Weight-band dosing tables: Simplifying paediatric ART

One of the obstacles to scaling up paediatric antiretroviral therapy (ART) coverage in resource-limited settings is the relative complexity of paediatric dosing. There is a need to simplify ART in order to facilitate treatment initiation and ongoing management of infants and children by health care providers, as well as to support adherence in the home. This article reviews the development of weight-band dosing tables as a strategy for simplifying the delivery of paediatric ART.

In 2007, only 8% of the estimated 1 800 000 children (0 -14 years) living with HIV in sub-Saharan Africa were receiving antiretroviral therapy (ART). Coverage will need to be expanded greatly if the goal of providing ART to 80% of children in need by 2010 is to be met. 1 Moreover, recent evidence highlights early initiation of ART as particularly critical for infants with HIV. 2 Clinical guidelines issued by the World Health Organization (WHO) now recommend immediate initiation of ART for all HIV-infected infants. 3 Paediatric ART management involves a complex process of interactions between patients, families, health care providers and the antiretrovirals (ARVs) themselves. Barriers to the delivery of effective treatment occur both within the health care system and in the home. These include delayed diagnosis, limited availability of health care providers trained in paediatric ART, few available paediatric ARV formulations, complicated regimens and dosing schedules, and poor palatability of some ARVs. Difficulties in the home include overcrowding, difficult work schedules of the parents and the stresses associated with parental disclosure in the home. Unlike adults, children require changes in antiretroviral dose as they grow and become older, and rely upon adult caregivers to administer medicines. 4 Children have traditionally been dosed according to body surface area (BSA) (e.g. zidovudine (AZT), didanosine (ddI), lopinavir/ritonavir (LPV/r)), weight (mg/kg) (e.g. stavudine (d4T), lamivudine (3TC), abacavir (ABC), nevirapine (NVP)) or dose per weight band (efavirenz (EFV)). Manufacturers' recommendations for some ARV drugs (e.g. LPV/r) include both BSA and weight-based dosing methods. 5 The calculation of BSA generally requires accurate measurement of the child's weight and length or height (ideally with a stadiometer or measuring box), and a normogram or mathematical formula (e.g. Mosteller formula). 6 For both BSA and mg/kg weight-based dosing approaches, the calculated dose of each ARV drug must be rounded up or down as a 'best-fit' dosage according to which solid or liquid formulations of the ARV drug are available. This may lead to confusion and uncertainty on the part of the prescriber.
In resource-limited settings (RLS), primary care doctors and nurses rather than paediatricians are responsible for the majority of paediatric ART initiation and follow-up. Lack of accurate measuring equipment and the relative complexity of BSA dosing may inhibit initiation of ART in infants and young children or mean that inappropriate doses are given. While the most 'accurate' dosing may be obtained with the use of liquid formulations, large volumes of solutions (which may require refrigeration, e.g. d4T solution) may be challenging for caregivers to administer to young children, particularly if palatability is poor (e.g. LPV/r).
ART simplification strategies are required to help health care providers manage ART in children, and to help caregivers and children adhere to therapy. Interventions include the use of adult or preferably paediatric fixed-dose combinations (FDCs), selection of doses based on weight band rather than individual mg/kg or BSA doses, prescription of pills or capsules rather than liquids, and identification of reliable oncedaily regimens.
Weight-band dosage tables assist health care providers by assigning a fixed dose of medication for a particular weight range (e.g. 1.5 ml of LPV/r solution for children weighing 4 -9.9 kg, or half a 150 mg 3TC tablet for children weighing 14 -19.9 kg). In large public sector WEIGHT-BAND DOSING TABLES: SIMPLIFYING PAEDIATRIC ART C l i n i C a l One of the obstacles to scaling up paediatric antiretroviral therapy (ART) coverage in resource-limited settings is the relative complexity of paediatric dosing. There is a need to simplify ART in order to facilitate treatment initiation and ongoing management of infants and children by health care providers, as well as to support adherence in the home. This article reviews the development of weight-band dosing tables as a strategy for simplifying the delivery of paediatric ART.

62
MOVinG FROM BSa anD WEiGHT-BaSED DOSinG aPPROaCHES TO WEiGHT-BanD DOSaGE TaBlES treatment programmes, tables can reduce the time and risk of dosing errors involved in calculating multiple ARV doses by weight or BSA. They can also facilitate easier checking of doses against weight gain by clinicians, nurses, pharmacists or adherence counsellors. Under-dosing of ARVs, due in part to a lack of regular dose adjustments for ongoing growth, has been described in a large cohort of children in the UK and Ireland. 7 A study in Thailand found that 17 of 18 doctors using a standardised drug dosage table avoided miscalculations and reported more confidence with prescriptions. 8 In 2006, the WHO published simplified weight-band dosing tables on all ARV drugs for which there were available paediatric indications, formulations or sufficient information and evidence to provide guidance on prescribing and doses. 9 Decisions about dosing were based upon manufacturers' information, ARV formulation choices, available data from clinical studies and expert paediatric pharmacology consultation. What was considered to be the 'optimal' dose for a particular weight band, given the limitations imposed by currently available drug formulations, was selected. Weightband doses were determined by using BSA values calculated from median height-for-weight from international growth charts using the following formula: The dosing tables are directed at RLS and are based on the following principles: 1. It is preferable to use one formulation or fixed combination of any given drug(s bands, accepting that some deviation from target dosing will occur. 11. Children have to be weighed at each clinic visit, and dose changes are required as children grow and/or gain weight. 12. When capsules are opened or tablets dissolved or crushed and added to food or liquid, it is important that the entire volume/amount of vehicle be taken to ensure administration of the full dose. The dosing tables are based on standardised weight bands starting from 5 kg body weight for the individual ARV drugs (excluding EFV, which starts from 10 kg) and 10 -14 kg for the fixed-dose combinations (AZT + 3TC, d4T + 3TC, AZT + 3TC + ABC, d4T + 3TC + NVP). The weight bands are in 1 kg divisions from 5 to 11 kg, and 2 -5 kg divisions from 12 to 35 or 40 kg.
Dosing tables may be adapted according to the specific drug formulations available to a regional or national treatment programme. For example, dual (d4T + 3TC) or triple (d4T or AZT + 3TC + NVP) FDCs play an important role in paediatric ART in many countries (e.g. Thailand and many African countries), but are not widely used in South Africa. Satisfactory early clinical and immunological outcomes have been described following the use of fractions of generic adult FDCs in children dosed according to a weight-band table method in Thailand. 10 In another project, a visual dosing aid (VDA) incorporating coloured dosing bands for five first-line ARV drugs was developed to assist clinicians in prescribing paediatric ART consisting of syrups, generic adult tablets or a combination. It compared well with generic paediatric FDC tablets and could help facilitate paediatric ART roll-out in RLS. 11 The South African national guidelines for the management of HIV-infected children (2005)  prior to the more widespread availability of FDCs. BSA dosing recommendations were converted to weightband doses using approximations of weight-for-age and height-for-age derived from standardised growth charts of girls from birth to 36 months and 20 years in the USA (National Center for Health Statistics). Dose for each weight band was based on the practicality of the available dosage forms for each drug, and practical storage and dosing instructions were included with the chart. The chart lacks dose recommendations for children weighing less than 5 -7 kg and for HIV-TB co-infection.
The Western Cape Antiretroviral Drug Dosing Chart for Children (2007), based on the 2006 WHO recommendations, was developed by the author and colleagues as a pilot project for the HIV Directorate of the Western Cape (South Africa) provincial health department. The chart was directed at both clinicians and pharmacists involved in prescribing and dispensing ART for children, and it was successfully piloted at a number of HIV clinics before being distributed across the province. It was produced as a laminated A4 colour copy. Standard first-and second-line ART regimens (including regimens compatible with rifampicin-based TB treatment) as well as general comments relating to storage, administration and common drug interactions and sideeffects of the individual ARV drugs were printed on the reverse side. Only ARV formulations available at public sector treatment facilities were included (no FDCs were available), and there was an emphasis on the early introduction of solid formulations where possible. Fractions of tablets (not less than half a tablet) were only incorporated for scored tablets. The target dose in mg/ kg or mg/m 2 for each drug as well as the formula for calculating BSA was included to facilitate comparison of dosing methods when necessary.
The following local practices were incorporated: 1. The off-label 'opened capsule' d4T dosing method whereby a d4T capsule is opened and the powder contents dispersed in a standardised volume of water and the required dose drawn up with a syringe and administered to the child (e.g. to administer a standardised d4T dose of 10 mg for the weight bands 7 -9.9 kg, a 20 mg d4T capsule is opened and dispersed into 5 ml of water and 2.5 ml is withdrawn with a syringe and administered and the remaining 2.5 ml is discarded). The rationale for this method is to minimise the use of stavudine solution (1 mg/ml when reconstituted), which is expensive, requires refrigeration, and results in relatively large medication volumes for administration to young infants at the usual dosing schedule (1 mg/kg/dose). There are now data based on high-performance liquid chromatography analysis of active drug concentration in dispersed capsule solutions, supporting the accuracy of this method for certain brands of d4T capsules. 14 2. Standardised doses of ritonavir used for pharmacological boosting of LPV/r (in order to achieve a ratio of 1:1) in children receiving concurrent LPV/r and rifampicin-based TB treatment. 15 3. Colour coding of ARVs on the chart corresponding to colour coding methods used in the pharmacy for medication containers and syringes to assist parents and caregivers with correct dosing. 4. The chart incorporated co-trimoxazole and multivitamin syrup dosing according to weight bands. 5. Since there were no standard weight-band dosing recommendations for infants weighing <5 kg in the WHO document (2006), it was recommended that a clinician experienced in ARV prescribing be consulted for such cases.
There are a number of approaches to validation of weight-band dosing. Direct methods include pharmacokinetic (Pk) studies and therapeutic drug monitoring. Indirect methods include comparison of weight-based with BSA doses, and safety and efficacy studies. Qualitative studies assessing the usefulness of a weightband dosing table to prescribing clinicians and dispensing pharmacists should be undertaken.
Differing growth rates and the prevalence of malnutrition could have a significant impact on the accuracy of weight-band dosing of drugs that are usually dosed by BSA. A study comparing the calculated BSA dose range with WHO weight-band doses for AZT, ddI, NVP and LPV/r using actual heights and weights of 601 children at the time of ARV initiation was undertaken at a tertiary hospital in South Africa in 2007. 16 The median age was 28 months (interquartile range 13 -62), 49% of children weighed <10 kg, and 59% and 63% of children had weight-for-age and height-for-age z-scores <-2 (moderate to severe underweight or stunting), respectively. Children with body weight <5 kg were excluded as weight-based dosing recommendations were unavailable for this category, and children <6 months of age were excluded as LPV/r BSA dosing recommendations are different in this age group. The BSA dose ranges used were AZT 180 -240 mg/m 2 , ddI 90 -120 mg/m 2 , NVP 160 -200 mg/m 2 , and L/r 230 -300 mg/m 2 .
Results are presented in Table I. The conclusion of this study was that the 2006 WHO simplified weight banddosing method effectively avoided under-dosing children in relation to existing BSA dose recommendations for AZT, NVP and LPV/r suspensions. However, the authors noted that the risk of over-dosing is greater with weight-band recommendations for existing capsule or tablet formulations of these ARVs. Further studies are recommended for the WHO weight-band dosing method.  A detailed review of Pk studies and clinical outcome, safety and efficacy studies undertaken in children treated with individual or FDC ARV drugs dosed according to weight bands is beyond the scope of this article. There are no reported Pk or clinical studies directly comparing weight band with mg/kg or BSA dosing approaches.

ValiDaTiOn OF WEiGHT-BanD DOSinG TaBlES
In July 2008, the WHO published a revised weightband dosing table for individual ARVs as well as dual and triple FDCs applicable to children >6 weeks of age, indicating the number of tablets or millilitres of solution to be administered twice daily by weight band from 3 kg to 34.9 kg (Fig. 1). 17 The