Self-Sampling in the Diagnosis of Recurrent Vulvovaginal Candidosis

Abstract

This study aimed to determine the accuracy and feasibility of self-sampling in patients suspected of having recurrent vulvovaginal candidosis (RVC). Of 441 patients with symptoms suggestive of RVC presenting during an 8-year period (January 2000 to December 2007) at a dermatology clinic, 277 were instructed to perform weekly vaginal self-sampling for a period of up to 8 weeks. Demographic charactervistics, medical history, physical examination, culture results, and therapeutic efficacy were analyzed with Fisher exact, χ test, or Student t test. When only considering the results of the culture taken at consultation, 17.1% (20/117) of RVC cases could be confirmed. Positive cultures from self-sampling confirmed another 97 cases of RVC (82.9%). The sensitivity of a single Candida culture ranged from 18% to 53%, depending on the cutoff level of growth intensity of the yeast recovered. Specificity ranged from 97% to 100%, and the positive predictive value ranged from 92% to 100%. The number of positive cultures obtained was not associated with the duration of earlier vaginal complaints or with the efficacy of prophylactic treatment. Prophylactic treatment was equally effective in patients taking fluconazole once (8/13, 61.5%) or twice (48/74, 64.9%) a month, but treatment regimes were not randomized. The diagnosis of RVC can be improved dramatically by self-sampling, enabling a sooner start of adequate treatment. Multiple positive cultures were not associated with disease of longer duration or more severe disease and did not influence the response to prophylactic treatment.