247. Evaluation of the T2Candida Panel as an Antifungal Stewardship Tool in Transplant and Non-Transplant Patients at a Tertiary Care Center

BACKGROUND: Invasive candidiasis (IC) has a mortality of >30% and likelihood of death increases 50% each day antifungal therapy (AF) is delayed. In patients with suspected IC early empiric AF improves outcomes but has the potential for overuse. Blood cultures (BC) detect ~50% of candidemia with a turn-around-time (TAT) of 2–5 days. T2Candida Panel (T2) an FDA-approved molecular diagnostic test, detects 5 Candida species directly in blood with specificity 99.4%, sensitivity 91.1%, and TAT of 3–5 hours. Our institutional guidelines permit the use of empiric AF in patients with suspected IC (Figure 1). We evaluated the utility of T2 as an AF stewardship tool to support these guidelines. METHODS: We reviewed patients that had T2 done January 2016 to May 2016 at Henry Ford Hospital, a 900-acute care bed teaching hospital in Detroit, MI. Patients with negative T2 [T2(−)] and negative concurrent BC [BC (−)] were evaluated. The primary endpoint was discontinuation (d/c) of AF after a T2 (−) result. Secondary endpoints were candidemia after d/c of AF and all-cause 30-day mortality. Comparative analyses were performed of transplant (txp) vs. non-transplant (non-txp) patients. Univariate analysis was done to determine the association of risk factors and outcomes. Multivariate regression using forward selection was used to model mortality risk. Time to d/c of AF were modeled using Kaplan–Meier estimators. RESULTS: 500 consecutive patients with T2 (−) and BC (−) results were identified. Patients on AF for prophylaxis or treatment of fungal infection were excluded. 472 patients (93 txp patients) were included in the analyses. Characteristics of the txp and non-txp patients are in Table 1. Median TAT in hours (hr) of T2 was 6 (±2) vs. 123 (±25) for final BC (−) result. 264/472 (56%) patients were initiated on empiric AF. In patients with T2 (−) result, AF were d/c within 7 days in 97%; time to d/c of AF was 72 hr in 50% txp patients and 48 hr in 50% of non-txp patients respectively (Figure 2). No episodes of candidemia were diagnosed after d/c of AF. All-cause mortality was lower in txp patients (14%) vs. non-txp patients (34%) (P = 0.0002). Likelihood of mortality did not increase after d/c of AF (OR1.3, 95% CI 0.913–2.064). CONCLUSION: T2 promotes d/c of empiric AF in txp and non-txp patients with suspected IC without a negative impact on clinical outcomes. [Image: see text] [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.