Use of Newer Assays for Syphilis

By Carol A. Kemper, MD, FACP

Clinical Associate Professor of Medicine, Stanford University, Division of Infectious Diseases, Santa Clara Valley Medical Center

Dr. Kemper does research for Abbott Laboratories and Merck. This article originally appeared in the December 2011 issue of Infectious Disease Alert. At that time it was peer reviewed by Timothy Jenkins, MD, Assistant Professor of Medicine, University of Colorado, Denver Health Medical Center. Dr. Jenkins reports no financial relationship to this field of study.

Source: Park IU, et al. Screening for syphilis with the treponemal immunoassay: Analysis of discordant serology results and implications for clinical management. J Infect Dis 2011;204:1297-1304.

Clinical laboratories in north america are increasingly switching to the newer treponemal-specific assays, such as the enzyme immunoassay (EIA) or chemiluminescence immunoassay (CIA) for syphilis testing. Because these assays are automated, they are cost-effective for clinical laboratories compared with the traditional RPR/VDRL testing. But whether they are cost-effective for clinical purposes is uncertain. While the EIA/CIA assays are highly sensitive (95%-99%) and specific (98%-99%), they may result in a larger number of false-positive or discordant results in a low prevalence population, which arguably increases the cost for treatment and follow-up, as well as potential over-treatment. Furthermore, data are lacking on the management of patients with discordant test results (CIA-positive/RPR-negative). The CDC presently recommends the use of a second treponemal test (e.g., FTA or TP-PA) for discordant specimens. Data suggest that, in a low-prevalence population, approximately 40% of these discordant specimens will be negative by a second treponemal-specific test; treatment may not be necessary for these individuals. For those who test positive by a second-treponemal-specific assay, physical findings and clinical risk factors should be assessed, and patients with risk factors should be offered antibiotic therapy, if not previously treated.

Researchers at Kaiser Permanente in San Francisco and Oakland, CA, examined the clinical characteristics of patients with discordant serology, who would not be identified by standard screening methods (i.e., CIA-positive, RPR-negative). A total of 21,623 assays were performed between August 2007 and October 2007, 439 (2%) of which were CIA-positive. Of these, 255/439 (58%) were RPR-negative. Of these 255 CIA-positive/RPR-negative discordant specimens, 184 (72%) were TP-PA-positive and 71 (28%) were TP-PA-negative. The CIA-positive/RPR-negative/TP-PA-positive group was significantly more likely to be male, men having sex with men (MSM), and HIV-positive compared with the TP-PA-negative group (all P < 0.001). They were also more likely to be African-American, and to have received previous treatment for syphilis (57% vs. 9%). However, even after excluding patients with history of syphilis, discordant TP-PA-positive patients were still more likely to be male, MSM, HIV-positive, and African American compared to those who were TP-PA-negative.

In addition, the median CIA test quantitative index was significantly higher in the CIA-positive/RPR-negative/TP-PA-positive group compared with the TP-PA-negative group (9.8 vs. 1.6), suggesting that the CIA index may be potentially useful in treatment decisions. Two-thirds of the TP-PA-negative patients had CIA index values ≤ 2, whereas only 9% of the TP-PA-positive patients had a result ≤ 2.

Several additional interesting observations were made from these data. One individual was initially CIA-positive/RPR-negative/TP-PA-negative but subsequently TP-PA-positive on follow-up testing, suggesting that he may have been in the "window period" and seroconverted the CIA test earlier than the TP-PA. On the other hand, 6 of 31 (23%) patients with discordant results who were initially TP-PA-negative, and who were followed with repeat testing, "seroreverted" their CIA result, suggesting their initial CIA test was falsely positive. Of the 28 pregnant woman identified with discordant serology, 12 (43%) were CIA-positive/RPR-negative/FTA-positive; 5 of these had a history of syphilis. Of the remaining 16 women who were CIA-positive/RPR-negative/TP-PA-negative, none had a prior history of syphilis.

Of the 255 patients with discordant serology, 100 (39%) were HIV-positive; most (86%) of these were CIA-positive/RPR-negative/TP-PA-positive. The use of the tests and interpretation of discordant test results in HIV-positive individuals presents a difficult challenge. Current guidelines recommend a lumbar puncture (LP) for HIV+ patients with latent syphilis of uncertain duration or late latent disease (although neurosyphilis is less likely in patients with an RPR titer ≤ 1:16). Because the CIA quantitative results are often masked, it is uncertain whether all of these patients require an LP. Further study is required to examine the utility of the CIA as a quantitative index.