Therapeutic drug monitoring in anti-tuberculosis treatment: a systematic review and meta-analysis

BACKGROUND: Therapeutic drug monitoring (TDM) may improve tuberculosis (TB) treatment outcomes, but there is little evidence to guide TDM in clinical practice.

DESIGN: We performed a systematic review and meta-analysis to summarise existing literature on TDM in first-line drugs.

RESULTS: We identified 41 studies that reported 2 h post-dose drug concentrations (C_2h) for first-line drugs and 12 studies that reported clinical outcomes. We pooled data by study quality, design, region, dosing modality and patient characteristics. The pooled proportion of subjects with low isoniazid C_2h was 0.43 (95%CI 0.32–0.55), 0.67 (95%CI 0.60–0.74) had low rifampicin C_2h, 0.27 (95%CI 0.17–0.38) had low ethambutol C_2h, and 0.12 (95%CI 0.07–0.19) had low pyrazinamide C_2h. Patients with diabetes had a non-significant increase in the proportion of subjects with low C_2h levels across all four drugs. Only three of 12 studies that examined clinical outcomes demonstrated an association between low C_2h and unsuccessful treatment outcomes.

CONCLUSION: Across a wide variety of studies, a high proportion of patients undergoing first-line anti-tuberculosis treatment had 2 h drug concentrations below the accepted normal threshold. These findings point to a discrepancy between accepted 2 h TDM thresholds and TB drug dosing recommendations.