[Brain abscess. Evaluation of prognostic factors: does the use of antibiotic prescribing protocols improve outcome?].

Background: the aim of this study was to evaluate prognostic factors in brain abscess (AB) and influence of management with antibiotic prescribing protocols (APP).

Patients and methods: observational study of a cohort of non-paediatric patients with BA admitted at a 944-bed hospital (1976-2005). Data collection from clinical records has been done according to a standard protocol. We analysed epidemiological, clinical, radiological, microbiological and laboratory data associated with mortality. From 1976 to 1983 (Period I), antibiotic treatment was not done according to any internal APP; from 1983 (Period II), antibiotic management was done according to a APP designed by infectious diseases specialists and neurosurgeons. Predictors of mortality were identified by univariate analysis. The influence of the use of APP in outcome was assessed.

Results: 104 patients with BA were included (mean age 45 years; range 12-86); presumed primary pathogenic mechanism of BA was identified in 89%; microbiologic diagnosis was made in 76%. Overall mortality was 16.3%. Factors statistically associated with higher mortality were: age > 40 years, ultimately fatal underlying disease, acute severe clinical condition at the onset of BA, altered mental status and inadequate empirical treatment; 33 patients were treated in Period I and 71 in Period II; no statistically significant differences were found between epidemiological, clinical, radiological or microbiological characteristics of the groups except for mean age (> 40 years in 36% and 62% respectively in Period I and II). Rates of resolution of BA were 60 vs. 77.4% (p < 0.05); relapses 21 vs. 7% (p < 0.05) and mortality 18 vs. 15.4% (p > 0.05), in Period I and II respectively.

Conclusions: main prognostic factors associated with mortality in patients with BA are age, rapidly fatal underlying disease, acute severe clinical condition at the onset of BA, altered mental status and inadequate empirical treatment. Empiric treatment according to APP was associated with greater resolution and lower relapse rates.