In September 2007, the province of Alberta, Canada, introduced reverse sequence syphilis screening (RSSS) using an enzyme immunoassay (EIA) (Architect Syphilis TP Chemiluminescent Microparticle Immunoassay, Abbott Laboratories, Abbott Park, IL) as the initial serologic screening test for all laboratories province-wide. The change was made to improve syphilis diagnostics due to a resurgence of infectious syphilis in the province that began in 2003 (1.4/100,000) and has continued until (8 years) the writing of this report (7.4/100,000 in 2009).1 Before this change, the traditional algorithm using a qualitative rapid plasma reagin (RPR) test (Macro-Vue RPR kit, Becton Dickinson Microbiology Systems, Ontario, Canada) was the initial screening test, except for high-risk patients with appropriate clinical history for whom RPR and Treponema pallidum passive particle agglutination (TP-PA) were specifically requested by health care providers. Quantitative RPR tests were reflexively performed on all reactive samples, which were confirmed by treponemal-specific tests, TP-PA, and fluorescent treponemal antibody absorption.
Because RPR titers can be slow to rise in early acute syphilis and can decrease over time and become nonreactive in untreated late syphilis, syphilis EIA was introduced to increase diagnostic sensitivity and specificity and to facilitate high throughput screening on an automated platform.2 In support of the implementation of RSSS in Alberta, Chuck et al demonstrated that such a change would be cost-effective with an incremental cost-effectiveness ratio of $461 (Canadian) per correct diagnosis.3 Under RSSS, all samples tested as indeterminate/positive by syphilis EIA are reflexively tested for quantitative RPR titers, and all samples from patients who are indeterminate/positive by EIA with no laboratory record of previously positive treponemal syphilis serology are reflexively tested with a confirmatory treponemal test, a line immunoassay (INNO-LIA Syphilis, Innogenetics NV, Ghent, Belgium).4,5
An analysis of routinely collected surveillance data from the Edmonton city region of Alberta (population 1.1 million, 2007) from 2004 to 2009 showed a significant rise in the diagnosis of late latent syphilis (LLS) cases but not in primary syphilis after the introduction of RSSS. With the reversal of the algorithm, the potential to identify additional cases of syphilis with discordant treponemal and nontreponemal tests (EIA+/RPR−) increases.6 Using laboratory-derived specimen data, others have reported an increase in positivity rates after moving to EIA screening from RPR,7,8 but those studies were limited by the lack of clinical information to determine how the rise in positive specimens translated into newly diagnosed cases. We sought to determine the impact of discordant results on new diagnoses of LLS after the introduction of RSSS in Edmonton, Canada.
Under Alberta's Public Health Act, a syphilis case investigation is triggered by the provincial sexually transmitted infections (STIs) program when results are received from the laboratory for all positive/reactive samples for syphilis (serology, fluorescent antibody test, dark-field microscopy, and/or polymerase chain reaction for Treponema pallidum). A partner notification nurse contacts both the health provider and the client to obtain the client's clinical history, including syphilis exposure, symptoms, and any previous treatment. Syphilis staging, treatment, and follow-up recommendations are provided by 1 of 4 STI Medical Consultants in the province once confirmatory serology is available. LLS is defined as a patient who has been asymptomatic during the previous 12 months with persistently reactive treponemal serology (regardless of nontreponemal serology reactivity), who does not report exposure to a sexual partner with infectious syphilis, and who has not been previously treated for syphilis.9 For newly diagnosed LLS cases in Alberta, the partner notification nurse arranges testing and treatment if needed for long-term sexual partners and children. Information on all newly diagnosed cases is entered into the electronic provincial Communicable Disease Reporting System.
Aggregated laboratory data were provided to assess testing volumes. Comparisons of demographics and risk behaviors of RPR reactive (RPR+) and RPR nonreactive (RPR−) newly diagnosed LLS cases were conducted using χ2 or Fisher exact test for proportions and Mann-Whitney U test for continuous variables. Data were analyzed using PAWS Statistics version 18.0 (SPSS Inc, Chicago, IL). This study was approved by the University of Alberta Health Research Ethics Board.
The number and proportion of primary and LLS syphilis cases by testing volume in the Edmonton region from January 2004 to April 2009 are presented in Figure 1. The figure illustrates the rise in LLS cases and the plateau of primary syphilis after the introduction of RSSS. Before RSSS, 97 cases (0.07%) of 147,460 specimens screened using RPR between January 2004 and August 2007 resulted in an LLS diagnosis; following the introduction of RSSS, 137 cases (0.14%) of 96,509 specimens screened using EIA between September 2007 and April 2009 resulted in an LLS diagnosis (P < 0.001). All annual LLS proportions were lower before the implementation of RSSS.
After the introduction of the RSSS, 3 cases of discordant primary syphilis (RPR−) were identified that would have been missed using RPR as the initial screening test. We also observed the proportion of discordant LLS cases increased significantly from 35.8% (n = 34) to 58.3% (n = 81, P < 0.001). Demographic and clinical information on all newly diagnosed LLS cases from Edmonton after the implementation of RSSS (between September 2007 and April 2009) was extracted from Communicable Disease Reporting System (Table 1). There were no significant differences between RPR− and RPR+ LLS cases in terms of gender, median age, percentage of pregnant women, and percent HIV seropositivity. The majority of the LLS cases were among individuals born outside of Canada, with a median of 5 years since immigration to Canada. Significant differences were found in the reason for syphilis testing between the 2 groups: fewer RPR− LLS cases reported immigration as the reason for testing, whereas more RPR− LLS cases reported STI screening as the reason.
In the period following the implementation of RSSS in Alberta, the proportion of LLS cases diagnosed doubled in a metropolitan area of the province, a greater increase than would be expected by the increase in testing volume. Reasons for the increase in testing volumes include annual increases in prenatal specimens, ranging from approximately 500 to 1500 samples per year during the study period, and possible increased public awareness of syphilis in response to the syphilis outbreak. However, the increase in LLS cases was likely due to the increased sensitivity of EIA as compared with RPR as the initial screening test2 and may reflect a better estimation of prevalent syphilis.
Most of the 81 RPR− LLS cases (58.3%) diagnosed during the post-RSSS period would not have been identified using the traditional algorithm of an RPR screen unless they were identified as high-risk patients by their health care provider and a TP-PA was performed as part of the screen. Investigations of discordant results required public health resources to determine whether the results reflected acute cases of infectious syphilis or LLS. The RPR+ LLS cases were more likely to report immigration as their reason for testing; all new immigration applicants to Canada undergo mandatory screening and follow-up for syphilis. The majority of RPR− LLS cases cited STI screening as the reason for testing, which may be related to an increased awareness of the syphilis outbreak.
The majority of LLS cases were diagnosed among foreign-born individuals. It is possible that some cases of nonvenereal trepanematoses were misclassified as LLS due to the inability of syphilis diagnostic tests to distinguish between syphilis and nonvenereal trepanematoses.10 Moreover, a previous study found that patients with periodontal disease carry oral treponemes that can be detected with syphilis serological assays.11 As well, some cases may have been misclassified as LLS due to the patient's inability to recall previous treatment or due to unavailable documentation of treatment in other countries. These limitations would be the same for both the traditional testing algorithm and RSSS thus unlikely to have any effect on the observed rise in LLS cases.
A Centers for Disease Control and Prevention evaluation of treponemal screening tests among 4 New York City laboratories showed that 56% of persons with positive treponemal EIA had negative nontreponemal test results and recommended the use of a second treponemal test, TP-PA or fluorescent antibody test, which found 83% of these patients tested positive by 2 assays.12 After in-house validation of the INNO-LIA, an immunoblot assay, the ProvLab used this test as the second treponemal test. Although some experts have raised concerns about the specificity of the treponemal antigen-based assays,13 false-negative TP-PA with positive fluorescent treponemal antibody absorption and INNO-LIA results have been reported.14 Centers for Disease Control and Prevention encourages clinicians to consider treatment of LLS for cases where the EIA is positive, RPR−, and the second treponemal test is reactive, unless the history or results of a physical examination suggests a recent infection.12 Treatment may reduce the chance of progression to tertiary complications.12 The risk in untreated patients has been estimated to be between 15% and 40%,15 although some experts believe this is an overestimation in the antibiotic era. Mother-to-child transmission of syphilis during pregnancy depends on a number of factors: syphilis stage, reactivity of nontreponemal tests, and gestational age at treatment. In one widely cited study, the estimated risk of mother-to-child transmission from untreated late latent cases was reported as 10%.16 Identifying and treating unrecognized infections among women of childbearing age (18 cases of the 81) has the potential to reduce vertical transmission in treated LLS cases.
Our study found an increase in newly diagnosed LLS cases after the introduction of RSSS. Public health investigations of discordant laboratory results propagated the detection and diagnosis of more LLS cases but did not have the same impact on identifying early acute cases. The impact of RSSS on diagnosing significantly more LLS cases aligns with previous reports of increased positivity among laboratory specimens. The increase in the percentage of RPR− cases with RSSS has a direct impact on public health programs that are required to investigate and provide appropriate treatment and follow-up of all newly diagnosed cases. It is important that sufficient resources are allocated for STI program delivery where such changes in diagnostic testing have been implemented or are being considered.
1. Government of Alberta. Notifiable sexually transmitted infections. 2009 annual report. Edmonton, Alberta: Alberta Health and Wellness; 2010.
2. Seña AC, White BL, Sparling PF. Novel Treponema pallidum
serologic tests: A paradigm shift in syphilis screening for the 21st century. Clin Infect Dis 2010; 51:700–708.
3. Chuck A, Ohinmaa A, Tilley P, et al.. Cost effectiveness of enzyme immunoassay and immunoblot for the diagnosis of syphilis. Int J STD AIDS 2008; 19:393–399.
4. Hagedorn HJ, Kraminer-Hagedorn A, De Bosschere K, et al.. Evaluation of INNO-LIA syphilis assay as a confirmatory test for syphilis. J Clin Microbiol 2002; 40:973–978.
5. Lam TK, Lau HY, Lee YP, et al.. Comparative evaluation of the INNO-LIA syphilis score and the MarDx Treponema pallidum
immunoglobulin G Marblot test assays for the serological diagnosis of syphilis. Int J STD AIDS 2010; 21:110–113.
6. Park IU, Chow JM, Bolan G, 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.
7. Mishra S, Boily MC, Ng V, et al.. The laboratory impact of changing syphilis screening from the rapid-plasma reagin to a treponemal enzyme immunoassay: A case-study from the Greater Toronto Area. Sex Transm Dis 2010; 37:1–7.
8. Owusu-Edusei K Jr, Peterman TA, Ballard RC. Serologic testing for syphilis in the United States: A cost-effectiveness analysis of two screening algorithms. Sex Transm Dis 2011; 38:1–7.
9. Disease Control and Prevention Branch. Alberta Case Definitions Manual. Edmonton, Alberta: Alberta Health and Wellness, 2008.
10. Larsen SA, Steiner BM, Rudolph AH. Laboratory diagnosis and interpretation of tests for syphilis. Clin Microbiol Rev 1995; 8:1–21.
11. Riviere GR, Wagoner MA, Baker-Zander SA, et al.. Identification of spirochetes related to Treponema pallidum
in necrotizing ulcerative gingivitis and chronic periodontitis. N Engl J Med 1991; 325:539–543.
12. CDC. Syphilis testing algorithms using treponemal tests for initial screening—Four laboratories, New York City, 2005–2006. Morb Mortal Wkly Rep 2008; 57:872–875.
13. Ho EL, Lukehart SA. Syphilis: Using modern approaches to understand an old disease. J Clin Invest 2011; 121:4584–4592.
14. Maple PA, Ratcliffe D, Smit E. Characterization of Treponema pallidum
particle agglutination assay-negative sera following screening by treponemal total antibody enzyme immunoassays. Clin Vaccine Immunol 2010; 17:1718–1722.
15. Singh AE, Romanowski B. Syphilis—a review with emphasis on clinical, epidemiologic and some biologic features. Clin Micro Rev 1999; 12:187–209.
16. Fiumara NJ. Syphilis in newborn children. Clin Obstet Gynecol 1975; 18:183–189.