摘要目的 比较Autof MS、Asta MicroIDsys和Bruker Biotyper三种商业化基质辅助激光解析电离飞行时间质谱系统对常见微生物的鉴定效能.方法 方法学评价研究.收集2011至2018年本实验室食品中分离菌株及2016至2018年中国人民解放军总医院临床常规分离菌株共169株,经生化鉴定、16S rDNA或ITS测序,分属39个属,95个种.采用扩展直接涂抹法平行使用三种质谱系统进行鉴定,其中革兰阴性细菌(G-)93株、革兰阳性细菌(G+)65株、酵母菌11株.采用SPSS 18.0软件对数据进行统计分析.结果 169株菌株质谱鉴定在种水平置信度得分和可信任得分水平中,Autof MS质谱系统、Asta MicroIDsys质谱系统和Bruker Biotyper质谱系统准确鉴定率分别为91.1％(154/169)、86.4％(146/169)和81.7％(138/169),Autof MS和Bruker Biotyper之间差异具有统计学意义;在属水平鉴定中,Autof MS质谱系统准确鉴定率为98.8％(167/169),Asta MicroIDsys和Bruker Biotyper质谱系统准确鉴定率均为97.0％(164/169),三种质谱系统之间差异无统计学意义.结论 三种质谱系统对常见微生物均有很好的鉴定效能.依据各质谱系统的鉴定分值评判原则,Autof MS质谱系统在种可信水平上的准确鉴定率较高.

abstractsObjective To compare the efficiency of domestic MALDI-TOF MS systems Autof MS, Korea Asta MicroIDsys and Bruker Biotyper for common microorganisms identification. Methods This is a methodological comparison study. A total of 169 strains were isolated either from food in our laboratory since 2011 to 2018 or clinical samples in Chinese PLA General Hospital since 2016 to 2018. A total of 39 genus, 95 species were identified through Vitek2 Compact combined with 16S rDNA or ITS sequencing. Among them, a total of 93 Gram-negative bacteria strains, 65 Gram-positive bacteria strains, and 11 yeast strains were identified by three MALDI-TOF MS systems parallelly, while using extended direct smear method for sample preparation. The SPSS 18.0 software was used for data Statistical analysis. Results By Mass spectrometry identification, when 169 strains were at the species level confidence score and acceptable score level, 91.12％ (154/169) was correctly identified to species level by Autof MS system, 86.39％ (146/169) by ASTA MS system, and 81.66％ (138 / 169) by Bruker Biotyper MS system. The difference of identification accuracy to species level between Autof MS and Bruker Biotyper MS was statistically significant. Besides, the accuracy of genus identi fi cation was 98.82％ (167 / 169) by Autof MS mass spectrometry system and 97.04％ (164 / 169) by both ASTA MicroIDsys and Bruker Biotyper mass spectrometry system. The differences of identification accuracy to genus level among the three MS systems were not significant. Conclusions All of the three MS systems have good identification capability for common microorganisms. Autof MS systems performed slightly better than Bruker Biotyper MS systems in species level identification.