Evaluation of a commercial broth microdilution panel for colistin susceptibility testing of clinical isolates of Escherichia coli and Klebsiella pneumoniae

Background: Colistin is among the last resort antibiotics for the treatment of infections caused by multidrug-resistant Gram-negative pathogens. Antimicrobial susceptibility testing of colistin is challenging due to its physico-chemical properties. Broth microdilution (BMD) is the recommended method for colistin susceptibility testing. However BMD is not practical for clinical microbiology laboratories as manual preparation of BMD plates is time-consuming and labor intensive. Recently, some more user-friendly BMD products with commercial panels have become available. Our objective was to evaluate the performance of a commercial broth microdilution (BMD) product [Sensititre (Thermo Fisher Scientific)] for colistin MIC determination by comparison with reference BMD method using a collection of E. coli and K. pneumoniae isolates. Methods: A total of 323 unique patient isolates (102 E. coli, 221 K. pneumoniae) were included in the study. Isolates were stored at -70°C and subcultured twice on sheep blood agar before testing. Colistin MICs of the isolates were determined using Sensititre (a premade BMD product with dried antibiotics) and an 'in-house prepared BMD panel prepared in accordance with CLSI guidelines' (reference method). MIC determination with Sensititre was performed according to manufacturer's instructions. The reference method was performed using untreated 96-well sterile polystyrene plates. Colistin MIC results were interpreted according to EUCAST breakpoints (susceptible, ≤ 2 mg/L; resistant, > 2 mg/L). Results: Overall susceptibility rate of isolates to colistin by reference BMD was 75.9%. Overall categorical agreement (CA), essential agreement (EA), very major error (VME), and major error (ME) rates for Sensititre were 98.5%, 72.5%, 3.8%, and 0.8%, respectively. The CA and EA between Sensititre and reference BMD for the isolates with reference colistin MICs close to the susceptibility breakpoint (2 - 8 mg/L) was 94.2% and 48.1%, respectively. Sensititre yielded a VME rate of 15% and ME rate of 0%, respectively, for this subset of isolates. Conclusions: In conclusion, Sensititre showed high CA but low EA with reference BMD for entire collection of isolates. The VME rate was just slightly above 3% and ME rate was acceptable. The rates of CA and EA were decreased and the rate of VME was increased when a subset consisting of more challenging isolates was used.