Vitek MS与Bruker MS对希木龙念珠菌复合体鉴定能力评估

Evaluation of the performance of Vitek MS and Bruker MS on the identification of Candida haemulonii species complex

目的评估中国医院侵袭性真菌监测网(CHIF-NET)2010-2015年收集的希木龙念珠菌复合体原始鉴定、Vitek MS和Bruker MS基质辅助激光解吸电离飞行时间质谱(MALDI-TOF MS)鉴定结果。方法收集CHIF-NET 2010-2015年40株希木龙念珠菌和5株假希木龙念珠菌,所有菌株均经过分子生物学方法进行复核鉴定确认。以复核鉴定结果作为"金标准",评估各医院原始常规方法鉴定结果,并评估Vitek MS和Bruker MS对希木龙念珠菌复合体的鉴定性能。结果与分子生物学复核鉴定结果相比,希木龙念珠菌的原始鉴定准确率为47.5%,分别有15%的菌株被显色培养,ATB Rapid ID32C,API 20C或Vitek 2Compact错误鉴定为奥默毕赤酵母和新型隐球菌;常规鉴定方法无法鉴定假希木龙念珠菌,其中3株被错误鉴定为希木龙念珠菌;与分子生物学鉴定结果相比,Vitek MS IVD2.0和Bruker MS DB 5 989对希木龙念珠菌的鉴定准确率分别为100.0%和57.5%。Vitek MS IVD 2.0数据库中无假希木龙念珠菌的参考谱图,故将5株菌全部错误鉴定为希木龙念珠菌,且可信度>90%。Vitek MS RUO数据库中无希木龙念珠菌复合体参考谱图,故无鉴定结果。Bruker MS DB 5989对假希木龙念珠菌的菌种鉴定准确率为20%。Bruker MS自建库后对所有菌株均能正确鉴定到种水平。结论该研究率先在国内评估了Vitek MS与Bruker MS鉴定希木龙念珠菌复合体的性能,对于少见菌种的鉴定可通过增加数据库质谱谱图提高质谱的鉴定能力。

Objective To evaluate the results of original identification,Vitek MS and Bruker MS identification of Candida haemulonii species complex collected from National China Hospital Invasive Fungal Surveillance Net(CHIF-NET)2010-2015.Methods A total of40Candida haemulonii and five Candida duobushaemulonii were collected from all hospitals in China as part of CHIF-NET2010-2015.The original identification results were evaluated by setting the review results as"gold-standard",and the performances of Vitek MS and Bruker MS matrix assisted laser desorption ionization time of flight mass spectrometry(MALDI-TOF MS)in the identification of the C.haemulonii species complex were observed.Results As compared with the molecular identification,the accurate rate of original identification of C.haemulonii was47.5%,while15%of the strains were mistakenly identified as Kodamaea ohmeri and Cryptococcus neoformans by CHROMagar chromogenic media,ATB Rapid ID32C,API20C or Vitek2Compact system.C.duobushaemulonii could not be identified by using routine identification methods.As for the identification of the C.haemulonii,the accurate rate of the Vitek MS IVD2.0and Bruker MS DB5989were100%and57.5%,respectively.There were no reference spectra for C.duobushaemulonii in the Vitek MS IVD2.0database,so it could not be identified by Vitek MS IVD2.0and misidentified all C.duobushaemulonii as C.haemulonii with confidence value>90%.There were no reference spectra for C.haemulonii complex in the Vitek MS RUO database,so this complex could not be identified by Vitek MS RUO.20%of C.duobushaemulonii were identified correctly to species level by Bruker MS DB5989.After complementation with the"in-house"database,all of the remaining isolates were correctly identified to species level.Conclusion This is the first study to evaluate the performances of Vitek MS and Bruker MS in the identification of C.haemulonii species complex in China.MALDI-TOF MS can provide rapid and accurate identification but is reliant on a robust mass spectra database.The performance of MALDI-TOF may be improved by adding mass spectral profiles into the current databases.