Nongonococcal urethritis (NGU) is among the most common of all clinical syndromes among patients presenting for sexual health or sexually transmitted disease (STD) clinical services, yet poorly understood.1–5 After 5 decades of inroads in understanding the etiologic roles of Chlamydia trachomatis, Mycoplasma genitalium, possibly some strains of Ureaplasma urealyticum, and occasionally herpes simplex virus (HSV), adenoviruses, and Trichomonas vaginalis, the etiology remains obscure in about half of all cases. This proportion is higher still in men with recurrent or persistent urethritis after antimicrobial treatment, and limited data are available on the likely etiologic differences in relation to vaginal, anal, or oral exposure. Not all NGU is necessarily caused by a specific pathogen. Some cases may have immunologic or other noninfective origins, and some symptoms suggesting urethritis are unaccompanied by objective evidence of inflammation. It has been hypothesized that combinations of normal bacterial flora, or dysbiosis that might follow exposure to a new microbiome, may be responsible for some cases, perhaps with little or no potential for harm to affected men or their partners.2–6 Not surprisingly in these circumstances, empiric treatment with azithromycin or doxycycline is inconsistently effective.5 And the clinical spectrum and appropriate management of the sex partners of men with NGU has been a black box.
Investigators at the Melbourne Sexual Health Centre (MSHC), Australia, and affiliated academic institutions have been productively probing the mysteries of NGU for over a decade. This issue of the journal contains 2 new contributions from them, based on retrospective review of MSHC clinical records. One is relatively straightforward and will directly influence clinical recognition and management of the small minority of NGU patients with herpetic urethritis.7 The other is potentially far reaching but controversial.8
The herpes article7 expands on the authors' observation that herpes is responsible for a small proportion of NGU1 and describes the clinical manifestations in 80 patients with NGU who had positive polymerase chain reaction tests for HSV, the largest case series to date. Compared with 80 men with urethral chlamydial infection, those with herpes more frequently had prominent dysuria, overt meatal inflammation, inguinal lymphadenopathy, constitutional symptoms and, not surprisingly, genital ulcers. Urethral discharge was observed by the current investigators in only 32% of those with herpetic NGU, compared with 69% for chlamydial urethritis.7 HSV1 and HSV2 accounted for 55 (68%) and 25 patients (32%), respectively. Nine (16%) of those with HSV1 urethritis but none with HSV2 had signs or symptoms suggestive of primary herpes (P = 0.03), including multiple, bilaterally distributed genital ulcers, bilateral inguinal lymphadenopathy, and constitutional symptoms. These clinical features are consistent with prior reports that urethritis occurs in up to 40% of primary genital herpes in men.9,10 They also accord with the much lower frequency of recurrent genital herpes due to HSV1 compared with HSV2,11 and it is likely that many of the NGU patients with HSV2 had longstanding infection, whereas most HSV1 infections probably were new. Data on past recurrent herpes are not reported. The retrospective case review design, plus the likelihood that clinicians selectively tested patients with signs or symptoms suggesting herpes—which were previously published from the same clinic1—likely affected the observed clinical spectrum. For example, MSHC clinicians may have selectively performed HSV polymerase chain reaction tests in men with prominent dysuria, meatitis, or scant discharge, or may have been more inclined to diagnose urethritis in those with typical mucocutaneous lesions. These are not serious flaws and the limitations are acknowledged by the authors, who properly cite the need for prospective studies and systematic HSV testing, including serology, to define the true spectrum of herpetic urethritis.
These results should directly influence clinical management of some men with NGU. The authors suggest that clinicians “may consider” empirical antiherpetic therapy, but I would go further and strongly recommend treatment for many patients. Herpetic urethritis often is excruciating or can rapidly become so; treatment efficacy depends strongly on early therapy, especially in persons with initial HSV infection; and acyclovir and related drugs are nearly free of toxicity and side effects.10 Accordingly, I urge a low threshold for such treatment (often with azithromycin or doxycycline) for all men with NGU manifested by prominent dysuria with scant discharge, overt meatal erythema or ulceration, or tender penile edema, and certainly those who also have inguinal lymphadenopathy or constitutional symptoms.9,10 Many such men will have adenoviral NGU,1 but they will not be harmed by antiherpetic therapy. It goes without saying that most of those with multiple mucocutaneous lesions typical for initial herpes should be treated regardless of accompanying urethritis.
In the other study, Ong et al8 reviewed the MSHC database to determine the prevalence of clinical manifestations in the female partners of men with NGU, to address the conundrum of management of the partners of men with NGU not associated with known causative pathogens. In a commentary in response to the Melbourne group's landmark study of NGU etiology,1 I lamented the absence of systematic research on the clinical correlates in partners of men with idiopathic NGU and other syndromes,4 and I thank the authors for crediting my remarks as part of the rationale for the present report.
The study design was insightful and innovative. They analyzed the case records of 1331 heterosexual couples in which both partners attended the clinic the same day, and compared findings in the female partners of 91 men with pathogen-free NGU with those in the partners of 176 men with urethral C. trachomatis infection, 26 with M. genitalium, and 652 with no urethritis. The results showed symptoms of pelvic pain, postcoital bleeding, and dysuria, but not vaginal discharge or vulvovaginal pruritus, to be equally or more prevalent in the partners of men with pathogen-free NGU compared with chlamydial or mycoplasmal NGU and men with no urethritis. One or more urogenital symptoms were present in 70% of the partners of men with idiopathic NGU compared with 57% and 62% of the partners of men with C. trachomatis or M. genitalium, respectively, and 44% of the partners of asymptomatic men without urethritis (adjusted odds ratio, 3.2 for pathogen-free NGU versus asymptomatic men; 95% confidence interval, 1.8-5.5). Symptoms and signs suggesting pelvic inflammatory disease (PID) also were more common in the partners of men with idiopathic NGU than the partners of asymptomatic men (adjusted odds ratio, 4.8; 95% confidence interval, 2.1-11.3). The authors conclude that the female partners of men with pathogen-free NGU are at increased risk of genital infection regardless of identification of a specific pathogen in the woman, that they be contacted to urge clinical evaluation, and that treatment with doxycycline or azithromycin be considered for all female partners, regardless of symptoms or signs of infection.
This is an important, uniquely designed study that finally starts our field along a path toward understanding the clinical counterparts in women of idiopathic NGU in men. Nevertheless, some important limitations—for the most part acknowledged and at least partly addressed by the investigators—dictate caution in interpreting the results. Most important, there is likely referral bias favoring women with genitourinary symptoms, who undoubtedly are more likely to seek care. Further, women may be more likely to perceive, acknowledge, or seek care for symptoms when their partners with NGU divulge their own symptoms. These factors favor selective clinic attendance by symptomatic women, and the results cannot safely be extended to other partners, likely the large majority, who lack symptoms and do not seek care. Further, in most NGU case series, including that reported from MSHC, 25% to 40% of NGU is attributed to C. trachomatis or M. genitalium and up to 60% are pathogen-free.1,2,5 The analyzed NGU cases reverse this ratio, with 69% chlamydial or mycoplasmal and 31% pathogen-free. Because the analytic frame required simultaneous attendance by partners, this distribution is consistent with a lesser likelihood of symptoms in the partners of men with idiopathic NGU, notwithstanding the primary results. In addition, the sample is small compared with the likely number of all men with NGU who attended the clinic during the analytic interval and not necessarily representative of them. The proportion of female partners of men with NGU captured by this study design is smaller still. The authors attempted to address these issues with an ancillary analysis of partners of men with STDs not associated with PID, that is, genital warts or candidiasis, but that analysis is also limited to women seeking STD-related health care and does not eliminate the problem of likely referral bias.
There may be ascertainment bias as well. It is likely that clinicians who evaluated partner pairs sometimes communicated with one another, and in some cases, the same clinician may have examined both partners, perhaps influencing perception and recording of examination findings and diagnosis. Additionally, most of the symptomatic women seeking care probably would be treated regardless of their partners' urethritis. And even a substantial prevalence of treatable genital infections does not necessarily imply that such manifestations would develop without treatment in the remainder. Only a prospective study could make that determination. Finally, the case definitions apparently did not distinguish which member of each partner pair was the index. How many women attended primarily because of their own symptoms, how many informed their partners, and how many of those men, and with what symptoms, accompanied them? How might this have influenced the results?
NGU is not a single clinical entity but a heterogeneous syndrome. Primary examples are the well-documented differences in etiologic spectrum between initial and recurrent NGU,5,12 which in many clinical settings represents a substantial minority of cases2 and, as documented by the Melbourne group,1 between NGU acquired by fellatio versus vaginal (and perhaps anal) intercourse. Ideally, a study such as this would stratify the index cases accordingly. As noted above, NGU sometimes might be a dysbiosis, with no pathogen likely to harm either affected men or their partners, and the present authors and others have described the lesser inflammatory signs and less risky sexual behavior patterns in men with idiopathic compared with chlamydial or mycoplasmal NGU.1,2 Among men with idiopathic NGU in Seattle, 38% and 34% had histories of prior NGU and diagnosed depression, respectively, compared with 16% to 17% of men with chlamydial urethritis,2 both factors elevating suspicion of noninfective etiologies of their current symptoms. In the same study, 46% of the men with pathogen-free NGU were judged to be at low risk for STD. In the present study, the men's urethral discharges were not characterized (amount, purulence) nor are data available on urethral leukocytosis. Notwithstanding the investigators' research1 that led to a decision to not routinely analyze urethral white blood cells at MSHC, most investigators consider this central to documenting urethritis,2,5 and many will view its absence as a problem that limits understanding of the results. The reported clinical spectrum in the female partners is incomplete, as well, probably also owing to limitations in MSHC's clinical database, and the lack of data on objective signs of vaginal discharge (amount, character, amine odor, microscopy) and of syndromes like mucopurulent cervicitis and bacterial vaginosis also limit interpretation. As in most clinical settings and guidelines,13–16 the diagnosis of PID was intentionally broad to maximize sensitivity of the findings but with a decrement in specificity.
At one level, the study is not controversial. Current standard practice is that when female partners of men with NGU attend for care before laboratory test results are available, most are treated with azithromycin or doxycycline. However, the US, UK, European, and Australian guidelines are vague, advising evaluation but with an apparent focus on treatment primarily to cover undiagnosed chlamydial infection.13–16 Despite the guidelines' wording in favor of routine examination and treatment, until now no available data have supported these practices, and another report from MSHC showed they are not consistently followed in Australia and New Zealand,17 which anecdotally seems to be the case in the United States as well. The authors recommend that current guidelines be modified (or perhaps strengthened), perhaps implying that specific programs be implemented and resources devoted to that end. Although they advise only “consideration” of antibiotic treatment, the overall context implies routine treatment of all partners regardless of clinical findings.
In my opinion, these take the interpretation too far. As examples of common exceptions, is treatment indicated for the partner whose only sexual contact with the index case was receptive fellatio, or the partner of a man with recurrent NGU after treatment of both partners, whose clinical examinations and diagnostic tests are normal? With or without this study, probably nobody would argue against advising men with NGU to inform and refer their female partners, but more assertive efforts probably should be limited to the partners of men with defined pathogens. Treatment might be offered as an option to partners of men with pathogen-free NGU, but I am inclined to believe it unnecessary. Finally, no data are available to judge whether partner treatment prevents transmission of urethritis to other partners or reinfection of the index case, primary considerations in partner management for other STDs. As important as these results are, they do not support establishing programs or devoting resources to actively notify all female partners of men with idiopathic NGU. But it is fervently hoped that this study will stimulate programmatic and other research to address these issues. Thanks to Dr. Ong and his colleagues for priming the pump!
1. Bradshaw CS, Tabrizi SN, Read TR, et al. Etiologies of nongonococcal urethritis: bacteria, viruses, and the association with orogenital exposure. J Infect Dis 2006; 193:336–345.
2. Wetmore CM, Manhart LE, Lowens MS, et al. Demographic, behavioral, and clinical characteristics of men with nongonococcal urethritis differ by etiology: a case-comparison study. Sex Transm Dis 2011; 38:180–186.
3. Horner P. The etiology of acute nongonococcal urethritis—the enigma of idiopathic urethritis? Sex Transm Dis 2011; 38:187–189.
4. Handsfield HH. Nongonococcal urethritis: a few answers but mostly questions. J Infect Dis 2006; 193:333–335.
5. Bachmann LH, Manhart LE, Martin DH, et al. Advances in the understanding and treatment of male urethritis. Clin Infect Dis 2015; 61(Suppl 8):S763–S769.
6. Manhart LE, Khosropour CM, Liu C, et al. Bacterial vaginosis-associated bacteria in men: association of Leptotrichia
spp. with nongonococcal urethritis. Sex Transm Dis 2013; 40:944–949.
7. Ong JJ, Morton AN, Henzell HR, et al. Clinical characteristics of herpes simplex virus urethritis compared with chlamydial urethritis among men. Sex Transm Dis 2017; 2:121–125.
8. Ong JJ, Sarumpaet A, Chow EPF, et al. Should female partners of men with nongonococcal urethritis (NGU), negative for Chlamydia trachomatis
and Mycoplasma genitalium
, be informed and treated? Clinical outcomes from a partner study of heterosexual men with NGU. Sex Transm Dis 2017; 2:126–130.
9. Corey L, Adams HG, Brown ZA, et al. Genital herpes simplex virus infections: clinical manifestations, course, and complications. Ann Intern Med 1983; 98:958–972.
10. Corey L, Wald A. Genital Herpes. Chap. 4 in Holmes KK, Sparling PF, Stamm WE, et al. [eds]. Sexually Transmitted Diseases, 4th edition. New York, McGraw-Hill, 2008:399–437.
11. Lafferty WE, Coombs RW, Benedetti J, et al. Recurrences after oral and genital herpes simplex virus infection. Influence of site of infection and viral type. N Engl J Med 1987; 316:1444–1449.
12. Wong ES, Hooton TM, Hill CC, et al. Clinical and microbiological features of persistent or recurrent nongonococcal urethritis in men. J Infect Dis 1988; 158:1098–1101.
14. Horner P, Blee K, O'Mahony C, et al.; Clinical Effectiveness Group of the British Association for Sexual Health and HIV. UK national guideline on the management of non-gonococcal urethritis. Int J STD AIDS 201627:85–96.
15. Horner PJ, Blee K, Falk L, et al. 2016 European guideline on the management of non-gonococcal urethritis. Int J STD AIDS 2016; 27:928–937.
17. Teague R, Fairley CK, Newton D, et al. How men with non-chlamydial, non-gonococcal urethritis are managed in Australasia. Int J STD AIDS 2008; 19:581–585.