Early versus delayed treatment of HIV infection. Zidovudine should be given before symptoms develop.

An understanding of the virology and pathogenesis of HIV infection provides a rationale for initiating early intervention with antiretroviral drugs. Even at the earliest stages of infection when HIV-infected patients are asymptomatic, viral replication is ongoing, particularly in lymphoid tissues. Initiation of antiretroviral therapy can reduce viral replication and delay disease progression. A possible objection to early intervention therapy with zidovudine is the risk of selecting out resistant isolates of HIV, which would be difficult to treat. In practice, zidovudine-resistant isolates occur significantly less frequently in patients with early-stage disease compared with those with late-stage HIV infection, thus supporting the early use of zidovudine; in addition, alternative therapies, active against zidovudine-resistant isolates, are available. Clinical trials with zidovudine in asymptomatic patients have differed in terms of length of follow-up, patient inclusion criteria, dosages and end-points. However, a number of conclusions are possible based on the results obtained: early intervention delays the progression of AIDS, delays the onset of symptomatic disease, has a favourable effect on surrogate markers of HIV infection and is well tolerated; it does not, however, seem to produce any benefit in terms of survival. It is this last point that has given rise to much of the controversy regarding early intervention with zidovudine in asymptomatic patients. Since the disease is progressive in nature with persistent and high levels of viral replication and as prolonging the period of relative health and quality of life when the patient is asymptomatic is desirable, the choice to treat before symptoms develop would appear to be the optimal therapeutic strategy.