Article Text
Statistics from Altmetric.com
In a recent paper published in this journal,1 the laboratory results from 1516 infants with laboratory-confirmed early neonatal infections were reported for the 27-month period up to March 2008 (approximately 674 cases per year). Six per cent of laboratory isolates from infants younger than 48 h were reported as resistant to empirical treatment with penicillin and gentamicin. This means that 40 infants (6% of 674 infants per year) would have received antibiotic treatment to which the causes of infection were resistant by laboratory tests. We do not have empirical data that show exactly how many of the 40 infants experienced adverse effects from receiving relatively ineffective antibiotics, but it is unlikely to be zero. Treatment of these 40 with more effective antibiotics requires that a much larger number are treated because we cannot identify those who will have antibiotic-resistant infections. The number treated could be all infants with suspected early sepsis or infants at particular risk of an adverse outcome, such as preterm infants.
If we consider that most infants treated empirically for suspected early sepsis receive antibiotics for less than 48 h, then the additional antibiotic costs of using—for example, cefotaxime and amikacin (for which the data shown suggest that there would be 100% susceptibility) would be lower than £5 per infant. If we estimate that the possibility of sepsis is investigated in 10% of UK live births (10% of 708 711 in 2008) and that all of these infants were given the more effective antibiotic regime, then the cost would be approximately £350 000. This figure can be compared with the Health and Safety Executive2 value of £1 million to prevent one death among 100 000. We could further reduce costs by only treating the higher-risk infants, such as the more preterm, with the more effective antibiotic regimes. The cost of selecting antibiotic-resistant microbes is probably not very significant in that the duration of use will be short for most of the patients, and there is little microflora in a newborn infant from which to select resistant strains. The authors should specify the criteria that justify the statement that current (antibiotic) guidelines for empirical therapy of neonatal sepsis are appropriate.
Footnotes
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.