5-YEAR PROSPECTIVE-STUDY OF HIV-INFECTION IN THE EDINBURGH HEMOPHILIAC COHORT

Objective - To identify measures of immune state that reflect the course of HIV related disease in order to predict deterioration of symptoms and assess response to treatment. Design - Five year longitudinal clinical and laboratory study. Setting - Regional haemophilia centre, university virology laboratory, and Medical Research Council laboratory. Patients - 32 Patients with haemophilia A exposed to a single batch of HIV contaminated factor VIII concentrate from the Scottish National Blood Transfusion Service in 1984 who were followed up regularly in Edinburgh (31) or abroad (one). Main outcome measures - Counts of circulating T cell subsets (CD4 and CD8); plasma β2 microglobulin, neopterin, and IgA concentrations; and delayed type hypersensitivity to multiple skin test antigens. Results - 18 Patients who seroconverted after exposure had received significantly more contaminated factor VIII than the 14 who did not (mean 43 (range 9-109) v 15 (3-30) phials, p < 0.01). The two groups were not distinguishable by other criteria before exposure. The group that seroconverted subsequently showed a progressive fall in mean circulating CD4 lymphocytes and an increase in plasma β2 microglobulin and neopterin concentrations. From 1987 patients in this group also showed an increase in mean circulating CD8 lymphocytes and in plasma IgA concentration, neither of which was seen in patients who did not seroconvert. Patients with HIV antibody who developed Centers for Disease Control category IV symptoms within five years after infection showed more extreme changes in all measures, except CD8 lymphocyte count, than those whose symptoms remained in categories II and III. Skin test reactivity declined to barely detectable levels in all patients positive for HIV antibody. Conclusions - Serial estimates of circulating CD4 lymphocytes and of plasma β2 microglobulin concentration are the most reliable measures of disease progression; of these, β2 microglobulin concentration seems to be the better predictor of impending serious symptoms. High IgA concentrations reflect rather than predict disease state. Individual variation in most measures is such that a wide range of measurements should be used in assessing the effects of trial treatment in HIV infected patients without symptoms.