Discussion

In addition to those previously reported in the current literature, we have presented a further eight cases where adenovirus was identified as the cause of urethritis and bilateral conjunctivitis. Specific clinical features have previously been recognised that should alert the clinician to considering adenovirus as the cause of NGU13 ,14 ,17 and our case series adds further support to this. Tables 1⇑–3 demonstrate the many similarities that occur in patients diagnosed with concurrent adenovirus infection of the urethra and conjunctiva.

Patients typically present with dysuria which can be severe,13–15 ,17 clear or mucoid urethral discharge and evidence of meatitis. Perhaps unexpectedly, constitutional symptoms were common, occurring in 20% of cases (table 2). In our eight cases, conjunctivitis was bilateral, spreading from one eye to the other 2–5 days thereafter (figure 1). Conjunctivitis was also bilateral in a further three cases but it was not commented on in the remaining 29 cases. All of our cases had more than five polymorphonuclear leucocytes per high-powered field and this was a consistent finding in other case reports (table 3). Azariah and Reid14 and Tønsberg and Hartgill18 comment on the identification of large numbers of mononuclear cells on urethral smear and postulate that this could indicate a viral aetiology. We did not look at this in our case series; however, if the patient is symptomatic but does not have microscopic evidence of urethritis, then finding mononuclear cells may alert the clinician to consider adenovirus as the cause.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Photograph of one of our cases presenting with right sided conjunctivitis.

Oral sex is the likely route of acquisition of urethral adenovirus infection.13–17 The virus can be transmitted by direct exposure to infected tissue. Furthermore, oral secretions infected with adenovirus contain very high levels of viral particles (100 000–1 000 000/mL). Of the case reports that stated the type of sex, including our eight cases, 100% of patients had received oral sex in the 2–28 days preceding the onset of symptoms. Adenovirus can also be spread by autoinoculation which could explain the development of conjunctivitis following initial urethral infection. Vaginal sex was also commonly reported (table 2). Adenovirus has been isolated from the female genital tract on several occassions7 ,12 ,21–23 and therefore provides another possible route for acquiring urethral infection.

Of the 40 adenovirus DNA positive samples, successful molecular serotyping was performed in 27 (table 3). Subgroup D was identified in 26 samples and there was just one subgroup B2 virus. Analysis of sequences of both hexon gene regions from an eye and a penile swab from one of our cases showed 100% similarity to adenoviruses in species D type 56. In a second case, samples from the eye, urine and urethra showed 99%–100% similarity to two adenoviruses in species D types 19 and 64. We were unable to discriminate further exactly which serotype was responsible for the patient’s symptoms on the basis of the sequence data obtained. Azariah and Reid also reported cross reaction of more than one adenovirus serotype in their study.14 Of note, type 64 has not been previously identified in the literature as causing urethritis and/or conjunctivitis. Further research is required to determine if infection with type 64 can result in urethritis and conjunctivitis. The commonest serotypes overall were 19 and 37. Subgroup D adenoviruses are well known to cause keratoconjunctivitis,3 and types 8, 19 and 37 have been responsible for large epidemics.24 This review supports the role of adenovirus subgroup D, and in particular types 19 and 37, as the cause of concurrent urethritis and conjunctivitis.

Subgroup B2 also has an affinity for the genital tract but is not thought to cause conjunctivitis. In another of our cases, a sequence from one of the targeted hexon gene regions from a conjunctival swab showed closest genetic similarity (99%) to two adenoviruses in species B2, types 11 and 35. Again, we were unable to distinguish which serotype was causative. This does however highlight another potential clinical disease that subgroup B adenoviruses are responsible for that has not previously been reported.

Adenovirus infections are endemic throughout the year. However, 14 (74%) of the cases where information was available occurred in the autumn and winter months in comparison with only 5 (26%) in the spring and summer (table 1). Although this does not provide conclusive evidence that adenovirus typically causes concurrent urethritis and conjunctivitis in the autumn and winter, it may be prudent for clinicians to have a high index of suspicion during this time.

Diagnosing adenovirus as the cause of NGU allows the clinician to reassure the patient that the urethritis and conjunctivitis are self-limiting. The mean length of time for symptoms to resolve was 14.2 days. Adenovirus is contagious, and therefore advice concerning good hand hygiene, avoidance of sharing towels and no sexual contact until symptoms have resolved should be provided. Artificial tears can help with the symptoms of conjunctivitis. Antibiotics are not indicated. Testing for adenovirus DNA should be considered in patients who have ongoing symptomatic NGU despite initial antibiotic treatment if not performed at the first visit.