OBJECTIVE: To evaluate a simpler criterion for bacterial vaginosis than the recommended criterion of Amsel. DESIGN: Cross-sectional study comparing diagnostic data with a recommended standard. SETTING: Danish general practice. PARTICIPANTS: 595 non-pregnant women aged 15-49 years who were gynaecologically examined, were divided into two groups in accordance with a complaint of vaginal discharge. MAIN OUTCOME MEASURE: The associations of pH > 4.5, homogeneity, amine odour, clue cells, and wet smear assessment of predominance of lactobacilli, Gardnerella vaginalis, motile rods with the criterion of Amsel. RESULTS: The four components of Amsel showed a considerable variation of association. Predominance of GV was more highly associated than several of the components. The association of a wet smear criterion composed of the presence of clue cells in combination with predominance of Gardnerella vaginalis and absence of lactobacilli was superior to other combinations. In the group without a complaint of vaginal discharge, preceding screening with use of a pH-measurement seemed advantageous. CONCLUSION: The wet smear criterion seems suitable for use in general practice. But before a definitive establishment, calculation of the reproducibility of interpretation of this criterion, preferably with a more explicit setting of the components, has to be carried out.
OBJECTIVE: To evaluate a simpler criterion for bacterial vaginosis than the recommended criterion of Amsel. DESIGN: Cross-sectional study comparing diagnostic data with a recommended standard. SETTING: Danish general practice. PARTICIPANTS: 595 non-pregnant women aged 15-49 years who were gynaecologically examined, were divided into two groups in accordance with a complaint of vaginal discharge. MAIN OUTCOME MEASURE: The associations of pH > 4.5, homogeneity, amine odour, clue cells, and wet smear assessment of predominance of lactobacilli, Gardnerella vaginalis, motile rods with the criterion of Amsel. RESULTS: The four components of Amsel showed a considerable variation of association. Predominance of GV was more highly associated than several of the components. The association of a wet smear criterion composed of the presence of clue cells in combination with predominance of Gardnerella vaginalis and absence of lactobacilli was superior to other combinations. In the group without a complaint of vaginal discharge, preceding screening with use of a pH-measurement seemed advantageous. CONCLUSION: The wet smear criterion seems suitable for use in general practice. But before a definitive establishment, calculation of the reproducibility of interpretation of this criterion, preferably with a more explicit setting of the components, has to be carried out.