| |
Lymphogranuloma Venereum in the United Kingdom
|
| |
| |
Clinical Infectious Diseases Jan 1, 2007;44:26-32
Helen Ward,1 3 Iona Martin,2 Neil Macdonald,1 Sarah Alexander,2 Ian Simms,1 Kevin Fenton,1 Patrick French,4 Gillian Dean,5 and Catherine Ison2
1HIV and STI Section and 2Sexually Transmitted Bacteria Reference Laboratory, Health Protection Agency Centre for Infections, 3Department of Infectious Disease Epidemiology, Imperial College London, and 4Mortimer Market Centre, Camden Primary Care Trust, London, and 5Claude Nicol Centre, Brighton and Sussex University Hospitals, Brighton, United Kingdom
ABSTRACT
Background. Over the past 2 years, lymphogranuloma venereum (LGV), caused by L serovars of Chlamydia trachomatis, has emerged as a significant problem among men who have sex with men (MSM). We report on, to our knowledge, the largest case series of LGV to date, with detailed epidemiological and clinical characteristics of the epidemic in the United Kingdom.
Methods. A national diagnostic service and surveillance system was established in October 2004. Cases were confirmed by the presence of C. trachomatis and an LGV serovar (L1, L2, or L3) from genotyping. For confirmed cases, an enhanced surveillance questionnaire was sent to the clinician.
Results.
Through February 2006, a total of 327 cases of LGV were confirmed. Cases were diagnosed across the United Kingdom, with the majority from London (71%) and Brighton (13%). Case reports were received for 282 MSM. The majority (96%) had proctitis, many with severe local and systemic symptoms. There was a high level of coinfection with human immunodeficiency virus (76%), hepatitis C (19%), and other sexually transmitted infections (39%). Nine cases of human immunodeficiency virus infection were diagnosed around the same time as LGV. Most cases were acquired within the United Kingdom, although patients with early cases were more likely to report contacts in The Netherlands.
Conclusions. We found a significant burden of this once-rare sexually transmitted infection among MSM in the United Kingdom. LGV may be contributing to the epidemic of human immunodeficiency virus infection by facilitating transmission. Further control efforts are required, including awareness campaigns, continued detailed surveillance, and expanded chlamydia testing among MSM.
"....This report is the first national picture of LGV in MSM and, to our knowledge, is the largest case series published to date. We have found a significant burden of this once-rare sexually transmitted infection among MSM in the United Kingdom.... The infection is present across the United Kingdom but is concentrated in areas with substantial numbers of MSM living with HIV infection.... Infection appears to have been acquired primarily through unprotected anal intercourse, often with multiple partners. The majority of patients were coinfected with HIV, and some patients had apparently acquired HIV infection and LGV at approximately the same time. High levels of coinfection with bacterial sexually transmitted infections and hepatitis C were also reported, similar to findings in The Netherlands.......A total of 228 (87%) of 263 patients with LGV cases presented because of symptoms....There was a high level of coinfection with HIV, hepatitis, and other sexually transmitted infections....The men reported a median of 3 sexual partners in the previous 3 months (range, 0-201 partners; mean, 9.6 partners)....Eighty patients (71%) reported meeting new partners at sex-on-premises venues or at sex parties, and 26 (23%) reported meeting new partners via the Internet. A small number of men reported occupations that could link widespread sexual networks, including sex work and travel-related work. The majority of men (201 [88%] of 252) reported unprotected anal sex; 188 (75%) reported unprotected receptive anal intercourse, and 139 (55%) reported unprotected insertive anal intercourse. Fisting was reported by 32 (32%) of 99 men, and use of sex toys was reported by 15 (19%) of 78, but there was a large amount of missing data for detailed sexual behavior.....To inform control measures, we need further information on the distribution of infection in MSM. We are currently conducting case-finding exercises in a small number of clinics to measure the prevalence of urethral and rectal chlamydia, the proportion of these strains that are LGV serovars, and the range of symptoms. Results from this study will help to inform future screening and diagnostic guidelines. Until these results are available, we recommend extended testing for chlamydia, including testing of rectal samples for men with proctitis and others who may have been exposed".
Introduction
Before 2003, lymphogranuloma venereum (LGV) was considered to be a "tropical" disease, rare outside of resource-poor countries. It is caused by Chlamydia trachomatis serovars L1-3 and classically causes inguinal syndrome [1]. In the past 2 years, LGV has emerged as a significant problem among men who have sex with men (MSM) in Europe and North America. After an initial case report from Rotterdam, The Netherlands, in 2003 [2], other outbreaks were reported in sexual networks of MSM in other European cities [3-7] and, subsequently, in North America [8, 9]. Although the term "outbreak" has been used to describe the rapid increase in case detection, some authors have questioned whether it is merely a rediscovery of an endemic disease [10, 11]. All case and cluster reports have documented the high level of coinfection with HIV, which has been attributed to particular sexual networks in which some MSM may preferentially chose partners of the same HIV status as themselves ("serosorting") [12]. However, it is also possible that the re-emergence of LGV cases is the result of a different clinical spectrum of disease in the context of HIV infection. In this article, we describe, for the first time, the detailed epidemiological and clinical characteristics of cases identified during the 18 months of our enhanced surveillance program and explore the extent to which this reflects an outbreak or an endemic disease. In addition, we compare cases by HIV status and discuss whether the high level of overlap is likely to be attributable to behavior or biology.
METHODS
In October 2004, the Health Protection Agency (HPA) Centre for Infections (London, UK) set up an enhanced surveillance system for LGV in response to the case reports from The Netherlands [13]. This included a reference laboratory diagnostic service and reporting system. An alert was issued through a letter to clinicians in genitourinary medicine and microbiologists that included a briefing about recent LGV cases in Europe, details of the investigation algorithm, and a summary of the enhanced surveillance protocol. This alert was synchronized with the release of a leaflet warning MSM about LGV and describing the symptoms, as well as a sector briefing, developed and distributed by the Terrence Higgins Trust Gay Men's Prevention Team, that was featured widely in the gay press. In December 2004, after 5 cases were confirmed, a national incident team was established that included microbiologists and epidemiologists from the national Centre for Infections, local public health specialists, clinicians, voluntary sector representatives, and a press officer. Clinicians were advised to perform a test for chlamydia for any man reporting sex with other men who had any signs of inguinal syndrome or proctitis. If chlamydia was confirmed in the local laboratory using a standard test, the sample could be forwarded to the Sexually Transmitted Bacterial Reference Laboratory for confirmation of C. trachomatis infection and, if test results were positive, genotyping to determine if the isolate was an LGV-associated serovar. The HPA reference service tested rectal specimens from MSM with anorectal symptoms (typically proctitis and rectal discharge) or urethral specimens from patients with inguinal lymphadenopathy and contacts of individuals with LGV who were known to be positive for C. trachomatis. Serological testing for C. trachomatis has been used in Europe and can suggest the possibility of LGV, but it does not confirm cases because of a lack of specificity and has not been used in the United Kingdom as part of the case definition. A small number of stored chlamydia specimens from patients who had presented with symptoms suggestive of LGV during the previous 12 months were tested.
The specimens received were predominantly from rectal sites but also included fresh dry swabs, unprocessed swabs in transport buffer, and residues from processed nucleic acid amplification test specimens. Dry swabs were hydrated with 500 L of phosphate-buffered saline and agitated on an orbital shaker at 150 rpm for 1 h. DNA was extracted from the samples using either the automated Roche MagnaPure (Roche Diagnostics) with a DNA isolation kit 1 or by a manual extraction procedure using the QIAamp Viral RNA Mini Kit (Qiagen). Both extractions were performed according to manufacturers' instructions.
The chlamydial status of specimens was determined using a Centers for Disease Control and Prevention in-house real-time PCR assay (C. Cheng, personal communication). This was run as a duplex, containing primers targeting the C. trachomatis cryptic plasmid, as well as primers targeting the human ribonuclease P gene that acted as an internal control. All reactions were performed using a Rotorgene-3000 (Corbett Research).
From October 2004 through September 2005, genotyping of the samples with results positive for C. trachomatis was performed using the methods of Lan et al. [14, 15]. After September 2005, samples were genotyped using the real-time PCR method of Morre et al. [16]. This PCR spans a deletion present in the pmpH gene of LGV-associated serovars and, consequently, generates a positive result only in LGV-associated serovars of C. trachomatis. All specimens that were confirmed to be LGV positive using this method were then retrospectively genotyped by the method of Lan et al. [14, 15], as described above.
The case definition used by the HPA is capable of confirming the presence of C. trachomatis and an LGV serovar (L1, L2, or L3) by genotyping. Cases confirmed to be LGV were reported to the local microbiologist and to the LGV surveillance team at the Centre for Infections. The latter issued an enhanced surveillance questionnaire for completion by the clinician to collect demographic information, medical and sexual history, and details of risk factors and relevant sexual contacts and networks in the United Kingdom and elsewhere. Data from the enhanced surveillance and laboratory were entered into a secure database. These were exported into Stata software, version 8 (Stata) for analysis and reporting. Simple univariate analysis was performed to compare HIV-positive individuals with HIV-negative individuals, using 2 tests, with statistical significance reported at the 5% level. All data were collected through routine diagnostic and surveillance public health functions; consequently, ethical committee approval and individual patient consent were not sought.
RESULTS
Diagnostic testing. From 4 October 2004 through 28 February 2006, the Sexually Transmitted Bacterial Reference Laboratory received 1408 specimens for LGV diagnostic testing. Chlamydia was not confirmed in 216 specimens (15.3%), 18 (1.3%) were equivocal, and 17 (1.2%) were not tested, because they were inhibited on 2 platforms. Seventy-four incorrect samples (5.3%) , such as serum samples, were referred but not tested. Of the remaining 1083 samples, 101 (9.3%) could not be classified, 327 (30.2%) were confirmed to be LGV positive (all serovar L2), and 655 (60.5%) were confirmed to be positive for other chlamydial serovars (D-K). Figure 1 shows the epidemic curve, together with the total number of specimens received, which has increased steadily. Cases were diagnosed across the United Kingdom, but the majority were from London (71%) and Brighton (13%).
Enhanced surveillance. Enhanced surveillance forms were received for 282 cases occurring in MSM. Cases were predominantly in men of white ethnicity (260 [95%] of the cases) with a median age of 38 years (range, 21-65 years). Nine men were reported to have had LGV twice during this period. Three report forms were also received for cases occurring in heterosexual men who had presented in 2005 with inguinal lymphadenopathy and urogenital syndrome (2 cases in London and 1 case in the north of England). All 3 patients reported sexual contacts abroad. These cases are excluded from the further analysis below
Clinical presentation. A total of 228 (87%) of 263 patients with LGV cases presented because of symptoms, 16 (6%) presented as contacts, 10 (4%) presented through referral (usually with symptoms), and 9 (3%) had cases that were detected during routine examination and screening for sexually transmitted infection or HIV infection. The majority (262 [96%]) had proctitis; of these patients, 34 also reported genital symptoms, and a further 12 men had genital symptoms alone (table 1). Six of these men had LGV diagnosed from rectal swab samples, despite having no rectal symptoms reported, and 3 men received a diagnosis based on the results of tests of urethral swabs or urine extracts. A significant proportion of the men had multiple local and systemic symptoms suggestive of severe disease. Eight men had no symptoms; these included 4 contacts of individuals with LGV and 4 men who were detected during a routine screen (1 as the result of having been sexually assaulted). The time from development of symptoms to presentation varied widely, from <1 day to 18 months (median duration, 12 days), but 46 (25%) of 182 patients took >1 month to present to health services after development of symptoms. In 16 cases, the clinician reported that the patient had been investigated by gastroenterologists for inflammation, including some patients who had been admitted to the hospital and others who had undergone surgical procedures. Some patients were mistakenly treated for inflammatory bowel disease before LGV was diagnosed [17].
Coinfection. There was a high level of coinfection with HIV, hepatitis, and other sexually transmitted infections. HIV infection was present in 214 (76%) of 280 cases, including 9 cases in which HIV infection was diagnosed at approximately the same time as LGV. Seventy-eight (38%) of the 205 patients for whom a date of HIV infection diagnosis was available had received a diagnosis of HIV infection relatively recently (i.e., in 2003 or after). Forty-one (19%) of 214 patients were positive for hepatitis C antibodies. Concurrent infection with other sexually transmitted infections was documented in 105 (39%) of 268 cases; these infections included gonorrhoea (54 cases), infectious syphilis (13 cases), other chlamydial infection (17 cases), hepatitis B (1 case), and genital herpes (6 cases). There was no difference between HIV-positive and HIV-negative men in terms of age, ethnicity, or reason for attending clinical health services (table 2). Of the patients with HIV infection, 45% were receiving HAART. HIV-positive men were significantly more likely to have PCR-positive hepatitis C and to report unprotected anal intercourse, compared with HIV-negative men. There was no significant difference between HIV-positive and HIV-negative men in terms of other sexual behavior reported.
Sexual networks. The men reported a median of 3 sexual partners in the previous 3 months (range, 0-201 partners; mean, 9.6 partners). The majority of cases (178) were thought to have been acquired in the United Kingdom; 48 men reported recent sexual contacts overseas, the most common places being The Netherlands, Spain, and other parts of Western Europe. Patients who presented in 2004 were significantly more likely than those who presented in 2005 to report sexual contacts in or from The Netherlands (7 [17.5%] of 40 patients vs. 2 [1.2%] of 167 patients; OR, 17.5; 95% CI, 4.48-88.02; P < .001). Information on venues for meeting new partners was available for 113 men. Eighty patients (71%) reported meeting new partners at sex-on-premises venues or at sex parties, and 26 (23%) reported meeting new partners via the Internet. A small number of men reported occupations that could link widespread sexual networks, including sex work and travel-related work. The majority of men (201 [88%] of 252) reported unprotected anal sex; 188 (75%) reported unprotected receptive anal intercourse, and 139 (55%) reported unprotected insertive anal intercourse. Fisting was reported by 32 (32%) of 99 men, and use of sex toys was reported by 15 (19%) of 78, but there was a large amount of missing data for detailed sexual behavior.
DISCUSSION
This report is the first national picture of LGV in MSM and, to our knowledge, is the largest case series published to date. We have found a significant burden of this once-rare sexually transmitted infection among MSM in the United Kingdom, with 327 confirmed samples in the first 17 months of surveillance. The clinical presentation is varied, but many patients have multiple symptoms indicative of severe disease, with some patients having developed invasive disease requiring surgical referral [17]. The infection is present across the United Kingdom but is concentrated in areas with substantial numbers of MSM living with HIV infection. Infection appears to have been acquired primarily through unprotected anal intercourse, often with multiple partners. The majority of patients were coinfected with HIV, and some patients had apparently acquired HIV infection and LGV at approximately the same time. High levels of coinfection with bacterial sexually transmitted infections and hepatitis C were also reported, similar to findings in The Netherlands [18].
This national diagnostic, enhanced surveillance system provides a comprehensive picture of diagnosed LGV, but this is still likely to be an underestimate of the true extent of the epidemic, for the following reasons. First, the national diagnostic service is currently only available for symptomatic MSM and their sexual contacts who have already received a diagnosis of chlamydia. Asymptomatic or mildly symptomatic cases will, therefore, be missed, as will patients presenting to clinics whose local laboratories refuse to test rectal specimens for chlamydia. More extensive rectal chlamydia screening would almost certainly uncover cases with fewer or no symptoms [19], but, as yet, this is not recommended in national guidelines. Second, some men may have conditions that remain undiagnosed if they present to specialists outside of genitourinary medicine, where LGV may not be considered within the differential diagnosis of proctitis. Third, despite awareness campaigns among community groups and health care professionals, a proportion of patients and clinicians will not be alert to the potential diagnosis. The clinical and epidemiological picture presented is also limited by the surveillance system, with case report forms being completed retrospectively; although some clinics were able to provide full data, others returned limited behavioral and contact data.
The results of this surveillance show many features of an outbreak. However, before October 2004, there was no national diagnostic facility and no systematic surveillance of LGV in the United Kingdom; reports of chancroid, donovanosis, and LGV were given as a single total when reported in the aggregate KC60 clinic returns [20]. These aggregate data give a picture of sporadic, imported infections, reports fluctuating between 50 and 70 cases during recent years (HPA, unpublished data). The recent increase is clearly, in part, an artefact of the change to surveillance. However, there are other factors suggesting an ongoing outbreak, including geographic clustering, common social venues, potentially linked sexual networks, and links to European locations that had previously reported cases. The serovar associated with the clusters in The Netherlands [21] has recently been shown to have existed in the United States in the 1980s [11], but the recent picture in Europe may indicate a shift in the epidemiological characteristics of the disease. This change could have been influenced by a number of factors, including the distribution of risk behaviors and sexual network structures. The resurgence of LGV has occurred against a background of increased prevalence of HIV infection, syphilis, and gonorrhoea in MSM and has been influenced by developments in the HIV epidemic and behavioral changes in MSM. There has also been an increase in traditional "sexual market places," such as sex-on-premises venues and a proliferation of Web sites that facilitate the acquisition of new sexual contacts, effectively joining previously isolated sexual networks and reducing the time required for epidemics to evolve.
We have identified a substantial association between LGV and HIV infection, hepatitis C, and, to a lesser extent, other sexually transmitted infections, which may reflect the dense sexual networks in which all of these infections are being transmitted. The overlap with HIV infection is higher than for recent outbreaks of syphilis; in this LGV case series, almost 80% of patients had HIV infection, compared with 40%-47% of MSM in recent syphilis outbreaks in the United Kingdom and 50%-60% in the United States [22, 23]. The higher coinfection rate may reflect increasing serosorting, with HIV-positive men preferentially meeting other HIV-positive men via the Internet, sex parties, or clubs, for whom the imperative to practice safer sex may have, until recently, been diminished. Alternatively, the overlap may reflect selection bias, with clinicians selectively testing HIV-infected men and these men being more aware of LGV as a result of the publicity generated through HIV-related organizations, such as the Terence Higgins Trust. Another explanation is that LGV is acting as an opportunistic infection and producing a different range of symptoms and signs in men with HIV infection. Further research is indicated to explore this.
We need to actively address the LGV epidemic to reduce further transmission; this is essential, because it represents a significant public health threat and burden in itself and will be a potent factor in increasing HIV transmission between serodiscordant partners. It may also potentiate the transmission of other organisms, such as hepatitis C, in HIV-positive partnerships. It is essential to work closely with community organizations to raise and maintain awareness, including outreach to key social venues, and to constantly remind clinicians to be alert to the possibility of LGV. This is a distinct and often-severe condition, and it requires accurate diagnosis to inform appropriate treatment.
To inform control measures, we need further information on the distribution of infection in MSM. We are currently conducting case-finding exercises in a small number of clinics to measure the prevalence of urethral and rectal chlamydia, the proportion of these strains that are LGV serovars, and the range of symptoms. Results from this study will help to inform future screening and diagnostic guidelines.
Until these results are available, we recommend extended testing for chlamydia, including testing of rectal samples for men with proctitis and others who may have been exposed. Systematic screening of MSM for rectal chlamydia is currently being discussed by clinicians and microbiologists in the United Kingdom, and recommendations will be based on the results of the case-finding exercise. Until diagnostic facilities for LGV can be expanded [19], we think that presumptive treatment of rectal chlamydia and proctitis in MSM with 3 weeks of therapy (100 mg of doxycycline administered orally twice daily for 3 weeks) is indicated. It is also essential that patients are followed up with a test-of-cure and that sexual partners are sought and treated. Where contacts are difficult to trace, local awareness campaigns must be initiated to identify risk networks.
|
|
| |
| |
|
|
|
