Objectives To examine lymphogranuloma venereum (LGV) trends over time among men who have sex with men (MSM) visiting the Amsterdam sexually transmitted infection (STI) clinic; to investigate anal LGV symptomatology; and to examine the positivity and characteristics of anorectal and inguinal LGV.
Methods We included MSM consultations from whom a swab (from anorectum, bubo or an genital ulcer) was taken for Chlamydia trachomatis (Ct) screening. Anorectal swabs were taken from all MSM who reported receptive anorectal intercourse in the preceding 6 months. Ct positive samples were further tested with a pmpH PCR to identify L-genovars. Patient symptoms, clinical and anoscopic inflammatory signs, and STI co-infections were noted; Gram-stained anorectal mucosal smears were examined.
Results Between January 2005 and June 2012, 48 570 consultations among MSM were conducted. In 3628/35 650 visits, anorectal Ct infections were diagnosed, including 411 anal LGV (1.2%). Moreover, 65/1649 genital ulcer swabs were Ct positive; 10 were inguinal LGV (0.6%) Since January 2011 a significant increase in the positivity of LGV occurred (p<0.0001). 89 (27.2%) anorectal LGV cases were asymptomatic. HIV prevalence among anorectal LGV cases was significantly higher (p=0.008) than among inguinal LGV cases. STI co-morbidity in anorectal LGV cases remained invariably high during the study period.
Conclusions Since January 2011, LGV positivity in MSM consultations in Amsterdam has risen significantly. The great majority comprise anal LGV; inguinal LGV is rare. Anal LGV is asymptomatic in a quarter of cases. In all MSM with anal Ct infections LGV should be excluded, irrespective of symptoms or inflammatory signs.
- CHLAMYDIA TRACHOMATIS
- EPIDEMIOLOGY (CLINICAL)
- LYMPHOGRANULOMA VENEREUM
Statistics from Altmetric.com
The epidemic of lymphogranuloma venereum (LGV) among men who have sex with men (MSM) in industrialised countries first became apparent in 2003.1 In a report from Rotterdam, the Netherlands, 13 MSM (most of them HIV infected) with severe LGV anorectal infections were presented.2 Subsequently, similar LGV proctitis cases were reported from major cities throughout Europe.3 LGV among MSM is associated with high-risk behaviour, which is reflected in the large proportion of sexually transmitted co-infections, like HIV (in up to 82.2% of the cases),4 hepatitis C and previous syphilis infections.5 LGV is an invasive ulcerative sexually transmitted infection (STI) caused by Chlamydia trachomatis (Ct) biovar L.6 Acute anorectal LGV is characterised by anorectal cramps, pain, bloody discharge and constipation. If left untreated, chronic disease can lead to irreversible anorectal strictures causing soiling, pain, constipation, and megacolon.7 ,8 LGV may also increase the risk for acquisition and transmission of other STIs, including HIV and blood-borne infections like hepatitis C virus.9 ,10
The LGV epidemic is ongoing and new cases are being reported every year, mainly from large urban centres throughout Europe, the USA and Australia.3 ,11 ,12 The vast majority of diagnosed LGV infections in this epidemic were anorectal; only few inguinal LGV infections have been reported. This clinical–epidemiological discrepancy leaves the mode of transmission between men unexplained.3 ,13
There is some debate whether anorectal LGV may have an asymptomatic course. We earlier reported that in a considerable proportion of consultations (up to 40.0% of the cases), patients reported few symptoms or had no physical abnormalities.5 ,14 In contrast to our findings, studies from the UK showed that only a small proportion of LGV cases were asymptomatic (6.0–17.2%).15 ,16
In this article we report on the positivity trends of LGV, the symptomatology of anorectal LGV, the relative frequencies of anorectal and inguinal LGV, and the differences in demographics and concurrent STIs between anorectal and inguinal LGV cases, among MSM visiting the STI clinic in Amsterdam, the Netherlands, between January 2005 and June 2012.
Study population and setting
The STI outpatient clinic of the Public Health Service in Amsterdam, the Netherlands, offers free-of-charge examination and treatment for sexually transmitted diseases (STDs). In the clinic an MSM client is defined as a man who reports sexual intercourse in the preceding 6 months with one or more men. In 2011 more than 40% of all MSM visiting STI clinics throughout the Netherlands were visitors of the STI clinic in Amsterdam.17 ,18 In Amsterdam, MSM are offered screening for gonorrhoea, syphilis, hepatitis B, chlamydia and HIV as described previously.19 Each visitor is seen by a trained nurse, who records possible STI related symptoms, inspects the ano-genital region, and collects samples for routine diagnostics. If the visitor reports receptive anal sex in the previous 6 months, anoscopy is performed. Anorectal symptoms (anorectal pain, discharge, soiling, blood loss, urge or constipation) and signs of anorectal mucosal inflammation (oedema, purulent discharge, mucosal fragility, ie, bleeding on manipulation, and ulceration) are recorded in the electronic patient file.
Swabs collected from the urethra, from the anorectal mucosa on anoscopy and from ano-genital ulcers, as well as bubo aspirates, were screened for Ct by nucleic acid amplification testing (Aptima Combo 2 system; GEN-PROBE, San Diego, California, USA). Since 2005, all anorectal mucosal, ulcer or bubo samples positive for Ct were tested further with a pmpH based in-house real-time PCR to discriminate between LGV and non-LGV genotypes, as described previously.20 ,21 If the pmpH test was inconclusive (mainly due to insufficient DNA), the result was considered negative for LGV. If an anorectal mucosa swab or an anorectal ulcer swab was positive for an LGV genotype, anorectal LGV was diagnosed. If a penile ulcer swab or a bubo aspirate was positive for an LGV genotype, inguinal LGV was diagnosed. Samples of the urethra positive for Ct were not tested further for LGV.
If the visitor reported anorectal symptoms or if signs of anorectal mucosal inflammation were observed on anoscopy, Gram-stained anorectal mucosal smears were made and examined by a trained laboratory technician to exclude gonorrhoea and to count the number of polymorphonuclear leucocytes per high power microscope field (PMNL/hpf). If more than 10 PMNL/hpf were recorded, a diagnosis of presumptive non-specific proctitis was made and doxycycline therapy commenced before the definite results became available.
HIV antibody rapid testing (Determine 1/2; Abbott Laboratories, Abbott Park, Illinois, USA) was offered to all clients, except to visitors already diagnosed with HIV. Reactive samples were confirmed by line immunoassay (Inno-Lia HIV I-II Score; Innogenetics, Ghent, Belgium). When a client declined to be tested, the HIV status was reported as ‘unknown’. Syphilis serologic screening to detect both recent and past infections was performed with a Treponema pallidum particle agglutination assay (TPPA, Fujirebio, Chuo-ku, Tokyo, Japan), followed by the rapid plasma reagin (RPR, bioMérieux, Marcy l'Etoile, France) in case of a TPPA titre of ≥1:80. In case of genital ulceration, dark-field microscopy to detect spirochaetes and PCR on ulcer swabs for T pallidum and herpes simplex virus (HSV type 1 and 2) were performed.22 All clinical findings, diagnoses and subsequent treatment were recorded in an electronic patient database along with patient characteristics and information on sexual behaviour.
Data from the electronic patient file were extracted for all MSM who visited the STI clinic in Amsterdam for screening between January 2005 and June 2012. Visitors could be diagnosed with LGV on more than one occasion. For the analysis of the symptomatology of anorectal LGV, men with anorectal gonorrhoea were excluded, since both infections may cause symptoms.
Fisher's exact test and the Mann–Whitney U test were used to compare groups. We analysed whether LGV positivity in all MSM consultations increased over the last six quarters, from January 2011 to June 2012, with the χ2 test for trend. Trends in the annual proportions of LGV cases with STI co-morbidity (HIV, TPPA and gonorrhoea) were analysed, for the whole study period, with the χ2 test for trend. All tests were two-sided and were considered significant if the p value was less than 0.05. All statistical analyses were performed using SPSS V.19. Since we used anonymised routine data, no ethical clearance for this study was needed.
Consultations of MSM
Between January 2005 and June 2012, 48 570 consultations involving an MSM visitor were done (figure 1). At 35 650 visits an anorectal swab was obtained. We diagnosed 3628 (10.2%) anorectal Ct infections, including 411 anorectal LGV infections (positivity 1.2%). In 1649 consultations genital ulcers were sampled. In total 65 were Ct positive; of these, 10 were L-biovar Ct positive. Therefore the inguinal LGV positivity among MSM was 0.6%.
Consultations with LGV infections
In the study period 411 anorectal LGV infections were diagnosed in 365 MSM and 10 inguinal LGV infections in 10 individual MSM. Of the men with an anorectal LGV infection, 46 MSM had a repeat infection; 34 were infected twice, 7 three times, and 5 more than three times. No individual had a dual anorectal and inguinal LGV infection.
Characteristics of anorectal and inguinal LGV consultations are compared in the online supplementary table S1. Overall, the consultations with anorectal and inguinal LGV were predominately of Dutch nationality (respectively, 69.6% and 50.0%; p=0.3) and the median age was, respectively, 41 (IQR 30.3–48.8) and 43 (IQR 35.0–46.0) years (p=0.9). The prevalence of HIV infection was significantly higher among MSM with anorectal (82.7%) compared to those with inguinal LGV (50.0%, p=0.008). TPPA serology was positive in, respectively, 68.1% and 60.0% (p=0.7). Concurrent gonorrhoea, urogenital and/or anorectal, was detected in 24.6% of anorectal LGV cases. No concurrent gonorrhoea was diagnosed among consultations with inguinal LGV (p=0.1).
LGV infections 2005–2012
The LGV positivity per quarter fluctuated between 0.1% and 2.5% in the period from 2005 to 2009 (figure 2, see online supplementary table S2). From 2010 onwards the LGV positivity was stable at around 1% per quarter, but more recently a sharp increase was observed (χ2 test for trend over six quarters 2011–2012; p<0.0001). In absolute numbers, more LGV cases (n=64) have been found in the first half of 2012 compared to all of 2011 (n=51).
Clinical presentation of anorectal LGV infections
For the analysis of clinical signs and symptoms of anorectal LGV, we excluded men diagnosed with concurrent anorectal gonorrhoea (n=84) since both infections may cause anorectal symptoms or inflammatory signs. In the remaining 327 anorectal LGV cases, anorectal symptoms were reported in 195 consultations (59.6%) (table 1). In 167 cases (51.1%), anorectal inflammatory signs were identified on proctoscopy. In 124 (37.9%) consultations, both inflammatory signs and anal symptoms were present; and in 89 (27.2%) consultations with an anal LGV infection, neither anorectal symptoms nor inflammatory signs were recorded. Overall there were 3628 consultations with a positive Ct diagnosis. When consultations with a concurrent anorectal gonorrhoea (n=640) were excluded, in 2193/2988 (73.4%) consultations with an anorectal Ct infection, neither anal symptoms nor inflammatory signs were recorded.
In 203/238 (85.3%) anorectal LGV cases with anorectal symptoms or inflammatory signs, light microscopic examination of Gram-stained anorectal smears revealed ≥10 PMNL/hpf (table 1).
Concurrent STI among LGV cases
Of 421 LGV cases, 345 (81.9%) were known to be HIV positive or tested HIV positive at the time of consultation; of 17 (4.0%) cases, the HIV status was unknown. Of all LGV cases, 286 (67.9%) were TPPA seropositive, indicating a previous syphilis or current syphilis infection at the time of consultation, and 101 (24.6%) had concurrent urogenital or anorectal gonorrhoea. During the study period, the proportion of men with LGV who also had HIV, TPPA positivity or gonorrhoea, remained consistently high, and no significant downward trends over time of co-infections among LGV cases were detected (χ2 test for trend for TPPA: p=0.8; for gonorrhoea: p=0.3); the proportion of cases with HIV co-infection increased over time (χ2 test for trend; p=0.04; figure 3 and see online supplementary table S3).
Our data show a significant increase of LGV positivity in MSM consultations at the Amsterdam STI clinic over the six quarters since January 2011. In absolute numbers we found more LGV in the first half of 2012 compared to all of 2011. The rise of LGV positivity cannot be attributed to changes in testing policy, since no changes were made during the study period. A similar increase in the number of LGV cases has been reported in the UK.23 Between October 2004 and March 2011, 14 250 specimens were received for confirmatory testing, of which 1665 were found to be LGV positive24; a third of these cases had been diagnosed since January 2010. The LGV endemic among MSM in Western society has been going on for almost 10 years and the recent increasing trends in Amsterdam and the UK indicate that the epidemic is not controlled and screening should be further intensified.
We found a high prevalence of concurrent STI (gonorrhoea, HIV, syphilis) among LGV cases. During the study period the proportion of concurrent STI among LGV cases remained high, with a small but significant increase in the proportion of concurrent HIV. This negates the idea that the LGV epidemic spreads from the core group of high-risk MSM identified at the start of the epidemic into a broader population. Two recent publications from France25 and the Netherlands26 described LGV in women caused by the L2b genovar characteristic for the epidemic in MSM. In conclusion, with only two female cases reported in the 9 years since the discovery of the epidemic, it seems that the L2b LGV epidemic is still largely confined to a population of high-risk MSM. Nonetheless, a spread of LGV into the broader population is still a possible threat.
With only 10 inguinal LGV cases over 7.5 years, the overt predominance of anorectal LGV infections (n=411) in Amsterdam remains unexplained. The preponderance of anorectal LGV cases has also been reported in other European studies.16 It has been suggested that anorectal LGV could be transmitted through sharing of toys and fisting during group sex activities,2 but this could not be confirmed in large systematic studies afterwards.5 ,27 Few cases of urethral LGV infection have been reported.15 ,28 Missed urethral LGV infections, and possibly pharyngeal LGV29 could form contributing sources responsible for anorectal LGV infections. As MSM with urethral Ct infections are not routinely tested for LGV, perhaps many urethral LGV diagnoses are missed. This could explain the discrepancy between the numbers of reported inguinal and anorectal LGV cases reported.
Compared to anorectal cases, significantly fewer cases of inguinal LGV were HIV infected and they less often had concurrent STIs. This could indicate that inguinal LGV cases show less high-risk behaviour compared to cases with anorectal LGV. Larger numbers of patients, recruited in multicentre studies, are required to analyse in greater depth these differences in risk factors between men with LGV at different anatomical locations.
The IUSTI/WHO guideline for LGV recommends screening for anorectal infections in MSM irrespective of anorectal symptoms.6 Yet, in many settings LGV testing is performed only in case of inflammatory anorectal signs and/or symptoms. Based on the low proportion of asymptomatic anorectal LGV cases found in the UK,15 ,16 the British Association for Sexual Health and HIV recommends LGV testing in case of an anorectal syndrome; that is, the diagnosis is based on clinical suspicion after the exclusion of other aetiologies of proctitis.30 Here, we found that in 27.2% of the anorectal LGV cases neither symptoms were reported nor were inflammatory signs noticed on routine anoscopy. This is in agreement with earlier reports by our group indicating that a sizeable proportion of anorectal LGV patients are asymptomatic.5 ,14 Asymptomatic anorectal LGV could be missed if screening is performed exclusively in men with anorectal symptoms and/or anoscopic proctitis. In settings where anoscopy is not performed routinely, the chance to miss LGV is even greater. In 40.4% of the anorectal LGV cases described here, no symptoms were reported on consultation. Light microscopic examination of anorectal smears is an inexpensive, easy and relatively sensitive test for the syndromic management of LGV infections in MSM.5 In 85.3% of the studied anorectal LGV cases with symptoms or signs, we detected more than 10 PMNL/hpf. Anorectal smears as screening tool could be of value, especially in settings where expensive serovar-L specific nucleic acid amplification testing is unavailable or too costly. Likewise, an HIV positive sero-status is a strong predictor for LGV in MSM with a proven anorectal Ct infection and can be considered in a syndromic algorithm.5
A limitation of our study is the lack of LGV testing of urethral Ct positive samples. Currently we are investigating the prevalence of urethral LGV infections in greater detail. Due to the retrospective nature of the study, consistent data on risk parameters like condom use and the number of partners were unfortunately not available for the whole study period.
In conclusion, the LGV epidemic among MSM in Amsterdam is ongoing, and seems to have accelerated in the past year and a half, with a higher number of cases detected in the first half of 2012 compared to the whole of 2011. To date, anorectal LGV is still an STI confined to MSM engaging in high-risk behaviour. Anorectal LGV forms the vast majority of cases; inguinal LGV is rare. The mode of transmission of LGV within the MSM community remains enigmatic; the discrepancy in numbers of anorectal and inguinal infections suggests that many LGV infections are missed. Control methods might have failed due to missed LGV cases like asymptomatic infections and infections at sites which are not routinely tested for LGV, such as the urethra.
Compared to anorectal LGV cases, cases with inguinal LGV are significantly less often co-infected with HIV and they tend to have fewer co-infections like gonorrhoea. Lower co-morbidity in inguinal LGV cases could be a reflection of lower-risk behaviour compared to anorectal LGV cases and requires further study. LGV is asymptomatic in a quarter of the consultations, therefore LGV should be excluded in all MSM with anorectal Ct infections, irrespective of anorectal symptoms or inflammatory signs.
Since 2005, lymphogranuloma venereum (LGV) has been prevalent among men who have sex with men (MSM) in Amsterdam, and the incidence has risen significantly since 2011.
LGV is associated with a high prevalence of sexually transmitted infections, including HIV, in 80% of the cases, suggesting that LGV remains confined to a high-risk population.
The great majority of LGV cases are anal LGV; inguinal LGV is rare, but perhaps many diagnoses are missed.
Since LGV is asymptomatic in 27.2%, it should be excluded in all MSM with anal Chlamydia trachomatis infections, irrespective of anal symptoms or inflammatory signs.
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.
Files in this Data Supplement:
- Data supplement 1 - Online supplement
Contributors NdV analysed the data and wrote the manuscript; MvR gathered the data from the electronic patient file and wrote the manuscript; MSvdL supervised the epidemiological analysis and wrote the manuscript; HdV designed and supervised the study, and wrote and approved the final manuscript.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
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.