Intrauterin device (IUD) application is a widely used effective, safe and economic method for family planning. However IUD use may cause certain changes in vaginal ecosystem and may disturb microflora leading to increased colonization of various opportunistic pathogen microorganisms. The aims of this study were (i) to detect the biofilm production characteristics of Candida spp. isolated from vaginal and IUD string samples of women with IUDs, and (ii) to investigate the relationship between biofilm production and antifungal resistance. A total of 250 women (mean age: 34.4 +/- 7.6 years) admitted to gynecology outpatient clinics with vaginal symptoms (discharge and itching) were included in the study. The patients have been implanted CuT380a type IUDs for a mean duration of 59.8 +/- 42.4 months. Without removing IUD, string samples were obtained by cutting and simultaneous vaginal swab samples were also collected. Isolated Candida spp. were identified by conventional methods and API 20C AUX (BioMerieux, Fransa) system. Minimal inhibitory concentrations (MIC) of fluconazole, itraconazole and amphotericin B were determined by broth microdilution method according to the CLSI guidelines. Biofilm formation was evaluated by crystal violet staining and XTT-reduction assays, and the isolates which yielded positive results in both of the methods were accepted as biofilm-producers. In the study, Candida spp. were isolated from 33.2% (83/250) of the vaginal and 34% (85/250) of the IUD string samples, C.albicans being the most frequently detected species (54 and 66 strains for the samples, respectively). The total in vitro biofilm formation rate was 25% (21/83) for vaginal isolates and 44.7% (38/85) for IUD string isolates. Biofilm formation rate of vaginal C.albicans isolates was significantly lower than vaginal non-albicans Candida spp. (14.8% and 44.8%, respectively; p = 0.003). Biofilm formation rate of C.albicans strains isolated from vaginal and IUD string samples were found as 14.8% (8/54) and 45.5% (30/66), with a statistically significant importance (p < 0.001). However, no statistically significant difference was detected for biofilm formation rates of non-albicans Candida spp. when sample types were considered [44.8% (13/29) and 42.1% (8/19), respectively; p > 0.05]. Fluconazole resistance was significantly higher in biofilm-producing vaginal Candida spp. than those of non-producers (52.4% vs. 16.1%; p = 0.001), however, itraconazole resistance was found similar in biofilm-producer and non-producer isolates (47.6% vs. 32.3%; p > 0.05). Resistance rates for both fluconazole and itraconazole were higher in biofilm-producers (39.5% and 52.6%, respectively), than those of non-producers (10.6% and 29.8%, respectively), representing a statistical significance (p = 0.002 and p = 0.03, respectively) for Candida spp. strains isolated from IUD string samples. The overall resistance rates of C.albicans and non-albicans Candida spp. against fluconazole, were determined as 15% and 54.2%, respectively, while those rates were 24.2% and 68.7%, respectively, against itraconazole. MIC value of amphotericin B for all of the Candid spp. isolates was <= 1.5 mu g/ml. In conclusion, the data obtained from this study revealed that Candida spp. may lead to vaginal infections by inducing biofilm formation in IUD strings and these biofilms may be related to resistance to antifungal agents. Thus, women using IUDs should be followed-up periodically for the development of biofilms in their IUD strings.