Inter-laboratory comparison of plasma catecholamine determinations using several different assays

Thirty-four laboratories performed altogether 41 plasma catecholamine assays on each of four samples. Various radioenzymatic assays, high performance liquid chromatographic (HPLC) assays with electrochemical detection and fluorimetric assays were used. There was reasonable agreement that the levels of adrenaline and noradrenaline in a basal plasma pool were about 0.2 and 1.8 nM, respectively. The levels in a pool of plasma obtained after exercise were about 0.7 and 11 nM, respectively. The study, however, revealed a sometimes considerable variability between methods as well as between laboratories using the same method. Results from duplicate determinations of noradrenaline suggest frequent problems with intra-laboratory reproducibility. Results concerning the recoveries of 0.7 or 3.0 nM adrenaline or 2.0 nM noradrenaline (added to the basal plasma pool) showed a rather frequent need for improved precision. Fluorimetric assays gave unacceptable results. Plasma free dopamine measurements showed a basal level of 0.1-0.2 nM with most HPLC assays and a tendency towards higher levels and greater scatter with radioenzymatic methods. On the whole, reverse phase HPLC methods and an inhomogeneous group of single-isotope derivative radioenzymatic assays showed the largest variability. Less variability was found with the radioenzymatic assay of Peuler & Johnson (1977), provided that a few obviously erroneous results were excluded. The smallest variability was found with microparticulate cation exchange HPLC. It is concluded that plasma catecholamine assays would benefit from better standardization and a continuous quality control. Problems associated with validation of new assays, as well as modifications of old assays are discussed.