• 
• 

Three Part Question

In [sexually active males having been diagnosed with gonococcal urethritis] is [the prevalence of co-infection with Chlamydia trachomatis][enough to justify epidemiological treatment]?

Clinical Scenario

A 21-year-old man has been diagnosed with gonococcal urethritis. You are advised to give treatment to cover chlamydia infection as well and you wonder if this is necessary.

Search Strategy

Medline - 1966 to June week 2 2006
Embase - 1980 to 2006 Week 24
CINAHL - 1982 to June Week 2 2006
The Cochrane Library 2006 Issue 2
Medline/Embase/CINAHL: [Exp Urethritis OR gonococcal.mp OR Gonorrhoea.MP OR neisseria gonorrhoeae.mp OR gonorrhea] AND [exp chlamydia OR exp chlamydia trachomatis OR non gonoccoccal urethritis.mp OR NGU.mp] AND [antibiotic$ OR exp Anti-bacterial agents] limit to Humans, English language and males.
Cochrane - (Neisseria gonorrhoeae [MESH] OR (gonorrhoea) OR (gonococcal) OR Chlamydia [MESH] OR (chlamydia trachomatis) OR (non-gonococcal)) AND (antibiotic)

Search Outcome

Altogether 274 papers were identified by Medline, 253 by Embase and 6 by CINAHL. Of these 2 were relevant from Medline, and a further 2 were relevant from Embase.
Cochrane: 161 were identified. None relevant.
One further article was found in a manual journal search.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Das S, Allan PS, Wade AA March 2002 UK	Case notes of patients diagnosed with gonorrhoea or chlamydial infection within the period March 1989 to February 2000	Retrospective review	Number of episodes of gonorrhoea	1250 episodes in 1175 individuals	Retrospective design Study predates implementation of NAAT methods for diagnosing chlamydia, therefore there may have been a proportion of false negatives in those tested for chlamydia. Demographic data of patients not shown.
			Number of episodes of chlamydia	4127 episodes in 3956 individuals
			Episodes of co-infection	332 episodes in 322 individuals
			Treatment of co-infections	235 (73%) treated seperately, ancillary treatment given in 19 (5%) cases. Treatment for both given in 68 (20%) cases as diagnosed at same time.
			Number of co-infections that missed treatment for chlamydia	10. 9 would have been treated if ancillary treatment given (0.2% of chlamydial infection, 2% of co-infection group)
Oriel JD, Reeve P, Thomas BJ, Nicol CS. April 1975 UK	44 men attending the Department of Venereology, St Thomas' Hospital, London between February 1 and April 30, 1973 with confirmed gonococcal urethritis.	Prospective Cohort	Men who had chlamydia yielded from urethral specimens	25% on first attendance, 34% in total.	Aims of study not clearly stated. Very old paper, methods of detecting chlamydial infection much less sensitive than today. Very small sample size. Results not discussed in relation to the epidemiological treatment of chlamydia, but with regards to incidence of post-gonococcal urethritis (PGU). Demographic data of patients not shown.
			Patients with significant urethritis (indicating PGU) 2 weeks after treatment	100% chlamydia positive v. 38% chlamydia negative.
Tapsall JW, Kinchington M. 1996 Australia	142 men and women attending a number of different clinics and hospitals in the Sydney area in the 21 months up to June 30 1994, who were tested for both N.gonorrhoeae and C.trachomatis, and had a positive result for either or both organisms.	Prospective Cohort	Rates of gonococcal infection	147 episodes in 142 patients. Males = 113 episodes in 111 patients. Urethra only site of gonococcal infection in 68.2%. Women = 34 episodes in 31 patients.	Australian study. Rates of infection and co-infection may be very different in the UK so can't generalise from study. Diagnosis of chlamydia relied on EIA which has been shown to have low sensitivity and so cases of chlamydia may have been missed in this study. Data on sexual orientation of patients not directly sought. However, male sample thought to contain a high proportion of homosexuals, giving a biased result. C.trachomatis has a longer incubation time than N.gonorrhoeae but samples taken at same time, with no follow up samples. Some infections may have been missed Demographic data of patients not shown.
			Co-infection with chlamydia	6.8%. 3.5% males v. 17.6% females (p<0.01)
			Ratio of gonococcal to chlamydial infections	1:1.6 (1:1.09 men v. 1:3.3 women, p=<0.001)
			Ratio of infections men:women	gonorrhoea 1:0.3 v. chlamydia 1:0.9 (p=<0.0001)
Charuwichitratrana S, Polnikorn N, Puavilai S, Limsuwan A. 1989 Thailand	120 male patients having had treatment for gonococcal urethritis between March-June 1987.	Prospective cohort.	Rate of co-infection with chlamydia	26.67%	Thai study. Rates of infection and co-infection may be very different in the UK so can't generalise from study. Methods of detecting chlamydial infection much less sensitive than today, some cases may have been missed. Patients tested for chlamydia after treatment with a variety of different antibiotics. Chlamydial infection may have been sensitive to some of these antibiotics.
			Difference in rates of chlamydia detection	29.5% with urethral symptoms v. 22.4% in those without symptoms. (chi squared = 0.447, p>0.5)
Creighton et al, February 2003, UK	All patients (17,854) with a new clinical problem attending the Caldecot Centre between 1 January 1998 and 31 December 1998.	Prospective Review	Infection rates	17,854 new attendances. 3.8% gonorrhoea, 8.1% chlamydia, 1.5% co-infected	Clinic has highest number of reported gonorrhoea and chlamydia cases in the country. May not be representative of the general population. Does not state how they distinguished which was the primary infection
			Rates of chlamydial co-infection in those attending with gonorrhoea	24.2% (95% CI, 22.3-26.2) heterosexual men v. 38.5% (95 CI, 35.9-31.1) women (p=<0.001)
			Rates of gonorrhoea co-infection in those attending with chlamydia	18.8% (95% CI, 17.3-20.3) heterosexual males v. 13.3% (95% CI, 12.2-14.4) women (p=0.002)
			Homosexual male infection rates	73% gonorrhoea. 6.9% co-infection with chlamydia. Lower than for heterosexual men (p=0.001)
			Median age of those with co-infection	22.4 yrs men v. 19.4 yrs women. Lower than for either infection alone (p=0.0001)
			Ethnic group and co-infection	Most cases of both occured in Black Carribeans. Differences in co-infection between ethnic groups statistically significant (p=0.024 as proportion of gonococcal infections, p<0.001 as proportion of chlamydia infections

Comment(s)

A number of papers, spanning several decades and from several different locations around the world, give differing impressions on the incidence of chlamydial co-infection in men with gonococca urethritis. Rates varied from between 6.8% (Tapsall) to 34% (Oriel). Two studies (Charuwichitratrana, Creighton) recommend the epidemiological treatment of chlamydia, two studies (Tapsall, Das) do not recommend it, and one study (Oriel) makes no recommendation. However all studies but one (Creighton) were dated in their methods of chlamydia diagnosis, and also did not compare rates for different sub-groups of the population. One study (Tapsall) was also fatally flawed in that it contained a high proportion of homosexual men, who are known to have a low incidence of chlamydial infection. It is also important to remember that chlamydia is a cause of significant morbidity if left untreated, especially when passed on to females.

Clinical Bottom Line

Epidemiological treatment for chlamydia should be given to heterosexual men in a high risk group i.e. under 25 and/or of Black Carribean ethnic origin. Wherever possible, all men presenting with gonococcal urethritis should be tested for chlamydial co-infection.

References

1. Das S, Allan PS, Wade AA. A retrospective study of the clinical effectiveness of the treatment of genital co-infection with N. gonorrhoeae and C. trachomatis in Coventry International Journal of STD & AIDS. 2002;13(3):178-80.
2. Oriel JD, Reeve P, Thomas BJ, Nicol CS. Infection with Chlamydia group A in men with urethritis due to Neisseria gonorrhoeae Journal of Infectious Diseases 1975;131(4):376-82.
3. Tapsall JW, Kinchington M. The frequency of co-infection with Neisseria gonorrhoeae and Chlamydia trachomatis in men and women in Eastern Sydney. Pathology 1996;28(1):84-87.
4. Charuwichitratrana S, Polnikorn N, Puavilai S, Limsuwan A. Prevalence of chlamydial infection in patients with gonococcal urethritis. Journal of the Medical Association of Thailand 1989;72(5):280-283
5. Creighton S, Tenant-Flowers M, Taylor CB et al. Co-infection with gonorrhoea and chlamydia: how much is there and what does it mean? Int J STD AIDS. 2003 Feb;14(2):109-13.