Evaluation of passive particle agglutination test for antibody to human immunodeficiency virus.

A gelatin particle agglutination assay was compared with indirect immunofluorescence by using 663 serum samples from acquired immunodeficiency syndrome and acquired immunodeficiency syndrome-related complex patients and asymptomatic male homosexuals in the United States and from hemophiliacs and healthy adult controls in Japan. The results showed that all 104 samples which were positive by indirect immunofluorescence were also positive by particle agglutination, while 5 additional samples were positive by particle agglutination only. The coincidence rate for antibody-positive and antibody-negative specimens was 99% (658 of 663) between particle agglutination and immunofluorescence. Four of the five samples which were positive by particle agglutination only were found by radioimmunoprecipitation to contain anti-env gene products of human immunodeficiency virus. Antibody titers of samples giving a positive reaction by particular agglutination varied from low (titer, 256) to remarkably high (256 X 10(5)). All specimens having particle agglutination titers of more than 10(5) were positive by immunofluorescence. A high correlation (r = 0.66) was observed between the titers of antibodies determined by particle agglutination and those determined by immunofluorescence. After fractionation of a serum sample from an individual at high risk by using high-performance liquid chromatography, it was shown that immunoglobulin M as well as immunoglobulin G human immunodeficiency virus antibody was detected by particle agglutination. Additional serum samples with a potential risk of giving false-positive results, such as heat-treated specimens, specimens containing antibodies to HLA, specimens containing auto-antibodies, and serum samples from individuals with a history of multiple blood transfusions, were shown to be clearly negative by particle agglutination.