Letters

Effectiveness of glucocorticoids in treating croup

Authors acknowledge Cochrane Collaboration

General practitioners must be ready to treat children

Suitable formulations of oral glucocorticoids are available in primary care

Children with croup should receive corticosteroids in primary care: results of audit

Authors acknowledge Cochrane Collaboration

EDITORIt has been brought to my attention that we omitted some key acknowledgments in our recently published systematic review of glucocorticoids for the treatment of croup.¹ The protocol for our review was registered with the Acute Respiratory Infection Review Group of the Cochrane Collaboration in 1997. Our review is currently under revision with the review group and will be published in the Cochrane Database of Systematic Reviews. The Acute Respiratory Infection Review Group was very helpful in searching its database of trials identified through handsearching of relevant journals and in providing insightful comments on our protocol.

One advantage of having our study published in the Cochrane Database of Systematic Reviews is that we will be able to update the review as new evidence is identified. The BMJ's policy of permitting duplicate publication with the Cochrane Collaboration is extremely important. I hope that other journals join in this support of the Cochrane Collaboration through like minded policies.

Terry P Klassen, chair. 
Department of Paediatrics, University of Alberta, 2C3.67 Walter C Mackenzie Health Sciences Centre, Edmonton, Alberta, Canada T6G 2R7 terry.klassen{at}ualberta.ca

General practitioners must be ready to treat children

EDITORThe meta-analysis by Ausejo et al is timely as the croup season approaches.¹ Despite the usefulness of corticosteroids having been recognised for many years, our experience is that children rarely receive them in any form before admission to hospital.

Geelhoed et al showed the benefit of a small single dose of dexamethasone for outpatient croup in 1996,² and Godden et al confirmed the usefulness of nebulised budesonide.³ Despite this, doctors still seem to be reluctant to treat mild croup, which usually lasts three to four days if left untreated.

Our review of the case notes of children admitted with croup between November 1996 and October 1997 identified 51 cases in a total paediatric population of 65 000. General practice referrals accounted for three fifths of admissions (31/51); none of the children admitted as general practice referrals had received either inhaled or systemic steroid apart from one child seen at a primary care casualty department, who received hydrocortisone.

Although formal croup scores were not recorded on admission, most cases were mild and 46 of the children were discharged within 16 hours of admissionthat is, the next morning if admitted at night or the same day if admitted in the early hours of the day. The cough lasted for over eight hours in 28 cases, which suggests that there is an opportunity for treatment if parents are encouraged to seek help. As discussed by Thomson in her commentary³ in referring to the paper of Geelhoed et al,² dexamethasone is as cheap as, easier to administer than, and as effective as nebulised corticosteroids. It should perhaps be added to the list of "black bag essentials"albeit that the main determinant of treatment or indeed admission to hospital may be the degree of parental anxiety generated by caring for a child with stridor in the middle of the night.

Angela J Tillett, specialist registrar in paediatrics. 
Department of Paediatrics, Addenbrooke's NHS Trust, Cambridge CB2 2QQ lawr{at}globalnet.co.uk

James D M Gould, consultant paediatrician. 
Department of Child Health, Ipswich Hospital NHS Trust, Ipswich IP4 5PD

Suitable formulations of oral glucocorticoids are available in primary care

EDITORThe meta-analysis of Ausejo et al shows that glucocorticoids improve the course of croup in children who are seen in emergency departments or admitted to hospital.¹ Many children with croup are not referred to hospital, so how are doctors in primary care to respond?

I see no reason to suppose that children with croup should not benefit from steroids just because they are in the community, but unfortunately dexamethasone is available only as tablets in the community in the United Kingdom. This is not a suitable formulation for small children.

Nebulised budesonide is available but is expensive and inconvenient and reaches its expiry date alarmingly fast. As a result in our practice we use soluble betamethasone tablets, which are equipotent to dexamethasone and can be dissolved in a drink. I have also observed that they have the advantage of tasting more palatable than soluble prednisolone tablets.

The paper that prompted our change in practice was published in the BMJ in 1996.² For the past two years we have used a single oral dose of 0.15 mg/kg (prescribed as 500 µg soluble betamethasone tablets to dissolve in a drink) to treat children in the community with croup.

Others who wish to use oral steroids for children with croup in the community may find this information on available formulations useful while we all wait for a much needed randomised controlled trial of glucocorticoids in croup in primary care.

Christopher Cates, general practitioner. 
Manor View Practice, Bushey Health Centre, London Road, Bushey, Hertfordshire WD2 2NN chriscates{at}emailmsn.com

Children with croup should receive corticosteroids in primary care: results of audit

EDITORCorticosteroids can reduce the symptoms and duration of stay in hospital for children with croup (acute laryngotracheobronchitis).¹ We would like to bring the results of our audit on the management of children with croup to the attention of colleagues working in primary care.

We studied 101 case records of children with croup assessed in Leicester Royal Infirmary (July to December 1998). Of 58 children referred by their general practitioners, 18 had received antibiotics, six salbutamol, and two steam inhalation. Steroids were not given to any of the children. Of the 101 children, 78 were admitted, and, of these, 56 stayed less than 24 hours. The need for inpatient care might have been avoided in this group if corticosteroids had been given earlier in the illness.

It is disappointing that none of the children received steroids before their arrival at the hospital. There is clear evidence that oral dexamethasone or nebulised budesonide results in clinical improvement in outpatients with mild to moderate croup, reducing the need for admission. ^{2 3} One oral dose of dexamethasone (0.15 mg/kg) is effective in reducing the need for further medical care in mild croup.⁴ Prednisolone has not been widely studied, but equivalent doses (1 mg/kg) would be useful.⁵

Well designed trials have found a dramatic reduction in symptoms of airway obstruction as early as one hour after treatment. Substantial gains are achieved by treating all children with croup with steroids: fewer transfers to intensive care, shorter hospital stays, and cost savings.

We recommend that all children with croup should be given one oral dose of dexamethasone or prednisolone as soon as the diagnosis is made. When the diagnosis is in doubt or the clinical condition warrants admission to hospital, children should be referred in the usual way.

S Jothimurugan, senior house officer. 
Jothidevi{at}hotmail.com

Zoaka Hassan, honorary clinical fellow. 
M Silverman, professor of child health. 
University Department of Child Health, Leicester Royal Infirmary, PO Box 65, Leicester LE2 7LX