1219. Diagnosing vaginal discharge syndrome and its potential impact on clinical practice in a regional hospital in Taiwan

Background Vaginal discharges are frequently encountered in clinical practice and usually managed empirically. We aimed to investigate the prevalence of bacterial vaginosis (BV), vulvovaginal candidiasis (VVC) and trichomoniasis among adult women of reproductive age and evaluate the appropriateness of empirical treatment. Methods Non-pregnant women between the age of 20 and 49 years who presented with vaginal discharge were prospectively enrolled in a teaching hospital since Oct 2018. Vaginal swabs were collected for determination of Nugent score, culture for Trichomonas vaginalis (TV) and Candida species, and multiplex polymerase chain reaction (PCR) for BV, VVC and TV. Demographics, symptoms, physical findings, and the empirical treatment were recorded. Results From Oct 2018 to May 2020, 172 women were included (median age, 37 years). The prevalence of laboratory confirmed BV, VVC, and TV was 21.5% (n=37), 20.3% (35), and 0.6% (1), respectively. Ten (5.8%) women had concurrent BV and VCC. Among 38 women who had bacterial vaginosis or trichomoniasis, only 8 (21.1%) received metronidazole empirically while more than half (11/19, 57.9%) of women who received metronidazole empirically did not have laboratory-confirmed bacterial vaginosis or trichomoniasis. Among 35 women who had candidiasis, 10 (28.6%) received antifungal agents. Antifungal agents were prescribed to more than two thirds (21/31, 67.7%) of women who did not have laboratory confirmed candidiasis. Overall, 58.7% (101/172) of empirical treatment was deemed optimal. Multiplex PCR test has an overall diagnostic accuracy of 86.0% (148/172) as compared to the composite gold standard. Conclusion The empirical treatment for vaginal discharge syndrome is suboptimal. Better diagnostic assays have a potential to improve clinical patient care. Disclosures All Authors: No reported disclosures