Diagnostic evaluation of the BD Affirm™ VPIII assay as a point-of-care test for the diagnosis of bacterial vaginosis, trichomoniasis and candidiasis in a population of pregnant women from South Africa.

Abstract

OBJECTIVE: Untreated Sexually Transmitted Infections (STIs) and Bacterial vaginosis (BV) pose a serious health risk to mother and child. Limited data exist on the use of the BD Affirm VPIII assay as a point-of-care test. This study compared the BD Affirm VPIII assay to the BD MaxTM Vaginal assay (reference test) for the detection of BV, Trichomonas vaginalis, and Candida spp. The prevalence of single and co-infections are also reported here. METHODS: The study enrolled 273 pregnant women from King Edward VIII hospital in Durban. Socio-demographic, sexual behaviour and clinical data were collected from all consenting women. The women provided two self-collected vaginal swabs for testing. The swabs were tested using the BD Affirm VPIII assay and the BD MaxTM Vaginal assay. The prevalence of BV, trichomoniasis and candidiasis was calculated as the percentage of women who tested positive for BV, T.vaginalis and Candida infection and 95% confidence intervals (CIs) were calculated for these percentages using the formulas for calculating CIs for proportions. The number of co-infections was calculated using chi-square analysis. The diagnostic accuracy of the BD AffirmTM VPIII assay compared to the BD Max assay was assessed through the calculation of sensitivity, specificity, Negative Predictive Value (NPV) and Positive Predictive Value (PPV) and their respective 95% confidence intervals. RESULTS: In this study population, 85% of the participants were unmarried; however, 84% reported having a regular partner, and 96.3% did not use a condom regularly. The prevalence of Bacterial Vaginosis, Candidiasis and Trichomoniasis was 49.4%, 57.2% and 10.3%, respectively. A large proportion of women (78.8%) in this study did not have a discharge despite being positive for one or more pathogens. The BD AffirmTM VPIII assay showed a moderate sensitivity (79.8%) and specificity (80.3%) for diagnosing BV in all participants. The assay had an excellent specificity for Candida and T. vaginalis of 97.4% and 100.0%; respectively, however, it exhibited poor sensitivities of 52.9% and 42.4%, respectively. CONCLUSION: Our findings show a higher prevalence of Bacterial Vaginosis in antenatal attendees than previously reported, while the prevalence of Candidiasis and Trichomoniasis was in keeping with previous reports. The high number of asymptomatic infections detected is of concern and indicates the need for the re-evaluation of the syndromic management approach, especially in the antenatal population. The BD AffirmTM VPIII assay was found to be unsuitable as a screening test for vaginal infections in pregnancy. The assay performed better as a confirmatory test and may serve useful if used in conjunction with other clinical parameters such as vaginal pH.