Background Infants living with HIV are at high risk of tuberculosis and death. Optimal antituberculosis therapy is essential for favourable clinical outcomes particularly in severely ill children. Using WHO-recommended weight-band dosing, younger children weighing <8kg are at risk of suboptimal exposures. We aimed to evaluate plasma concentration of first line antituberculosis drugs in infants with HIV.

Methods EMPIRICAL trial (#NCT03915366; EDCTP2-funded (RIA2017MC-2013)) is a randomized controlled trial evaluating empirical antituberculosis and cytomegalovirus treatment in infants with HIV hospitalized for severe pneumonia in 5 African countries. Eligible infants aged <1 year, weighing ≥3kg, on antituberculosis treatment had a blood sample taken 2-hours post-dose at days 30, 90 and 180 in a pharmacokinetic sub-study. Antituberculosis drugs were dosed according to WHO weight-bands using fixed-dose-combination dispersible tablets of rifampicin(15mg/kg)/isoniazid(10mg/kg)/pyrazinamide(35mg/kg) 75/50/150mg with ethambutol(20mg/kg) 100mg. Antiretroviral-naïve infants initiated treatment in accordance with national guidelines. We compared C2hr plasma concentrations for rifampicin, isoniazid, pyrazinamide, and ethambutol with published Cmax references.

Results Forty-nine infants of whom 21 were female, median (range) age 6.1(2.5–13.5) months and weighing 5.3(3.4–8.7) kg were included in the analysis of study day 30. The geometric mean (CV%) C2hr for rifampicin, isoniazid, pyrazinamide and ethambutol were 3.66(161) mg/L, 2.80(102) mg/L, 22.27(97) mg/L, and 0.56(101) mg/L, respectively. The C2hr values were substantially below adult reference Cmax for rifampicin [ref in adults (10 mg/kg dose): 8–24 mg/L] and ethambutol [ref: 2–6 mg/L], slightly lower for isoniazid [ref: 3–6 mg/L], and within range for pyrazinamide [ref: 20–60 mg/L].

Conclusion Plasma levels of first-line TB drugs in infants with HIV and severe pneumonia were low compared to adults. This is consistent with other studies showing that infants and younger children do not achieve adult references for first-line TB drugs at current recommended doses. Our data support considerations for optimising dosing of first-line TB drugs for infants.