THE EFFECT OF HIV DISEASE ON SERUM MARKERS OF HEPATITIS-DELTA INFECTION IN INTRAVENOUS DRUG-ABUSERS

The prevalence of serum markers of delta hepatitis was determined pros pectively in 82 intravenous drug abusers at various stages of human im munodeficiency virus (HIV) disease. Seventeen were HIV negative, 30 we re HIV positive without acquired immunodeficiency syndrome (AIDS) and 35 had been diagnosed as having AIDS, Antihepatitis D virus (HDV) in s erum was measured by a commercially available enzyme-linked immunosorb ent assay (ELISA) and also by solid phase capture radioimmunoassays (R IAs) for immunoglobulin G (IgG) and immunoglobulin M (IgM) anti-HDV. H DV antigen and hepatitis 8 virus (HBV) DNA were also measured. Hepatit is 8 surface antigen (HBsAg) and anti-HBs were determined by using a c ommercially available RIA. Anti-HDV (RIA) was only detected in serum t hat contained HBsAg. These anti-HDV (RIA) positive samples also tested positive with the commercial anti-HDV electroimmunoassay. In addition , the commercial anti-HDV ELISA detected anti-HDV in some serum sample s that were negative for HBsAg; these anti HDV-positive HBsAg-negative samples were frequently lipemic or contained rheumatoid factor. The p revalence of HBsAg and anti-HBs did not differ significantly with the stage of HIV disease. HBsAg was detected in 3 of 13 (23%) HIV-negative , 5 of 29 (17%) HIV-positive, and 4 of 18 (22%) patients with AIDS. Ig G and IgM anti-HDV (RIA) was positive in 2 of 3 HIV-negative and 4 of 5 HIV-positive pre-AIDS HBsAg-positive subjects. However, none of 4 AI DS patients had anti-HDV. The difference between AIDS and non-AIDS pat ients was statistically significant (Fisher's exact test, p = 0.03). H DV antigen was detected in serum from one AIDS patient. Our data sugge st that a commonly used commercial ELISA for anti-HDV is frequently as sociated with falsely positive results in lipemic serum or serum conta ining rheumatoid factor and that AIDS in subjects who are intravenous drug abusers is associated with significant reduction in the incidence of IgG and IgM anti-HDV in the serum. The low prevalence of anti-HDV in AIDS probably reflects defective anti-HDV production rather than re duced HDV replication and expression.