We evaluated the performance and analytical parameters of a one-step magnetic IRMA kit for the measurement of myosin in serum. The method uses two monoclonal antibodies selected for their high affinity to the heavy chains of human ventricular myosin. The first antibody is coupled to a magnetic solid phase and the second one is labeled with 125I. The working range of the IRMA (range of myosin concentrations measured with an imprecision < 10% CV) was 250-3600 microU/L and the sensitivity 20.8 +/- 7.2 microU/L. The between-assay variability, evaluated from replicate measurements in different runs of two serum pools was 14.6 CV% for the first pool (259.1 +/- 37.8 microU/L) and 14.3 CV% for the second pool (442.0 +/- 63.1 microU/L), respectively. To evaluate the clinical usefulness of myosin as a marker of myocardial cell damage, serum myosin was measured in patients with acute myocardial infarction (AMI) (n = 9) or subarachnoid hemorrhage (n = 20). The results obtained with the myosin assay were compared with those of two other markers considered specific for myocardial necrosis (CK-MB and myoglobin). In eight patients with AMI, serum myosin was elevated 24-36 hours after the onset of chest pain and reached a maximum at 4-7 days, returning to control levels at 8-11 days. The one remaining AMI patient showed two subsequent peaks in serum CK-MB and myoglobin concentrations (thus suggesting an extension of myocardial necrosis), the myosin concentrations reaching their peak only after 9 days.(ABSTRACT TRUNCATED AT 250 WORDS)