Evaluation of a positive yeast-in-blood service

The purpose of this study was to evaluate the effect of a novel "yeast-in-blood" surveillance service on the timeliness of antifungal chemotherapy for candidemia.

Two hundred seven blood cultures positive for Candida species between July 1, 1994, and February 15, 2009, as identified by the microbiology positive culture log, were included in this retrospective chart review. Patients with a positive culture before July 1, 2008, were evaluated by the yeast-in-blood surveillance service and included as intervention cases (IC). After this time, those occurring after discontinuation of the service were included as nonintervention cases (NIC). The primary outcome measure was the time to antifungal therapy from the time of blood culture draw. Secondary outcome measures included antifungal selection and time to antifungal prescription order from culture positivity.

Median time to therapy was 58.9±28.4 hours and 41.3±30.5 hours for NIC and IC, respectively (P=0.001). Median time to prescription order was 3.0±6.9 hours for NIC versus 1.9±3.9 hours for IC. Candida albicans was the predominant organism identified (45.3% of NIC and 54.6% of IC). Treatment agents included an azole in 57.4% of NIC and 47.4% of IC, an echinocandin in 33.3% and 27.3% and a polyene in 5.7% and 27.8%, respectively.

Time to therapy and time to prescription were shorter in those evaluated by the surveillance service. These data suggest that a yeast surveillance service improves antifungal medication therapy initiation.