Ferritin: another pregnancy-specific protein in human seminal plasma and amniotic fluid, as estimated by six methods

Large quantities of ferritin were found in both seminal plasma and amniotic fluid, but there was a wide spread of results for each of six new non-radioactive monoclonal assays designed to give the same results for ferritin in blood serum. Correlation of values between the various assays was also poor, but after the exclusion of outliers it was possible to calculate approximate ratios for the concentrations. For seminal plasma, the Amersham 'Amerlite' luminescence immunoassay gave the lowest concentrations at 81 +/- 56 ng ml-1 WHO 80/602 (mean +/- SD, n = 43). Relative to this value as 100%, the other assays gave:Flow 'Monoscan' colorimetric IEMA = 161%; Pharmacia-LKB 'DELFIA' fluoroimmunoassay = 174%, Walker 'Synelisa' colorimetric ELISA = 182%; Abbott 'IMx' MEIA = 207%, while the results from the Ramco colorimetric ELISA were much higher and not parallel to the other assays. For amniotic fluid, the Amersham 'Amerlite' LIA gave 127 +/- 95 ng ml-1 (n = 38), taken as 100%, and the ratios for the other assays were:Abbott also approximately 100% but with a wide spread; Walker 43%, Pharmacia-LKB 22% and Ramco 148%. The Flow assay produced 15 zero values from 38 samples, and values were also very low in another 17 samples, so that no comparison with the other assays was possible. An attempt was made to determine the proportion of the different types of isoferritins being estimated. The Pharmacia-LKB assay estimated 90% of the theoretical concentration for pure liver ferritin, 42% of that from the placenta, 26% of that from the heart and 13% of that from the spleen. The other assays gave different proportions.