OBJECTIVES: The aetiology of non-gonococcal urethritis (NGU) in a considerable proportion of men remains unaccounted for. We wished to investigate the possible aetiological role of bacterial vaginosis (BV), the commonest cause of abnormal discharge in women, in this condition. METHODS: We carried out two studies. In the first, case-control, study, we recruited men with and without NGU and examined their female partners for evidence of BV. The second, cohort design, study which ran concurrently with the first study involved recruiting women with and without BV and examining their male partners for evidence of NGU. The diagnoses of both NGU and BV were made microscopically to include symptomatic and asymptomatic individuals in both disease categories. RESULTS: In the case-control study 51 couples were recruited. Of these 39 men had NGU and 12 (31%) of their female contacts had BV. In contrast, of 12 men without NGU, only one (8%) of the female partners had BV (odds ratio 4.89, 95% CI: 0.51-42.27). When only Chlamydia trachomatis negative patients were considered, the odds ratio for an association between BV and NGU was increased to 6.77, 95% CI: 0.73-62.68). Thirty eight couples were recruited to the cohort design study. Of 17 women with BV, 12 (71%) of their male partners had NGU. In contrast, of 21 women without BV, seven (33%) of their male partners had NGU (p = 0.049, odds ratio 4.8). When only C trachomatis negative patients were considered, the significance of the association was increased (p = 0.037; odds ratio 5.42). CONCLUSIONS: An association exists between NGU and BV, and vice versa. If BV arises de novo the findings could help to explain the development of urethritis in stable sexual relationships.
OBJECTIVES: The aetiology of non-gonococcal urethritis (NGU) in a considerable proportion of men remains unaccounted for. We wished to investigate the possible aetiological role of bacterial vaginosis (BV), the commonest cause of abnormal discharge in women, in this condition. METHODS: We carried out two studies. In the first, case-control, study, we recruited men with and without NGU and examined their female partners for evidence of BV. The second, cohort design, study which ran concurrently with the first study involved recruiting women with and without BV and examining their male partners for evidence of NGU. The diagnoses of both NGU and BV were made microscopically to include symptomatic and asymptomatic individuals in both disease categories. RESULTS: In the case-control study 51 couples were recruited. Of these 39 men had NGU and 12 (31%) of their female contacts had BV. In contrast, of 12 men without NGU, only one (8%) of the female partners had BV (odds ratio 4.89, 95% CI: 0.51-42.27). When only Chlamydia trachomatis negative patients were considered, the odds ratio for an association between BV and NGU was increased to 6.77, 95% CI: 0.73-62.68). Thirty eight couples were recruited to the cohort design study. Of 17 women with BV, 12 (71%) of their male partners had NGU. In contrast, of 21 women without BV, seven (33%) of their male partners had NGU (p = 0.049, odds ratio 4.8). When only C trachomatis negative patients were considered, the significance of the association was increased (p = 0.037; odds ratio 5.42). CONCLUSIONS: An association exists between NGU and BV, and vice versa. If BV arises de novo the findings could help to explain the development of urethritis in stable sexual relationships.