Aminoxyl radicals (nitroxides) are a class of compounds with important biomedical applications, serving as antioxidants, spin labels for proteins, spin probes of oximetry, pH, or redox status in electron paramagnetic resonance (EPR), or as contrast agents in magnetic resonance imaging (MRI). However, the fast reduction of the radical moiety in common tetramethyl-substituted cyclic nitroxides within cells, yielding diamagnetic hydroxylamines, limits their use in spectroscopic and imaging studies. In vivo half-lives of commonly used tetramethyl-substituted nitroxides span no more than a few minutes. Therefore, synthetic efforts have focused on enhancing the nitroxide stability towards reduction by varying the electronic and steric environment of the radical. Tetraethyl-substitution at alpha position to the aminoxyl function proved efficient in vitro against reduction by ascorbate or cytosolic extracts. Moreover, 2,2,6,6-tetraethyl-4-oxo(piperidin-1-yloxyl) radical (TEEPONE) was used successfully for tridimensional EPR and MRI in vivo imaging of mouse head, with a reported half-life of over 80 min. We decided to investigate the stability of tetraethyl-substituted piperidine nitroxides in the presence of hepatic microsomal fractions, since no detailed study of their "metabolic stability" at the molecular level had been reported despite examples of the use of these nitroxides in vivo. In this context, the rapid aerobic transformation of TEEPONE observed in the presence of rat liver microsomal fractions and NADPH was unexpected. Combining EPR, HPLC-HRMS, and DFT studies on a series of piperidine nitroxides -TEEPONE, 4-oxo-2,2,6,6-tetramethyl(piperidin-1-yloxyl) (TEMPONE), and 2,2,6,6tetraethyl-4-hydroxy(piperidin-1-yloxyl) (TEEPOL), we propose that the rapid loss in paramagnetic character of TEEPONE is not due to reduction to hydroxylamine but is a consequence of carbon backbone modification initiated by hydrogen radical abstraction in alpha position to the carbonyl by the P450-Fe (V) =O species. Besides, hydrogen radical abstraction by P450 on ethyl substituents, leading to dehydrogenation or hydroxylation products, leaves the aminoxyl function intact but could alter the linewidth of the EPR signal and thus interfere with methods relying on measurement of this parameter (EPR oximetry).