eLetters to:

COMMENTARY: Iain Chalmers Why We Need to Know Whether Prophylactic Antibiotics Can Reduce Measles-Related Morbidity Pediatrics 2002; 109: 312-315 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

eLetters published:

Prophylactic antibiotics for uncomplicated measles

Frank Shann   (19 March 2002)

WHO expert advisers need to promulgate consistent advice, based on evidence

Iain Chalmers   (2 April 2002)

Prophylactic antibiotics for uncomplicated measles 19 March 2002
Frank Shann, Director of Intensive Care Royal Children's Hospital, Melbourne Send letter to journal: Re: Prophylactic antibiotics for uncomplicated measles shannf{at}cryptic.rch.unimelb.edu.au Frank Shann	Dr Chalmers argues, correctly, that it would be helpful to know whether we should give prophylactic antibiotics to children with measles. Presumably, he means children in developing countries where there is high morbidity and mortality from measles (1). WHO already recommends that children with measles be given antibiotics if they have otitis media, pneumonia, neurological complications, or malnutrition (2). This means that there is already a long list of indications for antibiotic treatment: otitis media, tachypnoea, subcostal retraction, nasal flaring, grunting, cyanosis, inability to feed, severe vomiting, convulsions, lethargy, coma, weight- for-height <70%, or edema of both feet (2). Almost all children who are admitted to hospital with measles will have at least one of the above indications for antibiotic therapy. The controversy about the routine use of prophylactic antibiotics for children with uncomplicated measles relates only to children who do not have otitis media, pneumonia, neurological complications, or malnutrition. Such children will have much lower morbidity and mortality rates than children who are admitted to hospital with measles. A controlled trial of routine antibiotic prophylaxis would need 9,200 children in each group to detect a 20% reduction in morbidity from 5% to 4% with a two-sided alpha error of 0.05 and a beta error of 0.1. It would be difficult to justify a trial of the effect of antibiotics on morbidity (rather than mortality) in communities where 10% or more of babies die before they are five years old. In my meta-analysis, all the controlled trials of prophylactic antibiotics for children with measles were performed before the WHO indications for antibiotic therapy had been clearly enunciated (3). Almost all the children in the trials were inpatients, and a high proportion will have had one or more of the WHO indications for antibiotic therapy. Despite this, only one of the 637 children in the control group died (1.6 per 1000 cases) compared to four of the 764 children given prophylactic antibiotics (5.2 per 1000 cases); the odds ratio was 4.0 (mid -P corrected 95% confidence interval 0.5 to 101.6). Dr Chalmers suggests that routine use of prophylactic antibiotics for measles will not contribute significantly to the development of antibiotic resistance. However, WHO strongly endorses the view that antibiotic use in people and animals in the community, as well as use in hospitals, increases the prevalence of resistance (4). A dose of measles vaccine costs the same as a course of five days of cotrimoxazole – about US$0.17 from UNICEF. Prophylactic cotrimoxazole for uncomplicated measles will have a very limited impact on total child mortality. In contrast, measles immunization reduces total mortality by about 50% in developing countries (probably because it reduces mortality from other diseases as well as measles) (5). It would be better to increase coverage with measles vaccine rather than divert resources into prophylactic antibiotics for children with uncomplicated measles. It would certainly be helpful to know whether routine prophylactic antibiotic therapy benefits children with measles when they do not have otitis media, pneumonia, neurological complications or malnutrition. However, only very limited resources are available to perform the large controlled trial that would be required, and more useful information is likely to be obtained from trials of vaccines in developing countries, and trials of the treatment of established pneumonia in the presence of antibiotic-resistant organisms. Frank Shann Intensive Care Unit, Royal Children’s Hospital, Melbourne, Australia 1. Chalmers I. Why we need to know whether prophylactic antibiotics can reduce measles-related morbidity. Pediatrics 2002109;312-315. 2. World Health Organization. Management of the child with a serious infection or severe malnutrition: guidelines for care at the first- referral level in developing countries. WHO/FCH/CAH/00.1. Geneva: WHO, 2000. 3. Shann F. Meta-analysis of trials of prophylactic antibiotics for children with measles: inadequate evidence. BMJ 1997;314:334-7. 4. World Health Organization. Global strategy for containment of antimicrobial resistance. WHO/CDS/CSR/DRS/2000.1. Geneva: WHO, 2000. 5. Aaby P, Samb B, Simondon F, Coll Seck AM, Knudsen K, Whittle H. Non- specific beneficial effects of measles immunisation: analysis of mortality studies from developing countries. BMJ 1995; 311: 481-485.
WHO expert advisers need to promulgate consistent advice, based on evidence 2 April 2002
Iain Chalmers, Researcher UK Cochrane Centre Send letter to journal: Re: WHO expert advisers need to promulgate consistent advice, based on evidence ichalmers{at}cochrane.co.uk Iain Chalmers	I am glad that Professor Shann acknowledges that prophylactic antibiotics should be given to some children with measles in whom there is no firm evidence of bacterial infection. I hope he will now ensure that his Cochrane review (1) is amended to make this clear, citing the evidence on which this recommendation is based. The section on measles in the WHO report to which he refers, and for which he was an adviser(2), mentions only Vitamin A as a treatment for measles. The issue of prophylactic antibiotics is not addressed; indeed the only explicit reference to antibiotics is the recommendation that benzylpenicillin should be given if measles is complicated by mouth ulcers. For advice on the management of other complications, readers are referred to other sections of the 180-page report. Thus, for example, those seeking advice on the management of measles croup are referred to the section on croup, where they are unhelpfully simply referred back to the section on measles! I made the point in my article that the trustworthiness of WHO recommendations depends on the scientific quality of the process on which they are based. Sadly, there is no reference to the research evidence supporting the recommendations in the WHO report cited by Professor Shann (2). Furthermore, the section on measles does not reflect the recommendation published by a WHO expert group which, among possible research projects on the treatment of measles, accorded highest priority to a controlled trial of prophylactic antibiotics in measles(3). It is disappointing that Professor Shann has chosen to ignore my reference to this WHO recommendation. It is clearly time for WHO expert advisers on the treatment of measles to make up their minds, preferably on the basis of evidence rather than opinion! References 1. Shann F, D'Souza RM, D'Souza R. Antibiotics for preventing pneumonia in children with measles (Cochrane Review). In: The Cochrane Library, Issue 1, 2002. Oxford: Update Software. 2. World Health Organization. Management of the child with a serious infection or severe malnutrition: guidelines for care at the first- referral level in developing countries. WHO/FCH/CAH/00.1. Geneva: WHO, 2000. 3. World Health Organisation. Clinical research on treatment of measles: report of a meeting. WHO/CDR/95.15, Geneva: WHO, 1995.