Initial Antituberculous Regimen with Better Drug Penetration into Cerebrospinal Fluid Reduces Mortality in HIV Infected Patients with Tuberculous Meningitis: Data from an HIV Observational Cohort Study

Tuberculous meningitis (TM) is the deadliest form of tuberculosis. Nearly two-thirds of HIV infected patients with TM die, and most deaths occur within one month. Current treatment of TM involves the use of drugs with poor penetration into the cerebro-spinal fluid (CSF). In this study, we present the mortality before and after implementing a new antituberculous regimen (ATR) with a higher drug penetration in CSF than the standard ATR during the initial treatment of TM in an HIV cohort study. The new ATR included levofloxacin, ethionamide, pyrazinamide, and a double dose of rifampicin and isoniazid and was given for a median of 7 days (interquartile range 6–9). The new ATR was associated with an absolute 21.5% (95% confidence interval (CI), 7.3–35.7) reduction in mortality at 12 months. In multivariable analysis, independent factors associated with mortality were the use of the standard ATR versus the new ATR (hazard ratio 2.05; 95% CI, 1.2–3.5), not being on antiretroviral therapy, low CD4 lymphocyte counts, and low serum albumin levels. Our findings suggest that an intensified initial ATR, which likely results in higher concentrations of active drugs in CSF, has a beneficial effect on the survival of HIV-related TM. 1. Introduction In 2011, there were 8.7 million incident cases of tuberculosis (13% of them in HIV infected patients) and 1.4 million deaths from tuberculosis (30% of them in HIV infected patients) [1]. With 25% mortality in non-HIV infected patients and 67% in HIV infected patients, tuberculous meningitis has the highest mortality among all forms of tuberculosis [2]. Moreover, tuberculous meningitis is more common in HIV infected patients and can comprise up to 19% of all cases of HIV-related tuberculosis [3, 4]. Currently, treatment of tuberculous meningitis involves the same drugs and doses as other forms of tuberculosis [5–7]. While isoniazid and pyrazinamide have good cerebrospinal fluid (CSF) penetration, rifampicin concentration in CSF may not reach the minimal inhibitory concentration for tuberculosis, and ethambutol and streptomycin have poor CSF penetration [2, 8]. Among second line drugs, levofloxacin, ethionamide and cycloserine have good penetration in CSF [8–10]. In a phase 2 randomized controlled trial investigating the safety of moxifloxacin and a higher intravenous dose of rifampicin during the first two weeks of treatment of tuberculous meningitis, the use of a higher dose of rifampicin was associated with a survival benefit [11]. These data suggest that increasing the CSF penetration of the initial treatment of