The determination of selenium in serum and urine by inductively coupled plasma mass spectrometry: comparison with Zeeman graphite furnace atomic absorption spectrometry

Abstract

An inductively Coupled Plasma Mass Spectrometric (ICPMS) method for the determination of selenium in both serum and urine is described. 78Se is used analytically in spite of 38Ar40Ar isobaric interference at mass 78. Initially 82Se was monitored but, limited isotope abundance and therefore limited detection capability for urine selenium precluded continued use. An ethanol–Triton X-100-nitric acid diluent was used to dilute serum and urine and enhance selenium ionization so that both serum and urine can be analyzed with the same calibration curve. Results derived by the ICPMS method were compared with Zeeman Graphite Furnace Atomic Absorption Spectrometry (ZGFAAS) using nickel as the matrix modifier. Detection limits for ZGFAAS and ICPMS using mass 78 are 2.9 and 0.25 μg/l, respectively. ICPMS and ZGFAAS instrument responses were recorded for additions of inorganic selenium, trimethylselenonium iodide, seleno-dl-methionine, and seleno-dl-cystine to urine and serum. ICPMS slopes for all compounds added to urine and serum were found to be nearly identical. ZGFAAS response for each compound was more variable than ICPMS. ZGFAAS response for trimethylselenonium iodide was approximately 3-fold lower than for the other compounds. ZGFAAS regression slopes and correlation coefficients were 0.72 and 0.8139 for reference urine samples. ICPMS regression slope and correlation coefficient vs. the reference target values were 0.95 and 0.9700 for the same urines. Regressions slopes and correlation coefficients for reference sera were 1.01 and 0.9912 for ICPMS and 1.12 and 0.9648 for ZGFAAS. We conclude that ICPMS produced more accurate results than ZGFAAS for selenium in serum and urine.