Detection of Trichomonas vaginalis DNA by use of self-obtained vaginal swabs with the BD ProbeTec Qx assay on the BD Viper system

Abstract

Trichomonas vaginalis is the most prevalent nonviral sexually transmitted infection worldwide, and improved diagnostic methods are critical for controlling this pathogen. Diagnostic assays that can be used in conjunction with routine chlamydia/gonorrhea nucleic acid-based screening are likely to have the most impact on disease control. Here we describe the performance of the new BD T. vaginalis Qx (TVQ) amplified DNA assay, which can be performed on the automated BD Viper system. We focus on data from vaginal swab samples, since this is the specimen type routinely used for traditional trichomonas testing and the recommended specimen type for chlamydia/gonorrhea screening. Vaginal swabs were obtained from women attending sexually transmitted disease or family planning clinics at 7 sites. Patient-collected vaginal swabs were tested by the TVQ assay, and the Aptima T. vaginalis (ATV) assay was performed using clinician-collected vaginal swabs. Additional clinician-collected vaginal swabs were used for the wet mount and culture methods. Analyses included comparisons versus the patient infection status (PIS) defined by positive results with the wet mount method or culture, direct comparisons assessed with κ scores, and latent class analysis (LCA) as an unbiased estimator of test accuracy. Data from 838 women, 116 of whom were infected with T. vaginalis, were analyzed. The TVQ assay sensitivity and specificity estimates based on the PIS were 98.3% and 99.0%, respectively. The TVQ assay was similar to the ATV assay (κ=0.938) in direct analysis. LCA estimated the TVQ sensitivity and specificity as 98.3 and 99.6%, respectively. The TVQ assay performed well using self-collected vaginal swabs, the optimal sample type, as recommended by the CDC for chlamydia/gonorrhea screening among women.

Figures

FIG 1

Four-way Venn diagram. All positive results are shown, regardless of the PIS definition. Each patient is represented only once. Overlapping ovals indicate patients with positive results by each of the assays represented by those ovals. For example, 75 patients had positive results with all 4 assays, 35 had positive results with all assays except the wet mount method, and 4 had positive results only with the two NAATs. The TVQ assay had 3 positive results that were not positive by any other assay, while the ATV assay had 8 such results.