RESUMO
OBJECTIVE: To correlate the symptoms, signs and clinical diagnosis in women with vaginal discharge, based on the combined weight of the character of the vaginal discharge and bedside tests, with the laboratory diagnosis. METHODS: Women presenting consecutively to the women's health center with vaginal discharge were interviewed and examined for assessment of the quantity and color of the discharge. One drop of the material was then examined for pH and the whiff test was done; a wet mount in saline and in 10% KOH was examined microscopically. The clinical diagnosis was based on the results of these assessments. Gram stain and cultures of the discharge were sent to the microbiology laboratory. RESULTS: One hundred and fifty-three women with vaginal discharge with a clinical diagnosis of vulvovaginitis participated in the study. Fifty-five (35.9%) had normal flora and the other 98 (64.1%) had true infectious vulvovaginitis (kappa agreement = 18%). According to the laboratory, the principal infectious micro-organism causing the vulvovaginitis was Candida species. Candida infection was associated with pH levels of less than 4.5 (p < 0.0001, odds ratio = 4.74, 95% confidence interval: 2.35-9.5, positive predictive value 68.4%). The whiff test was positive in only a small percentage of bacterial vaginosis (BV) (p = not significant (NS)). Clue cells were documented in 53.3% of patients with a laboratory diagnosis of BV (p < 0.02, positive predictive value 26.7%). CONCLUSIONS: The current approach to the diagnosis of vulvovaginitis should be further studied. The classical and time-consuming assessments were shown not to be reliable diagnostic measures.
