< Back to the previous page ADD TO BASKET

Current challenges in laboratory automation

Evaluation of Colibrí™ for the identification of Gram negative bacilli

April 19 • P1780

A. Roché1, G. Teissier1, R. Fournier1, P. Mion1, J. Bayette1

1) Département de microbiologie de Labosud, Montpellier, France

Background: Colibrí™ is a system which automatizes the picking of isolated colonies grown on petri dishes and allows the preparation of matrix-assisted laser desorption/ionization time of flight targets. Fast and reliable identification of microorganisms is a key task in microbiology to improve time to result.

Materials/methods: A total of 92 clinical urine specimens previously selected with Gram negative bacilli (Klebsiella spp. n=25, Enterobacter spp. n=16, Citrobacter spp. n=9, Proteus spp. n=19, other Proteae group n=7, Pseudomonas aeruginosa n=12, other species n=4) were inoculated onto CHROMID®CPS elite (bioMérieux) and processed into the WASPLab® system. One colony with size > 0.6mm was selected by a technician on the image and plates were transferred to Colibrí™ for the picking of the colonies, the automatic spotting and the matrix HCCA dispensing on disposable MBT and the matrix HCCA dispensing on disposable MBT Bio targets for identification with BRUKER MALDI biotyper®. In parallel, the spotting was realized manually with the same plate using the procedure of the laboratory. Identification were compared between manual and Colibrí™ spotting. In case of discrepancy spots were realized again with both methodology for confirmation.

Results: The comparison between Colibrí™ and the manual process was evaluated on a total of 114 results (including 22 retests). The concordance was 78.1% (n=89). The rate of discrepancies of 21.9% (n=25) concerned in majority Klebsiella species (n=9) and Proteus mirabilis (n=6) for which no peaks was observed with a Colibrí™ spotting and a score of identification > 2 with the manual methodology. Only 5 discrepancies were confirmed by the retest without any link to a specific specie. No errors of identification were observed with Colibrí™ compared to manual spotting.

Conclusions: Colibrí™ for automatic spotting of Gram negative bacilli on disposable MBT Bio targets can be implemented in microbiology laboratory for the diagnostic of urinary tract infections with a good confidence level. The use of formic acid treatment for mucous species used in the routine of the laboratory can reduce the number of unidentified isolates observed with Colibrí™.



Paper poster