The need for systematic therapeutic drug monitoring of isoniazid in tuberculosis: A real-world study.

Objectives Isoniazid is a primary first-line antituberculosis agent, exhibiting concentration-dependent bactericidal effect while also posing a risk of hepatotoxicity. Therapeutic drug monitoring could optimize exposure but remains underused. This study aimed to assess isoniazid pharmacokinetic variability in real-world settings and the proportion of patients achieving therapeutic concentrations. Methods We conducted a retrospective, single-centre study of adult tuberculosis patients receiving standard first-line therapy (isoniazid, rifampicin, pyrazinamide, ethambutol). Patients underwent fasted-state isoniazid monitoring via a 4-point limited-sampling protocol. We calculated AUC 0–24h and acetylation status (rapid: half-life <2.2 h; slow: ≥2.2 h). Expected exposure was defined by a C max of 3–6 mg/L, while target exposure was defined by an AUC 0–24h above 10.5 h·mg/L for bactericidal efficacy and below 21.8 h·mg/L to minimize the risk of hepatotoxicity. Results Among 109 patients (66.1% male, median age 40.7 y, 69.8% pulmonary tuberculosis), AUC 0–24h showed high interindividual variability (range 3.4–62.8 h·mg/L, coefficient of variation 59.4%), independent of weight-adjusted dosing (Pearson’s r = −0.016, P = 0.875). Slow acetylators predominated (59%). With a median dose of 4.1 mg/kg, 24 patients (22%) had subtherapeutic AUC 0–24h (<10.5 h·mg/L), while 52 (47.7%) exceeded the hepatotoxicity risk threshold (>21.8 h·mg/L). Nearly one-third had C max within the expected range but AUC 0–24h above the safety limit. Conclusions Real-world data reveal substantial isoniazid exposure variability, with frequent deviations from target AUC 0–24h . These findings advocate for systematic therapeutic drug monitoring and individualized dosing, prioritizing AUC 0–24h as a key monitoring parameter.