1878. Rapid Disc Diffusion Susceptibility Testing Directly from Blood Cultures with Gram Negative Bacilli is an Accurate Inexpensive Tool to Facilitate Prompt Antibiotic Stewardship
Session: Poster Abstract Session: Antibiotic Stewardship: Diagnostics
Saturday, October 29, 2016
Room: Poster Hall


Current national strategy addresses antibiotic stewardship in order to decrease selection of antibiotic resistance and aims to advance the development and use of rapid diagnostic tests for the characterization of antibiotic-resistant bacteria. The accuracy of rapid direct disc diffusion tests (RDDDT) has not been systemically assessed.  We conducted a pilot study to evaluate RDDDT as an antibiotic stewardship strategy.


In our RDDDT, 106 positive blood cultures (BACTEC) with gram negative bacilli (GNB) were directly inoculated by a swab to a MŸller-Hinton agar disc diffusion (DD) plate. The DD plates were read at 7AM and 3PM, plus at 16-24 hour incubation. The results of RDDDT were compared to standard DD (CLSI) and to Vitek tests the next day. Discrepancies were assessed as very major (VM), major (MA), minor (MI), or complete agreement. Time to the RDDDT result (not reported to the chart) was compared to the actual time of the routine standard report (Vitek). The actual time to optimization of antibiotic therapy was compared to the estimated time to optimization if RDDDT had been used.


106 patients with GNB bacteremia were included. 3761 and 3182 individual RDDDT readings were compared with standard DD and Vitek results, respectively. Compared to standard DD, RDDT had discrepancy rates of 0.1% VM, 1.4% MA, and 8.6% MI. Comparisons with Vitek were comparable (Table). Median time from positive culture to first interpretable RDDDT zone with our twice-a-day reading was 13+ hours, compared to 34+ hrs for the routine chart reports from Vitek. Among 93 patients whose antibiotic therapies were assessed clinically, 57  (61%) required a change in antibiotic regimen for optimization with median time of 55+ hours from the time of positive cultures. Optimization could have been accomplished at median time of 36+ hours sooner if RDDDT had been used.


High agreement rates of RDDDT with standard DD were observed, with only 0.1% VM  and 1.4% MA discrepancies. RDDDT could reliably and safely be utilized in practice, saving considerable time to optimization. Furthermore, RDDDT is simple, inexpensive, uses standard materials, and would be applicable worldwide, especially in areas where expensive molecular technologies are not available.

Hiroki Saito, MD, MPH, UC Irvine, Orange, CA, Kaye Evans, BA, Department of Pathology and Laboratory Medicine, University of California Irvine School of Medicine, Orange, CA, Ellena Peterson, PhD, Medical Microbiology, University of California Irvine Health, Orange, CA and Lauri Thrupp, MD, FIDSA, FSHEA, Medicine, University of California Irvine Health, Orange, CA


H. Saito, None

K. Evans, None

E. Peterson, None

L. Thrupp, None

Findings in the abstracts are embargoed until 12:01 a.m. CDT, Wednesday Oct. 26th with the exception of research findings presented at the IDWeek press conferences.