Article Text

Download PDFPDF
Original research
Clinical and laboratory features of 244 men with primary syphilis: a 5-year single-centre retrospective study
  1. Stefano Ramoni1,
  2. Giovanni Genovese1,2,
  3. Andrea Pastena1,
  4. Giovanni Casazza3,
  5. Giovanna Lunghi4,
  6. Angelo Valerio Marzano1,2,
  7. Marco Cusini1
  1. 1 Dermatology Unit, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy
  2. 2 Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Milan, Italy
  3. 3 Department of Biomedical and Clinical Sciences "L. Sacco", Universita' degli Studi di Milano, Milan, Italy
  4. 4 Microbiology and Virology Unit, Fondazione IRCCS Ca' Granda Ospedale Maggiore di Milano Policlinico, Milan, Italy
  1. Correspondence to Dr Marco Cusini, Dermatology Unit, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Via Pace, 9 - 20122 Milan, Italy; marco.cusini{at}


Background Syphilis incidence has exponentially increased in recent decades, particularly among men who have sex with men (MSM). Primary syphilis is characterised by a chancre appearing at the site of Treponema pallidum (TP) inoculation. Atypical morphological variants of syphilitic chancre are frequent. Clinical suspicion must be confirmed either by the demonstration of TP within the lesion through direct tests, such as dark field microscopy (DFM) or T. pallidum nucleic acid amplification technique (TP-NAAT), or by serological tests.

Objectives To analyse the clinical features, the sexual behaviour and the role of diagnostic tests in a cohort of men with primary syphilis in Milan.

Methods Epidemiological, clinical and laboratory data of male patients with primary syphilis seen at the STI Center of the University of Milan between 2015 and 2019 were retrospectively evaluated. Diagnosis was confirmed by at least one positive diagnostic test of either DFM, TP-NAAT or serology.

Results Among a total of 244 patients, 160 (65.6%) were MSM and 32 (13.1%) were living with HIV. One hundred twenty-four (50.8%) patients had a clinically atypical chancre. Chancres were exclusively extragenital in 30 (12.3%) patients, with MSM being more commonly affected (MSM vs heterosexuals: 16.3% vs 4.8%, respectively; p=0.012), and anal region the most frequently involved site. Chancres were multiple in 68/242 (28.1%) patients and morphologically atypical in 76/244 (31.1%). Diagnosis was obtained by (1) both serology and direct methods in 158/244 patients (64.7%), (2) serology solely in 47/244 (19.3%) and (3) direct methods solely in 39/244 (16%). DFM yielded positive results in 83/139 (59.7%) patients, while TP-NAAT gave positive results in 114/121 (94.2%) patients.

Conclusions Patients with primary syphilis frequently present with morphologically atypical chancres. Furthermore, MSM commonly exhibit extragenital involvement. A combined diagnostic approach including both direct and indirect tests is needed.

  • syphilis
  • serology
  • sexual health
  • clinical STI care
  • dermatology

Data availability statement

All data are available on reasonable request to the corresponding author (

Statistics from


Syphilis is an STI due to Treponema pallidum subspecies pallidum (TP) with mucocutaneous and systemic involvement characterised by a clinical course that encompasses three stages with asymptomatic latent periods between them.1

WHO estimated that in 2012 there were 5.6 million and 0.5 million new cases of syphilis globally and in Europe, respectively.2 Based on the Annual Epidemiological Report for 2018 of the European Center for Disease Prevention and Control (ECDC), the trend in syphilis rates has been on the rise since 2011 and up to 2017 in Europe, with more than two-thirds of cases with information on transmission category being reported in men who have sex with men (MSM).3 Consistent with these data, the Italian National Institute of Health (ISS) reported that the incidence of primary and secondary syphilis in Italy markedly increased in the period 2000–2017 compared with the period 1991–2000, particularly among MSM.4

The primary stage of syphilis is characterised by a chancre appearing at the site of TP inoculation.1 Classically, the syphilitic chancre is described as a solitary, painless, indurated, eroded dark red nodule associated with regional lymphadenopathy.5 It has been reported that about 50% of chancres are atypical in morphology, size, number, localization and/or course.6 The main atypical morphological variants include (1) papulonodular without erosion; (2) gangrenous, in which a pyogenic superinfection causes an extensive loss of substance; (3) diphtheroid, with a whitish pseudo-membrane covering the lesion; (4) rhagadiform; and (5) Follmann’s balanitis, characterised by diffuse inflammation and superficial erosions secreting a serous exudate. Other rarer clinical variants include proctitis, single/multiple erosions without infiltration and erythematous-scaling lesions. In case of multiple erosions, the chancre may take on a herpetiform aspect.6 Based on their maximum diameter, chancres may be considered dwarf (<5 mm), regular (≥5–15 mm) or giant (>15 mm).5 In approximately 40% of cases, primary syphilis can present with multiple chancres, which may result from either multiple inoculation during the same sexual intercourse or autoinoculation starting from the first chancre.7 8 Although genitalia are the most frequent localisation, extragenital sites, mainly anal region or oral cavity, may be involved, especially in MSM.9 In most cases, syphilitic chancres spontaneously heal in about 4 weeks, leaving no sequelae. Very rarely, scars and/or phimosis can occur.10 Clinical suspicion of primary syphilis must be confirmed either by the demonstration of TP within the lesion through direct tests, such as dark field microscopy (DFM) or T. pallidum nucleic acid amplification technique (TP-NAAT), or by serological tests.11

Although several single case reports of atypical chancres have been published,12–14 there are only limited data of primary syphilis deriving from extensive studies.9 15 The aims of this study are to provide an up-to-date epidemiological and clinical overview on primary syphilis presenting in Milan between 2015 and 2019 and to analyse, in a real-life setting, the role of direct diagnostic tests and serology.

Materials and methods

Patients and methods

We conducted a single-centre retrospective analysis of patients with primary syphilis who attended the STI Center of the University of Milan between 1 January 2015 and 31 December 2019.

The low number of women (n=1/245) diagnosed with primary syphilis during this period and the differences in terms of clinical manifestations between male and female patients with primary syphilis led us to include only male patients in the present study. Physical examination of patients, clinical description of chancres and DFM tests were performed by two dermatologists (SR and MC) with decades of experience in the diagnosis and treatment of STIs. Morphological variants of syphilitic chancre had been discussed and categorised before study initiation. In case of atypical lesions, SR and MC always consulted with each other to ensure consistency in reporting. All data were entered in the computerised database of our Department.

As shown in figure 1, primary syphilis was suspected in patients with genital and/or extragenital lesions presenting after sexual exposure and confirmed based on both clinical examination and laboratory criteria. Every patient was also tested for hepatitis C and HIV, while other investigations, such as urethral swabs for Chlamydia trachomatis and Neisseria gonorrhoeae and lesional swabs for herpes simplex virus (HSV) 1 and 2 or Chlamydia trachomatis serovars L1–L3, were performed depending on clinical signs and history. Patients were included in the study where there were (1) mucosal or cutaneous lesions compatible with syphilitic chancre and (2) positivity of at least one among the following tests: DFM, TP-NAAT or serology. In naïve patients, serology was considered positive in case of two different treponemal tests’ positivisation. Serological screening consisted of an ELISA for anti-TP IgM+IgG antibodies followed, in case of a positive result, by Treponema pallidum particle agglutination test to confirm the diagnosis. In patients with previous history of syphilis, serology was considered positive where there was more a than fourfold rise in rapid plasma reagin test titres. TP-NAAT was performed using a commercial CE-marked real-time PCR for the 47 kDa membrane lipoprotein of TP (tpp47 gene).

Figure 1

Diagnostic work-up of the Sexually Transmitted Infection Center of the University of Milan for patients with suspect primary syphilis. DFM, direct field microscopy; NAAT, nucleic acid amplification test.

Parameters collected included gender, age at diagnosis, sexual behaviour, HIV status, co-infections, chancre location, number of lesions, size of the lesion, chancre morphology (typical, papulonodular without erosion, gangrenous, diphtheroid, rhagadiform, Follmann’s balanitis, herpetiform, phimotic, erosive without infiltration, erythematous-scaling, proctitis), serology (positive, negative, not performed), DFM (positive, negative, not performed) and TP-NAAT (positive, negative, not performed). Due to the retrospective nature of the study, data about incubation period, lymphadenopathy and local pain were largely incomplete and, therefore, were not included.

Statistical analysis

Categorical variables are reported as count (percentage) and continuous variables as median (IQR). Statistical analyses were performed with Fisher’s exact test for categorical variables and with Wilcoxon two-sample test. Statistical significance was defined as p value ≤0.05. All statistical analyses were performed with SAS software V.9.4 (SAS Institute, Cary, NC, USA).


Patients and clinical findings

A total of 244 consecutive cases of primary syphilis diagnosed over a 5-year period were included (37 in 2015, 52 in 2016, 54 in 2017, 45 in 2018 and 56 in 2019).

Demographic, clinical and laboratory features are summarised in table 1.

Table 1

Demographic and clinical features of the 244 patients included in the study

The median age at diagnosis was 38 (IQR 30–48) years. Eighty-four (34.4%) patients were heterosexual and 160 (65.6%) were MSM. Thirty-two (13.1%) patients were HIV seropositive; HIV diagnosis was made at the same time as the syphilis diagnosis in two cases. Urethral swabs for Chlamydia trachomatis and N. gonorrhoeae were performed in 15 patients with urethral discharge, and were negative in all cases. Swabs for HSV 1/2 were performed in 22 patients due to the presence of multiple lesions reminiscent of genital herpes and swabs for Chlamydia trachomatis serovars L1–L3 were performed in six cases due to proctitis and/or marked lymphadenopathy suggestive of lymphogranuloma venereum (LGV). Two patients were co-infected by LGV and one by genital herpes caused by HSV 2. Genital warts were diagnosed simultaneously with primary syphilis in one patient. Taking into account site, morphology, number of lesions and size, 123/244 (50.4%) cases can be defined atypical for at least one of the aforementioned clinical features (figure 2).

Figure 2

Clinical features of syphilitic chancres. (A) Multiple eroded nodules on the glans and penile shaft. (B) Chancre of the upper labial mucosa. (C) Chancre of the finger. (D) Follmann’s balanitis. (E) Diphtheroid chancre of the pharynx. (F) Eroded nodule of the anal region.

Exclusively extragenital chancres were seen in 30 (12.3%) cases, and simultaneously genital and extragenital in two (0.8%) patients. The anal region was the most frequent extragenital site (n=21/244; 8.6%), followed by oral cavity (n=8/244; 3.3%), finger (n=1), intergluteal fold (n=1) and groin (n=1). Chancres were morphologically typical in 168/244 (68.9%) patients and atypical in the remaining 76 (31.1%) cases. The atypical morphological variants were as follows: papulonodular without erosion (n=27;11.1%), erosive without infiltration (n=11;4.5%), gangrenous (n=10;4%), rhagadiform (n=9;3.7%) herpetiform (n=6;2.5%), diphtheroid (n=6;2.5%), Follmann’s balanitis (n=2;0.8%), phimotic (n=2;0.8%) erythematous-scaling (n=2;0.8%) and proctitis (n=1;0.4%). Nine out of 21 (42.9%) anal chancres were rhagadiform.

Most patients (n=174/242; 71.9%) presented with a single lesion, while multiple chancres were observed in 68/242 (28.1%) cases; in two cases with Follmann’s balanitis, a numerical criterion was not applicable. Dwarf chancres were observed only in three (1.2%) patients, while lesions were regular in size in the remaining 241 (98.8%) patients.

Direct diagnostic tests and serology

Of the 139 (57%) patients who were tested with DFM, 83 (59.7%) tested positive. One hundred fourteen out of 121 (94.2%) patients tested positive with TP-NAAT. Serology was positive at first visit in 205 (84%) patients; among the 39 (16%) patients with negative serology at the first visit (pre-serological syphilis), 28 were re-tested and 15 (53.6%) became positive, while 13 (46.4%) remained negative on retesting after 2–4 weeks (table 1).

As shown in figure 3, syphilis was diagnosed only by means of serology in 47/244 (19.3%) patients. In this group of patients, direct tests were not performed in 40 patients, while 7 patients had negative TP-NAAT. For two of these patients with negative TP-NAAT, DFM was also negative. In patients with negative serology at first visit (pre-serological syphilis), DFM yielded a positive result in 19/39 (48.7%) cases, while it was negative in 8/39 (20.5%) patients and not performed in 12/39 (30.8%) patients; all 20 cases with negative DFM or where DFM was not carried out had positive TP-NAAT.

Figure 3

Diagnostic tests performed in our cohort of patients. (A) Direct tests stratified by serology results. (B) Direct tests’ results. DFM, direct field microscopy; TP-NAAT, T. pallidum nucleic acid amplification test.

In line with the diagnostic work-up carried out at our centre (figure 1), no patients with positive DFM (n=83) were tested with TP-NAAT and all patients with negative DFM (n=56) were tested with TP-NAAT. Among the negative DFM, only two (3.6%) patients were also negative with TP-NAAT, and the diagnosis was made with serology.

Comparisons between MSM versus heterosexuals and HIV-seropositive versus HIV-seronegative patients

As shown in table 2, extragenital sites were significantly more common in MSM as compared with heterosexuals (16.3% vs 4.8%, respectively; p=0.012). However, no significant differences in terms of lesion site were observed between HIV-seropositive and HIV-seronegative patients. Furthermore, no significant differences between MSM versus heterosexuals or HIV-seropositive versus HIV-seronegative patients were found in terms of median age, number of lesions (multiple vs single), clinical presentation (atypical vs typical) or serology at diagnosis (negative vs positive).

Table 2

Clinical and laboratory findings stratified for sexual behaviour and HIV status


In recent decades, the incidence of syphilis in Europe has exponentially increased.3 Our study of a large cohort of patients with primary syphilis showed syphilis in nearly 0.5% of all patients who attended between 2015 and 2019 the STI Center of the University of Milan, without significant differences in annual incidence.

The median age of our cohort was 38 (IQR 30–48) years, consistent with 2018 ECDC data, which reported the largest number of syphilis cases in the age group 25–44 years3 and with the data published by the ISS, which reported that the median age of patients with primary and secondary syphilis diagnosed in the period 1991–2017 was 36 (IQR 29–45) years.4

Recent surveillance data from Europe, England and Italy3 4 16 showed a significantly higher prevalence of syphilis among MSM. In line with the aforementioned data, MSM represent two-thirds of our study population. Furthermore, a steady increase in syphilis diagnoses was reported in the last years among MSM. According to the latest annual official data on STI released by Public Health England, syphilis diagnoses rose between 2018 and 2019 among MSM (3%; from 5684 to 5875) in England.16

About 10% of our cohort was living with HIV and most patients were on anti-retroviral treatment with suppressed viral load, and, consequently, were not considered at risk of transmitting HIV.17

It can be conjectured that oral–genital or oral–rectal intercourses, which are frequently reported as the only unprotected sex practices among MSM,18 might have played an important role for perpetuating syphilis transmission. In agreement with this hypothesis, several studies suggested that the increase of bacterial STIs such as syphilis among MSM could be due, in part, to HIV-exclusive risk reduction strategies, that is, engaging in oral sex over anal sex.18 19

The importance of a comprehensive screening with a multiplex NAAT for TP, C. trachomatis serovars L1–L3 and HSV 1/2 in patients with genital or anal ulcers suspect for primary syphilis was borne out by finding three patients co-infected in our cohort (two LGV and one HSV).

Atypical morphological features were found in about 30% of patients. Unlike Duncan et al 8 who described multiple lesions in about 40% of a cohort of 69 patients with genital chancres, we found only 30% multiple chancres. In comparison with the 5%–10% prevalence of extragenital chancres reported in the literature,20 21 we found higher percentage (12.3%) in our series. Nevertheless, it is conceivable that extragenital chancre prevalence may remain underestimated due to the painless nature and the difficulty in visualising lesions at particular anatomical sites. The expected association between extragenital chancre location and homosexual behaviour is likely related to anal sex.9

The high percentage of pre-serological syphilis (16% of our cohort) underlines the importance of direct diagnostic methods, especially DFM, to provide an early point-of-care diagnosis and reduce the spreading of the infection due to diagnostic delays. On the other hand, the diagnosis of primary syphilis was made only by serology in approximately 20% of patients.

It is well established that TP-NAAT has a higher sensitivity than DFM22 23; although we did not perform a real comparison between the two methods, our data are consistent with the literature, showing that almost all patients with negative DFM were positive when tested with TP-NAAT. TP-NAAT is suitable also for the oral cavity, and this method can also detect other pathogens simultaneously. However, it is more expensive and takes time for the results. Thus, DFM, which is cheaper and faster, maintains an important role in the point-of-care diagnosis of genital and erosive/ulcerative lesions of primary syphilis: while a negative result does not exclude a diagnosis of syphilis, a positive result provides prompt diagnosis for index case management and facilitates early prophylactic treatment of exposed sex partners.24 For extragenital lesions, notably in the oral cavity, NAAT remains the gold standard for lesion diagnosis and is not affected by the presence of saprophytic spirochaetes.11

In contrast to the results by Brischetto et al,25 who found that only 2/61 (3%) patients with positive TP-NAAT had negative serology at the time of diagnosis and suggested that TP-NAAT adds little clinical value over serology for the diagnosis of syphilis, around 15% of our patients showed negative serology at the first visit and were diagnosed thanks to a direct test. Thus, our study suggests that when primary syphilis is suspected, the best practice is that of a direct test together with serological investigations.26

While direct tests are essential for confirming diagnosis of primary syphilis in a timely manner, serology still has an important role to play in detecting people with early latent infection or those with occult primary syphilis.27

Interestingly, it can be speculated that early treatment reduces the possibility of a subsequent seroconversion, with about 50% of our patients with negative serology at the diagnosis remaining negative 1 month after treatment.

This study has potential limitations, primarily due to its retrospective nature. Moreover, the percentage of extragenital chancres identified might have been underestimated due to the difficulty to perform proctoscopy in our centre. Furthermore, we decided to not include women in our study due to the very low number of female patients with primary syphilis (n=1) observed at our STI centre in the 5-year period of the study. This finding may be explained by the high proportion of difficult-to-detect syphilitic chancres in females (such as the intravaginal or cervical),28 by females’ preference to attend gynaecologists (and not STI centres) in case of genital lesions and by a low index of suspicion among other specialties.

STI programmes should increase their focus on primary syphilis, which is often unrecognised and untreated.27 Indeed, primary syphilis plays a key role in sustaining syphilis transmission since the majority of cases most likely occur through sexual contact with syphilitic chancres.24 29

Primary syphilis must be suspected in any case of cutaneous or mucosal lesion that appears within 12 weeks after sexual exposure, especially among core transmission groups of MSM, in which a sexual network of infectious patients and susceptible sex partners maintains a persistent reservoir of infection.27

The epidemiological trend that started in the 2000s shows every sign of persistence2 30 and the rarest clinical manifestations will undoubtedly reappear. Secondary and also primary syphilis has to be considered a ‘great imitator’. Therefore, the expectation that syphilitic chancres will be ‘typical’ is unsafe, and syphilis must be considered in the differential diagnosis of all lesions, no matter how ‘atypical’. We must remain ever vigilant.

Key messages

  • Men who have sex with men represent an ‘at risk’ population for primary syphilis, which is frequently extragenital in this cohort.

  • Syphilitic chancres may be atypical for at least one parameter among morphology, dimension, site and number in about half of the patients.

  • Primary syphilis diagnostic work-up requires a combined approach including both direct tests, namely dark field microscopy and Treponema pallidum nucleic acid amplification technique, and serology.

Abstract translation

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 availability statement

All data are available on reasonable request to the corresponding author (

Ethics statements

Patient consent for publication


We acknowledge Derek and Jane Freedman for their support in English language editing.



  • SR and GG are joint first authors.

  • Handling editor Jo Gibbs

  • SR and GG contributed equally.

  • Contributors MC, SR and GG conceived the study and discussed the structure of the manuscript. AVM contributed to the concept of the study and helped with data analysis. SR, GG and MC wrote the manuscript, analysed the patient data and designed the concept of table and figures in close exchange with GC and AVM. GC provided biostatistical support. GL was responsible for laboratory testing. All authors contributed to the discussion and critically revised and edited the manuscript. All authors read and approved the final 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.

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.