Evaluation of an automated quantitative latex immunoturbidimetric non-treponemal assay for diagnosis and follow-up of syphilis: a prospective cohort study.

Purpose. We evaluated the Sekure rapid plasma reagin (RPR-S) (Sekisui Diagnostics) automated quantitative latex immunoturbidimetric assay performed on the SK500 clinical chemistry system for clinical appropriateness. Methodology. Syphilis-infected individuals and controls were recruited into a prospective cohort study conducted at a sexually transmitted infection clinic in Antwerp, Belgium. Sera collected at diagnosis (baseline) and at 3, 6, 9 and 12 months post-treatment were tested with RPR-S and Macro-Vue RPR card (RPR-C) (Becton Dickinson) assays; RPR-C was considered the reference test. IgG/IgM enzyme immunoassay and Treponema pallidum polA serum PCR results were consulted by discordancy at baseline. Categorical analyses were performed and correlations were assessed with (non)-linear regression. Post-treatment longitudinal serological evolution was evaluated. Results. A total of 463 samples from 120 new syphilis cases from a variety of clinical stages and 30 syphilis-negative controls were tested. Initially, there was a weak correlation between quantitative RPR-C/S (r=0.15). In 70 samples there was a strong suspicion of hook effect. Of these, 57/70 sera were retested with an extra dilution step, resulting in an average 12-fold increase in quantitative RPR-S results. After the extra dilution, the overall qualitative RPR-C/S agreement was 78.89 %, (kappa-coefficient: 0.484). Of the 92 discordant samples, 9 were from the baseline visit (RPR-C titre: 1-8), which could have led to possible missed diagnoses using the RPR-S. Conclusions. The sensitivity and accuracy of the RPR-S test requires improvement before it can be used to diagnose syphilis and evaluate treatment efficacy in clinical practice.