892. Clinical Impact of Early Antifungal De-escalation (≤ 2 days) based on Rapid Species Identification in Patients with Azole-susceptible Candidemia

Abstract Background Echinocandins are first-line therapy of candidemia with an option to de-escalate (DE) to oral azoles based on clinical response and susceptibilities. There are no universally accepted DE criteria but rapid diagnostic testing (RDT) allows for earlier Candida sp. identification. The objective of this study was to compare outcomes between early DE (≤ 2 days) and late DE ( > 2 days) using RDT as an antifungal stewardship strategy. Methods This was a retrospective study in adults with candidemia caused by C. albicans, C. tropicalis, and C. parapsilosis from 2017-2021. Patients with neutropenia or deep-seated/uncontrolled source were excluded. Primary outcome was 30-day global response (clinical and microbiological success). Secondary outcomes included clinical and microbiologic success, length of stay after candidemia, development of resistance, infection recurrence, and mortality. Comparisons were performed using Chi-squared or Fischer’s exact test for categorical variables and Student’s t test or Mann-Whitney U for continuous variables. A multivariable logistic regression model was constructed to identify independent factors for global response using DE strategy, ICU status, and considering all variables with a p-value < 0.1 on univariate analysis. Results 87 patients were included, 34 (39%) in early DE group and 53 (61%) in late DE group. Groups were well matched except early DE patients were less likely to be in the ICU at time of candidemia (18% vs 38%; p=0.079). C. albicans was the most common Candida sp. (54%) and a vascular catheter was the most common source (45%). Overall global response was 93% with an in-hospital mortality of 8%. At the time of DE, 45% were considered hemodynamically unstable. No difference in global response between early and late DE was identified (94% vs 96%, p=1.0). On multivariable analysis, no independent predictors were identified and late DE was not associated with improved global response [OR 0.82 (95% CI 0.08, 8.24)]. No differences were identified in secondary outcomes. Conclusion There were no differences in outcomes between early and late DE in the treatment of azole-susceptible candidemia. Early DE within 2 days based on rapid species identification should be considered as an antifungal stewardship strategy based on local susceptibilities. Disclosures All Authors: No reported disclosures.