How many of the 5 million children in England who receive oral penicillins each year really need them, and what are the effective doses? It is time to replace the rule of thumb used for dosing with up-to-date evidence, according to a paper published on behalf of the improving Children’s Antibiotic Prescribing UK Research Network (iCAP).
The authors used a literature and formulary search to find out where the current, sometimes confusing, recommendations come from. It appears that, ‘a general recommendation to use age banding for all antibiotics in children, irrespective of the type of penicillin or disease indication, was published in the BMJ in 1963…Critically, these recommendations have remained unchallenged and unchanged to this day’.
Does a big child = half an adult? The BNF recommendations to halve doses between successive age bands for children have remained unchanged, for penicillin V, flucloxacillin and amoxicillin since they first appeared in it in the 1960s and 1970s. Adult penicillin doses have increased substantially but those for children have not. The mg/kg doses may also now be even lower than they were, because of an increase in average children’s weights. When the authors analysed the actual doses that would be received today, they found that at the older age/weight ranges, the doses are ‘strikingly low’. Many of the older children’s medicines do have a limited evidence base, but these authors were surprised at how little recent evidence supports the current recommendations.
They warn of the risks of low dosing – for antimicrobial resistance, severe complications and need for retreatment, and conclude that work is needed to determine the effective doses for children of all weights and ages, and also to, ‘establish more clearly which children really need antibiotics in the era of pneumococcal conjugate (PCV 13), Haemophilus influenzae B, and meningitis C vaccines’.
Ahmed U et al. Dosing of oral penicillins in children: is big child=half an adult, small child=half a big child, baby=half a small child still the best we can do? BMJ 2011;343:d7803