October 26, 2008
A large study presented at the joint 2008 ICAAC/IDSA in Washington, DC found that people who delayed starting HIV treatment until their CD4 counts fell below 350 were 70% more likely to experience a new AIDS-defining illness or death, than compared to those who began at CD4 counts of 350-500. This finding adds to the growing body of evidence supporting earlier treatment of HIV disease.
The researchers looked at the medical records of over 8,000 people from 22 research cohorts throughout the US and Canada between 1996 and 2006. Of those, almost 2,500 started HIV treatment with CD4 counts of 350-500, with the rest delaying treatment until sometime after their CD4 counts fell below 350. Current US guidelines recommend treatment for anyone with a CD4 count below 350.
Overall, people who started with lower CD4 counts were 1.7 times as likely to experience an AIDS-defining illness or death. Interestingly, the researchers found that neither a history of injection drug use nor HCV co-infection changed the increased risk associated with delayed treatment, although they did lead to more frequent illness and death overall. Not surprisingly older age was independently associated with an increased likelihood of illness or death, with each decade adding a 60% risk.
The researchers had no way of directly measuring adherence to treatment in this study. They did look at the likelihood that people in each group had undetectable HIV levels once they began treatment and found no significant difference between the groups.
Being a retrospective, combined cohort analysis limits the strength of this studys observation. The presenters acknowledged this, and they said they took great pains to control for the kinds of biases that might influence their findings.
The same research team is analyzing this database to compare the rates of AIDS-defining illness and death between people who start HIV treatment with CD4 counts above and below 500. They expect to present those results soon.
In spite of the limitations of this kind of study, its findings are nonetheless important. Lacking a definitive, prospective, randomized study to answer this question, these studies may influence guidelines and prescribing practices.