Evaluation of resazurin-based assay for rapid detection of polymyxin-resistant gram-negative bacteria

Background Colistin resistance is considered a serious problem due to a lack of alternative antibiotics. The Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test is a resazurin reduction-based technique that relies on the visual detection of bacterial growth in the presence of a defined concentration of colistin. The aim of this study was to evaluate the performance of the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test in the detection of colistin susceptibility in common clinical Gram-negative bacteria. Results A total of 253 clinical isolates from a teaching hospital, including Acinetobacter baumanii (n = 58, 8 colistin-resistant), Pseudomonas aeruginosa (n = 61, 11 colistin-resistant), Klebsiella pneumoniae (n = 70, 20 colistin-resistant) and Escherichia coli (n = 64, 14 colistin-resistant) were tested in this study. The sensitivity and specificity of the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test compared to Broth microdilution method was 100 and 99%, respectively. Conclusions Our results suggest that Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test could be used as an accurate detection method for colistin resistance.


Background
Polymyxin E, also known as colistin is a multicomponent polypeptide antibiotic, which belongs to the group of polymyxin [1]. Polymyxin E was discovered in the 1940s; yet, later on, it was abandoned in clinical practice due to its increased nephrotoxicity. However, due to the increase of multidrug resistance (MDR) in Gram-negative bacteria, especially in the ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species), colistin has been applied in clinical practice for the last few years as the last resort treatment option [2,3]. Currently, colistin resistance is considered a serious problem, due to a lack of alternative antibiotics [4,5]. As for now, rapid identification of colistin resistance is considered essential for the effective control of MDR Gram-negative bacteria infection.
Broth microdilution (BMD) is the only reference method that has been recommended by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) and Clinical and Laboratory Standards Institute (CLSI) for the detection of minimum inhibitory concentrations (MICs) of colistin [6,7]. Nevertheless, colistin antimicrobial susceptibility testing is very challenging to perform [8,9]. For example, the operational steps of BMD are complex and time-consuming, making it unsuitable for clinical use [10]. Clinical microbiology laboratories are especially affected by the lack of an accurate, fast and easy-to-conduct method to test the colistin susceptibility [11][12][13]. Therefore, it is of great significance for clinical anti-infective treatment to develop and promote new, convenient, economical, rapid and accurate colistin sensitivity detection method.
In 2016, Nordmann et al developed the Rapid Polymyxins NP test for Enterobacteriaceae spp [14]. The method can be used to detect bacteria that can grow, metabolize glucose, and produce acid in the presence of polymyxin such as polymyxin B or colistin through color changes of PH indicators. However, one of the significant limitations when using this approach is that it cannot be applied for non-fermentative bacteria such as A. baumannii and P. aeruginosa. More recently, Lescat et al have developed a rapid resazurin-mucoid susceptibility test method called Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test, which can quickly detect the sensitivity of colistin for both Enterobacteriaceae spp and non-fermentative bacteria within 4 h [15]. The method is mainly based on detection of the strain viability by observing the color change of resazurin (an active colorant) from blue to purple or pink in the presence of colistin (3.75 mg/L).
In this study, we analyzed the performance of the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test in the detection of colistin susceptibility in 253 nonduplicate clinical Gram-negative isolates aiming to provide a basis for the popularization and application of a new method for rapid screening of colistin-resistant common clinical Gram-negative bacteria.

Results
The colistin MICs of the 253 Gram-negative isolates ranged from ≤0.06 to ≥32 mg/L. BMD results were used as a standard, and 53 colistin-resistant strains and 198 colistin susceptible strains were correctly detected by the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test. Very major errors (VME) and major errors (ME) corresponded to false-susceptible and false-resistant results, respectively [16]. There were only two ME in A. baumannii; details are shown in Tables 1 and 2. The specificity of A. baumannii was 96%; while the sensitivity and specificity of the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test to P. aeruginosa, K. pneumoniae and E. coli were 100% (Table 3).

Discussion
In this study, we described the diagnostic performance of the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test, a phenotypic method for differentiation between colistin-resistant strains and colistin-susceptible strains. Compared with the reference BMD, the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test showed accuracy in detecting the resistance to colistin. Besides, the method was fast, easy to perform, and the obtained data were easy to interpret. Rapid Polymyxin NP test makes up for the limitations of applicability in non-fermenters [14]. In our study, we examined it efficiency in detecting non-fermentative bacteria, but also fermentative bacteria, such as E. coli strains and K. pneumoniae strains. The results showed that the sensitivity and specificity of the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test to Enterobacteriaceae were 100%, which was consistent with a previous study [15]. In the present study, there were only two ME in colistin-susceptible A. baumannii strains. The categorical agreement for all tested isolates was 99.2% for the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test. In addition, the sensitivity and specificity were respectively 100 and 99%, which further suggested that this method is suitable for detecting fermentative bacteria.
So far, a number of studies have examined the mechanism of colistin resistance [17,18]. This study revealed that chromosome mutations of two-component regulatory systems (TCSs) and mcr-1, which were located in plasmid, were the main causes of colistin resistance in 53 strains. In addition, we were able to detect drug resistance without a difference. Therefore, compared with the Rapid Polymyxin NP test, the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test is suitable to be used in more scenes.
MicroScan Colistin Well is a newly developed kit for detection of colistin resistance in Gram-negative bacteria [19]. The principle of Rapid ResaPolymyxin Acinetobacter/ Pseudomonas NP test is based on the visual detection of the reduction of the resazurin reagent, a viability colorant that is observed by color change (blue to purple or pink). Interestingly, in the current study, no significant color changes were observed in colistin-resistant P. aeruginosa after the addition of the resazurin reagent for 1 h. After prolonging the observation time for another 1 h, the color changed from blue to purple. In other words, the results were not obtained until 2 h later in the study, while very obvious color changes were observed 15 min      after the addition of the resazurin reagent in the colistinresistant strains of A. baumanii, K. pneumoniae and E. coli, including 2 ME. This may be because the growth rate of P. aeruginosa is slower than that of Enterobacteriaceae, thus taking longer to decompose resazurin into fluorescent substance resorufin. It suggested that the observation time of the results of this experiment needed to be optimized according to the strain.  [20,21], while our method could only be used to screen colistin resistant strains with MIC ≥4 mg/L. Thirdly, the reading time of P. aeruginosa results was different from that reported by the inventors, requiring an additional 1 h of observation time.

Conclusion
The Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test has great stability and sensitivity in detection of colistin resistance in Gram-negative bacteria such as A. baumanii, P. aeruginosa, K. pneumoniae and E. coli strains. In addition, this method is fast and easy to perform. It can contribute in selecting more precise therapeutic choices, and optimizing antibiotic stewardship, and preventing the development of outbreaks with multidrug-resistant isolates. Nevertheless, the testing time of P. aeruginosa is longer than that reported by the inventor, so the observation time of this method needs to be further optimized.

Bacterial strains
A total of 253 nonduplicate clinical Gram-negative isolates including A. baumanii strains (n = 58), P. aeruginosa strains (n = 61), K. pneumoniae strains (n = 70) and E. coli strains (n = 64) were obtained from a teaching hospital in Wenzhou, China. Species identification was performed using the Matrix-Assisted Laser Desorption Ionization Time-Of-Flight Mass Spectrometry (MALDI-TOF MS, Bruker Daltonics, US). A total of 53 colistinresistant strains were selected from our previous studies and were detected by BMD, including 8 A. baumanii strains, 11 P. aeruginosa strains, 20 K. pneumoniae strains and 14 E. coli strains. In addition, 50 colistinsusceptible isolates of each four bacterial species mentioned above were randomly selected as the control group. E. coli ATCC 25922 and P. aeruginosa ATCC 27853 were used as control strains [6].

Antimicrobial susceptibility test
BMD was performed in triplicate. According to the EUCAST/CLSI joined guidelines [6,7], the clinical breakpoints for colistin provided for P. aeruginosa and A. baumanii were ≤ 2 mg/L (susceptible breakpoint) and ≥ 4 mg/L (resistant breakpoint) and Enterobacteriaceae are ≤2 mg/L (susceptible breakpoint) and > 2 mg/L (resistant breakpoint).   Fig. 1 Representative results of the Rapid ResaPolymyxin Acinetobacter/Pseudomonas NP test. Non-inoculated well is shown as the control of the medium and the color changed (first column). Negative, the tested isolate only grows in the absence of colistin (second column). Positive, the tested isolate grows in the presence and absence of colistin (third column)