Population pharmacokinetic modeling of pyrazinamide in children and adults with tuberculosis

STUDY OBJECTIVE: To determine population pharmacokinetic parameters of pyrazinamide after multiple oral doses given to children and adults with tuberculosis. DESIGN: Prospective, multiple-dose population pharmacokinetic study. SETTING: Five hospitals in the United States. PATIENTS: Sixty-seven adults and 23 children with active tuberculosis. INTERVENTION: The 90 patients received multiple oral doses of pyrazinamide as part of their treatment, based on the best clinical judgment of the attending physicians and in keeping with standard clinical practices at each institution. The patients also received other antituberculosis drugs empirically or based on in vitro susceptibility data. MEASUREMENTS AND MAIN RESULTS: Serum samples were collected over 12 hours after dosing and were assayed with a validated gas chromatography assay with mass selective detection. Concentration-time data were analyzed by using population methods. Pyrazinamide concentrations increased linearly with increasing oral doses (185-3550 mg). Median maximum serum concentration values were 41.0 microg/ml with daily dosing and 66.1 microg/ml with larger, twice-weekly dosing. Incomplete (18%) or delayed (30%) absorption was more common in children than in adults (1% for each). Pharmacokinetic parameters of pyrazinamide were independent of human immunodeficiency virus status and patient demographics, except for body weight. Population elimination half-life values in pediatric and adult patients were 3.5 and 6.0 hours, respectively. Median volume of distribution (L/kg) was 32% larger in children, and median clearance (L/hr/kg) was 106% larger in children, with a resultant median half-life 43% shorter in children. CONCLUSION: Pyrazinamide concentrations and most pharmacokinetic parameters were comparable to those previously published. Apparent half-life was somewhat shorter than that in previous reports. Compared with adults, absorption of pyrazinamide in children appeared more likely to be incomplete or delayed.