July 2, 2014
Starting treatment with a CD4 count between 350-499 cells/mm3, instead of delaying until it falls below 350, may reduce the risk of death by up to a third and lower the risk of AIDS progression by up to one half, according to a study published in the journal AIDS.
In preparation for a review of the 2013 World Health Organization (WHO) guidelines, researchers analyzed 21 observational studies and three randomized controlled trials (RCTs) to determine the benefits and risks of starting antiretroviral therapy (ART) at a CD4 count of 350-499, compared to starting below 350. The researchers found that there were significant advantages to starting ART at 350-499; the only disadvantage was a higher risk of lab abnormalities reported in one RCT.
The WHO currently recommends starting ART in low- to middle-income countries at a CD4 count of 350-499. By contrast, guidelines in a number of high-income countries, including the U.S., recommend that therapy be initiated as close to diagnosis as possible, even at CD4 counts of 500 or more.
The review showed that starting treatment at CD4 counts of 350-499 reduced the risk of death by a quarter according to one RCT (25% lower risk, hazard ratio [HR] 0.77, 95% confidence interval [CI] 0.34 to 1.76), and by almost a third according to combined data from 13 cohort studies (30% lower risk, HR 0.66, 95% CI 0.55 to 0.79), when compared to starting therapy only at CD4 counts below 350.
Two RCTs concluded that such earlier treatment initiation lowered the risk of progression to AIDS or death by half (RR 0.48, 95% CI 0.26 to 0.91), while combined data from nine cohort studies showed such a risk to be 30% lower (RR 0.70, 95% CI 0.40 to 1.24).
However, according to one RCT, people who started treatment at CD4 counts of 350-499 had an almost 50% higher chance of grade 3 or 4 lab abnormalities than those beginning ART only at lower CD4 counts (RR 1.49, 95% CI 1.25 to 1.77).
The researchers noted that there was less evidence for the benefits of starting ART at CD4 counts of 500 or more. They hope that the results from two ongoing trials, START and TEMPRANO, will definitively answer the "When to start?" question.
Barbara Jungwirth is a freelance writer and translator based in New York.
Follow Barbara on Twitter: @reliabletran.