Progression of HIV disease in a haemophilic cohort followed for 11 years and the effect of treatment

Abstract

Objective - To describe the progression of HIV disease in a haemophilic cohort and to show the influence of treatment. Design - 11 year longitudinal clinical and laboratory study. Setting - A haemophilia centre. Patients - 111 patients infected with HIV during October 1979 to July 1985. Main outcome measures - Symptoms of HIV infection, AIDS, and death. Interventions - 26 asymptomatic patients started taking zidovudine or placebo (1000 mg/day) during November 1988 to February 1990; 10 patients with CD4+ counts of 0·2×109/1 started zidovudine 500 mg/day during January to November 1990. 35 patients used pentamidine for primary or secondary prophylaxis. Results - At 11 years from seroconversion the estimated rate of progression to AIDS was 42% (95% confidence interval 27% to 57%); to symptoms 85% (75% to 95%); and to death 41% (25% to 57%). Progression to AIDS was significantly faster in patients aged 25 and over than in those aged less than 25 (relative risk 5·0 (2·4 to 10·4); p<0·00001) and in those with previous cytomegalovirus infection than in those not infected (relative risk 3·0 (1·4 to 6·8); p=0·006). 16 of 27 (59%) patients with p24 antigenaeinia developed AIDS compared with 17 of 84 (20%) patients without p24 antigen (p<0·001). The risk of progression to AIDS before 30 November 1988 in patients with CD4+ counts ≤0·2×109/1 was higher than after November 1988 (relative risk 1·9 (0·85 to 4·43); p=0·1). For 1989 and 1990 the observed cumulative numbers of AIDS cases (among 81 patients with sufficient CD4+ counts) were 22 and 25 compared with 29 and 37 predicted from the rate of fall of CD4+ counts up to the end of 1988 (p=0·03). Conclusion-Treatment seems to be reducing the progression of HIV disease in this haemophilic cohort.