Background: In 2010 the Clinical Labortory Standards Institute (CLSI) recommended to discontinue the use of phenotypic testing for Extended Spectrum β-lactamases (ESBL) and use a lower minimum inhibitory concentration (MIC) to determine therapy. The accuracy of automated susceptibility methods for FEP is thus crucial. We evaluated the accuracy of FEP susceptibility for CNSE on Vitek2, E-test in comparison with manual broth dilution (MBD).
Methods: Enterobactericeae with ceftriaxone MIC ≥2 µg/mL were collected 1/13 - 4/15. Susceptibility testing was done via Vitek2, E-test (BioMerieux, Durham, NC). Manual broth dilution (MBD) served as the reference standard. The presence of serine Class A (TEM-1/2, SHV-1, CTX-M-1/2/3/8/9/14/15/25/26/41, KPC, VEB, PER, GES) Class C cephalosporinase (ACC, FOX, MOX, DHA, CIT and EBC) were confirmed using multiplex PCR FEP susceptibility was interpreted according to 2016 CLSI guidelines. Minor error (mE), major error (ME), very major error (VME), categorical agreement (CA) were assessed according to the FDA definitions.
Results: 79 isolates were identified as CNSE of which 53 (82%) produced Class A (37 TEM, 17 SHV, 24 KPC, 15 CTX-M); 26 (12%) produced Class C of which 7 isolates also produced TEM. 46% (36/79), 56% (44/79), and 41% (32/79) were FEP susceptible Vitek2, Etest, and MBD, respectively. Comparing Vitek2 to MBD Class A: VME rate 13% (7/53), ME rate 2% (1/53), mE 32% (17/53); Class C: mE 31% (8/26). E-test compared to MBD Class A: mE 28% (15/53), VME 4% (2/53); Class C: mE 15% (4/26). Of the 15 CTX-Ms: 10 produced TEM or SHV of which 70% (7/10) were resistant to FEP; 1 produced TEM and SHV resistant to FEP; 3 only produced CTX-M of which 100% (3/3) were resistant to FEP.Of the 24 KPC: 12 produced TEM and SHV 75% (9/12) were resistant to FEP; 6 produced TEM or SHV 100% (6/6) were resistant to FEP; 6 only produced KPC 83% (5/6) were resistant to FEP by MBD.
Conclusion: Commercially available methods for FEP susceptibility testing are not without limitations in CNSE. E-test outperformed Vitek2 (fewer mE, VME). Isolates producing Class C had more accurate FEP susceptibility compared to Class A. Due to the performance issues of FEP susceptibility testing confirmatory ESBL testing may have an ongoing role for select CNSE.