International Journal of STD & AIDS >
Volume 19, Number 12 >
Pp. 805-809
Original research articles |
* Division of Gastroenterology, Department of Medicine Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, Room HG40, Toronto, Ontario M4 N 3M5;
Molecular Diagnostics for Special Bacteriology, National Microbiology Laboratory, Winnipeg, Manitoba;
Maple Leaf Medical, Toronto, Ontario;
Community Acquired Infections Division, Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada, Ottawa, Ontario;
** Hassle Free Clinic, Toronto, Ontario;
Division of Infectious Diseases Department of Medicine, Sunnybrook Health Sciences Centre, Toronto, Ontario;
Research Assistant, Sunnybrook Health Sciences Centre, Toronto, Ontario;
Division of Infectious Diseases Department of Medicine, University Health Network, Eaton 13-215, Toronto, Ontario;
*** Clinical Research Associate, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada;
London School of Hygiene and Tropical Medicine, London, UK;
Public Health Agency of Canada and, University of Ottawa, Ottawa, Ontario;
National Microbiology Laboratory, Public Health Agency of Canada, Winnipeg, Manitoba, Canada
Correspondence to: Dr Jill Tinmouth Email: jill.tinmouth{at}sunnybrook.ca
 | Summary |
|---|
 Top Summary INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: lymphogranuloma venerum • Chlamydia trachomatis • proctitis • prevalence • men who have sex with men
| INTRODUCTION |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
In industrialized countries, until recently, LGV was rarely described other than sporadic cases thought to be due to exposures in the tropics. In 2003, an outbreak of LGV was described in a group of 13 men who have sex with men (MSM) in Rotterdam, Holland.2 Subsequently, cases were reported in other European countries, largely among MSM.2–6 In response to this outbreak, the Public Health Agency of Canada (PHAC), in partnership with the Canadian provinces and territories, established an enhanced surveillance programme in February 2005 to rapidly identify and describe new cases of LGV in Canada. Between February 2005 and September 2007, there were 88 reported cases of LGV in Canada.7 Canadian and European cases were found to be infected with the same serovar (L2b), suggesting that the outbreak may have spread from Europe to Canada.5,6,8 In the recent outbreak, LGV appears to affect HIV-infected MSM disproportionately with 75% or more cases in both Europe and Canada reporting HIV co-infection.3,8
Of the 88 Canadian cases of probable or confirmed LGV, 85% had experienced proctitis, while only 44% reported a genital papule or lesion and 3.2% reported urethritis (Kropp, personal communication). A similar pattern has been reported in other Western countries. If transmission among Canadian cases is occurring via anal intercourse, the distribution of symptoms (genital symptoms much less frequent than rectal symptoms) in reported symptomatic cases raises the possibility that there may be a significant reservoir of sub-clinical genital LGV infection in Canada. In addition, MSM with asymptomatic rectal LGV may also be contributing to the Canadian outbreak.
The primary aim of this study was to determine if a reservoir of sub-clinical genital or rectal LGV C. trachomatis infection exists in MSM as such a reservoir might be contributing to the ongoing LGV transmission. Secondary aims were to measure the frequency of sub-clinical non-LGV C. trachomatis infection and identify risk factors associated with both types of C. trachomatis infection.
| METHODS |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This cross-sectional study was conducted at five sites in Toronto: two sites were HIV tertiary care centres, one was a primary care clinic serving a large MSM population, one was an STI clinic and one was a bathhouse frequented by MSM for sexual encounters. We recruited 25 subjects from each of the two HIV tertiary care HIV centres, 76 subjects each at the primary care clinic and the STI clinic and 51 subjects at the bathhouse. Subjects were enrolled between January 2006 and August 2006.
Subjects were men over 18 years of age without proctitis (based on anoscopic exam and symptoms) who reported having sex with other men. Individuals with symptoms such as rectal discomfort or bleeding were included as long as the rectal mucosa was normal on anoscopic exam, that is, there was no evidence of rectal ulceration, bleeding or mucopurulent discharge. Participants were enrolled if they had minimal non-rectal symptoms such as symptoms of urethritis, genital papules or swollen lymph nodes.
MSM who had prior infection with LGV and those without any sexual contacts in the year prior to enrolment were excluded.
All participants completed a detailed confidential questionnaire regarding: (i) rectal and urethral symptoms, current and in the past 90 days; (ii) sexual history; (iii) medical history; (iv) travel history and (v) demographics. All participants had an anoscopic exam. We collected three specimens from each participant: urine, rectal swab and serum. In subjects with positive serology but with negative rectal and urine specimens for C. trachomatis by nucleic acid amplification test (NAAT), we also collected a pharyngeal swab. All subjects received appropriate counselling regarding STIs and were educated about LGV.
Urine, pharyngeal and rectal swabs were tested for the presence of C. trachomatis at the National Microbiology Laboratory (NML) in Winnipeg, Manitoba using a commercially available chlamydia NAAT, Amplicor CT/NG Test (Roche Molecular Biochemicals, Laval, QC). C. trachomatis was also identified using a polymerase chain reaction (PCR) assay targeting the C. trachomatis 16s rRNA gene.9 DNA sequencing of the omp1 gene was performed on those specimens tested positive for C. trachomatis using a PCR assay to determine serovar. Amplification of the omp1 locus was performed using a nested PCR with primers CT1 and CT510 and D1 and D2;11 the latter were used in the second-round PCR and for DNA sequencing to determine serovars. Serovar classifications were considered ‘non-typable’ if both the commercial NAAT and the 16S PCR assays were positive, but the nested omp1 PCR was negative.
Serology was performed using a commercial kit (Focus Diagnostics Inc., Toronto, Ontario, Canada) and an in-house microimmunofluoresence (MIF) that utilized both the L2 LGV and non-LGV antigen acquired from University of Washington, Seattle, WA, USA.12
The presence of human globin gene in PCR assay was used to ensure rectal specimen adequacy in participants with positive serology but negative rectal swabs. The rectal swab was repeated if the globin PCR assay was negative.
A positive urine, rectal or pharyngeal swab for C. trachomatis by NAAT defined ‘active infection’. LGV was distinguished from non-LGV C. trachomatis infection based on the DNA sequencing results of the omp1 locus. Serology was considered positive if the LGV-specific IgG titre was greater than or equal to 1:256 or if the LGV-specific IgM titre was greater than or equal to 1:16, regardless of whether there was cross-reactivity with the non-LGV antigen. These parameters were selected as they reflect the criteria used by the Canadian public health authorities to support a probable diagnosis of LGV.13
Descriptive data and risk factors were reported as percentages or medians depending on the nature of the data. Univariate comparisons between those with and without active infection were done with the 
2 test or with Fisher's exact test depending on the cell size. Medians were compared using non-parametric methods. All tests were two-sided and were considered significant if the P value was less than 0.05.
We were not able to perform a formal sample size calculation as there were no data on the expected prevalence of LGV in high-risk Canadian populations. Therefore, this study used a convenience sample of 253. Given this sample size, we estimated the 95% CI around our estimate of prevalence to be ±1.3% if the true prevalence of LGV in the underlying population was 1%, while we expected the 95% CI around our prevalence estimate to be ±2.1% if the true prevalence was 3%.
The protocol was reviewed and approved by the Research Ethics Boards at Sunnybrook Health Science Centre (SHSC), University Health Network (UHN) and the PHAC. The study was carried out in accordance with the Declaration of Helsinki.
| RESULTS |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
|
Of our 253 participants, 12 (4.7%) were infected with non-LGV C. trachomatis (95% CI: 2.5–8.1%); there were 11 rectal and one urethral infections. Serovars included D (n = 5), G (n = 5) and J (n = 2). In those with rectal infection, none reported rectal pain and one reported rectal bleeding. Only the subject with urethritis had symptoms from the same site of infection. Individuals with non-LGV rectal or urethral C. trachomatis were not more likely to have positive serology (OR: 2.6, 95% CI: 0.5–12.6, P = 0.23) (Table 3). Finally, there was one positive pharyngeal sample that was non-typable.
|
|
| DISCUSSION |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The absence of LGV infection in our sample is reassuring, as it suggests that a large reservoir of genital or rectal LGV did not exist among asymptomatic MSM in Toronto at the time of our study. The prevalence of LGV in a cohort of minimally symptomatic high-risk persons has not yet been reported. However, in a cohort with known LGV infection from the UK, 3.2% were asymptomatic.14 Two other studies from the Netherlands of cohorts with known LGV infection did not describe the clinical presentations in detail, however, the range of reported rectal discharge varied widely, from 2% to 44%,15–17 possibly suggesting even higher rates of minimally symptomatic LGV infection. As these studies indicate that sub-clinical LGV infection has occurred in other countries, the lack of LGV cases in our cohort may be due to a lower prevalence of LGV infection in Canada relative to European countries, because the LGV outbreak in Toronto was waning at the time of our study,7 because of our small sample size or possibly, because of a combination of all three factors.
Eight percent of our sample had positive serology in the absence of active infection with LGV C. trachomatis, possibly representing either previous infection or false-positive findings. As there was no association between self-reported symptoms in the past 90 days and positive LGV serology, previous infection, if any, is likely to have been asymptomatic. However, these may also be false-positive findings as LGV serology is generally felt to be unreliable due to the cross-reactive nature of the antigens used in the serology test.
In fact, because of this cross-reactivity, L2 is often the sole antigen in serologic assays used to test all types of C. trachomatis, including non-LGV serovars.18 Up to 70% of serum samples positive for LGV serovars have been shown to cross-react with non-LGV serovars.19 In our sample, 50% of those who had an antibody response to the L2 antigen as defined above also had an antibody response to the non-LGV antigen (using a conservative cut-off titre of >1:256 for the non-LGV IgG titre). Our data corroborate work by others19 indicating that serology using IgG or IgM is highly unreliable to test active C. trachomatis infection, either LGV or non-LGV, in an asymptomatic individual. A recently published study suggests that it is possible to use IgA titres to differentiate between rectal infection with non-LGV and L2 LGV C. trachomatis,20 however, we did not measure IgA titres in our subjects.
We found a 5% rate of predominantly rectal sub-clinical non-LGV C. trachomatis. Other studies from the UK and the rest of Europe have reported that the prevalence of non-LGV C. trachomatis infection ranges from 8.6% to 13.3% in MSM.1,21–23 These studies were predominantly done in STI clinics or saunas (where presumably subjects were likely to have engaged in higher risk behaviours), and although we included patients from these settings, half of our patients were from lower risk settings (HIV clinics and general practice). Interestingly, we found that rates of infection with C. trachomatis were similar in subjects from low-risk and high-risk sites (4% and 6.3%, respectively). Therefore, if routine screening for C. trachomatis of asymptomatic MSM is undertaken, it should not be limited to high-risk settings, such as STI clinics, only.
In those studies that reported rates of infection by anatomic site, non-LGV C. trachomatis rectal infection was most common, ranging from 5.9% to 8% of those studied.1,21–25 Patients with rectal infection were more likely than those with urethral infections to be asymptomatic;21,22,24 in fact, in these studies, the majority (63–85%) of those with rectal infection were asymptomatic.23–25 In our study, all subjects with rectal non-LGV C. trachomatis infection were asymptomatic while the participant with urethral infection reported symptoms consistent with the site of infection.
We found, as have others, that a history of unprotected anal sex is associated with rectal C. trachomatis infection.21,22,25 Self-reported regular condom use reduces rectal C. trachomatis infection by as much as 50%.1 Our study is the first to report rectal enemas and drug insertion as risk factors for rectal non-LGV C. trachomatis infection, although another study did find an association between these factors and LGV proctitis.15
Our findings are of particular importance because of their relevance to HIV transmission. STIs promote the spread of HIV by facilitating HIV shedding and by increasing susceptibility. Seminal HIV viral load is approximately five-fold higher in the presence of gonococcal and/or chlamydial urethritis.26 In both heterosexuals,27,28 and MSM29,30 HIV acquisition is strongly associated with gonococcal, chlamydial, early syphilis and herpes simplex infections. Specifically, C. trachomatis infection has been shown to increase the risk of HIV transmission in MSM29 and in heterosexuals31 approximately four-fold.
In summary, we did not find a reservoir of sub-clinical LGV in our small sample of MSM in Toronto. However, we did find that the risk of sub-clinical rectal non-LGV C. trachomatis may be elevated among those MSM who in the past three months engaged in unprotected anal intercourse, used rectal enemas or rectal drugs. These findings are important as undetected non-LGV C. trachomatis may potentiate HIV transmission among MSM. In particular, because our small sample size is an important limitation to our study, further study is required to confirm our findings in other settings to identify high-risk subsets that might benefit from screening for non-LGV C. trachomatis, and to perform cost benefit analyses of screening these high-risk populations.
| ACKNOWLEDGEMENTS |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Meetings at which this work was presented in part: Canadian Association of HIV Research (CAHR), 16th Annual Canadian Conference on HIV/AIDS Research, 28 April 2007, Toronto, Ontario, Canada.
International Society for Sexually Transmitted Diseases Research (ISSTDR), 17th meeting, 29 July 2007–1 August 2007, Seattle, WA, USA.
| REFERENCES |
|---|
|
TopSummaryINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. A. Abdul-Sater, E. Koo, G. Hacker, and D. M. Ojcius Inflammasome-dependent Caspase-1 Activation in Cervical Epithelial Cells Stimulates Growth of the Intracellular Pathogen Chlamydia trachomatis J. Biol. Chem., September 25, 2009; 284(39): 26789 - 26796. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
