Department of Laboratory Medicine, University of California, San Francisco
In 2003, an outbreak of lymphogranuloma venereum (LGV) proctitis was detected in men who have sex with men (MSM) in Rotterdam, Netherlands. The outbreak was caused by LGV serovar 2, identified by nested polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP) analysis.1 The disease being seen was not the typical textbook presentation where LGV is described as a disease involving genital ulcers and inguinal lymphadenopathy (bubonic form). The bubonic form of LGV is endemic in some developing countries, and there was speculation that this unusual agent was introduced into the Netherlands by a traveler who acquired the infection in one of these countries. Concern has been raised about further spread of LGV into the European Community and the impact this ulcerative disease might have on HIV transmission. Further, United States STD control officers also expressed concern about the possible introduction of LGV into the United States. Thus, in response to the findings in Europe, CDC published a guideline of diagnostic criteria for LGV.2 The same was done by some state public health officials and in other countries. The lay press has also publicized the situation.
We would like to suggest that concern about the potential introduction of LGV into the United States ignores the real situation. That LGV could be an important cause of proctitis among MSM was highlighted in New York and Washington, D.C., more than 40 years ago.3,4 LGV has likely been here for many years, and the new efforts to find this infection will undoubtedly succeed, but to attribute the suddenly (to be) found infections to a newly introduced organism is just wrong. The LGV infections will be detected because laboratories will use diagnostic tests that can specifically identify the LGV agent.
To the best of our knowledge, there has been no routine surveillance for the presence of LGV in the United States. Certainly the disease is reportable, but only the classic bubonic form is really considered. However, after chlamydial cultures became regularly available in some settings, workers revisited the LGV problem in MSM. Quinn and colleagues5 reported isolation of LGV2 from ulcerative lesions in the intestinal tract of MSM in 1981 in Seattle, WA. At the time, we were offering cultures for C trachomatis to the medical community in San Francisco, CA. We identified many LGV isolates from rectal swabs and found that the typical classic presentation of LGV (inguinal buboes in the male) simply was not being seen. In other words, the proctitis syndrome appeared to be the major or perhaps the sole clinical finding.
Throughout these studies, our standard isolation procedures were used. The specimens were usually rectal swabs. But some of the specimens were collected using anoscopy to visualize lesions that were then swabbed or biopsied. Swabs were placed in an antibiotic-containing transport medium. This medium was then diluted (typically 10−1 and 10−2) and the dilutions were inoculated into cycloheximide-treated McCoy cells. Isolates were identified by the use of iodine or fluorescent antibody staining methods that identified inclusions as those of C trachomatis. The isolates were then identified as LGV strains by their ability to grow to high titer in these cells without the centrifugation that is needed to grow the trachoma biovar. We published several papers dealing with various aspects of clinical presentation and diagnosis of these infections. In one study that stressed the inadequacy of serodiagnosis, we noted that 2 of 6 rectal isolates were LGV2.6 Another report described clinical findings in 3 men whose biopsies yielded LGV2.7 In a review article, we reported isolation of Chlamydia trachomatis from rectal specimens of 27 of 127 (21.3%) MSM with proctitis, and 20 of 27 (74.1%) isolates were identified as LGV strains.8
Recently, following the publicity following the reports on the introduction of LGV into Netherlands, we reviewed results with 523 consecutive rectal samples that had been processed in this laboratory between 1980 and 1985. All the specimens had been collected in medical clinics (ambulatory care, emergency room, screening and acute care, etc) from symptomatic men. Chlamydia trachomatis was cultured from 101 (19.3%) of them. Of these, 68 (67.3%) were identified as the LGV biovar. Our identification of the LGV biovar using biologic properties was validated in the recent study by Caldwell et al.9 on the presence of tryptophan oxidase in genital chlamydia. They wanted to test LGV strains for the presence of the genes for the enzyme and we sent them 45 of our isolates. They genotyped the isolates and of the 45 isolates we identified biologically as LGV they identified 30 as LGV2, 14 as LGV1 (1 could not be typed).
At some time in 1985, we stopped offering free culture of rectal swabs. Many of the clinicians with whom we worked started treating the symptomatic patients with doxycycline on a syndromic basis.
We have not been performing routine diagnostic testing for MSM, but recently, in the process of evaluating diagnostic tests for extragenital chlamydial infection, we did isolation attempts on 205 MSM and isolated C trachomatis from the rectal swabs of 7%.10 Of the 14 isolates obtained in these studies, 5 were LGV strains. With a hiatus between the earlier studies and these recent studies, we cannot definitively say that LGV has been present in our community throughout this time period; however, it certainly seems likely.
Rather than be concerned about the introduction of LGV into the United States, we should recognize that it has probably been here all along. LGV is not an exotic tropical disease but one that is endemic in the United States. That it is not homogeneously distributed in the United States is suggested by the report from Bauwens et al.,11 who found only 12 of 767 (1.6%) of their rectal isolates obtained from at-risk MSM in Seattle from 1981 to 1991 to be LGV strains. Though their recovery rates were lower than those in San Francisco, they too were recovering LGV strains throughout the 1980s. That others are not reporting LGV is simply because the commonly used diagnostic tests for C trachomatis do not discriminate between the 2 biovars. The nucleic acid amplification tests target species-specific nucleic acid sequences that do not distinguish the invasive LGV biovar from the more common trachoma biovar where infection is restricted to columnar and squamocolumnar cells. Antigen detection methods target common antigens.
Relying on the textbook presentation of the bubonic form of the disease before thinking of LGV is a mistake. Clearly LGV proctocolitis is far more common among MSM. Why the bubonic form is not seen more often in these confirmed LGV infections is not clear, but disease presentation may be a function of lymphatic drainage from the initial infected site.
Instead of worrying about importation, we should focus on the substantial problems LGV presents. It is time to perform broad-based screening studies that include asymptomatic men and women to understand the epidemiology of these infections. Certainly such an invasive disease, capable of causing so much tissue destruction, has the potential to enhance transmission of HIV. We do not know the best way to diagnose LGV infections. New tools are needed. Questions about the best methods of treating LGV remain as there have never been adequate controlled treatment trials.
1. Nieuwenhuis RF, Ossewaarde JM, Gotz HM, et al. Resurgence of lymphogranuloma venereum in Western Europe: an outbreak of Chlamydia trachomatis
serovar L2 proctitis in Netherlands among men who have sex with men. Clin Infect Dis 2004; 39:996–1003.
2. Centers for Disease Control and Prevention. Lymphogranuloma venereum among men who have sex with men: Netherlands, 2003–2004. MMWR Morb Mortal Wkly Rep 2004; 53:985–988.
3. Grace AW. Anorectal lymphogranuloma venereum. JAMA 1943; 122:74–78.
4. Greaves AB. The frequency of lymphogranuloma venereum in persons with perirctal abscesses, fistulae in ano or both: with particular reference to the relationship between perirectal abscesses of lymphogranuloma origin in the male and inversion. Bull World Health Organ 1963; 29:797–801.
5. Quinn TC, Goodell SE, Mkrtichian E, et al. Chlamydia trachomatis proctitis. N Engl J Med 1981; 305:195–200.
6. Schachter J. Confirmatory serodiagnosis of lymphogranuloma venereum proctitis may yield false-positive results due to other chlamydial infections of the rectum. Sex Transm Dis 1981; 8:26–28.
7. Bolan RK, Sands M, Schachter J, Miner RC, Drew WL. Lymphogranuloma venereum and acute ulcerative proctitis. Am J Med 1982; 72:703–706.
8. Schachter J, Osoba AO. Lymphogranuloma venereum. Br Med Bull 1983; 39:151–154.
9. Caldwell HD, Wood H, Crane D, et al. Polymorphisms in Chlamydia trachomatis
tryptophan synthase genes differentiate between genital and ocular isolates. J Clin Invest 2003; 111:1757–1769.
10. Schachter J, Moncada J, Liska S, Klausner J. Performance of nucleic acid amplification tests (NAATs) for chlamydial and gonococcal infections of the oropharynx and rectum (Abstract C-277). Presented at the 104th General Meeting of the American Society for Microbiology; New Orleans, LA, May 23–27, 2004.
11. Bauwens JE, Lampe MF, Suchland RJ, Wong K, Stamm WE. Infection with Chlamydia trachomatis
lymphogranuloma venereum serovar L1 in homosexual men with proctitis: molecular analysis of an unusual case cluster. Clin Infect Dis 1995; 20:576–581.