Objective: To study the revalence of bacterial vaginosis (BV) in symptomatic and asymptomatic women in a community setting, to assess the validiy of the Pap smear in the detection of BV and to determine the sensitivity and specificity of clinical criteria for the diagnosis of BV.
Study design: Gram staining was performed on vaginal secretions collected from 301 women to determine the prevalence of BV. Each smear was assigned a score on the basis of the bacterial morphotypes identified. The bacterial pattern was also studied on 270 adequate, Pap-stained cervical smears and compared with the gold standard, the morphotypes on gram stain, to derive the validity of the Pap smear in diagnosing BV. Also each clinical criterion defined for diagnosis of BV was independently evaluated to determine their sensitivity and specificity.
Results: The overall prevalence of BV was 41.5% in the study subjects using Amsel's and Nugent's criteria. The Pap smear was 78.3% sensitive and 86.9% specific in detection of BV. Among the clinical criteria, the presence of clue cells on wet mounts was both highly sensitive (76.7%) and specific (92.4%) for diagnosing BV, while a positive whiff test had a specificity of 86.9% with sensitivity of 33.9%. Vaginal pH > 4.5 was sensitive (88.3%) but less specific (58.6%). The presence of a thin, homogeneous discharge clinging to vaginal mucosa had low sensitivity (56.3%) and specificity (48.9%). In this study, BV was prevalent in both symptomatic and asymptomatic women. The routine Pap smear was able to detect BV in the majority of women. Its utility can be greater in asymptomatic women, in whom this infection may go undetected otherwise, resulting in future complications related to reproductive health. Among Amsel's clinical criteria, the most sensitive and specific was the presence of clue cells in wet mounts. They can be detected easily by laboratory staff with little training in primary health care seting