We read with interest the paper from Sturm and colleagues in which they showed a PCR-based restriction fragment length polymorphism (RFLP) to differentiate the lymphogranuloma venereum (LGV) biovars from the trachoma and urogenital biovars (6). Recently an outbreak of rectal LGV infections among men having sex with men (MSM) was reported with both symptomatic and asymptomatic patients (5). Interesting from a clinical perspective is that the LGV-based genital ulcer disease (GUD) described by Sturm is not associated with human immunodeficiency virus (HIV), while the recent outbreak of rectal LGV infections in MSM is strongly associated with HIV seropositivity. Since current commercially available diagnostics do not offer the option of distinguishing the LGV biovars from the other biovars and the current ompA PCR-based RFLP analysis for specifically identifying all individual serovars is too time consuming (1, 2), we developed a real-time PCR for rapid and easy diagnosis of LGV infections which was used for clinical and case finding purposes in this outbreak among MSM (3). We used the polymorphic membrane protein H gene (pmp gene) as a PCR target because it has a unique gap in all LGV strains of Chlamydia trachomatis, compared to other serovars, which makes the PCR highly specific. The following primers and probes were selected: LGV-F (5′ CTG TGC CAA CCT CAT CAT CAA 3′), LGV-R (5′ AGA CCC TTT CCG AGC ATC ACT 3′), and LGV MGB-probe (6-FAM-CCT GCT CCA ACA GT). Reaction details are described elsewhere (3). This assay will be much easier than the CrP-PCR-D test followed by AccI restriction as reported by Sturm for the diagnosis of LGV-based GUD in South Africa. However, we realize that real-time PCR (TaqMan, LightCycler, or Rotorgene) is not readily available everywhere; therefore, we report here a simple adaptation of our real-time PCR, resulting in a potentially easier assay than the CrP-PCR-D. When only the PCR is performed using the above-described LGV-F and LGV-R primers (so without using the probe), a PCR product of 98 bp will be generated in the case of all serovars in the trachoma biovar (serovars D to K), while the presence of serovars in the LGV biovar (serovars L1, L2, L2a, and L3) will result in a fragment of 62 bp (98 − 36 bp). In this way no restriction is needed anymore, and the PCR product can be directly analyzed at the gel level. Since the fragment is also shorter than the CrP-PCR-D fragment of 552 bp, potentially better amplification will be accomplished and fewer adverse effects of inhibition or problems with DNA quality will be encountered.

Lastly, in contrast to Sturm and colleagues, we do not think that current textbooks are incomplete in the description of the clinical picture of LGV. Early symptoms with inconspicuous ulcers without inguinal buboes have been described before (4). It is probably due only to the recent advancements in diagnostic tests for LGV, as described, e.g., by Sturm and our group, that LGV cases can now be better detected during these early stages of the infection.

• Copyright © 2005 American Society for Microbiology