Etiology of Genital Ulcer Disease and Its Relationship to HIV Infection

IN INDIA AND SUB-SAHARAN AFRICA, where sexually transmitted diseases (STDs) are common, human immunodeficiency virus type-1 (HIV) is spread primarily by heterosexual contact.1 One theory to explain the extensive heterosexual transmission of HIV in these regions is that the presence of other STDs facilitates HIV transmission.2,3 Untreated STDs, especially genital ulcer disease (GUD), could enhance both the susceptibility of an uninfected person to acquiring an HIV infection as well as the infectivity of an HIV-infected individual.4 HIV nucleic acid is frequently detected in ulcer specimens from HIV-infected men5; a decreased rate of HIV nucleic acid detection was reported after successful treatment of these infections.6 In fact, some studies have shown that effective management of STDs can reduce the incidence of HIV infection by as much as 40%.7 The biologic mechanisms by which GUD and other STDs facilitate the transmission of HIV are incompletely understood. However, it is likely that macro or micro disruptions of the epithelium and the presence of CD4+ lymphocytes in genital ulcers are important factors in viral transmission. The primary etiologic agents of GUD in STD clinic attendees are Treponema pallidum, Haemophilus ducreyi, and herpes simplex virus type-2 (HSV-2); Chlamydia trachomatis and Calymmatobacterium granulomatis are less common causes of GUD. The proportion of genital ulcers that are caused by each of these agents varies both geographically and temporally.8–10 Conventional laboratory tests used for the diagnosis of GUD are relatively insensitive and are often not available in many clinics treating such patients. For these reasons, the World Health Organization (WHO) has recommended the use of syndromic STD management protocols, especially in those areas with limited resources.11 According to a WHO treatment algorithm for GUD, patients should be treated with antibiotics effective against both syphilis and chancroid and should be managed symptomatically for genital herpes only if they present with vesicular lesions or a history of vesicular lesions. Syndromic management of GUD does not include antiviral therapy; antimicrobial therapy is directed at the bacterial causes of genital ulcers. Some of the problems associated with syndromic management of GUD, however, include the observation that patients often do not present with the classic clinical signs because they are seen late in the course of disease, with atypical ulcers due to immunosuppression resulting from HIV infection, or with ulcers of mixed etiology.12,13 Better diagnostic tests, which are rapid, sensitive, and inexpensive, are urgently needed in resource-limited settings to assist in the diagnosis of GUD. Polymerase chain reaction (PCR) amplification tests have been developed for each of the known etiologic agents of GUD. A multiplex PCR (M-PCR) amplification assay, which can simultaneously detect the presence of herpes simplex viruses types-1 and -2, H. ducreyi, and T. pallidum in a single ulcer specimen, has also been developed and shown to be more sensitive than conventional laboratory tests for the diagnosis of GUD.10 Despite their sensitivity, nucleic acid amplification assays are not practical for the management of patients with GUD in most clinical settings because of the costs, time-to-results, and technical expertise required. Nevertheless, nucleic acid amplification tests will be very useful in a) determining the etiology of GUD in different populations and geographic areas, b) evaluating syndromic algorithms for GUD, c) investigating the epidemiology of GUD, d) assessing the accuracy of a clinical diagnosis or the performance of conventional laboratory tests, e) assessing appropriateness of therapy, and f) identifying new agents of GUD. In this issue of the journal, Risbud et al.14 screened 746 consecutive patients presenting to two STD clinics in Pune, India for the presence of genital ulcers. Genital ulcers were observed in 41% of these patients. This proportion is almost twice as high as what has been reported in South Africa15 and may reflect the inclusion of patients who were attending the clinics for other reasons, such as symptoms of urethritis. Risbud et al.14 used M-PCR to evaluate the 277 men and 25 women with genital ulcers; HIV seroprevalence among these patients was 22.2%. The etiology of GUD, as determined by M-PCR, was HSV-1/-2 (26%), H. ducreyi (23%), T. pallidum (10%), multiple infections (7%), and no etiology (34%). Results obtained with M-PCR were compared with those of a clinical diagnosis based on the appearance of the ulcer and with microbiologic evaluations which included darkfield microscopy and serology for syphilis. The results of this study emphasize two important points. First, they confirm previous studies, which demonstrated that clinical diagnosis was not an accurate method for the identification of GUD etiology, particularly in HIV-infected individuals.12,13 Second, the authors showed that the conventional tests used for the diagnosis of GUD were less sensitive, and often less specific, than M-PCR. Previous studies showed that the sensitivity of culture for HSV relative to that of M-PCR varied from 61% to 75%.5,10,13 HSV culture is generally not available and is relatively expensive. It was encouraging to see that the rapid antigen detection test, Herpcheck (Abbott Laboratories, Abbott Park, IL), is comparable to culture, with a sensitivity of 68.5% and a specificity of 99.5% when compared with M-PCR. Thus, Herpcheck may be suitable for use in many resource-limited settings. Darkfield microscopy and serology have been important methods used in the diagnosis of primary syphilis. The results of this study demonstrated that darkfield microscopy provided no added advantage in the diagnosis of primary syphilis. Moreover, the low sensitivity and specificity of darkfield microscopy that were observed in this study reinforce the need for adequate training of laboratory and clinical personnel. With limited resources, darkfield microscopy should not be promoted for developing countries. The usefulness of the results of syphilis serology as an adjunct in the diagnosis of primary syphilis depends on several factors. Prior or untreated latent syphilis can result in false-positive results in both the nontreponemal and treponemal tests. The nontreponemal test results obtained in this study suggest that more than 30% of patients with primary syphilis present before developing reaginic antibodies. Although treponemal tests, such as the FTA-ABS test, are more sensitive, they are more expensive and require training. However, a new rapid treponemal test, in a dipstick format, is under evaluation for use in resource-limited settings. Culture for H. ducreyi was not performed in this study; however, previous evaluations indicated that, compared with M-PCR, culture had a sensitivity of 74%.10,13 The high proportion of M-PCR negative results (34%) observed in this study was troubling. Previous studies indicated that the proportion of M-PCR negative specimens ranged from 6% to 20%.5,10,13 Specimen quality is undoubtedly an important issue. However, patient selection (all patients were screened for ulcers) and the presence of additional agents should also be considered. Herpes simplex virus was the most common agent identified by M-PCR and was detected in >31% of the specimens. The observation that patients with HSV were more likely to be HIV infected than were patients with syphilis or chancroid has important implications for the syndromic management of GUD in India as well as other countries where HIV infection is prevalent. Type-specific serologic assays for HSV-1 and -2 have been developed using glycoprotein G-1 (gG-1) and G-2 (gG-2) as the target antigen.16 Population-based seroprevalence studies have indicated that HSV-2 infections are far more widespread than previously recognized and that the great majority of people with serologic evidence of HSV-2 infection have no history of genital herpes.17 Many gG-2-seropositive persons intermittently reactivate HSV-2 on mucosal surfaces with small, often difficult-to-recognize lesions or lesions of short duration.6 HIV infection has been shown to increase genital shedding of HSV-2 as well as to increase the frequency of recurrence in gG-2-seropositive individuals.18,19 Conversely, as in other studies, these findings suggest that both initial and recurrent HSV-2 infections are strongly associated with HIV infection.13 Thus, HSV-2 may be responsible for a greater population-attributable risk of HIV infection than previously thought. Syndromic algorithms which include the management of genital HSV, condom promotion, and antivirals to suppress recurrent episodes may have a greater impact on HIV incidence than those that do not.