Introduction Syphilis is a sexually transmitted disease caused by Treponema pallidum, which produces 5.6 million new cases worldwide in 2012 and results in significant morbidity and mortality. Despite the availability of diagnostic tests and affordable treatment, the disease remains a global health problem. Detection of non treponemal and treponemal antibodies is the most reliable method for laboratory diagnosis of syphilis. Recently a chemiluminescence microparticle immunoassay (CMIA) has been introduced for the detection of treponemal antibodies and evaluation of its performance against reference methods is needed.

Methods Sera samples were tested with three syphilis serology tests: Venereal Disease Research Laboratory (VDRL) for non treponemal antibodies, fluorescent treponemal antibody absorption (FTA-ABS) test and CMIA. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and agreement (Cohen’s kappa coefficient) of CMIA was calculated against FTA-ABS (reference group) for the diagnosis of syphilis.

Results Sera samples from 80 patients with suspected syphilis were included in this study. According to serology results: 29 syphilis non reactive sera (VDRL, FTA-ABS and CMIA non reactive), 8 VDRL reactive but FTA-ABS and CMIA non reactive sera and 43 syphilis reactive sera including 10 samples with non reactive VDRL test but FTA-ABS and CMIA reactive (past or treated syphilis) and 33 samples with all reactive tests (ongoing syphilis). General agreement between FTA-ABS and CMIA was 91.2% with kappa coefficient 0.82. CMIA clinical sensitivity was 97.7%, clinical specificity was 83.8%, PPV was 87.5% and NPV 96.9%.

Conclusion CMIA has the advantages of automation and avoids the subjectivity of FTA-ABS test. According to our results, CMIA has a good agreement with FTA-ABS but has low specificity. Limitations of this test for the diagnosis of syphilis should be taken into account since confirmatory test like FTA-ABS must still be done along with VDRL test and viewed in the context of clinical presentation