INTESTINAL INVOLVEMENT BY lymphogranuloma venereum (LGV) strains of chlamydia was described in 1936.1 Although infection of the urethra by non‐LGV Chlamydia trachomatis was first described in 1972, infection of the rectum by these strains was not recognized until 1977.2–6 With technical improvements in culturing and serotyping of C. trachomatis,7–13 larger epidemiologic studies have become possible.14–16 In a previous study15 done in an urban sexually transmitted disease (STD) clinic during the period 1982 to 1984, serovars G and D were more frequently identified among homosexual men with rectal infection than among women with cervical infection, whereas serovar E isolates and C complex isolates were less frequent in homosexual men. These findings may reflect differences in the serovars that were circulating in different populations in the early 1980s. Alternatively, it may be that some serovars have biologic characteristics better suited to infection at specific anatomic sites. A study of women16 suggested that those with C. trachomatis infections of the serovar F and G groups had less frequent and less pronounced clinical manifestations of cervical infection (bleeding and mucopus) as compared with those having infections with serovar B and C complex strains. Among men with urethritis,14 a correlation between higher number of leukocytes on Gram stain and infection with the FG serovar has been reported, although no strong overall association was observed between the degree of acute inflammatory response and serovar. Earlier studies15,17,18 of C. trachomatis rectal infection among homosexually active men did not attempt to correlate particular signs (erythema, bleeding, mucopus), number of leukocytes, or inclusion counts with specific serovars. We thus undertook this cross‐sectional study using data from the Seattle King County Department of Public Health Sexually Transmitted Diseases at Harborview Medical Center (SKCDPH‐STD clinic at HMC) to examine whether rectal infection with different serovars was associated with differences in clinical presentation as measured by symptoms, signs, number of leukocytes, and inclusion counts.
All homosexually active men presenting to the SKCDPH‐STD clinic at HMC for routine screening, evaluation of symptoms, or epidemiologic treatment after presumed exposure to a sexually transmitted disease between January 1, 1982 and December 31, 1990 were eligible for the study. The SKCDPH‐STD clinic at HMC provides comprehensive STD‐related clinical services, including diagnosis, treatment, education, and risk‐reduction counseling. During the early 1980s 25 to 30% of visits to the clinic were made by homosexually active men. However, new‐problem visits to the clinic made by homosexual men declined from 4,142 in 1980 to 509 in 1988 (−88%).19
Patients presenting to the clinic were evaluated by clinicians according to standing clinical orders. Demographic (age, race), sexual, and medical history were taken by a medical specialist, and a genitourinary examination and collection of specimens for microbiologic studies were done. All patients were asked a standard series of questions regarding the reason(s) for visit; the presence, nature, and duration of genital and rectal symptoms; history of prior sexually transmitted diseases; recent sexual partners and practices; and history of allergy or recent use of antibiotics. All patients then underwent a standard physical examination that included urethral, rectal, and throat cultures for Neisseria gonorrhoeae and urethral and rectal cultures for C. trachomatis. Urethral Gram stains were done routinely. Anoscopy was performed, and a rectal Gram stain was obtained when rectal symptoms were reported. Rectal symptoms reported to the clinician included tenesmus, pruritis, discomfort, pain, discharge, and bleeding. Clinical findings recorded by the clinician included erythema (redness of the mucosa), bleeding (mucosa bled easily when swabbed with a cotton‐tipped swab), and mucopus (presence of thick yellow‐white purulent discharge). Proctitis was noted when there was inflammation of the rectal mucosa and evidence of redness, bleeding, or a purulent discharge. Laboratory records were used to identify all male patients who had rectal infection with C. trachomatis. Clinical and laboratory information was abstracted from the clinical or laboratory records for each man with rectal C. trachomatis infection and entered into a database for computer analysis.
Specimen Collection and Microbiologic Methods
Gram stains of rectal and urethral samples were examined for polymorphonuclear leukocytes and intracellular Gram‐negative diplococci. Pharyngeal cultures were obtained to determine whether N. gonorrhoeae and Neisseria meningitidis were present. Dark‐field examination was performed on scrapings of any genital or perianal lesions, and urethral and rectal specimens were obtained for C. trachomatis, N. gonorrhoeae, and herpes simplex virus culture. Stool specimens were examined for ova and parasites, and cultures for shigella, salmonella, yersinia, and campylobacter were obtained when signs or symptoms warranted.20,21
Isolation of N. gonorrhoeae
Specimens obtained from swabbing the rectum, urethra, and pharynx were streaked onto separate biplates containing modified Thayer‐Martin and enriched chocolate agar. The biplates were evaluated for the presence of N. gonorrhoeae according to previously described techniques.22
Isolation of C. trachomatis
Rectal and urethral specimens for C. trachomatis culture were obtained with a type III calcium‐alginate swab and were processed as previously described.23 Briefly, a 0.1‐ml aliquot of the clinical specimen was inoculated, and the remainder of the specimen was frozen at −70°C. All rectal specimens were sonicated before inoculation to reduce contamination by fecal microorganisms and to improve sensitivity.8 Chlamydial inclusions were detected by direct immunofluorescence after 48 to 72 hours of intracellular development as previously described.13 Specimens were viewed by inverting the plate and reading it at ×160, using a fluorescence microscope with vertical illumination. The total number of inclusions per microtiter was well enumerated. When inclusions were numerous, three fields were counted, and the total number of inclusions per well was estimated by multiplying the mean number of inclusions for the three fields by the number of microscope fields per well.
Positive C. trachomatis specimens identified through laboratory records were thawed and passaged in McCoy cell monolayers in 12‐mm shell vials until highly infectious (about 107 infectious units). Isolates were serotyped by a dot‐ELISA method using monoclonal antibodies specific for type‐ and subspecies‐specific epitopes of C. trachomatis as previously described12 or were serotyped using a simplified microtiter culture method for rapid immunotyping.24 Serovars were grouped in accordance with the results of Yuan et al.:25 B complex (serovars B, Ba, D, E); group FG (serovars F and G); and C complex (serovars A, C, H, I, J, K). Although LGV strains are classified in the B (L1, L2) and C (L3) complexes, they were kept separate for purposes of analysis. Isolates were typed without knowledge of inclusion counts, demographic characteristics, sexual behavior, or clinical and laboratory findings. Less than 5% of specimens were unrecoverable.24
Demographic, sexual, clinical, and laboratory characteristics of men with rectal chlamydial infection by B complex, group FG, or C complex were compared. Men with isolates of the B complex group were the reference group, and men with LGV strains were kept separate in these analyses. Comparisons of proportions and tests of association for categorical variables were analyzed by using the chi‐square or Fisher's exact tests. Unconditional logistic regression was used to calculate odds ratios (OR) and 95% confidence intervals (95% CI) and to adjust for potential confounding variables.26 All analyses were done with the Statistical Analysis System (SAS).
There were 764 positive rectal C. trachomatis cultures recorded in the laboratory logbook on 618 men. After exclusion of (1) 43 repeat positive samples obtained within 1 month of an initial positive culture and (2) 267 samples from men whose clinical records were not available because, in most cases, the men were examined in other health department clinics, 454 samples remained. Of the 454 samples, 310 were typed and 144 could not be revived for typing. Of the 310 isolates that were successfully typed, 5 were LGV strains and 305 were non‐LGV isolates. For comparative purposes, subjects were divided into groups on the basis of the serovar of their infecting isolate in accordance with the results of Yuan et al.25 Included in B complex were 164 isolates (53.7% of those typed) represented by 145 isolates of the D serovar (47.5%), and 19 of the E (6.2%). C complex comprised 55 isolates (18.0% of those typed) among which there were 3 (1%) I serovar isolates, 50 (16.4%) J isolates, and 2 (0.6%) K isolates. The FG group included 86 isolates (28.2% of those typed) with 27 (8.9%) F isolates and 59 (19.3%) G isolates. Five isolates were LGV‐associated serovars (L1 = 3; L2 = 2; L3 = 0). Trachoma‐associated serovars (A, B, Ba, and C) were not detected. The 144 men with rectal isolates that could not be revived for typing were also included as a separate subgroup in most analyses. Of the total 764 positive C. trachomatis rectal cultures identified in the laboratory logbook, 470 were immunotyped and 310 included in the analysis. The proportion of B, C complex, FG group isolates (as well as of D, E, I, J, K, F, and G) included in the study was similar to their proportion among all immunotyped rectal isolates.
Demographic and Sexual Behavior Characteristics
Compared to men with infections of the B complex type, men with C complex or FG isolates were not statistically different in age, race, reason(s) for visit, prior sexually transmitted diseases history, or report of practicing anal receptive intercourse (Table 1). More than 50% of the men were younger than 30 years of age, and 80 to 100% were white. Although men with B complex were more likely to be younger than men in other groups (mean age, 27.4 years versus 28.1 to 29.4 years), this difference was not statistically significant. Compared to men with infection of the B complex type, men with C complex isolates were more likely to report 0 to 1 sexual partner during the last month (P < 0.05); men with FG group isolates were more likely to have had greater than 4 sexual partners during the last month (P < 0.05); and men infected with an isolate that could not be typed were more likely to report a new partner during the last month (P < 0.01).
Clinical and Laboratory Characteristics
Among the 164 men with B complex, 55 with C complex, 86 with FG, and 144 with untyped isolates, respectively, there were 88 (53.6%), 37 (67.2%), 41 (47.6%), and 84 (58.3%) men without rectal symptoms (Table 2). Erythema, bleeding, or mucopus was identified among 29 men (17.7%) with B complex isolates, 15 men (17.4%) with FG, and 28 men (19.4%) with untyped isolates versus only 5 men (9.1%) with C complex isolates (P < 0.05). About 20% of men infected with B, FG, and untyped isolates and 10% of those with C isolates had proctitis. Only one man had concurrent conjunctivitis (and urethritis), and all three isolates from this patient were typed Ia of the C complex group. One man had a mixed infection (J + G). Concurrent chlamydial urethral infection was present among only 3.6 to 8.1% of men without LGV isolates.
Compared to men with isolates of the B complex, only men with isolates of the C complex were statistically different in terms of clinical presentation. This analysis was adjusted for age, year of examination, number of sex partners, and concurrent presence of N. gonorrhoeae. Men infected with C complex were less likely to report symptoms (OR: 0.4; 95% CI: 0.1‐0.8) and less likely to have clinical evidence of rectal erythema, bleeding, or mucopus (OR: 0.3; 95% CI: 0.1‐0.8). The mean duration of symptoms ± standard deviation (range, n) was 35.8 days, ±28.1 (1‐90, n = 9). Compared to men with B complex, men with C complex infections were also less likely to have proctitis (OR: 0.5; 95% CI: 0.1‐1.2); however, anoscopy was performed less frequently among men with C complex isolates (P < 0.05).
Men with isolates of the FG group and men with untyped isolates were not different from men infected with B complex for presence of symptoms, erythema, bleeding, or mucopus. There were no differences in number of leukocytes between the groups.
All five men with LGV strains had symptomatic infections and evidence of proctitis on examination.
Influence of Inclusion Count and Duration of Symptoms on the Association Between Chlamydial Serovars and Symptoms, Signs, and Laboratory Findings
Men infected with both N. gonorrhoeae and C. trachomatis were clinically similar to men infected only with N. gonorrhoeae and were more likely than men infected only with C. trachomatis to have rectal specimens with more than 500 inclusions (data not shown). Thus, men infected with both organisms were excluded from the following analyses. C. trachomatis inclusion counts were similarly distributed among men with B, C complex, and FG group isolates. The geometric mean number of inclusions per well (range, median) were: B complex, 74.9 (1‐9000, 80); C complex, 80.4 (1‐4,350, 82); and FG group, 67.3 (1‐15,000, 55).
Independent of the serovar, the mean duration of symptoms ± standard deviation (range, n) for men with less than 100 inclusions was 24.1 days ± 26.8 (range, 1‐120 days, n = 39), versus a mean duration of 11.4 days ± 8.6 (range, 1‐30, n = 25) for men with more than 100 inclusions (P = 0.008).
Because inclusion count correlated with duration of symptoms, we undertook an analysis of men stratified into two groups according to their rectal isolate inclusion count (<100 or ≥100 inclusions per well). Men with an unknown inclusion count were excluded, the proportion excluded being similar for all serovars.
The distribution of clinical and laboratory findings among men infected with B complex or FG group isolates stratified by <100 versus ≥100 inclusions is shown in Table 3. Although the number of men infected with C complex isolates was too small to permit a similar analysis, few had signs and symptoms of C. trachomatis rectal infection (data not shown). For men with more than 100 inclusions, selected clinical and laboratory findings were more common among men infected with FG group isolates (Table 3).
Among the 53 men with B complex and more than 100 inclusions, 6 (11.3%) had erythema or bleeding, 2 (3.8%) had mucopus, and 7 (13.2%) had at least one of these findings. Among the 25 men with isolates of the FG group and more than 100 inclusions, 5 (20%) had erythema or bleeding, 5 (20%) had mucopus, and 6 (24%) had at least one of these findings. Five men (9.4%) with B complex had proctitis versus 7 men (28%) with FG isolates (Table 3). Among men with more than 100 inclusions, men with isolates of the FG group versus men with B complex isolates were more likely to have mucopus (OR: 10.5; 95% CI: 1.2‐95.5), more than 15 polymorphonuclear leukocytes (OR: 19.2; 95% CI: 1.7‐219.8) and proctitis (OR: 4.2; 95% CI: 1.1‐16.7) (Table 4). In a subanalysis of men with more than 200 inclusions, men with FG versus those with B complex isolates were more likely to have proctitis (OR: 10.5; 95% CI: 1.7‐65.8). Odds ratio estimates were not affected by adjustment for report of a new partner in the last month, race, marital status, sexual preference, or the report of anal receptive intercourse during the last month.
Temporal Distribution of Serovar Groups
Nearly 60% of the men with rectal chlamydia infection were diagnosed in the 1982 to 1983 interval. However, the relative proportion of each serovar group of isolates remained nearly the same from 1982 to 1990. During each year, about 40% of isolates were B complex, 20% FG group, 10 to 20% C complex, and 20 to 30% were untyped. Over time rectal infection was consistently less frequent among homosexual men older than 35 years of age compared to those who were younger (P < 0.001).
In our previous, smaller study of rectal isolates collected from homosexual men and cervical specimens collected from heterosexual women between 1982 and 1984,15 we reported that serovars D and G were more frequently isolated from rectal than from cervical specimens and serovars E and F were more frequently isolated from cervical than from rectal specimens. It is of interest to note that the pattern of serovar distribution among rectal isolates was stable over the 9 years included in this study.
Previous findings have suggested that lower numbers of viable organisms are present in samples taken from the rectum than in those from other urogenital sites.15 The percentage of men with B, C, FG, and untyped isolates with more than 1,000 inclusions was, respectively, 8%, 14.8%, 8.5%, and 8.3% and was not significantly different by group. In a study of women with cervical infection,16 there were 41%, 50%, and 16%, respectively, of women with B, C, and FG isolates with more than 1,000 inclusions. Inasmuch as all specimens came from the same clinic, were collected by the same clinicians, and assayed in the same laboratory, variation in clinical and laboratory procedures does not explain these differences. Directed sampling of mucopurulent secretions or areas of inflammation may be more difficult in the rectum than the cervix. Estrogenic hormones accelerate the growth of C. trachomatis in cell culture, and thus estrogen in women may enhance adherence and growth of C. trachomatis.27 Alternatively, compounds present in fecal material could inhibit the growth of C. trachomatis, reducing inclusion counts in the rectum.
It is of importance to note that over the study period (1982‐1990), the number of men with rectal chlamydia infection seen in the clinic declined substantially (>80% decline) with a marked decrease after 1985.19 Since 1983, there have been similar substantial declines in reported cases of STDs in homosexually active men in several geographic areas.29–35 Recent analyses among homosexually active men attending the SKCDPH‐STD clinic suggest that the frequency of unsafe sex practices declined greatly in the period 1982 through 1988.19 This decline has been attributed to behavior changes in response to the acquired immunodeficiency syndrome epidemic.
Men infected with C complex strains of C. trachomatis were less likely to report symptoms and to have erythema, bleeding, or mucopus on examination. Because the mean number of inclusions in C complex isolates was comparable to the mean number in B and FG isolates, the C complex serovar does not appear to be less adapted than other serovars to the rectal environment. Anoscopy was performed less frequently among men with C complex isolates which might have resulted in less frequent detection of subtle signs. Because symptomatic men in the clinic were generally examined with anoscopy, it is likely that men with C complex isolates were less symptomatic than men with rectal infection by other C. trachomatis serovars.
Before men were stratified according to number of inclusions detected in rectal isolates, those having isolates of the FG group were not different from men infected with B complex in terms of symptoms, erythema, bleeding, mucopus, number of leukocytes, or proctitis. However, the mixing of more chronic infections with more recently acquired ones might have made it difficult to identify differences between the serovar groups. The stratified data showed that among men with more than 100 C. trachomatis inclusions, those infected with isolates of the FG group versus those infected with isolates of the B complex were more likely to present with mucopus, more than 15 leukocytes, and proctitis. A higher inclusion count could represent a greater dose, a more recent infection, or perhaps a greater susceptibility. Men with less than 100 inclusions might have more chronic and less symptomatic rectal infection.
Although the relationships between time of infection, immune response, duration of symptoms, inclusion count, and clinical manifestations associated with the different serovars remains to be clarified, our results are consistent with the hypothesis that there are pathogenic differences between non‐LGV C. trachomatis strains isolated in rectal infections. Studies in animals also have suggested that there are differences in virulence between strains in the DK serovar group.36 In a study of women, the FG serovar group versus the B or C complex groups was associated with less frequent manifestations of cervicitis and a smaller number of inclusions.16 In that study, 26 (96%) of 27 women with FG group isolates had an isolate of the F type, whereas among 86 men with FG group isolates included in our analysis, 59 (69%) were G and only 27 (29%) were F (Fisher's exact test, P < 0.001). It is possible that there are important biologic differences between the F and G serovars, even if they are recognized as being closely related.25 Recent data suggest that there are important genotypic differences within a particular serovar. For example, among women with serovar F isolates, women with isolates that vary genotypically from the prototype sequence are more likely than those with prototype isolates to have symptomatic and severe endometrial disease.37
In summary, our results show that compared to men with FG group or B complex rectal isolates, men with C complex isolates were least likely to report symptoms, have erythema, bleeding, and/or mucopus. In addition, among the strata of men with an inclusion count of more than 100, men infected with FG group isolates were more likely than men infected with B complex isolates to present with mucopus, more than 15 polymorphonuclear leukocytes, and proctitis. These results are consistent with the hypothesis that there may be intrinsic differences related to serovar in the pathogenicity of strains of C. trachomatis causing genital infection.
1. Bensaude R, Lambling A. Discussion on the aetiology and treatment of fibrous stricture of the rectum (including lymphogranuloma inguinale). Proc R Soc Med 1936; 29:1441–1460.
2. Quinn TC, Goodell SE, Mkrtichiane E, et al. Chlamydia trachomatis proctitis. N Engl J Med 1981; 305:195–200.
3. Dunlop EMC, Vaughan-Jackson JD, Darougar S, Jones BR. Chlamydia infections: incidence in “non-specific” urethritis. Br J Vener Dis 1972; 48:425–428.
4. Goldmeier D, Darougar S. Isolation of Chlamydia trachomatis from throat and rectum of homosexual men. Br J Vener Dis 1977; 53:184–185.
5. Quinn TC, Corey L, Chaffer RG, Schuffer MD, Brancato FP, KK Holmes. The etiology of anorectal infections in homosexual men. Am J Med 1981; 71:395.
6. Quinn TC, Stamm WE. Proctitis, proctocolitis, enteritis and esophagitis in homosexual men. In: Holmes KK, Mardh PA, Sparling PF, et al., eds. Sexually Transmitted Diseases, 2nd ed. New York: McGraw-Hill, 1990: chap 55.
7. Thompson SE, Washington AE. Epidemiology of sexually transmitted Chlamydia trachomatis infections. Epidemiol Rev 1983; 5:96–123.
8. Stamm WE, Quinn TC, Mkrtichiane E, Wang S-P, Schuffler MD, Homes KK. Chlamydia trachomatis proctitis. In: Mardh PA, Holmes KK, Oriel JD, Piot P, Schachter J, eds. Chlamydial infections. Amsterdam: Elsevier Biomedical Press, 1981:1114.
9. Wang SP, Kuo CC, Grayston JT. A simplified method for immunological typing of trachoma-inclusion conjunctivitis lymphogranuloma venerum organism. Infect Immun 1973; 7:356–360.
10. Kuo CC, Wang SP, Holmes KK, Grayston JT: Immunotypes of Chlamydia trachomatis isolates in Seattle, Washington. Infect Immun 1983; 41:865–868.
11. Wang SP, Kuo CC, Barnes RC, Stephens RS, Grayston JT. Immunotyping of Chlamydia trachomatis with monoclonal antibodies. J Infect Dis 1985; 152:791–800.
12. Barnes RC, Wang SP, Kuo CC, Stamm WE. Rapid immunotyping of Chlamydia trachomatis with monoclonal antibodies in a solid-phase enzyme immunoassay. J Clin Microbiol 1985; 22:609–613.
13. 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.
14. Batteiger BE, Lennington W, Newhall WJ, Katz BP, Morrison HT, Jones RB. Correlation of infecting serovar and local inflammation in genital Chlamydia infections. J Infect Dis 1989; 160:332–336.
15. Bames RC, Rompalo AM, WE Stamm. Comparison of Chlamydia trachomatis serovars causing rectal and cervival infections. Infect Dis 1987; 156:953–958.
16. Stamm WE, Stevens CE, Suchland R, et al. Association of infecting serovar with clinical manifestations in Chlamydia trachomatis genital infection in women. In: Bowie RB, Caldwell HD, Jones RB, eds. Chlamydial Infections. Proceedings of the Seventh International Symposium on Human Chlamydial Infections. Cambridge, UK: Cambridge University Press, 1990:303–306.
17. Rompalo AM, Price CB, Roberts PL, Stamm WE. Potential value of rectal screening cultures for Chlamydia trachomatis in homosexual men. J Infect Dis 1986; 153:888–892.
18. 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. J Infect Dis 1995;20:576–581.
19. Handsfield HH, J Schwebke. Trends in sexually transmitted diseases in homosexually active men in King county, Washington, 1980–1990. Sex Transm Dis 1990; 17(4):211–215.
20. Lennette EH, Schmidt NH. Diagnostic Procedures for Viral, Rickettsial and Chlamydial Infections, 5th ed. New York: American Public Health Association, 1979.
21. Blaser MJ, Berkowitz ID, LaForce FM, Cravens J, Reller LB, WangW-LL. Campylobacter enteritis: clinical and epidemiologic features. Ann Intern Med 1979; 91:179–185.
22. Handsfield HH, Knapp JS, Diehr PK, KK Holmes. Correlation of auxotype and penicillin susceptibility of Neisseria gonorrhoeae
with sexual preference and clinical manifestations of gonorrhea. Sex Transm Dis 1980; 7:15.
23. Yoder BL, Stamm WE, Koester CM, Alexender ER. Microtest procedure for isolation of Chlamydia trachomatis. J Clin Microbiol 1981; 13:1036–1039.
24. Suchland RJ, WE Stamm. Simplified microtiter cell culture method for rapid immunotyping of Chlamydia trachomatis. J Clin Microbiol 1991; 29(7):1333–1338.
25. Yuan Y, Zhang YX, Watkins NG, Caldwell HD. Nucleotide and deduced amino acid sequences for the 4 variable domains of the major outer membrane proteins of the 15 Chlamydia trachomatis serovars. Infect Immun 1989; 57:1040–1049.
26. Breslow NE, Day NE. Statistical methods in cancer research. The Analysis of Case-Control Studies, vol 1. Lyon, France: International Agency for Research on Cancer, 1980.
27. Bose SK, Goswani PC. Enhancement of growth of adherence and growth of Chlamydia trachomatis by estrogen treatment of HeLa cells. Infect Immun 1986; 11:265.
28. Bavoil P, Ohlin A, Schacter J. Role of disulfide bonding in outer membrane structure and permeability in Chlamydia trachomatis. Infect Immun 1984; 4:479–485.
29. Judson FN. Fear of AIDS and gonorrhea rates in homosexual men. Lancet 1983; 2:159–160.
30. Poulsen A, Ullman S. AIDS-induced decline of the incidence of syphilis in Denmark. Acta Derm Venereal (Stockh) 1985; 65:567–569.
31. Handsfield HH. Decreasing incidence of gonorrhea in homosexually active men: minimal effect on risk of AIDS. West J Med 1985; 143:469–470.
32. von Krogh G, Hellstrom L, Bottiger M. Declining incidence of syphilis among homosexual men in Stockholm. Lancet 1986; 2:920–921.
33. Carne CA, Weller IV, Johnson AM, et al. Prevalence of antibodies to human immunodeficiency virus, gonorrhea rates, and changed sexual behavior in homosexual men in London. Lancet 1987; 1:656–658.
34. Petersen CS, Sndergaard J, Wantzin GL. AIDS-related changes in pattern of sexually transmitted disease (STD) in an STD clinic in Copenhagen. Genitourin Med 1988; 64:270–272.
35. Landrum S, Beck-Sague C, Kraus S. Racial trends in syphilis among men with same-sex partners in Atlanta, Georgia. Am J Public Health 1988; 78:66–67.
36. Tuffrey M, Falders P, Gale J, Taylor-Robinson D. Salpingitis in mice induced by human strains of Chlamydia trachomatis. Br J Exp Pathol 1986; 67:605–616.
37. Dean D, Oudens E, Bolan G, Padian N, Schacter J. Major outer membrane protein variants of Chlamydia trachomatis
are associated with severe upper genital tract infections and histopathology in San Francisco. J Infect Dis 1995; 172:1013–1022.