Brain abscess: clinical experience and analysis of prognostic factors

Abstract

Background Over the past 2 decades, the diagnosis and treatment of brain abscess have been facilitated by a number of technological advancements, which have resulted in a significant improvement of outcome. The aim of this manuscript is to review our experience, to determine the factors related to the outcome, and to improve the therapeutic strategy for this disease. Methods From 1986 to 2002, 178 consecutive patients with bacterial brain abscess were treated at the National Taiwan University Hospital, Taipei, Taiwan. We reviewed their clinical presentation, bacteriology, treatment, and outcome retrospectively. Groups were compared by χ 2 test, Fisher exact test, or t test as appropriate. Multivariate logistic regression with backward selection was used to select the set of covariates that were independently associated with outcome. Results One hundred eleven patients (62%) had favorable outcome, 14 patients (8%) had severe disability, 9 patients (5%) became vegetative, and 44 (25%) died during hospitalization. Patients with better Glasgow Coma Scale (GCS) on admission, no underlying disease, positive culture, or surgical treatment were more likely to have a good outcome. Patients with nasopharyngeal carcinoma, acquired immunodeficiency syndrome, hematologic disease, deep-seated abscess, or medical treatment alone were more likely to have a poor outcome. Multivariate analysis revealed that only GCS, immunodeficiency, and presence of underlying disease related with outcome. Conclusions The poor prognostic factors of brain abscess are poor GCS, immunodeficiency, and presence of underlying disease. Aggressive treatment with surgery when indicated and careful management of specimen for culture might improve outcome.