What ARDS is
"ARDS is non-cardiogenic pulmonary oedema and hypoxaemia arising from direct or indirect lung injury — a common final pathway of many insults that almost always ends in hypoxaemic respiratory failure needing mechanical ventilation. Despite decades of trials, treatment remains largely supportive: protect the lung and treat the cause."
Summarised from Irwin & Rippe's Intensive Care Medicine; with Marino's The ICU Book and Oh's Intensive Care Manual.How the definition evolved
| Definition | Key idea |
|---|---|
| Ashbaugh & Petty (1967) | First description of the syndrome |
| AECC (1994) | ALI/ARDS split by P/F; limited by CXR reliability & PEEP effect |
| Berlin (2012) | Current mainstream: acute onset ≤1 week, bilateral opacities, not fully explained by cardiac failure/overload, P/F on PEEP ≥5. Drops the term "ALI". Severity by P/F. |
| New Global Definition (2023) | Broadens Berlin: adds a non-intubated category (HFNO ≥30 L/min or CPAP/NIV ≥5), allows SpO₂/FiO₂ ≤315 and lung ultrasound, and relaxes the PEEP rule for resource-limited settings. |
Berlin severity (measured on PEEP ≥5 cmH₂O)
| Severity | PaO₂/FiO₂ | Approx. mortality |
|---|---|---|
| Mild | 200–300 | ~27% |
| Moderate | 100–200 | ~32% |
| Severe | ≤100 | ~45% |
What triggers the lung injury
| Direct (pulmonary) injury | Indirect (extra-pulmonary) injury |
|---|---|
| Pneumonia (commonest overall) | Sepsis (commonest indirect) |
| Gastric aspiration | Severe non-thoracic trauma / shock |
| Pulmonary contusion | Acute pancreatitis |
| Inhalation injury / near-drowning | Massive transfusion / TRALI |
| Diffuse alveolar haemorrhage | Burns, fat embolism, drug reaction |
Always name the trigger. ARDS is a syndrome, not a diagnosis — the single most important therapeutic step is source control of the underlying cause (antibiotics for sepsis/pneumonia, etc.).
Diffuse alveolar damage — and the "baby lung"
Whatever the trigger, the histology converges on diffuse alveolar damage (DAD): injury to the alveolar–capillary barrier → protein-rich oedema, hyaline membranes, surfactant loss and microthrombi. Aerated lung shrinks to a small, compliant "baby lung"; the flooded, collapsed regions cause shunt → refractory hypoxaemia and stiff lungs.
| Phase | Timing | Hallmarks |
|---|---|---|
| Exudative | ~Days 1–7 | Epithelial/endothelial injury, neutrophils, interstitial & alveolar oedema, hyaline membranes; maximal hypoxaemia |
| Proliferative | ~Days 7–21 | Organisation of exudate, type-II pneumocyte & fibroblast proliferation, granulation tissue; recovery begins in most |
| Fibrotic | >2–3 weeks (can start early) | Collagen deposition, "honeycombing", ↓ compliance; predicts worse outcome & prolonged ventilation |
Because only the baby lung is available for ventilation, "normal" tidal volumes over-distend it → VILI. Lung-protective settings, PEEP (to recruit and prevent atelectrauma) and prone positioning all follow directly from this shunt-plus-baby-lung model.
Making the diagnosis
| Berlin criterion | Requirement |
|---|---|
| Timing | Within 1 week of a known insult or new/worsening symptoms |
| Imaging | Bilateral opacities on CXR/CT not fully explained by effusions, collapse or nodules (+ ultrasound in 2023 definition) |
| Origin of oedema | Not fully explained by cardiac failure or fluid overload (echo if no risk factor) |
| Oxygenation | P/F (or SpO₂/FiO₂) reduced, on PEEP/CPAP ≥5 |
Work-up
- Find the cause: cultures, respiratory panel, amylase/lipase, transfusion review, drug history.
- Exclude cardiogenic oedema: echocardiography ± natriuretic peptides (compared in §11).
- Bedside: lung ultrasound (B-lines, spared areas), ABG for P/F, assess compliance/driving pressure once ventilated.
Lung-protective ventilation
"The one intervention that consistently improves survival is a ventilator strategy that limits stretch: low tidal volume, low plateau pressure and low driving pressure. Everything else is adjunct."
Summarised from Irwin & Rippe; Marino; ARDSNet.Low tidal volume
6 mL/kg predicted body weight (start 8, reduce to 6; floor 4). Dose on height-based PBW, never actual weight.
Limit the pressures
Plateau ≤30 cmH₂O and driving pressure (Pplat − PEEP) <14–15 — the variable most tightly linked to survival.
Set PEEP
Use a PEEP/FiO₂ table; higher PEEP for moderate–severe disease. Titrate to oxygenation and lowest driving pressure, avoiding over-distension.
Permissive hypercapnia
Accept pH ~7.25–7.30 and a high CO₂ rather than raise Vt (avoid if raised ICP or severe pulmonary hypertension).
Target oxygenation, don't chase it
SpO₂ 88–95% / PaO₂ 55–80. Wean FiO₂ to avoid hyperoxia. Full detail on the Mechanical Ventilation page.
What to add — and the evidence
| Intervention | When | Evidence / caveat |
|---|---|---|
| Prone positioning ≥16 h/day | P/F <150 (moderate–severe) | PROSEVA — clear mortality benefit. Early and prolonged. |
| Conservative fluid strategy | Once shock resolved | FACTT — fewer ventilator days (no mortality harm). Keep the lung dry. |
| Neuromuscular blockade | Early severe ARDS / dyssynchrony | Short course; oxygenation ↑ (ACURASYS positive, ROSE neutral) — not routine for all. |
| Higher PEEP | Moderate–severe, recruitable | Benefit mainly in more severe disease; individualise. |
| Corticosteroids | Moderate–severe (e.g. dexamethasone); COVID-ARDS | DEXA-ARDS/CoDEX supportive; overall mixed — reasonable, not universal. |
| Veno-venous ECMO | Refractory hypoxaemia despite the above | EOLIA/CESAR — rescue in expert centres. |
| Inhaled pulmonary vasodilators (NO) | Refractory hypoxaemia / RV failure | Improve oxygenation transiently; no mortality benefit — rescue/bridge only. |
Aggressive recruitment manoeuvres with PEEP escalation increased mortality in the ART trial — avoid as a blanket strategy. High-frequency oscillation (OSCILLATE/OSCAR) is not beneficial.
The rest of the ICU bundle
| Domain | Approach |
|---|---|
| Treat the cause | Source control, timely antibiotics for sepsis/pneumonia |
| Sedation & comfort | Light, targeted; daily interruption where safe; analgo-sedation |
| Fluids & haemodynamics | Conservative once resuscitated; watch RV in severe ARDS/high PEEP |
| Nutrition & prophylaxis | Enteral feed; VTE and stress-ulcer prophylaxis |
| Prevent complications | Ventilator bundle (VAP), early mobilisation, glucose control |
Outcomes
- Mortality rises with severity (~27% mild → ~45% severe); most deaths are from the underlying illness/multi-organ failure, not refractory hypoxaemia alone.
- Survivors often have prolonged morbidity — ICU-acquired weakness, reduced diffusing capacity, cognitive impairment and reduced quality of life (long-term follow-up matters).
- Early fibroproliferation predicts worse outcomes and longer ventilation.
Common mistakes
Delivers huge volumes to a baby lung = volutrauma. Always use height-based PBW.
Normalising gases with big volumes/high FiO₂ harms the lung. Accept SpO₂ 88–95% and permissive hypercapnia.
Prone works when applied early for ≥16 h/day in P/F <150 — not as a last-ditch 2-hour trial.
A wet lung worsens oxygenation. Once perfusion is restored, run dry.
No amount of ventilator optimisation replaces treating the cause (sepsis, aspiration, pancreatitis).
Exam pearls
Q: Berlin criteria?
Onset ≤1 week; bilateral opacities; not fully explained by cardiac failure/overload; P/F on PEEP ≥5 — mild 200–300, moderate 100–200, severe ≤100.
Q: What did the 2023 global definition add?
A non-intubated category (HFNO ≥30 L/min or CPAP/NIV ≥5), SpO₂/FiO₂ ≤315, lung ultrasound, and relaxed PEEP rules for resource-limited settings.
Q: Histological hallmark and phases?
Diffuse alveolar damage — exudative (hyaline membranes) → proliferative → fibrotic.
Q: Interventions with a mortality benefit?
Low tidal volume/low driving pressure (ARDSNet) and prone positioning (PROSEVA). ECMO is rescue; NMBA and steroids are selective.
Q: Why is driving pressure important?
ΔP = Pplat − PEEP reflects tidal strain on the baby lung and is the ventilator variable most strongly linked to mortality; keep <14–15.
All the comparisons in one place
The distinctions examiners love, gathered at the end for quick revision.
Berlin (2012) vs New Global Definition (2023)
| Feature | Berlin 2012 | Global 2023 |
|---|---|---|
| Support required | Intubated, PEEP ≥5 | Also non-intubated (HFNO ≥30 L/min, CPAP/NIV ≥5) |
| Oxygenation index | PaO₂/FiO₂ only | PaO₂/FiO₂ or SpO₂/FiO₂ ≤315 |
| Imaging | CXR/CT | CXR/CT or ultrasound |
| Resource-limited | Not addressed | PEEP requirement relaxed |
Direct vs Indirect ARDS
| Feature | Direct (pulmonary) | Indirect (extra-pulmonary) |
|---|---|---|
| Insult | Injures alveolar epithelium directly | Injures endothelium via blood-borne mediators |
| Examples | Pneumonia, aspiration, contusion | Sepsis, pancreatitis, transfusion, trauma |
| Imaging | Often patchy/asymmetric consolidation | More diffuse, symmetric oedema |
| Recruitability | Often less recruitable | Often more recruitable (PEEP-responsive) |
ARDS vs Cardiogenic pulmonary oedema
| Feature | ARDS | Cardiogenic oedema |
|---|---|---|
| Mechanism | Capillary leak (high-permeability) | Raised hydrostatic pressure |
| Onset | Over hours–days with a trigger | Often abrupt |
| Heart / echo | Normal LV function, no volume overload | LV dysfunction / raised filling pressures |
| Fluid / oedema | Protein-rich; responds poorly to diuresis alone | Transudate; responds to diuresis/afterload reduction |
| Natriuretic peptides | Usually lower | Usually elevated |
DAD phases
| Feature | Exudative | Proliferative | Fibrotic |
|---|---|---|---|
| Timing | Days 1–7 | Days 7–21 | >2–3 weeks |
| Hallmark | Hyaline membranes, oedema | Fibroblast/type-II proliferation | Collagen, honeycombing |
| Clinical | Worst hypoxaemia | Recovery in most | ↓ compliance, prolonged ventilation |
References
- Irwin RS, Lilly CM, Mayo PH, Rippe JM (eds). Irwin & Rippe's Intensive Care Medicine. 9th ed. Wolters Kluwer; 2023.
- Marino PL. Marino's The ICU Book. 5th ed. Wolters Kluwer; 2025.
- Bersten AD, Handy JM (eds). Oh's Intensive Care Manual. Elsevier; 2026.
- ARDS Definition Task Force. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307:2526–2533.
- Matthay MA, Arabi Y, Arroliga AC, et al. A new global definition of ARDS. Am J Respir Crit Care Med. 2023;207:37–47.
- ARDSNet. Ventilation with lower tidal volumes for ARDS. N Engl J Med. 2000;342:1301–1308.
- Amato MBP, Meade MO, Slutsky AS, et al. Driving pressure and survival in ARDS. N Engl J Med. 2015;372:747–755.
- Guérin C, Reignier J, Richard JC, et al. (PROSEVA). Prone positioning in severe ARDS. N Engl J Med. 2013;368:2159–2168.
- National Heart, Lung, and Blood Institute ARDS Network (FACTT). Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006;354:2564–2575.
- Papazian L, et al. (ACURASYS); Moss M, et al. (ROSE). Neuromuscular blockade in ARDS. N Engl J Med. 2010;363:1107–1116 / 2019;380:1997–2008.
- Combes A, Hajage D, Capellier G, et al. (EOLIA). ECMO for severe ARDS. N Engl J Med. 2018;378:1965–1975.
- Grasselli G, Calfee CS, Camporota L, et al. ESICM guidelines on ARDS. Intensive Care Med. 2023;49:727–759.