Article Text

Download PDFPDF
Treponema pallidum PCR testing for diagnosis of mucocutaneous ulcers suspicious for syphilis
  1. Muhammad Hyder Junejo1,
  2. Mark Collery2,
  3. Gary Whitlock1,
  4. Alan McOwan1,
  5. Victoria Tittle1,
  6. Diarmuid Nugent1
  1. 1 56 Dean Street, Chelsea and Westminster Hospital NHS Foundation Trust, London, UK
  2. 2 Micropathology, Coventry, UK
  1. Correspondence to Dr Muhammad Hyder Junejo, 56 Dean Street, Chelsea and Westminster Hospital NHS Foundation Trust, London, UK; muhammad.junejo{at}


Background Primary syphilis is characterised by the appearance of an ulcerated lesion (chancre) on the anogenital or oral mucosa from which Treponema pallidum DNA may be detectable by PCR. Serological tests for syphilis may be non-reactive in early infection, even after the appearance of a chancre. We reviewed the use of a multiplex-PCR (M-PCR) test to determine the added value of T. pallidum DNA detection in the management of individuals presenting with mucocutaneous ulceration at a sexual health service in central London.

Methods We performed a cross-sectional analysis of all individuals with detectable T. pallidum DNA from September 2019 to April 2020. Electronic patient records were reviewed and concomitant results for treponemal serology and/or rapid plasma reagin (RPR) extracted, along with demographic data, history of syphilis and indices of sexual behaviour including number of sexual partners contacted. Any subsequent treponemal serology and RPR results were also reviewed.

Results M-PCR swab specimens were performed in 450 individuals, of whom 63 (14%) had detectable T. pallidum DNA; 60 of 63 (95%) were gay or bisexual men and 11 of 63 (17%) were living with HIV. A history of treated syphilis was present in 17 of 63 (27%). Same-day treponemal serology/RPR testing was performed in 58 of 63 (92%) patients. Of the 58 who had same-day syphilis serology/RPR, 9 (16%) had their syphilis infection confirmed by treponemal DNA PCR alone. A total of 165 partners were traced as contacts of infection, of whom 25 (15%) were contacts of individuals diagnosed by M-PCR testing alone.

Conclusion In individuals with T. pallidum PCR-positive lesions, around one in six in our cohort were negative on standard diagnostic serological tests for syphilis. Treponemal DNA testing is an important addition to serological assays in individuals with mucocutaneous ulceration who are at risk of recent syphilis infection and facilitates early diagnosis and contact tracing.

  • PCR
  • diagnosis
  • management
  • syphilis
  • skin ulcer

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Syphilis diagnoses in England have tripled from 2010 to 2019, disproportionately affecting gay, bisexual and other men who have sex with men (MSM), who accounted for almost 80% of cases in 2019.1

Classically, primary syphilis presents with a painless, solitary anogenital/oral chancre. This may progress to heterogeneous manifestations, including oral or genital mucous patches in early secondary syphilis, and/or asymptomatic latent infection which may result in long-term sequelae.2

Syphilis diagnostics are challenging and usually rely on serological assays alongside clinical presentation and prior serology results. Serological markers include specific treponemal tests such as treponemal enzyme immunoassay (EIA) or Treponema pallidum particle agglutination assay (TPPA), and non-specific tests such as venereal disease research laboratory and rapid plasma reagin (RPR). According to UK clinical guidelines,2 treponemal tests are preferred to screen for initial infection, but as these remain positive after treatment they cannot detect reinfection. Both treponemal and non-treponemal tests may be non-reactive for up to 2 weeks following appearance of a chancre.2 However, T. pallidum DNA is detectable by PCR from swab specimens taken from a lesion, offering the opportunity for laboratory-confirmed diagnosis in those with early infection and negative serology or without a diagnostic fourfold increase in serum RPR.3

56 Dean Street is a combined sexual health and HIV service in central London and part of Chelsea and Westminster NHS (National Health Service) Foundation Trust, a publicly funded provider of specialist medical care across multiple hospital and community sites in central and west London. At 56 Dean Street, 1420 syphilis infections were diagnosed in 2019,4 accounting for 17.8% of all syphilis diagnoses in MSM in England over the same time period.5 A combined treponemal IgG/IgM EIA is used to test those with no history of syphilis and confirmatory TPPA and quantitative RPR performed in those with a positive EIA. Quantitative RPR is used to test for reinfection in those with prior treated syphilis, alongside repeat TPPA testing. In individuals presenting with anogenital or oral ulceration, direct testing of lesion exudate using a multiplex-PCR (M-PCR) swab detects T. pallidum, Haemophilus ducreyi and herpes simplex virus (HSV) DNA. This may be employed in addition to serological tests for syphilis. We reviewed our use of M-PCR to determine the added value of T. pallidum DNA detection in diagnosing individuals with early syphilis presenting with oral or anogenital lesions.


The results of all M-PCR test requests from September 2019 until April 2020 were extracted from electronic patient records. A retrospective, cross-sectional analysis was performed of all individuals with detectable T. pallidum DNA. Concomitant treponemal serology and/or RPR results were extracted along with demographic data, clinical features and indices of sexual behaviour. The number of sexual partners notified as contacts of syphilis was recorded according to timeframes specified by UK national clinical standards.6 Any subsequent treponemal serology and RPR results performed within Chelsea and Westminster Hospital NHS Foundation Trust were also reviewed.

Specimen DNA extraction for M-PCR

Samples for M-PCR were collected with a swab and sent to an external provider for processing (Micropathology, Coventry, UK). Genomic DNA extraction was performed for T. pallidum using the polA gene as a target (estimated sensitivity 97.56% and specificity 100%).7 A 200 µL aliquot of swab resuspension solution was taken and extracted on the KingFisher Bio-Robot using an ‘in-house’ universal DNA extraction protocol developed with input from the manufacturer (Promega) based on the Maxwell HT DNA Kit with a final DNA eluate of 200 µL.

T. pallidum DNA assay

A nested PCR amplification was performed using primers as described elsewhere.8 Each first-round reaction mixture contained 1x PCR buffer, 1.5 mM magnesium chloride (MgCl2), 0.1 mM deoxynucleoside triphosphates (dNTPs), 1U SuperTherm Taq Polymerase (LPI, UK) and 0.5 µM primer, to which 20 µL nucleic acid extract was added. In the second-round reaction, 1 µL aliquot of first-round reaction was added to a second-round reaction mix containing 0.5 µM primers using the MyTaq HS reaction mix (Meridian Bioscience), containing per reaction 20 µL nuclease-free water (Severn Biotech), 5 µL MyTaq HS mix (made by adding 33 µL MyTaq HS enzyme to 1.8 mL of the supplied buffer) and 1 µL EvaGreen (Biotium). First and second reaction mixes were subjected to thermal cycling on a programme of initial heating to 95°C for 1 min 5 s, followed by 30 cycles of 95°C for 20 s, 65°C for 20 s and 72°C for 20 s. Second-round reaction mixes were subjected to thermal cycling on a programme of initial heating to 95°C for 1 min 45 s, followed by 30 cycles of 95°C for 20 s, 65°C for 20 s and 72°C for 20 s. Amplicons were detected from PCR reactions by agarose gel electrophoresis or by melting products and visualisation using a Roche 480 real-time thermal cycler. M-PCR tests received prior to 09:00 on a given day had a 48-hour turnaround time for results reporting.


A total of 450 M-PCR specimens were taken, of which 63 (14%) were positive for T. pallidum; 132 (29%) were positive for HSV 1/2. There were no cases of HSV/T. pallidum coinfection. Of 63 individuals with T. pallidum detected, 62 were male and 1 was female, with a median age of 35 years (IQR 30–42); 59 (93%) were MSM and 11 (17%) were living with HIV. Recreational drug use during sex was reported by 9 of 63 (14%) individuals, of whom 1 (11%) reported using injectable drugs. Prior treated syphilis was reported by 17 of 63 (27%); of the 63 individuals, 32 (51%) had a solitary ulcerative lesion, 30 (48%) had multiple lesions and 1 (1%) had a rash. Lesions with detectable T. pallidum DNA were located on the penis (n=39), perianus (n=17), scrotum (n=2), oropharynx (n=1), penis and perianus (n=1), penis and perineum (n=1), and palm (suspected secondary syphilis with excoriated rash) (n=1); in one case the site was undocumented.

Same-day syphilis serology or RPR was performed in 58 of 63 (92%) patients. Of five individuals in whom concurrent blood tests were not sent, one had a new positive EIA 5 days previously and was attending for treatment when the chancre was identified and M-PCR performed; three were recalled for antibiotic treatment and serology/RPR sent on the day of treatment; and one had empiric treatment for presumed primary syphilis on the day of M-PCR without confirmatory blood tests.

Of 58 patients who had syphilis serology/RPR testing on the day that their positive treponemal DNA specimen was collected, 9 (15%) had syphilis confirmed by treponemal DNA PCR alone (table 1). A total of 165 partners were informed as contacts of infection, of whom 25 (15%) were contacts of individuals diagnosed by M-PCR testing alone.

Table 1

Serology results in individuals with detectable M-PCR and negative concomitant Treponema pallidum serology and/or without diagnostic fourfold RPR increase


Around one in six individuals in our cohort with detectable T. pallidum DNA tested negative for syphilis on serological assays, according to standard diagnostic algorithms used to interpret results. Our data confirm that a negative EIA or absence of a fourfold increase in RPR cannot exclude primary syphilis, in keeping with the 80% sensitivity for serology observed in a recent analysis at another UK centre.9 Longitudinal follow-up showed individuals with DNA-positive lesions treated prior to EIA antibody positivity may be TPPA-negative following treatment (patients 2, 3 and 4 in table 1).

M-PCR testing allowed for laboratory-confirmed diagnosis in cases where serology responses had not yet evolved. This facilitated linkage of 25 additional contacts to testing and epidemiological treatment. Given that chancres are highly infectious, it is reasonable that contact tracing for individuals with PCR-positive lesions should be prioritised.

Empiric antibiotic treatment is sometimes used in individuals at risk of recent syphilis infection presenting with lesions suggestive of a chancre. Nevertheless, clinical appearances may be non-specific: HSV, chancroid, lymphogranuloma venereum, trauma and aphthous ulceration may all resemble primary syphilis. In an era of increased antibiotic stewardship, empiric treatment risks antimicrobial overuse particularly if employed for cases of equivocal appearance. Contact tracing based on presumed, rather than laboratory-confirmed, infection is also problematic. Dark-ground microscopy of exudate taken from a lesion offers the advantage of a point-of-care diagnosis, but is operator-dependent, has reduced sensitivity relative to PCR and cannot diagnose syphilis on samples taken from oral lesions.10

Our analysis has certain limitations. The absence of treponemal DNA testing does not imply all cases with negative serology would have been missed. Typical chancres can be identified by experienced clinicians based on their clinical presentation and dark-ground microscopy findings. Nonetheless, we feel that PCR testing may be particularly helpful where the diagnosis is less certain or in lower incidence settings with less familiarity of the clinical manifestations of syphilis. We did not review records of individuals presenting with ulceration who did not undergo M-PCR testing. In our review of subsequent treponemal serology tests in the nine individuals with confirmed syphilis infection on M-PCR testing alone, we cannot exclude the possibility of retesting, or indeed treatment of any subsequent reinfection, at healthcare facilities outside of our organisation.

We conclude that treponemal DNA testing is an important addition to serology in individuals with anogenital or oral lesions who are at risk of recent syphilis infection and facilitates early diagnosis and contact tracing.

Key messages

  • Serological tests for syphilis may be negative in early infection.

  • 9 of 58 (16%) individuals with positive Treponema pallidum DNA PCR had a negative concomitant treponemal serology or did not show a fourfold increase in rapid plasma reagin.

  • 25 of 165 (15%) partners informed were contacts of individuals with T. pallidum confirmed on PCR alone.

  • DNA testing is an important addition to serology in individuals with oral or anogenital ulceration who are at risk of recent syphilis infection.

Ethics statements

Patient consent for publication

Ethics approval

As a review of routine clinical practice, ethical approval was not required. The study was registered as an audit at Chelsea and Westminster Hospital NHS Foundation Trust. Reference number: WCSLA946.



  • Handling editor Federico Garcia

  • Contributors DN conceived the idea for the study. MHJ and DN developed the study design. MHJ carried out the notes review. MHJ, DN and MC drafted the study report. GW, AM and VT reviewed and edited the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.