The term “croup” refers to the clinical syndrome of a hoarse voice, harsh barking cough and inspiratory stridor.1-6 The commonest cause of this symptom complex is viral laryngotracheobronchitis (LTB). In line with common usage, our use of the term “croup” in this article implies viral LTB.
Other causes of upper-airway obstruction must be considered in the differential diagnosis (Box 1). The presence of biphasic stridor (unlike the variable inspiratory stridor seen in croup) should alert the clinician to the likelihood of fixed airway narrowing (eg, due to vocal cord palsy, subglottic haemangioma, subglottic stenosis).
Spasmodic croup, occurring in older, often atopic, children, is treated similarly to viral croup, as it is the severity of upper-airway obstruction at the time of presentation that determines the treatment.
Croup is a common cause of upper-airway obstruction in young children, occurring in about 2% of preschool-aged children annually.1 It mainly affects children aged 6–36 months, with a peak incidence at 12–24 months. There is a male predominance of 3:2.1 The condition is usually mild and self-limiting, although it may occasionally cause severe respiratory obstruction. Viral infection of the upper airway results in inflammation of the pharynx, larynx, trachea and bronchi. However, it is specifically subglottic inflammation (and resultant swelling) that compromises the airway in croup. Typically, viral croup develops over several days along with a concurrent coryzal illness. The symptoms of airway obstruction disappear over 3–5 days.3-5 A number of viruses may cause croup, the most common being parainfluenzavirus 1 and 2 and respiratory syncytial viruses.1-4
The use of corticosteroids to treat croup has dramatically reduced both the number of patients requiring hospital admission and the rates of endotracheal intubation.4-6 Here we evaluate the current evidence for treatment of croup.
Determining the degree of airway obstruction (based primarily on the history) is the most important consideration when assessing children with croup. As airway obstruction can worsen rapidly, repeated careful clinical assessment is essential. Various other factors, some unrelated to the immediate clinical features, may affect the likelihood of a child with croup being admitted to hospital (Box 2).
General appearance. A child who is agitated, is tiring from the effort of breathing, or whose level of consciousness is decreasing, must immediately be given supplemental oxygen via a facemask, nebulised adrenaline and systemic corticosteroids. The child should be closely observed, and nebulised adrenaline treatment should be repeated if there is no reduction in upper-airway obstruction within minutes. Such a child may rapidly deteriorate and require intubation.
Degree of respiratory distress. Stridor at rest, tracheal tug, chest-wall retractions, changing respiratory rate and pulse rate or palpable pulsus paradoxus indicate moderate to severe croup.
Cyanosis or extreme pallor indicates the need for immediate supplemental oxygen (repeated), nebulised adrenaline and intubation.
Oxygen desaturation, indicated by oximetry, is usually a late and unreliable sign of severity.12 Occasionally, it may be present in less severe croup.
Mild airway obstruction can be assumed when the child appears happy and is prepared to drink, eat, play and take an interest in the surroundings. Most children presenting to general practitioners will fall within this category. As illustrated in Scenario 1 (Box 3), there may be mild chest-wall retractions and mild tachycardia, but stridor at rest will not be present. The GP should reassure the parents, explain what to expect over the coming days, and give them an information sheet. A single dose of oral corticosteroid may be warranted, especially if the child has an additional factor that increases the likelihood of hospital admission (Box 2).
If a child with moderate airway obstruction begins to appear worried, preoccupied or tired, sometimes sleeping for short periods, the child will require treatment, close observation in hospital and frequent clinical review (Box 3, Scenario 2). Progression of signs will indicate the need for further treatment, which may include intubation.
Oximetry is never a substitute for good clinical assessment. Oxygen saturation may be close to the normal level in a child with severe croup, or, conversely, substantially lowered in a child with mild to moderate croup.12 This variability presumably relates to ventilation–perfusion mismatching caused by lower-airway disease.
Croup severity scores have been used in hospital-based clinical research studies to assess the suitability of patients for treatment in a standardised manner.7,8,14,15 They give a cumulative score, grading for the degree of stridor, chest-wall retractions, air entry, cyanosis, dyspnoea and level of consciousness. However, they are of limited value in clinical practice.
Major changes in croup management over the past 25 years include the introduction of nebulised adrenaline and the more liberal use of systemic and nebulised corticosteroids.9,10,16-26 Over the past decade, work has focused on corticosteroid treatment, and its utility can be seen in the levels of evidence available to ascertain its effectiveness (Box 4).
The commonly used medications for croup are summarised in Box 5. A flow chart summarising an approach to assessing and treating croup in the emergency department setting is presented in Box 6.
As illustrated in Scenario 1 (Box 3), mild croup is often less symptomatic during the daytime, only to become worse at night. However, the GP is often called upon to decide whether a young child warrants specific treatment for the potential deterioration that may occur subsequently. The treatment choice is simple: a dose of oral corticosteroid or not. The decision about whether to treat is less clearcut. In fact, there is no evidence available from studies in the general practice setting to guide the clinician, only recommendations based upon patient populations presenting to emergency departments. Such patients may not be representative of those presenting to GPs.13 Fortunately, the levels of evidence favouring the use of systemic corticosteroids in the emergency department setting are generally sound.
The mechanisms by which corticosteroids exert their effects are not known. It is presumed that corticosteroids possess rapidly acting anti-inflammatory or vasoconstrictive properties in the upper airway.8,14,19 Studies using oral dexamethasone,21,23,25 nebulised budesonide (2 mg/dose)15,18 and nebulised dexamethasone26 have demonstrated the efficacy of corticosteroids compared with placebo in relieving croup symptoms in the hospital setting. Commonly used alternatives to dexamethasone are prednisone or prednisolone.6,16
Oral corticosteroids are preferred in most paediatric emergency departments because they are inexpensive, easy to administer, readily available, and produce measurable improvements within hours (E2).17 Doses of dexamethasone ranging from 0.15 to 0.6 mg/kg have been shown to be similarly efficacious for treating croup (E2).25 For simplicity, and to avoid confusion when prescribing, commonly recommended doses of dexamethasone are 0.15–0.3 mg/kg and doses of prednisolone suspension or prednisone tablets are 1–2 mg/kg (E4).4
Two meta-analyses,3,6 which included 24 randomised controlled trials (RCTs), have shown clinically significant benefit (based on clinical croup severity scores) of systemic corticosteroids given at 6 and 12 hours, but not at 24 hours (E1). Reviewing the meta-analyses together, the quality of the studies included was considered good, although it was noted that the numbers of participants were generally small (median, 40).4 The possibility of selection bias was raised, as a funnel plot demonstrated few published studies showing no effect of systemic corticosteroids.4
The use of systemic corticosteroids has also been associated with a significant decrease in the number of adrenaline nebuliser treatments required and a reduced average length of stay in the emergency department (E1–E4).3,4,6,9,13,28 Furthermore, since the earlier use of systemic corticosteroids in children attending emergency departments was advocated, the need for endotracheal intubation has been reduced and the duration of intubation has fallen (E1, E2).13,16,28
A number of trials comparing the method of delivery of corticosteroids have shown that oral, intravenous, intramuscular and inhaled (nebulised) forms of corticosteroid are all superior to placebo (E1, E2).3,6,9,13,15-18,21-23,25,26 Oral and intramuscular dexamethasone 0.6 mg/kg are equally efficacious for relieving the symptoms of croup at 24 hours and 10 days (E2).22 In addition, oral dexamethasone 0.6 mg/kg has been shown to be superior to nebulised dexamethasone 160 μg or placebo in mild croup, treated in an emergency department setting, for reducing symptoms and rates of subsequent attendance for medical review within a week.9 Most paediatricians prefer the oral route, as it is inexpensive, easy to administer and kinder for the patient (E4).4
The use of steam inhalation for treating croup has been advocated since the 19th century,5 but the treatment has never been scientifically validated. Three studies evaluating the use of supersaturated cold air (“cold water fog”) to treat croup were unable to find evidence of benefit. One was an underpowered RCT in which 16 subjects with croup were randomly allocated to receive room air or a humidified atmosphere for 12 hours in hospital (E3).29 The second trial involved only 7 subjects, who showed no improvement in respiratory resistance after receiving 2 mL of nebulised cold sterile water (E4).30 The third trial, a double-blind RCT involving 71 children aged 3 months to 6 years with moderately severe croup, showed no benefit from mist (supersaturated air) therapy when given in addition to oral dexamethasone (0.6 mg/kg, given to all patients) in terms of croup score, oxygen saturation, heart rate or respiratory rate over 2 hours (E2).31
As illustrated in Scenario 2 (Box 3), a child with moderate to severe croup may present in the evening to an emergency department. In this situation, a dose of oral corticosteroid should be given immediately while the child is assessed to determine whether more rapid relief of the upper-airway obstruction is required.
Oxygen is the most important treatment for a child with moderate or severe croup who has considerable upper-airway obstruction with significant oxygen desaturation (Spo2 [pulsatile oxygen saturation] < 90%) (E4).11 There have been no RCTs of oxygen therapy. Oxygen is the initial treatment before the administration of pharmacological treatment in the hospital setting. Theoretically, the relief of hypoxia should increase respiratory muscle endurance, increase lung oxygen reserves and make intubation under anaesthesia, where needed, a safer procedure (E4).
A child with persisting inspiratory stridor at rest and marked chest-wall retractions has severe croup. Such a child, who may be severely obstructed without being centrally cyanosed, should receive immediate treatment with nebulised adrenaline (1:1000 dilution at a dose of 0.5 mL/kg of body weight to a maximum dose of 5 mL, delivered neat to the nebuliser bowl) (E1, E2).8,13,19,20,24 For a child still at home or in the GP’s surgery, this dose could be administered while awaiting the ambulance for transfer to hospital. In addition, a dose of oral corticosteroid should be given (E4).11 The child should be reassessed regularly.
The mechanism of action of nebulised adrenaline is uncertain. It has been suggested that nebulised adrenaline reduces bronchial and tracheal epithelial vascular permeability, thereby decreasing airway oedema, which, in turn, increases the airway radius and improves airflow (E4).4,8,19,20,24 The onset of action is rapid: double-blind RCTs of this treatment document a fall in croup symptom scores within 30 minutes (E2),8,24 and the effect lasts for about 2 hours (E2).4,8,24 The need for several doses of nebulised adrenaline in a short period of time highlights the need to consider urgent transfer to a paediatric centre and/or intubation (E4).4
The use of 2–4 mg of nebulised budesonide to treat croup (a treatment that began to attract attention during the 1990s) has been shown to be effective (E2).8,9,18 The onset of action is within 30 minutes,8 which compares favourably with systemically administered corticosteroids that have an effect within one hour (E2).21-23,25 (The response to systemic corticosteroids in croup at less than one hour after administration has not been assessed.)
Repeated doses of 2 mg nebulised budesonide 12th hourly for up to 48 hours in a hospital setting has been shown to reduce croup symptoms and relapse rates compared with placebo (E2).18 A recent meta-analysis of trials of inhaled corticosteroids for croup concluded that croup symptom scores were significantly lowered within 5 hours and that hospital admission rates were also reduced (E1).10 However, the cost of equipment and medication, the time needed for administration and the potential for the child to become upset when inhaled preparations are used mean that oral corticosteroids are the preferred option, unless a child repeatedly vomits the oral corticosteroid (E2, E4).4,9,11
Croup is common, but most children experience only mild symptoms. The question of whether all children presenting to a GP need an oral corticosteroid is controversial and untested. There are no RCTs on interventions for croup in primary care settings. If the primary purpose for giving oral corticosteroids to all children with croup in general practice were to prevent hospital admission (a readily measurable outcome), then the majority of children would be treated unnecessarily, as the risk of admission is low. However, if any subjective improvement, such as minimising cough and mild breathing discomfort, was considered an outcome justifying corticosteroid treatment, then one could argue that the risks of harm are low and the potential benefits to the child and parent warrant the treatment.
Children who present to emergency departments are routinely given a dose of corticosteroid to decrease croup severity, avoid admission to hospital, reduce the risk of admission to intensive care and reduce the risk of progression to endotracheal intubation. A minority with moderate to severe croup will need oxygen, nebulised adrenaline and repeated doses of corticosteroids. Only a small proportion of those presenting with moderate to severe croup will need intubation.
Perhaps the most important aspect of management in croup, which is beyond the scope of evidence-based practice, is to keep the young child near their parent for reassurance when assessing and treating the condition.
1: Causes of upper-airway obstruction in young children
Supraglottic |
Laryngeal/subglottic |
Tracheal |
|||||||||
Acute tonsillar enlargement (due to bacterial or viral infection) |
Viral croup |
Foreign body |
|||||||||
Foreign body |
Spasmodic croup |
Bacterial tracheitis |
|||||||||
Retropharyngeal abscess |
Foreign body |
Tumour (eg, anterior mediastinal lymphoma) |
|||||||||
Epiglottitis (rare) |
Bacterial tracheitis |
Trauma (eg, haematoma) |
|||||||||
Acute angioedema |
Laryngomalacia ± viral infection (in infants) |
Congenital abnormality |
|||||||||
|
Diphtheria |
|
|||||||||
|
Thermal/chemical injury |
|
|||||||||
|
Intubation trauma |
|
|||||||||
|
Laryngospasm (associated with neural disorder, hypocalcaemia, or reflux) |
|
|||||||||
2: Factors increasing the likelihood of hospital admission after initial treatment for croup3,4,6-11
History of severe obstruction before presentation
History of previous severe croup or known structural airway anomaly (eg, subglottic stenosis)
Age less than 6 months
Significant degree of respiratory distress (stridor at rest)
Inadequate fluid intake
Parental anxiety
Proximity of home to the hospital/transport issues
Re-presentation to the emergency department within 24 hours
Poor response to initial treatment
Uncertain diagnosis
3: Presentation with croup: two typical case scenarios
Scenario 1: A child with mild croup presents to a general practitioner at 10 am
“James”, a 20-month-old boy, is brought to a GP with a barking cough, slight fever and hoarse voice. He has had a restless night. There is a family history of croup but this is James’s first episode of croup and his mother is anxious. There is no significant past medical history. On examination, James is playful, smiling and interactive. He has a typical “seal bark” cough but no audible stridor. His chest is clear on auscultation, there is no tracheal tug or chest-wall recession, and his respiratory rate is normal. He has mild croup and needs no immediate treatment. His mother asks whether James will get worse tonight. The GP replies that it is possible, and discusses with James’s mother the arguments for and against giving a single dose of oral corticosteroid.
For corticosteroid:
It will reduce even mild symptoms.
It will reduce the risk of hospital admission or reattendance.
It is relatively safe as a single dose.
It is easy to administer.
Against corticosteroid:
Mild symptoms do not require any treatment.
Before the popularisation of corticosteroids for treating croup, the risk of hospitalisation was low (about 1 in 80).1
The reduction in risk of reattendance within a week after a single dose of corticosteroid given in an emergency department is small. (In one study, 12 children needed to be treated with dexamethasone to avoid 1 reattendance within a week.13)
Parents may inappropriately use any remaining prednisolone to treat other illnesses.
Scenario 2: A child with moderate to severe croup presents to a hospital emergency department at 10 pm
“Simon”, a 23-month-old thriving boy, presents to the emergency department of the local hospital with viral croup. He has been hospitalised for croup on three previous occasions, twice having been admitted to the intensive care unit of a paediatric teaching hospital. Between bouts of croup, he is well. He has become agitated since waking 30 minutes earlier. He has loud inspiratory stridor at rest, is using his sternomastoid and abdominal muscles to aid breathing, and has obvious tracheal tug with palpable pulsus paradoxus. His parents gave him a dose of oral prednisolone earlier in the evening.
Simon needs immediate treatment with oxygen, nebulised adrenaline and further systemic corticosteroids.
4: Rating the evidence for recommendations
Our recommendations are based on the following levels of evidence, simplified from the National Health and Medical Research Council’s “Quality of evidence ratings”.27
E1 (Level I): Evidence from systematic review or meta-analysis of all relevant randomised controlled trials (RCTs).
E2 (Level II): Evidence from at least one well designed RCT.
E3 (Level III): Evidence from well designed cohort or case–control studies.
E4 (Level IV): Consensus opinion of respected authorities.
5: Pharmacological treatment of croup in the emergency department*
Medication |
Croup grade |
Dose |
Comments |
||||||||
Oxygen |
Severe (Spo2† < 90%), very severe with central cyanosis |
Minimum flow rate 4 L/min, given via facemask |
Give nebulised adrenaline and systemic corticosteroids as well. |
||||||||
Systemic corticosteroids (oral) |
Moderate, severe |
Dexamethasone 0.15–0.3 mg/kg OR prednisone/prednisolone 1–2 mg/kg |
Onset of action is within 1 hour. Treatment can be repeated after 12–24 hours. Intramuscular route can be used for a child who will not swallow or vomits repeatedly (this is uncommon). |
||||||||
Nebulised corticosteroids (if child repeatedly vomits oral corticosteroids) |
Moderate, severe |
Budesonide 2 mg (4 mL) undiluted |
Onset of action is within 30 minutes. Treatment can be repeated 12th hourly for 2 days. |
||||||||
Nebulised adrenaline 1:1000 |
Moderate, severe |
0.5 mL/kg, up to maximum of 5 mL nebulised |
Onset of action is within minutes. Give corticosteroids as well. Repeat doses may be needed. |
||||||||
* Regular review is essential to monitor response to therapy. † Pulsatile oxygen saturation. |
|||||||||||
- 1. Denny FW, Murphy TF, Clyde WA Jnr, et al. Croup: an 11-year study in a pediatric practice. Pediatrics 1983; 71: 871-876.
- 2. Marx A, Torok TJ, Holman RC, et al. Pediatric hospitalizations for croup (laryngotracheobronchitis): biennial increases associated with human parainfluenza virus 1 epidemics. J Infect Dis 1997; 176: 1423-1427.
- 3. Kairys SW, Olmstead EM, O’Connor GT. Steroid treatment of laryngotracheitis: a meta-analysis of the evidence from randomized trials. Pediatrics 1989; 83: 683-693.
- 4. Klassen TP. Croup: a current perspective. Pediatr Clin North Am 1999; 46: 1167-1178.
- 5. Skolnik NS. Treatment of croup: a critical review. Am J Dis Child 1989; 143: 1045-1049.
- 6. Ausejo M, Saenz A, Pham B, et al. The effectiveness of glucocorticoids in treating croup: meta-analysis. BMJ 1999; 319: 595-600.
- 7. Chin R, Browne GJ, Lam LT, et al. Effectiveness of a croup clinical pathway in the management of children with croup presenting to an emergency department. J Paediatr Child Health 2002; 38: 382-387.
- 8. Fitzgerald DA, Mellis CM, Johnson M, et al. Nebulized budesonide is as effective as nebulized adrenaline in moderately severe croup. Pediatrics 1996; 97: 722-725.
- 9. Geelhoed GC, Macdonald WB. Oral and inhaled steroids in croup: a randomized, placebo-controlled trial. Pediatr Pulmonol 1995; 20: 355-361.
- 10. Griffin S, Ellis S, Fitzgerald-Barren A, et al. Nebulised steroid in the treatment of croup: a systematic review of randomised controlled trials. Br J Gen Pract 2000; 50: 135-141.
- 11. Fitzgerald DA, Mellis CM. Management of acute upper airways obstruction in children. Mod Med Aust 1995; 38: 80-88.
- 12. Stoney PJ, Chakrabarti MK. Experience of pulse oximetry in children with croup. J Laryngol Otol 1991; 105: 295-298.
- 13. Osmond M. Croup. Clin Evid 2002 Dec; (8): 319-329.
- 14. Westley CR, Cotton EK, Brooks JG. Nebulized racemic epinephrine by IPPB for the treatment of croup: a double-blind study. Am J Dis Child 1978; 132: 484-487.
- 15. Husby S, Agertoft L, Mortensen S, Pedersen S. Treatment of croup with nebulized steroid (budesonide): a double blind, placebo controlled study. Arch Dis Child 1993; 68: 352-355.
- 16. Tibballs J, Shann FA, Landau LI. Placebo-controlled trial of prednisolone in children intubated for croup. Lancet 1992; 340: 745-748.
- 17. Klassen TP, Craig WR, Moher D, et al. Nebulized budesonide and oral dexamethasone for treatment of croup. JAMA 1998; 279: 1629-1632.
- 18. Roberts GW, Master VV, Staugas RE, et al. Repeated dose inhaled budesonide versus placebo in the treatment of croup. J Paediatr Child Health 1999; 35: 170-174.
- 19. Kelley PB, Simon JE. Racemic epinephrine use in croup and disposition. Am J Emerg Med 1992; 10: 181-183.
- 20. Prendergast M, Jones JS, Hartman D. Racemic epinephrine in the treatment of laryngotracheitis: can we identify children for outpatient therapy? Am J Emerg Med 1994; 12: 613-616.
- 21. Cruz MN, Stewart G, Rosenberg N. Use of dexamethasone in the outpatient management of acute laryngotracheitis. Pediatrics 1995; 96: 220-223.
- 22. Donaldson D, Poleski D, Knipple E, et al. Intramuscular versus oral dexamethasone for the treatment of moderate-to-severe croup: a randomized, double-blind trial. Acad Emerg Med 2003; 10: 16-21.
- 23. Super DM, Cartelli NA, Brooks LJ, et al. A prospective randomised double-blind study to evaluate the effect of dexamethasone in acute laryngotracheitis. J Pediatr 1989; 115: 323-329.
- 24. Waisman Y, Klein BL, Boenning DA, et al. Prospective randomised double-blind study comparing L-epinephrine and racemic epinephrine aerosols in the treatment of laryngotracheitis (croup). Pediatrics 1992; 89: 302-306.
- 25. Geelhoed GC, Macdonald WBG. Oral dexamethasone in the treatment of croup: 0.15 mg/kg versus 0.3 mg/kg versus 0.6 mg/kg. Pediatr Pulmonol 1995; 20: 362-368.
- 26. Luria JW, Gonzalez-del-Rey JA, DiGiulio GA, et al. Effectiveness of oral or nebulized dexamethasone for children with mild croup. Arch Pediatr Adolesc Med 2001; 155: 1340-1345.
- 27. National Health and Medical Research Council. How to use evidence: assessment and application of scientific evidence. February 2000. Table 1.3: Designation of levels of evidence. Available at: www.nhmrc.gov.au/publications/pdf/cp69.pdf (accessed Aug 2003).
- 28. Geelhoed GC. Sixteen years of croup in a Western Australian teaching hospital: effects of routine steroid treatment. Ann Emerg Med 1996; 28: 621-626.
- 29. Bourchier D, Dawson KP, Fergusson DM. Humidification in viral croup: a controlled trial. Aust Paediatr J 1984; 20: 289-291.
- 30. Lenney W, Milner AD. Treatment of acute viral croup. Arch Dis Child 1978; 53: 704-706.
- 31. Neto GM, Kentab O, Klassen TP, Osmond MH. A randomized controlled trial of mist in the acute treatment of moderate croup. Acad Emerg Med 2002; 9: 873-879.
Abstract
Croup affects about 2% of preschool-aged children every year.
Most children have mild croup and are managed at home, often after review by a general practitioner, who may decide that a single dose of oral corticosteroid is indicated (eg, if a risk factor for hospital admission exists).
A minority of children develop moderate or severe croup. They should be reviewed in an emergency department and may need hospital admission.
More liberal use of systemic corticosteroids for croup (in both primary care and emergency department settings) has been associated with reduced rates of hospital admission, reduced admissions to the intensive care unit and a reduced need for endotracheal intubation.
We discuss the assessment and evidence-based management of a child with mild croup presenting to a GP and a child with moderately severe croup presenting to an emergency department.
We present a flow chart summarising an approach to assessing and treating croup in the emergency department.