Endothelin-1 (ET-1) is a potent vasoconstrictor peptide apparently involved in a number of vascular diseases in man. Nonetheless, its determination in biological samples is difficult, and data on plasma or urine concentrations are controversial. We investigated different sample preparation procedures as well as different radioimmunoassays for their influence on ET-1 measurement. Recovery of ET-1 depended on the extraction procedure, the type and size of the extraction columns and on the biological matrix itself. Incomplete removal of matrix components by the extraction leads to the formation of particulate matter in the evaporated eluate. When dissolved in assay buffer, ET-1 was found to be absorbed to and only partly released from these particulates, so that it was not accessible for measurement in a radioimmunoassay. This was the case for all extraction procedures investigated except for that involving acetic acid. HPLC analysis of spiked samples revealed that ET-1 is in part degraded during extraction, most probably to Meth-sulphoxide ET-1. Dilution curves of synthetic pure ET-1 standards from different suppliers, prepared either in plasma with subsequent extraction or in assay buffer of the respective radioimmunoassay, resulted in considerable differences in the measured values for ET-1-immunoreactivity. Every radioimmunoassay tested had a specific pattern of recognizing different synthetic ET-1 standards. In conclusion, the measurement of ET-1-immunoreactivity is strongly influenced by the experimental conditions of sample preparation as well as by the radioimmunoassay employed.