Literature review > Issue 7 > Review on Alary  et al. 

This study is a part of a series of studies that have been reported investigating STI prevalences among female sex workers and their male clients in Benin, West Africa. In this particular paper the objectives were to look at male clients for risk factors for urethral infections and also to re-evaluate the leucocyte esterase dipstick as a screening tool using PCR as a more sensitive gold standard.

The participants were male clients who were planning to visit female sex workers in Cotonou, Benin. Mobile clinics were set up in the brothel or bar and prospective clients were asked to provide a urine sample before visiting the prostitute and after the encounter were asked a questionnaire and to undergo a clinical examination. LED testing was done immediately on site, HIV EIA testing was done locally and urine samples for PCR were shipped to Canada for testing. The authors give no indication of time/temperature for storage of urine samples before processing, method of shipment or any quality assurance measures taken to assess the methodology and sensitivity of testing procedures.

Of the 404 men who gave a urine sample 330 answered the questionnaire and 298 agreed to a physical examination. In this situation, a bar or brothel, approaching these men with no added incentives to persuade them to answer questions and to undergo a physical examination must have been extremely difficult and given these circumstances they did well to get so many responders.

The data is presented to answer the two questions of the objectives, risk factors for infection with chlamydia and gonorrhoea and the performance of the LED tests. The risk factor results show risks for infection with CT and NG being lack of condom use with FSWs and a history of STI; there were no measured risks for Trichomonas vaginalis infection. Most, over 80%, of urethral infections were detected in asymptomatic men. For the LED evaluation the strips were 100% sensitive and specific for symptomatic patients - which should be the case, but 49.4% sensitive, 94.7% specific, with a positive predictive value of 37.5% in asymptomatic patients. Given that symptomatic patients would receive syndromic treatment, the most important group would be the asymptomatic group and this is where a rapid test would be most useful. The authors discuss that the results for the LED strip detection are improved by using more sensitive detection methods as a gold standard comparison. In addition, as the samples are from West African men, the LED is not complicated by the presence of schistosoma, which is known to give false positive results. This again makes the LED more useful. They discuss that a PPV as low as 37.5% could be useful, as this would equate to the same level of sensitivity as that used for syndromic management treatment in women with genital discharge.

The use of LED strips to help case detection in urethral infections has been considered in many studies in developing countries and most authors have found that the low sensitivity and positive predictive value make its use questionable [1-4]. In this study the use of PCR increases the sensitivity but the PPV is still low and the economics and logistics of using the test in routine clinics would have to be justified. Although the test strip is rapid and easy to read, to use the test in a rural clinic would involve the logistics of collecting clean urine samples, training staff to use and interpret results, and local availability of LED strips. The costs involved to establish its use would have to be calculated against extra numbers of cases detected, possible over treatment, and costs of undetected cases, which may lead to complications or transmission [5]. As well as the economics, a few further calculations would help to decide its use; a calculation of the likelihood ratio [6] for a positive test at the prevalence of 7.7% shows that there is a 11.2% probability of urethritis, whereas a negative test shows a 0.54% probability. If the prevalence was higher the test would be more useful.

In conclusion, the authors have demonstrated that in sex worker clients the majority of urethral infections are asymptomatic and STI case finding is important in this high risk population. The risk factor scores and LED could be utilised for case finding but the value of using LED with a low PPV would have to be assessed against logistics and costs.

References:

1. Patrick DM, Rekart ML, Knowles L. Unsatisfactory performance of the leucocyte esterase test of first voided urine for rapid diagnosis of urethritis. Genitourin Med 1994;70:187-90.

2. Sellors JW, Mahony JB, Pickard L, et al. Screening urine with a leukocyte esterase strip and subsequent chlamydial testing of asymptomatic men attending primary care practitioners. Sex Transm Dis 1993;20:152-6.

3. Mayaud P, Changalucha J, Grosskurth H, et al. The value of urine specimens in screening for male urethritis and its microbial aetiologies in Tanzania. Genitourin Med 1992;68:361-5.

4. Tyndall MW, Nasio J, Maitha G, et al. Leukocyte esterase urine strips for the screening of men with urethritis-use in developing countires. Genitourin Med 1994;70:3-6.

5. Vuylsteke B, Meheus A. STD syndrome management. In: Dallabetta G, Laga M, Lamptey P (Eds.). Control of Sexually Transmitted Diseases: A Handbook for the Design and Management of Programs. AIDSCAP/Family Health International, Arlington, 1996, pp 151-168.

6. Sackett DL, Haynes RB, Guyatt G, Tugwell P. Interpretation of diagnostic data. In Clinical Epidemiology: a basic science for clinical medicine. Boston: Little, Brown and company 1991;69-152.

about SDI | newsletters | grants | publications | literature reviews

WHO Home - WHO Search - TDR Home - SDI Home - SDI Contact us
(c) WHO/OMS 2001