Literature review > Issue_1 > Review Rodriguez et al. 

Because of its protean clinical manifestations and the lengthy periods of asymptomatic infection, serological testing is widely relied upon for diagnosis of syphilis. Traditionally, serological diagnosis is based upon the results of a combination of nontreponemal and treponemal tests. Because the nonspecificity of the inexpensive nontreponemal screening tests can lead to biological false positive reactions, the treponemal tests have been used in many countries as confirmatory tests. The treponemal tests have evolved from the cumbersome and insensitive T. pallidum immobilization test, to immunofluorescence, agglutination, and ELISA tests that utilize whole intact or sonicated rabbit-propagated T. pallidum organisms, and now to enzyme immunoassays using individual or combinations of recombinant T. pallidum antigens. With each step of this evolution, there has been an attempt to decrease subjectivity in the readout, increase sensitivity and specificity, and maximize opportunities for automation.

This article describes the evaluation of the Bioscreen anti-T. pallidum enzyme immunoassay (EIA), manufactured in Cuba, with sera from 78 patients with various stages of syphilis, 125 persons with other conditions (some of which may lead to false positive results in the nontreponemal tests), and with 164 blood donors. The Bioscreen is an enzyme immunoassay that measures reactivity against TmpA, a T. pallidum lipoprotein antigen originally described by Van Embden's group in 1985 [1]. Others have evaluated TmpA as an antigen for serodiagnosis of syphilis and have found that it is promising and that it is generally equivalent to the TPHA in sensitivity and specificity. In a study of 938 sera, Ijsselmuiden et al. [2] demonstrated 99.6% specificity, and sensitivity figures that ranged from 76% for primary syphilis to 100% and 98% for secondary and early latent syphilis, respectively. Antoni et al. [3] demonstrated that the N-terminal 19 amino acids of TmpA are the immunodominant epitope on this molecule and might provide a very specific antigen for a diagnostic test. One interesting observation concerning TmpA is that antibody titers to this antigen decline rapidly following treatment for syphilis [2, 4], thus perhaps making possible the differentiation of past versus current syphilis infection.

While the current study by Rodriguez et al. confirms that the Bioscreen EIA has a sensitivity and specificity equivalent to the T. pallidum hemagglutination assay (TPHA), there are some concerns about extrapolation of the results to an actual clinical setting. Obviously, the results obtained are wholly dependent upon the sera that are selected for testing. Seventy-eight sera from persons with syphilis were tested, including 28 who were categorized as having the early latent stage. No criteria for diagnosing syphilis or determining stage were provided; this is of particular concern for those stated to have early latent syphilis. A strict definition of early latent syphilis requires clear evidence of primary or secondary manifestations, or negative syphilis serology, within one year prior to diagnosis. Further, no patients with late latent syphilis were included in the study, and this group often comprises a majority of syphilis cases that are diagnosed in a clinical setting. There is no indication whether the subjects with syphilis had been previously treated with antibiotics, which may alter reactivity in some tests. In the group of sera from subjects with other conditions or from blood donors, no criteria were provided to determine whether any of these subjects may have also had past or current syphilis. This is critical if accurate specificity values are to be obtained.

The limited numbers of primary syphilis sera tested in this study also impair the reader's ability to evaluate the test. The authors accurately point out that the sensitivity of serological tests varies during early primary syphilis, yet 37 of the 38 primary syphilis sera tested in this study were reactive in the VDRL or RPR test, a clear indication that these sera were not obtained from patients with early primary syphilis. If more RPR-negative early primary syphilis patients had been evaluated in this study, the overall sensitivity of the EIA would certainly have been lower than the reported 93.3% (my calculation gives a sensitivity of 71/78 or 91%).

This study demonstrates that the Bioscreen test performs as one would expect, based upon historical reports examining TmpA antigen. However, the search should certainly continue for new and better antigens for syphilis diagnosis, particularly for antigens that are recognized early during primary infection. The publication of the T. pallidum genome will make the identification of new recombinant T. pallidum antigens much easier, and it is anticipated that a new generation of treponemal tests will result.

about SDI | newsletters | grants | publications | literature reviews

WHO Home - WHO Search - TDR Home - SDI Home - SDI Contact us
(c) WHO/OMS 2001