et al (ART). JAMA. 2017;318(14):1335–1345. DOI 10.1001/jama.2017.14171. PMID 28973363.

6Prone positioning

Current approach (2026)
Guideline: strong — severe ARDS (ATS 2017)RCT: mortality benefit
In moderate–severe ARDS (PaO₂/FiO₂ <150) on lung-protective settings, apply early, prolonged prone positioning — ≥16 hours/day. One of the few interventions with a clear mortality benefit. Pre-empt the main hazards (ETT/line dislodgement, pressure injury, facial edema) and screen for relative contraindications (unstable spine/pelvis, open abdomen, raised ICP, uncontrolled shock).
Evidence — old → new
pre-13
Earlier prone trials. Improved oxygenation but not survival — proning was often late, brief, or used in less severe patients.
2013
PROSEVA (the practice-changer). RCT, 466 patients with severe ARDS (P/F <150, FiO₂ ≥0.6, PEEP ≥5); prone ≥16 h/day vs supine. 28-day mortality 16.0% vs 32.8% (P<0.001); HR for death 0.39 (95% CI 0.25–0.63). 90-day mortality 23.6% vs 41.0% (HR 0.44). Complications similar.

What changed. Early initiation + long sessions + severe-ARDS selection + lung-protective ventilation converted prone from an oxygenation trick into a mortality-reducing therapy.

Significance. Large absolute mortality reduction at low cost; needs a trained team and airway/line vigilance.

In practice

In genuine P/F <150 ARDS, prone early and commit to ≥16 h sessions rather than brief trials. The main barrier is operational (staffing, safe turning), not evidence.

Sources (1)
  • Guérin C, et al (PROSEVA). Prone positioning in severe ARDS. N Engl J Med. 2013;368(23):2159–2168. DOI 10.1056/NEJMoa1214103. PMID 23688302.

7Neuromuscular blockade

Current approach (2026)
RCT: no routine benefit (ROSE)Selected severe: benefit (ACURASYS)
Do not routinely paralyze. Reserve a short (≤48 h) cisatracurium course (ACURASYS/ROSE regimen: 15 mg IV bolus, then 37.5 mg/h for 48 h) for selected severe ARDS with refractory hypoxemia, severe ventilator dyssynchrony, or high driving pressure despite optimized sedation. Paralyze only with deep sedation and analgesia confirmed first and maintained throughout (to prevent awareness), a secured airway, and full ventilatory control; reassess daily and stop early.
Evidence — old → new
2010
ACURASYS (positive signal). RCT, 340 patients with severe ARDS (P/F <150); cisatracurium 48 h vs placebo. Adjusted 90-day mortality favored cisatracurium (HR 0.68, 95% CI 0.48–0.98; P=0.04); 28-day 23.7% vs 33.3% (P=0.05). No increase in ICU-acquired paresis.
2019
ROSE (no benefit). RCT, 1,006 patients (P/F <150, PEEP ≥8); cisatracurium 48 h + deep sedation vs usual care + lighter sedation, both on high PEEP. 90-day in-hospital mortality 42.5% vs 42.8% (P=0.93); stopped for futility, with more cardiovascular adverse events.

What changed. With a contemporary high-PEEP strategy and lighter sedation, routine early paralysis adds no survival benefit (ROSE), tempering ACURASYS. The role narrows to rescue/selected use.

Significance. Avoid blanket paralysis and the deep sedation it requires; use NMB purposefully and briefly.

In practice

Avoid routine paralysis. Reserve short-course cisatracurium for when dyssynchrony or refractory hypoxemia defeats good sedation, then stop early and reassess.

Sources (2)
  • Papazian L, et al (ACURASYS). N Engl J Med. 2010;363(12):1107–1116. DOI 10.1056/NEJMoa1005372. PMID 20843245.
  • NHLBI PETAL Network (Moss M, et al; ROSE). N Engl J Med. 2019;380(21):1997–2008. DOI 10.1056/NEJMoa1901686. PMID 31112383.

8Corticosteroids

Current approach (2026)
Guideline: conditional (SCCM 2024)RCT: mortality/VFD benefit (DEXA-ARDS)
In moderate-to-severe ARDS (PaO₂/FiO₂ ≤ 200) identified early and without a strong contraindication, consider early IV dexamethasone — 20 mg once daily on days 1–5, then 10 mg once daily on days 6–10 (stop at extubation if sooner). Benefit is greatest when started early; do not initiate high-dose corticosteroids late (>2 weeks) in unresolving ARDS. Before starting: ensure the precipitant is being treated (source control + antimicrobials) — avoid in active uncontrolled infection, use caution in viral pneumonias (e.g., influenza), and screen for Strongyloides in endemic exposure. Monitor glucose, secondary infection, and ICU-acquired weakness. Guideline note: SCCM 2024 conditionally recommends corticosteroids in ARDS but does not specify a molecule, dose, or duration; the regimen above reflects DEXA-ARDS.
Evidence — old → new
2006
LaSRS (cautionary baseline). RCT, 180 patients with ARDS ≥7 days; methylprednisolone vs placebo. 60-day mortality unchanged (29.2% vs 28.6%; P=1.0); starting steroids >14 days after onset was associated with higher mortality.
2007
Meduri (early signal). RCT (n=91), early severe ARDS; methylprednisolone 1 mg/kg/day vs placebo. ICU mortality 20.6% vs 42.9% (P=0.03), faster extubation.
2020
DEXA-ARDS (the modern regimen). Open-label RCT, 277 patients, moderate–severe ARDS; dexamethasone 20 mg d1–5 → 10 mg d6–10 vs routine care. Ventilator-free days +4.8 (95% CI 2.57–7.03; P<0.0001); 60-day mortality 21% vs 36% (absolute −15.3%; P=0.0047).
2020–21
COVID-19 (CoDEX, RECOVERY). CoDEX (n=299): same regimen, +ventilator-free days (6.6 vs 4.0; P=0.04). RECOVERY (n=6,425): dexamethasone 6 mg (PO/IV) once daily up to 10 days, 28-day mortality 22.9% vs 25.7% (RR 0.83, 0.75–0.93) — benefit on ventilation (29.3% vs 41.4%) or oxygen (23.3% vs 26.2%), none without respiratory support (17.8% vs 14.0%).
TrialPopulationRegimenKey result
DEXA-ARDS (2020)mod–severe ARDSDexamethasone 20 mg ×5 d → 10 mg ×5 d↑VFD; 60-d mortality 21% vs 36%
RECOVERY (2021)COVID-19 on O₂/ventilationDexamethasone 6 mg/d (PO/IV) ≤10 d28-d mortality 22.9% vs 25.7%
Meduri (2007)early severe ARDSMethylprednisolone 1 mg/kg/d↓ICU mortality 20.6% vs 42.9%
LaSRS (2006)persistent ARDS ≥7 dMethylprednisolone (late)No benefit; harm if >14 d

What changed. From “controversial, harmful if late” (2006) to “early dexamethasone improves ventilator-free days and mortality in moderate–severe ARDS” (2020), with COVID-19 confirming benefit in hypoxemic failure on support.

Significance. Among the few drugs with RCT-level mortality signals in ARDS — but conditional on timing, severity, and avoiding late initiation; agent/dose are not interchangeable across populations.

In practice

Start dexamethasone early in clear moderate–severe ARDS once source control is adequate; avoid initiating steroids in the late fibroproliferative phase. Use RECOVERY’s 6 mg regimen for COVID-related hypoxemia and the DEXA-ARDS regimen for non-COVID moderate–severe ARDS.

Sources (5)
  • Steinberg KP, et al (LaSRS). N Engl J Med. 2006;354(16):1671–1684. DOI 10.1056/NEJMoa051693. PMID 16625008.
  • Meduri GU, et al. Chest. 2007;131(4):954–963. DOI 10.1378/chest.06-2100. PMID 17426195.
  • Villar J, et al (DEXA-ARDS). Lancet Respir Med. 2020;8(3):267–276. DOI 10.1016/S2213-2600(19)30417-5. PMID 32043986.
  • Tomazini BM, et al (CoDEX). JAMA. 2020;324(13):1307–1316. DOI 10.1001/jama.2020.17021. PMID 32876695.
  • RECOVERY Collaborative Group. N Engl J Med. 2021;384(8):693–704. DOI 10.1056/NEJMoa2021436. PMID 32678530.

9Fluid management

Current approach (2026)
RCT: more ventilator-free days (FACTT)No mortality difference
After initial resuscitation and once shock has resolved, use a conservative fluid strategy (net even-to-negative balance). It shortens ventilation and ICU stay without harming non-pulmonary organs.
Evidence
2006
FACTT. RCT, 1,000 patients; conservative vs liberal fluids over 7 days. 60-day mortality similar (25.5% vs 28.4%; P=0.30), but 7-day balance −136 mL vs +6,992 mL and more ventilator-free days (14.6 vs 12.1; P<0.001), without more shock or dialysis.

What changed. Established that “keeping the lungs dry” after resuscitation is safe and shortens ventilation.

Significance. A free, low-risk intervention; the caveat is timing — it applies after, not during, resuscitation of shock.

In practice

Once the shock phase is over, consider active de-resuscitation toward an even/negative balance when hemodynamics allow. The key judgment is confirming resuscitation is truly complete before restricting fluids.

Sources (1)
  • NHLBI ARDS Network (Wiedemann HP, et al; FACTT). N Engl J Med. 2006;354(24):2564–2575. DOI 10.1056/NEJMoa062200. PMID 16714767.

10Oxygenation targets

Current approach (2026)
RCT: no target superiorRCT: harm signal — ultra-conservative (LOCO2)
Target a reasonable SpO₂ of ~92–96% (PaO₂ ~60–80 mm Hg); avoid sustained hyperoxia and, in ARDS, avoid an ultra-conservative target (SpO₂ 88–92% / PaO₂ 55–70 mm Hg), which signalled harm. No single oxygen target has proven superior.
Evidence
2020
ICU-ROX. RCT, 1,000 ventilated patients; conservative vs usual oxygen. No difference in ventilator-free days; 180-day mortality 35.7% vs 34.5%.
2020
LOCO2 (ARDS). RCT, 205 patients; conservative (PaO₂ 55–70 mm Hg, SpO₂ 88–92%) vs liberal. Stopped early for safety — 90-day mortality 44.4% vs 30.4% (difference 14.0% [0.7–27.2]) with mesenteric ischemia events.
2022
PILOT. RCT, 2,541 patients; SpO₂ targets 90% vs 94% vs 98%. No difference in ventilator-free days or mortality.

What changed. Enthusiasm for tight “conservative” oxygen faded: LOCO2 showed possible harm in ARDS, and ICU-ROX/PILOT found no benefit of any specific target — supporting a moderate range over extremes.

In practice

Aim for SpO₂ ~92–96% in most ARDS, avoid both hyperoxia and tight 88–92% targets, and don’t chase a precise number — the trials show the target band, not a single value, is what matters.

Sources (3)
  • ICU-ROX Investigators (Mackle D, et al). N Engl J Med. 2020;382(11):989–998. DOI 10.1056/NEJMoa1903297. PMID 31613432.
  • Barrot L, et al (LOCO2). N Engl J Med. 2020;382(11):999–1008. DOI 10.1056/NEJMoa1916431. PMID 32160661.
  • Semler MW, et al (PILOT). N Engl J Med. 2022;387(19):1759–1769. DOI 10.1056/NEJMoa2208415. PMID 36278971.

11Refractory hypoxemia & ECMO

Current approach (2026)
RCT: selected benefit (EOLIA / CESAR)
For very severe / refractory ARDS meeting EOLIA-type thresholds — PaO₂/FiO₂ <50 for >3 h, or <80 for >6 h, or pH <7.25 with PaCO₂ ≥60 mm Hg for >6 h — refer early to an ECMO center. This presumes lung-protective ventilation, adequate PEEP, and a proning trial are already in place; venovenous ECMO is appropriate for selected patients at experienced centers.
Evidence — old → new
2009
CESAR. UK RCT, 180 patients; referral to an ECMO center vs conventional. Survival without severe disability at 6 months 63% vs 47% (RR 0.69, 95% CI 0.59–0.97; P=0.03). Caveats: ~75% of referred actually received ECMO; control care not protocolized.
2018
EOLIA. RCT, 249 patients with very severe ARDS; immediate VV-ECMO vs conventional (crossover allowed). 60-day mortality 35% vs 46% (RR 0.76, 95% CI 0.55–1.04; P=0.09); 28% of controls crossed over (57% of those died). A Bayesian reanalysis found a high posterior probability of benefit.

What changed. From “ECMO is rescue” toward “early VV-ECMO is reasonable in selected very severe ARDS.” EOLIA was statistically negative but, with CESAR and the Bayesian reanalysis, supports timely referral.

Significance. Outcomes depend on early referral, case selection, and center experience; high crossover makes the point estimates conservative.

In practice

Refer to the ECMO center early rather than waiting for a crash — selection and timing drive outcomes, and refractory patients do worse when referred late.

Sources (2)
  • Peek GJ, et al (CESAR). Lancet. 2009;374(9698):1351–1363. DOI 10.1016/S0140-6736(09)61069-2. PMID 19762075.
  • Combes A, et al (EOLIA). N Engl J Med. 2018;378(21):1965–1975. DOI 10.1056/NEJMoa1800385. PMID 29791822.

12Oxygenation & non-invasive support (pre-intubation)

Current approach (2026)
RCT: fewer intubations (HFNO; awake prone)
In acute hypoxemic respiratory failure not yet intubated, high-flow nasal oxygen (HFNO) is first-line non-invasive support; in COVID-19 on HFNO, add awake prone positioning. Track the ROX index (SpO₂/FiO₂ ÷ respiratory rate) to anticipate HFNO failure; NIV (especially helmet) is an option in selected patients but needs close monitoring and low tidal volumes. Monitor closely and avoid delaying intubation.
Evidence — old → new
2015
FLORALI. RCT, 310 patients with non-hypercapnic AHRF (P/F ≤300); HFNO vs standard O₂ vs NIV. Intubation at 28 days not different overall (38% vs 47% vs 50%; P=0.18), but HFNO had more ventilator-free days and lower 90-day mortality (HR for death 2.01 standard vs HFNO; 2.50 NIV vs HFNO). In P/F ≤200, intubation lower with HFNO.
2021
Awake proning (Ehrmann meta-trial). 1,126 patients with COVID-19 AHRF on HFNO; awake prone vs standard. Treatment failure (intubation or death by 28 days) 40% vs 46% (RR 0.86, 95% CI 0.75–0.98); HR for intubation 0.75, no harm signal.

What changed. HFNO displaced NIV as default non-invasive support in hypoxemic (non-hypercapnic) failure; COVID-19 added awake proning as an evidence-based adjunct on HFNO.

Significance. These reduce intubation in selected patients but require vigilant monitoring; persistent high work of breathing should prompt timely intubation.

In practice

Use HFNO as the default bridge in hypoxemic failure, adding awake proning in cooperative COVID patients — but set a clear deadline and intubate rather than prolong a failing trial.

Sources (2)
  • Frat JP, et al (FLORALI). N Engl J Med. 2015;372(23):2185–2196. DOI 10.1056/NEJMoa1503326. PMID 25981908.
  • Ehrmann S, et al. Awake prone positioning for COVID-19 AHRF: a meta-trial. Lancet Respir Med. 2021;9(12):1387–1395. DOI 10.1016/S2213-2600(21)00356-8. PMID 34425070.

13Inhaled pulmonary vasodilators

Current approach (2026)
RCT/meta: no mortality benefitHarm: renal impairment (iNO)
Use inhaled nitric oxide (iNO) or inhaled epoprostenol only as short-term rescue for refractory hypoxemia (a bridge to proning/ECMO), not as routine therapy. They transiently improve oxygenation but do not reduce mortality.
Evidence
2016
Inhaled NO (Cochrane meta-analysis). Across RCTs, iNO produced a transient improvement in oxygenation at 24 h but no mortality benefit (longest follow-up 38.2% vs 37.5%; RR 1.04) and a significant increase in renal impairment (RR 1.59).

Significance. Oxygenation gains do not translate into survival, and iNO carries renal harm — confining these agents to rescue use while definitive therapies (proning, ECMO) are arranged.

In practice

Reach for inhaled vasodilators only to buy time in refractory hypoxemia, watch renal function, and don’t expect a survival benefit — prioritize proning and ECMO referral. Inhaled epoprostenol avoids iNO’s renal signal and is cheaper; with iNO watch for methaemoglobinaemia and avoid abrupt discontinuation (rebound).

Sources (1)
  • Gebistorf F, et al. Inhaled nitric oxide for ARDS (Cochrane Review). Cochrane Database Syst Rev. 2016;(6):CD002787. DOI 10.1002/14651858.CD002787.pub3. PMID 27347773.

14HFOV & ECCO₂R (not routine)

Current approach (2026)
RCT: harm — HFOV (OSCILLATE)RCT: no benefit — ECCO₂R (REST)
Do not use high-frequency oscillatory ventilation (HFOV) routinely — it does not help and may increase mortality (ATS 2017: strong against). Extracorporeal CO₂ removal (ECCO₂R) to enable ultra-low tidal volumes is not recommended outside trials — no benefit and more harm.
Evidence
2013
HFOV — OSCILLATE & OSCAR. OSCILLATE (n=548) was stopped early: in-hospital mortality 47% vs 35% with HFOV (RR 1.33, 95% CI 1.09–1.64). OSCAR (n=795) found no difference (41.7% vs 41.1%). HFOV does not improve — and may worsen — outcomes.
2021
ECCO₂R — REST. RCT, 412 patients; ECCO₂R-facilitated lower tidal volume vs standard care. 90-day mortality 41.5% vs 39.5% (ns), fewer ventilator-free days, and more serious adverse events (31% vs 9%, including intracranial hemorrhage). Stopped early for futility.
In practice

Skip HFOV and routine ECCO₂R. If conventional lung-protective ventilation and proning fail, the next step is ECMO referral — not these.

Sources (3)
  • Ferguson ND, et al (OSCILLATE). N Engl J Med. 2013;368(9):795–805. DOI 10.1056/NEJMoa1215554. PMID 23339639.
  • Young D, et al (OSCAR). N Engl J Med. 2013;368(9):806–813. DOI 10.1056/NEJMoa1215716. PMID 23339638.
  • McNamee JJ, et al (REST). JAMA. 2021;326(11):1013–1023. DOI 10.1001/jama.2021.13374. PMID 34463700.

15Where it’s heading (2026)

Subphenotypes / endotypes. Hyper- vs hypo-inflammatory ARDS respond differently to PEEP, fluids, and possibly steroids; biomarker-guided (“predictive enrichment”) trials are the main frontier.

Personalized mechanics. Driving pressure and mechanical power as titration targets, and individualized PEEP, are moving toward randomized testing.

ECMO selection & definition validation. Refining who benefits after EOLIA; prospective validation of the 2023 Global Definition (HFNO/SpO₂-based).

16Bottom line — the current ARDS bundle

  1. Treat the cause (source control + antibiotics) and apply the supportive bundle.
  2. 6 mL/kg PBW, Pplat ≤30; minimize driving pressure (lower is better).
  3. Adequate PEEP titrated to oxygenation; no aggressive recruitment maneuvers.
  4. Prone ≥16 h/day if P/F <150.
  5. Conservative fluids once shock has resolved.
  6. Early dexamethasone in moderate–severe ARDS; avoid late initiation.
  7. Cisatracurium only for selected severe/dyssynchronous patients, short course.
  8. Target SpO₂ ~92–96%; avoid hyperoxia and ultra-conservative oxygen.
  9. Refer early for VV-ECMO in refractory very severe ARDS.
  10. Pre-intubation: HFNO ± awake proning, with a low threshold to intubate.
  11. Avoid HFOV and routine ECCO₂R; inhaled vasodilators only as rescue.

17FAQ

What tidal volume should be used in ARDS?
6 mL/kg predicted body weight (range 4–8) with plateau pressure ≤30 cmH₂O, minimizing driving pressure (≤15 cmH₂O). Low tidal-volume ventilation reduced mortality in ARMA/ARDSNet (31.0% vs 39.8%).
Does prone positioning reduce mortality in ARDS?
Yes, in moderate-to-severe ARDS (PaO₂/FiO₂ <150). In PROSEVA, ≥16 h/day of prone positioning reduced 28-day mortality (16.0% vs 32.8%).
Which corticosteroid regimen is used for ARDS?
Dexamethasone 20 mg IV once daily on days 1–5, then 10 mg once daily on days 6–10 (DEXA-ARDS). Start early; avoid initiating steroids late (>2 weeks) in unresolving ARDS.
What PEEP strategy is recommended in ARDS?
Adequate PEEP titrated to oxygenation and compliance (a higher-PEEP table is reasonable in moderate–severe ARDS). Avoid aggressive recruitment maneuvers with decremental PEEP titration, which increased mortality in ART.
Should ARDS patients be routinely paralyzed?
No — routine early neuromuscular blockade showed no survival benefit in ROSE. Reserve a short (≤48 h) cisatracurium course for selected severe ARDS with refractory hypoxemia or dyssynchrony (ACURASYS).
When is ECMO indicated in ARDS?
For very severe / refractory ARDS (e.g., PaO₂/FiO₂ <80 despite optimization). Refer early to an ECMO center; venovenous ECMO benefits selected patients (EOLIA; CESAR).
What oxygen-saturation target should be used in ARDS?
Aim for SpO₂ ~92–96% (PaO₂ ~60–80 mm Hg). No single target proved superior (ICU-ROX, PILOT), and an ultra-conservative target (SpO₂ 88–92%) signalled harm in ARDS (LOCO2). Avoid sustained hyperoxia.
Is inhaled nitric oxide useful in ARDS?
Inhaled nitric oxide transiently improves oxygenation but does not reduce mortality and increases renal impairment (Cochrane). Use it only as short-term rescue for refractory hypoxemia, not routinely.

19Guidelines & evidence basis

Badges summarize each recommendation’s basis: Guideline: strong Guideline: conditional RCT benefit RCT neutral RCT harm. Guideline strengths use the GRADE system and are quoted from the documents below.

Guidelines referenced (4)
  • Fan E, et al. ATS/ESICM/SCCM Clinical Practice Guideline: Mechanical Ventilation in Adult Patients with ARDS. Am J Respir Crit Care Med. 2017;195(9):1253–1263. DOI 10.1164/rccm.201703-0548ST. PMID 28459336. — strong: low tidal volume (4–8 mL/kg PBW) & prone in severe ARDS; conditional: higher PEEP & recruitment maneuvers; strong against routine HFOV.
  • Qadir 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. DOI 10.1164/rccm.202311-2011ST. PMID 38032683. — conditional: corticosteroids, VV-ECMO (selected severe), NMBA (early severe), and higher PEEP without recruitment maneuvers; strong against prolonged recruitment maneuvers.
  • Grasselli G, et al. ESICM guidelines on ARDS: definition, phenotyping and respiratory support strategies. Intensive Care Med. 2023;49(7):727–759. DOI 10.1007/s00134-023-07050-7. PMID 37326646. — 21 GRADE recommendations across HFNO, NIV, tidal volume, PEEP/recruitment, prone, neuromuscular blockade, and extracorporeal life support.
  • Chaudhuri D, et al. 2024 Focused Update: Corticosteroids in Sepsis, ARDS, and Community-Acquired Pneumonia. Crit Care Med. 2024;52(5):e219–e233. DOI 10.1097/CCM.0000000000006172. PMID 38240492. — conditional recommendation to administer corticosteroids in ARDS; no specific molecule, dose, or duration recommended.

Disclaimer. This page is an educational summary of published evidence for clinicians and is not a substitute for individual clinical judgment or institutional protocols. Doses and thresholds are taken from the cited trials; verify against current labeling, local guidelines, and patient-specific factors before applying. “In practice” notes are general practical guidance. Recommendation strengths are quoted from the cited guidelines (GRADE); reviewed by the DosePilot Medical Team (Jun 2026). All citations verified via PubMed/Europe PMC on 2026-06-22.