Effect of cytochrome P450 induction on phosphorus metabolites and proton relaxation times measured by in vivo 31P-magnetic resonance spectroscopy and 1H-magnetic resonance relaxometry in human liver.

Experimental and clinical studies have led to the hypothesis that the phosphodiester signal obtained by 31P magnetic resonance (MR) spectroscopy may be a specific marker for the hepatic induction of oxidative metabolism (P450 induction) by phenobarbitone or ethanol. Systematic studies in humans are lacking. Therefore, we studied 10 volunteers who received rifampin (600 mg/d) for 6 days, resulting in a documented induction of oxidative metabolism as measured by an increase in urinary 6-beta-hydroxycortisol output in all volunteers (P = .0004). 31P-MR spectroscopy and 1H-MR relaxometry were performed before and after hepatic P450 induction. As shown by 31P-MR spectroscopy, the median phosphomonoester concentration (PME) relative to nucleoside triphosphate (NTP) increased by 21% from 0.63 (range, 0.40-0.89) before induction to 0.76 (0.49-1.67) after induction (P = .0451). The median level of phosphodiesters (PDE) relative to NTP increased by 28% from 4.82 (3.41-6.67) before induction to 6.18 (4.63-11.63) after induction (P = .0091). An increase in the level of inorganic phosphates (Pi) relative to NTP was observed, but changes were not significant. As shown by 1H-MR relaxometry, a nonsignificant trend of the liver parenchyma to shorter relaxation times was observed after P-450 induction. In conclusion, both PME/NTP and PDE/NTP ratios (measured by in vivo 31P-MR spectroscopy) increased significantly after hepatic induction with rifampin. Further clinical studies with 31P-MR spectroscopy must take into account the potential effects of P450-inducing agents.