THE ASSOCIATION BETWEEN sexually transmitted diseases (STD) that cause genital ulceration and human immunodeficiency virus (HIV) infection has been suggested in several studies among a number of diverse populations.1–5 Epithelial discontinuity due to genital ulceration has been hypothesized to increase the efficiency of HIV transmission and acquisition.2 Many characteristics associated with increased risk for genital ulcer disease (GUD), particularly bacterial infections such as syphilis or chancroid, are likewise associated with an increased risk for prevalent HIV infection.6–10 These characteristics include drug use, prostitution or prostitute exposure, and multiple casual sexual partners.
Although the associations between genital ulceration and HIV risk can be shown consistently, the relative risks for HIV vary widely. Reasons for this variability may be that GUDs caused by different agents are differentially associated with varying biologic risks for HIV, or may be due to the differences in the definition of GUDs. References to GUD have included self‐report of prior genital ulcers, clinician‐observed genital ulceration, and serologic evidence of diseases such as syphilis or infection with herpes simplex virus type II (HSV‐2).1,3,11,12 Each of these “definitions,” however, is subject to different biases. Reliance on self‐reported versus observed GUD, for example, results in detection of GUD at different stages in the natural history of the disease, as well as recall bias. Infection‐specific techniques limit the spectrum of diseases included in the definitions. In cross‐sectional studies, serologic tests usually cannot determine the temporal sequence of disease acquisition, and for treatable diseases such as syphilis, serologic tests may become negative after therapy.13 Self‐reports of previous infections may be inaccurate because of misidentification of lesion etiology or failure to appreciate acquisition of infection. Therefore, finding different results in studies evaluating the HIV‐GUD interaction is not surprising.
To evaluate further this complex issue, we studied several definitions of GUD and their association with prevalent HIV infection in patients attending an STD clinic, focusing on how different definitions might lead to different strengths of association and possibly influence counseling messages.
As part of a prospective study to evaluate transmission and acquisition of STD, consecutive STD clinic patients and their sexual partners were actively recruited for study enrollment. This study was approved by the institutional review board of the Johns Hopkins Medical School and has been described previously.14 Informed consent was obtained from all study participants. After informed consent, each participant was administered a 40‐minute questionnaire by trained study clinicians. The clinician‐administered questionnaire included a detailed history of the onset, severity, and possible self‐treatment of genital ulcerations or presenting problems, current sexual practices, numbers and types of sexual partners, and past history of STD and drug and alcohol use. All participants were questioned regarding history of specific STD (e.g., syphilis, herpes, chancroid) and more general STD categories (genital ulcers).
All patients were evaluated for prevalent STD, which included a directed physical examination, urethral or cervical specimen collection for Gram stain evaluation, and vaginal secretion collection and processing for light microscopic evaluation. All subjects had urethral or cervical cultures obtained for Neisseria gonorrhoeae using modified Thayer‐Martin media15 and for Chlamydia trachomatis according to standard procedures.16 Phlebotomy was performed for routine syphilis serologic testing (rapid plasma reagin [RPR] test [Macro‐Vue, BBL, Cockeysville, MD] and fluorescent treponemal antibody absorption test [FTA‐ABS; Zeus Scientific, Inc., Raritan, NJ]), as well as HIV and HSV testing. In patients with observed genital lesions, genital ulceration was defined on the basis of superficial epithelial disruption; thus, verruciform lesions such as venereal warts were not considered genital ulcers. Exudate from the base of all lesions was examined under the dark‐field microscope for detection of spirochetes. Genital lesions were cultured for Hemophilus ducreyi using biplates containing Mueller‐Hinton agar base supplemented with 5% horse blood, 1% IsovitaleX, and 3 μg/ml vancomycin on one side, with Thayer‐Martin agar base supplemented with 2% bovine hemoglobin, 5% fetal calf serum, 1% IsovitaleX, and 3 μg/ml vancomycin on the other side.17
Herpes simplex virus cultures were collected routinely, and positive isolates were typed using standard procedures.18 Serologic evidence of HSV‐2 infection was determined by Western blot.18 HSV‐2 lysate from standard HSV‐2 strains was separated using gel electrophoresis and transferred to nitrocellulose strips. These strips were incubated with the patients' serum and then developed using a biotinylated horseradish peroxidase system. Serologic evidence of HSV‐2 was based on identification of the 92‐kilodalton gG glycoprotein band pattern, which has been shown previously to be specific for HSV‐2.19
All patients with genital ulcerations had serum specimens screened on‐site for evidence of syphilis on the day of presentation using the RPR card test. FTA‐ABS testing was done on RPR‐reactive specimens and on RPR‐nonreactive specimens if syphilis was clinically suspected or if genital ulceration was observed. HIV testing was performed on consenting participants using a commercially available enzyme‐linked immunosorbent assay (ELISA) for screening. Serum samples reactive in the ELISA test were confirmed by Western blot analysis (DuPont, Wilmington, DE) as previously described.20
Data Collection and Statistical Analysis
Version 6.08 of SAS‐Windows software (SAS, Inc., Cary, NC) was used for data analysis. Unadjusted odds ratios (OR) and 95% confidence intervals (CI) were computed separately by sex to assess univariate associations between relevant behavioral and clinical factors and prevalent HIV infection. To determine the independent effect of each definition of GUD on risk of prevalent HIV infection, five multiple logistic regression models were constructed, each differing from the others only in the definition of GUD used. The first model used observed genital ulceration, the second used self‐reported history of syphilis, the third used reactive serologic tests (reactive RPR card with reactive FTA‐ABS test) for syphilis, the fourth model used serologic evidence of HSV‐2 infection, and the fifth model used culture‐proven herpes genital ulcer. Other risk factors for HIV infection identified by univariate analysis were included in each model, namely, history of parenteral drug use, self‐reported history of same sex partner (men only), source of income (wages vs. nonwage income), history of incarceration, history of crack or other nonparenteral cocaine use, current cigarette smoking, and history of gonorrhea.
Demographic, Behavioral, and Clinical Characteristics
Between June, 1990 and August, 1991, 964 patients were enrolled into the study. Sixty (6.2%) patients refused HIV testing. Of the 904 participants tested for HIV, complete interview, clinical, and laboratory data were available for 863 (96%); these individuals comprise the subjects for this analysis (Table 1). More men (481) than women (382) entered the study, which reflected the general demographic profile for the clinic's populations. Most participants were young, African‐American residents of Baltimore City. Although participants of each gender reported a similar number of sexual partners in the 30 days before clinic presentation (median = 1), compared with women, men reported more lifetime partners (median = 20 vs. 6), more frequent same‐sex partner exposure (9% vs. 3%), more parenteral drug use (15% vs. 9%), more nonparenteral cocaine use (31% vs. 18%), and more prior syphilis (11% vs. 5%). In addition, prevalent HIV infection was more common in male participants (12% vs. 5%). Both observed genital ulcers and serologic evidence of syphilis also were present more frequently among male than among female participants (24% vs 11%, and 16% vs. 11%, respectively). If male patients with any same‐sex partner exposure during their lifetime were excluded, genital ulcers were observed in 23% of male patients compared with 10.8% of female patients (P < 0.001), but there were no gender differences in serologic evidence of syphilis (men = 13%, women = 11%, P = 0.25). Among patients with GUD, there was no gender difference between the percentage of patients with HSV culture‐positive genital lesions (men = 36%, women = 33%).
Predictors of Prevalent Human Immunodeficiency Virus Infection in Univariate Analysis
We first selected which factors clinicians commonly use to predict prevalent HIV infection among patients presenting for STD evaluation. Patient self‐reported history, laboratory test results that were available to examiners while the patient was in the clinic (“stat laboratory” test results), and observed physical findings were evaluated for their relationship to a positive HIV serologic tests (Table 2). Among men, the likelihood of HIV seropositivity at study enrollment was significantly elevated if the patient reported ever having a same‐sex partner, using injectable drugs, prostitute exposure, current cigarette smoking, having been incarcerated, or having no wage‐derived income. For women, historical factors associated with increased risk for prevalent HIV infection included intravenous, crack, or other nonparenteral cocaine use, and current cigarette smoking. Self‐reported STDs associated with prevalent HIV infections were a history of syphilis among both men and women. In addition, risk for prevalent HIV infection was also significantly elevated among participants of either sex with active syphilis diagnosed while they were in the clinic, based on detection of spirochetes on dark‐field microscopy examination or with a reactive “stat” serum RPR card test.
Predictors of Prevalent Human Immunodeficiency Virus Infection by Multivariate Analysis
To evaluate the relationship of different definitions of GUD to prevalent HIV infection while controlling for other risk factors, we constructed five multivariate models. The five models differed from one another solely in the method by which GUD was defined: model 1 used observed genital ulceration on physical examination; model 2 used a self‐reported history of syphilis; model 3 used serologic evidence of syphilis (a reactive RPR card test with confirmatory FTA‐ABS test); model 4 used serologic evidence of previous HSV‐2 infection; and model 5 used culture‐proven HSV lesions. In three of five models, the definition of genital ulceration used remained an independent predictor of HIV seropositivity after controlling for other characteristics associated with increased risk (Table 3). The relative odds for HIV seropositivity also varied as much as threefold, depending on which definition of GUD was used: acute genital ulceration (OR = 2.0; 95% CI = 1.0–3.9), history of syphilis (OR = 6.0; 95% CI = 2.8–12.7), reactive RPR card test and FTA‐ABS test (OR = 3.7; 95% CI = 1.9–7.0), serologic evidence of HSV‐2 (OR = 1.2, CI = 0.7–2.1), and observed HSV‐2 culture‐positive genital ulcerations (OR = 1.0, CI = 0.4–4.2).
To further assess whether significant associations were gender specific, men and women were analyzed separately. Other than observed genital ulceration among women, GUD definitions consistent with syphilis (models 2 and 3) remained significantly associated with prevalent HIV infection. Observed GUD and a history of syphilis were more strongly associated with HIV seropositivity for men (OR = 2.5 and 7.4, respectively) than for women (OR = 1.1 and 3.7, respectively). A reactive RPR card test with confirmatory reactive FTA‐ABS test was also strongly associated with HIV infection for men (OR = 4.2) and for women (OR = 3.0).
Relation of Syphilis Screening Tests With Past History of Syphilis Infection
The logistic regression models (models 2 and 3) using different definitions of syphilis showed more than 60% variation in HIV risk. Additional information such as historical data, dark‐field microscopy result, and FTA‐ABS testing of patients with nonreactive RPRs were used further to evaluate and understand this discrepancy. Among 863 patients who had HIV serologic testing and complete demographic information, 8.6% (74) gave a history of syphilis infection. The history of syphilis could be validated by a reactive RPR card test with confirmatory FTA‐ABS test in only 59% (44) of these participants. In the remainder, 18% (13) had only a reactive FTA‐ABS test and 23% (17) had both nonreactive RPR card and FTA‐ABS tests. Conversely, when serologic test results were considered, 9.5% (75) of the 789 participants who did not give a history of syphilis had reactive RPR card and confirmatory FTA‐ABS tests. These patients included 38 (51%) patients who presented with primary or secondary syphilis, 8 (10%) patients determined to have early latent syphilis, and 6 (8%) patients identified as syphilis contacts; and 23 (31%) patients with syphilis of unknown duration. Serologic screening evidence of syphilis and past syphilis histories, therefore, did not correlate in 12% of the study population (105 patients), and may possibly contribute to observed differences in the strength of association between different definitions of syphilis and prevalent HIV infection. Almost three quarters of these patients (75/105) had no known history of syphilis and reactive RPR and FTA‐ABS serologies. Each of the remaining 30 patients in whom syphilis history and serologic test results did not agree gave a history of prior syphilis and had nonreactive RPR card tests; in 13 of these, the history was substantiated by a reactive FTA‐ABS test, whereas in the remaining 17, the FTA‐ABS test was nonreactive.
Patients attending STD clinics comprise populations who are often engaged in sexual activity associated with increased risk for HIV infection, and should be counseled and offered HIV screening. Within this group are subpopulations at still greater risk for HIV infection who may benefit from carefully directed interviews and more intensive risk‐reduction counseling. Particularly high‐risk groups include homosexually active men, parenteral drug users, individuals who practice unsafe sex with multiple or anonymous partners or with parenteral drug‐using partners, and individuals with GUD.21–29 People practicing high‐risk behaviors may be identified in a variety of different ways, and different approaches are likely to modify the proportion of at‐risk individuals identified as well as the observed risk for infection. Drug use, for example, may vary with regard to the route of ingestion (inhalation, subcutaneous injection, intravenous injection), the type of drug used, the frequency of ingestion, or past history or current use.25–30 Similarly, homosexual activity may refer to recent sex exclusively with same‐sex partners, sex ever with a same‐sex partner, or any variety of different sex acts between partners.31 In this study, evaluation of five different definitions of genital ulceration demonstrated that, regardless of the definition used, syphilis was significantly associated with prevalent HIV infection, even after correction for behavior, but genital herpetic infection was not.
We found that observed genital ulceration and definitions of syphilitic GUD were associated with prevalent HIV infection while controlling for other risk behaviors. Each GUD definition had a different strength of association. Self‐reported history of syphilis was most strongly associated with prevalent HIV infection. Regardless of gender, patients with a history of syphilis were more than six times more likely to be HIV infected than those with no history of syphilis. If the patient had serologic evidence of syphilis, his or her likelihood of HIV infection was elevated nearly fourfold. Thus, although each of these variables measured the same disease, the strengths of association with HIV infection varied considerably. Potential reasons for the observed discrepancy include undiagnosed syphilis, loss of serologic reactivity to Treponema pallidum antibody after therapy, and poor recall or misunderstanding of prior syphilis diagnosis. Although each of these factors may have contributed to variation in risk for HIV observed in this study, it is difficult to assess their relative contribution. Nonetheless, among 74 people with a history of syphilis, 30 (41%) had nonreactive RPR card tests and would have been undetected by routine syphilis serologic screening procedures. Efforts to corroborate self‐reports of prior syphilis using the FTA‐ABS test, which may remain reactive longer than the RPR test, were successful in 13 of 30 patients. In the remainder, although syphilis history could not be confirmed serologically, it is possible that at least some had syphilis in the past and became seronegative after effective therapy. In a study by Romanowski and colleagues, over 25% of patients treated for primary or secondary syphilis seroreverted to nonreactive screening and confirmatory FTA‐ABS tests.13 This trend also has been demonstrated to be more pronounced in HIV‐infected people.32
Similarly, of the 119 patients with serologic evidence of syphilis, 75 (63%) had no knowledge of the diagnosis. Although most of these patients (46) were diagnosed with syphilis at the time of their enrollment visit to the clinic, in some patients the reactive serologic tests likely represented previously diagnosed infection, which might diminish the ability to demonstrate associations between syphilis and HIV infection.
Although several studies have demonstrated significant associations between serologic evidence of HSV‐2 infection and HIV risk,9,10,33 we did not find any contribution of genital herpes to HIV risk in this study using serologic evidence of prior HSV‐2 infection, and HSV culture‐proven genital lesions as definitions. The prevalence of HSV‐2 antibodies in our study population was high. Overall, 57.4% of the study subjects had HSV‐2 antibodies, and only 4.7% gave a history of genital herpetic infection. This is in accordance with previously published data from these STD clinics,33 which reported a seropre‐valence of 56.6% for HSV‐2 and 72% for HSV‐1. Furthermore, culture‐proven HSV‐2 genital ulcerations at presentation were not associated with prevalent HIV infections in our study population. In another study of the etiology of GUD34 among this clinic population, syphilis was the leading cause of genital ulcerations (52%) among HIV‐infected men, followed by culture‐proven herpetic ulcers (17%). In contrast, HSV caused 50% of the GUD observed in HIV‐uninfected men, whereas syphilis was identified as the cause of GUD in 21%.34 No cases of chancroid ulcerations have been diagnosed in these STD clinics since 1990 to the present. Nonetheless, we expected to observe an association between HSV ulceration and HIV, as has been reported.10,11,33 HSV recurrences, however, are expected to be more common among HIV‐infected individuals as their CD4 levels decline below 200 cells/mm3.35 In patients with higher CD4 levels, this association may not be seen.
Thus, although GUD is strongly associated with prevalent HIV infection, the strength of the association depends on the definition used. For clinicians dealing with patients at risk for HIV, evaluation should include several definitions of GUD to maximize risk ascertainment. Using a single definition, such as history of syphilis alone, may be suboptimal to identify potential HIV‐infected patients and to direct appropriate counseling and testing messages. Similarly, researchers evaluating potential risk factors for HIV should consider the various shortcomings of different definitions of genital ulceration in cross‐sectional and retrospective studies.
1. Greenblatt RM, Lukehart SA, Plummer, et al. Genital ulceration as a risk factor for human immunodeficiency virus infection. AIDS 1988; 2:47-50.
2. Simonsen JN, Cameron W, Gakinya MN, et al. Human immunodeficiency virus infection among men with sexually transmitted diseases: Experience from a center in Africa. N Engl J Med 1988; 319:274-277.
3. Plummer FA, Simonson JN, Cameron DW, et al. Co-factors in female-male transmission of human immunodeficiency virus type 1. J Infect Dis 1991; 163:233-239.
4. Cameron DW, Lourdes JD, Gregory MM, et al. Female to male transmission of human immunodeficiency virus type 1: Risk factors for seroconversion in men. Lancet 1989; 2:403-407.
5. Mehendale SJ, Rodriques JJ, Brookmeyer RS, et al. Incidence and predictors of human immunodeficiency virus type I seroconversion in patients attending sexually transmitted disease clinics in India. J Infect Dis 1995; 172:1486-1491.
6. Mertens TE, Hayes RJ, Smith PG. Epidemiologic methods to study the interaction between HIV infection and other sexually transmitted diseases. AIDS 1990; 4:57-65.
7. Laga M, Nzila N, Goeman J. The interrelationship of sexually transmitted diseases and HIV infection: Implications for the control of both epidemics in Africa. AIDS 1991; 5(Suppl 1):S55-S63.
8. Quinn TC, Glasser D, Cannon RO, et al. Human immunodeficiency virus infection among patients attending clinics for sexually transmitted diseases. N Engl J Med 1988; 318:197-202.
9. Nelson KE, Vlahov D, Cohn S, et al. Sexually transmitted diseases in a population of intravenous drug users: Association with seropositivity to human immunodeficiency virus (HIV). J Infect Dis 1991; 164:457-463.
10. Rolfs RT, Goldbery M, Sharrar GR. Risk factors for syphilis: Cocaine use and prostitution. Am J Public Health 1990: 80;863-857.
11. Stamm WE, Handsfield HH, Rompalo AM, Ashley RL, Roberts PL, Corey L. The association between genital ulcer disease and acquisition of HIV infection in homosexual men. JAMA 1988; 260:1429-1433.
12. Holmberg SD, Stewart JA, Gerber RA, et al. Prior herpes simplex virus 2 infection as a risk factor for HIV infection. JAMA 1988; 259:1048-1050.
13. Romanowski B, Sutherland R, Fick GH, Mooney D, Love EJ. Serologic response to treatment of infectious syphilis. Ann Intern Med 1991; 114:1005-1009.
14. Upchurch DM, Weisman CS, Shepherd M, et al. Interpartner reliability of reporting of recent sexual behaviors. Am J Epidemiol 1991; 134:1159-1166.
15. Thayer JD, Martin JE. Improved medium selective for cultivation of Neisseria gonorrhea
and Neisseria meningitidis.
Health Laboratory Reports 1966; 81:559-566.
16. Stamm WE, Tam M, Koester M, Cles L. Detection of Chlamydia trachomatis
inclusions in McCoy cell cultures with fluorescein-conjugated monoclonal antibodies. J Clin Microbiol 1983; 17:666-668.
17. MacDonald K, Cameron DW, Irungu G, et al. Comparison of Sheffield media with standard media for the isolation of Haemophilus ducreyi.
Sex Transm Dis 1989; 16:88-90.
18. Corey L, Adams HG, Brown ZA, Holmes KK. Genital herpes simplex virus infections: Clinical manifestations, course, and complications. Ann Intern Med 1983; 98:958-972.
19. Ashley RI, Miltoni J, Lee F, Nahmias A, Corey L. Comparison of Western blot and glycoprotein G-specific immunodot enzyme assay for detecting antibodies to herpes simplex virus types 1 and 2 in human sera. J Clin Microbiol 1988; 26:662-667.
20. Centers for Disease Control and Prevention. Interpretation and use of Western blot assay for serodiagnosis of human immunodeficiency virus 1 infections. MMWR 1989; 38(S-7):1-7.
21. Centers for Disease Control and Prevention. Human immunodeficiency virus infection in the United States: A review of current knowledge. MMWR 1987; 36(Suppl):1-48.
22. Darrow WW, Exhenberg DF, Jaffe HW, et al. Risk factors for human immunodeficiency virus (HIV) infections in homosexual men. Am J Public Health 1987; 77:479-483.
23. Holmes KK, Karon JM, Kreiss J. The increasing frequency of heterosexually acquired AIDS in the United States, 1983-88. Am J Public Health 1990; 80:858-862.
24. Friedland GH, Klein RS. Transmission of the human immunodeficiency virus. N Engl J Med 1987; 31:1125-1135.
25. DesJarlais DC, Friedman SR, Stoneburner RL. HIV infection and intravenous drug use: Critical issues in transmissions dynamics, infection outcomes, and prevention. Rev Infect Dis 1988; 10:151-158.
26. Chaisson RE, Moss AR, Onishi R, Osmond D, Carison JR. Human immunodeficiency virus infection in heterosexual intravenous drug users in San Francisco. Am J Public Health 1987; 77:169-172.
27. Shoenbaum EE. Hartel D, Selwyn PA, et al. Risk factors for HIV-1 infection in intravenous drug abusers. N Engl J Med 1989; 321:874-879.
28. Vlahov D, Munoz A, Anthony JC, Cohn S, Celentano DD, Nelson KE. Association of drug injection patterns with antibody to human immunodeficiency virus, type 1 among intravenous drug users in Baltimore. Am J Epidemiol 1990; 132:847-856.
29. Chaisson RE, Bacchetti P, Osmond D, Brodie B, Sande MA, Moss AR. Cocaine use and HIV infection in IVDUs in San Francisco. JAMA 1989; 261:561-565.
30. Celentano DD, Vlahov D, Menon AS, Polk BF. HIV knowledge and attitudes among IVDUs comparisons to the U.S. population and by drug use behaviors. Journal of Drug Issues 1991; 21:647-661.
31. Blumstein P, Schwarts P. American Couples. New York: Maron; 1983.
32. Haas JS, Bolan G, Larsen SA, Clement MJ, Bacchetti P, Moss AR. Sensitivity of treponemal tests for detecting prior treated syphilis during human immunodeficiency virus infection. J Infect Dis 1990; 162:862-866.
33. Hook EW III, Cannon RO, Nahmias AJ, et al. Herpes simplex virus infection as a risk factor for human immunodeficiency virus infection in heterosexuals. J Infect Dis 1992; 165:251-255.
34. Koutsky LA, Stevens CE, Holmes KK, et al. Underdiagnosis of genital herpes by current clinical and viral-isolation procedures. N Engl J Med 1992; 326:1533-1539.
35. Erlich KS, Mills J, Chatis P, et al. Acyclovir-resistant herpes simplex virus infections in patients with the acquired immunodeficiency syndrome. N Engl J Med 1989; 320:293-296.
International Conference on Emerging Infectious Diseases March 8–11, 1998 Atlanta, GA
The International Conference on Emerging Infectious Diseases will be convened to (1) encourage the exchange of scientific and public health information on global emerging infectious disease issues, (2) highlight programs and activities that address emerging infectious disease threats, (3) identify program gaps, (4) increase emerging infectious disease awareness in the public health and scientific communities, and (5) enhance partnerships in addressing emerging infectious diseases.
The meeting will consist of plenary sessions and symposia with invited speakers as well as oral and poster presentations based on the submission of an accepted abstract. Major topics will include current work on surveillance, epidemiology, research, communications and training, and prevention and control of emerging infectious diseases as well as topics related to emergency preparedness and response. Abstracts should address new, re‐emerging, or drug‐resistant infectious diseases that affect human health including foodborne diseases, tropical diseases, sexually transmitted diseases, infectious respiratory diseases, infectious diseases transmitted by animals and arthropods, infections acquired in health care settings, infectious diseases of infants and children, infectious diseases in immunodeficient persons, infectious diseases in minority and other at‐risk populations, blood safety, and xenotransplantation.
Deadline for submission of abstracts is October 31, 1997.
Registration will be limited to 2,500 participants.
Additional information on abstract submission, registration, and exhibit opportunities can be obtained by sending an e‐mail message to email@example.com or by calling 202–942–9248. Proceedings of the conference will be published in the Emerging Infectious Diseases journal.