Brain abscesses - the Groote Schuur experience, 1993 - 2003

Abstract

Objective: To review management and outcome of patients with brain abscess treated at Groote Schuur Hospital (GSH) between 1993 and 2003.

Patients and methods: Case notes, radiological results and laboratory records were reviewed retrospectively for 121 patients at GSH who underwent a neurosurgical procedure for treatment of a brain abscess between 1993 and 2003. Patients not treated surgically were excluded. Follow-up with serial computed tomography (CT) scans, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels and the temperature chart were used to determine the duration of parenteral antibiotic treatment and the need for repeat surgical evacuation of pus by either aspiration or excision of the capsule.

Results: The mean patient age was 33 years, with a male-to-female ratio of 5:1. Headache, depressed level of consciousness and pyrexia were the commonest presenting symptoms. Other symptoms included seizures and hemiparesis. The frontal lobe was the commonest site (44%); the majority of abscesses occurred as a result of infection following trauma. Mastoiditis (21%) and sinusitis (8%) were the second and third most common causes. Organisms were identified in 81% of cases; polymicrobial infections occurred in half of these. Thirty-three different organisms were identified, the majority of which were Gram-positive bacteria. Anaerobes were present in 23%, while methicillin-resistant Staphylococcus aureus was identified in 1 patient. Nocardia was seen in 3 patients, 2 of whom were diabetic. The average duration of parenteral antibiotic therapy was 15 and 19 days for excision and aspiration respectively. Accurate ESR records were available in 75% of cases, with a positive predictive value in 81%, while CRP (in use since 1999) was helpful in 92% of cases.

Outcome: Sixteen patients (13%) died, 12 of whom had been admitted with a Glasgow Coma Score (GCS) of less than 4T/15. Thirteen patients developed epilepsy. On discharge, patients continued to take oral antibiotics for another 4 weeks, at the end of which they were reviewed at the neurosurgery outpatient department.

Conclusion: Compared with previous studies from this hospital, mortality and morbidity have been diminishing progressively. GCS at the time of admission remains the most important prognostic factor.

South African Journal of Surgery Vol. 43(3) 2005: 79-82