Laboratory identification and clinical characteristics of Streptococcus bovis/Streptococcus equinus complex bacteremia: a retrospective, multicenter study in Hiroshima, Japan

Table 2 Comparison of identification results of 39 isolates of SBSEC using the MALDI Biotyper and VITEK2 compact

SBSEC type	Reference (sodA), n	MALDI Biotyper			VITEK2
SBSEC type	Reference (sodA), n	Identified with highest score as^a	Correct identification to the subspecies level	Correct identification to the SBSEC species level	Identified with high probability as^b	Low discrimination	Not identified	Correct identification to the subspecies level	Correct identification to the SBSEC species level
SGSP	29	29 SGSP	29 (100)	29 (100)	20 SGSP	5 (17) ^c	4 (14)	20 (69)	24 (83)
SGSG	5	4 SGSG, 1 SGSP	4 (80)	5 (100)	5 SGSG	0 (0)	0 (0)	5 (100)	5 (100)
SL	4	3 SL, 1 SGSP	3 (75)	4 (100)	1 SL, 1 SGSP	2 (50)^d	0 (0)	1 (25)	4 (100)
SISI	1	1 SE	0 (0)	1 (100)	None	1 (100)^e	0 (0)	0 (0)	1 (100)
Total	39	–	36 (92)*	39 (100)^†	–	8 (21)	4 (10)	26 (67)*	34 (87)^†

SBSEC Streptococcus bovis/equinus complex; SGSP Streptococcus gallolyticus subspecies pasteurianus; SGSG Streptococcus gallolyticus subspecies gallolyticus; SL Streptococcus lutetiensis; SISI Streptococcus infantarius subspecies infantarius; SE Streptococcus equinus
*MALDI Biotyper was significantly more accurate in identifying the SBSEC isolates at the subspecies level compared with the VITEK2 (P = 0.010)
^†There were no significant differences in the rates of correct identification of the SBSEC isolates at the species level between the MALDI Biotyper and the VITEK2 (P = 0.055)
^aIdentified with the highest score (score > 2.0)
^bIdentified with high probability (> 95%)
Details of “Low discrimination” are as follows
^cSGSG (50%)/SGSP (50%) n = 1; SGSG (33%)/SGSP (33%)/SISI (33%) n = 1; SL (33%)/SGSG (33%)/SGSP (33%) n = 1; SL (51%)/SGSG (49%) n = 1; Lactococcus raffinolactis (34%)/SL (33%)/SGSG (33%) n = 1
^dSL (51%)/SGSG (49%) n = 1; SL (50%)/SGSP (50%) n = 1
^eSL (50%)/SISI (50%) n = 1

ISSN: 1471-2334