icon-folder.gif   Conference Reports for NATAP  
48th Annual ICAAC / IDSA 46th Annual Meeting
October 25-28, 2008
Washington, DC
Back grey_arrow_rt.gif
Risk of Death 70% Higher When Delaying ART With CD4s at 351 to 500
  48th ICAAC
October 25-28, 2008
Washington, DC
Mark Mascolini
Compared with people who began antiretroviral therapy (ART) within 1.5 years of having a CD4 count between 351 and 500, those who put off treatment longer had a 70% higher risk of death, according to results from a prospective US-Canadian cohort study. Neither hepatitis C virus (HCV) coinfection nor injecting drug use contributed to the higher death risk with delayed treatment, though both HCV and injecting drugs heightened the risk of death.
This analysis of the North American AIDS Cohort Collaboration on Research and Design involved 8374 US or Canadian residents with a CD4 count between 351 and 500 during follow-up between 1996 and 2006 in one of the collaborating cohorts. No one had clinical signs of AIDS, and no one had ever swallowed an antiretroviral. The investigators compared all-cause mortality in people who began treatment within 1.5 years of a CD4 count between 351 and 500 and people who did not start antiretrovirals within 1.5 years. The deferred-treatment group could include people who never began treatment or who died before starting therapy.
A cohort study like this has one obvious disadvantage compared with a randomized trial. In an observational study people decide on their own whether to begin treatment, basing their decision on a complex array of factors beginning with their clinician's advice, which is itself a potent bias. Observational studies have another less well recognized disadvantage: lead-time bias. In plain English, in a study like this investigators lack data for clinical events that happen during the treatment-deferral period. Savvy investigators, like those running this study, can make subtle adjustments to minimize the impact of lead-time bias, but they can't entirely erase the impact of this bias on their results.
The cohorts included 2473 people who did begin HAART within 1.5 years of reaching a 351-to-500 CD4 count and 5901 who did not. Median age of the groups stood at about 39 years, and just under 40% of each group were white. Median first CD4 count in the cohort was about 430 and median viral load about 15,000 copies. A slightly higher proportion in the deferred-treatment group had HCV infection (34% versus 27%) or a history of injecting drug use (21% versus 16%), but the investigators did not believe the scale of those differences is clinically meaningful.
Median CD4 count just before starting HAART measured 421 in the early group and 275 in the late group. The early starters began treatment at a median date of January 2000, compared with May 2001 in the deferred group. Types of regimens begun did not differ much between groups.
Among the 2473 people who started treatment sooner, 221 died from any cause. Among the 5901 who deferred treatment, 446 died. Statistical analysis adjusted for multiple death risk factors and stratified by cohort and calendar year determined that people who delayed antiretroviral therapy had a 70% higher risk of dying (relative hazard [RH] 1.7, 95% confidence interval (CI) 1.4 to 2.1, P < 0.001). Every additional 10 years of age independently raised the risk of dying 60% (RH 1.6, 95% CI 1.5 to 1.8, P < 0.001). Every 100-cell higher initial CD4 count trimmed the death risk 10%, but that advantage fell short of statistical significance (95% CI 0.7 to 1.0, P = 0.083).
Viral load had no impact on risk of death in this analysis. The disadvantage of deferring therapy held true in a subgroup of people with a known initial viral load. The 70% higher death risk with deferred treatment also held when the researchers restricted the analysis to people with a known drug-injecting status and adjusted the analysis for injecting drug use. Restricting the analysis to people with a known HCV status and adjusting for HCV also failed to budge the 70% higher risk with deferred treatment. However, a history of injecting drug use or HCV infection each individually made death more likely.
The investigators believe their findings support beginning antiretroviral therapy at a CD4 count between 351 and 500, higher than currently recommended in most guidelines. Because of the size of this cohort, the researchers argue that their findings apply to similar populations who have the same access to antiretroviral therapy.
Definitive advice on when to start antiretrovirals cannot come from a cohort study in which people are not randomized to start or delay treatment. As these investigators stress, early starters may have a survival advantage, for example, because they are more likely to have the motivation and resources to care for their health better than later starters. START, a trial that randomizes antiretroviral-naive people to begin therapy with more than 500 CD4s or to delay until they reach a count of 350, will begin early in 2009.
1. Kitahata MM, Gange SJ, Moore RD. Initiating rather than deferring HAART at a CD4+ count between 351-500 cells/mm3 is associated with improved survival. 48th Annual International Conference on Antimicrobial Agents and Chemotherapy (ICAAC). October 25-28, 2008. Washington, DC. Abstract H-896b.