Visualization of the retinal microvasculature by targeted dye delivery.

Although fluorescein angiography has proven to be an important tool in the diagnosis and management of retinal vascular diseases, it is subject to certain limitations, namely the presence of the choroidal background, which usually precludes a detailed examination of the retinal microvasculature. Moreover, the inability to repeat the bolus reduces the chance of obtaining high-quality photographs of early phases, and does not allow for a complete binocular examination or for testing the response to induced physiologic changes. We have developed a method of targeted dye delivery that consists of encapsulating the dye in lipid vesicles, injecting them intravenously, and causing them to release their contents locally when a short heat pulse is induced in a retinal artery by a laser. This method was applied in the rhesus monkey in order to visualize the retinal microvasculature. A well-defined bolus and absence of background fluorescence permitted both following of the dye front through the vasculature and clear imaging of the capillary network over the whole posterior pole. The bolus delivery could be repeated as many as 100 times in 45 min without significant loss of contrast. The comparison of these results with conventional fluorescein angiography illustrated the advantage of the new method. The examination of the safety of the delivery system indicates that there is no major obstacle to the eventual application to humans.

Quantitative analysis of retinal hemodynamics using targeted dye delivery.

A new method designed to allow repeated mapping of retinal hemodynamics on a macro- and microcirculatory level was evaluated in the primate eye. The method, called "targeted dye delivery," consists of encapsulating a fluorescent dye in temperature-sensitive liposomes, injecting the liposomes systemically, and using a light pulse from an argon laser to release a bolus of dye in a targeted retinal vessel. The follow-up of the well-defined dye front thus generated allows calculation of the blood flow and capillary transit time. Evaluation of targeted dye delivery in a monkey indicated that centerline blood velocity and the vessel diameter can be measured with a reproducibility of 10% and 4%, respectively, in vessels that are 40 microns and larger. These measurements yielded flow values that had a reproducibility of 10% on the same day and 13% on different days. The normalization of flow rate by the vessel diameter was consistent with theoretic estimates and promises to be a circulation indicator independent of variations between individual and species. The transit time across capillary beds at different locations was found to be similar, thus indicating that the method could be used to evaluate the local viability of the microcirculation.

Feasibility of targeted drug delivery to selective areas of the retina.

A new method was developed to deliver locally a bolus dose of a drug to the retinal vasculature. The targeted delivery system was based on encapsulating the drug in heat-sensitive liposomes, which are injected intravenously and lysed in the retinal vessels by a heat pulse generated by a laser. To test if substances delivered in the vessels could also penetrate into the surrounding tissue, 6-carboxyfluorescein was encapsulated in liposomes and used as a marker for drug penetration. Moderate argon laser pulses were applied to the retinal vessels of Dutch pigmented rabbits to induce breakdown of the blood-retinal barrier (BRB). A suspension of liposomes at a dose of 2 ml/kg body weight, corresponding to a carboxyfluorescein dose of 12 mg/kg, was injected into the ear vein. The dye was released from the liposomes proximal to the damaged portion of the vessel. Fundus fluorescein angiograms were recorded with a video camera and digitized for subsequent image analysis. The penetration of carboxyfluorescein into the retinal tissue was evaluated by comparing the fluorescence intensity of the area around the damaged vessel with that of an adjacent control area. The dye penetration increased with the numbers of laser applications (P less than 0.001). The leakage was localized distally to the released site and was restricted to areas with a disrupted BRB. The mass of carboxyfluorescein that penetrated gradually spread with time. Both veins and arteries could be used for the targeted delivery. These results indicated that this delivery system, which is fully controllable by laser through the pupil, can deliver drugs inside the vasculature and into the retinal tissue wherever the BRB is disrupted.

Occlusion of retinal vessels using targeted delivery of a platelet aggregating agent.

Local laser targeted delivery of a platelet aggregating agent to occlude retinal and choroidal vessels was evaluated in rabbits and rats. Liposomes containing adenosine diphosphate (ADP) were administered intravenously and an argon laser was used to lyse the liposomes in main retinal arteries. Control vessels were treated with the same energy of laser without administering ADP. Fluorescein angiography performed 2 weeks later showed that all the control vessels were perfused. Ninety percent of the ADP-treated arteries showed complete or partial occlusion. Successful occlusion increased with the laser energy and decreased with increasing vessel diameter. Histopathology showed that occlusion was achieved in retinal as well as choroidal vessels. The inner retina remained relatively unaffected at the treatment site but the outer retina was thermally damaged. These preliminary results suggest that targeted delivery of a platelet aggregating agent holds promise for occluding vessels in the fundus.