The effects of desferrioxamine on thrombus formation in injured microvessels of the rabbit ear Ichiro Hashimoto, Hideki Nakanishi, Yoshitaka Shono, and Shinji Tanaka Department of Plastic and Reconstructive Surgery, University Hospital, The University of Tokushima School of Medicine, Tokushima, Japan Abstract:We investigated the effects of the iron chelator desferrioxamine (DFX) on thrombus formation in the arterioles and venules of the rabbit ear chamber. Thrombi were induced by irradiation with filtered light in combination with a fluorescent dye. The occlusive thrombus formation time was significantly extended by DFX. The mor-phological composition of thrombi in the arterioles and venules was different. In the arterioles, the thrombi consisted of platelet aggregation, but in the venules, platelets and leukocytes accumulated on the endothelium. This suggests that hydroxyl radicals may be important mediators in this model, as DFX is known as a hydroxyl radical scav-enger. Furthermore, the components of thrombi in the arterioles and venules in the skin microvascular system may be different. J. Med. Invest. 46:200-204, 1999 Keywords:rabbit ear chamber, desferrioxamine, thrombus formation, microcirculation INTRODUCTION Desferrioxamine (DFX) is an iron chelator used for the treatment of acute iron overload states and is known to act as a hydroxyl radical scavenger (1, 2). Recently it has also been reported that DFX protects the endothelial cells of heart and lung (3, 4) and can also shield skin flaps from necrosis (5, 6).On the other hand, it has been shown that DFX does not benefit the endothelium or improve sal-vage of skin necrosis (7). In flap surgery, thrombus formation is one of the causes of flap necrosis. The effect of DFX on thrombosis is, however, not clear. In this study, we used the rabbit ear chamber (REC) as a model of skin microcirculation and observed the thrombus formation resulting from endothelial damage due to dye/light reaction. The effects of DFX on this damage were also examined. Since this thrombus model is usually used for mesenteric vessels, and as there have been no reports on thrombus formation in the skin, the morphological features are also described. MATERIALS AND METHODS Animals/ Rabbit Ear Chamber A transparent round-table chamber (rabbit ear chamber:REC) was implanted in the earlobe of Japanese white rabbits weighing between 2.7 and3.2kg (Fig.1) (8). Five to seven weeks after im-plantation, a complete microvascular network had formed in the chamber, and arterioles 20-30μm in diameter and venules 45-55μm in diameter were selected for the experiments (Fig. 2). Photochemically induced thrombosis model No anesthesia was needed for the following studies. Thrombi were produced in the vessels with the dye/light method reported by Sato and Ohshima(9, 10). Briefly, a solution of fluorescein sodium(2.5%, 2ml/kg;Wako Pure Chemical Industries, Ltd., Osaka, Japan) was injected through a vein of the contralateral ear. After confirming the appear-ance of fluorescence in the microvessels in the REC, filtered light (400-500nm light passed through a fluorescein isothiocyanate excitation filter from a high pressure mercury lamp) was irradiated onto the vessels through an objective lens (x20) (Fig. 3). The diameter of the irradiated spot was130μm on the focal plane. The light intensity was controlled at 0.5mW under the objective lens. Irradiated ves-sels were continuously monitored and recorded with an intravital microscope-television system in all cases and the time course of thrombus formation was calculated with the aid of these records. In all cases, the intervals "ti" (from irradiation to the onset of thrombus formation) and "ts" (from irradiation to complete obstruction of blood flow) were also cal-culated. Experimental Groups and Administration of DFX Thirty rabbits were prepared and divided into three groups of 10 animals each. In the control group, 10cc of saline was injected via the contralateral ear15minutes before aggregation was induced. In the two DFX treatment groups, 100mg/kg of DFX (Novartis Pharma Co., Tokyo, Japan) and10ml of saline, or 200mg/kg of DFX and10ml of saline was injected in the same way. Statistical Analysis The Mann Whitney U-test was used for statistical analysis of the results. P values less than 0.05were considered statistically significant. RESULTS When the filtered light was irradiated onto the vessels after confirmation of fluorescence in the microvessels in the RECs, microthrombi began to stick to the wall, and continued to accumulate on the inside of the vessels. Eventually the microvessels were totally occluded by the thrombi and the blood flow was completely blocked. The composition of the thrombi in the arterioles and venules was dif-ferent. Morphologically, the thrombus formation in the arterioles began with platelet aggregation (Fig. 4), but in the venules, platelets, erythrocytes and leukocytes started to gather and adhere to the endothelium from the start (Fig. 5). This difference between thrombosis in arterioles and venules was observed in all groups. In arterioles and venules, the thrombus forma-tion times, ti and ts, were extended as a result of DFX (Fig. 6). In the arterioles, ti and ts of the control group were 25.2±2.7 (mean±S.D.) and 84.3±4.6sec and those of the DFX100mg/kg group were 36.5±5.2 (p<0.05) and 95.7±9.2 sec (not signifi-cant). The times of the DFX200mg/kg group were 45.7±6.0 (p<0.01) and 111.0±8.0 sec (p<0.05). In the venules, ti and ts of the control group were 25.0±2.0 and104.9±5.6 sec and those of the DFX100mg/kg group were 38.3±7.4 (not significant) and 128.3±12.6 sec (not significant). The times of the DFX200mg/kg group were 43.7±7.6 (p<0.05) and135.1±11.7 sec (p<0.05). DISCUSSION Different methods to induce thrombosis in vivo are presently available. However, there are few reports on the application of thrombus models to the REC. Fleming and Seiler have evaluated the antithrombotic efficacy of some drugs by using a laser-induced thrombosis model in the REC (11, 12). In their model using irradiation by ruby laser to vessels, it is thought that the localized thermal effect on the red cells as a result of absorption of the ruby laser evoked the formation of a thrombus, which consisted of platelets accumulated around a core of one or two damaged red cells (13). It is, however, generally accepted that the first step of thrombogenesis is damage to endothelial cells. Therefore, their model differs from the physiologi-cal mechanism of thrombus formation. We applied the fluorescent dye/light method, which Sato and Ohshima developed for mesenteric vessels, to the REC. It has been reported that the mechanism of thrombus formation in the in vivo model of the dye/light reaction does not consist of a localized thermal effect but of endothelial dam-age due to forms of active oxygen generated with excited dyes (14, 15). However, details of the mech-anism are not clearly understood at present. Our data showed that DFX lengthened the throm-bus formation time in the arterioles and venules in a dose dependent manner. DFX is known to act as a hydroxyl radical scavenger (1, 2). With almost the same dye/light method as described here, it has been shown that, in the pial arterioles of the mouse, the hydroxyl radical scavengers dimethyl sulfoxide (DMSO) and glycerol are effective inhibitors of platelet aggregation (15). Hydroxyl radicals may therefore be important mediators of the skin micro-vascular responses in this model. The maximum clinical daily dose of DFX is80mg/kg for humans. As no statistically significant difference could be identified in the DFX 100 mg/kg group, the anti-thrombotic efficacy of this agent may not be very strong. However it has been reported that a drug mixture containing a small dose of DFX (20mg/kg) improves experimental flap survival (16, 17). With that method, a mixture including a small amount of each drug can be effective and DFX can be used in a clinically viable manner. Recently, it has been found that after a prolonged ischemic phase, reestablishment of the vascular supply causes tissue injury which is called ischemia reperfusion injury. In this injury, reactive oxygen species such as the superoxide anion radical (O2 -), hydrogen peroxide (H2O2), and the hydroxyl radical (OH·) lead to endothelial cell injury which results in increased capillary permeability, thrombus forma-tion, and microcirculatory failure. As the thrombi in our model may be due to endothelial injury by hydroxyl radicals, we are convinced that this model very closely resembles such clinical conditions (15, 18). In this study, it was observed that thrombus for-mation in the arterioles consisted of platelet aggre-gation, but in the venules both platelets and leukocytes gathered and adhered to the endothelium. When the same model as used in this study was applied to mesenteric vessels, platelet aggregation was observed in the venules (9, 10). This difference in composition of the thrombus between REC and the mesenteric vessels may be caused by charac-teristics of the organ. In the skin microcirculation, WBC contributed to thrombus formation only in the venules but not in the arterioles. It is not clear whether the leukocyte sticking in thrombus is caused by slow blood flow velocity or by the func-tional features of venous endothelial cells. Further studies are needed to investigate this point. REFERENCES 1. Gutteridge JM, Richmond R, Halliwell B:Inhibition of the iron-catalysed formation of hydroxyl radicals from superoxide and of lipid peroxidation by desferrioxamine. Biochem J184:469-472, 1979 2. Gutteridge JM, Rowley DA, Halliwell B: Superoxide-dependent formation of hydroxyl radicals and lipid peroxidation in the presence of iron salts. Detection of 'catalytic' iron and anti-oxidant activity in extracellular fluids. Bio-chem J206:605-609, 1982 3. Sellke FW, Shafique T, Ely DL, Weintraub RM:Coronary endothelial injury after cardio-pulmonary bypass and ischemic cardioplegia is mediated by oxygen-derived free radicals. Circulation88 (part 2):395-400, 1993 4. Pickford MA, Green CJ, Sarathchandra P, Fryer PR:Ultrastructural changes in rat lungs after 48h cold storage with and without reper-fusion. Int J Exp Pathol 71:513-528, 1990 5. Green CJ, Dhami L, Prasad S, Healing G, Shurey C :The effect of desferrioxamine on lipid peroxidation and survival of ischaemic island skin flaps in rats. Br J Plast Surg 42:565-569, 1989 6. Angel MF, Mellow CG, Knight KR, O'Brien BM:The effect of deferoxamine on tolerance to secondary ischaemia caused by venous obstruction. Br J Plast Surg42:422-424, 1989 7. Nishikawa H, Manek S, Fryer PR, Green CJ: Reperfusion injury in rat adipomusculocutaneous flaps:an ultrastructural and microangiographic study. Br J Plast Surg44:500-505, 1991 8. Asano M, Ohkubo C, Sasaki A:A modified rabbit ear chamber and an example of its application for intravital-microscopic study on acute effects of topical thermal stimulation. Jpn J Pharmacol48:149-152, 1988 9. Sato M, Ohshima N:Platelet thrombus induced in vivo by filtered light and fluorescent dye in mesenteric microvessels of the rat. Throb Res35:319-334, 1984 10. Sato M, Ohshima N : Hemodynamics at stenoses formed by growing platelet thrombi in mesenteric microvasculature of rat. Microvasc Res31:66-76, 1986 11. Seiler SM, Buchanan JO, Schumacher WA, Zavoico GB, Gamberdella M, Fleming JS, Meanwell NA:2-[3-[2-(4,5-diphenyl-2-oxazolyl) ethyl] phenoxy] acetic acid (BMY42393):2).Oral activity and efficacy in animal models of arterial thrombosis. Throb Res 74:125-133, 1994 12. Fleming JS, Buchanan JO, Seiler SM, Meanwell NA:Antithrombotic activity of BMY-43351, a new imidazoquinoline with enhanced aqueous solubility. Throb Res 63:145-155, 1991 13. Grant L, Becker FF:Mechanisms of inflamma-tion.1. Laser-induced thrombosis, a morphologic analysis. Proc Soc Exp Biol Med 119:1123-1129, 1965 14. Herrmann KS:Platelet aggregation induced in the hamster cheek pouch by a photochemical process with excited fluorescein isothiocyanate-dextran. Microvasc Res26:238-249, 1983 15. Rosenblum WI, El Sabban F : Dimethyl sulfoxide (DMSO) and glycerol, hydroxyl radical scav-engers, impair platelet aggregation within and eliminate the accompanying vasodilation of, injured mouse pial arterioles. Stroke 13:35-39, 1982 16. Lepore DA, Knight KR, Bhattacharya S, Ritz M, Robbins SP, Seig P, Morrison WA, O'Brien BM:Drug mixture which improves survival of ischemic rabbit epigastric skin flaps. Micro-surgery 15:685-692, 1994 17. Kohout M, Lepore DA, Knight KR, Van GJ, O'Brien BM, Stewart AG and Morrison WA: Cool perfusion solutions for skin flaps:a new mixture of pharmacological agents which im-proves skin flap viability. Br J Plast Surg 48:132-144, 1995 18. Matsuno H, Uematsu T, Umemura K, Takiguchi Y, Asai Y, Muranaka Y, Nakashima M : A simple and reproducible cerebral thrombosis model in rats induced by a photochemical reaction and the effect of a plasminogen-plasminogen activator chimera in this model. J Pharmacol Toxicol Methods29:165-173, 1993 Received for publication June 17, 1999;accepted July 22, 1999. Address correspondence and reprint requests to Ichiro Hashimoto, M.D., Department of Plastic and Reconstructive Surgery, University Hospital, The University of Tokushima School of Medicine, Kuramoto-cho, Tokushima770-8503, Japan and Fax:+81-88-633-7297.