Clinical evaluation of the Vaginal Panel Realtime PCR Kit (Vircell, SL) for the diagnosis of vaginitis

Abstract Vaginitis is a common clinical disorder for which women seek medical care. A clinical evaluation of the Vaginal Panel Realtime PCR kit (qPCR test) compared with reference methods was performed. A total of 1011 vaginal swab specimens were analyzed. The reference methods were Gram strain-based Nugent score combined with Gardnerella culture for bacterial vaginosis (BV) diagnosis and culture for Candida species and Trichomonas vaginalis detection. Candida species identification was performed using matrix-assisted laser desorption ionization-time of flight mass spectrometry. The agreement between both methods for BV was 93.11% with a Kappa value of 0.82 (95% confidence interval 0.781–0.860). The sensitivity, specificity, positive predictive value and negative predictive value of the qPCR test compared to the reference method for Candida species were 96% (95% confidence interval 92.8%-98.1%), 98.4% (95% confidence interval 97.2%-99.2%), 95.3% (95% confidence interval 92.0%-97.3%) and 98.7% (95% confidence interval 97.6%-99.3%), respectively. The qPCR test yielded a positive result for T. vaginalis in twelve of the specimens however the parasite was not microscopically observed in any cultured specimens. The Vaginal Panel Realtime PCR kit showed an optimal agreement compared to reference methods for vaginitis diagnosis, even improving the detection of T. vaginalis .