We have developed a covalent antithrombin-heparin (ATH) complex with advantages compared to non-covalent antithrombin:heparin (AT:H) mixtures. In addition to increased activity, ATH has a longer intravenous half-life that is partly due to reduced plasma protein binding. Given ATH's altered clearance, we investigated biodistribution of ATH in vivo. ATH made from either human plasma-derived AT (pATH) or recombinant human (produced in goats) AT (rhATH) was studied. 125I-ATH + unlabeled carrier was injected into rabbits at different doses. 131I-labeled albumin was administered just before sacrifice as a marker for trapped blood in tissues. Immediately after sacrifice, animal components were removed, weighed, and subsamples were counted for gamma-radioactivity. Percent recoveries of ATH in various organs/compartments at different time points were calculated, and kinetic distribution plots generated. At saturating doses, early disappearance of rhATH from the circulation was much faster than pATH. Co-incident with clearance, 26 +/- 3% of dose for rhATH was liver-associated compared to only 3.7 +/- 0.5% for pATH by 20 min. Also, at early time periods, >60% of all extravascular ATH was liver-associated. Analysis of the vena cava and aorta suggested that vessel wall binding might also account for initial plasma loss of rhATH. By 24 h, most of pATH and rhATH were present as urinary degradation products (51 +/- 3% and 63 +/- 8%, respectively). In summary, systemic elimination of ATH is greatly influenced by the form of AT in the complex, with liver uptake and degradation playing a major role.