Article Text


Epidemiology oral session 10: Men who have sex with men
O1-S10.03 Lymphogranuloma venereum, HIV and high-risk behaviour: findings from LGV enhanced surveillance and mathematical modelling
  1. M Ronn1,
  2. G Garnett1,
  3. G Hughes2,
  4. C Ison2,
  5. I Simms2,
  6. S Alexander2,
  7. H Ward1
  1. 1Imperial College London, London, UK
  2. 2Health Protection Agency, UK


Background Lymphogranuloma venereum (LGV) re-emerged in industrialised countries in the early 2000s and most transmission is assumed to occur among high-risk HIV-positive men who have sex with men (MSM). As the epidemic matures, we would expect to see declining HIV prevalence among cases as the transmission moves outside the initial core groups, however LGV cases continue to present with high levels of HIV-positivity. We have previously conducted a systematic review and a meta-analysis of literature published on LGV and found an eightfold greater likelihood of HIV-positivity among LGV cases when compared to controls with non-LGV chlamydia. In this study we investigate the role of high-risk behaviour, serosorting and differences in transmission probability as contributing factors to the persistent association between HIV and LGV.

Methods Data from LGV Enhanced Surveillance in the UK were used to investigate the relationship between HIV-positivity and LGV acquisition by setting HIV-positivity as the outcome in logistic regression. A simple deterministic compartmental model was developed, stratified by HIV-status and risk behaviour, to simulate the spread of infection among MSM in a large metropolitan population. Assortative mixing for HIV status and high-risk behaviour were controlled independently of each other see Abstract O1-S10.03 figure 1.

Abstract O1-S10.03 Figure 1

Flowchart illustrating the model structure.

Results Up to date approximately 81% of the 1500 confirmed LGV cases have had surveillance forms returned with above 70% HIV-prevalence among cases over time. HIV-positive LGV cases were more likely to report unprotected anal intercourse and greater number of partners than HIV-negative LGV cases in surveillance data. The mathematical model indicates that extreme levels of serosorting would be needed to explain the high prevalence of HIV in the presence of high-risk HIV-negative group. Increase in the number of HIV-positive individuals, higher transmission probability in HIV-positive and serosorting together were better able to reproduce the strong association between HIV and LGV.

Conclusions Differential acquisition of LGV occurs according to HIV status. LGV acquisition among HIV-positives may be due to serosorting together with an increased transmission probability due to reduced condom use and sexual practices, and potential biological susceptibility. Findings suggest that HIV-positive MSM need targeted health promotion in view of their different risk behaviours.

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.