Management of patients with Streptococcus milleri brain abscesses

We evaluated the efficacy of cefotaxime in the management of brain abscesses caused by Streptococcus milleri. Twenty two patients with a S. milleri brain abscess were treated with metronidazole and cefotaxime, in accordance with recent recommendations by the British Society Of Antimicrobial Chemotherapy (BSAC). Seven patients who had Glasgow Coma Scales < or =11 also received rifampicin and high dose cefotaxime. The clinical response of the patients was determined.

A retrospective study at the Queen Elizabeth Hospital, Birmingham covering the period April 1996-March 2004 was carried out. Neurosurgical and anti-microbial therapeutic approaches were reviewed. Any evidence of improvement of clinical features and radiological disappearance of brain abscesses were determined.

Outcome was assessed using the Glasgow Outcome Score (GOS) at 3 and 6 months from the time of surgical intervention. Eighteen patients (82%) had a good outcome by 6 months, with an outcome score of 4-5. Thirteen patients resumed normal life despite minor deficits (GOS 5), while a further five patients had moderate disability though remained independent (GOS 4). One patient had a GOS of 3 and there were three deaths (14). The minimum time to radiological resolution of the abscess was within 1 month in six cases (27) These all represented solitary lesions that required a single drainage procedure in conjunction with 4 weeks of intravenous cefotaxime and metronidazole. Ten cases (45%) had resolution within 4 months and a further three cases took at least 6 months from the time of surgery to show radiological clearance.

This cohort of patients responded favourably to the guidelines recommended by the BSAC. This was confirmed by the Glasgow Outcome Score (GOS 4-5) at 6 months review. Cefotaxime at a higher dose with rifampicin was prescribed for patients presenting with a decreased conscious level (GCS 8-11), subsequent failure of anticipated clinical improvement or clinical deterioration. There was no clinically significant difference in GOS between the two treatment groups. An algorithm for management of brain abscess is presented, based on our clinical experience and review of the literature.