LYMPHOGRANULOMA VENEREUM (LGV) has been considered a sporadic disease in Europe and North America, occurring endemically only in parts of Africa, Latin America and Asia.1 However, at the end of 2003, there was an outbreak of LGV in men who have sex with men (MSM) in Rotterdam.2 Subsequent reports from other European cities, such as Hamburg,3,4 Paris,5,6 London,7 Stockholm,8 and Zurich,9 and from North America10 as well as Australia,11 indicated the emergence of a new epidemic in this high-risk group.
According to these reports, affected MSM suffered primarily from severe proctitis, and a high proportion was also infected with HIV. Most of these patients were infected by C. trachomatis L2b biovar, a variant that was first identified in patients from Amsterdam.12
LGV is an invasive disease caused by Chlamydia trachomatis biovars L1 to L3. Unlike other anogenital C. trachomatis infections, LGV preferably affects lymphatic tissue after invasion through an epithelial surface.1 Depending on the site of inoculation, patients usually suffer from acute inguinal lymphadenitis with abscess formation (inguinal syndrome), or acute hemorrhagic proctitis (anorectal syndrome).13 If left untreated, patients develop persistent infections with chronic inflammation, which may lead to formation of fistulae and strictures. Eventually, severe disfiguring and restricting conditions such as genital elephantiasis, esthiomene, and so called “frozen pelvis” with infertility may develop.14,15
In this article, we report about an outbreak of LGV in Vienna and analyze clinical manifestations, as well as socioepidemiologic and behavioral data. We define novel variants of the genotype L2 of C. trachomatis, evaluate homologies to previously defined European strains, and correlate the clinical symptoms with different variants of the L2 strain.
This study was conducted at the Department of Dermatology, Medical University of Vienna, and the Outpatient Center for the Diagnosis of Infectious Venerodermatological Diseases, Vienna. During the period of November 2005 through November 2006, 24 white MSM either with anorectal or inguinal symptoms and their partners were examined for C. trachomatis infection. As a control group, 10 asymptomatic HIV+MSM with similar sociodemographic characteristics were evaluated for urethral and anal C. trachomatis infection. The control patients were age-matched HIV+MSM with high-risk sexual behavior (>5 partners within previous 6 months, unprotected sexual intercourse) and anorectal swabs were obtained as preventive medical checkup routinely in our HIV outpatients’ center.
The study was performed according to the Declaration of Helsinki. All patients declared their informed consent about possible scientific use of the data obtained.
A standardized questionnaire was used in 13 out of 15 patients diagnosed for LGV after the diagnosis was established and before therapy was started to interview the patients about their sexual behavior including condom use and sexual techniques, history of sexually transmitted infections (STI), HIV infection status, number of partners within the last 6 months, and travel habits. The external anogenital area was examined to define clinical manifestations (mucopurulent or bloody anal discharge, perianal erythema, ulcer, bubo) possibly related to LGV, and a routine STI screening was performed including urethral and anal swabs for C. trachomatis testing, gram staining, gonococcal culture and syphilis serology. Symptoms reported by the patients were recorded and pain was classified by the patients as mild, moderate, severe, or none. In 5 patients with severe rectal symptoms a sigmoidorectoscopy was performed. Punch biopsies were obtained for histopathologic examination, and rectal swabs for chlamydia testing. In patients with inguinal abscess formation material was obtained by aspiration after abscess incision, and anorectal involvement examined by proctoscopy. To further evaluate extent of disease, a computer tomography (CT) scan of the pelvis was performed in these patients.
To detect C. trachomatis DNA in clinical specimens, the automated Cobas TaqMan CT system (Roche Diagnostics) and CT ProbeTec-ET (Becton-Dickinson) were used according the manufacturers’ instructions. To specify the C. trachomatis biovars, ompA-VS4 DNA fragments were amplified by PCR using primers MB4 and MF44, followed by DNA sequence analysis as described previously.16 For subtyping L2 strains, the ompA-VS1/VS2 region was analyzed by sequencing PCR amplimers obtained with primers MF21L (CGACCGTGTTTTGCAAACAGATGT) and MB22L (CATCCACATTCCCACAAAGCT). These primers include some modifications to MF21 and B22 originally described by Dean et al.16 to optimize homology to the L2 reference sequence. Sequence data were obtained by direct sequencing of PCR products from both directions using upstream and downstream PCR primers as sequencing primers. Control swabs were obtained 3 to 4 weeks after the end of the treatment period and analyzed by the Cobas TaqMan CT system for C. trachomatis DNA.
From November 2005 through November 2006, 22 patients or their partners with symptoms of proctitis, and 2 patients with inguinal purulent lymphadenitis tested positive for C. trachomatis DNA of anal swabs and inguinal aspirates, respectively. Urethral swabs were negative for C. trachomatis in all patients. Table 1 summarizes the laboratory results of these 24 subjects. C. trachomatis infection was diagnosed by DNA detection in rectal swabs or inguinal aspirates using PCR (TaqMan, Roche) and subsequent confirmation by another nucleic acid amplification test (NAAT) for chlamydia (SDA, ProbeTec, Becton-Dickinson) in all cases. Several studies have shown good agreement between different NAATs for C. trachomatis.17–20 Discrepant findings were also described and may result from cases of low bacterial load, predominantly in asymptomatic individuals. The samples from symptomatic patients in our study probably contain high amounts of bacteria, unlikely to cause discrepant results in different NAATs. To identify the genotypes, the VS4 region of the ompA gene was PCR-amplified followed by DNA-sequencing. Of the 24 samples, 15 (62.5%) were determined as biovar L2, 7 (29.2%) from patients with proctitis were C. trachomatis biovars D, F, or G, and 2 samples (8.3%) could not be classified due to lack of amplification of ompA sequences. All patients lived in Vienna, Austria, and were successfully treated with doxycycline 100 mg twice daily for 3 weeks, as clinical symptoms disappeared and control swabs were negative after the treatment period. Chlamydia testing of asymptomatic patients with demographic characteristics similar to the LGV patients revealed negative results (data not shown).
Clinical Symptoms of LGV Patients
Thirteen (86.7%) LGV patients presented with colorectal and 2 (13.3%) with inguinal disease. Primary symptoms and levels of pain are listed in Table 1. The most common symptoms included severe mucopurulent and/or bloody rectal discharge. In 5 patients with severe anorectal symptoms, sigmoidorectoscopy revealed extensive, sharply demarcated shallow ulcers covered with fibrin and purulent exudates, as well as erythema, consistent with a diagnosis of acute hemorrhagic proctitis (Fig. 1A, B). Histopathology of punch biopsies revealed dense lymphocytic infiltration of the lamina propria (Fig. 1C). The CT scan of these patients showed diffuse thickening of the rectal mucosa and enlarged regional lymph nodes without signs of perirectal abscess formation or ascites (Fig. 1D). Three patients suffering from particularly severe proctitis reported systemic symptoms, such as fever and malaise, indicating a systemic inflammatory immune response. The level of pain (no pain-0, mild-1, moderate-2, or severe-3) was classified as moderate to severe by all patients, only 2 patients were without pain and had hardly any symptom (Table 1). Inguinal lymph node enlargement was not found in any patient with anorectal symptoms.
Two patients presented with unilateral painful swelling of the groin. These patients sonographically had extensive lymphadenopathy with abscess formation (bubo, Fig. 1E). Incision resulted in substantial draining of pus. Both patients with the inguinal syndrome had neither anal discharge nor signs of anorectal involvement as controlled by proctoscopy, although a rectal swab from 1 of the patients tested positive for C. trachomatis biovar L2 (Table 1). None of the LGV patients had urethritis as judged by negative Gram stain and Chlamydia PCR of urethral smears.
In addition to the VS4 region of the ompA gene required to identify the L2 biovar, the VS1 and VS2 region was sequenced in samples from 11 LGV patients (73.3%). The obtained DNA sequences were compared with the prototype sequences of LGV genotypes L2, L2a, and L2b by BLAST algorithm in GenBank. Eight samples (72.7%) displayed nucleotide sequences identical to strain L2b (Table 1), a C. trachomatis genovariant described during a recent outbreak of LGV in Amsterdam for the first time.12 In addition, we could identify 3 new variants of biovar L2. When compared with prototype L2, one variant had a point mutation at base position 973 in VS2 of L2 reference sequence 434/Bu (M14738), where cytosine was replaced by thymidine (C973T). This codon change predicted an amino acid substitution by isoleucine for threonine (I325T) (Fig. 2). This variation was identified in 2 patients and we propose to designate this as a new genovariant L2c. Another novel variant with the proposed designation L2d was detected in 1 patient with a mutation in VS1 at base position 730 with cytosine substituted by thymine (C730T). This change predicts a change from valine to alanin (V243A) (Fig. 2). A proposed further new variant L2e was detected in another patient, with a variation at position 1427 in VS4 resulting in replacement of cytosine by thymidine (C1427T) (Fig. 2), predicted to be silent at the aa level (Thr475Thr). However, in this patient sequence data of VS1 and VS2 regions were not available due to insufficient amplification. The novel sequences were deposited in GenBank (Accession Numbers: L2c: EF460796, L2d: EF460797, L2e: EF460798).
STI History and HIV Infection
All 15 patients with LGV had a history of previous and/or concurrent STI. These included 11 patients with HIV (73.3%), 4 patients with gonorrhea (26.7%), 3 patients with syphilis (20.0%) and 2 patients with hepatitis B virus infection (13.3%) (Table 1). Eight patients (53.4%) were coinfected with gonorrhea or syphilis at the time of the diagnosis of LGV (Table 1).
The HIV status was known in 13 patients, 11 (84.6%) were HIV-positive at the time point of LGV diagnosis. More than two thirds (8 [72.3%]) of HIV-infected patients were in CDC-stages A1 or A2 with a mean number of 469.5 (±180.4) CD4+ T-cells per milliliter at the time of diagnosis. Only two of the patients currently received highly active antiretroviral therapy (HAART) (Table 1).
All patients were MSM of white origin with a mean age of 41.3 (±10.4). The socioepidemiologic questionnaire was obtained from 13 out of the 15 LGV patients. Each of them reported unprotected insertive and receptive anal sexual intercourse, regardless of their own HIV status or the HIV status of their partners. The high-risk behavior is also illustrated by the mean number of 5.7 ± 5.6 (range, 1–21) partners within the preceding 6 months (Table 1) and the limited use of condoms: 6 (46.2%) of the patients always practiced unprotected anal sex, and 7 (53.8%) subjects reported occasional condom use. Anal fisting and the use of sex toys were denied by the patients. Two of our patients who were diagnosed at the beginning of this endemic reported sexual, mostly anonymous contacts either in Germany (Hamburg and Berlin) or the United Kingdom (London) (Table 1), cities with reported outbreaks of LGV.21,22 The patients who presented later on denied sexual contacts outside Vienna, where they had presumably acquired their infection. None of the patients had traveled outside Europe during the 6 months before diagnosis.
In this study, we diagnosed a substantial number of patients with LGV, which has not been detected in Vienna in the past 30 years. Interestingly, we identified different L2 sequence variants of ompA, which may reflect LGV strains involved in the Vienna outbreak of this study that are different from L2b detected in Amsterdam.
The patients’ characteristics of those with anorectal involvement and those with an inguinal syndrome did not differ with regard to age, reported sexual practices, or the C. trachomatis L2 genotypes involved. One possibility for either rectal or inguinal LGV might be that the infection was acquired via different sexual practices, such as fisting or sharing sex toys,2 at distinct sites of the genital tract, which was denied by our LGV patients. Another mechanism for inguinal abscess formation might be a lymphatic spread of the pathogens from the anal mucosa, as the primary site of infection, to the inguinal lymph nodes. Despite the fact that all patients had practiced unprotected insertive and receptive anal sexual intercourse, urethral infections for C. trachomatis L2 were not detected. We speculate that negative results from urethral swabs might be due to transient urethritis that spontaneously resolved following transmigration of the pathogens from the epithelium to the lymphatic tissue. However, we cannot rule out the possibility that C. trachomatis occurs isolated in the ejaculate of LGV patients and might thus be transmitted by sexual intercourse, not detectable by urethral swabs.
The recent outbreak of LGV was dominated by the C. trachomatis variant L2b. This genotype was first described in samples of patients in The Netherlands,12 and subsequently found in France,23 Germany,3 Canada,10 and Australia.11 We could detect the same L2b genotype in 8 patients. However, 3 new variants of the L2 biovar were found in 4 patients, indicating different sources of infection in the outbreak in Vienna. L2b differs from L2 by 2 point mutations in ompA. The same mutations were detected in L2c and L2d. In addition, 1 more point mutation was found in L2c and L2d. L2e differs from L2b by 1 mutation in VS4; however, in this case, we had no sequence data of VS1 and VS2, which contains the 2 variant sites between L2 and L2b. The detection of additional point mutations in ompA of L2c and L2d may indicate that L2b may have been imported to Europe some time ago and since then has undergone some mutational change resulting in different regional variants. This possible local variation needs to be confirmed in further studies on LGV patients from different European areas. In 1 patient (patient 23) who had almost no symptoms, a new L2 sequence, designated variant L2d, was identified. In his infected partner (patient 24), who was also almost asymptomatic, we did not obtain sufficient amounts of DNA to perform sequence analysis. It may be speculated that he was infected with the same variant, causing atypically mild symptoms. However, due to the limited number of patients, no firm conclusions about an association of specific genovariants and distinct clinical symptoms can be drawn. A detailed analysis of the variant involved in the LGV clusters described previously9,22 may help in understanding the epidemiology of LGV.
The high coinfection rate with HIV (73.3%) and other STI (53.3%), as shown for other LGV cohorts,21–23 may be a result of serosorting, i.e., HIV-positive men preferentially engage with other HIV-positive men,24 for whom the practice of safer sex may appear less relevant. Another explanation could be that bacterial infections of the genital tract such as C. trachomatis or Neisseria gonorrhoeae are associated with higher HIV load in genital secretion,25 suggesting that LGV might also facilitate HIV transmission. It appears less likely that LGV represents an opportunistic infection with more severe symptoms in HIV-infected patients, as there was no association between the stages of the HIV infection and clinical symptoms. The pathogenetic significance of the C. trachomatis isolates in HIV infected MSM could also be demonstrated by negative results of anal swabs in asymptomatic HIV+MSM, arguing against the possibility of bystander functions of these pathogens. Alternatively, the increase of STI in HIV-infected MSM in developed countries may result from improved health conditions due to HAART, eventually resulting in higher sexual activity. Finally, HIV-infected persons are generally affected by STI at higher rates26,27 when compared with the HIV-noninfected population. Thus, HIV might tag a sexually highly active population pursuing hig-risk sexual practices.
This report illustrates the further spread of LGV within western Europe. Increased awareness and adaptation of current screening policies in European countries are urgently required to get the recent LGV outbreak under control.
1. Mabey D, Peeling RW. Lymphogranuloma venereum. Sex Transm Infect 2002; 78:90–92.
2. Nieuwenhuis RF, Ossewaarde JM, Gotz HM, et al. Resurgence of lymphogranuloma venereum in Western Europe: an outbreak of Chlamydia trachomatis
serovar L2 proctitis in The Netherlands among men who have sex with men. Clin Infect Dis 2004; 39:996–1003.
3. Meyer T, Arndt R, von Krosigk A, et al. Repeated detection of lymphogranuloma venereum caused by Chlamydia trachomatis
L2 in homosexual men in Hamburg. Sex Transm Infect 2005; 81:91–92.
4. v Krosigk A, Meyer T, Jordan S, et al. [Dramatic increase in lymphogranuloma venereum among homosexual men in Hamburg]. J Dtsch Dermatol Ges 2004; 2:676–680.
5. Halioua B, Bohbot JM, Monfort L, et al. Ano-rectal lymphogranuloma venereum: 22 cases reported in a sexually transmited infections center in Paris. Eur J Dermatol 2006; 16:177–180.
6. Herida M, Sednaoui P, Couturier E, et al. Rectal lymphogranuloma venereum, France. Emerg Infect Dis 2005; 11:505–506.
7. Macdonald N, Ison C, Martin I, et al. Initial results of enhanced surveillance for lymphogranuloma venereum (LGV) in England. Euro Surveill 2005; 10:E050127 5.
8. Berglund T, Bratt G, Herrmann B, et al. Two cases of lymphogranuloma venereum (LGV) in homosexual men in Stockholm. Euro Surveill 2005; 10:E050303 3.
9. Gebhardt M, Goldenberger D. Lymphogranuloma venereum (LGV) serotype L2 in Switzerland, 2003–2005. Euro Surveill 2005; 10:E051222 4.
10. Kropp RY, Wong T. Emergence of lymphogranuloma venereum in Canada. CMAJ 2005; 172:1674–1676.
11. Stark D, van Hal S, Hillman R, et al. Lymphogranuloma venereum in Australia: anorectal Chlamydia trachomatis
serovar L2b in men who have sex with men. J Clin Microbiol 2007; 45:1029–1031.
12. Spaargaren J, Fennema HS, Morre SA, et al. New lymphogranuloma venereum Chlamydia trachomatis
variant, Amsterdam Emerg Infect Dis 2005; 11:1090–1092.
13. Perine PL, Stamm WE. Lymphogranuloma venereum. In: Holmes KK, ed. Sexually Transmitted Diseases: MCG (Manual). New York: McGraw-Hill; 1998:423–32.
14. Gupta S, Ajith C, Kanwar AJ, et al. Genital elephantiasis and sexually transmitted infections—revisited. Int J STD AIDS 2006; 17:157–165; quiz 66.
15. Aggarwal K, Jain VK, Gupta S. Bilateral groove sign with penoscrotal elephantiasis. Sex Transm Infect 2002; 78:458.
16. Dean D, Stephens RS. Identification of individual genotypes of Chlamydia trachomatis
from experimentally mixed serovars and mixed infections among trachoma patients. J Clin Microbiol 1994; 32: 1506–1510.
17. Chernesky MA. Chlamydia trachomatis
diagnostics. Sex Transm Infect 2002; 78:232–234.
18. Gaydos CA, Theodore M, Dalesio N, et al. Comparison of three nucleic acid amplification tests for detection of Chlamydia trachomatis
in urine specimens. J Clin Microbiol 2004; 42:3041–3045.
19. Van Dyck E, Ieven M, Pattyn S, et al. Detection of Chlamydia trachomatis
and Neisseria gonorrhoeae
by enzyme immunoassay, culture, and three nucleic acid amplification tests. J Clin Microbiol 2001; 39:1751–1756.
20. Marshall R, Chernesky M, Jang D, et al. Characteristics of the m2000 automated sample preparation and multiplex real-time PCR system for detection of Chlamydia trachomatis
and Neisseria gonorrhoeae
. J Clin Microbiol 2007; 45:747–751.
21. Bremer V, Meyer T, Marcus U, et al. Lymphogranuloma venereum emerging in men who have sex with men in Germany. Euro Surveill 2006; 11.
22. Ward H, Martin I, Macdonald N, et al. Lymphogranuloma venereum in the United kingdom. Clin Infect Dis 2007; 44:26–32.
23. Herida M, de Barbeyrac B, Sednaoui P, et al. Rectal lymphogranuloma venereum surveillance in France 2004–2005. Euro Surveill 2006; 11.
24. Truong HH, Kellogg T, Klausner JD, et al. Increases in sexually transmitted infections and sexual risk behaviour without a concurrent increase in HIV incidence among men who have sex with men in San Francisco: a suggestion of HIV serosorting? Sex Transm Infect 2006; 82:461–466.
25. Cohen MS, Hoffman IF, Royce RA, et al. Reduction of concentration of HIV-1 in semen after treatment of urethritis: implications for prevention of sexual transmission of HIV-1. AIDSCAP Malawi Research Group. Lancet 1997; 349:1868–1873.
26. Buchacz K, Greenberg A, Onorato I, et al. Syphilis epidemics and human immunodeficiency virus (HIV) incidence among men who have sex with men in the United States: implications for HIV prevention. Sex Transm Dis 2005; 32:S73–S79.
27. Simms I, Fenton KA, Ashton M, et al. The re-emergence of syphilis in the United Kingdom: the new epidemic phases. Sex Transm Dis 2005; 32:220–226.