Debate continues over when HIV-positive adults should consider starting treatment, taking account of factors such as their physical health and the presence of any symptoms, their CD4 count and their viral load. Different people have varying views over how to use viral load testing to assess how well an individual has responded to treatment, and when it may be time to switch to new drugs.

For HIV-positive children, there is even more uncertainty about using surrogate markers to guide treatment choices. In the past it used to take months to establish which children born to HIV-positive women were themselves infected with HIV, but infected children can now be diagnosed very early in life. Since most infected children are exposed around the time of delivery, this means that it is possible to offer anti-HIV treatment very soon after infection. However, this does depend on the mother knowing her own HIV status during pregnancy.

Treatment is recommended where the infant has symptoms of immune deficiency or signs of a weakened immune system. US treatment guidelines endorse antiretroviral therapy in HIV-infected infants aged less than twelve months, due to the high rates of disease progression and death in this group. However, this view is based on expert assessment of the risks (e.g. side effects, inadequate dosing leading to resistance) versus the benefits, rather than clinical data. Whether treatment is the best course of action for every asymptomatic infant is unclear. There is some evidence that treating asymptomatic infants may undermine the development of effective immune responses to HIV - the opposite effect to early treatment in adults (Luzuriaga) - although in others it may prevent illness and death.

In Europe, PENTA guidelines suggest that all symptomatic children and infants should receive treatment. For asymptomatic children the decision to start treatment should be individualised on the basis of a child's risk of progression to AIDS or death based on CD4 percentage and age. A risk of progression to AIDS of >10% and a risk of death in the next 12 months of >5% is considered to be a trigger to start treatment. A calculator is available online at http://www.ctu.mrc.ac.uk/penta/hppmcs/calcProb.htm. These guidelines are based on a meta-analysis of pooled data from 3941 children (Dunn 2003).

In infants below the age of 1 year the PENTA guidelines suggest that it may be more appropriate to consider treatment based on a six month risk of AIDS or death. However the guidelines also note that many doctors would prefer to treat all asymptomatic children below the age of 1 year.

Despite the absence of a randomised study looking at the best time to commence therapy in HIV-infected infants, there are data which suggest that very early treatment of newborns is feasible and modestly effective. The observational PENTA 7 study looked at 20 infants who began treatment with d4T/ddI/nelfinavir before three months of age. This study found incomplete viral suppression in 70% of the infants after 72 weeks of follow-up, with 30% of the babies developing some drug resistance (PENTA Study Group 2004). This result is inferior to that observed in adults following treatment for primary infection, although it should be noted that this particular combination is no longer a preferred first-line combination.

For children over one year of age, decisions about treatment are generally based on immune function, although some experts recommend treatment of all HIV-infected children and infants, regardless of immune function. In part, this is because CD4 counts vary with age, so it's difficult to use them as a guide to treatment decisions in children.

In the first few months after birth the CD4 cell count may be above 2000 cells/mm³ (compared with 500 to 1200 cells/mm³ in uninfected adults), and it will decline as the child grows older, although it will not reach the adult level until late childhood. This makes it difficult to decide when the CD4 cell count is declining, unless the fall in CD4 cell count is very fast and dramatic. The CD4 cell counts at which children experience opportunistic infections can be much higher than those in adults. For more details of CD4 cell counts in children, see the CDC surveillance definition in the section Children: Centers for Disease Control and Prevention classification, 1994 in Introduction to HIV and AIDS: Definitions.

The viral load marker is also regarded as a less reliable marker of disease progression in children and infants than in adults. Viral load is very high during the first months of life, just as it is in recently infected adults, but it seems to take longer to decline again to the 'set-point', and there are few data on viral load later in childhood. A review of 70 children enrolled for the Penta trial of AZT in asymptomatic children found that viral load was significantly higher in children under the age of two, and viral load continued to decline up to the age of four years old. However, in both adults and children, suppressed viral load significantly reduces the risk of disease progression in the short-to-medium term (Palumbo 1998).

References

Luzuriaga K. Early therapy of vertical human immunodeficiency virus type 1 (HIV-1) infection: control of viral replication and absence of persistent HIV-1-specific immune responses. J Virol 74: 6984-6991, 2000.

Palumbo PE et al. Predictive value of quantitative plasma HIV RNA and CD lymphocyte count in HIV-infected infants and children. JAMA 279: 756-761, 1998.