Article Text
PDF
Abstract
Background Chancres, the hallmark of primary syphilis, are classically described as single, painless ulcers at the site of Treponema pallidum inoculation. We aimed to determine the frequency of painful or multiple anogenital lesions of primary syphilis among men, whether there was concurrent herpes simplex virus (HSV) infection and whether HIV status altered clinical presentations.
Methods This study was conducted among men with T. pallidum PCR-positive lesions, attending a clinic in Melbourne, Australia, between 2009 and 2014. Lesions were also tested with HSV PCR, and syphilis serology undertaken.
Results 183 men with T. pallidum PCR-positive primary anogenital lesions were included. 89% were men who have sex with men, and 10.9% were heterosexual. 38 men (20.8%) were HIV positive. Anal lesions were more common in HIV-positive men (34.2%) than in HIV-negative men (11.6%). Primary lesions were frequently painful (49.2%) or multiple (37.7%), and infrequently associated with HSV (2.7%). Of 37 men with both painful and multiple primary lesions, only 8% had concurrent HSV. Presentation was not significantly altered by HIV status.
Conclusions Primary syphilis lesions are often painful and/or multiple in the absence of herpes coinfection, and may be clinically misdiagnosed.
- GENITAL ULCERS
- SYPHILIS
- PCR
Statistics from Altmetric.com
Background
Globally, syphilis remains a major and resurgent public health problem with high incidence rates in many settings, including among men who have sex with men (MSM).1 ,2 Left untreated, syphilis can lead to further transmission, morbidity and enhanced HIV transmission.3
Primary syphilis is characterised by the presence of a chancre, which has been classically described as a single, indurated painless ulcer at the site of Treponema pallidum inoculation. However, in clinical practice, the lesions of primary syphilis can present atypically. Chancres may, for example, present as multiple and/or painful ulcers, which are features that would usually suggest an alternative diagnosis of genital herpes simplex virus (HSV) rather than primary syphilis.4–6 The extent to which lesions of primary syphilis may be multiple or painful is not clear, as to our knowledge, there has been no previous published study that has specifically aimed to ascertain this using sensitive T. pallidum PCR testing, while excluding concurrent HSV coinfection.
The aim of this study was to describe among men with T. pallidum PCR-positive anogenital lesions of primary syphilis, the extent to which lesions were painful or multiple, whether these were attributable to concurrent genital HSV infection and whether the clinical presentation was altered by concurrent HIV infection.
Methods
Study setting
This study was undertaken between November 2009 and October 2014 at the Melbourne Sexual Health Centre, the main public sexually transmitted disease clinic in Melbourne, Australia. The clinic operated a walk-in sexually transmitted infection (STI) clinic and an outpatient HIV service. Patients with anogenital symptoms suggestive of a possible STI were triaged into the STI clinic for assessment and management by a specialist sexual health physician. Physicians administered treatment for syphilis or herpes on the day of presentation based on their presumptive clinical diagnosis.
Case definition
Men with anogenital lesions were routinely tested for syphilis and genital HSV using T. pallidum PCR and HSV PCR with a swab collected from the lesions. In addition, men had serological testing for syphilis. Specimens were obtained for dark ground microscopy (DGM) examination for motile treponemes in selected cases, where clinicians deemed sufficient lesion exudate could be obtained. In the outpatient HIV service, serological screening for syphilis was routinely performed with the blood tests taken for HIV monitoring.7 Men were included in the study if they had primary anogenital syphilitic lesions that were positive for T. pallidum on PCR, at least one serological test for syphilis was reactive and HSV PCR testing from the lesion undertaken. A flow chart illustrating the inclusion process is shown in figure 1.
Inclusion criteria flowchart. HSV, herpes simplex virus; Tp, Treponema pallidum.
The clinical characteristics of primary lesions, including reported pain from lesions, tenderness of lesions on physical examination, number of lesions and treatment prescribed at the time of presentation, were recorded in the clinic electronic record. The number of lesions noted on clinical examination was categorised into single lesions versus multiple lesions. Single lesions included solitary typical chancres, ulcers, macules, papules, erosions and fissures. Multiple lesions included cases where more than one lesion was present. This included a small number of cases, where there was a wider area of dermatosis. The clinical characteristics were compared between HIV-positive and HIV-negative men.
Two sexual health physicians (JMT, ID) independently reviewed the electronic record for each case to confirm it fitted the definition of a primary syphilitic lesion. Where agreement could not be reached, a third physician (MC) adjudicated the diagnosis. All cases were consistent with the 2014 Centre for Disease Control (CDC) primary syphilis case classifications. Laboratory diagnoses were consistent with both the CDC and the Australian Public Health Laboratory Network case classifications for syphilis.8 ,9
All patients were classified as having either a first episode of primary syphilis or primary syphilis reflecting reinfection following a previous syphilis infection. Cases were considered as first episodes of primary syphilis where there was a positive T. pallidum PCR result from a mucocutaneous lesion, and the serology was positive on rapid plasma reagin (RPR), ELISA immunoassay (EIA) or T. pallidum particle agglutination assay (TPPA) testing with no previous history of syphilis. Cases were considered as reinfections where the patient had a positive T. pallidum PCR result, and the RPR titre had risen fourfold (two dilutions) after an earlier episode of successfully treated syphilis.
Laboratory methods
Laboratory testing was performed by the Victorian Infectious Diseases Reference Laboratory (VIDRL). The T. pallidum PCR assay used was a VIDRL in-house TaqMan real-time PCR assay targeting the polA gene, an assay in routine diagnostic use since 2004.10 Specimens were tested in duplicate with a threshold cycle cut-off of 38. DGM was performed on site by laboratory staff of the Microbiology Diagnostic Unit, University of Melbourne, experienced in T. pallidum identification by DGM.
All cases had EIA total antibody, TPPA and RPR performed. Fourteen cases did not have EIA IgM performed. In every case, the EIA (total and IgM), TPPA and RPR results were from testing of the same sera, which were collected at or closest to the date of sampling the T. pallidum PCR-positive lesions, (range: between 12 days before and 16 days after T. pallidum PCR sampling). Sera were tested using RPR (Macro-Vue RPR card; Becton Dickinson, Sparks, Maryland, USA), TPPA (Serodia TPPA; Fujirebio, Tokyo), a recombinant total antibody EIA (Trepanostika TP recombinant;11 Bio-Merieux, Boxtel, the Netherlands) and whole-cell lysate IgM EIA (Mercia Syphilis IgM; Microgen Bioproducts, Camberley, Surrey) until December 2011 when the assay was discontinued, and replaced with the Biorad Syphilis IgM EIA. This combination of four different serological tests was used as part of an algorithm to improve sensitivity of syphilis detection.
Testing for HSV was undertaken using a VIDRL in-house multiplex real-time TaqMan PCR assay targeting HSV 1 and 2 glycoprotein B that also detects varicella zoster virus and cytomegalovirus. This assay was validated on the previous assay in use by VIDRL.12 ,13
Statistical analysis
The characteristics of primary lesions—their location, number of lesions and pain/tenderness—were compared between HIV-positive and HIV-negative men. The results of laboratory investigations were also compared according to HIV status.
As TPPA and EIA total antibody markers usually remain positive after the first syphilis infection, TPPA and EIA total antibody results were analysed for first syphilis infections only. RPR and EIA IgM results were analysed separately for first syphilis infections and syphilis reinfections.
Treatment history was ascertained as an indication of what diagnoses the sexual health physician had considered: the proportion of men who were treated for syphilis with benzathine penicillin or treated for herpes with antivirals on the day of presentation was determined.
Categorical variables were compared using the χ2 test and ORs, while continuous variables were compared using the Mann–Whitney U test.
The details and protocol of this study were submitted to the Alfred Hospital Research Ethics Committee. Ethical approval was granted (project number: 119/13).
Results
Patient characteristics
There were a total of 183 men with T. pallidum PCR-confirmed primary syphilis lesions included in the study: 163 (89.1%) were MSM and 20 (10.9%) were heterosexual. The median age among men was 35 years.
Thirty-eight men (20.8%) were HIV positive: 37 MSM and 1 heterosexual man. Among men who were HIV positive, the median CD4 count was 640 (range 19–959). Twenty-seven (71.1%) were on antiretroviral therapy, and 20 (74%) of these had a viral load of <50 copies/mL.
Clinical characteristics
The clinical manifestations of primary syphilis among the men according to HIV status are shown in table 1. The 183 men had T. pallidum PCR-positive lesions at 185 sites: penile (n=152), anal (n=30), scrotal (n=2) and oropharyngeal (n=1). There were two men who had anal lesions and a second T. pallidum PCR-positive lesion at another body site: one penile and one oropharyngeal. The proportion of men who had anal lesions was significantly higher in HIV-positive men compared with HIV-negative men (34.2% vs 11.6%, p=0.0008; OR 3.95, 95% CI 1.70 to 9.14). All heterosexual men had penile lesions with no anal lesions.
Clinical characteristics of Treponema pallidum PCR-positive primary anogenital syphilis lesions among men by HIV status
Of the 183 men, 49.2% (OR 0.70, 95% CI 0.34 to 1.44) had painful/tender primary lesions, 37.7% (OR 0.91, 95% CI 0.44 to 1.89) had multiple primary lesions and 20.2% (OR 1.07, 95% CI 0.44 to 2.57) had painful/tender and multiple lesions. Only 2.7% had a concurrently positive HSV PCR result from the same lesion(s). Of the 37 men who had both painful and multiple primary syphilitic lesions, only 8% had a concurrent positive HSV PCR. Cases with a positive HSV PCR result are detailed in online supplementary table S1.
Of the 30 men who had primary syphilis at the anus, 29 also had anal swabs collected for Chlamydia trachomatis testing by strand displacement assay. Four men tested positive for rectal chlamydia, none of whom had lymphogranuloma venereum-associated genotypes.
The proportion of primary lesions that were painful/tender was similar between HIV-positive and HIV-negative men (42.1% vs 51.0%, p=0.3). The proportion of primary lesions that were multiple was also similar between HIV-positive and HIV-negative men (39.5% vs 36.5%, p=0.8).
Pathology results
The results of laboratory testing for men in the study according to HIV status are detailed in table 2. DGM was performed on specimens obtained from 118 primary lesions with 74 (62.7%) of these having motile treponemes consistent with T. pallidum.
Laboratory results for men with Treponema pallidum PCR-positive primary anogenital syphilis lesions according to HIV status
For first syphilis infections, the proportion of men in the study who had positive serology results were: TPPA 96.7%, EIA total antibody 88.7%, RPR 79.5% and EIA IgM 75.2%. HIV-negative men were more likely to test positive for TPPA than HIV-positive men. For syphilis reinfections, the proportion of men who had positive serology results for RPR and EIA IgM were 90.6% and 17.9%, respectively. HIV-negative men were more likely to have positive EIA IgM. In 14 cases, EIA IgM was not performed: 4 in HIV-positive men and 10 in HIV-negative men.
Treatment
On the day of presentation, 71% of men received treatment for presumptive syphilis prescribed by the sexual health physician. This included 59% who were treated for syphilis with no antiviral for herpes and 12% treated for both syphilis and herpes. The remaining 29% men were not treated for syphilis. This included 13% who were treated for herpes only and 16% who received no treatment for either syphilis or herpes. All men with positive DGM received syphilis treatment on the day of attendance (table 3). Among the 44 men who had negative DGM result, 30 received syphilis treatment on the day of attendance. Among the 65 men with no DGM performed, 31 received treatment for syphilis on the day. Men who presented with a single lesion were significantly more likely to be treated for syphilis on the day of presentation (OR 1.95, 95% CI 1.02 to 3.73). Those who presented with painless lesions were marginally more likely to be treated for syphilis on the day of presentation (OR 1.89, 95% CI 0.99 to 3.62).
Number of patients receiving treatment for syphilis on the day of first presentation
Discussion
In this study of T. pallidum PCR-confirmed cases of primary anogenital syphilis in men, tender or painful lesions were common, as were multiple lesions, in the absence of HSV coinfection. Painful or tender primary lesions were present in about half of the men, while about one-third had multiple lesions. These were seen in both HIV-positive and HIV-negative men, with no significant difference in the frequency of painful or multiple primary lesions between the two groups. Nearly, one-third of the men were not presumptively treated for syphilis on the day of presentation: treatment was less likely to be given if there were multiple or painful lesions present.
Primary syphilis is commonly described as a single, painless lesion.14 ,15 Previous papers have reported multiple primary syphilis lesions in 47% of cases and painful lesions in 25%–35% of cases.4–6 However, these studies had smaller numbers of subjects, diagnoses were made without the benefit of T. pallidum PCR testing, and two instances were from the pre-HIV era.4 ,5 Our study identified primary syphilis lesions using T. pallidum PCR, which is substantially more sensitive than DGM.16 The lower yield of T. pallidum from DGM in our study is consistent with this. Previous evidence for differences in the clinical presentations of primary syphilis between HIV-positive and HIV-negative men has been mixed.17–20 There were no differences in the rates of painful or multiple lesions in our study between HIV-positive and HIV-negative groups.
Most of the primary lesions included in this study were genital; however, in 16% of cases, the site of the lesion was anal. The significantly higher rate of anal lesions in HIV-positive men compared with HIV-negative men might be explained by either greater receptive sex or higher rates of unprotected receptive anal sex among HIV-positive MSM because of serosorting.21 Around one in six men had primary syphilis following a previous syphilis infection, with a higher rate of syphilis reinfection among HIV-positive men.
A strength of this study is that in addition to the use of sensitive T. pallidum PCR, all cases had HSV PCR testing and contemporaneous serological testing for syphilis, thus, minimising the potential for false positive T. pallidum PCR results. There are a number of limitations to consider when interpreting the data. We have only included men with T. pallidum PCR-positive lesions—it is possible that some men with minimal or very atypical signs of primary syphilis may not have been tested with T. pallidum PCR. The physicians administering treatment at this clinic were all specialist sexual health physicians, experienced in the management of syphilis and herpes. It is likely that painful and multiple anogenital lesions would be even more likely to be diagnosed as herpes in other settings, where testing for syphilis by T. pallidum PCR is not readily available, and healthcare providers have less experience in the management of STI.
A significantly lower rate of IgM EIA positivity was observed in HIV-positive men compared with HIV-negative men; however, some men were not tested for IgM, and larger studies are required to verify this. No direct comparison of the two IgM assays used in this study could be performed, after the first assay became unexpectedly unavailable in Australia. Eleven per cent of men with first primary syphilis infections had negative EIA total antibody results. Furthermore, 18.6% of men who had either a first primary infection or a primary reinfection were negative on RPR testing, indicating that neither traditional nor reverse syphilis screening algorithms are perfect.22
A central principle that has underpinned the efforts to control syphilis is earlier detection and treatment of infectious syphilis cases to prevent further transmission of infection. This has included attempts to improve the frequency of screening in high-risk MSM, including HIV-positive MSM.7 ,23 As the lesions of early syphilis are highly infectious, public health control measures should include health promotional messages to high-risk populations encouraging recognition of potential symptoms of syphilis, as well as education of healthcare providers to improve awareness and accuracy of diagnosis of syphilis lesions. One challenge is that for many primary care providers, syphilis is still an uncommon clinical presentation, and misdiagnosis can lead to delays in treatment.24 The findings of our study indicate that health promotion to populations at risk of syphilis and training of healthcare providers managing syphilis should highlight the atypical way in which primary syphilis can present. For healthcare providers, this should include advice to consider T. pallidum PCR testing, where available, and syphilis serology in patients at risk for syphilis presenting with anogenital lesions, where a clinical diagnosis of anogenital herpes is considered.
Key messages
Treponema pallidum PCR is useful for the diagnosis of anogenital lesions, and may be positive prior to any serological markers.
Primary syphilis lesions are frequently painful/multiple in the absence of herpes simplex coinfection, and may be clinically misdiagnosed.
The frequency of painful or multiple lesions in primary syphilis is not significantly altered by HIV status.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Data supplement 1 - Online table
Footnotes
A subset of this data was presented in poster format at the Royal College of Pathologists of Australasia, Pathology Update Conference, Melbourne, Australia February 2015.
Handling editor Jackie A Cassell
Contributors JMT conceived the idea for the study with DEL. JMT, DEL, MC and CKF planned the study. DEL and FA contributed pathology data. ID contributed clinical data. JMT, MC and ID provided opinion on syphilis staging. All authors contributed to the writing and review of the manuscript.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.