The term "D-dimer assay" suggests, that these assays report the concentrations of the end-stage degradation product of crosslinked fibrin. This hardly occurs in patients. Degradation products of cross-linked fibrin rather occur with a wide range of molecular weights, and comprise various numbers of the D-dimer motif. Moreover, the numerical values obtained with different "D-dimer" assays vary widely. The variations are probably due to differences in reactivity of the various monoclonal antibodies used with the various D-dimer containing degradation products as they occur in patients; and to the various calibrators with a value assigned by the manufacturers. For the reasons indicated above a calibrator in the strict sense (e.g., pure D-dimer) is not feasible. This study shows that: it appears feasible to generate a conversion factor for each of the "D-dimer" assays studied, which will make the widely varying results obtained with these kits comparable; that the conversion factors can be based on a pool of real patient samples; and that the conversion factors are pool-independent. The next and final step in this study is to prepare and make available an international reference material for use by manufacturers and others facing a comparison problem. This is being carried out, in collaboration with the NIBSC (Dr. P. Gaffney), and the material is expected to be available in early 1997.