While all direct thrombin inhibitors (DTIs) produce their anticoagulant effect by inhibiting thrombin, there are differences in their mode of action, not only in protease generation/inhibition, but also in modulating cellular mechanisms such as the release of endothelial mediators including nitric oxide (NO) and prostacyclin. Preliminary studies have revealed that argatroban treatment, unlike bivalirudin and lepirudin, results in vasomodulation and other pharmacological effects unrelated to thrombin inhibition [

Pathophysiol Haemost Thromb
;Suppl 3:
]. To validate this observation primates (Macaca mulatta; n=6/group) were treated with 100 to 500 μg/kg iv BID of these three DTIs. Blood samples were drawn at 5, 15, 30, 60, 120, 240 and 360 minutes. Total NO levels were measured by two different methods. The Griess reaction was performed by a chromogenic substrate method (R&D Systems, Minneapolis, MN) based on the two-step diazotization reaction where NO2 is acidified to produce a nitrosating agent, which is complexed with sulfanilic acid to produce a diazonium ion, which is then coupled with chromogenic reagents. A chemiluminescent assay was also used. Plasma NO reacts with oxygen to form nitrite. Using the Ionics Instrument 208I (Sievers Instrument, Boulder, CO), the nitrite is converted into NO by chemical reactions, and the resulting NO level is determined using a highly sensitive gas phase. DTIs did not have any non-specific interfering affect on the assays. Primates treated with argatroban showed a dose- and time-dependent increase in plasma NO levels. Baseline levels ranged from 4–8 μM (mean ± SD = 6.1 ± 1.8 μM), whereas 5 min after the administration of argatroban NO levels were 6–11 μM (7.2 ± 1.8 μM) for 100 μg/kg, 8–14 μM (11.8 ± 2.6 μM) for 250 μg/kg, and 11–19 μM (15.7 ± 3.1 μM) for 500 μg/kg argatroban. This increase in NO was observed for 30 min. The area under the curve (0–6 hrs) revealed a dose-dependent increase in a non-linear fashion. No difference between the two methods was noted. For lepirudin and bivalirudin, the NO levels did not differ from baseline during the study period, and no differences between these treatment groups were observed. Flow cytometric studies showed comparable inhibition (29–37%) of tissue factor mediated platelet activation 5 min after dosing for all DTIs. However, the % P-selectin expression was lower in argatroban treated primates than with lepirudin and bivalirudin treatment (17.1 ± 5.2 vs 29.3 ± 7.1; p <0.026). The 5 and 60 min argatroban samples assayed for 6-keto-PGF1 alpha showed 20–30% higher levels (not significant) compared to lepirudin and bivalirudin. These results support the hypothesis that argatroban is distinct from lepirudin and bivalirudin in that it produces an increase in NO and prostacyclin which may partly contribute to its therapeutic efficacy by stabilizing platelets.

Author notes

Corresponding author