RESUMO
BACKGROUND: The aim of the study was to establish a new syphilis test algorithm using Architect Syphilis TP (Abbott Japan, Japan: AST), a fully automated treponemal antibody test, as a screening test in a university hospital laboratory. We evaluated performance characteristics of AST in various patient groups. METHODS: A total of 1,357 serum samples obtained from patients at a university hospital from June to August, 2008 were categorized into checkup, preoperative, other diseases, diagnosis (clinically suspected of syphilis), and follow up groups. We compared the results of AST with those of RPR (N=1,276) or Treponema pallidum hemagglutination assay (TPHA, N=81). Samples with discrepant results between RPR or TPHA and AST were retested by fluorescent treponemal antibody absorption test (FTA-ABS) and all patients' clinical records were thoroughly reviewed. RESULTS: The positive rate of AST was significantly higher than that of RPR in preoperative and other diseases groups and was the same as that of RPR in diagnosis group. There were no significant differences in check up and follow up groups. The results of AST showed 97.4% (1,243/1,276) and 97.5% (79/81) concordance rates with those of RPR and TPHA, respectively. Among 26 RPR-AST discrepant and FTA-ABS confirmed cases, there were 20 RPR false-negatives, 4 RPR false-positives, 1 AST false-negative, and 1 AST false-positive. CONCLUSIONS: Based on the results and literature review, we established a new syphilis test algorithm using AST as a screening test, which would be helpful for detection of more syphilis patients including latent infections.
