Comparison of clinical and CSF profiles in children with tuberculous and pyogenic meningitis; role of CSF protein: glucose ratio as diagnostic marker of tuberculous meningitis.

OBJECTIVES To compare the clinical and laboratory features of tuberculous meningitis with bacterial meningitis and to determine if protein glucose ratio in the cerebrospinal fluid can be predictive of the condition. METHODS It was a retrospective descriptive study at the Paediatric Ward of Aga Khan University Hospital, Karachi, for which records of 46 patients with tuberculous meningitis and 33 with bacterial meningitis were reviewed. All records related to the study period that was from January 2005 to October 2010. The records were retrieved in December 2010. Tuberculous meningitis was defined as illness > or = 14 days, basal enhancement or tuberculoma on computerised tomography scan and response to anti-tuberculous therapy. Pyogenic meningitis was defined by the presence of a pathogenic isolate inthe cerebrospinal fluid (bacterial culture) or by positive latex particle agglutination or a pathogen on cerebrospinal fluid gram stain and clinical improvement with antibiotics. Logistic regression was used to model the probability of having tuberculous meningitis. To define the optimal protein-glucose ratio, cutoff point for the diagnosis of tuberculous meningitis, a receiver operating characteristic curve was applied. All analysis were done using SPSS 16. RESULTS Features predictive of a diagnosis of tuberculous meningitis were protein:glucose ratio of > or = 2 (OR 21 95% CI 4.7-93); cerebrospinal fluid total leukocyte count < 800 (OR 58, 95% CI 5-649); and the presence of hydrocephalus (OR 19, 95% CI 3.3-109). CONCLUSION A set of simple clinical, laboratory and radiological criteria can help in predicting tuberculous meningitis. The value of cerebrospinal fluid protein:glucose ratio needs to be validated in larger studies with bacteriologically-confirmed cases of tuberculous meningitis.