In a study to determine the reliability, sensitivity, and specificity
of the Chiron RIBA HIV-1/HIV-2 Strip Immunoblot Assay (RIBA HIV-1/2 SU
) for confirmation of human immunodeficiency virus type 1 (HIV-1) and
HIV-2. antibodies, 1,263 serum samples from various populations in the
United States, Caribbean, Africa, India, and Thailand were evaluated
by RIBA HIV-1/2 SIA, and the results were compared with those obtained
by an HIV-1 Western blot (immunoblot) assay. All sera were tested by
HIV enzyme immunoassay, RIBA HIV-1/2 SIA, and Western blotting, Sample
s with discrepant results were further tested by an HIV-1 and/or HIV-2
immunofluorescent-antibody assay and HIV-1 p24 antigen assay, The RIB
A HIV-1/2 SIA detected all 17 HIV-1 and HIV-2 dually reactive serum sa
mples, all 215 HIV-2-positive serum samples, and 480 of 481 HIV-1-posi
tive serum samples for a sensitivity of 99.8%. Of 548 negative samples
, 523 were RIBA HIV-1/2 SIA negative, for a specificity of 95.4%, with
22 (4%) samples interpreted as indeterminate and 3 (0.6%) interpreted
as falsely positive, Western blotting detected 391 of 548 negative sa
mples (specificity, 71.4%), with 152 (27.7%) samples interpreted as in
determinate and 5 (0.9%) interpreted as falsely positive. In conclusio
n, the RIBA HIV-1/2 SIA had a sensitivity comparable to that of Wester
n blotting and could discriminate HIV-1 from HIV-2 in one blot, provid
ing a cost advantage, Because of its high degree of specificity, the R
IBA HIV-1/2 SIA further reduced the number of indeterminate results fo
und by Western blotting, providing a more accurate means of assessing
seronegative individuals.