Four-Laboratory Validation Study of the Determination of Chloramphenicol in Veal Calf Urine

Abstract A four-laboratory validation study of a method for the quantitation and confirmation of low part-perbillion levels of chloramphenicol extracted from veal calf urine was done. With this method, chloramphenicol, derivatized to the bis(trimethylsilyl) ether, was quantitated by gas chromatography with electron capture detection (GC–ECD) and confirmed by selected-ion monitoring in a negative ion chemical ionization gas chromatograph-mass spectrometer (GC–NICI–MS). Four analysts from 4 laboratories participated in the portion of the study devoted to chloramphenicol quantitation. Every analyst analyzed 5 sets, one set per day, on 5 different days. Each set included 6 samples consisting of a blank, 2 fortified samples, 2 incurred urine samples, and one duplicate. Thus, each analyst worked on a total of 30 samples. Chloramphenicol concentrations ranged from 0 to 9.7 ppb. All data were reported to 0.1 ppb. Coefficients of variation for distribution, CVd, ranged from 8.82 to 14.14% and for precision, CVr, ranged from 9.15 to 14.80%. Three analysts from 3 laboratories also participated in the confirmatory portion of the study, which was carried out to test whether the NICI–MS method developed earlier for higher concentrations of chloramphenicol and for an extract of muscle tissue could be applied to lower levels of chloramphenicol and to an extract prepared from urine. The extracts of 10 of the 30 samples were designated for confirmation only, but were obtained by the same procedure as extracts used for the quantitation part of the study. The confirmation part of the study failed to determine whether NICI–MS is still valid for these levels and this matrix because source temperature lower than that recommended by the originating laboratory as used by all collaborating laboratories. The spectra generated by collaborators were different from those expected.