Respiratory failure in children

Contents

Classification

Type I — Hypoxaemic

  • PaO₂ < normal for age (or SpO₂ <94% on air)
  • Normal or low PaCO₂
  • Mechanism: V/Q mismatch, intrapulmonary shunt, impaired diffusion
  • Examples: pneumonia, ARDS, pulmonary oedema

Type II — Hypercapnic (Ventilatory Failure)

  • PaCO₂ >6.5 kPa (>50 mmHg) with respiratory acidosis (pH <7.35)
  • Mechanism: inadequate alveolar ventilation — muscle weakness, central drive failure, severe obstructive disease with air trapping
  • Examples: neuromuscular disease, severe bronchiolitis, status asthmaticus, central hypoventilation

Mixed (Type I + Type II)

Common in children as respiratory muscle fatigue supervenes during prolonged Type I failure.

Causes by Age

Age group Common causes
Neonate Respiratory distress syndrome (surfactant deficiency), transient tachypnoea of the newborn, meconium aspiration, congenital diaphragmatic hernia, pneumonia, congenital heart disease
Infant (1–12 months) Bronchiolitis (RSV — most common), pneumonia, pertussis, congenital heart disease, BPD (bronchopulmonary dysplasia) exacerbation
Toddler / pre-school Viral croup, asthma (first presentation), pneumonia, foreign body aspiration, epiglottitis (rare — vaccinated), anaphylaxis
School age Asthma, pneumonia, pleural effusion, myocarditis with pulmonary oedema, neuromuscular disease (SMA, DMD exacerbation)
Adolescent As per adult: asthma, pneumonia, PE (rare), anorexia-related respiratory failure, neuromuscular

Assessment of Respiratory Failure

Work of Breathing Markers

In children (unlike adults), the chest wall is highly compliant — increased work of breathing is most evident as:

  • Tachypnoea — most sensitive sign; exceeds age-specific norms early
  • Subcostal, intercostal, suprasternal recession — soft tissue retracts inwards during inspiration due to high negative intrathoracic pressure
  • Nasal flaring — particularly in infants
  • Grunting — expiratory grunting (auto-PEEP effect from partially closed glottis) indicates severe respiratory failure, particularly in neonates
  • Head bobbing — in infants with severe respiratory distress; accessory muscle use
  • Paradoxical chest movement (see-saw) — in neuromuscular weakness; chest falls while abdomen rises in inspiration

Signs of Impending Respiratory Arrest

  • Decreasing respiratory rate (exhaustion — not improvement)
  • Decreasing work of breathing despite worsening gas exchange (false reassurance)
  • Reduced consciousness, inability to maintain alertness
  • Pale or dusky appearance, central cyanosis
  • Weak cry, inability to drink/feed

Respiratory Support Escalation

1. Supplemental Oxygen

  • Low-flow nasal cannula: up to 2L/min in infants (higher flows cause discomfort and ineffective delivery)
  • Face mask with non-rebreather (NRBM): 10–15L/min; achieves FiO₂ 0.6–0.7

2. High-Flow Nasal Cannula (HFNC / Optiflow)

  • Warmed, humidified high-flow O₂ (up to 2–3 L/kg/min); generates low levels of CPAP (1–5 cmH₂O)
  • Widely used in bronchiolitis and viral pneumonia in infants
  • PARIS RCT (Franklin, NEJM 2018): in bronchiolitis, HFNC at ED presentation reduced escalation to ICU-level care compared to standard O₂, though no difference in the primary outcome (duration of O₂ requirement)
  • Current practice: HFNC as a step below CPAP/NIV; reduces intubation rates in bronchiolitis

3. CPAP / BiPAP (Non-Invasive Ventilation)

  • CPAP: continuous positive airway pressure; splints alveoli open; reduces FRC loss; mainly for Type I failure
  • BiPAP: inspiratory + expiratory pressure support; used for Type II failure (neuromuscular disease, bronchiolitis with hypercapnia, obstructive sleep apnoea)
  • Masks: infant-specific masks or nasal prongs; careful fitting critical

4. Intubation and Mechanical Ventilation

Indications:

  • GCS ≤8 or rapidly falling consciousness
  • Exhaustion — decreasing respiratory effort despite adequate support
  • Haemodynamic instability
  • Failure of non-invasive support (persistent hypoxia SpO₂ <90% on max FiO₂, rising PaCO₂ with acidosis)
  • Airway compromise (stridor, neck mass, epiglottitis)

Paediatric intubation considerations:

  • Uncuffed vs cuffed ETT: cuffed ETTs are now used down to neonates in most UK centres; appropriate sizing critical (age/4 + 4 for uncuffed; age/4 + 3.5 for cuffed)
  • Cricoid is the narrowest airway in young children (vs subglottis in adults)
  • RSI: atropine 20 mcg/kg (prevents bradycardia from suxamethonium in children <1 year); fentanyl 3 mcg/kg; propofol or ketamine; suxamethonium or rocuronium
  • Apnoeic oxygenation: nasal cannula O₂ 1L/kg/min during laryngoscopy extends the safe apnoea time

Lung-Protective Ventilation

Same principles as adults (ARDS Network): tidal volume 6 mL/kg ideal body weight; PEEP sufficient to maintain oxygenation; plateau pressure <30 cmH₂O; permissive hypercapnia if needed to limit lung injury.

Paediatric ARDS (PARDS)

Paediatric Berlin Definition (PALICC 2015)

Unlike adults (Berlin definition requires PaO₂/FiO₂ ratio), PARDS uses the oxygenation index (OI) or OSI (SpO₂-based) to accommodate non-invasively ventilated patients and infants in whom ABG is not always obtained:

  • OI = (FiO₂ × mean airway pressure × 100) / PaO₂
  • OSI = (FiO₂ × mean airway pressure × 100) / SpO₂ (for non-ventilated or SpO₂-based monitoring)
Severity OI OSI
Mild 4–8 5–7.5
Moderate 8–16 7.5–12.3
Severe ≥16 ≥12.3

Management follows similar principles to adult ARDS: lung-protective ventilation, prone positioning, careful fluid balance.

High-Frequency Oscillatory Ventilation (HFOV)

HFOV delivers very small tidal volumes (~1–3 mL/kg) at very high rates (3–15 Hz) superimposed on a constant mean airway pressure (MAP). It reduces volutrauma and maintains lung volume.

Evidence in Adults

The OSCILLATE trial (see Journal Club) randomised 571 adults with moderate-severe ARDS to HFOV vs conventional ventilation (NEJM 2013). HFOV significantly increased in-hospital mortality (47% vs 35%). HFOV was subsequently abandoned as a rescue strategy in adult ARDS.

Evidence in Children

The adult evidence should not be extrapolated to paediatrics:

  • No large RCT of HFOV in children has been conducted
  • UK PICU practice: HFOV is used as rescue therapy for severe PARDS (OI ≥16) not responding to conventional lung-protective ventilation and adjuncts
  • Observational data in children suggest HFOV can improve oxygenation; evidence of mortality benefit is inconclusive
  • Some infants and neonates have different physiology — smaller airways, surfactant deficiency, and congenital lung anomalies may respond differently to HFOV

Current Practice (UK PICU)

HFOV is considered a rescue strategy in severe PARDS after failure of:

  • Optimised conventional ventilation (low TV, PEEP titration)
  • Prone positioning
  • Inhaled nitric oxide

Specific Conditions

Bronchiolitis

  • Peak incidence 1–3 months; RSV most common pathogen
  • Supportive: O₂, HFNC, NG feeding; antibiotics not indicated (viral)
  • Intubation for: apnoeas, severe hypoxia, exhaustion
  • No evidence for bronchodilators, nebulised hypertonic saline (in hospital), or corticosteroids

Viral Croup (Laryngotracheobronchitis)

  • Dexamethasone 0.15–0.6 mg/kg IM/oral (single dose): evidence-based; reduces oedema; single dose as effective as multiple doses
  • Nebulised adrenaline: temporary relief for severe croup (Westley croup score ≥8); effect wears off in 2 hours — patient must be observed for rebound
  • Heliox: may reduce turbulent flow in severe croup; limited evidence; not standard

Status Asthmaticus

  • Salbutamol 2.5–5 mg nebulised or 15 mcg/kg IV bolus (severe)
  • IV magnesium sulfate 40 mg/kg (max 2 g): bronchodilation; evidence from RCTs
  • IV aminophylline: refractory cases; monitor for arrhythmia; limited evidence in paediatrics
  • Ketamine infusion: bronchodilator + analgesic + anaesthetic; useful if intubation needed
  • Heliox: reduces turbulent flow; bridge to other therapies
  • Intubation in asthma is high-risk: severe air trapping + haemodynamic compromise on induction; prepare for peri-intubation cardiovascular collapse

Viva Questions

1. How does the assessment of respiratory failure differ between children and adults?

Children have more compliant chest walls than adults — the ribs are cartilaginous and the intercostal muscles relatively weak in infants. This means that negative intrathoracic pressure during inspiration causes soft tissue recession (subcostal, intercostal, suprasternal) that is visible and provides an important clinical sign of increased work of breathing. Adults develop less visible recession because their stiffer chest wall resists inward movement. This means that in children, recession is a key clinical sign of severity — its presence, location, and degree must be assessed. Grunting in infants indicates severe respiratory failure (glottis partially closed on expiration to generate auto-PEEP and maintain FRC). The age-specific normal ranges for respiratory rate are critical — what constitutes tachypnoea in a 2-month-old (>60/min) is very different from a 10-year-old (>25/min). Paradoxically, a child who appears less distressed after a period of escalating distress may simply be exhausted — a deceasing respiratory rate with worsening gas exchange is a pre-arrest sign. Assessment of respiratory failure in children requires combining vital signs, work of breathing signs, mental state, and ability to feed/cry — a composite picture rather than relying on any single parameter.

2. Can the OSCILLATE trial evidence on HFOV in adults be applied to paediatric practice?

No, not directly. OSCILLATE (2013) showed HFOV increased mortality in adult moderate-severe ARDS compared to conventional ventilation (47% vs 35%), and this led to abandonment of routine HFOV in adult ARDS. However, there are important reasons why paediatric physiology and practice differ. First, no equivalent large RCT exists in children — we simply do not have paediatric-level evidence from which to draw the same conclusions. Second, paediatric lung disease has different aetiologies (bronchiolitis, viral ARDS, surfactant deficiency in neonates) with different pathophysiology from adult ARDS. Third, in the OSCILLATE trial, the control arm used a sophisticated lung-protective ventilation protocol — many paediatric practice comparisons are with less-optimised conventional ventilation. In current UK paediatric ICU practice, HFOV is a rescue strategy for severe PARDS (OI ≥16) failing optimal conventional management, prone positioning, and inhaled NO — not a first-line approach. The adult data justifiably prevents HFOV from being a default strategy, but paediatricians appropriately retain it as a salvage option for the most severely affected children.

3. A 6-week-old infant with bronchiolitis is admitted with a respiratory rate of 80, subcostal recession, SpO₂ 88% on 2L/min O₂ via nasal cannula. Describe your management escalation.

Start with immediate supplemental O₂ escalation: increase to high-flow nasal cannula (HFNC) at 2 L/kg/min (so approximately 6–8 L/min for this infant) — warmed and humidified; this generates a small amount of CPAP, reduces work of breathing, and improves oxygenation. Target SpO₂ 94–98%. NG tube: feeding is impaired by breathing difficulty — do not try to feed orally; NG feeds or IV dextrose to prevent hypoglycaemia. Regular reassessment: if SpO₂ improves to >94%, recession decreases, RR improves — continue on HFNC with close monitoring. If inadequate response (SpO₂ <92% on maximum HFNC, increasing recession, decreasing RR from exhaustion, GCS fall): escalate to CPAP via infant prongs or mask (CPAP 6–8 cmH₂O with 0.4–0.5 FiO₂). Blood gas if still deteriorating — check PaCO₂ (rising CO₂ indicates ventilatory failure); if PaCO₂ rising with acidosis: intubation is needed. Call PICU early — do not wait for arrest; earlier transfer to PICU with ongoing HFNC support is safer than emergency transfer in respiratory arrest. Antibiotics are not indicated in pure viral bronchiolitis unless co-infection is suspected (fever, focal consolidation). No bronchodilators routinely — not shown to benefit.