Article Text

Download PDFPDF

Short report
Clinical factors associated with syphilis concordance in men in sexual partnerships: a cross-sectional couples study
  1. Janet M Towns1,2,
  2. Sarah Huffam1,2,
  3. Eric P F Chow1,2,
  4. Ian Denham1,
  5. Lei Zhang1,2,3,4,
  6. Andrea Ryan1,
  7. Clare Elizabeth Bellhouse1,2,
  8. Melanie Bissessor1,
  9. Catriona S Bradshaw1,2,
  10. Tiffany Phillips1,
  11. Stephen Graves5,
  12. Christopher K Fairley1,2,
  13. Marcus Y Chen1,2
  1. 1 Melbourne Sexual Health Centre, Alfred Health, Melbourne, Victoria, Australia
  2. 2 Faculty of Medicine, Nursing and Health Sciences, Monash University Central Clinical School, Melbourne, Victoria, Australia
  3. 3 Faculty of Medicine, Nursing and Health Sciences, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
  4. 4 Research Center for Public Health, Tsinghua University, Beijing, China
  5. 5 Australian Rickettsial Reference Laboratory, Barwon Health, Geelong, Australia
  1. Correspondence to Dr Janet M Towns, Melbourne Sexual Health Centre, Alfred Health, Carlton, VIC 3053, Australia; jtowns{at}


Background Syphilis infections continue to increase among men who have sex with men (MSM) in many countries, with rates often higher among HIV-positive MSM. There is limited understanding of the risk and determinants of syphilis transmission between men. We aimed to examine the concordance of early syphilis infection between male sexual partners and clinical factors associated with transmission.

Methods Men attending Melbourne Sexual Health Centre with their male partners, where at least one was diagnosed with early syphilis, were identified from linkage of partner records between March 2011 and April 2016. Early latent syphilis was defined as a new asymptomatic syphilis presentation of less than 2 years’ duration. Associations between concordance and potential risk factors were examined using Fisher’s exact test.

Results Among 43 couples (86 men) identified, there were 13 couples (26 men) where both were diagnosed with early syphilis, representing a concordance rate of 30.2% (95% CI 17.2% to 46.1%). Among the 13 concordant couples, 5 men had primary syphilis (4 penile, 1 anal), 11 secondary syphilis (8 generalised rash, 3 penile, 2 anal, 1 oral lesion) and 10 early latent infections. Concordance was higher among couples where at least one partner had secondary syphilis compared with couples where neither partner had secondary syphilis (53% (9/17) vs 15% (4/26), P=0.016). Furthermore, concordance was higher among couples where one was HIV positive compared with couples where both were HIV negative (62% (5/8) vs 23% (8/35), P=0.042).

Conclusions There was an overall concordance rate of 30%. Higher concordance rates for early syphilis infection between male sexual partners were associated with HIV and secondary syphilis.

  • HIV
  • syphilis
  • gay men

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 remains a major public health problem among men who have sex with men (MSM). Syphilis rates in MSM are in many cases higher among HIV-positive men, who have disproportionately accounted for repeat syphilis infections.1 Untreated syphilis infection can lead to serious morbidity and increased HIV risk.

There are limited data on the risk of syphilis transmission between men or factors that influence transmission. Most studies on syphilis transmission are more than three decades old.2–5 Only one study, from 1983, specifically examined syphilis among male partners of men.6 In this study, 76 index men with primary or secondary syphilis had 98 male sexual contacts, with 48 (49%) of contacts diagnosed with early syphilis. No data on factors associated with transmission were presented.6

The aim of this study was to determine the concordance of early infectious syphilis between men within sexual partnerships and clinical factors associated with this.


Study setting

This was a retrospective study of routinely collected clinical data from the Melbourne Sexual Health Centre between March 2011 and April 2016. The centre consists of a walk-in STI clinic and an outpatient HIV clinic. All asymptomatic MSM were routinely screened for syphilis. MSM who presented with symptoms suggestive of syphilis were serologically tested for syphilis and those with mucocutaneous lesions routinely had a swab for Treponema pallidum PCR (Tp PCR) collected. In the outpatient HIV service, syphilis serology was performed as part of routine HIV monitoring. Patients attending the clinic completed a computer-assisted self-interview (CASI) asking whether their partner had attended on the same day, enabling partners to be identified.

Case definition

Men within sexual partnerships were identified using CASI. Male partners were those who attended the clinic together either on their day of syphilis testing, or on the day of syphilis treatment. Couples were included in the study if one or both partners were diagnosed with early infectious syphilis.

A review of medical records was performed to identify the clinical characteristics and pathology results. All syphilis infections were confirmed serologically as early syphilis: either T. pallidum seroconversion within 24 months or, in the case of a syphilis reinfection, a fourfold increase in rapid plasma reagin (RPR) titre. Some men with lesions had concurrent positive Tp PCR; however, all were also serologically confirmed cases. Two sexual health physicians (JMT and ID) independently reviewed each record to confirm the stage of syphilis infection. Men were classified as having primary, secondary or early latent syphilis. Staging was according to the Australian Public Health Laboratory Networks syphilis case classifications.7 In Australia, early latent syphilis is defined as a new asymptomatic syphilis presentation of less than 2 years’ duration. The duration of early latent syphilis infections was determined from the medical records. Couples were classified as having either concordant or discordant syphilis infection.

Laboratory methods

Serological and Tp PCR testing was performed by the Victorian Infectious Diseases Reference Laboratory. Prior to January 2016, all sera were tested with RPR (Becton Dickinson), T. pallidum particle agglutination (TPPA) assay (Fujirebio) and a recombinant total antibody ELISA immunoassay (EIA) (Trepanostika EIA, BioMerieux). Sera were also tested for T. pallidum IgM with whole-cell lysate IgM EIA (Mercia) until December 2011 when the assay was discontinued, and replaced with the Bio-Rad Syphilis IgM EIA. From January 2016, the BioMerieux EIA was replaced with a LIASON Treponema screen (DiaSorin), an automated chemiluminescence immunoassay (CLIA). Due to improved sensitivity of the laboratory syphilis testing algorithm with the CLIA assay, the RPR, TPPA and IgM assays were then only performed if sera were screened positive on the CLIA assay or there was a contemporaneously positive Tp PCR result. The Tp PCR assay used was one targeting the polA gene.8

Statistical analysis

Clinical characteristics and laboratory results for each couple were compared between the concordant and discordant couples to identify factors associated with transmission. These included: anatomical site and type of lesions (oral or anogenital lesions or generalised rash), HIV status, RPR titre, PCR result, stage of infection and whether infection was a first or reinfection. As data on sexual behaviour were not systematically collected, no behavioural associations were investigated. Associations between concordance and potential factors were examined using Fisher’s exact test. Due to the small sample size, multivariate regression analysis was not undertaken. SPSS V.22 was used for statistical analyses. This paper was checked against the Reporting of Studies Conducted Using Observational Routinely Collected Health Data guidelines.


During the study period, 43 couples (86 men) were identified where at least one partner was diagnosed with early syphilis. This included 30 couples (60 men) with discordant results and 13 couples (26 men) where both were infected with syphilis, representing a concordance rate of 30% (95% CI 17% to 46%).

Of the 56 cases of syphilis, there were 10 (18%) men with primary syphilis, 19 (34%) men with secondary syphilis and 27 (48%) men with early latent syphilis. Of the men diagnosed with early latent syphilis, 18 (67%) were of less than 1 year’s duration based on previous negative syphilis serology. The clinical characteristics and laboratory results of all men with syphilis are shown in online supplementary table 1.

Supplementary file 1

There were 13 men with a history of previous syphilis infection, 11 of whom had serological evidence of syphilis reinfection: all had a fourfold rise in RPR titre on parallel testing.

Eight of the 86 men (9%) were HIV positive. A contemporaneous CD4 count was known for seven men. The median CD4 count was 525 and 6 men had an HIV viral load of <100 copies/mL on antiretroviral therapy. Seven of the eight HIV-positive men were diagnosed with syphilis, including three with secondary syphilis and four with early latent syphilis. There were no couples where both partners were HIV positive.

Among the 13 concordant couples (26 men), 5 men had primary syphilis, 11 secondary syphilis and 10 early latent infection. The characteristics and laboratory results for couples concordantly infected are shown in table 1.

Table 1

Characteristics and laboratory results for male sexual couples concordantly infected with early syphilis

Concordance was higher among couples where at least one partner had secondary syphilis compared with couples where neither had secondary syphilis (53% (9/17) vs 15% (4/26), P=0.016). Five of the eight HIV-positive men were in syphilis concordant relationships, and three in discordant relationships. Concordance was higher among couples where one was HIV positive compared with couples where both were HIV negative (62% (5/8) vs 23% (8/35), P=0.042).


In this study of male sexual partners, the concordance rate for syphilis infection within partnerships where at least one man had early syphilis was 30%. Concordance between male sexual partners was more likely when at least one partner had secondary syphilis, or when one partner was HIV positive. To our knowledge, this is the only study to identify factors associated with a greater likelihood of concordance between men. There is only one other study published that estimates infection rates between men within sexual partnerships.6 This previous study also did not include index cases with early latent syphilis as we have.

There are a number of limitations to this retrospective study. Sample size was limited to 43 couples. We were unable to determine the number or types of specific sexual acts that took place between sexual partners or duration of sexual relationships. The direction of transmission could not be determined. Furthermore, we could not determine when transmission occurred or exclude infection of either man from a third person. Concordance may be biased upwards if men were more likely to attend the clinic because of symptoms.

The precise mechanism of syphilis transmission between MSM is not fully understood. T. pallidum can be identified in primary and secondary lesions but transmission is also believed to occur during early, asymptomatic infection. In one study, 40% of HIV-positive MSM with syphilis had oral T. pallidum detected without lesions, which supports the potential oral transmission of syphilis.9 That secondary syphilis was associated with a greater likelihood of syphilis concordance in our study supports the hypothesis that secondary syphilis is a relatively contagious stage, notwithstanding the limitations discussed above.

The association of syphilis concordance with HIV positivity in our study could have resulted from bias associated with HIV-positive MSM being over-represented among syphilis infections. But hypothetically, immune suppression associated with HIV could predispose to greater forward syphilis transmission and/or risk for acquisition. Impaired immune responses and clearance of T. pallidum have been shown in HIV-positive individuals.10 While HIV positivity was associated with a higher syphilis concordance, because of limited numbers, we could not determine which factors explained this observation. Further research is required to clarify the precise routes of transmission of syphilis between men and the role of asymptomatic treponemal shedding. Such information would help inform biomedical and other interventions aimed at curtailing the syphilis epidemic among MSM.


Afrizal and Jun Kit Tze, IT Department, Melbourne Sexual Health Centre.



  • Handling editor Jackie A Cassell

  • Contributors CKF conceived the idea for the study. MYC and SH developed the concept and planned the study. SH, MB, CSB and AR performed data collection and analysis. JMT, ID, MYC and SG provided clinical advice. EPFC and LZ provided statistical advice. JMT wrote the first draft and took responsibility for the manuscript. All authors contributed to the writing and review of the manuscript and have approved the final manuscript.

  • Funding EPFC is supported by the National Health and Medical Research Council (NHMRC) Early Career Fellowships (No 1091226). JMT is supported by the Australasian Chapter of Sexual Health Medicine (Royal Australasian College of Physicians) Research Entry Grant.

  • Competing interests EPFC reports grants from the National Health and Medical Research Council (NHMRC) and grants from Merck & Co, outside the submitted work.

  • Ethics approval Alfred Health Human Ethics Research Committee (project number: 108-15).

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