In reply: Acquired drug-resistance because of pharmacokinetic variability in a young child with tuberculosis.

Abstract

Reply: We appreciate the interest of Pasipanodya and colleagues in our case report and are aware of their important work in this area. We concur with some points they raise. We agree that direct observation of inadequate therapy, as in our case, will not result in optimal outcomes and is not in itself a cure-all. We agree with the importance of the pharmacokinetics of tuberculosis (TB) drugs, particularly in children. Major changes occur in the pharmacokinetics of drugs as children grow and develop, especially in the first few years of life.1 This is highly relevant to TB treatment, as children have lower drug exposures when given the same mg/kg dose as adults of many TB drugs,2 and updated World Health Organization (WHO) guidelines now recommend higher doses of the first-line TB drugs in children.3 We do not agree that our report provides conclusive support for routine individualized dosing of TB drugs in children. In our case report, many factors other than pharmacokinetic variability contributed to the poor outcome. The child had extensive pulmonary disease and should have received ethambutol, which may have provided additional protection against resistance acquisition. Drugs were under-dosed; using local weight-banded dosing based on new WHO recommendations, this child should have received isoniazid 90 mg (11.4 mg/kg), rifampin 90 mg (11.4 mg/kg), pyrazinamide 250 mg (31.6 mg/kg), compared with the much lower mg/kg doses the child actually received. The 2-month mycobacterial culture and drug-susceptibility result showing isoniazid resistance should have prompted a change in treatment. This child’s poor treatment response should have resulted in additional investigations or referral. These failings are at a health system level, and appropriate application of existing tools, without the need for individualized dosing, may have averted this poor outcome. We would be concerned with the feasibility of routine dose individualization given the weak health system described in our case, which hardly seems capable of supporting individualized drug dosing for all children with TB. Finally, the use of crushed or split adult tablets adds another challenge to accurate dosing and may reduce drug palatability. The child in our case was frequently vomiting his medications, further contributing to the inadequacy of his treatment. Child-friendly, palatable TB medications are an urgent priority. We are not in principle against individualized dosing and care, but given the likely cost and workload associated with routine therapeutic drug monitoring and individualized TB drug dosing, a pragmatic randomized controlled trial evaluating the impact, cost effectiveness, and feasibility of such a strategy in resource-limited high TB burden settings would be very informative. Early assessment of the WHO newly recommended doses of first-line drugs for young children found the resulting concentrations to fall well within the usually suggested range for efficacy.2 We would argue that in pediatrics the priority is to understand TB drug pharmacokinetics in children of all ages, and designing optimized, pragmatic, feasible dosing strategies to provide the best possible, safe drug exposure in children. Our case also argues for greatly increased efforts at health system strengthening to optimize all aspects of TB care for children. Anthony J. Garcia-Prats, MD Peter R. Donald, MD Desmond Tutu TB Centre Department of Paediatrics and Child Health Faculty of Medicine and Health Sciences Stellenbosch University Tygerberg, South Africa H. Simon Schaaf, MMed (Paed), MD (Paed) Desmond Tutu TB Centre Department of Paediatrics and Child Health Faculty of Medicine and Health Sciences Stellenbosch University Tygerberg Children’s Hospital Tygerberg, South Africa