Validation of the Bacterial Meningitis Score in adults consulting at an emergency department: a retrospective multicentric study.

Abstract

The Bacterial Meningitis Score (BMS) is recommended by pediatric academic societies to rule out the diagnosis of bacterial meningitis. The aim of this study was to evaluate the performance of the BMS to identify adults at no risk for bacterial meningitis. We conducted a multicentric retrospective study including adults who consulted the emergency department (ED) for meningitis [cerebrospinal fluid (CSF) white blood cells ≥5/mm with a ratio of white blood cells/red blood cells <1:900) during a 4-year period. The BMS variables were: CSF positive Gram stain, CSF absolute neutrophil count ≥1000 cells/μL, CSF protein ≥80 mg/dL, peripheral blood absolute neutrophil count ≥10 000 cells/μL, and seizures. Bacterial meningitis was defined for patients who had a lumbar puncture with CSF pleocytosis and positive bacterial analysis of CSF. The primary endpoint was the sensitivity of the BMS to rule out bacterial meningitis in adults. The secondary outcome was to assess the rate of patients for whom antibiotics could have been avoided using the BMS and the diagnostic performance of procalcitonin in patients with a BMS ≥1. Of 930 patients with meningitis, 626 were included in the analysis, and 27 (4.3%) were diagnosed with bacterial meningitis. A total of 384/626 (61.3%) patients had a BMS = 0, and none presented bacterial meningitis. BMS sensitivity was 100% [95% confidence interval (CI), 87.2-100%], and its negative predictive values were 100% (95% CI, 98.8-100%). Among patients with a diagnosis of meningitis in ED, a BMS of 0 may safely rule out bacterial meningitis.