The customer stated that they received erroneous results for three samples from the same patient tested for the elecsys tsh assay (tsh), elecsys ft3 iii (ft3), and the elecsys ft4 ii assay (ft4) on a cobas e 411 immunoassay analyzer (e411).This medwatch will apply to the ft4 assay.Please refer to the medwatch with (b)(6) for information concerning the tsh assay and refer to the medwatch with (b)(6) for information concerning the ft3 assay.The values highlighted in yellow are erroneous.The samples were initially tested for tsh and ft4 at the customer site on the e411 analyzer.The customer's e411 values for the (b)(6) 2017 sample were reported outside of the laboratory.It was asked, but it is not known if erroneous results were reported outside of the laboratory for the (b)(6) 2017 samples.The samples were repeated for tsh and ft4 on a lumipulse presto ii analyzer.Every other week, the customer tests tsh and ft4 on the e411 in parallel to a lumipulse presto ii analyzer at a branch laboratory.In order to fit the measurements from the lumipulse analyzer, the customer changes the settings on their e411 analyzer to multiply values by a coefficient.The following coefficients are applied in the e411 software: for tsh: result × 0.819, for ft4: result × 0.862 + 0.06.The samples dated (b)(6) 2017 were also provided for investigation where they were tested for tsh, ft4, and ft3 on a second e411 analyzer.No adverse events were alleged to have occurred with the patient.The e411 analyzer used at the customer site was serial number (b)(4).The e411 analyzer used for investigation was (b)(4).Ft4 reagent lot number 265631, with an expiration date of september 2018 was used on this analyzer.Calibration and quality controls run for investigation were ok.In reference to the mathematical differences of values generated with different assays from different vendors, different values may be generated due to the different antibodies used, the different assay setups, and differences in the standardization methodology/materials used.In reference to the difference in values from different vendors compared to their respective normal reference ranges, the normal reference ranges of assays from different vendors may be calculated using different calculation methodology and different patient cohorts are used.In this context, it also needs to be taken into account that differences in the normal reference ranges can occur, based upon age, gender and medication taken.Reduced tsh and ft4 values were generated for the first measurement of the (b)(4) 2017 sample on the e411 analyzer used for investigation.An analyzer specific issue or a sample specific issue such as a sample clot or foam/bubbles on the sample surface may have led to a lower amount of sample being pipetted.A general reagent issue can most likely be excluded.
|