Background: To evaluate the accuracy and performance of Autof MS 1000 mass spectrometer in bacterial and yeasts identification., a total of 2342 isolates were routinely obtained from microbial cultures of clinical specimens (i.e., blood, cerebrospinal fluid, respiratory tract, Puncture fluid, wound, stool and urine) from January to December 2019 in Henan Provincial People’s Hospital. Then, the repetitive strains of the same patient were excluded. The two mass spectrometry methods by Autof MS 1000 and Bruker biotyper systems and the classical biochemical VITEK 2/API 20C AUX Systems testing were used for analysis. The strains with different results by the above three test methods were identified by 16S rDNA and other specific genes sequencing.
Results: The accuracies of isolates identification at species level by Autof MS 1000 and Bruker biotyper systems were 98.9% and 98.5%, respectively. The genus level by Autof MS 1000 and Bruker biotyper systems were 99.7% and 99.4%, respectively. The Autof instrument was not significantly different from the Bruker instrument in identification accuracy at both the species and genus levels. The frequencies of unreliable identification were 1.1% (26/2342) for Autof ms1000 and 1.5% (34/2342) for Bruker biotyper, respectively. The results of in vitro experiments demonstrated that the coincidence rate of the Autof ms1000 mass spectrometers in the identification for five types of bacteria is over 93%, the identification error is less than 3%, and the no identification rate is 0. It was indicated Autof ms1000 identification was an acceptable method.
Conclusions: Autof MS 1000 mass spectrometer can be utilized to identify clinical isolates. An upgrade database is recommended to correctly identify rare strains.