Predictors of long-term neurological sequelae of tuberculous meningitis: a multivariate analysis

Abstract

There is paucity of studies on predictors of long-term sequelae of tuberculous meningitis (TBM). We report the neurological sequelae of TBM at 1 year and their predictors. Patients with TBM who were followed up for 1 year were included. The diagnosis of TBM was based on clinical, cerebrospinal fluid (CSF) and computed tomography (CT) scan findings. Detailed neurological examinations at admission and at 1 year were carried out. All the patients received four-drug antitubercular therapy. The frequency of sequelae at 1 year were noted and the role of various demographic (age, sex, duration of illness, BCG vaccination), clinical (weakness, seizure, extra central nervous system tuberculosis, Glasgow Coma Scale (GCS) score, cranial nerve palsy, stage, corticosteroid, drug-induced hepatitis, shunt surgery), and laboratory findings (erythrocyte sedimentation rate (ESR), CSF cell and protein, CT scan evidences of hydrocephalus, basal exudates, infarctions and tuberculoma) at presentation were evaluated employing logistic regression analysis. Sixty-five patients with TBM were included in this study whose age ranged between 13 and 80 years (mean 33.2), 27 of whom were females. Complete neurological recovery at 1 year occurred in 21.5% patients only although about 50% were independent for activities of daily living. Neurological sequelae were observed in 78.5% patients, which included cognitive impairment in 55%, motor deficit in 40%, optic atrophy in 37% and other cranial nerve palsy in 23%. On logistic regression analysis, focal motor deficit at admission was the most important predictor of neurologic deficits at 1 year. GCS score predicted the cognitive and motor sequelae. Neurological sequelae at year occurred in 78.5% patients with TBM in the form of cognitive impairment, motor deficit and optic atrophy. Sequelae were common in patients who had focal motor deficit and altered sensorium at admission.