Intended for healthcare professionals



BMJ 1995; 310 doi: (Published 07 January 1995) Cite this as: BMJ 1995;310:4
  1. David Isaacs, Clinical associate professor
  1. Department of Immunology and InfectiousDiseases, Royal Alexandra Hospital for Children, Camperdown, NSW 2050, Australia

    Tachypnoea (>50 breaths/min) warrants admission to hospital

    Bronchiolitis is a pathological description that has come to be used as a clinical diagnosis. It is primarily a disease of the small airways, causing these to be obstructed by inflammatory exudate. More than 70% of cases are caused by respiratory syncytial virus, which in temperate climates results in a sharp winter epidemic lasting two to five months.1 Bronchiolitis is a disease of infancy, characterised by cough, fever, tachypnoea, diffuse crackles, hyperinflation, and chest retraction. Wheezes are a less constant feature,1 2 3 and bronchiolitis should be distinguishable clinically from infantile asthma by the presence of widespread crackles. Unfortunately, the diagnostic criteria for bronchiolitis have varied considerably, with consequent blurring of the distinction between it and asthma.4

    Over 95% of infants have been infected with respiratory syncytial virus by the end of their second winter; 40% of the infections in infancy affect the lower respiratory tract,1 2 3 4 5 although only about 1% of these children will need admission to hospital.1 The overall mortality from primary infection in previously healthy infants is low and has been estimated at from 1 in 5000 to 1 in 20000.6 The mortality among children admitted to hospital with respiratory syncytial virus infection is about 1% and is about 3.5% for those with underlying cardiac or chronic lung disease.7 Other high risk groups for severe infection are babies born before term1 4 and children with congenital or acquired immunodeficiency.8 9

    Transplacental maternal antibody confers at best partial immunity, so, although all adults have antibody to respiratory syncytial virus, babies can develop severe infection from birth. However, a “honeymoon period” exists up to 4 weeks of age, during which infection is relatively uncommon, perhaps because of some relative protection from maternal antibody or decreased exposure.1 The peak incidence is 2–5 monthsof age. If babies become infected in the first month, and particularly if they were born before term, apnoea may be the first sign of illness.10

    Reinfections with respiratory syncytial virus, of decreasing severity, occur throughout life.11 Although reinfections virtually never cause bronchiolitis, they are epidemiologically important in forming a reservoir of infection so that infants are infected for the first time by a school age sibling or an adult with a cold.6 The main mode of transmission of the virus to infants is probably through direct inoculation of nasal secretions on the hands of infected children or adults,2 although spread by fomites may also be important.12 Spread of infection through droplets seems to be less important.1 13

    The risk of the lower respiratory tract being affected in respiratory syncytial viral infection is increased by overcrowding, day care, and parental smoking and is reduced by breast feeding for longer than one month.4 14 Children admitted to hospital with bronchiolitis due to respiratory syncytial virus have about a 1 in 2 risk of later recurrent wheezing1 4 15 : some develop classic asthma, while others have bronchial hyperreactivity even after symptoms have resolved.16 It is unclear whether pre-existing atopy predisposes to severe bronchiolitis and later asthma or whether infection with respiratory syncytial virus damages the bronchial mucosa; allows the entry of, and sensitisation to, inhaled allergens; and thus “causes” asthma. Children infected with respiratory syncytial virus who are not admitted to hospital are not at increased risk of asthma.1 4

    The treatment of bronchiolitis is largely supportive. In severe cases treatment with oxygen can be lifesaving. Most studies, including some recent ones, have suggested that bronchodilators have no role and might even be deleterious.17 18 However, a trial of nebulised salbutamol for wheezy infants with a diagnosis of bronchiolitis, half of whom werepositive for respiratory syncytial virus, showed improvement, even in those under 6 months old.19 These varying results may be due to different diagnostic criteria, but they suggest that when wheeze predominates, a trial of nebulised salbutamol is indicated. Corticosteroids are of no benefit in bronchiolitis.4 Systemic bacterial superinfection is rare, even in severe infections with respiratory syncytial virus20; and antibiotics are not routinely indicated.

    In general practice, assessment of severity is critical. Mild cases, which do not necessitate admission to hospital, will generally occur in babies aged 3 months or more, born at term, who are feeding well and whose respiratory rate is <50 breaths/min. Moderate cases necessitate admission to hospital or at least management at home with small, frequent feeds and frequent observations. In these cases the babies are tachypnoeic (50–70 breaths/min), with only slight difficulty with feeds. In severe cases the babies are highly tachypnoeic (>70 breaths/min) or have apnoea, do not feed well, and need urgent admission. Clinical predictors of severity are tachypnoea (>70 breaths/min), age less than 3 months, preterm delivery (particularly before 34 weeks), and an “ill” or “toxic” appearance.21 Babies with difficulty in feeding or apnoea should be admitted, and it is probably wise to admit babies with underlying heart, lung, or immune problems who develop bronchiolitis. Cyanosis is a late and sinister sign.

    Pulse oximetry, if available, is the best predictor of the severity of the disease and need for oxygen.21 22 Given its valuable role in assessing children with bronchiolitis and, to a lesser extent, other respiratory diseases, general practitioners should strongly consider buying a pulse oximeter. The mortality associated with bronchiolitis is low, but babies can deteriorate rapidly and clinical skills are necessary to reduce the mortality and morbidity still further.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    View Abstract