• Predictors of CD4 at HAART Initiation (Poster 567)
  Authored by M. Jonsson Funk on behalf of the CASCADE Collaboration
  

Dr. Funk and colleagues presented data from the CASCADE collaboration, which is a consortium of combined data from over 8,000 HIV seroconverters from 21 different clinical cohorts in Europe, Canada and Australia. The aim was to identify which demographic or regional variables in terms of CD4 count are applied in the decision to start highly active antiretroviral treatment (HAART). They included all patients from these combined cohorts who had started HIV treatment and had a CD4 count and viral load available prior to starting treatment.

There were 1,366 patients who started HAART and met the study criteria. The median age at seroconversion was 28 years. The median year of seroconversion was 1992. Twenty-one percent of the group were women. The median viral load was 4.5 logs/mL.

Of those who were completely naive to treatment, the mean CD4 count when starting treatment was 356. Between 1996-99 the CD4 count range was 350-400. However, since 1999, the CD4 count at which treatment was initiated has decreased, so that in 2002 the average was 247.

In those patients who had received some previous HIV treatment, but not HAART, the mean CD4 count when starting HAART was 284. Also, between 1996-99 the CD4 count range was 198-360, and again in 2002 the mean CD4 at which HAART was started decreased to 190. Interestingly, patients who had acquired HIV infection through heterosexual contact or injection drug use started HAART later and had mean CD4 counts at treatment initiation of 231 and 208 respectively.

Women were more likely to start HAART earlier than men, with a mean CD4 count of 308. There were also significant geographic differences in the CD4 count at which HAART was initiated, ranging from 138-562. In some cases, there were significant differences between clinics within the same country.

In general, HAART was initiated earlier during first three years that HAART was available, which reflects the general thinking at the time. Now, with the latest information about the long-term side effects of HAART and newer guidelines about CD4 count and viral load thresholds reflecting the limited risk of disease progression above a CD4 count of 350, the CD4 count at which most patients start treatment has decreased.

In patients who began treatment prior to the HAART era, the overall reduced CD4 count at the time of HAART initiation reflects more advanced disease because of the limited potency of the single or double combination regimens prevalent in the 1980s and early 1990s.

As the authors suggest, the CD4 count differences seen for HAART initiation in different geographic locations are probably the result of:

• differences in cohort composition,

• country or clinic specific guidelines for when HAART should be initiated,

• reimbursement issues, and

• emerging data or

• changes in expert opinion on what the CD4 threshold to initiate HIV treatment should be.