Table 3.

Errors and categorical agreement of 550 drug-species combination after comparing results of MIC values obtained from Vitek2 and zone diameters from rapid antimicrobial susceptibility testing (RAST) directly feasible on primarily sterile body fluids inoculated in blood culture bottles. For interpretation, EUCAST’s breakpoints for RAST (version 1.0, 1.1) for positive blood cultures were utilized on a total of 345 positive blood culture bottles inoculated with primarily sterile body fluids at the Department for Infectious Diseases at the University Hospital Heidelberg, Germany. Inhibition zones from RAST were compared to MIC results obtained from Vitek2. Very major errors (VME), major errors (ME) and minor errors (MinE) were calculated. RAST guidelines only contain an ‘area of technical uncertainty’ (ATU) where interpretation as susceptible, intermediate or resistant is not permitted. For determining VME and ME, isolates with ATU interpretation were excluded. MinE were only determined for susceptible and resistant RAST response to intermediate Vitek results (S, susceptible; R, resistant; I, susceptible, increased exposure; ATU, area of technical uncertainty)

	RAST interpretation			Vitek MIC results			Errors			Categorical agreement
	S	ATU	R	S	I	R	VME	ME	MinE
Cefoxitin (n = 197)	92.4% (182/197)	≤ 1% (1/197)	7.1% (14/197)	92.9% (183/197)	/	7.1% (14/197)	/	/	/	99.5% (195/196)
Ampicillin (n = 91)	44.0% (40/91)	/	56.0% (51/91)	47.3% (43/91)	/	52.7% (48/91)	/	7.0% (3/43)	/	96.7% (88/91)
Vancomycin (n = 91)	69.2% (63/91)	2.2% (2/91)	38.6% (26/91)	75.8% (69/91)	/	24.2% (22/91)	/	5.8% (4/69)	/	94.4% (84/89)
Piperacillin/tazobactam (n = 57)	56.1% (32/57)	28.1% (16/57)	15.8% (9/57)	78.9% (45/57)	/	21.1% (12/57)	/	/	/	100% (41/41)
Ciprofloxacin (n = 57)	56.1% (32/57)	19.3% (11/57)	24.6% (14/57)	71.9% (41/57)	7.0% (4/57)	21.1% (12/57)	/	4.9% (2/41)	4.3% (2/46)	91.3% (42/46)
Meropenem (n = 57)	94.7% (54/57)	/	5.2% (3/57)	96.4% (55/57)	1.8% (1/57)	1.8% (1/57)	/	4.3% (2/55)	1.8% (1/57)	94.7% (54/57)