Pharmacokinetics of First-Line Antituberculosis Drugs in HIV-Infected Children with Tuberculosis Treated with Intermittent Regimens in India

The objective of this report was to study the pharmacokinetics of rifampin (RMP), isoniazid (INH), and pyrazinamide (PZA) in HIV-infected children with tuberculosis (TB) treated with a thrice-weekly anti-TB regimen in the government program in India. Seventy-seven HIV-infected children with TB aged 1 to 15 years from six hospitals in India were recruited. During the intensive phase of TB treatment with directly observed administration of the drugs, a complete pharmacokinetic study was performed. Drug concentrations were measured by high-performance liquid chromatography. A multivariable regression analysis was done to explore the factors impacting drug levels and treatment outcomes. The proportions of children with subnormal peak concentrations (C-max) of RMP, INH, and PZA were 97%, 28%, and 33%, respectively. Children less than 5 years old had a lower median C-max and lower exposure (area under the time-concentration curve from 0 to 8 h [AUC(0-8)]) of INH (C-max, 2.5 versus 5.1 mu g/ml, respectively [P = 0.016]; AUC(0-8), 11.1 versus 22.0 mu g/ml . h, respectively [P = 0.047[) and PZA (C-max, 34.1 versus 42.3 mu g/ml, respectively [P = 0.055]; AUC(0-8), 177.9 versus 221.7 mu g/ml . h, respectively [P = 0.05]) than those more than 5 years old. In children with unfavorable versus favorable outcomes, the median C-max of RMP (1.0 versus 2.8 mu g/ml, respectively; P = 0.002) and PZA (31.9 versus 44.4 mu g/ml, respectively; P = 0.045) were significantly lower. Among all factors studied, the PZA C-max influenced TB treatment outcome (P = 0.011; adjusted odds ratio, 1.094; 95% confidence interval, 1.021 to 1.173). A high proportion of children with HIV and TB had a subnormal RMP C-max. The PZA C-max significantly influenced treatment outcome. These findings have important clinical implications and emphasize that drug doses in HIV-infected children with TB have to be optimized.