The analysis of positron tomographic studies of 3,4-dihydroxyphenylethylamine (dopamine) metabolism in which [18F]6-fluoro-L-3,4-dihydroxyphenylalanine (F-dopa) is used as a tracer is confounded by the presence of [18F]6-fluoro-3-O-methyl-L-3,4-dihydroxyphenylalanine (OMFD). This labeled molecule, formed by the action of peripheral catechol-O-methyltransferase on F-dopa, crosses the blood-brain barrier and contributes to the radioactivity measured by the tomograph. Corrections for this radioactivity in the brain have been proposed. They rely upon the assumption that regional variations in the handling of this molecule by the brain are negligible. Although this assumption is pivotal for the proper quantification of dopamine metabolism using F-dopa, the distribution and kinetics of OMFD have never been studied in humans. We present results in humans that show that there is little selective regional 18F accumulation in the brain, that the distribution volume of OMFD is close to unity, and that a single, reversible compartment is adequate to model the measured time course of radioactivity after an OMFD injection. Analysis of plasma samples for labeled metabolites showed that more than 95% of the radioactivity was associated with OMFD at all times. Our results for OMFD kinetics are in accord with published results obtained in nonhuman primates and for the bidirectional transport of large neutral amino acids across the blood-brain barrier measured using a synthetic amino acid.(ABSTRACT TRUNCATED AT 250 WORDS)