Screening for Syphilis Infection in Pregnant Women: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force

Importance  The incidence of syphilis and congenital syphilis in the United States has increased after reaching historic lows in the early 2000s.

Objective  To systematically review literature on the effectiveness and harms of screening for syphilis in pregnancy and the harms of penicillin treatment in pregnancy to inform the US Preventive Services Task Force.

Data Sources  MEDLINE, PubMed, and the Cochrane Central Register of Controlled Trials for relevant English-language literature, published from January 1, 2008, to June 2, 2017. Ongoing surveillance was conducted through November 22, 2017.

Study Selection  Studies conducted in countries categorized as “high” or “very high” on the Human Development Index that explicitly addressed 1 of 3 a priori–defined key questions.

Data Extraction and Synthesis  Independent critical appraisal and data abstraction by 2 reviewers. Data from included studies were narratively synthesized without pooling data.

Main Outcomes and Measures  Incidence of congenital syphilis; any harms of screening or penicillin treatment in pregnancy.

Results  Seven studies in 8 publications were included. One observational study evaluated the implementation of syphilis screening in pregnancy in 2 441 237 women in China. From 2002 to 2012, screening for syphilis in all pregnant women increased from 89.8% to 97.2%, and the incidence of congenital syphilis decreased from 109.3 to 9.4 cases per 100 000 live births. Five studies (n = 21 795) evaluated the false-positive findings of treponemal tests and 1 study (n = 318) evaluated the false-negative findings of nontreponemal tests. These studies found that false-positives with treponemal-specific enzyme or chemiluminescent immunoassays were common (46.5%-88.2%), therefore warranting reflexive (automatic confirmatory) testing for all positive test findings. One study (n = 318) found no false-negatives with treponemal tests, and 1 study (n = 139) demonstrated the prozone phenomenon (false-negative response from high antibody titer) with rapid plasma reagin screening using undiluted samples (2.9%). No studies were identified for harms of penicillin in pregnancy.

Conclusions and Relevance  Screening for syphilis infection in pregnant women is associated with reduced incidence of congenital syphilis, and available evidence supports the need for reflexive testing for positive test results.

In the United States, the rate of reported congenital syphilis was 15.7 cases per 100 000 live births in 2016, the highest rate reported since 2001.^1,2 Congenital syphilis is an infectious disease caused by the vertical transmission of Treponema pallidum; thus, prevention and detection of congenital syphilis depend on the identification of syphilis in pregnant women. Two screening protocols are commonly used: the traditional screening algorithm (ie, nontreponemal testing with reflex to treponemal testing) and the reverse sequence screening algorithm (ie, treponemal testing with reflex to nontreponemal testing) (Figure 1).^3,4 Untreated syphilis in pregnancy carries significant risk for stillbirth or fetal loss, premature birth, low birthweight, congenital syphilis, and neonatal death.^5,6 Parenteral benzathine penicillin G is the only recommended antibiotic for preventing maternal transmission of syphilis to the fetus and treating fetal syphilis infection.⁷ This evidence review was completed to inform the US Preventive Services Task Force (USPSTF) in the update to its 2009 “A” recommendation to screen all pregnant women for syphilis.^8,9

Because this topic represents well-established, evidence-based standards of practice, the USPSTF commissioned a targeted review using an updating process known as “reaffirmation,” which aims to identify “new and substantial evidence sufficient enough to change the prior recommendation.”^10,11 As such, only the interval evidence for targeted key questions from the previous systematic review is included. After members of the USPSTF were consulted, an analytic framework and 3 key questions (KQs) were developed to guide the evidence update (Figure 2). Detailed methods and results, including evidence to address the effect of repeat testing for syphilis in the third trimester, at delivery, or both, are available in the full evidence report at https://www.uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/syphilis-infection-in-pregnancy-screening1.

MEDLINE, PubMed (publisher-supplied references only), and the Cochrane Central Register of Controlled Trials were searched from January 1, 2008, to June 2, 2017 (eMethods in the Supplement). In addition to these searches, reference lists of existing reviews and primary studies were scanned. Searches were limited to articles published in English. Active surveillance via article alerts and targeted searches of high–impact factor journals to identify major studies published in the interim was conducted through November 22, 2017.

Two reviewers independently reviewed 453 unique citations and 34 full-text articles against a priori inclusion criteria (Figure 3; eTable 1 in the Supplement). For all KQs, studies conducted in countries categorized as “high” or “very high” on the Human Development Index were included. For evidence on the benefits of screening for syphilis in pregnancy (KQ1), randomized or nonrandomized controlled intervention studies and large before-after or ecologic studies reporting the association of implementing a screening program with the incidence of congenital syphilis and other adverse outcomes in pregnant women with syphilis were included. For evidence on the harms of screening (KQ2), studies in pregnant women that reported psychosocial harms, stigma, and screening test inaccuracy (ie, false-positive or false-negative results) were included. For KQ1 and KQ2, studies of screening for syphilis in asymptomatic pregnant women using either traditional or reverse sequence algorithms were selected. Studies of screening tests not currently used in US primary care settings and studies in women living with HIV were excluded. For evidence on the harms of treatment (KQ3), studies of penicillin treatment for syphilis in pregnant women that reported any maternal or neonatal harms were included.

Independent critical appraisal of included trials was conducted by 2 reviewers using predefined criteria (eTable 2 in the Supplement), with disagreements resolved by discussion. Each study was rated as good, fair, or poor quality. A good-quality study met all quality criteria. A fair-quality study failed to meet at least 1 criterion but had no known issue that would invalidate its results. Studies were rated as poor quality if they had major risk of bias. No studies were excluded for poor quality. Important study and participant characteristics and outcomes were abstracted and subsequently checked by a second reviewer for accuracy and completeness.

Data from 7 studies (reported in 8 publications) were summarized in a narrative format, with an accompanying summary table for KQ2.

Key Question 1. Does screening for syphilis in pregnant women reduce the incidence of congenital syphilis in newborns?

One fair-quality observational study, which used both historical and geographic comparators, was designed to evaluate the implementation of free syphilis screening (with follow-up and treatment) among all pregnant women living in the region of Shenzhen, China.¹² This study was included in the last evidence update to support the 2009 recommendation statement; however, results from longer-term follow-up have since been published. All pregnant women from January 2002 to December 2012 in 90 hospitals in Shenzhen (n = 2 441 237) were offered syphilis and HIV screening. A nontreponemal test was used to screen for syphilis, with reflex to treponemal testing if the test result was positive. Women testing positive for syphilis by serology were given follow-up visits and treatment (including health education), partner notification, and the opportunity to terminate their pregnancy. Women who chose to continue their pregnancies were treated with 3 injections of penicillin G (2.4 million units intramuscularly) at weekly intervals.

From 2002 to 2012, 8455 of the 2 441 237 pregnant women screened tested positive for syphilis.¹² The timing of screening pregnant women was not reported; however, the mean gestational week in which treatment occurred was 26.5 weeks (SD, 11.2 weeks; range, 3-43 weeks). The trend over the 10 years of observation of the timing of screening, treatment, or both (eg, if screening, treatment, or both occurred earlier in pregnancy in later years) was not reported. From 2002 to 2012, screening for syphilis in all pregnant women increased from 89.8% to 97.2%, and the incidence of congenital syphilis decreased from 109.3 to 9.4 cases per 100 000 live births. During this same period, in pregnant women infected with syphilis, the incidence of all adverse outcomes declined from 42.7% to 19.2%; incidence of congenital syphilis declined from 11.7% to 3.2%; and incidence of stillbirth or fetal loss declined from 19.0% to 3.3%. Although this study does not include an historical comparator (ie, a time point before implementation of the screening program) because the screening program was initiated in 2001 and screening commenced in 2002, the authors also report the incidence of congenital syphilis in Shenzhen compared with the national incidence. From 2002 to 2012, the incidence of congenital syphilis in China increased from 5.9 to 97.4 cases per 100 000 live births, while incidence of congenital syphilis specifically in Shenzhen decreased from 109.3 to 9.4 cases per 100 000 live births. No P values are reported for any of these comparisons or trends of outcomes. Despite methodological limitations (with both the historical and geographic comparisons) and concerns about applicability to US practice (timing of screening; treatment options, including termination of pregnancy and use of erythromycin for women with penicillin allergies), this study provides observational evidence that screening for, coupled with treatment of, syphilis in pregnancy is associated with a decrease in incidence of congenital syphilis.

Key Question 2. What are the harms of screening for syphilis in pregnant women?

Five new studies (n = 21 795) that reported on false-positive results of treponemal tests were identified,^13-17 1 of which also reported on false-negative results,¹⁷ along with 1 new study (n = 318) that reported on false-negative results of nontreponemal testing (Table 1).¹⁸ No studies were found that addressed the diagnostic inaccuracy of the entire screening algorithm or other potential harms of screening for syphilis in pregnant women. Four large, fair-quality retrospective studies evaluated the proportion of false-positive results using treponemal-specific enzyme immunoassays (EIAs) or chemiluminescent immunoassays (CIAs) in screening pregnant women for syphilis.^13-16 These studies found that false-positives with EIA or CIA were common (46.5 to 88.2%), therefore warranting reflexive (automatic confirmatory) testing for all positive CIA or EIA test results. None of the studies reported confidence intervals for false-positives.

One fair-quality prospective study evaluated the diagnostic accuracy of CIA and the T pallidum particle agglutination assay (TPPA) in 318 pregnant women.¹⁷ This study had only 1 positive result from CIA testing and 2 positive results from TPPA testing and therefore could not provide robust estimates of false-positives. This study found no false-negatives for any test.

One fair-quality retrospective study (n = 139) evaluated the prozone phenomenon using rapid plasma reagin (RPR) testing.¹⁸ The prozone phenomenon occurs when undiluted serum containing a high titer of nonspecific antibody (as may occur in secondary syphilis) produces a false-negative result attributable to a large quantity of antibodies occupying all the antigen sites (preventing flocculation).³ This study repeated RPR testing in discordant samples (RPR-negative/TPPA-positive) using diluted serum and found that 2.9% of discordant samples had a false-negative RPR result attributable to the prozone phenomenon.

Key Question 3. What are the harms of treatment of syphilis with penicillin during pregnancy to pregnant women or newborns?

No studies directly examining the harms of penicillin in pregnancy and meeting the inclusion criteria were identified. In particular, no studies were found that addressed the risk of the Jarisch-Herxheimer reaction or serious adverse events in women with a history of penicillin allergy.

The findings of this brief evidence review support the understanding that screening for syphilis early in pregnancy reduces congenital syphilis and also support the need for reflexive testing to investigate initial positive EIA/CIA test results in reverse sequence screening (Table 2). Screening for syphilis at the first prenatal visit to prevent congenital syphilis is standard of care and legally mandated in most US states.¹⁹ Observational evidence not included in this review supports the effectiveness of identification and treatment of syphilis in pregnancy to avoid adverse outcomes of pregnancy and specifically supports identification and treatment as early as possible in pregnancy (as opposed to in the third trimester or at delivery).^5,12,20

This update includes longer-term follow-up from an observational study evaluating the implementation of syphilis screening in more than 2 million pregnant women in Shenzhen, China, demonstrating an approximate 11-fold decrease in the incidence of congenital syphilis over 10 years. Screening for syphilis using treponemal and nontreponemal tests in combination is feasible for mass screening and provides a presumptive laboratory diagnosis of syphilis with high accuracy and reliability. Because of the false-positive test results with initial treponemal testing (ie, CIA or EIA) and a negative RPR and TPPA result in low-risk patients or low-prevalence populations, clinician education on the reverse sequencing algorithm and interpretation and limitations of syphilis serologic test results in general is critical to avoid overdiagnosis or underdiagnosis and treatment errors. Evidence from this review confirms concern for false-positives with treponemal-specific screening tests in low-risk pregnant females when the RPR result is negative and the prozone phenomenon has been ruled out, supporting the rationale for treponemal reflexive testing. Penicillin G is generally accepted to be effective and safe for use in pregnancy. Observational data support the effectiveness of benzathine penicillin G in preventing congenital syphilis when the mother is treated early in pregnancy, and serious harms are uncommon^5,6; however, good-quality evidence in pregnant women is lacking.²¹ The Jarisch-Herxheimer reaction, which can induce early labor or cause fetal distress in pregnant women, albeit rarely, is more common in primary and secondary syphilis during pregnancy and cannot be mitigated with a different choice of antibiotic.

This review was intended to support the USPSTF reaffirmation process and thus includes only the interval evidence accrued since the last recommendation in 2009. The review was scoped to identify evidence that could result in a change in this recommendation and therefore has some notable exclusions. First, it did not include studies addressing the effectiveness of screening or early prenatal care in low- or middle-income countries, because these studies were less applicable to prenatal care in the United States. Second, the review did not address the comparative screening accuracy of traditional vs reverse sequence algorithm testing; however, to our knowledge, no studies have compared these 2 testing algorithms in prenatal care. Third, the benefit of penicillin G for the treatment of syphilis is well established, so new evidence for this question was not included. Fourth, because the review was primarily focused on screening, it did not address the efficacy of alternative antibiotic treatments (eg, ceftriaxone) in pregnant women (with or without penicillin allergies).

Screening for syphilis infection in pregnant women is associated with reduced incidence of congenital syphilis, and available evidence supports the need for reflexive testing for positive test results.

Corresponding Author: Jennifer S. Lin, MD, Kaiser Permanente Research Affiliates Evidence-based Practice Center, Center for Health Research, Kaiser Permanente Northwest, 3800 N Interstate Ave, Portland, OR 97227 (jennifer.s.lin@kpchr.org).

Accepted for Publication: May 18, 2018.

Author Contributions: Dr Lin had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: Lin, Eder.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: Lin, Eder.

Obtained funding: Lin.

Administrative, technical, or material support: Eder, Bean.

Supervision: Lin.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: This research was funded under contract HHSA290201200015I, Task Order 5, from the Agency for Healthcare Research and Quality (AHRQ), US Department of Health and Human Services, under a contract to support the USPSTF.

Role of the Funder/Sponsor: Investigators worked with USPSTF members and AHRQ staff to develop the scope, analytic framework, and key questions for this review. AHRQ had no role in study selection, quality assessment, or synthesis. AHRQ staff provided project oversight, reviewed the report to ensure that the analysis met methodological standards, and distributed the draft for peer review. Otherwise, AHRQ had no role in the conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript findings. The opinions expressed in this document are those of the authors and do not reflect the official position of AHRQ or the US Department of Health and Human Services.

Additional Contributions: We gratefully acknowledge the following individuals for their contributions to this project: Smyth Lai, MLS, Katherine Essick, BS, and Shannon Robalino, MSLS (Kaiser Permanente Center for Health Research); Peter Miksovsky, MD (Northwest Permanente); Tina Fan, MD, MPH (AHRQ); and current and former members of the USPSTF who contributed to topic deliberations. USPSTF members, peer reviewers, and federal partner reviewers did not receive financial compensation for their contributions.

Additional Information: A draft version of this evidence report underwent external peer review from 5 content experts (Robert Phillips Heine, MD, Duke University School of Medicine; Jeanne S. Sheffield, MD, Johns Hopkins School of Medicine; and 3 individuals from the US Centers for Disease Control and Prevention). Comments from reviewers were presented to the USPSTF during its deliberation of the evidence and were considered in preparing the final evidence review.

Editorial Disclaimer: This evidence report is presented as a document in support of the accompanying USPSTF Recommendation Statement. It did not undergo additional peer review after submission to JAMA.