OBJECTIVE: Tuberculous meningitis (TBM) and its complications continue to have devastating neurological consequences for patients. Budgetary constraints, especially in developing countries, have made it necessary to select patients for shunting who are likely to experience good recoveries. To date, the value of cerebrospinal fluid shunting for human immunodeficiency virus (HIV)-positive patients with TBM has not been clearly established. METHODS: Thirty patients with TBM and hydrocephalus were prospectively evaluated. Coincidentally, one-half of the patients were HIV-positive. All patients underwent uniform treatment, including ventriculoperitoneal shunt placement and antituberculosis treatment. CD4 counts were measured for all patients. Outcomes were assessed at 1 month. RESULTS: No complications related to shunt insertion were noted. The HIV-positive group fared poorly (death, 66.7%; poor outcome, 64.7%), compared with the HIV-negative group (death, 26.7%; poor outcome, 30.8%). Despite cerebrospinal fluid shunting, no patient in the HIV-positive group experienced a good recovery (Glasgow Outcome Scale score of 5). This is in contrast to the six patients (40%) in the HIV-negative group who, with the same treatment, experienced good recoveries (Glasgow Outcome Scale scores of 5) at discharge (P<0.14). No patient (either HIV-positive or HIV-negative) who presented in TBM Grade 4 survived, whereas no HIV-positive patient who presented in TBM Grade 3 survived. A significant relationship was noted between CD4 counts and patient outcomes (P<0.031). CONCLUSION: In the absence of obvious clinical benefit, HIV-positive patients with TBM should undergo a trial of ventricular or lumbar cerebrospinal fluid drainage, and only those who exhibit significant neurological improvement should proceed to shunt surgery.