NONGONOCOCCAL urethritis (NGU) is one of the most common sexually transmitted diseases (STD) in men attending STD clinics and accounts for an estimated 250,000 to 300,000 initial office visits to physicians annually in the United States.1Chlamydia trachomatis is responsible for 30% to 40% of NGU cases,2 but the etiologies of the remainder are less well defined. Ureaplasma urealyticum may cause 10% to 40% of cases,2–4 although conflicting data exist.5,6Mycoplasma genitalium is a recently discovered organism that has been implicated in 10% to 50% of cases.7 Several other pathogens, such as herpes simplex virus (HSV), Trichomonas vaginalis, Haemophilus species, and adenovirus together account for fewer than 10% of cases.8–11 However, no etiology can be discerned for 20% to 50% of all cases.12–13 Moreover, it remains unclear whether nonchlamydial NGU causes complications in the infected patients or whether there are any associated clinical manifestations or complications in patients' sex partners.
Few studies have addressed the mode of acquisition of nonchlamydial NGU, and the clinical approach and recommendations for partner management generally are based on the tacit assumption that the syndrome is acquired during vaginal intercourse. However, three studies have addressed fellatio as a risk factor for chlamydianegative NGU, suggesting that oral bacteria or other oral factors might cause urethritis in either heterosexual or homosexual men.14–16 In addition, anecdotal reports by experienced clinicians in our STD clinic suggest that some men presenting with urethritis have no recent sexual exposure except for fellatio. A better understanding of risks and predictors of NGU may help guide studies of the etiology, prevention, and management of this common syndrome. Accordingly, this study aimed to test the hypothesis that insertive oral sex is a risk factor for nonchlamydial NGU in heterosexual men and to compare this risk factor with other demographic and behavioral variables. For comparison, predictors of chlamydial and gonococcal urethritis were also analyzed.
Study Design and Population
A retrospective case-control study was conducted among heterosexual men who attended either of the two STD clinics operated by the Seattle-King County Department of Public Health from January 1993 through December 1994. The STD clinics use a computerized medical record containing approximately 380 variables from each patient's chart. To be included in the study, each patient had to indicate to the examining clinician that he was sexually active during the past 2 months. The analysis was limited to exclusively heterosexual men, based on the patient's self-report of sex only with women within the past year, and only each patient's first visit during the study period for assessment of a new STD episode or for screening was analyzed. In addition, only those men from whom urethral specimens were collected for Gram's stain, culture for Neisseria gonorrhoeae, and culture for C. trachomatis were included. Because the etiology and epidemiology of NGU that persists or recurs promptly after antibiotic therapy appear to be different than for initial or isolated episodes of NGU,17 men who gave histories of prior NGU during the study interval (1993 to 1994) or had medical record documentation of a prior NGU diagnosis during the study period were excluded.
All demographic, historical, and behavioral data are obtained from the patients' self-reports to the examining clinicians. Sexual practices in the prior 2 months are recorded as yes/no responses to questions about participation in oral insertive (fellatio), oral receptive, anal insertive, anal receptive, and vaginal insertive intercourse. History of condom use during vaginal intercourse in the preceding 2 months is recorded as always (100% of exposures), most of the time (50% to 99%), sometimes (1% to 49%), or never. For the multivariate model, condom use “always” was compared with all other response patterns because the “most” and “sometimes” categories are highly subjective; it was reasoned that condom use “always” might be a more objective response.
For the purposes of this study, nonchlamydial NGU was defined by the presence of ≥5 polymorphonuclear leukocytes (PMNs) per 1000x microscope field seen on a Gram-stained smear of secretions obtained by urethral swab, plus either a history of compatible symptoms (dysuria or urethral discharge) or the finding of abnormal urethral exudate on physical examination. C. trachomatis was excluded by culture and N. gonorrhoeae by both culture and Gram's stain. In addition, two men with positive urethral cultures for T. vaginalis and 28 men with concurrent clinical diagnoses of first-episode genital herpes, which sometimes is accompanied by herpetic urethritis, were excluded.
Chlamydial urethritis was diagnosed on the basis of a positive urethral culture for C. trachomatis, and gonorrhea was defined by isolation of N. gonorrhoeae by culture. A nonurethritis control group was defined as men who had neither symptoms nor signs of urethritis. This group included men with urethral Gram-stained smears with < 5 PMNs/1000x field or a negative urine leukocyte esterase dipstick test (if urethral PMNs were not recorded) and absence of abnormal urethral discharge on physical examination. Men with incomplete evidence of urethritis, such as having a symptom of urethritis or an abnormal discharge without PMNs seen on a Gram-stained smear, were excluded from both the case and control groups. Pharyngeal cultures for N. gonorrhoeae and rectal cultures for both N. gonorrhoeae and C. trachomatis were performed if indicated by the exposure history or symptoms.
C. trachomatis was isolated by cell culture with identification by fluorescent monoclonal antibody, as previously described.18N. gonorrhoeae was isolated by culture on Thayer-Martin medium and identified by carbohydrate utilization reactions.19 Serological testing for syphilis and human immunodeficiency virus (HIV) was performed using standard methods.
Univariate and multivariate analyses were performed with the aid of SPSS (SPSS Inc., Chicago, IL)20 and Epi Info (USD Inc., Stone Mountain, GA).21 The chi-square test was used for all univariate comparisons of dichotomous variables, and continuous variables were analyzed by the Mann-Whitney U test. Logistic regression was used to evaluate demographic and behavioral predictors of urethritis while controlling for other factors. A backward stepwise procedure was used to choose the final model by eliminating variables not associated with the outcome of interest (P > 0.1).
During the study period, there were 14,527 visits by 11,407 men for evaluation of a new clinical problem or STD screening. There were 4,848 unduplicated heterosexual men who met the criteria for inclusion, of whom 58.1% presented to the STD clinic because of symptomatic complaints and 34.2% requested STD screening without specific symptoms. The remainder presented for diverse reasons, such as sexual exposure to an infected person. Table 1 summarizes the demographic and sexual behavior characteristics of the study population. The median age was 27 years, with a range of 13 to 75 years; 47.9% were white and 36.8% were African-American. These distributions are similar to those of the entire clinic population (data not shown).
Of the 4,848 men in the study population, 836 (17.2%) had nonchlamydial NGU, 307 (6.3%) had urethral chlamydial infection, 311 (6.4%) had urethral gonorrhea, and 2,175 (44.9%) had no urethritis and were analyzed as controls (Table 2). An additional 1,219 men (25.1%) had symptoms or signs suggestive of urethritis but had negative cultures and did not meet the case definition for nonchlamydial NGU and were excluded from subsequent analyses; these patients had demographic characteristics similar to those of the nonchlamydial NGU patients and controls. Forty-two men had positive urethral cultures for both C. trachomatis and N. gonorrhoeae; they are included in the gonorrhea group. Among the entire population of 4,848 men, 328 patients (6.8%) had genital warts, 276 (5.7%) had genital herpes, and 10 (0.2%) were HIV-seropositive. No STD diagnosis was established in 2,327 men (48.0%).
Nonchlamydial, Nongonococcal Urethritis
Table 3 compares the demographic and behavioral characteristics of the nonchlamydial NGU patients and the nonurethritis control patients. The nonchlamydial NGU group was comprised of more African-Americans than the controls and had slightly more sexual partners in the preceding 2 months. The nonchlamydial NGU patients reported always using condoms less frequently than the controls, although use “sometimes” was more frequent. Men with nonchlamydial NGU were less likely to report insertive oral sex than were controls (51.8% vs. 59.4%, P = 0.0001). Nine men with nonchlamydial NGU (1.1%) reported insertive oral sex and no vaginal intercourse compared with 31 control patients (1.4%) (P not significant). Of the nine men with nonchlamydial NGU and oral sex as their only sexual exposure, two had coexistent fungal balanitis and one had folliculitis at the base of the penis.
The multivariate model of risk factors for nonchlamydial NGU included all epidemiologically and biologically plausible variables found to be risk factors (P < 0.1) by univariate analysis. These variables included African-American race, having ≥two sex partners in the prior 2 months, condom use always, fellatio, and insertive anal sex. As shown in Table 4, African-American race was the strongest independent predictor of nonchlamydial NGU, with an odds ratio of 3.71. Having ≥two sex partners in the prior 2 months was also an independent risk factor for nonchlamydial NGU (OR 1.45), and condom use “always” was associated with a reduced risk of nonchlamydial NGU (OR 0.59). Borderline negative associations with nonchlamydial NGU were found for fellatio and insertive anal sex. In men who always used condoms, being African-American and having ≥two sex partners were the only independent risk factors for nonchlamydial NGU (data not shown). Excluding patients with histories of NGU before the study period also did not substantially change the results. Comparing the excluded men who had partial signs and symptoms of urethritis with nonurethritis controls showed no association with orogenital exposure, and did not change the risk factors found for nonchlamydial NGU using the formal definition (data not shown).
Table 5 shows the results of the univariate and multivariate analyses comparing men with chlamydial infection with the control group. Significant independent risk factors for chlamydial urethritis included African-American race (OR 2.93), having ≥two sex partners (OR 1.54), and age ≤24 years (OR 2.69). Fellatio was negatively associated with chlamydial infection (OR 0.69), and condom use “always” was weakly protective (OR 0.67).
Men with gonococcal urethritis had a mean age of 28.1 years, and 73.6% were African-American. As shown in Table 6, independent risk factors for gonorrhea included African-American race (OR 11.42), having ≥two sex partners (OR 2.27), and age ≤24 years (OR 1.62); condom use “always” (OR 0.31) and fellatio (OR 0.59) were negatively associated with gonorrhea. Although being circumcised appeared to protect against gonorrhea in the univariate analysis, there was no apparent effect after controlling for the other independent variables.
The epidemiology and risk factors for chlamydial and gonococcal urethritis have been well studied, but less is known about the causes and predictors of chlamydia-negative NGU. Independent associations of nonchlamydial NGU were found with number of sex partners and failure to consistently use condoms, supporting the sexually acquired nature of nonchlamydial NGU. The association of nonchlamydial NGU with African-American race, as for other STDs, also supports this conclusion. However, we did not confirm the hypothesized association between oral sex and nonchlamydial NGU.
Using the same database analyzed in the present study, Lafferty et al16 reported that oral insertive sex was independently associated with nonchlamydial NGU (OR 2.2, 95% CI 1.3 to 3.7) in homosexual or bisexual men. Hernández-Aguado et al14 studied 56 men with NGU who had negative cultures for C. trachomatis and U. urealyticum at a university-based STD clinic in Seville, Spain, and compared them with 64 controls who had urethral infection with either or both organisms. Insertive oral intercourse was found to be a risk factor for nonchlamydial, nonureaplasmal NGU both in the entire population (OR 11.4, 95% CI 2.4 to 75.1) and in the same population when controlled for homosexual exposure (OR 8.9, 95% CI 2.2 to 35.4). However, many of the men in that study did not have oral sex as their only route of sexual exposure. McGowan et al reported their findings in 51 men who had a history of insertive oral sex as their only unprotected exposure, suggesting that oral sex was the likely route of acquisition of urethritis.15 These investigators also found that some NGU patients harbored urethral bacteria consistent with oropharyngeal flora. Our study was compromised by the fact that almost all the heterosexual men in our population who acknowledged insertive oral sex also had vaginal intercourse in the preceding 2 months, making it difficult to separate the independent contributions of these practices. By contrast, many of the homosexually active men studied by Lafferty et al16 participated in insertive oral but not insertive anal intercourse. Although there were nine men in our analysis who appeared to have no behavioral risk factor for nonchlamydial NGU other than oral insertive sex, they comprised only 1.1% of the men with NGU, a proportion that did not differ significantly from controls. In addition, comparing patients who participated in oral plus vaginal sex with those who participated in vaginal sex only did not change the results (data not shown).
Reported condom use with every episode of vaginal intercourse in the preceding 2 months appeared to be protective against both nonchlamydial NGU and gonorrhea in this study and also demonstrated a trend toward protection against chlamydial infection. These findings confirm those of others,22–24 and there is little doubt that intact condoms, properly used, protect against STDs transmitted between mucosal surfaces. Nevertheless, Zenilman et al25 and Lafferty et al16 found no apparent protective effect of reported condom use in heterosexual and homosexual men, respectively. The apparent discrepancies may depend on the population studied and the extent to which social desirability may affect subjects' responses. Individuals at particularly high risk for STDs and populations subject to societal prejudices may be especially prone to giving socially desirable responses to behavioral questions. This may explain why condom use appeared not to be protective in the Baltimore STD Clinic,25 where most patients were minorities with a high prevalence of substance abuse, or among homosexually active men in our clinic.15 By contrast, the heterosexual men in our study, from diverse socioeconomic backgrounds, may be less likely to give socially desirable responses about condom use.
Several of this study's findings are consistent with those of previous studies. For example, African-American men were more likely to have all types of urethritis than their white counterparts. Also, young age has been associated with chlamydial infection and less strongly with gonorrhea.26–29 The consistency of these findings with expectations and prior studies provides limited assurance that our database and analysis are valid.
This study has several limitations. Most important, a retrospective, case-control study was conducted and the database was not specifically designed to address our hypothesis. The data were collected by several clinicians who used a standardized medical record but did not use standardized wording or styles when conducting their patient interviews. Some cases of chlamydial infection may have been missed by using the culture isolation technique; more sensitive DNA amplification methods now are available.30–32 Our data set did not permit looking for other etiologies of NGU, such as Ureplasma urealyticum and other genital mycoplasmas. STD clinic attendees clearly are not representative of the general population. Also, our STD clinic population may not be representative of other STD clinics or non-STD clinical settings; there is a higher ratio of NGU to gonococcal urethritis among patients seen by office practitioners than in STD clinics.33 On the other hand, although studying a high-risk population may bias results away from the null hypothesis, our control group was selected from this same high-risk population.
In summary, oral sex was not found to be an independent risk factor for nonchlamydial NGU. However, because of the nature of our clinic database and the limitations of the study, this result does not exclude such an association, and the data in men who have sex with men are compelling.14–16 It is plausible that commensal oral bacteria might cause urethritis.15 Alternatively, it is conceivable that noninfectious oral factors may cause urethral inflammation, although the response of most cases of nonchlamydial NGU to antibiotic therapy makes this hypothesis unlikely. Finding a significant association between oral sex and nonchlamydial NGU in a population of heterosexual men may be more difficult than in homosexual men, who are more likely to have oral sex as a single behavioral risk factor. In future studies it might be useful to select men who had insertive oral sex without recent vaginal or insertive anal intercourse. Future investigation also should include both epidemiological research designed to overcome problems common to retrospective studies, such as the inability to evaluate a rare exposure, selection bias, and recall bias, after bias, as well as microbiologic studies, comparing the urethral flora of infected men with the oral flora of their partners.
1. Division of STD Prevention. Sexually Transmitted Disease Surveillance, 1995. US Department of Health and Human Services, Public Health Service. Atlanta, GA: Centers for Disease Control and Prevention; 1996:29.
2. Bowie WR. Urethritis in males. In: Holmes KK, Mårdh P-A, Sparling PF, et al, eds. Sexually Transmitted Diseases. 2nd ed. New York: McGraw-Hill; 1990:627-639.
3. Bowie WR, Wang S-P, Alexander ER, et al. Etiology of nongonococcal urethritis: evidence for Chlamydia trachomatis and Ureaplasma urealyticum.
J Clin Invest 1977; 59:735-742.
4. Bowie WR, Floyd JF, Miller Y, Alexander ER, Holmes J, Holmes KK. Differential response of chlamydial and ureaplasma-associated urethritis to sulphafurazole (sulfisoxazole) and aminocyclitols. Lancet 1976; 2:1276-1278.
5. Lackey PC, Ennis DM, Cassell GH, Whitley RJ, Hook EW III. The etiology of nongonococcal urethritis (abstract). Infectious Diseases Society of America, 33rd Annual Meeting, San Francisco, CA, September 16-18, 1995.
6. Paavonen J, Kousa M, Saikku P, Vesterinen E, Jansson E, Lassus A. Examination of men with nongonococcal urethritis and their sexual partners for Chlamydia trachomatis
and Ureaplasma urealyticum.
Sex Transm Dis 1978; 5:93-96.
7. Taylor-Robinson D. Infections due to species of Mycoplasma
an update. Clin Infect Dis 1996; 23:671-684.
8. Holmes KK, Handsfield HH, Wang SP, et al. Etiology of nongonococcal urethritis. N Engl J Med 1975; 292:1199-1205.
9. 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.
10. Sturm AW. Haemophilus influenzae
and Haemophilus parainfluenzae
in nongonococcal urethritis. J Infect Dis 1986; 153:165-167.
11. Swenson PD, Lowens MS, Celum CL, Hierholzer JC. Adenovirus types 2,8, and 37 associated with genital infections in patients attending a sexually transmitted disease clinic. J Clin Microbiol 1995; 33:2728-2731.
12. McCutchan JA. Epidemiology of venereal urethritis: comparison of gonorrhea and nongonococcal urethritis. Rev Infect Dis 1984; 6:669-688.
13. Lefevre J-C, Lepargneur J-P, Bauriaud R, Bertrand M-A, Blanc C. Clinical and microbiologic features of urethritis in men in Toulouse, France. Sex Transm Dis 1991; 18:76-79.
14. Hernández-Aguado I, Alvarez-Dardet C, Gili M, Perea EJ, Camacho F. Oral sex as a risk factor for chlamydia-negative ureaplasma-negative nongonococcal urethritis. Sex Transm Dis 1988; 15:100-102.
15. McGowan I, Radcliffe KW, Bingham JS. Non-gonococcal urethritis in men practicing “safe” sex. Genitourin Med 1991; 67:70-71.
16. Lafferty WE, Hughes JP, Handsfield HH. Sexually transmitted diseases in men who have sex with men: transmission of urethral infections by fellatio and implications for STD/HIV prevention. Sex Transm Dis 1997;24:272-278.
17. Wong ES, Hooten TM, Hill CC, McKevitt M, Stamm WE. Clinical and microbiological features of persistent or recurrent nongonococcal urethritis in men. J Infect Dis 1988; 158:1098-1101.
18. 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.
19. Bonin P, Tanino TT, Handsfield HH. Isolation of Neisseria gonorrhoeae
on selective and nonselective media in a sexually transmitted disease clinic. J Clin Microbiol 1984; 19:218-220.
20. SPSS for Windows: Advanced Statistics, Release 6.0. SPSS Inc., Chicago, IL, 1993.
21. Epi Info Version 6.0. USD, Inc. Stone Mountain, GA, 1993.
22. Cates W Jr., Holmes KK. Re: Condom efficacy against gonorrhea and nongonococcal urethritis. Am J Epidemiol 1996; 143:843-844 (letter).
23. Barlow D. The condom and gonorrhoea. Lancet 1977; 2:811-813.
24. Weir SS, Feldblum PJ, Zekeng L, Roddy RE. The use of nonoxynol-9 for protection against cervical gonorrhea. Am J Public Health 1994; 84:910-914.
25. Zenilman JM, Weisman CS, Rompalo AM, et al. Condom use to prevent incident STDs: the validity of self-reported condom use. Sex Transm Dis 1995; 22:15-21.
26. Handsfield HH, Jasman LL, Roberts PL, Hanson VW, Kothenbeutel RL, Stamm WE. Criteria for selective screening for Chlamydia trachomatis
infection in women attending family planning clinics. JAMA 1986; 255:1730-1734.
27. Ramstedt K, Forssman L, Giesecke J, Granath F. Risk factors for Chlamydia trachomatis
in 6,810 young women attending family planning clinics. Int J STD AIDS 1992; 3:117-122.
28. Stergachis A, Scholes D, Heidrich FE, Sherer DM, Holmes KK, Stamm WE. Selective screening for Chlamydia trachomatis
infection in a primary care population of women. Am J Epidemiol 1993; 138:143-153.
29. Hook EW III, Holmes KK. Gonococcal infections. Ann Intern Med 1985; 102:229-243.
30. Lee HH, Chernesky MA, Schachter J, et al. Diagnosis of Chlamydia trachomatis
genitourinary infection in women by ligase chain reaction assay of urine. Lancet 1995; 345:213-216.
31. Bauwens JE, Clark AM, Loeffelholz MJ, Herman SA, Stamm WE. Diagnosis of Chlamydia trachomatis
urethritis in men by polymerase chain reaction assay of first-catch urine. J Clin Microbiol 1993; 31:3013-3016.
32. Bianchi A, Scieux C, Brunat N, et al. An evaluation of the polymerase chain reaction Amplicor Chlamydia trachomatis
in male urine and female urogenital specimens. Sex Transm Dis 1994; 21:196-200.
33. Gale JL, Hinds MW. Male urethritis in King County, Washington, 1974-75: I. Incidence. Am J Public Health 1978; 68:20-25.