High-Dose Oral Rifampin in Adults With Tuberculous Meningitis

Key Takeaways

• Adults assigned to the higher cumulative rifampin dose did not show a survival advantage at 6 months.
• Drug-induced liver injury was reported more often with intensified rifampin, and no deaths from liver injury were reported.
• The authors concluded that benefit was not evident and that possible harm could not be excluded.

An adult tuberculous meningitis trial tested intensified oral rifampin against standard dosing during initial therapy for tuberculous meningitis. Both groups received the same backbone regimen, with rifampin dose intensification as the only difference during the first 8 weeks. Higher cumulative rifampin exposure did not improve 6-month mortality over the standard-dose regimen. Drug-induced liver injury was reported more often with intensified rifampin, and the investigators concluded that benefit was not shown while possible harm could not be excluded.

This double-blind, randomized, placebo-controlled clinical trial enrolled adults with tuberculous meningitis in Indonesia, South Africa, and Uganda, including participants with and without HIV coinfection. The intention-to-treat population included 499 participants, with 249 assigned to the high-dose group and 250 to standard-dose therapy; 304 were living with HIV, and 428 had definite or probable disease. All participants received standard daily isoniazid, rifampin 10 mg/kg, ethambutol, and pyrazinamide as the backbone regimen. The intervention group received additional rifampin to a cumulative dose of 35 mg/kg for 8 weeks, whereas the comparison group received matched placebo for 8 weeks. Participants then continued standard therapy for the remainder of the 9- to 12-month course, and 6-month mortality was the prespecified primary outcome.

By 6 months, 109 participants in the high-dose rifampin group and 100 in the standard-dose group had died. The corresponding Kaplan-Meier mortality estimates were 44.6% and 40.7%, with a hazard ratio for death of 1.17. The 95% confidence interval ranged from 0.89 to 1.54, and the reported P value was 0.25. Among participants who died within 6 months, median time to death was 13 days in the intensified group and 24 days in controls. The interquartile ranges were 4 to 39 days and 6 to 56 days, and there was no mortality benefit with dose intensification.

Drug-induced liver injury was reported in 8.0% of the high-dose group and 4.4% of the standard-dose group. This was the notable adverse event difference reported between groups, and fatal liver toxicity was not seen. No deaths from drug-induced liver injury were reported during follow-up. The investigators concluded that no evidence of beneficial effect was observed and that a harmful effect could not be ruled out.