There are no atypical complaint trends for the carbon dioxide assay that require further investigation.Labeling is adequate and describes how co2 will decrease naturally when samples are exposed to air.The rate of co2 loss has been characterized and is present in labeling.A review of the customer's architect instrument logs shows the lower results, although several of the level 2 control results did fall below the expected range.When this occurs, the patient samples are flagged with a cntl result flag to notify the operator.Such flags are visible on the patient results reviewed.A review of the calibration data shows multiple calibrations in the month of (b)(6) and into (b)(6) 2018.In (b)(6), the calibrations show an increase in absorbance that indicates the calibrators are increasing in relative co2 concentration rather than decreasing as expected.This increase in the apparent concentration causes the calibration curve to depress further with each successive calibration, thus elevating the reported concentration of co2 in the qc and samples.The calibrators are susceptible to co2 loss just as patient samples and qc samples.The increase in absorbance change is clearly noted in the log review.However, the mechanism of the increase is not known.The customer recalibrated again in late october and the absorbance change was lower than previous calibrations; back to more typical values.The qc and patient values also returned to expected ranges and are higher as expected.This appears to be an isolated event with respect to the calibrators; possibly handling or contamination.No product deficiency was identified.
|