Background: Zidovudine and didanosine are antiretroviral drugs used for human immunodeficiency virus (HIV)-infected children with dose recommendations based on body surface area calculations. Although weight and height can both be measured, it may be impractical to expect providers in resource-limited settings to estimate accurately body surface area.
Methods: We developed an antiretroviral dosing chart based on authoritative sources for brand name drugs in weight bands (ie, 5-6.9, 7-9.9, 10-11.9, 12-14.9, 15-16.9, 17-19.9, 20-24.9, 25-29.9, 30-34.9 and 35-40 kg) to assist proper dosing of antiretrovirals for HIV-infected children in resource-limited settings. For drugs dosed by body surface area, we estimated likely weights and heights for age using standardized US growth charts for girls from which doses in weight bands were calculated. For this analysis, we calculated the difference between weight-based doses and body surface area-based doses for zidovudine 10 mg/mL oral solution, zidovudine 100-mg capsules, and didanosine 25, 50 and 100-mg chewable tablets using actual heights and weights from HIV-infected children in Africa and Romania.
Results: We used 1752 observations from 826 HIV-infected children (48% girls) from 9 countries. A total of 454 observations were in children <20 kg and 1298 ≥20 kg. For those <20 kg, the median difference of the weight-based dose as compared with the body surface area-based dose for zidovudine solution was -6.4% (range, -22.6, +13.7), zidovudine capsules +3.1% (range, -38.8, +44.7), didanosine chewable tablets +0.7% (range, -24.4, +22.5); for those ≥20 kg for zidovudine solution was 0.0% (range, -16.4, +11.8), zidovudine capsules +7.6% (range, -16.4, +36.9) and didanosine chewable tablets +1.2% (range, -16.4, +14.1). The dose precision for children <20 versus ≥20 kg was different for zidovudine solution (P < 0.001) and zidovudine capsules (P < 0.001), but not didanosine chewable tablets. The frequency that weight-based dose was more than 20% less than the body surface area-based dose for those <20 kg was 1.3% for zidovudine solution, 27.2% for zidovudine capsules and 4.9% for didanosine chewable tablets. For those ≥20 kg, the weight-based dose was never more than 20% less than the body surface area-based dose.
Conclusion: Dosing zidovudine and didanosine by weight band provides reasonably precise dosing as compared with body surface area-based doses. However, use of zidovudine capsules in children <20 kg results in under dosing by >20% in many instances. Didanosine chewable tablets allow for higher dosing precision compared with zidovudine capsules because of increased flexibility in the dosage form. Solid dosage forms of antiretroviral medications designed specifically for children are urgently needed.