Intensified Treatment of Tuberculous Meningitis in Adults: A Systematic Review and Meta-analysis

Abstract

Background: Tuberculous meningitis (TBM) remains the deadliest form of tuberculosis. Inadequate penetration of rifampicin and ethambutol into the brain and cerebrospinal fluid (CSF) may contribute to mortality. Over the last decade, research has focused on "intensified" treatment (higher-dose first-line drugs or addition of second-line drugs with good CSF penetration). This systematic review and meta-analysis evaluates the impact of intensified TBM treatment on mortality, disability, and safety.

Methods: A systematic literature search was conducted of clinical trials examining intensified TBM treatments compared with a rifampicin-based standard-of-care regimen in adults. Odds ratios (ORs) were calculated using a random-effects model with mortality as the primary outcome, with OR <1 indicating lower mortality. Disability and safety were examined as secondary outcomes. Subgroup analyses included (1) higher-dose rifampicin, (2) addition of fluoroquinolones, and (3) addition of linezolid.

Results: Ten trials meeting eligibility criteria, involving 1369 participants, were included. Higher-dose rifampicin (n = 1050; OR, 0.86; 95% CI, 0.54-1.35; P = .50), adjunctive fluoroquinolones (n = 1115; OR, 0.85; 95% CI, 0.56-1.27; P = .42), and linezolid (n = 79; OR, 0.73; 95% CI, 0.22-2.43; P = .61) did not significantly reduce TBM mortality. Due to heterogeneity in disability and safety endpoints, secondary outcomes could not be meta-analyzed.

Conclusions: Current clinical trial evidence does not support the use of intensified TBM treatment in adults. However, these analyses are limited by diverse TBM case definitions, absence of MRC grading at enrollment, variable rifampicin dosing, limited data on linezolid and higher-dose isoniazid, and heterogeneous disability and safety outcomes. Use of uniform case definitions and consistent endpoints is essential to standardize data.

Keywords: TBM; intensified; treatment; tuberculous meningitis.

Figure 1.

Flowchart of the study selection process. This flowchart of study selection illustrates the total number of records identified through the literature search, followed by records remaining after duplicates were removed, followed by records screened by title/abstract. The records excluded exit the flowchart with reasons for exclusion given. Full texts remaining were assessed for eligibility, and the number of full-text articles excluded is given along with reasons for exclusion. The end of the flowchart is the total number of studies included in the evidence synthesis (n = 10). Abbreviations: IV, intravenous; PK/PD, pharmacokinetic/phamacodynamic.

Figure 2.

Funnel plot of studies included in the meta-analysis. This funnel plot of the studies included in the evidence synthesis shows 2 studies outside of the 95% confidence interval with a balanced number of studies within the figure.

Figure 3.

Forest plot of comparison between intervention vs control by subgroups. A, Use of high-dose rifampicin. B, Addition or substitution of fluoroquinolones. C, Addition of linezolid. D, High-dose isoniazid and ethambutol. This forest plot summarizes mortality in the intervention vs control arms, overall showing no statistically significant difference in mortality in trials assessing higher-dose rifampicin, fluoroquinolones. Abbreviations: CIP, ciprofloxacin; LVX, levofloxacin; LZD, linezolid; MXF, moxifloxacin; RIF, rifampicin. , or linezolid.