British Columbia's Experience after Implementation of the Treponema pallidum Reverse Algorithm and PCR Detection, 2015 to 2020

Abstract

British Columbia (BC) implemented the syphilis reverse screening algorithm and Treponema pallidum PCR testing in 2014. We summarize the performance characteristics of the algorithm, together with PCR direct detection, and report on syphilis cases identified from 2015 to 2020. Prior to 2015, samples for syphilis diagnosis were first screened by rapid plasma reagin (RPR). As of 2015, sera were screened by the Siemens Advia Centaur syphilis assay (enzyme immunoassay [EIA]). Positive and equivocal samples were reflex tested by a T. pallidum passive particle agglutination assay (TPPA) and RPR. We used T. pallidum DNA PCR on clinical samples and restriction fragment length polymorphism analysis to identify azithromycin resistance mutations. Case/epidemiological data were obtained from the BC surveillance system. Of 1,631,519 samples screened by the EIA, 72,492 (4.4%) were positive and 187 (<0.1%) were equivocal. Of EIA-positive/equivocal samples, 10.6% were false positive, and false positivity was higher at lower EIA indices. The reverse algorithm detected 4,693 late latent syphilis cases that likely would have been missed by RPR screening. PCR had a very high sensitivity of 100% versus 52.9% and 52.4% for dark-field (DF) and immunofluorescence (IF) microscopy, respectively. The azithromycin resistance mutation A2058G was identified in 96% of PCR-positive samples, and A2059G was identified in 4%. Annually, there were 944 to 1,467 syphilis cases, with 62% in men who reported male sexual partners. The reverse algorithm had a low false-positive rate and very few equivocal screening results but did identify previously undiagnosed late latent syphilis cases. PCR was more sensitive than both DF and IF microscopy for direct diagnosis and enabled monitoring for azithromycin resistance. IMPORTANCE In this study, we summarize the performance characteristics of the algorithm, together with PCR direct detection and epidemiological analysis, and report on syphilis cases identified from 2015 to 2020. This allowed us to paint a complete picture of the outcome of the utilization of the reverse algorithm for diagnosing syphilis cases. The study clearly showed that the reverse algorithm had a low false-positive rate and very few equivocal screening results but did identify previously undiagnosed late latent syphilis cases. PCR was more sensitive than both DF and IF microscopy for direct diagnosis and enabled monitoring for azithromycin resistance.

Keywords: PCR; reverse algorithm; screening; syphilis.

FIG 1

Syphilis serology reverse screening algorithm, BCCDC Public Health Laboratory, 2015 to 2020. *, these latent infections would likely have been undetected and not treated when using rapid plasma reagin (RPR) as a screen test; **, the RPR titer is ≥4-fold higher than that of a previous specimen. EIA, enzyme immunoassay screen test; TPPA, treponemal passive particle agglutination assay.

FIG 2

TPPA and RPR results stratified by syphilis EIA signal index, 2015 to 2020. Note that false-positive result classification is based on interpretation for an individual sample; some of these may represent early infection or treated cases from the remote past.