Comparative Evaluation of Two Automated ID/AST Systems and Mikrolatest Kit in Assessing the In Vitro Colistin Susceptibility of Carbapenem-Resistant Enterobacteriaceae Isolates: A Single-Center Exploratory Study from North India.

Aims: To generate preliminary data about comparative evaluation of two automated ID/AST systems and Mikrolatest kit in determining in vitro colistin susceptibility of carbapenem-resistant Enterobacteriaceae spp. Materials and methods: Twenty-three carbapenem-resistant Escherichia coli and Klebsiella pneumoniae and two carbapenem-sensitive multidrug-resistant E. coli isolates obtained from various clinical samples of inpatients were included in the study. Species-level identification and antibiotic susceptibility testing (AST) of test isolates was performed using BD phoenix and MicroScan WalkAway 96 Plus automated systems. Identity was reconfirmed by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF MS). Additional colistin susceptibility testing was performed using Mikrolatest MIC colistin susceptibility testing kit (reference method).
Results: Results showed that 16% isolates (27.3% [3/11] K. pneumoniae and 7.1% [1/14] E. coli) exhibited in vitro colistin resistance by the reference method. While the categorical agreement between BD Phoenix M50 ID/AST system and reference test w. r. t in vitro colistin susceptibility results was 100% and 92.9% for K. pneumoniae & E. coli, respectively, it was much lower between MicroScan WalkAway 96 plus ID/AST system and the latter. Almost perfect agreement (96%; kappa: 0.834) was observed between BD Phoenix M50 system and reference method. Conclusions: The results of this study are preliminary and cannot be generalized. Multicentric studies with large sample sizes should be conducted throughout the country to gain a deeper understanding of the subject under consideration. Copyright:

INTRODUCTION

Polymyxins are well-established antibiotics that have recently regained significant interest as a consequence of the increasing incidence of infections due to multidrug-resistant (MDR) Gram-negative bacteria. Considering the paucity of novel antibiotics, colistin is currently often the only effective antibiotic agent available against MDR organisms in developing countries like India.[1]

Increasing use of colistin for MDR Gram-negative bacterial infections has led to the emergence of colistin resistance globally. As human infections with colistin-resistant Gram-negative bacteria are associated with higher patient mortality, polymyxin (including colistin) susceptibility testing is now a major challenge.[1] In 2016, the joint Polymyxin Breakpoints Working Group recommended the reference broth microdilution (rBMD) method as the reference test for determining colistin susceptibility.[2] However, a major disadvantage of conventional BMD method is that it is labor intensive. In order to overcome this problem, several automated bacterial identification and antibiotic susceptibility testing (ID/AST) systems were introduced in the market which offer a quick alternative to the conventional technique. Recently, some BMD-based commercial kits have also been made available, which are relatively easier to use.

Data on comparative evaluation of commercially available BMD-based colistin susceptibility testing kits is limited. This study was planned with the aim of generating laboratory data about the utility of two automated ID/AST systems and Mikrolatest MIC colistin susceptibility testing kit in determining in vitro colistin susceptibility of carbapenem-resistant Enterobacteriaceae (CRE).

MATERIALS AND METHODS

An exploratory study was conducted in a tertiary care teaching hospital located in Uttarakhand from January to May 2021 after obtaining approval from the institute ethics committee (letter no. AIIMS/IEC/20/60 dated 08/02/2020). The main objective of this study was to calculate percentage categorical agreement between BD phoenix M50 (Becton, Dickinson and company, Franklin Lakes, New Jersey, USA), MicroScan WalkAway 96 Plus automated ID/AST systems (Beckman Coulter, Inc., Brea, California, USA), and Mikrolatest MIC colistin susceptibility testing kit (Transasia Bio-Medicals, Mumbai, India) in CRE clinical isolates.

Twenty-three carbapenem-resistant Gram-negative bacterial isolates (namely, Escherichia coli and Klebsiella pneumoniae) obtained from various clinical samples such as blood, pus, urine, drain fluid, and central venous pressure (CVP) tip of inpatients were included in the study. Additionally, two carbapenem-sensitive MDR E. coli isolates were also included in the study.

All clinical samples were subjected to culture as per standard guidelines.[3] Isolated bacterial colonies obtained on solid culture media were subjected to Gram staining. Species-level identification was done by using matrix-assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF MS) (Bruker Daltonik GmbH, Bremen, Germany) as per manufacturer's guidelines. Antibiotic susceptibility testing (AST) of all isolates was performed using BD phoenix NMIC-500 and reconfirmed by MicroScan WalkAway 96 Plus Neg combo 72 panels, respectively. Additional colistin susceptibility testing of these isolates was performed using Mikrolatest kit as per manufacturer's guidelines. This kit is designed to test the susceptibility of Gram-negative bacteria to colistin on the basis of MIC determination. The test is based on rehydration of antibiotics in the wells with cation adjusted Mueller Hinton broth and bacterial suspension. This testing kit was used as the reference method to determine colistin minimum inhibitory concentration (MIC) results.

AST results (including those of colistin) were interpreted and bacterial isolates were categorized as Susceptible (S) and Resistant (R) as per the European Committee on Antimicrobial Susceptibility Testing (EUCAST) 2021 guidelines.[4] Discrepancies in colistin susceptibility results were categorized in the form of very major errors (VMEs) and major errors (MEs), respectively. E. coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853, and mcr-1–positive E. coli NCTC 13846 (Kwik Stik; Microbiologics, St. Cloud, Minnesota, USA) were used as control strains for all three aforementioned test methods. Bruker mass calibration standard was used for performing quality check in MALDI-TOF MS.

RESULTS

Bacterial isolates: Twenty-five Gram-negative bacteria (E. coli = 14 and K. pneumoniae = 11) were isolated from various nonrepetitive clinical samples received in microbiology department. No discrepancy in bacterial identification results was observed between automated ID/AST systems and MALDI-TOF MS, respectively. High resistance rates in the range of 64%–100% to aminoglycosides, fluoroquinolones, penicillins, β-lactam/β-lactamase inhibitor combinations, monobactams, cephalosporins, folate pathway inhibitors, tetracyclines, and glycylcyclines, respectively, were observed among all test isolates. The β-lactamase enzyme profile was determined by BD phoenix NMIC-500 panel. Seven and four K. pneumoniae isolates produced Ambler class D and B carbapenemase enzymes, respectively. While majority of the E. coli (10/14) produced Ambler class B carbapenemase enzymes, extended-spectrum β-lactamase (ESBL) and Ambler class D carbapenemase enzyme production was observed in three and one of these test isolates, respectively.

Colistin resistance: The test isolates were classified as colistin susceptible (MIC ≤2 mg/L) or resistant (MIC ≥4 mg/L). Table 1 shows the colistin resistance pattern (expressed as percentage) of all test isolates along with categorical agreement between each automated ID/AST system and reference method separately. With respect to colistin susceptibility, higher categorical agreement was observed between BD phoenix M50 ID/AST system and Mikrolatest kit, compared to that between MicroScan WalkAway 96 Plus ID/AST and reference method, respectively. The coefficient of agreement (kappa) among the three test methods for determining in vitro colistin susceptibility was calculated by using GraphPad Prism 9.1.2 (226) software (San Diego, CA, USA). While almost perfect agreement was observed between BD phoenix M50 ID/AST system and Mikrolatest kit (96%; kappa: 0.834; 95% confidence interval [CI]: 0.521–1.000), it was moderate (76%; kappa: 0.444; 95% CI: 0.122–0.767) when the results of MicroScan WalkAway 96 Plus ID/AST system were compared with those of the reference method. Fair agreement (72%; kappa: 0.340; 95% CI: 0.028–0.651) was observed between the two automated ID/AST systems.

Table 1

Percentage colistin resistance pattern of 25 test isolates by BD phoenix M50 and MicroScan WalkAway 96 Plus ID/AST systems and Mikrolatest MIC colistin susceptibility testing kit, respectively, along with categorical agreement

Organisms	T1 n/n (%)	T2 n/n (%)	T3 n/n (%) (reference method)	Errors				Categorical agreement (%)
				ME (%)		VME (%)		T1 versus T3	T2 versus T3
				T1 versus T3	T2 versus T3	T1 versus T3	T2 versus T3
Klebsiella pneumoniae	3/11 (27.3)	6/11 (54.5)	3/11 (27.3)	0	27.3	0	0	100	72.7
Escherichia coli	0/14 (0)	4/14 (28.6)	1/14 (7.1)	0	21.4	0	0	92.9	78.6

EUCAST=European Committee on Antimicrobial Susceptibility Testing, ME=Major error (colistin resistant by T1/T2 and sensitive by T3), n=number of resistant isolates, n=total number of isolates, T1=Test 1 (BD phoenix M50 ID/AST system), T2=Test 2 (MicroScan WalkAway 96 Plus ID/AST system), T3=Test 3 (Mikrolatest MIC colistin susceptibility testing kit), VME=Very major error (colistin sensitive by T1/T2 and resistant by T3). Colistin susceptibility of ATCC 25922 E. coli (MIC=0.25–2 mg/L), ATCC 27853 Pseudomonas aeruginosa (MIC=0.5–5 mg/L), and mcr-1–positive NCTC 13846 E. coli (MIC=4 mg/L, this MIC value is valid for ≥90% of isolates and should only on occasion be 2 or 8 mg/L) was ascertained using EUCAST 2021 guidelines

DISCUSSION

In the present study, BD Phoenix NMIC-100 was used for detecting carbapenemase-producing organisms (CPO). Majority of the CRE isolates (14/23; 60.9%) were found to produce Ambler class B carbapenemases as detected by BD Phoenix NMIC-500 panel. Production of Ambler class D carbapenemases was observed in 34.8% (8/23) of the carbapenem test isolates. These findings are in concordance with the available literature. New Delhi metallo-β-lactamase-1 (NDM-1) or bla, belonging to Ambler class B carbapenemase, has been the predominant gene encoding for carbapenem resistance in Enterobacterales in India.[5] However, recent Indian studies have indicated increasing occurrence of oxacillinase enzymes like bla, belonging to Ambler class D carbapenemases, among CRE isolates.[67] Recent performance evaluation studies of BD Phoenix NMIC-500 panel have shown promising results for accurate detection of CPO.[89] Therefore, the findings of this study w. r. t. CPO detection may be considered significant.

The categorical agreement between BD Phoenix M50 ID/AST system and the Mikrolatest MIC colistin susceptibility testing kit w. r. t in vitro colistin susceptibility test results was 100% and 92.9% for K. pneumoniae and E. coli, respectively. Overall, only one error (one ME for E. coli) was observed. The categorical agreement for colistin susceptibility test results between MicroScan WalkAway 96 plus ID/AST system and Mikrolatest kit was 72.7% and 78.6% among K. pneumoniae and E. coli isolates, respectively. Totally six errors (three MEs each for K. pneumoniae and E. coli, respectively) were observed. These findings are contrary to our previous study in which 100% categorical agreement was obtained between the MicroScan system and Mikrolatest kit.[10] To the best of our knowledge, performance evaluation studies on the BD Phoenix system and Mikrolatest kit are currently not available. However, a recent study by Hong et al.[11] highlighted the fact that the colistin AST results obtained using the BD Phoenix system were comparable to those of rBMD method. In this study, the categorical agreement rates between the results of the Phoenix system and reference methods were more than 90% for most antibiotics. VME (false-susceptible) and ME (false-resistant) rates were less than 3% for all the antibiotics tested in this study. Minimum inhibitory concentration (MIC) values for carbapenem and colistin determined by the Phoenix system were highly correlated with those of dilution methods, exhibiting 99.2%, 96.7%, and 98.5% of the agreement rate within onefold dilution difference for imipenem, meropenem, and colistin, respectively.[11]

Some of the lacunae of this study were as follows: (1) Use of a commercially available BMD-based kit as the reference method, keeping the ease of testing in mind. In the absence of the true “gold standard” test, the best available test may be used as the “reference standard.” This is likely to introduce some errors in calculation, resulting in what has been called as “reference standard bias.”[12] (2) Use of EUCAST guidelines for categorizing test isolates as colistin “susceptible” or “resistant.” There is no existing clinical and/or pharmacokinetic and pharmacodynamic (PK/PD) data suggesting that any colistin or polymyxin B MIC predicts clinical efficacy, and therefore, using the term “susceptible” is still controversial.[13] (3) The CPO results of BD Phoenix NMIC-500 panel were not confirmed using gold standard molecular assays. Although this panel provides advantage in that the carbapenemase test is automated and the results can be obtained within 6 h, the low specificity in CREs needs to be improved.[8] (4) Small sample size is another limitation.

We would like to conclude by stating that the results of this study are preliminary and cannot be generalized. Multicentric studies with large sample sizes should be conducted throughout the country in order to gain a deeper understanding of the subject under consideration.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.