Print E-mail
Volume 43, Number 2, April 2010

Prognostic Factors of Tuberculous Meningitis in Adults: A 6-Year Retrospective Study at a Tertiary Hospital in Northern Taiwan

Po-Chang Hsu, Chien-Chang Yang, Jung-Jr Ye, Po-Yen Huang, Ping-Cherng Chiang, Ming-Hsun Lee*

Received: February 17, 2009    Revised: April 25, 2009    Accepted: June 11, 2009   


Corresponding author:

Division of Infectious Diseases, Department of Internal Medicine, Chang Gung Memorial Hospital, 5 Fu-Shin Street, Gueishan 333, Taoyuan, Taiwan. 



Background and purpose: 

To investigate the clinical features, laboratory test results, imaging data, and prognostic predictors of tuberculous meningitis (TBM) in adults.



We retrospectively reviewed 108 adult patients with a diagnosis of TBM over a 6-year period. Patients were divided into “definite” and “probable” groups, depending on the diagnosis made by (1) positive culture, or polymerase chain reaction, of Mycobacterium tuberculosis (TB) from the cerebrospinal fluid (CSF); or (2) the isolation of TB elsewhere, or chest radiography consistent with active pulmonary TB, or imaging studies of the brain consistent with TBM, or clinical improvement on treatment. These two groups were compared for their clinical features, images, laboratory test results, and 9-month mortality rates to identify prognostic predictors.



Compared with the “probable” group (n= 62), the “definite” group (n= 46) had a higher mortality rate (50.0% vs. 30.6%, p= 0.041) and more consciousness disturbance (78.3% vs. 51.6%, p= 0.005), hydrocephalus (63.4% vs. 40.7%, p= 0.029) and isolation of TB from extra-CSF specimens (41.3% vs. 22.6%, p= 0.037). Old age (p= 0.002), consciousness change (p= 0.032), and hydrocephalus (p= 0.047) were poor prognostic indicators in the “definite” group as assessed by univariate analysis. Severity of TBM at admission and delayed anti-TB therapy resulted in a poor prognosis for all patients. Multiple logistic regression analysis showed that old age and hydrocephalus were independent factors for mortality. Adjunctive steroid therapy over 2 weeks improved survival in both the “definite” (p= 0.002) and “probable” (p= 0.035) groups, but more than 4 weeks of use had no significant effect on mortality. Steroid treatment, therefore, may improve the outcome of patients with TBM.



Old age, advanced stage of TBM at admission, hydrocephalus, and positive TB culture or polymerase chain reaction of CSF are factors associated with a poor prognosis for TBM. Early diagnosis and treatment, including short term steroid use, are mandatory for clinical care of adult patients with TBM.


Key words:

hydrocephalus, prognostic factors, steroid, tuberculous meningitis