SCREENING PROGRAMS for the detection and treatment of Chlamydia trachomatis infection in women have been associated with marked declines in chlamydial prevalence and a reduced incidence of symptomatic pelvic inflammatory disease. 1–7 Screening programs have focused mainly on women because most complications occur in this population, and 70% to 90% of infected women are asymptomatic. However, recent studies suggest that 40% to 60% of infected males also lack symptoms. 8–11 Failure to identify such infections in men presumably allows for the maintenance of a large untreated reservoir of chlamydial infection in the population. 12 However, until accurate urine-based testing for C trachomatis became available, efforts to screen and treat males have been hampered by the need to obtain diagnostic specimens using urethral swabs.
Nucleic acid-amplification tests (NAAT) for C trachomatis, including ligase chain reaction (LCR), polymerase chain reaction, and transcription-mediated amplification, are sensitive and specific for the diagnosis of chlamydial urethral infection in males. 10,11,13–16 The additional yield in number of infected persons identified using NAAT compared with culture appears to be greater among males with urethral infection than among women with cervical infection. 10,11,13 Most importantly, first-catch urine specimens provide a more acceptable testing method for asymptomatic males. Given these considerations, screening programs for chlamydial infections in men using urine NAAT are being implemented. 17–20 However, NAAT are considerably more expensive than less-sensitive tests, and the cost of universal testing of men using NAAT is prohibitive for most clinical facilities, including public clinics. Thus, selective screening criteria for NAAT testing in sexually transmitted disease (STD) clinic settings are needed. Although a number of studies have established prevalence thresholds for cost-effective universal screening of women at 3% to 6%, appropriate thresholds for the screening of men have not been established.
During two study periods, we performed both urine LCR and urethral culture for chlamydia in male STD clinic clients to define how best to use the urine LCR assay in a clinic with moderate chlamydia prevalence.
The study included heterosexual men who denied symptoms of urethral discharge or dysuria from October 1, 1995 to April 30, 1997 at an urban public health STD clinic. In addition, all male clients presenting for new problems between May 1, 1997 and October 31, 1997 (including those with symptoms) were recruited. Clinicians instructed subjects to provide the first 10 to 30 ml of voided first-catch urine in a plastic cup for C trachomatis LCR testing and leukocyte esterase (LE) testing. Two urethral swabs were collected. The first swab, which was inserted 1 cm to 2 cm into the distal urethra, was used for isolation of C trachomatis in tissue culture. The second swab, which was collected from the distal 1 cm of the urethra, was used for Gram stain and Neisseria gonorrhoeae culture. Ligase chain reaction was performed according to the manufacturer’s instructions (Abbott Laboratories, Chicago, IL) and culture for C trachomatis was performed by inoculating specimens into cyclohexamide-treated McCoy cells grown in 96-well microtiter plates, as previously described. 21 Inclusions were identified at 72 hours with fluorescent staining of elementary bodies. Leukocyte esterase testing was performed by clinicians or laboratory personnel using a dipstick (Miles, Elkhart, IN). Gram-stained smears of urethral swab samples were performed by clinicians, and LE tests of first-catch urine specimens were performed either by clinicians or by the laboratory personnel who performed urine LCR testing. A positive LE test result was defined as a reading of 1+ or greater. Symptoms considered suggestive of chlamydial infection included reports of urethral discharge or dysuria; men without these symptoms were defined as asymptomatic. Males with a positive culture for N gonorrhoeae (performed only for men with urethral discharge on examination) were included and accounted for in the analyses.
For all subjects, chlamydia prevalence as defined by results of urine LCR and urethral culture was stratified by results of the urethral Gram stain. Statistical associations for dichotomous variables were assessed by chi-square analysis and continuous variables were compared with the t test. Univariate and multivariate associations of chlamydial infection were analyzed, and logistic regression was used to perform multivariate analyses. All independent variables (age, number of sex partners in the previous 2 months, number of new sex partners in the last 2 months, condom use at last sex, and either LE test or urethral Gram stain results) were entered into the model, and P < 0.05 was designated as the level of significance. Interactions between independent covariates were assessed.
Among asymptomatic males, the yields of three approaches to screening for chlamydial infection were considered: (1) test all men using urine LCR, (2) test all men using urethral culture, and (3) use predictors for selecting men for urine LCR testing, including urethral Gram stain results, urine LE, age, and historical risk factors.
Characteristics of Subjects and Results of LCR Testing
During the 6-month period (May–October 1997), 668 symptomatic and 2,366 asymptomatic men were seen, of whom 626 (94%) and 1,528 (64.6%) had urine collected for LCR, respectively. In addition, 690 asymptomatic men who reported having sex with women only were specifically recruited and tested with urine LCR from May 1, 1996 to April 30, 1997. Asymptomatic men who had urine LCR performed during the two periods (2,196) did not differ in number of total or new sex partners in the previous 2 months, but those recruited during the earlier period were slightly younger (mean age, 29.7 years versus 32.0 years;P < 0.001) and had a higher prevalence of chlamydia as measured by either urine LCR or urethral culture (6.1% versus 4.0%;P = 0.003). The two groups differed in their report of sex with men, given the specified recruitment for the earlier time period; of subjects seen from May 1, 1996 to April 30, 1997, 15.9% reported sex with men.
In total, 71.8% of asymptomatic men presenting for new problems during the study period were included in the study. Reasons for the remaining 28.2% not being tested for chlamydial infection included visit for HIV test only (29.6%), referral for treatment with a known positive test result or contact with an STD (3.4%), and failure of the clinician to obtain a urine specimen (67%). Men not tested with urine LCR were not more likely to have urethral discharge on examination, but were less likely to have Gram stain of urethral secretions performed (P = 0.04) and to report a new sex partner in the past 60 days (P = 0.05).
Among subjects who had both urine LCR and urethral culture, chlamydia prevalence as defined by either test showing a positive result was 7.4% (5.5% among 1,625 asymptomatic men, 12.8% among 569 symptomatic men). Urethral Gram stain results were available for 1,581 (97.3%) of asymptomatic men and 555 (97.5%) of symptomatic men who had both tests performed. Table 1 depicts the chlamydia prevalence by test type among these men stratified by symptoms, urethral Gram stain findings, and results of culture for N gonorrhoeae, and shows the additional yield of urine LCR over culture for each subgroup. Urethral cultures for N gonorrhoeae, which were performed on 524 symptomatic men (92%), were positive in 68 (13%). Use of LCR yielded an increase in overall detection of C trachomatis of 39% and increased chlamydia detection in all subgroups. The relative increase in prevalence using LCR was highest among men with > 5 polymorphonuclear leukocytes per high-power (1000×) field (PMN/HPF) on urethral Gram stain regardless of symptoms, and in symptomatic men with gonorrhea diagnosed by urethral culture. Of note, prevalence as defined by urine LCR in asymptomatic men with > 5 PMN/HPF was 20.1%. Of all 2,194 men tested with both urethral culture and LCR, 14 had a positive result by culture and a negative result by urine LCR; of these 14 men, 7 (50%) were asymptomatic and 7 (50%) were symptomatic. On urethral Gram stain, five men had < 5 PMN/HPF and eight men had >5 PMN/HPF (one subject did not undergo Gram stain analysis).
Among men with positive cultures for C trachomatis, the number of inclusion-forming units in culture was strongly associated with the likelihood of a positive LCR result (P = 0.001, chi-square test for trend) (Table 2). Similarly, there was a trend toward the increased positivity of urine LE and urethral Gram stain with increasing inclusion-forming units; however, the number of cultures with >1000 inclusion-forming units was small, thereby reducing the precision of the latter estimates.
Specific Findings Among Asymptomatic Men
Among the 2,196 asymptomatic men tested with urine LCR, prevalence was highest among men 19 to 20 years (10.4%), regardless of collection of specimen for culture. In univariate analysis, age of 25 years or younger conferred a threefold risk of infection; neither number of sex partners in the previous 2 months nor condom use during last sex were significantly associated (Table 3). Report of a new partner in the previous 2 months was associated with a reduced likelihood of infection as detected by LCR. Both a positive urine LE test result and ≥ 5 PMN/HPF on urethral Gram stain were strongly associated with infection. Neither antibiotic use in the previous month nor prior history of nongonococcal urethritis or gonorrhea was associated with increased risk of infection (data not shown). In multivariate analysis, age of younger than 25 years and positive urine LE test result or ≥5 PMN/HPF on urethral Gram stain remained significantly associated with an increased risk of infection, conferring threefold and nearly sevenfold risks, respectively (Table 3). Report of a new partner in the past 2 months remained associated with a reduced risk.
Urethral Inflammation as a Predictor of Infection
Table 4 compares the performances of the urine LE test and the urethral Gram stain in predicting chlamydial infection as detected by urine LCR in the subgroup of men who reported no male sex partners. The urine LE test was performed for 709 (37.8%) of these men, and urethral Gram stain for 1,423 (75.7%). Asymptomatic men for whom an LE test or Gram stain was performed did not differ from those who did not undergo such analyses in age, chlamydia prevalence, report of new sex partners, or the presence of urethral discharge on examination. The positive predictive value of the urine LE test for a finding of ≥ 5 PMN/HPF on Gram stain was 53.3% (57/107 men); among 661 men who had < 5 PMN/HPF on Gram stain, 515 had a negative urine LE test result (negative predictive value, 77.9%). The specificity and positive predictive value of the Gram stain were higher than urine LE in all age groups. Sensitivity of both tests in predicting a positive LCR test result was highest in younger men (< 25 years). The positive predictive value of both LE and Gram stain was lower in men older than 30 years (5.1% and 10.9%, respectively), but the negative predictive value was higher (98.6% and 98.9%, respectively). The performance of urine LE in predicting chlamydial infection was not affected by whether the test was performed by clinicians or a laboratory technician (data not shown).
Performance of Screening Strategies Among Asymptomatic Men
The results of various screening strategies using LCR testing in asymptomatic men are summarized in Table 5. In our study population, simply testing all men younger than 25 years would have detected 56% of all infections and required the screening of 29% of all men, with an undetected prevalence of 2.7% among men 25 years or older. Ligase chain reaction testing of men who were younger than 25 years or had a positive urine LE result would have resulted in the detection of 84% of the 38 chlamydial infections that occurred in the group for which both tests were performed, and would have required the screening of 51% of men. Limiting LCR testing to asymptomatic men with a positive urine LE test result in the younger age group would have detected fewer infections (42% of all infections), but would have required testing only 8% of the population. Substituting urethral Gram stain findings as a measure of inflammation in these approaches did not change the results. Chlamydia prevalence in men 25 years or younger with < 5 PMN/HPF on urethral Gram stain was 3.5%, similar to that among men who had a negative urine LE test result. Increasing the age cutoff for screening to 30 years or younger significantly increased the proportion of infections detected (78%) and would have required the testing of 53% of subjects. Incorporating behavioral risks (new or increasing numbers of recent sex partners, condom use at last sex) did not improve the performance of any screening algorithm (data not shown).
The use of urine LCR afforded a substantial overall increase in the diagnosis of C trachomatis infection (39%) in men attending our STD clinic. This increase was highest in men without urethral inflammation as detected by urine LE or Gram stain (< 5 PMN/HPF), regardless of the presence of symptoms. In these men, the use of urine LCR increased the detection of C trachomatis by 79% to 90%. The increased detection of chlamydia by LCR considerably exceeds that previously identified in studies of women or symptomatic men (10–15%). 10,11,13–16 These findings are not surprising, however, because the chlamydia culture was the least sensitive in urethral specimens from asymptomatic men. Inclusion counts in cell culture from urethral samples obtained from asymptomatic men are 10-fold to 100-fold lower than those found in specimens from symptomatic men or from women. Additionally, inhibition of the LCR reaction may be greater in endocervical specimens than in male urine specimens as tested here. The presence of inhibition in some male urine specimens, however, is suggested by the culture-positive, LCR-negative specimens observed in our study, most of which had low inclusion-forming units.
Although the increased yield of urine LCR testing may be highest in men without urethral inflammation, infections in these men are also likely characterized by low organism burdens; whether such infections are less transmissible than those characterized by symptoms is not clear. In our study, screening asymptomatic men with either the urine LE test or urethral Gram stain performed best among men older than 30 years, for whom the negative predictive value of both tests was > 98%. We were unable to define sensitive and specific selective screening criteria for men 30 years or younger beyond screening for urethral inflammation; 17% to 30% of younger men with inflammation detected by either urine LE or urethral Gram stain, respectively, had C trachomatis detected by urine LCR.
Reasoning that symptomatic men would in general be tested or treated empirically for C trachomatis, we focused on selective screening approaches in asymptomatic men. Our data indicate that among asymptomatic men, the value of the urine LE and urethral Gram stain tests are modified by age. The positive predictive values of both tests decreased markedly and the negative predictive values increased in men older than 25 years. This finding likely reflects both a higher chlamydia prevalence in younger men and perhaps an increase in other etiologies of urethral inflammation among older men. One practical implication of our data is that because the additional specificity afforded by the urethral Gram stain over the urine LE test was highest in asymptomatic older men, the urine LE test may be a reasonable substitute in asymptomatic younger men if performance of the urethral Gram stain is not possible. The urethral Gram stain may remain the test of choice in the assessment of symptomatic men, particularly to look for evidence of infection with N gonorrhoeae.
Unexpectedly, asymptomatic subjects who reported one or more new sex partners in the previous 2 months were marginally less likely to have chlamydial infection as defined by urine LCR test, even when other factors were controlled for. This finding contrasts with that reported in most studies of risk factors for chlamydial infection in women. 12 We hypothesize that the asymptomatic men in our study may have had a more remote acquisition of C trachomatis than might be reflected in their recent sexual history. Further, male STD clients reporting higher numbers of partners may be more likely to access STD clinics for care and may consequently be more likely to have tests for C trachomatis performed, resulting in a lower likelihood of infection at subsequent evaluation.
Implementation of selective screening protocols for chlamydial infection in women have been associated with both a reduced prevalence of C trachomatis and a reduced likelihood of symptomatic pelvic inflammatory disease in the year after screening. 2–6 Less information is available regarding the epidemiologic characteristics of chlamydial infection among asymptomatic men and the effects of screening among men, largely because access to populations of men willing to undergo urethral swab sampling has been limited. Young age and report of fewer than five lifetime sex partners have been associated with an increased likelihood of chlamydial infection in male STD clinic attendees, 22,23 but among 100 asymptomatic adolescent males attending an adolescent medicine clinic, the only significant independent predictor of infection was a positive urine LE test result. 24 Further, the age-specific prevalence of C trachomatis infection may peak at an older age in males than in females, 17,18 necessitating the use of higher age cutoffs for selective screening recommendations in males. Other studies have also suggested that urine LE testing or other measures of pyuria to select men for NAAT testing may be relatively insensitive in predicting chlamydial infection. 17,24–30 However, the low cost of the LE test and the high cost of NAAT may render urine LE screening cost-effective at a low chlamydia prevalence, even with its low sensitivity and positive predictive value. 30–33 Others have also found that urine LE testing may be more sensitive in predicting chlamydial infection in younger men. 29
Limitations of this study include the small numbers of men with chlamydial infection in some subgroups, and that urine LE testing was not performed in all subjects. The selection of clients for LE testing may have been influenced by factors not included in our database (i.e., perceived risk of STD by clinicians, subjects’ willingness to have urethral swabs obtained). However, the performance of urine LE in predicting C trachomatis infection did not differ when subgroups in which only one test was performed (LCR or culture) were analyzed, and asymptomatic men who had the LE test or Gram stain performed did not differ from those who did not under go such analyses in age, chlamydia prevalence, or number of new sex partners. Neither did all subjects receive both LCR and urethral culture, as originally planned. There may have been undefined biases inherent in the clinicians’ decision to test with both modalities. Finally, our conclusions about selective screening criteria may not be generalizable to populations other than STD clinic attendees.
In conclusion, the relative increase in detection of chlamydia by urine LCR over culture was highest in men with no urethral inflammation evident on Gram stain, regardless of the presence of symptoms. Among culture-positive men, there was a direct relation between increasing inclusion count and a positive LCR test result, and a similar relation was observed between positive LCR and urine LE or PMNs detected on urethral Gram stain. The positive predictive value of the urine LE test in predicting chlamydial infection was highest in younger men. In men 30 years or older, a negative urine LE test result or < 5 PMN/HPF on urethral Gram stain supports not testing for C trachomatis. Although an age of younger than 30 years was associated with a higher prevalence of chlamydial infection in asymptomatic men, no additional selective screening criteria had both high sensitivity and positive predictive value in this group. However, sensitivity was improved by combining younger age and urethral inflammation with direct testing. If testing all men 30 years and younger is not economically feasible, lowering the age cutoff for screening to 25 years and younger or requiring urethral inflammation as detected by either the urine LE test or urethral Gram stain could help to prioritize urine LCR testing.
In devising selective screening criteria for asymptomatic males, clinics need to consider their age-specific chlamydia prevalence, whether urethral Gram stain is routinely performed, and the resources available for diagnostic testing. Prioritizing the use of the LCR test also depends on whether the goal is to maximize the likelihood of making a specific diagnosis of chlamydia in asymptomatic men (even though these men might be treated empirically if they have urethral inflammation) or to expand testing in men who would not otherwise be empirically treated for urethritis. One could argue that different public health agendas would be addressed with both goals. The former goal could result in more intensive targeting of female partners of men if they receive a specific chlamydia diagnosis, whereas the latter would obviously result in detection of infection in men who would not otherwise be empirically treated.
1. Cates W, Wasserheit JN. Genital chlamydial infections: epidemiology and reproductive sequelae. Am J Obstet Gynecol 1991; 164: 1771–1781.
2. Addiss DG, Vaughn ML, Ludka D, Pfister J, Davis JP. Decreased prevalence of Chlamydia trachomatis
infection associated with a selective screening program in family planning clinics in Wisconsin. Sex Transm Dis 1993; 20: 28–35.
3. Katz BP, Blythe MJ, Van der Pol B, Jones RB. Declining prevalence of chlamydial infection among adolescent girls. Sex Transm Dis 1996; 23: 226–229.
4. Marrazzo JM, Fine D, Celum CL, DeLisle S, Handsfield HH. Selective screening criteria for chlamydial infection in women: a comparison of CDC screening recommendations with age and risk-based criteria. Fam Plann Perspect 1997; 29: 158–162.
5. Scholes D, Stergachis A, Heidrich FE, Andrilla H, Holmes KK, Stamm WE. Prevention of pelvic inflammatory disease by screening for cervical chlamydia infection. N Engl J Med 1996; 334: 1362–1366.
6. Hillis SD, Nakashima A, Amsterdam L, et al. The impact of a comprehensive chlamydia prevention program in Wisconsin. Fam Plann Perspect 1995; 27: 108–111.
7. Centers for Disease Control, Prevention. Chlamydia trachomatis
genital infections—United States, 1995. MMWR Morb Mortal Wkly Rpt 1997; 46: 193–198.
8. Pack RP, DiClemente RJ, Hook EW III, Oh MK. High prevalence of asymptomatic STDs in incarcerated minority male youth: a case for screening. Sex Transm Dis 2000; 27: 175–177.
9. Jones CA, Knaup RC, Hayes M, Stoner BP. Urine screening for gonococcal and chlamydial infections at community-based organizations in a high-morbidity area. Sex Transm Dis 2000; 27: 146–151.
10. Quinn TC, Welsh L, Lentz A, et al. Diagnosis by AMPLICOR PCR of CT in urine samples from women and men attending sexually transmitted disease clinics. J Clin Microbiol 1996; 34: 1401–1406.
11. Toye B, Peeling RW, Jessamine P, Claman P, Gemmill I. Diagnosis of Chlamydia trachomatis
infections in asymptomatic men and women by PCR assay. J Clin Microbiol 1996; 34: 1396–1400.
12. Marrazzo JM, Stamm WE. New approaches to the diagnosis, treatment, and prevention of chlamydial infection. In: Remington JS, Swartz MN, eds. Current Clinical Topics in Infectious Diseases. 18th ed. Boston: Blackwell Science, 1998: 37–59.
13. Chernesky MA, Jang D, Lee H, et al. Diagnosis of Chlamydia trachomatis
infections in men and women by testing first-void urine by ligase chain reaction. J Clin Microbiol 1994; 32: 2682–2685.
14. Mahony JB, Luinstra KE, Sellors JW, Jang D, Chernesky MA. Confirmatory polymerase chain reaction testing for Chlamydia trachomatis
in first-void urine from asymptomatic and symptomatic men. J Clin Microbiol 1992; 30: 2241–2245.
15. 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.
16. Stary A, Tomazic-Allen S, Choueiri B, Burczak J, Steyrer K, Lee H. Comparison of DNA amplification methods for the detection of Chlamydia trachomatis
in first-void urine from asymptomatic military recruits. Sex Transm Dis 1996; 23: 97–102.
17. Marrazzo JM, White CL, Krekeler B, et al. Community-based urine screening for Chlamydia trachomatis
with a ligase chain reaction assay. Ann Intern Med 1997; 127: 796–803.
18. Cohen, DA, Nsuami M, Etame RB, et al. A school-based chlamydia control program using DNA amplification technology. Pediatrics 1998; 101:E1.
19. Rietmeijer CA, Yamaguchi KJ, Ortiz CG, et al. Feasibility and yield of screening urine for Chlamydia trachomatis
by polymerase chain reaction among high-risk male youth in field-based and other nonclinic settings. Sex Transm Dis 1997; 24: 429–435.
20. Gunn RA, Podschun GD, Fitzgerald S, et al. Screening high-risk adolescent males for Chlamydia trachomatis
infection. Sex Transm Dis 1998; 25: 49–52.
21. Stamm WE, Tam MR, Koester M, Cles LD. Detection of Chlamydia trachomatis
inclusions in McCoy cell cultures with fluorescein-conjugated monoclonal antibodies. J Clin Microbiol 1983; 17: 666–668.
22. Rietmeijer CA, Judson FN, Van Hensbroek MB, Ehret JM, Douglas JM Jr. Unsuspected Chlamydia trachomatis
infection in heterosexual men attending a sexually transmitted diseases clinic: evaluation of risk factors and screening methods. Sex Transm Dis 1991; 18: 28–35.
23. van Duynhoven YT, van de Laar MJ, Fennema JS, van Doornum GJ, van Den Hoek JA. Development and evaluation of screening strategies for Chlamydia trachomatis
infections in an STD clinic. Genitourin Med 1995; 71: 375–381.
24. Johnson J, Neas B, Parker DE, Fortenberry JD, Cowan LD. Screening for urethral infection in adolescent and young adult males. J Adolesc Health 1994; 15: 356–361.
25. Anestand G, Berdal BP, Scheel O, Mundal R, Odinsen O, Skaug K. Screening urine samples by leukocyte esterase tests and ligase chain reaction for chlamydial infections among asymptomatic men. J Clin Microbiol 1995; 33: 2483–2484.
26. Braverman PK, Biro FM, Brunner RL, Gilchrist MJ, Rauh JL. Screening asymptomatic adolescent males for chlamydia. J Adolesc Health Care 1990; 11: 141–144.
27. Werner MJ, Biro FM. Urinary leukocyte esterase screening for asymptomatic sexually transmitted disease in adolescent males. J Adolesc Health 1991; 12: 326–328.
28. McNagny SE, Parker RM, Zenilman JM, Lewis JS. Urinary leukocyte esterase test: a screening method for the detection of asymptomatic chlamydial and gonococcal infections in men. J Infect Dis 1992; 165: 573–576.
29. Patrick DM, Rekart ML, Knowles L. Unsatisfactory performance of the leukocyte esterase test of first voided urine for rapid diagnosis of urethritis. Genitourin Med 1994; 70: 187–190.
30. Bowden FJ. Reappraising the value of urine leukocyte esterase testing in the age of nucleic acid amplification. Sex Transm Dis 1998; 25: 322–326.
31. Randolph AG, Washington AE. Screening for Chlamydia trachomatis
in adolescent males: a cost-based decision analysis. Am J Public Health 1990; 80: 545–550.
32. Shafer M-A, Schachter J, Moncada J, et al. Evaluation of urine-based screening strategies to detect Chlamydia trachomatis
among sexually active asymptomatic young males. JAMA 1993; 270: 2065–2070.
33. Genç M, Ruusuvaara L, Mårdh P-A. An economic evaluation of screening for Chlamydia trachomatis
in adolescent males. JAMA 1993; 270: 2057–2064.