Definition and Overview
Respiratory Failure
Respiratory failure in children is classified by mechanism:
| Type | Mechanism | Example |
|---|---|---|
| Type I (hypoxaemic) | $V/Q$ mismatch, shunt, diffusion impairment | PARDS, bronchiolitis, pneumonia |
| Type II (hypercapnic) | Alveolar hypoventilation | Status asthmaticus, neuromuscular disease, upper airway obstruction |
| Mixed | Combined | Severe bronchiolitis, late-stage ARDS |
Paediatric ARDS (PARDS): Montreux/PALICC-2 Definition
The paediatric-specific ARDS definition differs from the Berlin adult criteria in important ways:
- Age: excludes patients with perinatal-origin lung disease (e.g. neonatal RDS, MAS)
- Timing: within 7 days of a known clinical insult
- Imaging: new infiltrate(s) consistent with acute parenchymal disease on CXR or CT; bilateral infiltrates not required in mild disease
- Oxygenation: uses $\text{OI}$ (oxygenation index) and $\text{OSI}$ (oxygen saturation index) in addition to $PaO_2/FiO_2$ ratio, because arterial blood gas is not always available in children
$$\text{OI} = \frac{FiO_2 \times \text{MAP} \times 100}{PaO_2}$$
$$\text{OSI} = \frac{FiO_2 \times \text{MAP} \times 100}{SpO_2}$$
| Severity | On invasive MV | OI | OSI |
|---|---|---|---|
| Mild | Yes | 4-8 | 5-7.5 |
| Moderate | Yes | 8-16 | 7.5-12.3 |
| Severe | Yes | $\geq 16$ | $\geq 12.3$ |
- Non-invasive ventilation (NIV) and high-flow nasal cannula (HFNC) patients may be classified as "at-risk" using $SpO_2/FiO_2$ ratio
- Left atrial hypertension must not be the primary cause (exclude cardiogenic pulmonary oedema), though PARDS can coexist with cardiac disease
Pathophysiology
Acute Lung Injury Cascade
- A direct (pulmonary) or indirect (systemic) insult triggers neutrophil and macrophage activation
- Inflammatory mediator release damages the alveolar-capillary membrane
- Protein-rich exudate floods the alveolar space, surfactant is inactivated or diluted
- Alveolar collapse, $V/Q$ mismatch, and intrapulmonary shunt drive refractory hypoxaemia
- Decreased compliance increases the work of breathing; the lung becomes stiff and heterogeneous
Paediatric Considerations
- Children have a higher chest-wall compliance relative to lung compliance; this predisposes to airway closure and de-recruitment at low lung volumes
- Smaller absolute lung volumes mean tidal volume excess causes greater volutrauma per unit lung mass
- Surfactant deficiency in prematurity amplifies injury; conversely, children $>$2 years have relatively more alveolar reserve than neonates
- Common precipitants in the PICU: viral pneumonia (RSV, influenza, SARS-CoV-2), bacterial sepsis (pneumococcal, staphylococcal), aspiration, near-drowning, immunosuppression
Clinical Features and Diagnosis
Recognition of Respiratory Failure
Early recognition is essential because respiratory arrest without prior intervention carries very high mortality in children.
Signs of increased work of breathing:
- Tachypnoea (rate varies by age: $>60$ breaths/min in neonates, $>50$ in infants, $>40$ in toddlers, $>30$ in older children)
- Subcostal, intercostal, sternal recession; nasal flaring; grunting
- Use of accessory muscles; head bobbing in infants
- Paradoxical see-saw breathing indicates impending failure
Signs of inadequate respiratory effort or fatigue:
- Declining respiratory rate, apnoea
- Decreasing effort despite worsening hypoxaemia
- Altered consciousness, inability to maintain airway
Haemodynamic impact: Hypoxaemia drives tachycardia; severe respiratory failure causes bradycardia and shock (pre-arrest pattern).
Chest Radiograph in PARDS
- Bilateral or unilateral (in mild) alveolar infiltrates not attributable purely to effusion, collapse, or nodules
- Severe PARDS: near-complete opacification ("whiteout")
- Distinguish from cardiomegaly and vascular redistribution of cardiac pulmonary oedema
Investigation and Monitoring
| Investigation | Purpose |
|---|---|
| Arterial blood gas | Quantify $PaO_2$, $PaCO_2$, pH; calculate OI |
| $SpO_2$ continuous | OSI-based severity if ABG unavailable |
| CXR (portable) | Confirm infiltrates, tube/line position |
| Full blood count, cultures, CRP, procalcitonin | Identify infectious aetiology |
| Respiratory virus PCR (NPA or BAL) | Targeted therapy; cohorting |
| Echocardiography | Exclude cardiogenic cause; assess RV function and pulmonary pressures |
| Lung ultrasound | Bedside; consolidation, B-lines, effusions |
| Bronchoscopy / BAL | Immunocompromised patients; atypical organisms |
Monitoring targets during MV for PARDS:
- $SpO_2$ 92-97% (target 88-95% acceptable in severe disease to limit $FiO_2$ toxicity)
- $PaCO_2$ 45-60 mmHg permissive hypercapnia (pH $>7.20$), unless contraindicated (intracranial hypertension, pulmonary hypertension, severe cardiac disease)
- Plateau pressure $\leq 28$ cmH$_2$O (up to 29-32 if chest wall compliance reduced, e.g. obesity, ascites)
- Driving pressure $\leq 15$ cmH$_2$O
Management
Step 1: Respiratory Support Escalation
Non-invasive support (mild/at-risk PARDS or pre-intubation stabilisation):
- High-flow nasal cannula (HFNC): flow rates up to 2 L/kg/min in infants, up to 50-60 L/min in older children; generates modest CPAP effect and washes dead space
- CPAP / BiPAP: appropriate for alert, cooperative children who can protect their airway
- Intubate if: progressive hypoxaemia on NIV/HFNC, fatigue, obtundation, apnoea, haemodynamic instability, inability to protect airway
Intubation:
- Pre-oxygenate maximally; in children with poor reserve this window is narrow
- ETT size (cuffed): $\text{age}/4 + 3.5$ mm ID (cuffed); verify with leak test and end-tidal CO$_2$
- RSI with ketamine 1-2 mg/kg IV or fentanyl 1-2 micrograms/kg IV + rocuronium 1.2 mg/kg IV (or suxamethonium 2 mg/kg IV in infants, 1.5 mg/kg in older children if no contraindication)
- Two-person technique, video laryngoscope available; have surgical airway backup plan declared before induction
Step 2: Lung-Protective Ventilation (Core Strategy)
The principle is to avoid volutrauma, barotrauma, atelectrauma, and oxygen toxicity while maintaining adequate gas exchange.
| Parameter | Target |
|---|---|
| Mode | Volume-controlled or pressure-controlled (both acceptable) |
| Tidal volume ($V_T$) | 5-8 mL/kg ideal body weight (IBW); 5-6 mL/kg in severe PARDS |
| PEEP | Moderate-to-high per $FiO_2$/PEEP table; typically 8-15 cmH$_2$O |
| Plateau pressure ($P_{plat}$) | $\leq 28$ cmH$_2$O |
| Driving pressure | $\leq 15$ cmH$_2$O |
| Respiratory rate | Age-appropriate; up to 35 breaths/min tolerated if no air-trapping |
| $FiO_2$ | Wean to $\leq 0.6$ as soon as targets met; avoid $>0.8$ prolonged |
| $PaCO_2$ / pH | Permissive hypercapnia if pH $>7.20$; manage acidaemia before paralysis |
PEEP titration:
- Use lung recruitment manoeuvres cautiously (risk of haemodynamic compromise)
- Titrate PEEP to best compliance or oxygenation; the $FiO_2$/PEEP table approach adapted from ARDSNet provides a pragmatic starting point
- Higher PEEP may worsen haemodynamics: monitor cardiac output/perfusion closely, especially in right heart dysfunction
IBW for tidal volume calculation:
- Boys: $IBW = 50 + 0.91 \times (\text{height in cm} - 152.4)$ kg (not appropriate below ~12 years)
- In younger children, use measured weight or length-based estimates; IBW approximates actual weight in healthy children of normal habitus
Step 3: Adjunctive Measures for Refractory Hypoxaemia
In escalating order of invasiveness:
- Optimise ventilator settings (PEEP, $V_T$, FiO$_2$, I:E ratio)
- Neuromuscular blockade (NMB): reduces ventilator dyssynchrony and patient self-inflicted lung injury (P-SILI); cisatracurium infusion (0.1-0.2 mg/kg/hr) or vecuronium; reassess daily for continued need
- Prone positioning: 16+ hours/day; evidence in adult severe ARDS; used in severe PARDS; requires experienced nursing and appropriate weight/size; contraindications include open abdomen, unstable spine, elevated ICP
- Inhaled nitric oxide (iNO): 10-20 ppm; improves $V/Q$ matching and reduces pulmonary vascular resistance; use in PARDS with concurrent pulmonary hypertension or refractory hypoxaemia as a bridge; does not improve survival but may allow time for lung recovery or ECMO cannulation
- High-frequency oscillatory ventilation (HFOV): may be considered as rescue in severe PARDS where conventional ventilation cannot maintain acceptable plateau pressures; set mean airway pressure above the closing pressure; evidence base is mixed; requires expertise
- ECMO (VV-ECMO): for severe, refractory PARDS (OI $>40$ despite optimal management); circuit sized by weight; flow rates 80-150 mL/kg/min; cannulation strategy varies by age and centre expertise; continue lung rest ventilation on ECMO; ECMO does not treat the underlying lung disease but provides time for recovery
Step 4: Supportive and Disease-Specific Management
Fluid strategy:
- After initial resuscitation (septic shock: 10-20 mL/kg isotonic crystalloid bolus, reassess; avoid aggressive fluid loading in PARDS without haemodynamic indication)
- Conservative fluid balance once haemodynamically stable reduces duration of ventilation
- Target euvolaemia; consider furosemide 0.5-1 mg/kg IV if fluid overloaded
Sedation and analgesia:
- Titrate to adequate comfort and synchrony; avoid over-sedation impairing respiratory drive recovery
- First-line: morphine 10-30 micrograms/kg/hr IV infusion or fentanyl 1-2 micrograms/kg/hr; midazolam 0.05-0.2 mg/kg/hr IV or dexmedetomidine 0.2-0.7 micrograms/kg/hr
- Use validated pain/sedation scales (COMFORT-B, FLACC, Richmond Agitation-Sedation Scale adapted for children)
- Sedation holidays once FiO$_2$ $<0.6$ and haemodynamically stable; daily wake-up trials unsafe in severe PARDS on NMB
Nutrition:
- Enteral nutrition within 24-48 hours of intubation if haemodynamically stable; gastroparesis is common, use prokinetics and post-pyloric feeding if needed
- Target 1.5-2 g/kg/day protein; calorimetry preferred if available
Antibiotics: treat identified or strongly suspected infection promptly by weight; broaden empirically in immunocompromised patients
Surfactant: not routinely recommended in PARDS; may be considered in specific contexts (post-RSV ARDS in infants, direct inhalation injury) but evidence is inconsistent
Corticosteroids: not standard; may benefit children with specific aetiologies (immune-mediated pneumonitis, ARDS in the context of haematological malignancy); early use in COVID-19 associated severe PARDS supported by extrapolation from adult data
Complications and Special Considerations
Ventilator-Induced Lung Injury (VILI)
Four overlapping mechanisms:
- Volutrauma: overdistension from excessive $V_T$
- Barotrauma: high peak/plateau pressures causing pneumothorax, pneumomediastinum
- Atelectrauma: repetitive opening and closing of unstable lung units
- Biotrauma: mechanical stress generates inflammatory mediators that amplify systemic injury
PARDS Phenotypes
Recognition is emerging that PARDS is heterogeneous; hyperinflammatory and hypo-inflammatory subtypes (similar to adult ARDS phenotyping) may respond differently to PEEP and fluid strategies. This is not yet actionable at the bedside but is exam-relevant.
Pulmonary Hypertension Complicating PARDS
- Common in severe PARDS; assess by echo
- iNO, avoid hypoxia, hypercapnia, and acidaemia (pulmonary vasoconstrictors)
- Milrinone 0.25-0.75 micrograms/kg/min IV improves RV contractility; avoid in hypotension without adjunctive vasopressor
The Immunocompromised Child
- Avoid unnecessary bronchoalveolar lavage delays; test broadly for bacterial, viral, fungal, and pneumocystis causes simultaneously
- Lower threshold for non-invasive support before intubation; once intubated, mortality is significantly higher
- Engage haematology/oncology early for disease-specific immunosuppression adjustment
Neonatal Interface
- Neonatal ARDS is recognised as a distinct entity; born at $\geq$36 weeks gestation presenting within 72 hours; separate from classic neonatal RDS
- Surfactant is indicated in RDS of prematurity; less clear evidence in term neonatal ARDS
- Lung-protective targets apply but $V_T$ 4-6 mL/kg with higher PEEP tolerance in neonatal circuits
- iNO indicated for hypoxic respiratory failure with pulmonary hypertension in near-term and term neonates (standard of care)
PICU and SPPE Viva Considerations
Approach to the viva question on PARDS management:
- State the PALICC-2 criteria confidently, including OSI/OI thresholds
- Distinguish mild/moderate/severe and match escalation accordingly
- Always frame ventilation decisions around lung protection (low $V_T$, moderate PEEP, low driving pressure)
- Know your adjuncts in order: NMB, prone, iNO, HFOV, ECMO
- Articulate the ECMO trigger (OI $>40$, failure of optimised conventional management) and the role of your state/regional ECMO service and retrieval pathway
- Acknowledge family communication and goals-of-care conversations, particularly for children with severe PARDS and malignancy
Common viva traps:
- Using adult $PaO_2/FiO_2$ thresholds without acknowledging OI/OSI in children
- Forgetting IBW for $V_T$ calculation in an obese adolescent
- Proposing surfactant universally without caveats
- Not addressing pulmonary hypertension when asked about refractory hypoxaemia