Abstract

Bacterial vaginosis is a disease affecting many women all over the world, being asymptomatic or symptomatic and producing obstetric and gynaecological complications. The aim of this study was to compare three methods for the diagnosis of bacterial vaginosis. For this cross-sectional descriptive study, vaginal swabs from 101 women attending Máximo Gómez Clinic in Holguín, Cuba, from April to May 2011 were used. One clinical and two laboratory methods were applied. The specimens were processed according to the standard ethical and laboratory protocols. Kappa index was used to determine the concordance between two methods at once, and categories from three methods were compared and analysed. A total of 37 (36.6%), 29 (28.7%) and 42 (41.6%) patients were diagnosed with BV according to Amsel’s, Nugent’s and Claeys’ methods, respectively. Kappa index between methods was 0.6872 (Amsel-Nugent), 0.7522 (Nugent-Claeys) and 0.8134 (Amsel-Claeys). As predictor of BV, from Amsel’s criteria, homogeneous white-greyish discharge was the least specific and whiff test was the most specific, while whiff test was the least sensitive and pH exceeding 4.5 was the most sensitive. The diagnosis of BV by the three methods shows moderate concordance, being Claeys’ method the most effective. Nugent’s method seems to make a sub-diagnosis of the disease and Amsel’s method contains criteria with poor sensitivity and specificity.

Key words: Bacterial vaginosis, polybacterial disease, kappa index, diagnosis, Amsel’s method, Nugent’s method, Claeys’ method