Evaluation of four commercial immunoassays for the detection of type specific antibodies to herpes simplex virus type 1 and 2

Paired sera from 40 patients with culture proven isolates from genital lesions (16 HSV-1 and 24 HSV-2) were tested. Four commercial enzyme immunoassays were used, two tests using type specific glycoprotein G antigens (Test A from Gull/Premier - Meridian Diagnostics and Test B from MRL Diagnostics) and two other tests using whole antigen from type 1 and type 2 strains (Test C from Virion-Serion and Test D from Genzyme Virotech). All sera were analysed in one test run for each EIA and in accordance with the instructions from the manufacturers. In cases with discrepancies between serological results and the primary typing of isolates, the strains were retyped using different type-specific monoclonal antibodies and by PCR using type-specific primers and probes. A complete agreement between all tests was found in six cases (15%) only. As expected, the two tests based on gG antigens showed good specificity (100%), but they lacked some sensitivity. Test A based on affinity purified gG antigen had a sensitivity of 86% and 73% against HSV-1 and HSV-2 antibodies, respectively. Test B usin g recombinant gG antigen missed none of the HSV-1 infections, b ut had a sensitivity of 85% against HSV-2 antibodies. Test A missed 25% (4/16) of the cases with proven HSV-1 infec tion and 21% (5/24) of the cases with HSV-2 infections verified by culture. Test B missed none of the HSV-1 infections, but 17% (4/24) of the HSV-2 infections. The sensitivity of the tests using whole antigen (Test C and Test D) was good (100%), but they were not able to discriminate betwe en type 1 and 2 infections.