Acute otitis media in children is importan

EDITOR—Cates's paper made the point that children with acute otitis media who are not particularly ill often get better quickly; they may be managed with analgesics and a deferred prescription for an antibiotic, which is not always redeemed.1 Unfortunately, the content of the paper did little to justify its title. Nowhere does Cates make any reference to the diagnostic criteria for otitis media in children or specify whether the same diagnostic criteria were used in both his practice and the control practice—surely the essence of a control group. We are not even told the age range of the children.

The notion of reducing antibiotic use in children with acute otitis media is not at all evidence based; indeed, the evidence is highly controversial if strict criteria for the diagnosis are upheld. This has been the subject of several meta-analyses, which have universally shown a positive if marginal advantage for the use of antibiotics in the primary management of children with acute otitis media.2 3 How children with acute viral upper respiratory tract infections with some associated otalgia should be managed is an entirely different issue.

Any attempt to improve diagnostic accuracy in childhood middle ear infections is to be welcomed, but it is singularly unhelpful if journal editors publish papers alleging to deal with acute otitis media when the meaning of that term is not specified. I would quibble with the use of the terms “acute otitis media,” “evidence based approach,” and “controlled” in Cates's paper.

I have a niggling feeling that this (admittedly worthwhile) audit would not have been published in the Papers section of the journal. It seems a shame that less rigorous scientific criteria should be applied to the peer review process for a general practice paper as would be the case with laboratory based or hospital research. If this is editorial practice then it does no credit to research endeavours in general practice and should be declared. I would be interested to see the reviewer's comments as to whether the entry criteria for this study were sufficiently rigorous to justify the term acute otitis media; I would also be interested to know whether the paper was reviewed by somebody who deals daily with pyogenic middle ear infections in children and with the often catastrophic consequences of such infections. The view that acute suppurative middle ear disease is a benign self limiting condition seems to be taking hold; often it is not.

Both acute otitis media and unexplained otalgia in children are important entities, and confusing the two simply muddies the waters.

Author's reply

EDITOR—Children with inflamed eardrums are usually described as having acute otitis media in day to day primary care, and the same diagnostic label was used in the trials included in the Cochrane review.1 The challenge that we face is to reduce unnecessary antibiotic use for the majority of these children, whose infection will resolve quickly (with or without antibiotics), while at the same time trying to avoid an increase in suppurative complications.

Using a handout to explain to parents that most children will get better without antibiotics and giving a deferred prescription may have met both objectives. The paper showed a large reduction in antibiotic use, and my practice partners and I have not seen any children with suppurative complications since our change in approach; nor have we noticed any increase in our follow up workload.

I certainly would not wish to imply that the potential complications of otitis media are trivial, but in our experience they are rare. In 1985 van Buchem et al reported the results of a trial in 4860 children with acute otitis media in primary care in Holland; no children were initially given antibiotics and only two developed mastoiditis, both responding to amoxicillin.2

Over 200 practices in the United Kingdom have requested a copy of our handout (available on the BMJ ‘s website3), and many have expressed an interest in trying to replicate our results. A before and after study of complications arising in children in these practices (compared with control practices that do not change policy) would have much greater power to detect any change in the incidence of these rare events than this study did. Clarke and his specialist colleagues might be well placed to carry out such a study.

Clarke says that we should have defined strict diagnostic criteria in our study. We chose not to do so because the children that we see with inflamed eardrums do not fall neatly into the two categories that he advocates. Moreover, the odds ratio analysis does not require an identical diagnostic threshold in each practice.

If further research identifies better ways to predict which children with acute otitis media will develop complications then it may be possible to target antibiotics more precisely. Until that time we commend our changed policy as a successful evidence based approach to the initial management of children with acute otitis media who are not unduly ill.