PCR in the Bacterial Vaginosis Diagnostic Algorithm

Bacterial vaginosis (BV) is the most common vaginal infection worldwide. The objective of this study was to apply a PCR-based method for detection of BV-associated organisms in symptomatic and asymptomatic women and to assess the need of PCR in the diagnostic algorithm of BV. Two vaginal samples were taken from 98 women (74 symptomatic, 24 controls). Amsel criteria and Nugent scoring were used together with tests such as: vaginal pH, Gram staining, routine culture, culture on dextrose agar with Gentamycin and Chromagar candida, and PCR for identification of Gardnerella vaginalis, Atopobium vaginae, Eggerthella-like bacterium, Leptotrichia, BVAB1, BVAB2, and Megasphera type1. In the symptomatic group 17 swabs were with vulvovaginal candidiasis (VVC); 24 with BV, 8 with intermediate BV, 13 with co-infections, and 12 with other infections. In the control group 2 swabs were with VVC, 1 with BV and 1 with other infection. Using PCR in the group with complaints G. vaginalis was found in 59 samples (79.7%), A. vaginae in 21 (28.4%), Eggerthella-like bacterium in 15 (20.3%), Leptotrichia in 21 (28.4%), BVAB1 in 3 (4.1%), BVAB2 in 16 (21.6%), Megasphera type1 in 25 (33.8%). In the control group G. vaginalis was identified in 4 samples, A. vaginae in 2, PCR for other bacteria remained negative. Eggerthella-like bacterium, Leptotrichia, BVAB2, and Megasphera type1 could be used as strong markers of BV meanwhile G. vaginalis could not. PCR-based technique is sensitive and specific, but a combined approach is needed in the diagnosis of vaginal discharge conditions.