A customer in (b)(6) reported the occurrence of a misidentification of citrobacter freundii qc sample (aglae) as citrobacter werkmanii in association with the vitek ms identification system.On (b)(6) 2016 - four (4) test slide spots were processed; each provided a result of citrobacter werkmanii.Expected result was citrobacter freundii.On (b)(6) 2016 - fine tuning of the instrument performed.On (b)(6) 1206 - following a fine-tuning procedure, two (2) spots were processed (one strain isolated from cos media, and one from cps media).The strain from cos gave one result of citrobacter freundii and one low discrimination result (50/50) of citrobacter freundii/werkmanii; cps gave a low discrimination result (50/50) of citrobacter freundii/werkmanii.On (b)(6) 2016 - evaluation of mzml and ecal logs from the vitek ms instrument indicates the tuning is within expected criteria.Good signal observed, high number of peaks observed.Spectra very clean, indicating good customer spotting technique.Confirmatory testing via vitek 2 gn id test kit obtained a result of citrobacter freundii.There is no indication or report from the hospital or treating physician to biomerieux that the occurrence led to any adverse event related to the patient's state of health.There is no patient directly associated with the qc sample.Culture submittal has been requested from the customer by biomerieux for internal investigation.Biomerieux internal investigation has been initiated.
|
An internal biomereiux investigation was performed.Misidentification on external qc strain algae (urine sample): vitek® ms ivd v2: citrobacter werkmanii or low discrimination between citrobacter; werkmanii and citrobacter freundii; vitek® 2 : (on gs and cps media): c.Freundii 95%.Vitek® ms was performed on (b)(6).Note that the misidentification (c.Werkmanii) was only obtained on (b)(6).Fine tuning dates: (b)(6) + increase of the threshold offset (0.08 to 0.11) on (b)(6).Culture conditions: for the first test performed on (b)(6) (misidentification): the strain was isolated on a cps and a cos media; however, the customer did not remember which media the strain had been picked from to perform the vitek® ms id.These two (2) culture medias are compatible with vitek® ms use.For other tests: labid contgs = sample grown on blood agar plate = > results obtained: single choice to c.Freundii or low discrimination between citrobacter werkmanii and citrobacter freundii labid controlecpsbis and controlecps = sample grown on cps agar plate ("controlecps" and "controlecpsbis" are from the same cps agar plate) = > results obtained: single choice to c.Freundii or low discrimination between citrobacter werkmanii and citrobacter freundii.No patient result was impacted as it is an external quality control.Investigation general conclusion: the customer obtained a misidentification on (b)(6) (c.Werkmanii instead of c.Freundii).After several repeats, this misidentification was never reproduced and either a low discrimination between c.Freudii and c.Werkmanii was produced, or the correct result c.Freundii.Based on the investigation: the spotting technique can be excluded: the spectra are very close to each other with good signal according to the dendrogram and the mass analysis.The fine tuning is not the root cause of this issue since all the criteria were met each time before and after the fine tuning.The culture media is not the root cause as the customer used two (2) different culture media (cps and blood agar) and obtained the correct identification on both.Since cps and blood agar are both compatible with the vitek® ms, there is no restriction to use them.The customer's mzml files have been reprocessed with next databases v3.0 and v3.1.Results are improved: two (2) misidentifications are changed to low discrimination results.However, a lot of low discrimination results remain.During the investigation, it was detected that the customer was using non-sterile tips.If customer uses non sterile tips or if they are touched with hands, customer could introduce contamination and it could lead to bad results/bad performances.Workflow user manual 4501-2233 - c clearly indicates to use "sterile colorless pipette tips without filter" (page 2-4).In conclusion, the most probable root cause of the misidentification is a mixed culture on the plate or a contamination during the test.(b)(4).
|