Intended for healthcare professionals


Effect of isoniazid prophylaxis on mortality and incidence of tuberculosis in children with HIV: randomised controlled trial

BMJ 2007; 334 doi: (Published 18 January 2007) Cite this as: BMJ 2007;334:136

Pharmacokinetics of isoniazid formulation needs to be addressed

It is very encouraging to see that prescribing isoniazid can reduce
tuberculosis in HIV infected children, yet 5/132 children randomised to
isoniazid still developed tuberculosis (1). Despite the participants of
this study being infants and young children, Zar et al (1) used tablets.
Whilst Zar et al (1) report dosing of halved/quartered tablets to achieve
a variability of just 20 percent (10mg/kg +/- 2mg/kg), no mention of the
weight of split tablet fractions is made to confirm this. Halved tablet
weights typically vary by greater than 20 percent (2), the weight of
quartered tablets is likely to vary even more.

Modified adult formulations of isoniazid have a significantly lower
bioavailability than licensed liquid preparations in children (3). Zar
et al (1) do not report how doses were administered, but it is well
recognised that many children under 12 years are unable to swallow whole
solid dose forms (4). If the halved/quartered tablets were crushed,
patients are unlikely to have received the intended dose.

When the amount of test drug received by participants cannot be
ascertained on reading the published study, how is one to judge the
validity or reliability of the findings? Logistic or cost concerns often
mean tablets must be used for children in healthcare systems which lack
resources, but reports of tablet fraction weights and basic
pharmacokinetic data are still required. Whilst the results of this study
suggest manipulated tablets of isoniazid reduce tuberculosis incidence, it
is likely that further reduction could be made if formulation was
optimised, perhaps by standardising the administration method and
validating it with pharmacokinetic data.

1. Zar HJ, et al. Effect of isoniazid prophylaxis on mortality and
incidence of tuberculosis in children with HIV: randomised controlled
trial. BMJ 2007;334(7585):136-9.

2. Teng J, et al. Lack of medication dose uniformity in commonly split
tablets. J Am Pharm Assoc 2002;42(2):195-9

3. Notterman DA, et al. Effect of dose formulation on isoniazid
absorption in two young children. Pediatrics 1986;77(6):850-2.

4. Czyzewski DI, et al. Teaching and Maintaining Pill Swallowing in HIV-
Infected Children. AIDS Read 2000;10(2):88-94.

Competing interests:
Researcher in paediatric pharmacokinetics

Competing interests: No competing interests

21 January 2007
Joseph F Standing
Ph.D Student
London School of Pharmacy, University of London, WC1N 1AX