Gas chromatography-mass spectrometry (GC-MS) analysis after heptafluorobutyric anhydride (HFBA) derivatization was one of the published methods used for the quantification of ephedrine (EP) and pseudoephedrine (PE) in urine. This method allows the clear separation of the derivatized diastereoisomers on a methyl-silicone-based column. Recently the authors came across a human urine sample with apparently high levels (µg/ml) of EP and PE upon initial screening. However, duplicate analyses of this sample using the HFBA-GC-MS method revealed an unusual discrepancy in the estimated levels of EP and PE, with the area response ratios of EP/PE at around 29% on one occasion and around 57% on another. The same sample was re-analyzed for EP and PE using other techniques, including GC-MS after trimethylsilylation and ultra-high-performance liquid chromatography-tandem mass spectrometry. Surprisingly, the concentration of EP in the sample was determined to be at least two orders of magnitude less than what was observed with the HFBA-GC-MS method. A thorough investigation was then conducted, and the results showed that both substances could interconvert during HFBA derivatization. Similar diastereoisomeric conversion was also observed using other fluorinated acylating agents (e.g. pentafluoropropionic anhydride and trifluoroacetic anhydride). The extent of interconversion was correlated with the degree of fluorination of the acylating agents, with HFBA giving the highest conversion. This conversion has never been reported before. A mechanism for the interconversion was proposed. These findings indicated that fluorinated acylating agents should not be used for the unequivocal identification or quantification of EP and PE as the results obtained can be erroneous.
Copyright © 2012 John Wiley & Sons, Ltd.