King's College London

Research portal

Clinical and Microbiologic Features Guiding Treatment Recommendations for Brain Abscesses in Children

Research output: Contribution to journalArticle

Susanna Felsenstein, Bhanu Williams, Delane Shingadia, Lucy Coxon, Andrew Riordan, Andreas K. Demetriades, Chris Chandler, Sanj Bassi, Eirini Koutoumanou, Simon Stapleton, Mike Sharland, Penelope A. Bryant

Original languageEnglish
Pages (from-to)129-135
Number of pages7
JournalPediatric Infectious Disease Journal
Volume32
Issue number2
DOIs
StatePublished - Feb 2013

King's Authors

Abstract

Background: There are no guidelines for the management of brain abscesses in children, and there is a paucity of recent data describing clinical and microbiologic features. We aimed to identify factors affecting outcome to inform antibiotic recommendations.

Methods: From 1999 to 2009, 118 children presented with brain abscesses to 4 neurosurgical centers in the United Kingdom. Clinical, microbiologic and treatment data were collected.

Results: The commonest preceding infection was sinusitis, with 59% of all children receiving antibiotics before diagnosis. Nonspecific symptoms were common, with only 13% having the triad of fever, headache and focal neurological deficit. Time between symptom onset and diagnosis varied widely (median, 10 days; range, 0-44). Magnetic resonance imaging was more frequently diagnostic than computed tomography. The most frequent organisms were Streptococcus milleri (38%), except after penetrating head injury or neurosurgery, for which Staphylococcus aureus was most common. The commonest empiric antibiotics were ceftriaxone/cefotaxime and metronidazole, which offered effective antimicrobial therapy in up to 83% of cases. Metronidazole added benefit in a maximum of 7% of cases, with ceftriaxone/cefotaxime alone sufficient in at least 76% and in all cases with cyanotic congenital heart disease or meningitis. A carbapenem would have been effective in 90%. The case fatality rate was 6% (33% in the immunocompromised). Long-term neurological sequelae affected 35%. Age younger than 5 years and a Glasgow Coma Scale score

Conclusions: We recommend ceftriaxone/cefotaxime and metronidazole as empiric treatment, although metronidazole may be unnecessary in many cases, with antistaphylococcal cover in cases of head trauma. Meropenem potentially would be a better choice in the immunocompromised. A prospective study of intravenous and oral treatment guided by clinical improvement is required beause 1-2 weeks of intravenous antibiotics during a total of 6 weeks may be sufficient in children.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454