BackgroundEarly and adequate antimicrobial therapy improves clinicaloutcomes in bloodstream infections. Yet the time to resultfor phenotypic antimicrobial susceptibility testing (AST)can take more than 48 hours, requiring long delays intailored and appropriate therapy. The Accelerate Pheno™system (Accelerate) provides organism identification (ID)in approximately 90 minutes and MIC-­based antimicrobialsusceptibility results within 7 hours, directly from positiveblood cultures. This study compared traditional methodsto the Accelerate Pheno™ system for ID and AST ofGram-­negative organisms isolated from positive bloodcultures.MethodsBlood cultures that flagged positive on the VersaTREK®system (aerobic and anaerobic bottles included) withGram-­negative organisms were tested prospectively withboth the Accelerate Pheno™ system (v1.2.0.69 software)and conventional methods (VITEK® MS and VITEK® 2 fororganism ID: Vitek® 2, E-­test, and disk diffusion for AST).CLSI 2016 standard interpretations were used. Thefollowing Gram-­negative targets were included in the finalcomparison: Escherichia coli, Klebsiella spp.,Enterobacter spp., Proteus spp., Citrobacter spp., Serratiamarcescens, Pseudomonas aeruginosa andAcinetobacter baumannii. Organism identification,susceptibility interpretations, and time to result werecompared for this study.ResultsOne hundred nineteen positive blood cultures were tested.Ninety-­one (46 prospective and 45 seeded) of the onehundred-­nineteen isolates tested were included in the finalanalysis. Reasons for exclusion include organism not onpanel, duplicate specimens and analysis failures. Resultsshowed 97.1% agreement in ID between the AcceleratePheno™ system and conventional methods. AST resultsshowed an overall essential agreement of 93.8% andoverall category agreement of 94.5%. Organismidentification was available on average 29 hours earlierwith the Accelerate Pheno™ system, and AST resultswere available on average 54 hours earlier with theAccelerate Pheno™ system.ConclusionsThe Accelerate Pheno™ system exhibited excellentconcordance with traditional culture based identificationand AST. The rapid time to AST results afforded by theAccelerate Pheno™ system provides informationnecessary to optimize patient therapy in a timely mannerand thus potentially positively impact patient outcome.

Abstract (Revised) Introduction

Summary• Results showed 97.1% agreement in ID between the Accelerate Pheno™ system and conventional methods.

• AST results showed an overall essential agreement of 93.8% and overall category agreement of 94.5%.

• Organism identification was available on average 29 hours earlier with theAccelerate Pheno™ system, and AST results were available on average54 hours earlier with the Accelerate Pheno™ system compared totraditional methods.

The authors would like to thank Accelerate Diagnostics, Inc. forproviding evaluation equipment and reagents for this study.

Acknowledgements

References1. Antibiotic Resistance Threats in the United States 2013-­CDC2. Yokota, et al. PloSOne. 20143. Septicemia in U.S. Hospitals, 2009 Healthcare Cost andUtilization Project (HCUP) Statistical Briefs

• The rise of antibiotic-­resistant bacteria represents a serious threat to public health and the economy, and antibiotic use is the primary driver of resistance development (1).

• In the presence of sepsis or septic shock, each hour delay in administration of appropriate antimicrobials is associated with a measurable increase in mortality (2).

• Septicemia was the sixth most common principal reason for hospitalization (2.1 percent of all hospitalizations). Nearly one out of every 23 patients in the hospital (4.2 percent) had septicemia (3).

• Phenotypic antimicrobial susceptibility results are necessary for optimization of antimicrobial therapy.

• A variety of molecular platforms are available that can identify the presence of known resistance genes. No molecular method can determine whether an isolate is in fact susceptible to a given drug.

• The Accelerate Pheno™ system provides phenotypic based AST results from positive blood cultures within 8 hours.

• Rapid AST results provide information physicians need to optimize antimicrobial therapy for septic patients in a timely manner.

Workflow

Results

Universal bacteria

Target E. coli

E. Coli ID Channel

Identification by FISH ELECTRO-­KINETIC CONCENTRATION (EKC)

Specimen Prep GEL ELECTRO-­FILTRATION (GEF)

E. faecalis vs. Linezolid MIC of 32 called “Resistant”

MIC-­based Susceptibility

Enterobacter spp. vs. Meropenem MIC of 2 called “Susceptible”SINGLE-­CELL ANALYSIS

Time to Identification and ASTTime to Identification Time to AST

Traditional Methods* 30 hours (9.1 hr to 61 hr)

62 hours (29.4 hr to 80 hr)

Accelerate PhenoTM System 1 hr 20 minutes 6 hr 36 minutes (8 hr total time)

*Traditional Methods for Identification include VITEK® MS and VITEK® 2*Traditional Methods for AST include VITEK® 2 AST, E-­Test, and KB disk diffusion

Susceptibility ResultsAST PERFORMANCE

Antibiotic EA [95% CI] CA [95% CI] VME ME

Amikacin 95% [90.2-­99.8%] 98.8% [96.3-­100%] 0 0

Cefepime 91.5% [85.4-­97.5%] 90.2% [83.8-­96.7%] 0 1

Ceftazidime 82.1% [73.5-­90.6%] 84.6% [76.6-­92.6%] 1 1

Ceftriaxone 98.6% [95.7-­100%] 95.7% [90.8-­100%] 0 0

Ciprofloxacin 97.6% [94.3-­100%] 98.8% [96.4-­100%] 0 0

Ertapenem 97.1% [93.1-­100%] 97.1% [93.1-­100%] 0 0

Gentamicin 98.8% [96.3-­100%] 100% [100-­100%] 0 0

Meropenem 92.7% [87-­98.3%] 95.1% [90.5-­99.8%] 1 0

Piperacillin-­Tazobactam 92.2% [85.6-­98.8%] 88.1% [80.3-­95.8%] 0 1

Tobramycin 93.5% [88-­99%] 96.1% [91.8-­100%] 0 0

Totals 93.8% [92.1-­95.6%] 94.5% [92.9-­96.1%] 2 3

ID PERFORMANCEMicrobe True

PosFalse Pos

True Neg

False Neg Sensitivity [95% CI] Specificity [95% CI]

Acinetobacterbaumannii 4 0 102 0 4/4 100% [100-­100%] 102/102 100% [100-­100%]

Citrobacter spp. 7 0 99 0 7/7 100% [100-­100%] 99/99 100% [100-­100%]

Enterobacter spp. 10 0 96 0 10/10 100% [100-­100%] 96/96 100% [100-­100%]

Escherichia coli 24 0 81 1 24/25 96% [88.3-­100%] 81/81 100% [100-­100%]

Klebsiella spp. 22 0 82 2 22/24 91.7% [80.6-­100%] 82/82 100% [100-­100%]

Proteus spp. 11 0 95 0 11/11 100% [100-­100%] 95/95 100% [100-­100%]

Pseudomonas aeruginosa 14 0 92 0 14/14 100% [100-­100%] 92/92 100% [100-­100%]

Serratia marcescens 10 0 96 0 10/10 100% [100-­100%] 96/96 100% [100-­100%]

A total of 106 positive blood cultures were tested by the Accelerate PhenoTMsystem. Using traditional* methods for organism ID as the gold standard,diagnostic sensitivity and specificity study results were:

Evaluation of the Accelerate Pheno™ System for the Rapid Identification and Susceptibility Testing of Gram-­Negative Rods from Positive Blood CulturesC. Johnson Ph.D.1,2, A. Davis1, R. Riordan1, A. Rivera-­Acosta1, M. Castillo De Tunon1, J. Dunn Ph.D. 1,2 and P.A. Revell, Ph.D.

1,2

Saturday-­450

Department of Pathology1, Texas Children's Hospital, Department of Pathology and Immunology2, Baylor College of Medicine

46 prospective and45 retrospective positive blood culture bottles on VersaTREK® (aerobic and anaerobic bottles)

VITEK® MS for organism IDPlated on culture media

VITEK® 2 for AST(CLSI 2016 standard

interpretations were used)

Aliquot 0.5 mL into Sample Vial

Load Sample Vial, Cassette and Cartridge;; Press Start

Blood agar Chocolate agar

Principle

Identification Results

Clone kill curvesClone intensity profiles Clone kill curvesClone intensity profiles

Sample is loaded into GEF well. The gel contains pores smaller than bacterial cells.

Postive charge is applied and debris migrates into gel leaving bacteria behind.

Electrical field is briefly reversed and bacteria move to center for ease of retrieval.

Accelerate Workflow

Conventional Methods