Variability in serial CD4 counts and relation to progression of HIV-I infection to AIDS in haemophilic patients. Transfusion Safety Study Group.

OBJECTIVE--To examine the CD4 count and its near term changes relative to progression to AIDS within 30 months and to subsequent CD4 counts. DESIGN--Longitudinal clinical and laboratory study. SETTING--Haemophilia treatment centres in six large American cities. PATIENTS--555 people with congenital clotting disorders who were infected with HIV, initially without AIDS, and seen at follow up for 6-30 months in 1986-9. MAIN OUTCOME MEASURES--Absolute CD4 counts and incidence of AIDS. RESULTS--Outset CD4 count and age were independently related to progression to AIDS (p less than 0.0001 and p less than 0.005 respectively). Patients with CD4 counts of 0.30-0.49 x 10(9) cells/l had an age adjusted risk of AIDS within 30 months of only 9% that of patients with counts less than 0.20 x 10(9)/l. Children under 10 years old had only 16% of the CD4 adjusted risk of AIDS of people aged greater than or equal to 45 years. Analysis of 149 patients' CD4 counts at the beginning and end of two successive six month intervals showed an average decrease of 11% in each six months regardless of the outset count (greater than or equal to 0.20 x 10(9)/l). For individual patients the decrease in the second six month period was unaffected by the decrease in the first six month period. CONCLUSIONS--Antiviral treatment of asymptomatic people, particularly children, with CD4 counts greater than or equal to 0.3 x 10(9)/l is questionable if predicted on near term progression to AIDS. Because of individual CD4 count variability and the low rate of progression to AIDS near term declines in individual CD4 counts are a poor index for identifying people who will rapidly progress to AIDS.