Performance Characteristics of the Reverse Syphilis Screening Algorithm in a Population With a Moderately High Prevalence of Syphilis

Transcription

1 Performance Characteristics of the Reverse Syphilis Screening Algorithm in a Population With a Moderately High Prevalence of Syphilis Angela R. Rourk, Frederick S. Nolte, PhD, and Christine M. Litwin, MD CME/SAM From the Department of Pathology and Laboratory Medicine, Medical University of South Carolina, Charleston. Key Words: Syphilis; TP-PA; RPR; Reverse algorithm; Treponema pallidum Am J Clin Pathol November 2016;146: DOI: /AJCP/AQW182 ABSTRACT Objectives: With the recent introduction of automated treponemal tests, a new reverse syphilis algorithm has been proposed and now used by many clinical laboratories. We analyzed the impact of instituting the reverse screening syphilis algorithm in a laboratory that serves a geographic area with a moderately high prevalence of syphilis infection. Methods: Serum samples sent for syphilis testing were tested using a treponemal enzyme immunoassay (EIA) as the screening assay. EIA reactive samples were tested by rapid plasma reagin (RPR) and titered to end point if reactive. RPR nonreactive samples were analyzed by the Treponema pallidum particle agglutination test (TP-PA). Pertinent medical records were reviewed for false-reactive screens and samples with evidence of past syphilis infection. Results: Among 10,060 patients tested, 502 (5%) were reactive on the initial EIA screen. The RPR was reactive in 150 (1.5%). TP-PA testing determined that 103 (1.0%) were falsely reactive on initial EIA screen. The reverse screening algorithm, however, identified 242 (2.4%) with evidence of latent, secondary, or past syphilis, 21 of whom had no or unknown prior treatment with antibiotics. Conclusions: Despite a 1.0% false-reactive rate, the reverse syphilis algorithm detected 21 patients with possible latent syphilis that may have gone undetected by traditional syphilis screening. Upon completion of this activity you will be able to: describe the traditional and reverse sequence syphilis screening algorithm. interpret the results from the reverse syphilis screening algorithm. discuss the advantages and limitations of the reverse syphilis algorithm. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit TM per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Exam is located at Since the causative agent of syphilis, Treponema pallidum, cannot be cultured in vitro, the laboratory diagnosis of syphilis relies on serologic methods. 1 The traditional approach for the serologic diagnosis of syphilis is the algorithm in which serum samples are initially tested for the presence of nontreponemal antibody, such as the rapid plasma reagin (RPR). 2 If reactive in undiluted serum samples, the samples are diluted serially to determine the end-point titer. Due to the potential for false-reactive results with nontreponemal tests, reactive samples are confirmed with a treponemal assay (eg, fluorescent treponemal antibody absorption test [FTA-ABS] or T pallidum passive particle agglutination test [TP-PA]). With the recent introduction of the automated enzyme immunoassay (EIA) and the chemiluminescent immunoassay (CLIA) treponemal tests to the clinical laboratory, a new syphilis algorithm has been proposed and now used by 572 Am J Clin Pathol 2016;146: American Society for Clinical Pathology, All rights reserved. For permissions, please journals.permissions@oup.com

2 many clinical laboratories. The reverse testing algorithm uses the treponemal EIA or CLIA instead of a nontreponemal test as the initial screening assay. 3,4 If reactive, a nontreponemal test (RPR) is performed and an end-point titer is determined if reactive. In the case of a nonreactive nontreponemal test, an alternative treponemal test is recommended to determine if the screening treponemal test was falsely reactive. The Centers for Disease Control and Prevention (CDC) recommends that serum samples nonreactive by the nontreponemal test be analyzed by the TP- PA assay. 3 A nonreactive TP-PA result would indicate that the results of the rapid treponemal screen were falsely reactive, while a reactive TP-PA would support an interpretation of either past, treated syphilis or late/latent syphilis. In addition, since approximately 30% of those with early primary syphilis will have nonreactive nontreponemal test results, 2 a few reactive TP-PA results may represent cases of very early syphilis since treponemal antibodies are detected slightly before nontreponemal antibodies. The false-reactive rate of the rapid treponemal EIA/CLIA screening assays varies based on the prevalence of syphilis in the population being tested, with low-prevalence areas giving the highest false-positive rates. Despite an increased false-positive rate in low-prevalence areas, the reverse algorithm can detect late syphilis that can be missed with the traditional algorithm. 5,6 Many clinicians, however, are presented with challenging clinical decisions when serum samples from their patients are confirmed as treponemal test reactive but nontreponemal nonreactive. This study is a detailed analysis of the impact of instituting the reverse screening syphilis algorithm in a laboratory that serves a geographic area with a high prevalence of syphilis infection and how we managed the change from the traditional to the reverse screening algorithm. Materials and Methods Human Serum Samples Between July 22, 2013, and February 16, 2015, a total of 11,577 serum samples from 10,060 patients were sent to the Medical University of South Carolina (MUSC) immunology laboratory for reverse algorithm syphilis testing. Procedures were followed in accordance with ethical standards established by MUSC in accordance with the Helsinki Declaration of The protocol used was approved by the MUSC Institutional Review Board (no ) to meet the Health Information Portability and Accountability Act guidelines. Specimens were usually run within a few hours of receipt for the BioPlex (Bio-Rad, Hercules, CA) immunoglobulin G (IgG) testing and then stored at 2 Cto8 Cforapproximately 24 hours prior to RPR and TP-PA testing. Following patient sample testing, relevant clinical information, including reason for testing, final clinical interpretation of results, and antibiotic treatment, was obtained from the electronic medical record on patient samples that were BioPlex IgG reactive and RPR nonreactive, including TP-PA reactive, TP-PA equivocal, and TP-PA nonreactive patients. Treponemal Screening Test All samples submitted for reverse syphilis algorithm screening were tested using the BioPlex 2200 syphilis IgG multiplex flow immunoassay (Bio-Rad Laboratories, Hercules, CA) according to the manufacturer s instructions. The kit uses three different sets of beads coated with recombinant proteins from Treponema pallidum (15kDa, 17kDa, and 47 kda) as antigen. In brief, samples are added to reaction wells containing bead reagent and sample diluent. After two incubations and their corresponding washing steps, the beads are read by a flow cytometry based detector. Each analyte is quantified and compared with a preestablished calibration curve. Data are calculated as relative fluorescence intensities and converted to a fluorescence ratio using an internal standard bead. The fluorescence ratio is compared with a calibration curve to determine the analyte concentration in antibody index (AI) units. Samples less than 0.9 AI are classified as nonreactive, 0.9 to 1.1 as equivocal, or greater than 1.1 as reactive. Nontreponemal Tests All samples testing reactive by the BioPlex assay were tested by RPR with the 18-mm circle quantitative card test according to the manufacturer s protocol (Cardinal Health, Dublin, OH). Any samples reactive by RPR were titered to end point. Samples were reported as nonreactive or with the highest titer dilution that produced a reactive result. A few select cerebrospinal fluid (CSF) samples were tested by the VDRL assay (Becton Dickinson, Sparks, MD) for confirmation of the diagnosis of neurosyphilis. The VDRL assay was run according to the manufacturer s protocol. Samples were reported as nonreactive or with the highest titer that produced a reactive result. Treponemal Confirmatory Tests All samples that were BioPlex reactive but RPR nonreactive were further tested by Serodia TP-PA (Fujiribio Diagnostics, Seguin, TX), a gelatin particle agglutination assay. This assay was performed according to the manufacturer s protocol. The results were read as nonreactive, equivocal, or reactive. Medical Staff Communication and Ordering Practices In July 2013, a notice was sent to all clinicians announcing the implementation of the reverse screening algorithm American Society for Clinical Pathology Am J Clin Pathol 2016;146:

3 Rourk et al /REVERSE SYPHILIS SCREENING ALGORITHM that included the rationale for the change and how to order and interpret the results of the tests. The electronic medical record in use was EPIC (Verona, WI). Upon implementation, clinicians continued to order the RPR as a syphilis screening test. These orders were cancelled with the following comment: Routine syphilis screening should now be ordered as SYPH IgG REF (LAB894). Please add this test to your preference list. Order RPR only if patient was previously reactive. This request was converted to SYPH IgG REF for your convenience. From July 9, 2014, to November 1, 2014, we ed clinicians who continued to order RPR as initial screening tests, reinforcing the proper sequence of tests and asking why they continued to order RPR for screening. As a result of this survey, we decided to implement an EPIC Alternative Selection tool that was triggered when a clinician ordered either the RPR or BioPlex IgG test on January 29, 2015, that communicated the information above. Statistical Analysis Statistical calculations were performed using an Excel spreadsheet (Microsoft, Redmond, WA). Results Reactive (n = 502 [5%]) Reactive; syphilis likely (n = 150 [29.9%]) Reactive; late, treated, or resolved syphilis (n = 242 [68.7%]) RPR Syphilis IgG Nonreactive; syphilis unlikely (n = 9,558 [95%]) Nonreactive (n = 352 [70.1%]) TP-PA Equivocal; retest if clinically indicated (n = 7 [2.0%]) Nonreactive; false positive syphilis IgG; syphilis unlikely (n = 103 [29.3%]) Figure 1 Reverse syphilis algorithm results. The syphilis immunoglobulin G (IgG) was used for the initial screening test, the rapid plasma reagin (RPR) as the nontreponemal test, and the Treponema pallidum particle agglutination (TP-PA) for confirmation of syphilis IgG reactive/rpr nonreactive patients. Reverse Algorithm Screening Results Figure 1 illustrates the reverse algorithm study design along with the test results and clinical interpretations. Percent RPR reactive RPR nonreactive, TP-PA nonreactive RPR nonreactive, TP-PA reactive >8.0 BioPlex Antibody Index Figure 2 BioPlex syphilis immunoglobulin G (IgG) percent distribution versus rapid plasma reagin (RPR) and Treponema pallidum particle agglutination (TP-PA) result. Percent distribution of syphilis IgG reactive patients with the following results: RPR reactive, RPR nonreactive/tp-pa nonreactive, and RPR nonreactive/tp-pa reactive vs BioPlex IgG antibody index from 0.9 to more than 8.0. Among 10,060 patients tested, 502 (5%) patients were reactive on initial screen. Of the 502 patients with BioPlex IgG screening reactive results, the RPR was also reactive in 150 (29.9%) and nonreactive in 352 (70.1%) serum samples for an overall positivity rate for active syphilis of 1.5%. Among the 352 BioPlex reactive/rpr nonreactive serum samples, TP-PA was reactive in 242 (68.7%), equivocal in seven (2.0%), and nonreactive in 103 (29.3%). The overall falsereactive rate for the BioPlex IgG screen was 1.0%. Six patients who tested RPR reactive received additional testing of their CSF by VDRL. All CSF samples were reactive, and the patients were confirmed to have a diagnosis of neurosyphilis. Prior to the institution of the reverse algorithm, the reactivity rate of the screening RPR was 2.0% with a falsereactive rate of 22.0%. BioPlex Syphilis IgG AI Value Distributions Percent distribution of BioPlex IgG AI in increments of 1.0 to more than 8 AI for patients with active syphilis, probable past infections, and false-reactive results is presented in Figure 2. Patients with acute syphilis (RPR reactive) had higher BioPlex AI values at initial screen (mean [SD], 7.8 [0.95]). A small peak representing a small percentage of patients with acute syphilis who had an AI closer to the cutoff ( AI) was observed. 574 Am J Clin Pathol 2016;146: American Society for Clinical Pathology 574

4 Table 1 Clinical Data for Patients With a False-Reactive Screening Result (n ¼ 103) Diagnosis/Reason for Testing No. of Patients Antibody Index, Mean (SD) HIV or other infectious disease a (1.7) Kidney/liver disorders/pretransplant (2.5) Pregnancy (1.6) STI/routine screen (1.8) Autoimmune disease b (1.9) Total (1.95) HIV, human immunodeficiency virus; STI, sexually transmitted infection. a Includes HIV (n ¼ 25), genital herpes (n ¼ 1), hepatitis C virus (n ¼ 3), and chlamydia (n ¼ 1). b Systemic lupus erythematosus (n ¼ 1), autoimmune thyroiditis (n ¼ 2), rheumatoid arthritis (n ¼ 1), and hyperparathyroidism (n ¼ 1). Patients with false-reactive results (RPR nonreactive/ TP-PA nonreactive) had lower BioPlex AI values at initial screen (mean [SD], 2.4 [1.95]) (P <.0001). A broad distribution of BioPlex AI values through the entire reactive range was observed in patients with evidence of past syphilis infection (RPR nonreactive/tp-pa reactive) with a mean (SD) of 6.1 (2.45) (P <.0001). Lower values up to an AI of 5.0 predicted a high percentage of nonreactive TP-PA (false reactives), while AI values higher than 5.0 predicted less false reactives and TP-PA reactivity. At AI values of 5.1 to 6.0, the false-reactive rate for the BioPlex dropped from 38.7% to 8.7% but then increased to 13.6% at AI values of 7.1 to 8.0. At AI values of more than 8.0, the false-reactive rate dropped to 1.2%, with most specimens either RPR reactive or RPR nonreactive/tp-pa reactive. False-Reactive Cases Table 1 lists the reasons for screening for syphilis or other major diagnoses that were identified in the 103 patients who had false-reactive screens with the BioPlex IgG along with the mean antibody index. Most patients (n ¼ 59) were seen as part of a pretransplant evaluation or part of a routine screen for infectious diseases such as human immunodeficiency virus (HIV) (n ¼ 25), hepatitis C virus (n ¼ 3), Chlamydia trachomatis (n ¼ 1), and herpes simplex virus (n ¼ 1). Other conditions noted were pregnancy (n ¼ 20) and autoimmune diseases (n ¼ 5) such as systemic lupus erythematosus, autoimmune thyroiditis, and rheumatoid arthritis. Nineteen patients were screened for syphilis as part of an evaluation for sexually transmitted infections. The mean AI values for all patients were low and ranged from 2.0 to 2.8. Past Syphilis and Latent, Untreated Syphilis Cases The clinical data for patients who were BioPlex IgG reactive/rpr nonreactive/tp-pa reactive are listed in Table 2. The highest numbers of patients with both treated and untreated/unknown treatment were in the HIV-positive group, followed by patients who were screened for the pretransplant examination. Two patients were tested for syphilis because of a rash. One of the patients reported a rash on the hands and feet lasting for 4 months. This case was considered secondary syphilis by the clinicians despite inconsistent history provided by the patient. The second patient had a fleeting penile rash. It was unclear whether the rash was related to the positive syphilis serology. Five newborn infants born to mothers with syphilis were BioPlex IgG reactive/rpr nonreactive/tp-pa reactive in our study. One newborn infant was initially RPR reactive (titer 1:4) before the initiation of the study and then subsequently had nonreactive RPR results with reactive BioPlex IgG and TP-PA on follow-up testing (Table 2). This infant was born to an RPR reactive mother (titer 1:8). The mother was treated 1 month prior to delivery. A second newborn child was initially RPR negative, BioPlex positive, and TP-PA positive. The mother was reported as RPR reactive and treated with penicillin prior to delivery from another facility (RPR titer unknown). The newborn died of prematurity at 6 days of age with a diagnosis of respiratory distress syndrome and possible sepsis. The third infant was born to a mother who was RPR reactive (titer 1:1). The mother was treated with penicillin during pregnancy. The other two infants were born to mothers with a diagnosis of latent syphilis at the time of delivery. Both mothers did not recall a past infection or treatment for syphilis in the past. Medical Staff Communication and Ordering Practice Immediately after implementation of the reverse screening algorithm, the laboratory was cancelling and reordering 445 tests/month, which steadily declined to 120 tests/month prior to our s to clinicians who continued to order RPR for screening. We sent 225 s and received 46 (20%) replies. The most common reasons given for the incorrect orders were that they were unaware of the change and RPR was still in their EPIC preference list and that order sets had not been changed to reflect current policy. Oddly enough, no disagreement with the rationale for the reverse sequence algorithm was expressed. Several clinicians mentioned that an EPIC Alternative Selection tool would help guide the correct ordering of both tests. After the campaign, we saw another steady decrease in the number of incorrect orders. We finally hardwired the correct ordering practices through implementation of an EPIC Alternative Selection tool. Thereafter, the number of incorrect orders again steadily declined and remained stable at an average of 19 tests/month through March American Society for Clinical Pathology Am J Clin Pathol 2016;146:

5 Rourk et al /REVERSE SYPHILIS SCREENING ALGORITHM Table 2 Clinical Data for BioPlex IgG Reactive/RPR Nonreactive/TP-PA Reactive Diagnosis/Reason for Testing Discussion Past, Treated Syphilis A number of studies have examined the impact of reverse syphilis screening on the diagnosis of syphilis, including studies of areas with either low or high prevalence for syphilis. South Carolina is considered a moderately highprevalence area for syphilis infections. The rate of primary and secondary syphilis was 5.7 per 100,000 in The state ranks 11th in rates of syphilis among 50 states. 7 Nine cases of congenital syphilis were reported from 2009 through Previous studies have demonstrated a high percentage of false-reactive tests using EIA/CLIA methods in the reverse syphilis screening algorithm. 3,5,8,9 Reverse screening with EIA/CLIA often yields a higher false-reactive rate than traditional testing does in populations with both a low and high prevalence of syphilis. 3,5 Our false-reactive rate of 1.0% is more consistent with rates seen in populations with a low prevalence of syphilis. In our study, among 502 serum samples that were reactive with the syphilis screening IgG EIA, 352 were nonreactive by nontreponemal testing for a discordant rate of 70.1%. This is a higher discordant rate than the 2008 CDC study reporting a 56.0% discordance 10 and the 2011 CDC study reporting a 56.7% discordance. 3 Furthermore, when our discordant serum samples were tested with TP-PA, 103 patients were nonreactive at a rate of 29.3%. Similarly, in the 2011 CDC study, the overall percentage of patients with nonreactive TP-PA or FTA-ABS tests was 31.6%. However, the false-reactive rate was 2.9 times greater in the low-prevalence population than in the high-prevalence population (40.8% vs 14.1%) in that study. 3 Several groups have reported that the BioPlex syphilis IgG AI can help identify false-reactive results and reduce the need for confirmatory testing. Recently, Yen-Lieberman et al 11 reported that the AI of the BioPlex syphilis IgG could be used to identify likely false-reactive results and reduce the need for confirmatory EIA testing. In their study, they demonstrated that specimens with AIs of 6.0 or higher were always Interpretation, No. Latent Syphilis, Untreated or Unknown Past Treatment HIV infection Kidney/liver disorders/pretransplant STI screen Cognitive disorder History of syphilis infection in mother Penile or cervical lesion/discharge Rash Total In Utero Exposure to Mother With Syphilis HIV, human immunodeficiency virus; IgG, immunoglobulin G; RPR, rapid plasma reagin; STI, sexually transmitted infection; TP-PA, Treponema pallidum particle agglutination. reactive when tested with another EIA and proposed that only specimens with a BioPlex AI of less than 6.0 be confirmed with a second test. Their study, however, was performed in a low-prevalence population. Loeffelholz et al 12 demonstrated that an AI of 8 or higher was 99% predictive of a reactive TP- PA test in several different populations and could potentially be used to eliminate unnecessary TP-PA tests. Our results were very similar to Loeffelholz et al, 12 with only 1.2% of specimens with an AI of 8 or higher found to be falsely reactive. If laboratories choose to use the BioPlex AI as part of their algorithm, they must independently verify the performance characteristics of this modification of the manufacturer s recommendation because the optimal AI cutoff may vary by patient population and confirmatory test used. More than half of the patients with false-reactive BioPlex IgG results were being screened for syphilis because of HIV infection, other infectious diseases, or pretransplant workup. Twenty patients with false-reactive results with the BioPlex IgG were screened because of pregnancy/miscarriage, and five had a history of autoimmune disease. Pregnancy has been well recognized as a cause of false-reactive nontreponemal tests. However the data are limited regarding the causes of false-reactive treponemal tests in pregnancy. 13,14 A recent study by Mmeje et al 15 examining discordant syphilis results using the reverse syphilis algorithm found that most pregnant women with discordant serology were CLIA reactive/rpr nonreactive/ TP-PA nonreactive on initial screen. More than half of these patients (53%) became CLIA nonreactive on retesting. In our study, the institution of the reverse syphilis algorithm led to the discovery of 21 cases of syphilis (8.7% of the EIA reactive/rpr nonreactive/tp-pa reactive cases) in which the diagnosis was either latent syphilis or untreated/ unknown treatment of past syphilis (Table 2). The reverse algorithm is beneficial for the diagnosis of patients, especially those who have early or late infections with nonreactive nontreponemal tests. In our study, there were many more cases of detection of late infections, but two cases 576 Am J Clin Pathol 2016;146: American Society for Clinical Pathology 576

6 involving presentation with rash (see Table 2) may have been due to an early syphilis or secondary syphilis. Because screening with treponemal-based tests cannot distinguish between treated and untreated syphilis, medical history of past treatment for syphilis will be important in determining the diagnosis. False-reactive treponemal EIA screening tests will create problems with repeated testing, especially in low-risk populations with low pretest probabilities of syphilis. The data presented on the number of false-reactive results in an area with a moderately high prevalence of syphilis continue to support the importance of performing a second treponemal assay (eg, TP-PA) when the results of the reverse syphilis screen and RPR are discordant. 3 Our institution decided to use the BioBlex IgG for our syphilis screening assay based on ease of use, high throughput, and automation. Other syphilis screening assays are available, such as the LIASON T pallidum specific assay (Diasorin, Stillwater, MN), that detect both a treponemal immunoglobulin M (IgM) antibody in addition to an IgG antibody. Since the BioPlex screening assay does not detect IgM treponemal antibodies, it may not detect very early cases of syphilis. An IgM assay for treponemal syphilis antibodies is under development for the BioPlex instrument. Future studies by clinical laboratories should include the detection of IgM treponemal antibodies in the BioPlex assay to determine whether this significantly improves the sensitivity of the assay without a loss of specificity. Our study also demonstrated that laboratories must be prepared to invest significant resources to educate clinicians and modify their ordering practices when changing from the traditional to the reverse screening algorithm. The first month after the change, we cancelled and converted nearly 100% of the syphilis screening test orders. That process, coupled with an to clinicians who continued to order RPR as the initial test, resulted in a slow but steady decline in inappropriate screening test orders. Finally, we hardwired the correct ordering practices through leveraging the capabilities of our electronic medical record. Corresponding author: Christine M. Litwin, MD, Dept of Pathology and Laboratory Medicine, Medical University of South Carolina, 171 Ashley Ave, Charleston, SC 29425; litwinc@musc.edu. This study was supported by the Medical University of South Carolina. References 1. Wassermann A. Eine serodiagnostische reaktion bei Syphilis. Dtsch Med Wochenschr. 1906;32: Larsen SA, Steiner BM, Rudolph AH. Laboratory diagnosis and interpretation of tests for syphilis. Clin Microbiol Rev. 1995;8: Centers for Disease Control and Prevention. Discordant results from reverse sequence syphilis screening five laboratories, United States, MMWR Morb Mortal Wkly Rep. 2011;60: Sena 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: Binnicker MJ, Jespersen DJ, Rollins LO. Direct comparison of the traditional and reverse syphilis screening algorithms in a population with a low prevalence of syphilis. J Clin Microbiol. 2012;50: Gratrix J, Plitt S, Lee BE, et al. Impact of reverse sequence syphilis screening on new diagnoses of late latent syphilis in Edmonton, Canada. Sex Transm Dis. 2012;39: South Carolina 2015 state health profile. gov/std. Accessed February 4, Lipinsky D, Schreiber L, Kopel V, et al. Validation of reverse sequence screening for syphilis. J Clin Microbiol. 2012;50: Marangoni A, Moroni A, Accardo S, et al. Laboratory diagnosis of syphilis with automated immunoassays. J Clin Lab Anal. 2009;23: Centers for Disease Control and Prevention. Syphilis testing algorithms using treponemal tests for initial screening four laboratories, New York City, MMWR Morb Mortal Wkly Rep. 2008;57: Yen-Lieberman B, Daniel J, Means C, et al. Identification of false-positive syphilis antibody results using a semiquantitative algorithm. Clin Vaccine Immunol. 2011;18: Loeffelholz MJ, Wen T, Patel JA. Analysis of Bioplex syphilis IgG quantitative results in different patient populations. Clin Vaccine Immunol. 2011;18: Watson-Jones D, Gumodoka B, Weiss H, et al. Syphilis in pregnancy in Tanzania, II: the effectiveness of antenatal syphilis screening and single-dose benzathine penicillin treatment for the prevention of adverse pregnancy outcomes. J Infect Dis. 2002;186: Nandwani R, Evans DT. Are you sure it s syphilis? A review of false positive serology. Int J STD AIDS. 1995;6: Mmeje O, Chow JM, Davidson L, et al. Discordant syphilis immunoassays in pregnancy: perinatal outcomes and implications for clinical management. Clin Infect Dis. 2015;61: American Society for Clinical Pathology Am J Clin Pathol 2016;146: