Vaginal swabs versus lavage for detection of Trichomonas vaginalis and bacterial vaginosis among HIV-positive women

OBJECTIVES: Cervicovaginal lavage (CVL) is often used for research and may be easier and more accurate than vaginal swabs as a specimen collection method. GOAL: The goal of this study was to compare (CVL) with vaginal swabs for the detection of bacterial vaginosis (BV) and Trichomonas vaginalis (TV). STUDY: CVL and vaginal swabs were collected from 216 HIV-infected women. Clinical assessments were made using wet mount for TV and Amsel's criteria for BV through CVL and swab collection methods. Laboratory gold standards used were Nugent's criteria for BV and InPouch (Biomed Diagnostics, San Jose, CA) culture for TV collected by swab. RESULTS: The prevalence by gold standards for BV was 49.3% and for TV was 25.2%. Sensitivities for direct microscopy versus culture for TV were 72.2 for CVL and 52.8 for vaginal swab (P <0.05). Sensitivities for Amsel's versus Nugent's criteria for BV were 36.2 for CVL and 34.0 for vaginal swab (P <0.80). Kappa scores of agreement between CVL and vaginal swabs for BV and TV were excellent for both. CONCLUSION: CVL was comparable to vaginal swabs as a specimen collection method for these 2 lower genital tract infections and may be superior for the diagnosis of TV.