Real-Time Imaging of Platelet-Rich Thrombi in Thick-Walled Blood Vessels Using Near-Infrared Fluorescence Light.

Direct visualization of thrombus formation in real-time is limited by the fact that the vasculature is a closed system. Imaging of thrombi within arterioles and venules of small animals has been accomplished using fluorescent dyes that emit light in the visible spectrum. However, these agents are not able to penetrate large, thick-walled blood vessels and are therefore unable to detect thrombi in clinically important sites such as coronary, carotid, and femoral vessels. Near infrared light (NIR, 700–900 nm) demonstrates enhanced tissue penetration and less light scatter than visible light. By labeling platelets with IR-786, a heptamethine indocyanine-type fluorophore that emits NIR light, we have developed a reagent capable of detecting thrombi within thick-walled vessels of large animals. Optimal platelet loading occurred at 2 μM IR-786, resulting in an accumulation of 3 ×10 6 molecules/platelet. Normal platelet function was demonstrated by platelet aggregation and P-selectin surface expression studies. Clearance studies in 35 Kg Yorkshire pigs showed that IR-786-labeled platelets circulated for >2.5 h following infusion. IR-786-labeled platelets were next used in conjunction with an integrated NIR fluorescence video imaging system to digitally record thrombus formation in large vessels of pigs in real-time. Initially FeCl 3 (oxidant injury) was utilized to induce clot formation in surgically exposed femoral arteries and IR-786-labeled platelets were shown to accumulate at the injury site with a signal to background ratio of 4.4±1.7. Thrombus growth was similarly detected and quantitated in coronary, carotid, and iliac arteries and veins. To assess potential applications of IR-786-labeled platelets for thrombus detection, thrombi were monitored in real-time, and quantified with respect to size and kinetics after electrocautery-induced injury to vessels, cutaneous incisions, intravascular stent insertion and introduction of embolic coils. Thrombolysis of formed IR-786-platelet-rich clots was also evaluated by the infusion of streptokinase and heparin, demonstrating dissolution of thrombi. Two color studies using IR-786-labeled platelets in conjunction with methylene blue, a near-infrared fluorescent blood pool agent used to assess vessel patency, showed that whilst the blood pool agent was more effective for assessing vessel patency, IR-786-labeled platelets were more sensitive for detecting and localizing thrombi within vessels. In addition, IR-786-labeled platelets were able to distinguish between actively growing and stabilized thrombi. IR-786-labeled platelets not only provide a useful tool to study mechanisms of thrombus formation but also afford a means of testing new anti-thrombotics and intravascular devices in vessels approximating the size of those in which clinically relevant thrombosis occurs. In addition, these pre-clinical studies indicate the utility of IR-786-labeled platelets as a contrast agent to detect intraoperative thrombosis, a complication of approximately 1–5% of vascular surgeries.