Abstract
Sir: Greene et al. 1 describes a case of rectal lymphogranuloma venereum (LGV) infection diagnosed from 2003. Following the initial 2003 report of a cluster of rectal LGV infections among men who have sex with men (MSM) in Amsterdam, LGV was identified as an important public health issue in the USA and the Western Europe. 2 Initially, discussion referred to a pathogen newly reintroduced from endemic tropical areas but questions arose as to whether rectal LGV infection had been consistently present, but undiagnosed, in developed countries. 3 6 Yet the epidemiology of LGV in endemic tropical regions, including parts of Asia, the Caribbean and South America, is poorly defined. 7 Screening studies of supposedly endemic areas have not been conducted, and the syndromic management strategy for STDs used in many low-income countries precludes definitive diagnosis of chlamydia infections or LGV subtyping. A previous analysis of urogenital specimens collected from men and women at high-risk for HIV and STD transmission in the coastal region of Peru found no evidence of LGV in the population. 8 However, rectal samples were not collected in the study and the presence of rectal LGV could not be assessed.
To determine the prevalence of LGV infection among MSM in Peru, we screened a convenience sample of 559 MSM attending the Alberto Barton STD clinic in Lima/Callao from May to December 2007. All participants underwent swab collection from urethral, pharyngeal and rectal sites for STD screening. All specimens with chlamydia identified by nucleic acid testing (Roche Cobas Amplicor, Roche Diagnostics, Indianapolis, IN, USA) underwent additional realtime polymerase chain reaction (RT-PCR) testing (LightCycler, Roche Diagnostics) to identify the characteristic polymorphic membrane protein gene (PmpH) observed in LGV subtypes of chlamydia (serovars L1-L3). A total of 28 chlamydia-positive samples (8 pharyngeal, 13 urethral and 7 rectal) were collected from 27 MSM, indicating an overall chlamydia prevalence of 5.0% (95% confidence interval [CI] = 3.4-7.0%). None (0/28) of the chlamydia strains were positive by RT-PCR for LGV infection (0%, 95% CI = 0-9.8%). Confirmation of assay validity was provided by the inclusion of positive and negative control samples.
Although LGV is thought to be common in tropical areas, no cases of LGV infection from any anatomic site were identified during screening of a high-risk MSM population in Lima, Peru. The absence of observed infection in this population could be due to the limited number of samples available for testing or may reflect the vaguely defined global epidemiology of LGV infection in developing areas. Peru is composed of three different geographic regions (coast, mountains and rainforest), each with distinct cultures and epidemiological patterns of infection. As participants were recruited from the urban, coastal region of the country, it is still possible that LGV infection could be endemic in high-risk populations of Peru's tropical Amazon River basin region. Additional epidemiological surveys are necessary to determine the prevalence of LGV strains in diverse regions of Latin America and to accurately define the global epidemiology of LGV infection.