The initial reporter stated they received discrepant results for three patient samples tested with the elecsys prolactin gen.2 assay on two cobas e 411 immunoassay analyzers.The results obtained on the e411 analyzers did not agree with results obtained using an unknown chemiluminescence immunoassay (clia) method.No incorrect results were reported outside of the laboratory.The first sample was tested on e411 analyzer serial number (b)(4), resulting in a prolactin value of 66.52 ng/ml.When tested using the clia method, the result was 44.4 ng/ml.The second sample was tested on e411 analyzer serial number (b)(4), resulting in a prolactin value of 20.03 ng/ml.When tested using the clia method, the result was 12.7 ng/ml.The third sample was tested on e411 analyzer serial number (b)(4), resulting in a prolactin value of 38.33 ng/ml.When tested using the clia method, the result was 28.0 ng/ml.All three samples were tested on another e411 analyzer (serial number unknown) and the results were reproducible when compared to results from e411 analyzer serial number (b)(4).No specific values were provided.
|
The first patient sample resulted in the following additional test data: b12 = 582.9, cort = 13.44, ferr = 97.14, ft3 = 3.80, ft4 = 1.17, insulin = 7.19, pth = 53.19, tsh = 2.03, vitd = 22.54.No units of measure were provided for these tests.This patient was born on (b)(6) 1982.Medwatch field a2.Has been updated.The second patient sample resulted in the following additional test data: fsh = 54.41, ft4 = 1.03, lh = 26.04, prog = < 0.050, tsh = 2.83, vitd = 48.56.No units of measure were provided for these tests.This patient was born on (b)(6) 1976.The third patient sample resulted in the following additional test data: fsh = 4.03, ft4 = 1.21, toxigg = 0.247, toxigm = 0.204, tsh = 1.44, vitd = 34.41.No units of measure were provided for these tests.This patient was born on (b)(6) 2000.The provided calibration data showed that signals were within expected ranges, but at the lower end.The provided quality control data was within range.Quality control data from (b)(6) 2021 was not provided.Other assays measured for the patient samples showed reagent short alarms.A roche field representative noticed this and decided to check the probe.A liquid level detection circuit board was replaced.Performance testing was run.Product labeling states: "in case of implausible high prolactin values a precipitation by polyethylene glycol (peg) is recommended in order to estimate the amount of the biological active monomeric prolactin." the customer did not perform peg treatment of the affected samples.The investigation could not identify a product problem.
|