Assessment of the cytochrome P-448 dependent liver enzyme system by a caffeine breath test.

[1-Methyl-14C], [3-Methyl-14C] and [7-Methyl-14C] caffeine were used to investigate demethylation in control rats, and in rats pretreated with phenobarbital or 3-methylcholanthrene, by a 14CO2-exhalation test. Compared to controls, pretreatment with phenobarbital did not enhance demethylation of any of the labelled caffeines. In contrast, induction by 3-methylcholanthrene, presumably of cytochrome P-448, resulted in highly significant increases in peak 14CO2 exhalation rates, 14CO2 disappearance constants and areas under the exhalation rate - time curves. Based on these results, [7-methyl-14C] and [3-methyl-14C] caffeine were chosen for assessing the feasibility of a caffeine breath test in man, using 5 normal volunteers and 2 patients with compensated liver cirrhosis. 14CO2 exhalation curves in cirrhotics were clearly different from those in normal volunteers, being characterised by a slower rise and a lower specific activity of exhaled 14CO2. Since the variability of the levels of the specific activity in subjects with normal livers suggested the influence of extraneous factors, a second group of normal volunteers, smokers and nonsmokers, was investigated. With either labels, the average 14CO2 exhalation rate was doubled in smokers. From these studies in rats and preliminary results in man it is concluded that specifically labelled caffeine is a suitable and promising substrate for studying demethylation by breath analysis. Presumably, caffeine represents a safe and sensitive indicator of the activity of the cytochrome P-448 system.