In March 1986, we began a 6-month short course trial of therapy for tuberculous meningitis, in which 28 patients were analyzed. The diagnosis was based on the following cerebrospinal fluid test results: in 53.5% of the cases, Mycobacterium tuberculosis was identified by direct smear; in 57%, culture in Löwenstein-Jensen medium was positive; in 83.3%, the detection of anti-bacille Calmette-Guérin (BCG) antibodies by enzyme-linked immunosorbent assay was positive; and in 74%, the dosification of adenosine deaminase activity was positive. In addition, in 21.4% of the cases, the diagnosis was established by means of autopsy findings. Moreover, the diagnosis was supported by bacteriological analyses from another tissue or body fluids. Despite the administration of an antituberculous therapy, 32.4% of the patients died: all of the decreased had reached the last stage of the disease by the beginning of treatment. Sixteen percent of the patients who survived after more than 18 months of follow-up after therapy had ended suffered neurological sequelae. With the 6-month therapeutic regimen, the morbidity/mortality is similar to that found in the longer-course therapies. The latter regimen is therefore thought to be a good and acceptable therapeutic option for the treatment of tuberculous meningitis.