The purpose of this study was to verify the statistical validity of the score proposed by Thomé et al. for the differential diagnosis between bacterial and viral meningitis and to study the utility of two new parameters (CRP and the patient's age). A retrospective review of 136 consecutive cases of meningitis was made. The cases were classified into three groups according to the culture results and the use or not of intravenous antibiotics. There were 20 cases of bacterial meningitis, 60 non-bacterial and 56 cases of meningitis of uncertain etiology. Considering only the patients in the first two groups, the capacity for differential diagnosis between bacterial and non-bacterial meningitis of the 8 parameters in the original score, the CRP and the patient's age was analyzed with a numeric value of 0, 1, or 2 assigned to each parameter. Finally, the results of applying the score that includes the two new parameters with the original score are compared. We found that all parameters showed statistical significance for the differential diagnosis between bacterial and viral meningitis. The resulting score can be used in order to decide the need for intravenous antibiotic therapy, with only a few cases being uncertain. The score with the two new parameters correctly classified 78 of the 80 cases, while leaving only two case uncertain as compared to the four that remained unclassified with the original score. We conclude that the score is a useful instrument in the differential diagnosis between bacterial and viral meningitis. Furthermore, the introduction of CRP and the patient's age improves the diagnostic value of the test.