Comparative evaluation of the QMAC-dRAST V2.0 system for rapid antibiotic susceptibility testing of Gram-negative blood culture isolates

Highlights

• Time necessary to obtain AST result using the QMAC-dRAST V2.0 is 7 h.

• QMAC-dRAST and disk diffusion method results show 92.9% agreement.

• Repeatability using reference strains was [94–100]% of R/S/I categorical agreement.

• Easy-to-use system but the prototype tested would benefit from being optimized.

Abstract

To comparatively evaluate the performance of the rapid antimicrobial susceptibility testing (AST) system QMAC-dRAST V2.0 and of standard disk diffusion in agar, AST was performed directly from 100 positive blood culture bottles with Gram-negative bacilli. AST results provided by QMAC-dRAST showed 92.9% agreement with disk diffusion method results. Discrepancies observed between results obtained with QMAC-dRAST and disk diffusion method conducted to 10 very major errors (0.8%, S with QMAC-dRAST vs R with disk diffusion method), 40 major errors (3.2%, R vs S, respectively), 15 minor errors (1.2%, S vs I or I vs R, respectively) and 23 very minors errors (1.8%, I vs S or R vs I, respectively). For very major and major errors, in only 36% of the cases did the repeat QMAC-dRAST confirm the initial result, whereas a repeat AST using disk diffusion method confirmed the initial result in 92% of cases. AST results obtained using microdilution in liquid medium confirmed those obtained with QMAC-dRAST and disk diffusion method in 4% and 89%, respectively. Repeatability and reproducibility tests performed on QMAC-dRAST using reference strains showed 94% to 100% of R/S/I categorical agreement.

Introduction

Bloodstream infections (BSI) are associated with significant morbidity and remain a major cause of death (Goto and Al-Hasan, 2013). The management of patients with BSI and sepsis critically requires prompt intervention, and early pathogen-adapted antibiotic therapy is of crucial prognostic value for these patients (Kumar et al., 2006; Timsit et al., 2014; Lodise et al., 2018). Inappropriate initial empirical therapy has been frequently observed, varying from 8.7% to 31.3% of reported cases, and is associated with prolonged hospital stays and higher mortality (Shorr et al., 2011; Retamar et al., 2012; Battle et al., 2017). Furthermore, the worldwide spread of multi-drug resistance has led to a change in BSI epidemiology. In 2017 in Europe, 14.9% of Escherichia coli isolates from blood culture were resistant to third-generation cephalosporins, ranging from 5.9% to 41.3% according to the country (ECDC, 2018). In this context, physicians are tempted to use broadest-spectrum empirical treatment regimens, which lead to excessive prescription of the corresponding antibiotics and contribute to the selection of resistant strains. Conversely, if a narrower-spectrum empirical regimen were used, physicians would fear the risk of treatment failure due to a possibly existing resistance of the causative agent. In both situations, adaptation of empirical therapy is required after antimicrobial susceptibility testing (AST).

Rapid bacterial identification and knowledge of resistance traits is therefore crucial in cases of bacteremia. In recent years, new technologies have been implemented in microbiology laboratories allowing reduction of the sample-to-answer turnaround time (TAT) needed for identification and AST. The use of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) resulting in rapid bacterial identification on day one allows for an early choice of appropriate antimicrobial therapy ahead of AST (Seng et al., 2009; Verroken et al., 2015; French et al., 2016). Genetic methods are useful for resistance gene detection but currently do not target all specific resistance determinants and pathogens susceptible to be involved in BSI, while the costs incurred are often prohibitive of their widespread use (Ginn et al., 2017).

AST covering a large number of antibiotics remains the cornerstone of the adjustment of empirical therapy. Usually, AST results obtained with standard methods are available within 18 to 24 h after detection of positive blood culture bottle (PBCB), but methods aimed at reducing TAT have been increasingly described The Vitek-2 Compact system (bioMérieux, Marcy l'Etoile, France) and the Phoenix system (Becton Dickinson [BD], Sparks, MD, USA) yielded identification and AST results within 4 to 14 h and 9 to 15 h, respectively, depending on the species (Gherardi et al., 2012). The disk diffusion method on Mueller-Hinton Rapid-SIR allowed AST reading within 6 to 8 h and demonstrated a significant impact on the use of appropriate antibiotic therapy (Périllaud et al., 2019; Pilmis et al., 2019). A novel methodology, based on automated microscopy for the analysis of bacterial growth in agarose, has become available recently with the Accelerate Pheno system® (Accelerate Diagnostics, Tucson, AZ) (Marschal et al., 2017; Lutgring et al., 2018; Charnot-Katsikas et al., 2018; Pantel et al., 2018) and with the QMAC-dRAST system (QuantaMatrix, Seoul, Republic of Korea) (Choi et al., 2017; Huh et al., 2018) that provided AST results within 7 to 8 h and 6 h, respectively, after Gram staining of samples from PBCBs. Since the initial description of the QMAC-dRAST system by the QuantaMatrix team in 2017 (Choi et al., 2017), only one evaluation of its performance has been published that focused on staphylococcal and enterococcal clinical isolates from PBCBs and on subcultured colony isolates (Huh et al., 2018). Two further Korean studies assessed the gain afforded by this system to clinical practice in the selection of optimal targeted antibiotic therapy for patients with positive blood cultures (Kim et al., 2018; Kim et al., 2019).

The aim of this study was to compare the QMAC-dRAST V2.0 results to those obtained using the standard disk diffusion method, for PBCBs harboring Gram-negative bacteria, to assess the repeatability, the reproducibility and the correctness of the QMAC-dRAST results, and to evaluate the turnaround time (TAT) of the QMAC- dRAST process.