• Reported incidence varies widely: approximately 10–80 per 100,000 person-years depending on the population studied and definition used
• The LUNG SAFE study (2016, multinational ICU study) found that ARDS accounted for ~10% of all ICU admissions and ~23% of mechanically ventilated patients
• Likely underdiagnosed: the same LUNG SAFE study found that clinicians failed to recognise ARDS in ~40% of cases at onset

• Overall ICU mortality remains 35–46% despite advances in supportive care
• Mortality by severity (Berlin Definition):
  • Mild: ~27%
  • Moderate: ~32%
  • Severe: ~45%
• In-hospital mortality for severe ARDS can exceed 50%
• Death is usually from multiorgan failure (especially sepsis-driven) rather than from refractory hypoxaemia alone

• More common in older adults (median age ~60 years)
• Slightly more common in males
• Higher incidence and worse outcomes in patients of lower socioeconomic status and certain ethnic groups

• Sepsis is the single most common cause of ARDS (~30–40% of cases) ^[2]
• Pneumonia, aspiration, trauma, pancreatitis, massive transfusion

• High tidal volume mechanical ventilation (ventilator-induced lung injury — VILI)
• Excessive fluid resuscitation
• Blood product transfusion (TRALI) ^[3]
• Certain drugs and oxygen toxicity

The alveolar-capillary membrane is the site of gas exchange and has the following layers (from air to blood):

1. Alveolar epithelium — two cell types:
   • Type I pneumocytes (~95% of alveolar surface area): thin, flat cells optimised for gas exchange
   • Type II pneumocytes (~5% of surface area): cuboidal cells that:
     • Produce and secrete surfactant (reduces surface tension, prevents alveolar collapse)
     • Act as progenitor cells to regenerate Type I cells after injury
     • Actively transport sodium and fluid OUT of the alveolar space (alveolar fluid clearance)
2. Basement membrane — shared between epithelium and endothelium
3. Interstitial space — thin, contains collagen and elastin
4. Capillary endothelium — single layer of endothelial cells

• Gas exchange: O₂ diffuses from alveolus → blood; CO₂ from blood → alveolus. The total diffusion distance is only ~0.2–0.5 μm.
• Fluid balance: The alveolar epithelium is normally very tight (tight junctions). Small amounts of interstitial fluid are drained by lymphatics. Active sodium transport by Type II cells keeps the alveolar space dry.
• Surfactant: a complex of phospholipids (mainly dipalmitoylphosphatidylcholine) and surfactant proteins (SP-A, SP-B, SP-C, SP-D) that reduces surface tension at the air-liquid interface. By LaPlace's Law (P = 2T/r), if surface tension (T) is high and alveolar radius (r) is small (at end-expiration), the pressure needed to keep alveoli open becomes very high → without surfactant, alveoli collapse (atelectasis) ^[4].

In ARDS, the alveolar-capillary membrane is destroyed:

• Endothelial injury → ↑ vascular permeability → protein-rich fluid leaks into interstitium and alveoli
• Epithelial injury → loss of barrier function, loss of fluid clearance, loss of surfactant production
• Result: flooded, collapsed, non-functional alveoli → profound hypoxaemia

The insult primarily damages the alveolar epithelium.

The insult primarily damages the pulmonary capillary endothelium via circulating mediators.

ARDS commonly occurs as part of MODS in the context of sepsis ^[5]. The lung is often the first organ to fail because of its unique position — it receives the entire cardiac output and has the body's largest capillary bed, making it maximally exposed to circulating inflammatory mediators.

^[5]

Pathophysiology: diffuse alveolar damage (DAD) ^[2]

Neutrophilic infiltration of lung → diffuse alveolar damage with increased permeability of alveolar-capillary membrane → alveolar and interstitial oedema ^[1]

The pathological hallmark of ARDS is diffuse alveolar damage (DAD), which proceeds through distinct phases:

Multiple mechanisms combine:

The hallmark of ARDS hypoxaemia is that it is shunt-predominant — this is why giving 100% O₂ doesn't fix it. The shunted blood never sees the oxygen. This is fundamentally different from, say, COPD (V/Q mismatch predominant), where supplemental O₂ is very effective.

This distinction is clinically critical:

The Berlin Definition replaced the older 1994 American-European Consensus Conference (AECC) definition. It removed the term "acute lung injury (ALI)" and instead classifies ARDS by severity.

PaO₂/FiO₂ ratio (P/F ratio): This is the key metric. A normal P/F ratio is ~400–500 (e.g., PaO₂ ~100 mmHg on FiO₂ 0.21). A P/F ratio ≤ 300 indicates significant oxygenation impairment. The lower the ratio, the worse the shunt and the higher the mortality.

The 2023 updated global definition expanded the Berlin criteria to improve recognition in resource-limited settings:

• SpO₂/FiO₂ ratio can replace PaO₂/FiO₂ when ABG is unavailable
  • Mild: SpO₂/FiO₂ > 315 with SpO₂ ≤ 97%
• Bilateral opacities: can be detected by lung ultrasound (not just CXR/CT)
• PEEP requirement: minimum PEEP of 5 cmH₂O OR high-flow nasal cannula (HFNC) ≥ 30 L/min now accepted
• Non-intubated ARDS: explicitly recognised (patients on HFNC or NIV can have ARDS)

This is a clinically and pathophysiologically meaningful classification:

The typical clinical trajectory:

1. Precipitating event (Day 0): e.g., onset of sepsis, pneumonia, aspiration
2. Latent period (hours): Patient may appear well initially or have symptoms of the underlying condition
3. Acute onset (6–72 hours): Progressive dyspnoea, tachypnoea, hypoxaemia refractory to supplemental O₂
4. Mechanical ventilation (often within 24–48 hours): Required in most moderate-severe cases
5. Exudative phase (Days 1–7): Maximal oedema, worst gas exchange
6. Fibroproliferative phase (Days 7–21): If surviving — either improvement or progression to fibrosis
7. Resolution or chronic phase ( > 21 days): Recovery with variable residual impairment, or persistent fibrosis

The key concept is intrapulmonary shunting:

• In normal lungs, all perfused alveoli are ventilated → V/Q matching → efficient gas exchange
• In ARDS, large areas of lung are flooded or collapsed → blood flows past non-ventilated alveoli → deoxygenated blood enters the pulmonary veins
• This is a true shunt (V/Q = 0) → cannot be corrected by ↑ FiO₂ alone (because the blood never contacts oxygen)
• This is why ARDS patients need PEEP — to recruit collapsed alveoli and convert shunt into ventilated units

The lung receives the entire cardiac output → every inflammatory mediator, activated neutrophil, or toxin in the bloodstream passes through the pulmonary capillary bed. The capillary bed has an enormous surface area (~70 m²), making it a massive "filter" that traps activated neutrophils. This is why the lung is often the first organ to fail in sepsis/MODS ^[5].

The balance between resolution and fibrosis depends on:

• Severity and duration of initial injury
• Effectiveness of alveolar fluid clearance (Type II cell function)
• Balance of pro-fibrotic (TGF-β, PDGF) vs anti-fibrotic mediators
• Ventilator-induced lung injury (VILI) from mechanical ventilation
• Superimposed infections
• Genetic factors

Multiple mechanisms:

1. Hypoxic pulmonary vasoconstriction (HPV): Normal reflex to divert blood away from hypoxic alveoli, but in ARDS, so much lung is hypoxic that HPV becomes widespread → ↑ pulmonary vascular resistance
2. Vascular compression: Oedema and alveolar distension compress small pulmonary vessels
3. Microvascular thrombosis: Intravascular coagulation (fibrin, platelet microthrombi) obstructs vessels
4. Vascular remodelling: In the fibroproliferative phase, intimal fibrosis and medial hypertrophy of pulmonary arterioles
5. Result: Right ventricular failure (acute cor pulmonale) — occurs in ~25% of ventilated ARDS patients

• Hong Kong has unique experience with ARDS from viral epidemics:
  • SARS (2003): SARS-CoV caused severe ARDS with high mortality; taught valuable lessons about infection control and ICU management
  • COVID-19 (2020–2023): SARS-CoV-2 pneumonia was a major cause of ARDS locally; COVID-ARDS has distinct features (preserved compliance early, high thrombotic burden, prolonged course)
• COVID-ARDS phenotypes:
  • Type L (early): Low elastance (preserved compliance), low V/Q ratio, low lung weight, low recruitability
  • Type H (later): High elastance (stiff lungs), high right-to-left shunt, high lung weight, high recruitability — more "classic" ARDS

• Sepsis (particularly from pneumonia, urinary tract, biliary sources)
• Aspiration pneumonia (common in elderly with poor dentition, stroke, dementia)
• Acute pancreatitis (biliary > alcoholic in HK)
• TRALI (with the blood transfusion service's universal leucodepletion, incidence has decreased but remains a consideration) ^[3]

The Berlin Definition is the universally accepted diagnostic framework for ARDS. It replaced the older 1994 AECC definition by eliminating the category of "acute lung injury (ALI)" and introducing severity stratification that directly correlates with mortality.

Think of the Berlin Definition as answering four questions simultaneously:

The PaO₂/FiO₂ ratio (P/F ratio) is the cornerstone metric. Let's understand it:

• PaO₂ = arterial partial pressure of oxygen (measured on ABG)
• FiO₂ = fraction of inspired oxygen (0.21 for room air, 1.0 for 100% O₂)
• Normal P/F ratio ≈ 400–500 (e.g., PaO₂ ~100 mmHg on FiO₂ 0.21 = 100/0.21 ≈ 476)
• P/F ≤ 300 = significant oxygenation impairment
• P/F ≤ 200 = moderate ARDS (roughly equivalent to PaO₂ < 80 mmHg on 40% O₂)
• P/F ≤ 100 = severe ARDS (roughly equivalent to PaO₂ < 100 mmHg on 100% O₂ — the lung is barely exchanging gas)

Why use a ratio rather than PaO₂ alone? Because PaO₂ depends on FiO₂ — a PaO₂ of 80 mmHg on room air is normal, but a PaO₂ of 80 mmHg on 100% O₂ indicates catastrophic gas exchange failure. The ratio normalises for the amount of O₂ being delivered.

Why require PEEP ≥ 5 cmH₂O? Without PEEP, collapsed alveoli stay collapsed, worsening the shunt and artificially lowering PaO₂. A minimum of 5 cmH₂O PEEP standardises the measurement by keeping at least some recruitable alveoli open. This prevents over-diagnosis of ARDS in patients who simply need a little PEEP to recruit normal lung.

The 2023 update expanded the Berlin criteria to improve recognition in resource-limited settings and non-intubated patients:

Exclusion of cardiogenic pulmonary oedema requires ^[9]:

Cardiac examination showing ABSENCE of:

• Elevated JVP
• S3 or S4 gallop
• New or changing murmur

CXR showing ABSENCE of:

• Pulmonary venous congestion
• Kerley B lines
• Cardiomegaly
• Pleural effusions

Echocardiography showing ABSENCE of:

• Aortic or mitral valve dysfunction
• Severe diastolic dysfunction
• Reduced LVEF

Exclusion of other causes of hypoxaemic respiratory failure may require tracheobronchial aspiration, bronchoalveolar lavage (BAL), and flexible bronchoscopy ^[9]

The ABG is the single most important investigation in ARDS — it confirms hypoxaemia, calculates the P/F ratio, and identifies acid-base disturbances.

Stepwise ABG interpretation ^[19][20]:

Calculating the A-a gradient:

• A-a gradient = PAO₂ − PaO₂
• PAO₂ = (FiO₂ × [Patm − PH₂O]) − (PaCO₂ / 0.8)
• On room air at sea level: PAO₂ = (0.21 × [760 − 47]) − (PaCO₂ / 0.8) ≈ 150 − (PaCO₂ × 1.25)
• Normal A-a gradient ≈ (Age/4) + 4, or roughly < 15 mmHg in young adults
• In ARDS, the A-a gradient is markedly elevated (often > 40 mmHg) because of intrapulmonary shunting

Pulse oximetry is a non-invasive measurement of SaO₂ but pCO₂ cannot be reflected → consider ABG when patient has respiratory distress ^[20]. In other words, SpO₂ alone cannot tell you about ventilation, acid-base status, or the degree of shunting. Always get an ABG in suspected ARDS.

Role: First-line imaging — part of the Berlin Definition criteria.

CXR is always done to detect any obvious lung abnormalities ^[20]. In ARDS, look for:

LVEDP increase → blood backflow → more pressure in pulmonary vasculature → upper lobe venous diversion ^[21]. This pattern of cephalisation of pulmonary vessels is characteristic of cardiogenic pulmonary oedema, NOT ARDS. In ARDS, the vascular pattern is typically normal or obscured by the diffuse opacities.

Pitfalls of CXR in ARDS:

• CXR lags behind clinical deterioration by 12–24 hours — a patient can have severe hypoxaemia before the CXR looks bad
• Supine AP films (common in ICU) magnify the heart size and can overestimate congestion
• CXR is normal/near normal in some causes of hypoxaemia → ABG is still indicated if desaturation or clinical distress → consider PE (most important), lower airway diseases, ILD, other shunts ^[20]

Role: Not required for Berlin Definition diagnosis, but provides:

• More sensitive detection of bilateral opacities (especially early or subtle ARDS)
• Better characterisation of distribution pattern (helps differentiate direct vs indirect ARDS)
• Identification of underlying cause (e.g., pneumonia, lung abscess, empyema)
• Exclusion of alternative diagnoses (PE with CTPA protocol, ILD pattern, lymphangitic carcinomatosis)
• Guidance for management (recruitable vs non-recruitable lung on CT)

Role: Critical for excluding cardiogenic pulmonary oedema ^[9].

Important nuance: ARDS and cardiac dysfunction CAN co-exist. Sepsis can cause both ARDS and septic cardiomyopathy. The Berlin Definition says "not fully explained by cardiac failure" — partial contribution is allowed. Clinical judgement is required.

Additional echo findings in ARDS:

• Acute cor pulmonale: RV dilatation with interventricular septal bowing into LV (D-shaped LV) — occurs in ~25% of ventilated ARDS patients from high PEEP and hypoxic pulmonary vasoconstriction
• This is a complication of ARDS, not a differential diagnosis

Role: Helps distinguish cardiogenic from non-cardiogenic pulmonary oedema.

BNP is released by heart stress → not specific for heart failure; there are non-cardiac causes including lung disease, renal impairment, and anaemia ^[10]. So interpret with clinical context. A very low BNP is most useful (high negative predictive value for HF).

Increasingly used at bedside in ICU:

The 2023 Updated Global Definition now accepts bilateral B-lines or consolidation on lung USS as imaging evidence of ARDS, particularly useful in resource-limited settings or when CXR is impractical.

Role: Not diagnostic of ARDS, but essential to exclude cardiac causes of bilateral pulmonary oedema.

Role: Historically used to measure PCWP (pulmonary capillary wedge pressure) to distinguish cardiogenic (PCWP > 18 mmHg) from non-cardiogenic (PCWP ≤ 18 mmHg) pulmonary oedema.

• The Berlin Definition no longer requires routine PA catheterisation
• However, it can be useful in difficult cases where the distinction between cardiogenic and non-cardiogenic oedema is unclear
• In ARDS: PCWP ≤ 18 mmHg (normal left atrial pressure)
• In cardiogenic pulmonary oedema: PCWP > 18 mmHg (elevated left atrial filling pressure)

Why did the Berlin Definition drop this requirement? Because PA catheterisation is invasive, carries risks (arrhythmia, pneumothorax, PA perforation), and echocardiography can provide equivalent information non-invasively in most cases.

In ARDS specifically, PFTs are not performed during the acute illness (the patient is too sick and often intubated). However, they may be relevant in:

• Recovery phase: documenting residual restrictive defect or ↓ DLCO
• Differential diagnosis: if considering ILD → reduced FVC and ↓ DLCO ^[24]
• Assessment of underlying disease: e.g., in anti-synthetase syndrome with ILD presenting as acute respiratory failure

Supplemental O₂ should be supplied to all patients. Oxygenation should be monitored continuously with pulse oximetry. Intubation and mechanical ventilation may be required ^[27][28]

• Immediate high-flow O₂ — target SpO₂ ≥ 88–95% initially
• Sit patient upright (if haemodynamically stable) — improves FRC and V/Q matching
• Prepare for escalation — have intubation equipment at bedside

CXR and ABG should be obtained following initial stabilisation to evaluate for ARDS ^[27][28]

Assessment of perfusion: measure BP for hypotension, look for signs of end-organ hypoperfusion ^[27][28]

If the patient is in septic shock (the most common cause of ARDS):

• Early goal-directed therapy (EGDT) targets ^[27][28]:
  • Urine output ≥ 0.5 mL/kg/hour
  • MAP ≥ 65 mmHg
  • CVP 8–12 mmHg (if central access available)
  • ScvO₂ ≥ 70% (if central access available)
• Fluid resuscitation with crystalloids (30 mL/kg within first 3 hours for sepsis)
• Vasopressors (noradrenaline first-line) if MAP < 65 mmHg despite fluids

In ARDS, the lung is heterogeneously affected:

• Dependent (posterior) regions: consolidated, flooded, non-recruitable
• Non-dependent (anterior) regions: aerated, relatively normal, but much smaller than in a healthy lung
• The aerated lung available for ventilation is only ~1/3 of normal — this has been called the "baby lung"

If you ventilate with normal tidal volumes (8–10 mL/kg), the entire volume goes into this small "baby lung" → overdistension → stretching of alveolar walls → mechanical injury → release of inflammatory mediators → worsening of the original lung injury. This is volutrauma.

Simultaneously, if alveoli repeatedly collapse during expiration and re-open during inspiration (tidal recruitment/derecruitment), the shear forces at the interface between open and collapsed alveoli cause further injury. This is atelectrauma.

Both processes amplify the inflammatory cascade → cytokine release into the systemic circulation → biotrauma → contributing to MODS and death.

The landmark ARDSNet ARMA trial (2000) showed that low tidal volume ventilation (6 mL/kg PBW) reduced mortality from 39.8% to 31.0% compared to traditional tidal volumes (12 mL/kg PBW). This is one of the strongest findings in critical care medicine.

Suggested initial ventilator settings for ARDS ^[30]:

Calculating predicted body weight (PBW):

• Male: PBW (kg) = 50 + 0.91 × (height in cm − 152.4)
• Female: PBW (kg) = 45.5 + 0.91 × (height in cm − 152.4)
• Example: For a 170 cm male → PBW = 50 + 0.91 × 17.6 = 66 kg → TV = 6 × 66 = 396 mL (~400 mL)

NIV is less efficacious or even harmful in ARDS ^[1][30][31]:

Why is NIV problematic in ARDS?

• ARDS patients have high minute ventilation demands → NIV may not deliver adequate support
• NIV cannot guarantee controlled tidal volumes → risk of patient self-inflicted lung injury (P-SILI) from large spontaneous breaths
• High transpulmonary pressure swings during spontaneous breathing with NIV may worsen lung injury
• Cannot protect the airway if patient deteriorates rapidly
• Delays intubation → associated with worse outcomes

Contraindications for NIV include: unable to protect airway (unconscious, confusion), haemodynamic instability, high aspiration risk (vomiting, excessive secretions), recent facial/upper GI/upper airway surgery/trauma/burns ^[1]

When might NIV/HFNC still be considered?

• Mild ARDS (P/F 200–300) with close monitoring
• HFNC (high-flow nasal cannula) at ≥ 30 L/min — the FLORALI trial showed HFNC may reduce intubation rates in hypoxaemic respiratory failure; accepted in the 2023 updated ARDS definition
• Patient decided not for intubation (palliative)
• Must be in a monitored setting with rapid access to intubation if patient deteriorates

What it is: Turning the patient face-down for ≥ 16 hours per day while mechanically ventilated.

Why it works (from first principles):

• In the supine position, the weight of the heart and mediastinum compresses the dependent (posterior) lung → more atelectasis dorsally
• In prone position, this compressive effect is redistributed → the previously dependent lung opens up
• V/Q matching improves because perfusion distribution is less affected by position change than ventilation → more alveoli are now ventilated that were previously collapsed
• More homogeneous distribution of transpulmonary pressure → less VILI
• Improved secretion drainage from dependent airways

Evidence: The PROSEVA trial (2013) showed that early prone positioning (within 36h of ARDS onset) for ≥ 16h/day in severe ARDS (P/F < 150) reduced 28-day mortality from 32.8% to 16.0% (absolute risk reduction ~17%).

Prone positioning is the second most impactful intervention in ARDS after lung-protective ventilation. It should be considered standard of care in severe ARDS.

What: Short-term (48h) infusion of a neuromuscular blocking agent (typically cisatracurium) in early severe ARDS.

Why it helps:

• Eliminates patient-ventilator dyssynchrony → prevents large spontaneous breaths that cause P-SILI
• Reduces oxygen consumption by respiratory muscles
• Improves chest wall compliance → lower plateau pressures
• May have direct anti-inflammatory effects

Evidence: The ACURASYS trial (2010) showed improved 90-day survival with 48h cisatracurium in early severe ARDS. The later ROSE trial (2019) did NOT confirm this benefit with a lighter sedation strategy — the current consensus is to use NMBA selectively (e.g., when P-SILI is suspected, when proning is difficult, or when ventilator dyssynchrony persists despite deep sedation).

Rationale: In ARDS, the alveolar-capillary membrane is leaky. Any excess intravascular fluid will leak into the alveoli and worsen pulmonary oedema. A conservative (restrictive) fluid strategy after initial resuscitation reduces the hydrostatic driving force for oedema.

Evidence: The FACTT trial (2006) showed that a conservative fluid strategy (targeting CVP < 4 or PCWP < 8) improved oxygenation and shortened duration of mechanical ventilation compared to a liberal strategy, without increasing organ failure.

Current evidence (2025):

• Moderate certainty evidence suggests that early, prolonged, low-dose corticosteroids (e.g., dexamethasone 20 mg/day for 5 days then 10 mg/day for 5 days, or methylprednisolone 1–2 mg/kg/day tapered over 2–3 weeks) may reduce mortality and duration of mechanical ventilation in moderate-severe ARDS
• The DEXA-ARDS trial (2020) and the COVID-19 RECOVERY trial (dexamethasone in COVID-ARDS) both supported benefit
• Do NOT start corticosteroids after day 14 of ARDS — late initiation may increase mortality

Agents: Inhaled nitric oxide (iNO) or inhaled epoprostenol (prostacyclin).

Mechanism: Selectively vasodilate pulmonary vessels in ventilated (aerated) alveoli → redirects blood flow to ventilated regions → improves V/Q matching → improves oxygenation.

Evidence: Improve oxygenation temporarily but no proven mortality benefit. Used as rescue therapy or bridge to other interventions.

Bridge to transplantation by extracorporeal life support (ECLS) i.e. ECMO ^[2]

What: An external circuit drains deoxygenated venous blood → passes through an oxygenator membrane → returns oxygenated blood to the venous system → effectively acts as an "external lung" while the native lung heals.

Why: Allows the ventilator settings to be drastically reduced ("lung rest") → minimises VILI → gives the lung time to heal.

Evidence: The EOLIA trial (2018) showed a trend towards improved mortality with VV-ECMO in severe ARDS, with significant crossover benefit — current consensus is to consider ECMO at experienced centres for refractory severe ARDS.

What: Transient application of high airway pressures (e.g., sustained inflation at 35–40 cmH₂O for 40 seconds, or staircase recruitment) to open collapsed alveoli.

Rationale: If collapsed alveoli can be opened and then kept open with adequate PEEP, oxygenation improves.

Evidence: Mixed results — the ART trial (2017) showed that aggressive recruitment manoeuvres with high PEEP may increase mortality. Current approach is to use gentle, individualised recruitment with careful monitoring for barotrauma and haemodynamic compromise.

• Goal: maintain comfort while allowing the ventilator to deliver lung-protective settings
• Strategy: target light sedation (Richmond Agitation-Sedation Scale [RASS] 0 to −2) when possible
• Daily sedation interruption ("sedation holidays") to reduce duration of MV
• Exceptions: deep sedation may be needed when proning or using NMBA

• ARDS patients are critically ill, immobilised, and often have systemic inflammation → high VTE risk
• LMWH (enoxaparin 40 mg SC daily) or UFH (5000 U SC Q8–12H) unless contraindicated
• Mechanical prophylaxis (graduated compression stockings, intermittent pneumatic compression) if anticoagulation contraindicated

• ICU patients on mechanical ventilation are at high risk of stress-related GI haemorrhage
• PPI (pantoprazole 40 mg IV daily) or H₂ receptor antagonist (ranitidine 50 mg IV Q8H)
• Only while on mechanical ventilation; discontinue when feeds established

• Early enteral nutrition (within 24–48h of ICU admission) is preferred
• Trophic feeds initially (10–20 mL/h), advanced as tolerated
• Parenteral nutrition only if enteral route fails or contraindicated
• Avoid overfeeding (excess CO₂ production worsens ventilatory demand)

• Target blood glucose 6–10 mmol/L (avoid tight control < 6 — NICE-SUGAR trial showed ↑ mortality)
• Insulin infusion for critically ill patients with hyperglycaemia

• As soon as haemodynamically stable and not deeply sedated
• Passive → active range of motion → sitting → standing → walking
• Reduces ICU-acquired weakness, delirium, and length of stay

This is the most important iatrogenic complication — and understanding it is the entire rationale for lung-protective ventilation.

Types of VILI:

This is why lung-protective ventilation (low TV, adequate PEEP, Pplat ≤ 30 cmH₂O) is the cornerstone of ARDS management — it doesn't treat the disease; it prevents the treatment from making things worse.

A direct consequence of positive pressure ventilation + damaged, heterogeneous lung tissue.

Barotrauma occurs in ~10% of mechanically ventilated ARDS patients, even with lung-protective strategies. The heterogeneous lung means some regions are overdistended while others are collapsed — the overdistended regions are at highest risk of rupture.

This is the feared long-term pulmonary complication.

Why it occurs: In the fibroproliferative phase ( > 7 days), fibroblasts proliferate within the alveolar space and interstitium. If the balance tips towards fibrogenesis (driven by TGF-β, PDGF, CTGF) rather than resolution, dense collagen is deposited throughout the lung parenchyma.

Clinical consequences:

• ↓ Lung compliance → persistent respiratory failure → difficulty weaning from mechanical ventilation
• ↓ Gas exchange surface area → chronic hypoxaemia
• Restrictive lung physiology on PFTs: ↓ FVC, ↓ TLC, ↓ DLCO
• Traction bronchiectasis on HRCT
• Pulmonary hypertension from vascular destruction and remodelling

Risk factors for fibrotic progression:

• Prolonged ARDS (especially if > 14 days of mechanical ventilation)
• Severe initial injury
• VILI (ongoing mechanical damage perpetuates the fibrotic cycle)
• Late steroid initiation (after day 14 — may paradoxically worsen fibrosis)
• Genetic predisposition

Prognosis: Most ARDS survivors show significant improvement in lung function over 6–12 months. However, some have persistent restrictive defects and impaired DLCO for years.

Incidence: ~10–30% of intubated ARDS patients develop VAP.

Why ARDS patients are particularly vulnerable:

• Endotracheal tube bypasses upper airway defences (mucociliary clearance, cough reflex)
• Pooling of oropharyngeal secretions above the cuff → micro-aspiration around the cuff
• ARDS itself impairs local pulmonary immune defences (alveolar macrophage dysfunction, surfactant loss)
• Prolonged mechanical ventilation = prolonged exposure to nosocomial organisms
• Sedation and immobility suppress cough reflex

Common organisms:

• Early-onset ( < 5 days): Streptococcus pneumoniae, Haemophilus influenzae, MSSA
• Late-onset ( > 5 days): MDR organisms — Pseudomonas aeruginosa, MRSA, Acinetobacter baumanii ^[32]

Clinical features: Fever, rigors, cough with purulent sputum, ↓ air entry, crackles, bronchial breathing ^[7]; new or worsening CXR infiltrates; ↑ WCC; ↑ tracheal secretions

Diagnosis: Clinical suspicion + quantitative or semi-quantitative cultures of lower respiratory tract specimens (endotracheal aspirate or BAL)

Prevention:

• Head of bed elevation ≥ 30–45°
• Oral hygiene with chlorhexidine (controversial in 2025 — some guidelines no longer recommend)
• Subglottic secretion drainage (specialised ETT with suction port above cuff)
• Minimise duration of mechanical ventilation (daily spontaneous breathing trials)
• Avoid unnecessary re-intubation
• Sedation interruption protocols

Complications of pneumonia include respiratory failure, septicaemia with multi-organ failure, lung abscess, and parapneumonic effusion ^[33]. VAP superimposed on ARDS dramatically worsens prognosis.

ARDS patients are at high risk for VTE/PE because of:

• Critical illness → endothelial activation, hypercoagulable state
• Immobilisation
• Central venous catheters
• Pulmonary microvascular thrombosis (part of ARDS itself — in-situ thrombosis within the damaged pulmonary vasculature)

Prevention with pharmacological VTE prophylaxis (LMWH or UFH) is essential in all ARDS patients unless contraindicated.

Incidence: ~20–25% of mechanically ventilated ARDS patients.

Why it occurs (from first principles):

The right ventricle (RV) is a thin-walled, compliant chamber designed to pump against the normally low-resistance pulmonary circulation. In ARDS, multiple factors conspire to increase RV afterload:

1. Hypoxic pulmonary vasoconstriction (HPV): widespread hypoxia in ARDS → generalised pulmonary arteriolar constriction → ↑ pulmonary vascular resistance (PVR)
2. Microvascular thrombosis: intravascular coagulation and fibrin deposition in pulmonary capillaries
3. Vascular compression: oedematous alveoli and high PEEP compress small pulmonary vessels
4. Vascular remodelling: in the fibroproliferative phase, intimal fibrosis and medial hypertrophy of pulmonary arterioles
5. Mechanical ventilation: high PEEP and high mean airway pressures increase intrathoracic pressure → compress alveolar vessels → ↑ PVR

The thin-walled RV cannot cope with the acutely increased afterload → it dilates → the interventricular septum bows into the LV ("D-shaped LV") → this impairs LV filling → ↓ cardiac output → haemodynamic collapse.

Clinical features:

• Hypotension, tachycardia
• ↑ JVP, ↑ CVP
• Hepatomegaly (acute congestion)
• Echo: RV dilatation, interventricular septal bowing, ↓ TAPSE, ↑ RV/LV ratio

Management:

• Optimise PEEP (avoid excessive PEEP which worsens RV afterload)
• Prone positioning (improves RV function by improving oxygenation and reducing PVR)
• Inhaled pulmonary vasodilators (iNO, inhaled prostacyclin) — selectively reduce PVR
• IV inotropes (dobutamine, milrinone) if haemodynamically significant
• Avoid fluid overloading the RV

ARDS patients frequently have haemodynamic instability from:

• Underlying cause (septic shock is the most common)
• Positive pressure ventilation → ↑ intrathoracic pressure → ↓ venous return → ↓ preload → ↓ cardiac output
• High PEEP amplifies this effect → "PEEP-induced hypotension"
• RV failure (see above)
• Sedation/anaesthesia → vasodilation
• Dehydration from conservative fluid strategy

Incidence: ~40–70% of ARDS patients develop AKI.

Mechanisms:

• Pre-renal: hypotension/shock → ↓ renal perfusion
• Intrinsic: sepsis-mediated tubular injury; nephrotoxic drugs
• Positive pressure ventilation: ↑ intrathoracic pressure → ↓ venous return → ↑ renal venous pressure → ↓ renal perfusion pressure; also activates RAAS → sodium and water retention
• Conservative fluid strategy: may occasionally undershoot and cause pre-renal AKI (this is a trade-off — too much fluid worsens pulmonary oedema; too little fluid impairs renal perfusion)

Management: Monitor UO and creatinine closely; adjust nephrotoxic drugs; renal replacement therapy if indicated (fluid overload, refractory hyperkalaemia, uraemia, severe metabolic acidosis) ^[34]

Common in sepsis-driven ARDS. DIC is a consumptive coagulopathy:

• Systemic activation of coagulation → widespread microthrombi → consumption of clotting factors and platelets → paradoxical bleeding
• Presentations: bleeding from IV sites, catheters, mucosal surfaces; ischaemia of extremities (microvascular thrombosis); petechiae
• Lab: ↓ platelets, ↑ PT/aPTT, ↓ fibrinogen, ↑ D-dimer ^[35]
• Management: treat the underlying cause (most important); supportive with platelet transfusion, FFP, cryoprecipitate as needed ^[35]

Includes two overlapping entities:

Risk factors:

• Prolonged mechanical ventilation ( > 7 days)
• Corticosteroid use (especially when combined with NMBA)
• Neuromuscular blockade
• Sepsis/MODS
• Hyperglycaemia
• Immobility

Clinical impact:

• Difficulty weaning from mechanical ventilation (diaphragmatic weakness)
• Prolonged rehabilitation
• Persistent functional impairment months to years after ICU discharge

Prevention: Minimise sedation (daily sedation interruption), minimise NMBA duration ( ≤ 48h), early mobilisation, glucose control, adequate nutrition.

Incidence: 50–80% of mechanically ventilated patients.

Mechanisms: Multifactorial — sedative/opioid exposure, sleep disruption, metabolic derangements, sepsis, hypoxia, social isolation.

Impact: Associated with ↑ mortality, prolonged MV, prolonged ICU stay, long-term cognitive impairment.

Prevention/management: ABCDEF bundle (Assess pain, Both SAT + SBT, Choice of sedation, Delirium monitoring, Early mobilisation, Family engagement).

Mechanism: Critical illness → splanchnic hypoperfusion + corticosteroid use + coagulopathy → erosion of gastric mucosa → bleeding.

Prevention: PPI or H₂ receptor antagonist while mechanically ventilated.

• Catheter-related bloodstream infection (CRBSI): from central venous catheters
• Catheter-associated urinary tract infection (CAUTI): from indwelling urinary catheters
• Clostridioides difficile colitis: from broad-spectrum antibiotic exposure
• Prevention: aseptic insertion, daily review of line necessity, prompt removal of unnecessary devices

Most ARDS survivors recover near-normal lung function by 12 months. The DLCO is the most sensitive residual abnormality and the last to normalise.

• Incidence: ~25–30% of ARDS survivors at 1 year
• Manifestations: memory deficits, impaired executive function, ↓ processing speed, ↓ attention
• Mechanisms: hypoxic brain injury, sedative exposure, delirium during ICU stay, neuroinflammation
• Comparable severity to mild traumatic brain injury or early Alzheimer's disease

• Muscle wasting (from immobility, catabolism, steroids, NMBA)
• Weight loss
• Reduced 6-minute walk distance (may persist for > 5 years)
• Many survivors unable to return to previous occupation at 1 year