d e c e m b e r 2 0 0 9 T H E S O U T H E R N A F R I C A N J O U R N A L O F H I V M E D I C I N E When to initiate antiretroviral therapy (ART) in child- ren and adults has been determined by disease stage and CD4 count or percentage. The starting criteria were based on cohort data and clinical experience. The main groups determining paediatric guidelines are from the USA, Europe (PENTA - Paediatric European Network for Treatment of AIDS) and the World Health Organization (WHO).1-3 With the advent of effective triple therapy from 1995, guidelines have adapted to new data. Ini- tial staging clinical and immunological criteria were those of the Centers for Diseases Control and Preven- tion (CDC).4 The staging criteria were based on North American experience and were not always appropriate for Africa. For example, failure to thrive, bronchiecta- sis and pulmonary tuberculosis were not adequately addressed. Although the CDC classification included CD4 deple- tion, it was only with the HIV Prognostic Markers Col- laborative Study (HPMCS), representing the combined data of 3 941 HIV-infected infants and children from the USA and Europe in the pre-highly active antiret- roviral therapy (HAART) era, that the relationship be- tween age and CD4 percentage became clearer.5 The study showed that the younger the infant, the higher the CD4 percentage or absolute count at which there was a high risk of disease progression or death within the next 12 months. As the CD4 count normally de- clines with age until 5 years of age and the CD4 per- centage remains stable, the CD4 percentage is used as a guideline until then (in the absence of lymphopenia). However, the CD4 count is an extremely accurate pre- dictor of outcome.6 These findings were subsequently incorporated into all paediatric guidelines. The establishment of a WHO classification system in 2006 helped to focus on conditions more frequently encountered in Africa.7 The system has 4 stages, with ART recommended for stages 3 and 4. The special vul- nerability of young infants to rapid disease progres- sion and death was not fully appreciated at this time, although allowance was made for initiating ART in HIV antibody-positive infants with severe disease before confirmation by HIV DNA polymerase chain reaction (PCR). The decision on when to initiate ART in infants and child- ren has always been influenced by fears of long-term toxicity or antiretroviral (ARV) resistance if therapy is started too soon. Also, ARV choices for children are far more limited than for adults. Second-line therapy is es- pecially unsatisfactory. However, there is increasing realisation, especially from adult data, that delaying ART until the CD4 cell count falls below 350 cells/µl is associated with increased morbidity and mortality.8,9 ART is therefore recom- mended at higher CD4 thresholds than previously. Mortality in HIV-infected infants is exceedingly high. The first data came from a study of combined outcome in nine vertical transmission prevention (VTP) studies conducted in sub-Saharan Africa.10 Thirty-five per cent of HIV-infected infants had died by 1 year of age and 52.5% by 2 years of age. The ZVITAMBO study in Zim- babwe showed a similar but slightly higher mortality in the first 2 years of life.11 For both studies, co-tri- WHEN TO START ANTIRETROVIRAL THERAPY IN INFANTS AND CHILDREN C L I N I C A L Mark F Cotton, FCPaed (SA), PhD, DTM&H, DCH (SA), Cert ID (SA) Helena Rabie, FCPaed (SA), MMed (Paed) Department of Paediatrics and Child Health and KID-CRU, Tygerberg Children’s Hospital and Faculty of Health Sciences, Stellenbosch University, Tygerberg, W Cape Ute Feucht, FCPaed (SA), MMed (Paed), Dip HIV Man (SA), CAHM Department of Paediatrics, Kalafong Hospital and University of Pretoria Avy Violari, FCPaed (SA) Perinatal HIV Research Unit, Chris Hani Baragwanath Hospital, Soweto, Johannesburg We review the background and key studies that inform decisions on when to initiate antiretroviral therapy (ART) in infants and children. The World Health Organization staging system from 2006 was based on conditions com- monly seen in Africa and provided an impetus for advancing ART in children. Because of poor predictive value of CD4 counts in infancy and inability to predict risk of death or disease progression, we recommend initiating ART in all infants under a year of age. CD4 thresholds for initiating therapy decline as children become older. WHO stage 3 and 4 should trigger ART regardless of CD4 count. Over 5 years of age, a CD4 count <350/µl requires ART. 50 INFANTS BELOW 12 MONTHS OF AGE T H E S O U T H E R N A F R I C A N J O U R N A L O F H I V M E D I C I N E d e c e m b e r 2 0 0 9 moxazole prophylaxis for prevention of Pneumocystis pneumonia (PCP) was not given. The Children with HIV Early Antiretroviral (CHER) trial is the first ARV trial to prospectively evaluate ART strat- egies to inform ARV guidelines.12 Before this, limited cohort studies favoured early ART. For example, Faye et al. reported an improved outcome in 40 infants com- mencing ART before 6 months of age compared with 43 starting later. In the early group there was no dis- ease progression versus 7 events, including 3 cases of encephalopathy, in those treated later.13 CHER commenced in July 2005 in two South Afri- can sites, the Perinatal HIV Research Unit (PHRU) in Soweto and the Children’s Infectious Diseases Clinical Research Unit (KID-CRU) at Tygerberg in the Western Cape. The hypothesis of the study is that early ART will have long-term benefit by delaying the need for con- tinuous therapy. At the time of planning and initiation, the standard practice in infants was to initiate ART for a low CD4 percentage (<20% until August 2006 and 25% thereafter, in the first year of life) or evidence of severe clinical disease.3,14 In the first part of the study, infants with baseline CD4 >25% were randomised to deferred ART (arm 1), where ART was initiated once treatment criteria were reached (CD4 <20% until Au- gust 2006 and <25% thereafter or severe HIV disease), or early ART commencing before 12 weeks of age until either the first (arm 2) or second (arm 3) birthdays, with interruption of ART until indicated through CD4 deple- tion (<20% after the first birthday) or clinical disease progression. For the first year of the study, deferred ART (arm 1) was compared with early ART (arms 2 and 3). On 20 June 2007, after the study had been open for 2 years, the Data Safety Monitoring Board for CHER, noting significantly improved survival in subjects in the early ART arms, recommended that no infants be ran- domised to deferred ART and that data until this time point be released.12 The difference in mortality and dis- ease progression is shown in Fig. 1. A key finding in CHER was high early mortality, mainly in the deferred arm, which diminished as the infants be- came older (Table I). Also of note, however, is that mor- tality was also higher early on in the early ART group, diminishing over time in the first year of the study. Subsequently, ARV guidelines for children through- out the world incorporated the findings of CHER and recommended that all HIV-infected children below 12 months of age start ART irrespective of CD4 status or disease progression.1, 2,15 More recently, a study by Bourne and colleagues, using birth and death data from Statistics South Africa be- tween 1997 and 2002, showed a peak in post-neonatal mortality, rising each year directly proportional to the rising HIV antenatal seroprevalence.16 These data are shown in Fig. 2. Also of note is that neonatal deaths are not addressed. It is possible that HIV may contribute significantly to neonatal deaths as well. These data, together with CHER, illustrate the enor- mous dilemma and problems for child health pro- grammes. In CHER the diagnostic HIV DNA PCR was done from 4 weeks of age with a 7-day turnaround time. In the public sector, the PCR is done at 6 weeks of age, to coincide with the 6-week immunisation visit. Turnaround time for the test results and then commu- nication to the caregiver all take time during a period of great vulnerability. One of the main problems is identifying HIV-infected children early and getting them into care. Because of 51 Fig. 1. Deferred ART is associated with a) a higher probability of death or b) first HIV event in the CHER trial. Death rate per 100 person- years (95% CI) Weeks on study* Early ART Deferred ART 0 - 13 9.8 (3.6 - 21.4) 40.6 (21 - 71) >13 - 26 3.7 (0.4 - 13.2) 19.7 (6.4 - 45.9) >26 - 52 3.1 (0.4 - 11.2) 6.8 (0.8 - 24.7) *Infants were <12 weeks of age at study entry.12 TABLE I. HIGH EARLY MORTALITY RATE IN THE CHER STUDY Fig. 2. Early peak in infant mortality proportional to rising antenatal HIV seroprevalence, 1997 - 2002 (reproduced with permission from AIDS16). d e c e m b e r 2 0 0 9 T H E S O U T H E R N A F R I C A N J O U R N A L O F H I V M E D I C I N E 52 the importance of the caregiver in giving ARVs, previ- ous guidelines emphasised adequate preparation and training over a number of weeks. Because of the new recognition of early infancy as a period of high mortal- ity, there is a need to ‘fast-track’ preparation and also to continue training and support after having initiated ARVs. Practical difficulties are illustrated by experience in the Tshwane district. In 2009, 82% of children newly referred to the Kalafong ART site had stage 3 or 4 dis- ease, the majority being referred from the in- and out- patient service. Of the 14 000 - 22 000 HIV-infected children thought to be in Tshwane, only between 3 000 and 4 000 are receiving ART or are in care. In this group there is less certainty on when to initiate ART. Ideally, decisions must be based on randomised studies. The PREDICT is underway in Thailand and Cam- bodia and will assess the CD4 thresholds for therapy in children between 1 and 12 years of age. All guidelines recommend using absolute CD4 counts from 5 years of age. For simplicity, the WHO regards 1 - 5 years of age as a single group, while the PENTA guidelines have 1 - 3 and 3 - 5 years of age. This group of children also represents a wide age spectrum. Clini- cal disease stage is extremely important and the ma- jority of children are symptomatic. The dissociation of CD4 from disease severity has been well documented in South African children.17 The HPMCS data, although showing a low risk of disease progression for higher CD4 percentages, still showed notable death and dis- ease progression with CD4 >30% up to 6 years of age.5,6 Initial data suggested that young children show excel- lent immune restoration once ART is initiated, but more recent studies suggest a poorer response if therapy is delayed to CD4 <15%. Initial reports also suggested that viral suppression was difficult to achieve, espe- cially in younger infants with extremely high viral loads. However, potent regimens and a better understanding of pharmacokinetics have improved outcomes. Importantly, health care providers caring for children often do not fully appreciate the extent of end-or- gan damage. Mild cognitive and behavioural changes, milder forms of chronic lung disease, renal and hepatic disease and poor growth may easily be overlooked, leading to irreversible organ damage, growth failure and delayed puberty. In the 3C4KIDS, 2 510 children over a year of age from Africa and Brazil contributed 357 deaths and 3 769 child-years at risk.18 None had access to ART, with 81% follow-up occurring on co-trimoxazole. The 12-month risk of death was higher than in the HPMCS for CD4 count and percentage. Most importantly, when anae- mia and failure to thrive were included, the CD4 thresh- olds for predicting death or severe disease were much higher (Fig. 3). Both of these conditions are markers for advanced HIV disease. Most importantly, there was a real risk of death at high CD4 counts and percentages. For these reasons, our guidelines favour earlier ART be- tween 1 and 5 years of age, in keeping with current US recommendations.2 There has been much progress in refining and improving ARV guidelines for children. More changes can be ex- pected as knowledge advances. Most important, how- ever, is the realisation that each child is unique and his or her individual and family circumstances need to be considered for maximum benefit. REFERENCES 1. The PENTA 2009 guidelines for the use of antiretroviral therapy in paediatric HIV-1 infection. HIV Medicine 2009; 10: 591-613. 2. Working Group on Antiretroviral Therapy and Medical Management of HIV- Infected Children. Guidelines for the Use of Antiretroviral Agents in Pediatric HIV Infection; 2008. 28 February 2008. http://aidsinfo.nih.gov/ContentFiles/ PediatricGuidelines.pdf (accessed 1 March 1 2008). 3. World Health Organization. Antiretroviral Therapy of HIV Infection in Infants and Children in Resource-Limited Settings: Towards Universal Access. Recommendations for a Public Health Approach. Geneva: WHO, 2006. 4. Centers for Disease Control and Prevention. Revised classification system for human immunodeficiency virus infection in children less than 13 years of age. MMWR Morb Mortal Wkly Rep 1994; 43: 1-10. 5. HIV Paediatric Prognostic Markers Collaborative Study Group. Short-term risk of disease progression in HIV-1-infected children receiving no antiretroviral therapy or zidovudine monotherapy: a meta-analysis. Lancet 2003; 362: 1605-1611. 6. HIV Prognostic Markers Collaborative Study. Predictive value of CD4 count and viral load for disease progression in untreated HIV-infected children. AIDS 2006; 20: 1289-1294. 7. World Health Organization. WHO Case Definitions of HIV for Surveillance and Revised Clinical Staging and Immunological Classification for HIV-Related Disease in Adults and Children. Geneva: WHO, 2006. 8. Kitahata MM, Gange SJ, Abraham AG, et al. Effect of early versus deferred Fig. 3. Estimated risk of death within 12 months for a child on co-trimoxazole prophylaxis by age, weight-for-age z- score and haemoglobin and CD4 percentage or count. (a) CD4% at 2 years of age; (b) CD4% at 5 years of age; (c) CD4 count at 2 years of age; and (d) CD4 count at 5 years of age (reproduced with permission from AIDS18 ). CHILDREN >12 MONTHS OF AGE CONCLUSION T H E S O U T H E R N A F R I C A N J O U R N A L O F H I V M E D I C I N E d e c e m b e r 2 0 0 9 antiretroviral therapy for HIV on survival. N Engl J Med 2009; 360: 1815-1826. 9. Sterne JA, May M, Costagliola D, et al. Timing of initiation of antiretroviral therapy in AIDS-free HIV-1-infected patients: a collaborative analysis of 18 HIV cohort studies. Lancet 2009; 373: 1352-1363. 10. Newell ML, Coovadia H, Cortina-Borja M, Rollins N, Gaillard P, Dabis F. Mortality of infected and uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet 2004; 364: 1236-1243. 11. Marinda E, Humphrey JH, Iliff PJ, et al. Child mortality according to maternal and infant HIV status in Zimbabwe. Pediatr Infect Dis J 2007; 26(6): 519-526. 12. Violari A, Cotton MF, Gibb DM, et al. Early antiretroviral therapy and mortality among HIV-infected infants. N Engl J Med 2008; 359(21): 2233-2244. 13. Faye A, Le Chenadec J, Dollfus C, et al. Early versus deferred antiretroviral multidrug therapy in infants infected with HIV type 1. Clin Infect Dis 2004; 39(11): 1692-1698. 14. World Health Organization. Scaling up antiretroviral therapy in resource limited settings: guidelines for a public health approach. April 22, 2002. http://www. helid.desastres.net/?e=d-0who--000--1-0--010---4-----0--0-10l--11es-5000- --50-packc-0---01131-010-muLPqGf842f94771000000004ab19d67-0-0-0&a= d&c=who&cl=CL2.1.5&ld=Js2888e (accessed 18 November 2009). 15. World Health Organization. Report of the WHO Technical Reference Group, Paediatric HIV/ART Care Guideline Group Meeting WHO Headquarters, Geneva, Switzerland; 2008 June 18, 2008. http://www.who.int/hiv/pub/paediatric/WHO_ Paediatric_ART_guideline_rev_mreport_2008.pdf (accessed 19 June 2008). 16. Bourne DE, Thompson M, Brody LL, et al. Emergence of a peak in early infant mortality due to HIV/AIDS in South Africa. AIDS 2009; 23(1): 101-106. 17. van Kooten Niekerk NK, Knies MM, Howard J, et al. The first 5 years of the family clinic for HIV at Tygerberg Hospital: family demographics, survival of children and early impact of antiretroviral therapy. J Trop Pediatr 2006; 52(1): 3-11. 18. Duong T, Gibb DM, Dunn DT, et al. (3Cs4Kids). Markers for predicting mortality in untreated HIV-infected children in resource-limited settings: a meta-analysis. AIDS 2008; 22: 97-105. 53