Article Text

Original research
Lymphogranuloma venereum genovariants in men having sex with men in Italy
  1. Antonella Marangoni1,
  2. Claudio Foschi1,
  3. Federico Tartari2,
  4. Valeria Gaspari2,
  5. Maria Carla Re1
  1. 1 Microbiology, DIMES, University of Bologna, Bologna, Emilia-Romagna, Italy
  2. 2 Dermatology, University Hospital of Bologna Sant'Orsola-Malpighi Polyclinic, Bologna, Emilia-Romagna, Italy
  1. Correspondence to Dr Claudio Foschi, Microbiology, DIMES, University of Bologna, Bologna, Italy; claudio.foschi2{at}


Objectives Lymphogranuloma venereum (LGV) is an STI caused by Chlamydia trachomatis serovars L1-L3. In Europe, the current epidemic is caused mainly by L2b genovariant, although increasing cases associated with other L2 variants have been reported. Here, we assessed the distribution of rectal LGV genovariants among men having sex with men (MSM) in Italy.

Methods From 2016 to 2020, all the anorectal swabs collected from MSM attending the STI Clinic of St. Orsola-Malpighi Hospital in Bologna and positive for C. trachomatis were stored. LGV infection was confirmed by a pmpH PCR, and, subsequently, a fragment of the ompA gene was amplified and sequenced. Sequences were aligned to reference strains representing different LGV variants.

Results LGV cases accounted for one-third of all chlamydial rectal infections with a total prevalence of 4.1% (76/1852). Total number of LGV cases per year remained constant. LGV was mainly found in symptomatic patients (>65%), older than 30 years, with a high burden of other STIs (63.7% HIV-positive, 35.5% with concurrent rectal gonorrhoea, 19.7% with early syphilis). A decreasing trend in HIV-LGV co-infection was noticed over time. Three main LGV genovariants were detected (L2f, 46.1%; L2b, 23.0%; L2-L2b/D-Da, 16.9%), together with other known L2b variants (mainly L2bV2 and L2bV4). Two novel L2b ompA variants with non-synonymous single-nucleotide polymorphisms were found. Over time, the percentage of L2f cases dropped gradually, with a significant increase in L2-L2b/D-Da cases (p=0.04).

Conclusions In our area, LGV is endemic among MSM with different circulating genovariants. Active surveillance and genotyping programmes are needed to reduce re-establishing of LGV infection.

  • lymphogranuloma venereum
  • Chlamydia trachomatis
  • homosexuality

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. The ompA sequences of the two new variants found in this study were deposited in GenBank under the accession number MT707652 (AM/8/17) and MT707653 (CF/185/18).

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.


Lymphogranuloma venereum (LGV) is a systemic STI caused by Chlamydia trachomatis serovars L1-L3.1 2

Since 2003, outbreaks of LGV have spread across Europe and other high-income countries, mainly among HIV-positive men who have sex with men (MSM), leading to an increase awareness for this disease.3–6 Current LGV cases are essentially manifested by proctitis (‘anorectal primitive syndrome’), characterised by anorectal ulceration, tenesmus, anal pain and mucous or bloody discharge.7 8

The current LGV epidemic in western Europe is caused by the L2 biovar of C. trachomatis with the predominance of the L2b genovariant.9 10

Recently, however, some countries have been reporting increasing cases associated with other L2 variants.11–14 As an example, between March 2009 and November 2011, during an LGV outbreak in Madrid, Spain, two different L2 variants were detected, namely L2b and L2/L2f.11 Similarly, in France, during a 6-year period (2010–2015), the prevalence of L2 genotype has increased over time, and four new genetic L2b ompA variants were described.12

Moreover, between 2017 and 2018, a Portuguese study reported an ongoing LGV outbreak (25 cases detected) caused by a recombinant strain with a rather unique genome make up, characterised by a L2-L2b/D-Da hybrid profile.15 This outbreak, mostly affecting HIV-positive MSM engaging in high-risk sexual practices, was mainly characterised by a symptomatic anorectal syndrome.

This whole scenario highlights the need to improve LGV genotyping, as well as to revise laboratory methods to target LGV-specific genome loci for detecting, monitoring and controlling LGV outbreaks.

Currently, in Italy, no active surveillance programme for LGV detection is available and many laboratories still do not have genetic tools to discriminate LGV from other C. trachomatis genovars. For this reason, so far, only a few data on the characteristics of LGV epidemic have been reported in Italy,2 16–19 and almost no study has investigated the circulating LGV genovariants in depth.20

Therefore, in this study, we assessed the characteristics of rectal LGV cases and the distribution of LGV genovariants in an MSM population attending an STI Outpatients Clinic in the North of Italy, in a 4-year period.


Study population and sample collection

From 1 June 2016 to 31 May 2020, all the anorectal swabs (E-Swab, Copan, Brescia, Italy) collected from MSM patients attending the STI Outpatients Clinic of St. Orsola-Malpighi University Hospital in Bologna (Italy) and positive for C. trachomatis were stored.

The criteria for collecting samples were based on clinical symptoms (eg, anal pain, anorectal discharge, tenesmus) or on sexual risk behaviours (ie, undefined history of unsafe anal intercourse).

Personal data, information about anorectal symptoms and data about other STIs (gonorrhoea, HIV, HCV, syphilis) were recorded from LGV-positive patients.

Laboratory investigations

LGV cases were laboratory-confirmed in a two-step protocol. Anorectal swabs were first processed by Versant CT/GC DNA 1.0 Assay (Siemens Healthcare Diagnostics, Terrytown, USA), a commercial real-time PCR test simultaneously detecting the presence of C. trachomatis and/or Neisseria gonorrhoeae DNA.21 C. trachomatis-specific PCR primers and probe were designed to target the GenBank nucleic acid sequence of the 7.5 kb cryptic plasmid. This molecular assay proved to be extremely sensitive in the detection of extragenital C. trachomatis infections, with a limit of detection of 10 copies/mL.21

When the samples yielded positive amplifications for C. trachomatis, LGV infection was confirmed with a pmpH real-time PCR, as previously described.22 This assay allows the distinction between L and non-L chlamydial serovars, on the basis of a unique 36 base pair deletion region of pmpH gene that only occurs among LGV strains.22

The samples were then stored at – 20°C until further identification of LGV genovariants. A fragment of the ompA gene (ca 1070 bp) was subsequently amplified by hemi-nested PCR using a previously described protocol.23 24 Part of this amplicon (ca 1000 bp) was sequenced in both directions and aligned to reference sequences from GenBank representing different LGV variants: L2a (GenBank accession number: AB915594); L2 (AM884176); L2b (AM884177); L2c (EF460796); L2d (EF460797); L2e (EF460798); L2f (EU676181); L2g (EU676180); L2bV1 (JX971936); L2bV2 (KU518893); L2bV3: (KU518894); L2bV4 (KU518892); L2-L2b/D-Da (MN094864).

Statistical analysis

All statistical analyses were performed by using GraphPad Prism software ( Analysis of variance test was used to compare quantitative data, whereas categorical data were analysed with χ2 test or Fisher’s exact test. Statistical significance was determined at p<0.05.

Data availability

The ompA sequences of the two new variants found in this study were deposited in GenBank under the accession number MT707652 (AM/8/17) and MT707653 (CF/185/18).


Rectal infection prevalence

During the study period, a total of 1852 rectal swabs for C. trachomatis detection were collected from MSM. Overall, the prevalence of chlamydial rectal infections was 12.2% (226/1852). LGV cases (n=76) accounted for about one-third (33.6%) of all chlamydial infections with a total prevalence of 4.1% (76/1852). Although the total number of LGV cases per year remained constant (about 19 cases/year), a marked decrease in LGV prevalence over time was noticed (ranging from 10.3% in the first year analysed to 2.8% in the last). In parallel, rectal gonorrhoea prevalence gradually decreased from 22.4% (period 2016–2017) to 10.7% (period 2019–2020), with a total prevalence of 14.1% (261/1852).

LGV cases

During the study period, a total of 76 cases of rectal LGV infection were detected. The mean age of positive patients was 38.6±10.6 years (min 20–max 65; median: 37). Clinical symptoms were present in 68.4% (52/76) of patients. Among the LGV-infected MSM, 63.7% (44/69) were HIV-positive, whereas for seven of them (9.2%) the HIV status was unknown. In five patients (5/44; 11.3%), the diagnosis of HIV was simultaneous to LGV infection. Excluding the 5 naïve individuals, 28 HIV-positive subjects (28/39; 71.8%) had a well-controlled infection (no detectable viral load), whereas in the remaining cases the viral load ranged between 20 and 25 000 copies/mL (mean±SD: 2849±7413.5). Moreover, in our series, 35.5% (27/76) of LGV-positive patients showed a concurrent rectal gonorrhoea and eight of them (10.5%) were positive for HCV infection. Early and latent syphilis was detected in 19.7% (15/76) and 7.9% (6/76) of LGV cases, respectively.

Considering the trend of LGV cases over time (table 1), no significant variations in clinical (eg, rectal symptoms) or epidemiological (eg, mean age, association with other bacterial STIs) characteristics were noticed. We only observed a decreasing trend, even though not significant, in HIV-LGV co-infection over the years (p=0.07).

Table 1

Clinical and epidemiological characteristics of LGV-positive patients stratified by year analysed

Moreover, a peak in the presence of rectal symptoms among LGV-positive patients was found in the period 2018–2019 (85%). Nevertheless, in the same year, a marked increase (50%) in LGV gonorrhoea co-infection was noticed compared with other periods, making it difficult to discriminate whether this increase is mainly due to LGV or gonorrhoea.

LGV genovariants

Of the 76 LGV-positive samples, 65 (85.5%) were successfully sequenced and genotyped. An L genovar was confirmed for all specimens by ompA sequencing. Most samples (30/65; 46.1%) had ompA sequences identical to L2f; 15 specimens (15/65; 23.0%) showed sequences identical to L2b and 11 (11/65; 16.9%) identical to the hybrid L2-L2b/D-Da. Moreover, we identified three L2bV2 (4.6%), three L2bV4 (4.6%), one L2bV1 (1.5%). In the remaining two cases, we found two novel L2b ompA variants that had non-synonymous single-nucleotide polymorphisms in the variable domain II, compared with nucleotide sequence of L2b reference strain (L2b/UCH-1/proctitis; GenBank accession no. AM884177). One variant (AM/8/17; GenBank accession no: MT707652) was characterised by a C→A point mutation at position 464 (His165Asn). The other one (CF/185/18 GenBank accession no: MT707653) showed a A→G substitution at position 462 (Asn164Ser).

As shown in table 2, no significant associations between LGV variants and clinical/epidemiological features (eg, presence of rectal symptoms, association with other STIs) were found.

Table 2

Clinical and epidemiological characteristics of LGV-positive patients stratified by LGV variants detected

Interesting data emerged from the analysis of the distribution of LGV variants over time (figure 1). Although not statistically significant (p=0.2), an increase of L2b cases was observed over the years (11.1% in the period 2016–2017 and 10.5% in 2017–2018 vs 25% in 2018–2019 and 31.5% in 2019–2020). Conversely, the percentage of L2f cases tended to drop gradually, going from 55.5% in 2016–2017 to 26.3% in 2019–2020 (p=0.09). The first case of the hybrid L2-L2b/D-Da was detected in May 2018, with a significant increase of cases over time (0% in the period 2016–2017 vs 31.5% in the period 2019–2020; p=0.04).

Figure 1

Distribution of lymphogranuloma venereum (LGV) variants over time (2016–2020), according to sequencing results of the ompA gene. Percentages are calculated as follows: number of cases belonging to a specific variants/total number of LGV cases in 1-year period analysed (reported in brackets). L2b variants: in 2016–2017 one L2b new variant; in 2017–2018 one L2bV4, one L2b new variant and one L2bV1; in 2018–2019 one L2bV4 and two L2bV2; in 2019–2020 one L2bV2 and one L2bV4. nt, not typable.

Finally, it is worth mentioning that all L2bV2 variants (n=2) were detected between April and July 2019.


In this study, we mainly assessed the distribution of rectal LGV variants in a MSM population in a 4-year period (2016–2020). To the best of our knowledge, this report represents one of the most comprehensive studies about the circulation of LGV genovariants in Italy.2 16 17 20 25

At first, we confirmed that the total prevalence of rectal LGV remained significant in the MSM population (4.1%), indicating that this infection is still endemic in both Italy and other European countries.18 26–28

In line with other national/international reports, LGV cases were mainly detected in symptomatic patients (>65% of symptomatic proctitis), older than 30 years, with a strong association with other STIs, namely HIV (63.7% of co-infections), rectal gonorrhoea (35.5%) and early syphilis (19.7%).18 28

In our setting, LGV prevalence tended to decrease over time, going from 10.3% in the 2016–2017 to 2.8% in 2019–2020. Changes in screening programme policies could probably explain this finding. We can speculate that in the first years, rectal swabs were mainly collected from high-risk subjects (eg, HIV-positive MSM with high number of sexual partners reporting fisting and sex toys sharing; sexual contacts with LGV-positive people), whereas, more recently, rectal sampling was presumably also performed in lower-risk subjects. Indeed, over time, we observed a reduction in LGV-HIV co-infections, indicating a possible shift of LGV in lower-risk groups of MSM.

Other interesting data emerged from LGV genotyping, obtained by ompA gene sequencing. We found a co-circulation of three main genovariants (ie, L2f, 46.1%; L2b, 23.0%; the hybrid L2-L2b/D-Da, 16.9%), together with other known L2b variants (mainly L2bV2 and L2bV4).

In agreement with recent epidemiological investigations, we confirmed a significant and constant presence of LGV cases due to variants other than L2b, in our country.11 12

As for other European countries,11 12 these data suggest that the co-circulation of three different LGV genovariants in Bologna, Italy, could be the result of three independent introductions.

Bologna is a high-density urban area, characterised by a significant presence of people coming from different European countries, as well as by a high flow of immigrants from worldwide regions. Thus, the creation of specific sexual networks and different dynamics of transmission could lead to a complex epidemiological scenario, where different LGV genovariants co-exist and spread in the population.

Nevertheless, further in-depth studies are needed to identify the distinct introduction pathways for the different LGV genovariants in Italy.

Interestingly, we report, for the first time, the presence of the hybrid L2-L2b/D-Da outside Portugal. This novel genovariant was described during a proctitis-associated LGV outbreak mainly in HIV-positive MSM, starting from April 2017.15 This hybrid strain presents a classical LGV chromosome backbone, while harbouring most of the major antigen (major outer membrane protein (MOMP)) from a non-LGV strain. Thus, it has been hypothesised that the genomic characteristics of this unique recombinant LGV variant may have implications on its transmission, tissue tropism and virulence.15

In our series, the hybrid L2-L2b/D-Da was found from May 2018, with increasing cases over time.

As previously suggested,15 the circulation of this strain among high-risk groups of MSM (eg, multiple partners, engaging in chemsex, involvement in international sexual networks) could have facilitate the spread of the hybrid L2-L2b/D-Da abroad.

Moreover, during the study, we found two novel L2b ompA variants never described before (ie, AM/8/17 and CF/185/18): both were characterised by non-synonymous single-nucleotide polymorphisms, leading to amino acid changes in variable domain II of the MOMP. This site has been already described as a common antigenic domain for C. trachomatis.1 12 14

Patient characteristics (eg, presence of symptoms, association with other STIs) did not differ with regard to the different LGV genovariants. These results are concordant with those of Peuchant et al,12 whereas disagree with those of Rodriguez-Dominguez et al, showing less aggressive symptoms in L2/L2f-infected patients compared with L2b-positive subjects.11

When looking to the presence of LGV cases over the years, we found significant changes in the distribution of LGV genovariants. We observed an increase of L2b and L2-L2b/D-Da cases, with a contemporary reduction of L2f prevalence. Besides the presence of peculiar sexual networks and transmission dynamics, this aspect could also be related to a different fitness, selection pressure and pathobiology of the various genovariants.

Our report provides a strong contribution for the worldwide picture of LGV molecular epidemiology, allowing to a deeper comprehension of the circulating genovariants and the associated clinical/epidemiological features. Moreover, our results underline the need to implement measures such as screening, active surveillance and genotyping programmes, to reduce/avoid re-establishing of LGV and prevent long-term complications.29 Finally, this study constitutes a warning for public health preventive strategies, considering the alarming STI burden found in the MSM population.30

The lack of clinical and epidemiological data among LGV-negative subjects is a major limitation of this work. Therefore, additional studies will help to better describe LGV epidemiology in our setting, as well as to assess LGV prevalence and predictors among different subgroups of patients (eg, asymptomatic subjects, high-risk/low-risk groups).

Moreover, further investigations, including the heterosexual population and other anatomical samples (ie, urogenital and pharyngeal specimens) will be needed to better understand LGV transmission dynamics and to set a tailored approach for LGV prevention and control.

Key messages

  • In Italy, lymphogranuloma venereum (LGV) is endemic among men having sex with men (MSM) with different circulating genovariants, mainly L2f, L2b and the hybrid L2-L2b/D-Da.

  • There is a need for active surveillance and genotyping programmes to reduce re-establishing of LGV infection.

  • Public health preventive strategies are urgently needed considering the alarming STI burden found in the MSM population.

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 relevant to the study are included in the article or uploaded as supplementary information. The ompA sequences of the two new variants found in this study were deposited in GenBank under the accession number MT707652 (AM/8/17) and MT707653 (CF/185/18).

Ethics statements

Ethics approval

The Ethical Committee of St. Orsola-Malpighi Hospital approved the study protocol (78/2017/U/Tess) and all the subjects gave written informed consent.


The authors would like to thank Dr Gianluca Cafasso for providing excellent technical support during the study.


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.


  • Handling editor Jason J Ong

  • Contributors AM, CF and MCR conceived and designed the experiments. VG and FT enrolled the patients and collected samples. AM and CF performed the experiments and analysed the data. AM, CF and MCR wrote the paper. All the authors read, reviewed 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.