Detection of candida by polymerase chain reaction vs microscopy and culture in women diagnosed as recurrent vulvovaginal cases

Abstract

The objectives were to compare the outcome of polymerase chain reaction (PCR), culture and microscopy of introital and vaginal samples for detection of candida in women with a history of recurrent vulvovaginal candidosis (RVVC). One hundred and three women with a history of RVVC, i.e. with at least four episodes of the condition in the previous year and who consulted with complaints consistent with a new episode, were studied. Introital and vaginal swabs were cultured on Sabouraud and CHROMagar. Isolated strains were identified on the chromogenic agar and by API 32C kits and by Vitek automatized identification system (BioMérieux, France). PCR for detection of Candida spp. was performed on vaginal lavage fluid. There was a complete agreement in the recovery rate when using the two agar media. However, complete concordance was not obtained between culture and PCR. In 32 (43.8%) of 73 women both PCR and culture were positive, and in 17 (23.3%) both were negative. In 15 (20.5%), only cultures proved positive, while in another 13 (17.8%) patients only the PCR was positive. Four of the PCR-negative became PCR-positive on retesting after dilution of the sample to try to reduce any potential putative PCR inhibitors. In the 47 PCR-positive women, 26 (55.3%) of the introital smears and 31 (65.9%) posterior vaginal smears showed candida morphotypes. In the 25 PCR-negative women, the corresponding figures were 9 (36.0%) and 17 (68.0%), respectively. The PCR test identified Candida albicans in 21 (87.5%) instances, but failed to do so in three (12.5%) cases in which the metabolic/assimilation tests were positive for this species. The corresponding figures for non-albicans Candida spp. were four (57%) and three (43%), respectively. The study of women with RVVC did not show any uniform agreement between the different diagnostic methods used to identify candida in genital samples or for speciation of yeast isolates. Thus reliance only on microscopy, culture or PCR may lead to inaccurate results.