Endovascular removal of intravascular thrombus using the AngioJet rheolytic thrombectomy (RT) system has been shown to be clinically effective. This system also permits the concomitant infusion of thrombolytic agent followed by thrombectomy, thus creating a novel strategy known as pharmacomechanical thrombectomy (PMT). Although these interventions have gained wide clinical application, little is known regarding the vessel wall response following thrombectomy therapy. The aims of this study were to assess the effect of thrombectomy interventions on endothelial function in a porcine model of deep venous thrombosis (DVT) and to evaluate the effect of nitric oxide (NO) precursor L-arginine on endothelial function following thrombectomy therapy. Deep vein thrombosis was created in bilateral iliac veins by deploying a self-expanding stent-graft incorporating an intraluminal stenosis from a groin approach. Five pigs underwent sham operation. Following 14 days of DVT, animals were randomized to three groups: the first group received RT treatment (RT group, n = 5); the second group received pharmacomechanical thrombectomy (PMT) with tissue plasminogen activator (alteplase 10 mg; PMT group, n = 5); and the third group received PMT with tPA plus intravenous L-arginine (20 mmol/l) (arginine group, n = 5). Iliac vein patency was evaluated by venography and intravascular ultrasound at 1 week. Nitric oxide level was determined by a chemiluminescent assay of the nitrite/nitrate metabolites (NO(x)). Thrombogenicity was evaluated by radiolabeled platelet and fibrin deposition. Veins were harvested and evaluated with light microscopy and scanning electron microscopy (SEM). Endothelial function was evaluated using organ chamber analysis. The luminal areas in the sham, RT, PMT, and arginine groups were 34 +/- 10 mm(2), 21 +/- 13 mm(2), 35 +/- 18 mm(2), and 37 +/- 16 mm(2), respectively. All iliac veins remained patent at 2 weeks. No difference in endothelial cell structure was observed between the three treatment groups by means of light microscopic or SEM examination. A decrease in platelet deposition occurred in the arginine group compared to the RT and PMT groups (P < 0.05). The arginine group also showed a greater endothelium-dependent relaxation compared to the RT or PMT groups in response to A23187, bradykinin, and ADP (P < 0.05). Local NO(x) level was higher in the arginine group than in the RT or PMT group (2.6 +/- 0.6 micromol/l versus 0.3 +/- 0.1 micromol/l and 0.3 +/- 0.2 micromol/l; P < 0.01). AngioJet RT and PMT interventions resulted in similar attenuated endothelium-dependent vasoreactivity and morphologic effect. L-Arginine supplementation preserves endothelial vasoreactivity and reduces platelet deposition following PMT in iliac DVT. Additionally, L-arginine enhances NO production at sites of venous thrombosis. The NO precursor L-arginine may have a therapeutic potential in preserving endothelial function following mechanical thrombectomy.