- Email: [email protected]
Reduction of the extent of ischemic skeletal muscle necrosis by perfusion with oxygenated perfluorocarbon
SCIENTIFIC PAPERS
Reduction of the Extent of Ischemic Skeletal Muscle Necrosis by Perfusion With Oxygenated Perfluorocarbon* Chittur Mohan, MD, Mark (Jennaro, MD, Corrado Marini, MD, Enrico Ascer, MD, Brooklyn,NewYork
Oxygenated perfluorocarbon emulsion has been shown to preserve feline cerebral function after ischemia. The postulated protective effects of perfluorocarbons include improvement of blood theology and prevention of neutrophil adherence by nonchemical inhibition of surface receptors. In this study, we used a well-described graeilis muscle model to investigate whether oxygenated perfluorocarbon can minimize skeletal muscle necrosis by mitigating the degree of leukosequestration. In eight adult mongrel dogs, both gracilis muscles were weighed and then subjected to 6 hours of normothermic ischemia followed by 4 8 hours of normothermic reperfnsion. However, one randomly selected side (experimental side) was infused with oxygen (02) Fluosol-DA 20% (4.4 4- 0.2 mL O2/100 m L ) intra-arterially at 12 mL/min for 4 0 minutes immediately after ischemia. Muscle biopsy specimens were obtained before ischemia and after 1 hour and 4 8 hours of reperfnsion to estimate myeloperoxidase (MPO) activity, a m a r k e r of neutrophil infiltration. After 4 8 hours, both gracilis muscles were harvested and weighed in all animals. Muscle necrosis was measured by serial transections, nitroblue tetrazolium staining, and computerized planimetry. The transmuscular oxygen tension (pO2) of the gracilis muscle on the experimental side increased from 2 to 4 nun Hg during iscbemia to 315 4- 5 0 nun Hg during 02 Fluosol-DA 20% infusion. The percentage of muscle necrosis on the control side was 4 8 . 0 8 % 4- 8.46%, compared with 2 7 . 6 2 % 4- 6 . 9 6 % on the experimental side (p < 0 . 0 0 1 ). MPO activity was significantly higher at 4 8 hours of reperfnsion compared with pre-ischemic and 1-hour r e p e d u s i o n values ( 5 . 4 6 4- 1.52 U/mg tissue protein versus 0 . 0 6 4- 0.01 U/mg tissue protein and 0 . 1 6 4- 0 . 0 6 U/mg tissue protein, respectively, in the control group; 1.78 4- 0.60 U/mg tissue protein versus 0 . 1 6 4- 0.08 U/mg tissue prorein and 0.27 4- 0 . 1 0 U/mg tissue protein, respectively, in the experimental group, p < 0 . 0 5 ) . HowFrom the Divisionof Vascular Surgery,MaimonidesMedicalCenter, SUNY Health ScienceCenterof Brooklyn,Brooklyn,New York.This work was supportedby a grant from the MaimonidesMedical Center Researchand DevelopmentFoundation. *This work has been designated the 1992Peter B. SamuelsPrize Essayby a resident.Dr. Mohan is the recipientof the award. Requests for reprints should be addressed to EnricoAscer, MD, MaimonidesMedicalCenter,DivisionofVascularSurgery,4802Tenth Avenue,Brooklyn,New York 11219. Presentedat the 20th Annual Meeting of the Societyfor Clinical VascularSurgery,Orlando, Florida,March 25-29, 1992. 194
ever, MPO activity at 48 hours of reperfusion in the experimental group was significantly lower than in the control group ( p < 0 . 0 5 ) . There was no difference in the percentage of weight gain between the control and the experimental groups ( 3 8 . 3 1 % 49.36% and 2 8 . 3 4 % 4- 7.35%, respectively, p > 0 . 0 5 ) . These data show that perfluorocarbons minimize the extent of skeletal muscle necrosis in this canine model. Based on our data on MPO activity, we believe that the protective effect of perfluorocarbons is in part due to the decreased leukosequestration in the muscle during the periods of ischemia and reperfusion. The potential for clinical application of this new treatment modality is very attractive, since high mortality and morbidity rates continue to be reported in patients with acute limb ischemia. h e management of acute limb ischemia continT to be a challenging problem for the vascular surgeon. Advances in surgical technology and techniques have enabled surgeons to restore blood flow to the ischemic limb in most circumstances. Additionally, pharmacologic manipulation of the cellular response and understanding of the role of biochemical mediators have contributed to new approaches in the treatment of the acute ischemic limb. However, despite all recent advances in the management of acute vascular occlusion, the optimal treatment of reperfusion injury is still not defined. The pathophysiologic derangements that occur after restoration of blood flow to an ischemic limb, known as the reperfusion syndrome, continue to be responsible for the very high mortality rate associated with acute vascular occlusion [1,2]. The beneficial effects of reperfusion after prolonged skeletal muscle ischemia are negatively counterbalanced by the detrimental effects of the "no-reflow" phenomenon, which further impairs perfusion, and by the activation of cellular and biochemical mediators caused by the reperfusion itself [3]. Microcirculatory perfusion failure, the hallmark of the no-reflow phenomenon, is the result of increased blood viscosity [4,5], which causes stasis and intravascular clotting and narrowing of the arteriolar and capillary lumina [6-9]. Neutrophil activation, aggregation, and infiltration into the ischemic skeletal muscle play a dominant role in causing cellular damage: activated neutrophils produce oxygen free radicals through the N A D P H (reduced nicotinamide-adenine dinucleotide phosphate) oxidase enzyme system [10-13]. The CD1 !/CD18 receptors on the surface of neutrophils appear to play a determinant role in promoting adherence of
THE AMERICAN JOURNAL OF SURGERY VOLUME 164 SEPTEMBER 1992
PERFLUOROCARBONS IN SKELETAL MUSCLE ISCHEMIA
neutrophils to the endothelial surface. Adherence of neutrophils is a prerequisite for leukosequestration into the ischemic muscle, as evidenced by the fact that in vivo administration of monoclonal antibodies directed against the CD11 receptors significantly reduces neutrophil accumulation [14]. Since oxygenated perfluorocarbon emulsion (oxygen [02] Fhosol-DA 20%) can mitigate the effects of repotfusion after ischemia in a feline model, possibly by improving blood rhcology and preventing neutrophil adherence by nonchemical inhibition of their surface receptors [15,16], we designed a study to investigate whether 02 Fhosol-DA 20% can minimize the extent of skeletal muscle necrosis in a canine model of ischemia-reperfusion injury. MATERIALS AND METHODS Eight mongrel dogs, weighing 25 to 28 kg, were used for this study. All the animals were premedicated with intramuscular acepromazine (5 mg/kg) and then anesthetized with intravenous pentobarbital (30 mg/kg). After endotracheal intubation, they were ventilated with room air at a tidal volume of 15 mL/kg. The respiratory rate was adjusted to maintain carbon dioxide (CO2) tension between 35 and 45 mm Hg. Supplemental doses of intravenous pentobarbital were given as needed to maintain effective general anesthesia. A rectal probe was inserted to monitor the temperature, which was kept between 37~ and 39~ with a heating pad. Maintenance intravenous fluid (lactated Ringer's solution) was infused at the rate of 5 mL/kg/h, and any blood loss was replaced with the same solution in a 3:1 ratio. The bilateral isolated gracilis muscle preparation, as previously described, was used in this study [17]. Gracilis muscles were completely isolated on their major vascular pedicle by dividing the minor vascular pedicle and the collaterals arising from the underlying adductor magnus muscle. In addition, the gracilis nerve and the medial and lateral tendinous insertions were transected to eliminate all collateral vessels. All the branches of the gracilis vessels, as well as the parent femoral vessels, were divided. After complete vascular isolation, the muscles were weighed with a suspension spring balance. Original resting muscle length was restored by suturing the tendons back to their original position. The gracilis muscle on each side underwent 6 hours of complete normothermic ischemia by occlusion of the vascular pedicle with a microvascular clamp. Effective ischemia was documented by monitoring the transmuscular oxygen tension (pOe) with a Kontron transcutaneous pOE electrode (#530, W & W Electronics AG Basel, Munchenstein, Switzerland); ischemia was confirmed by a decrease of the transmuscular pOe from 75 to 85 nun Hg, at baseline, to 2 to 4 mm Hg after the application of the microvascular clamp. Preparation of oxygenated Fluosol-DA 20%: Fhosol-DA 20% (intravascular perfluorochemical emulsion, The Green Cross Corporation, Osaka, Japan) was prepared and oxygenated in the following manner. Fluosol consists of three separate parts, which were mixed before use. The main constituents are as follows: (1) the
Fhosol emulsion (porfluorodecalin, perfluorotri-n,propylamine, and poloxamer 188); (2) solution 1 (sodium chloride and potassium chloride); and (3) solution 2 (sodium chloride, dextrose, magnesium chloride, and calcium chloride). The additive solutions serve to adjust the pH, ionic strength, and osmotic pressure in the final 20% emulsion and were added separately and sequentially before administration. The Fhosol emulsion supplied as a 400-mL volume in a 500-mL flexible plastic container was stored at - 2 0 ~ before use. The additive solutions were stored at room temperature. The frozen Fluosol emulsion was placed in a protective plastic bag and thawed in a warming water bath (37~ The bag was gently agitated during the thawing period. After the emulsion had thawed, it was removed from the water bath. Solutions 1 and 2 were added separately and sequentially to the thawed Fluosol emulsion under aseptic conditions. The contents of the bag were mixed thoroughly but gently by inversion. The final preparation resulted in a 500-mL volume in the bag. With the use of the aseptic technique, a 14-gauge needle was inserted into the puncture zone of the emulsion bag's port. An oxygenation catheter was gently advanced through the needle into the bag until the distal tip of the catheter was positioned near the bottom of the bag. The bag of Fhosol was vented to rclicvc excess pressure during the oxygenation procedure by inserting a 18-gauge sterile needle into one of the ports of the bag. A 0.2-#m filter was connected to the hub of the vent needle. The oxygenation catheter was connected to a 100% oxygen source tank via a 0.2-#m filter. The gas was allowed to bubble through the Fluosol at a rate of 2 L/min for 15 minutes. The oxygen content of the O2 Fluosol-DA 20% was measured with a Lex-OE-COnapparatus (Lexington Instruments Corporation, Waltham, MA). This was measured to be 4.4 4- 0.2 mL O2/100 mL. Experimental protoeoh One randomly chosen side in each animal was selected as the experimental side, and the contralateral side served as the control. Each side underwent 6 hours of normothermic ischemia followed by normothermic reperfusion for 48 hours. However, the gracilis muscle on the experimental side was perfused with 02 Fhosol-DA 20% intra-arterially at a rate of 12 mL/min for 40 minutes immediately following the ischemic period before normothermic blood reperfusion was begun. Muscle biopsy specimens (I00 to 200 rag) were obtained before ischemia and after 1 hour and 48 hours of reperfusion on both the control and experimental sides to measure myeloperoxidase (MPO) activity, a marker of neutrophil infiltration. The biopsy specimens were immediately frozen in liquid nitrogen and stored at -80~ All muscles were harvested at the end of the experiment and weighed, and the percentage of muscle necrosis was calculated. MPO assay: Skeletal muscle biopsy specimens of approximately I00 to 200 mg were homogenized in 2 mL of an ice-cold solution of I00 mM sodium chloride, 20 mM NaPO4, and 15 mM sodium-EDTA (cthylenediaminetetraacetic acid) (pH 4.7) using a tissue homogenizer and centrifuged at 20,000g for 15 minutes at 4 ~ The supernatant, which contained the water-soluble heine proteins
THE AMERICAN JOURNAL OF SURGERY
VOLUME 164 SEPTEMBER 1992
195
MOHAN ET AL
TABLE 1
Percentage of Muscle Necrosis and Muscle Weight Gain Control Side Muscle necrosis Muscle weigh~ gain
(%)
48.08 --- 8.46 38.31 +- 9.36
Experimental Side
(%)
27.62 -- 6.96* 28.34 ~ 7.35
*Significantly lower than control side by paired t-test (p < 0.001).
T A B L E II
Myeloperoxidase Activity (Units/mg Tissue Protein)
Preischemia 1-hour reperfusion 48-hour repen'usion*
Control Side
Experimental Side
0.06 __. 0.01 0.16 -~ 0.06 5.46 ~ 1.52
0.16 +- 0.08 0.27 _ 0.10 1.78 + 0 . 6 0 T
multiplied by the weight of the slice to estimate the weight of necrotic tissue. The sum of the necrotic tissue for each of the six slices was defined as the percentage of muscle necrosis for each muscle. All animals received humane care and treatment in compliance with the Guidefor the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication No. 85-23, revised 1985). Statistical analysis: Results are expressed as mean 4standard error of the mean (SEM). The differences in muscle necrosis and MPO activity at various time intervals between the two groups were compared for statistical significance using a paired t-test. The values of the MPO activity at various time intervals within each group were analyzed with analysis of variance.
RESULTS Percentage of muscle necrosis (Table I): The transmuscular pC2 of the gracilis muscle on the experimental side increased from 2 to 4 mm Hg during the ischemic period to 315 4- 50 mm Hg during 02 Fluosol-DA 20% infusion. The percentage of muscle necrosis, as deterhemoglobin and myoglobin, was decanted, and the pellet mined by NBT staining, was 48.08% 4- 8.46% in control was resuspended in ice-cold 50-mM potassium phosphate muscles, as compared with 27.62% 4- 6.96% (p <0.001) buffer containing 0.5% hexadecyltrimethylammonium in muscles perfused with intra-arterial O2 Huosol-DA bromide, pH 5.4. This suspension was rehomogenized 20% (experimental group). MPO activity: As shown in Table IL MPO activity in and sonieated in an ice bath for 10 seconds, freeze-thawed three times, and centrifuged at 20,000g for 15 minutes at the control group increased significantly from a pro-is4~ The supernatant was collected and assayed for total chemic level of 0.06 4- 0.01 U/mg tissue protein to 5.46 41.52 U/mg tissue protein at 48 hours of reperfusion (p protein content and MPO activity. Enzyme activity was measured by monitoring the <0.05). However, there was no significant change at 1 hour of reperfusion (0.16 4- 0.06 U/mg tissue protein) H202-dependent oxidation of 3,3',5,5'-tetramethylbenzidine (TMB) [18]. A spectrophotometer (Spectronic 601, compared with the pre-ischemic value (0.06 4- 0.01 Milton Roy Co., Rochester, NY) was used to monitor the U/mg tissue protein). A similar trend was seen in the generation of oxidized TMB at 655 nm. The reaction experimental group (0.16 4- 0.08 U/mg tissue protein at mixture contained 40 t~L of supernatant, 1.6 mM of pre-ischemia compared with 0.27 4- 0.10 U/mg tissue TMB, 0.3 mM of H202, 80 mM of sodium phosphate protein at 1 hour of reperfusion and 1.78 4- 0.60 U/mg buffer (pH 5.4), 8% N,N-dimethylformamide, and 40% tissue protein at 48 hours of reperfusion). Nevertheless, Dulbecco's phosphate-buffered saline in a total volume of the rise at 48 hours of reperfusion was significantly lower 1 mL. Activity was expressed as the initial velocity of in the experimental group compared with the control absorbance increase at 655 nm/min/mg of tissue protein. group (p <0.05). Graeilis muscle weight: The gracilis muscle weight One unit of enzyme activity was defined as the amount of MPO that produced an absorbance change of 1.00D/ increased significantly in both the control and experimental groups. In control muscles, it rose from a baseline min/mg of tissue protein during incubation at 37~ Protein content was measured with the Coomassie blue value of 94 4- 3 g to 130 4- 9 g (p <0.05). A similar trend method of Bradford [19]. This assay measures the was seen on the Huosol-treated side, where it increased change in absorbance at 595 nm and was done with a from 94 4- 4 g at baseline to 120 4- 6 g at the end of Gilford Response II UV/VIS spectrophotometer (CIBA reperfusion (p <0.05). However, the percentage inComing Diagnostics Corp., Gilford Systems, Oberlin, creases in the control and experimental groups were not significantly different (38.31% 4- 9.36% versus 28.34% 4OH). Muscle necrosis measurement: After removal, mus- 7.35%, p >0.05). cles were sectioned perpendicularly to their long axis into six equal slices. The slices were weighed and incubated in COMMENTS Despite all the treatment modalities available and a 0.5% nitroblue tetrazolium (NBT) solution for 20 minutes at 250C [20]. Areas of viable muscle were stained better understanding of the pathophysiologic mechablue, and necrotic areas remained unstained. The total nisms of reperfusion injury, acute skeletal muscle isarea at risk and the area of necrosis were determined on chemia continues to be a significant cause of mortality both sides of each slice by computerized planimetry, and and morbidity [1,2]. The optimal management to prevent the means of each were calculated [2I]. The ratio of reperfusion injury continues to be a challenge for surmean nonviable area to total area, in each slice, was geom. *Significantly higher than preischemia and 1-hour reperfusion in both groups oy analysis of variance (p < 0.05). TSignificantly lower than control side by paired t-test (p < 0.05).
196
THE AMERICAN JOURNAL OF SURGERY
VOLUME 164 SEPTEMBER 1992
PERFLUOROCARBONS IN SKELETALMUSCLEISCHEMIA
There are two major factors that seem to play a role in the pathogenesis of reperfusion syndrome: the no-reflow phenomenon, which causes the myocytes to progress from a reversibly injured state to an irreversible one, and direct cellular damage that follows the restoration of blood flow. Some of the pathogenetic factors of the no-reflow phenomenon include fibrin deposition and thrombus formation in the microcirculation, cellular edema occurring during ischemia, capillary collapse, and endothelial cell damage and dysfunction [3,22-24]. Alternatively, cellular injury is mediated by production of oxygen free radicals predominantly by the activated neutrophils [10-13]. Therefore, any attempt at mitigating ischemic injury should address both the effects of the no-reflow phenomenon and the injury related to reperfusion itself [3]. Impairment of the microcirculation after acute ischemia may be a determinant factor in the pathogenesis of infarction. Microcirculatory failure occurs secondarily to stasis and intravascular clotting due to increased blood viscosity [4,5] and capillary narrowing [6-9]. Blood viscosity relates directly to a concentration of plasma protein and of formed elements of the blood [25]. The concentration of these osmotically active substances is increased during ischemia by the shift of inorganic ions, namely, sodium, from the extracellular to the intracellular compartment [26,27]. Because ischemia paralyzes the sodium-potassium (Na +/K +) pump, the active transcellular transport of electrolytes is impaired [28]; therefore, water migrates from the intravascular to the intracellular compartment, causing increased blood viscosity and increased tissue pressure. The increased blood viscosity in conjunction with a spastic response of the microcirculation may be responsible for progressive and irreversible ischemia. Perfluorochemical emulsion (Fluosol-DA 20%), an acellular blood substitute, can exert a protective effect on the ischemic myocardium by improving myocardial blood flow and can actually reduce infarct size [29-32]. Some of the beneficial effects appear to be due to inhibition of neutrophil chemotaxis and lysosomal enzyme release [31]. Intracoronary perfusion of oxygenated perfluorochemical has also been shown to reduce the no-reflow injury in myocardial ischemia models, in addition to causing relative preservation of endothelial cell integrity [33]. Fluosol-DA 20% significantly decreased the percentage of gracilis muscle necrosis after 6 hours of normothermic ischemia followed by reperfusion. Although neutrophil infiltration is known to continue for 48 hours after ischemia [34,35], we found that a limited period of Fluosol-DA 20% infusion (40 minutes) in the ischemic muscles, before blood reperfusion, was effective in reducing the extent of muscle necrosis. Obviously, it would b e important to evaluate whether a prolonged infusion of perfluorocarbon could reduce the extent of muscle necrosis even further. Of interest, although oxygenated Fluosol-DA 20% increased the oxygen tension of the gracilis muscle to higher values (315 q- 50 mm Hg) than are obtained during reperfusion with blood, this did not affect
its capability to minimize muscle necrosis. Hence, one can infer that an hyperoxic environment, in the absence of cellular elements, does not play a pathogenetic role in muscle necrosis. Cytotoxic oxygen free radicals are the product of activated neutrophils exposed to oxygen during reperfusion. This study also confirms the predominant role played by activated neutrophils in the production of oxygen free radicals. Since there was no significant difference in the percentage weight gain in experimental and control muscles, we believe that perfluorocarbon infusion probably does not affect vascular permeability after ischemia-reperfusion. However, because weight gain is not as sensitive a methodology as measurement of radioactive-labeled albumin leakage, we cannot draw scientific conclusions regarding the effect of perfluorocarbon on vascular permeability. The beneficial effects of oxygenated perfluorocarbon emulsion infusion seem to be neutrophil mediated. Since the MPO activity at 48 hours of reperfusion was significantly lower than that of the control side (1.78 4- 0.6 U/ mg tissue protein versus 5.46 4- 1.52 U/mg tissue protein), it is possible that there is a reduction in neutrophil sequestration into the ischemic muscle due to their action of inhibiting neutrophil surface receptors (CD11/CD18 molecules). The surface receptors of neutrophils (CD11/CD18 molecules) are necessary for their adherence to the endothelium, which, in turn, appears to be a prerequisite before their migration into the ischemic skeletal muscle. Another aspect that requires investigation is the effect of perfluorocarbon emulsion on cytokine production after skeletal muscle ischemia and reperfusion. We have previously shown that there is increased production of interleukin-1 when skeletal muscle is subjected to prolonged ischemia followed by reperfusion [36]. Since interleukin-1 is known to be an important mediator that causes expression of CD 11/ CD 18 surface receptors of neutrophils, we believe that it is important to study the effect of perfluorocarbons on interleukin-1 release in ischemia-reperfusion. The inhibitory effect of perfluorocarbons on the surface receptors of neutrophils may be mediated indirectly via its effect on interleukin-1 production. Based on the results of this study, we conclude that intra-arterial infusion of oxygenated perfluorocarbon emulsion at the time of reperfusion appears to be a promising novel approach to decrease the extent of muscle necrosis after prolonged skeletal muscle ischemia followed by reperfusion. This approach is attractive because it is applicable to the clinical situation in which patients are first seen after the onset of ischemia and, therefore, cannot be pretreated. In the clinical arena, the only therapeutic modalities applicable to patients are those that can be instituted at the time of restoration of blood flow. REFERENCES 1. Haimovici H. Metabolic complications of acute arterial occlusions. J Cardiovasc Surg (Torino) 1979; 20: 349-57.
2. HaimoviciH. Arterial embolismwith acute massiveischemic myopathy and myoglobinuria. Surgery 1960; 116: 414-21.
THE AMERICAN JOURNAL OF SURGERY VOLUME 164 SEPTEMBER 1992 197
MOHAN ET AL
3. Quinones-Baldrich WJ. The role of fibrinolysis during reperfusion of isehemic skeletal muscle. Microeirc Endothelinm Lymphatics 1989; 5: 299-314. 4. Little JR. Implanted device for middle cerebral artery occlusion in conscious cats. Stroke 1977; 8: 258-60. 5. Boschenstein FK, Reilly JA, Yahr MD, Correil JW. Effect of low molecular weight dextran on cortical blood flow. Arch Neurol 1966; 14: 288-93. 6. Little JR. Modification of acute isehemia by treatment with mannitol. Stroke 1978; 9: 4-9. 7. Ames A III, Wright RL, Kowada M, Thurston JM, Majno G. Cerebral isohemia. The no-reflow phenomenon. Am J Pathol 1968; 52: 437-53. 8. Little JR. Microvascular alterations and edema in focal cerebral isohemia. In: Pappius H, Feindel W, editors. Dynamics of brain edema. New York: Springer-Verlag, 1976: 236-43. 9. Chiang J, Kowada M, Ames A III, Wright RL, Majno G. Cerebral isehemia III. Vascular changes. Am J Pathol 1968; 52: 455-76. 10. Rornson J, Hook BG, Kunkel SL, Abrams GD, Sehork MA, Luchessi BR. Reduction of the extent of ischemic myocardialinjury by ncutrophil depletion in the dog. Circulation 1983; 67:1016-23. 11. Engler R, Dahlgren M, Morris D, Pcterson M, Schmid-Schonbein G. Role of leukoeytes in response to acute myocardial isehemia and reflow in dogs. Am J Physiol 1986; 251: H314-22. 12. MeCord JM. Oxygen-derived free radicals in post-ischemic tissue injury. N Engl J Med 1985; 312: 159-63. 13. Noveili GP, Livi P, Ghinussi L, Lisi L, Brunelleschi S, Fantozzi R. Superoxide generation by granuloeytes during superior mesenteric artery occlusion shock in rabbits. In: Novelli GP, Ursini F, editors. Oxygen radicals in shock. International workshop. Basel: Karger, 1986: 587-93. 14. Simpson P J, Todd RF, Fantone JC, et al. Reduction of experimental canine myocardial reperfusion injury by a monoclonal antibody (anti-Mol, anti-CD 1lb) that inhibits leukocyte adhesion. J Clin Invest 1988; 81: 624-9. 15. Osterholm JL, Alderman JB, Triolo AJ, D'Amore BR, Williams HD, Frazer G. Severe cerebral isehemia treatment by ventriculosubarachnoid perfusion with an oxygenated fluorocarbon emulsion. Neurosurgery 1983; 13: 381-7. 16. Peerless SJ, Ishikawa R, Hunter IG, Peerless MJ. Protective effect of Fluosol-DA in acute cerebral isehemia. Stroke 1981; 12: 558-63. 17. Kuzon WM, Walker PM, Mickle DAG, Harris KA, Pynn BR, McKee NH. An isolated skeletal muscle model suitable for acute ischemia studies. J Surg Res 1986; 41: 24-32. 18. Suzuki K, Ota H, Sasagawa S, Sakatani T, Fujikura T. Assay method for myeloperoxidase in human polymorphonuclear leukocytes. Anal Biochem 1982; 132: 345-52. 19. Bradford M. A rapid and sensitive method for the quantitation
198
of microgram quantities of protein utilizing the principle of proteindye binding. Anal Biochem 1976; 72: 248-54. 20. Novikoff AB, Shin W, Drucker JJ. Staining results with two tetrazolium salts. J Biophys Biochem Cytol 1961; 9: 47-61. 21. Labbe R, Lindsay T, Walker PM. The extent and distribution of skeletal muscle necrosis after graded periods of complete ischemia. J Vase Surg 1987; 6: 152-7. 22. Chait LA, May JW Jr, O'Brian BM, Hurley JV. The effects of the perfusion of various solutions on the no reflow phenomenon in experimental free flaps. Hast Reconstr Surg 1978; 61: 421-30. 23. Leaf A. Cell swelling. A factor in isehemic tissue injury. Circulation 1973; 48: 455-8. 24. Stroek PE, Majno G. Mierovaseular changes in experimental tourniquet isehemia. Surg Gynecol Obstet 1969; 129: 1213-23. 25. Wells RE. Rheology of blood in the microvaseulature. N Engl J Med 1964; 270: 832-9. 26. Edstrom RFS, Essex HE. Swelling of the brain induced by anoxia. Neurology (Minneapolis) 1956; 6:118-24. 2 7 . Van Harreveld A. Changes in volume of cortical neuronal elements during asphyxiation. Am J Physiol 1957; 191: 233-42. 28. Leaf A. On the mechanism of fluid exchange of tissues in vitro. Biochem J 1957; 62: 241-8. 29. Forman MB, Bingham SE, Kopelman HA, et al. Reduction of infarct size with intracoronary perfluorochemical in a canine preparation of reperfusion. Circulation 1985; 71: 1060-8. 30. Forman MB, Puett DW, Bingham SE, et al. Preservation of endothelial cell structure and function by intracoronary perfluorochemical in a canine preparation of reperfusion. Circulation 1987; 76: 469-79. 31. Bajaj AK, Cobb MA, Virmani R, Gay JC, Light RT, Forman MB. Limitation of myocardial reperfusion injury by intravenous perfluorochemicals; role of ncutrophil activation. Circulation 1989; 79: 645-56. 32. Forman MB, Puett DW, Wilson BH, et al. Beneficial longterm effect of intracoronary perfluorochemieal on infarct size and ventricular function in a canine reperfusion model. J Am Coil Cardiol 1987; 9: 1082-90. 33. Kolodgie FD, Virmani R, Farb A. Limitation of no reflow injury by blood-free reperfusion with oxygenated perfluoroehemical (Fluosol-DA 20%). J Am Coil Cardiol 1991; 18: 215-23. 34. Rubin BB, Smith A, Liauw S, Isenman D, Romasehin AD, Walker PM. Complement activation and white cell sequestration in postisehemic skeletal muscle. Am J Physiol 1990; 259: H525-31. 35. Mohan C, Ascer E, Marini CP, Maughan R, Gennaro M. Does iloprost mediate thromboxane release and polymorphonuclear leukocyte sequestration in ischemic skeletal muscle? J Cardiovase Surg 1992 (in press). 36. Aseer E, Mohan C, Gennaro M, Cupo S. Interleukin-1 and thromboxane release after skeletal muscle ischemia and reperfusion. Ann Vase Surg 1992; 6: 69-73.
THE AMERICAN JOURNAL OF SURGERY VOLUME 164 SEPTEMBER 1992
Reduction of the Extent of Ischemic Skeletal Muscle Necrosis by Perfusion With Oxygenated Perfluorocarbon* Chittur Mohan, MD, Mark (Jennaro, MD, Corrado Marini, MD, Enrico Ascer, MD, Brooklyn,NewYork
Oxygenated perfluorocarbon emulsion has been shown to preserve feline cerebral function after ischemia. The postulated protective effects of perfluorocarbons include improvement of blood theology and prevention of neutrophil adherence by nonchemical inhibition of surface receptors. In this study, we used a well-described graeilis muscle model to investigate whether oxygenated perfluorocarbon can minimize skeletal muscle necrosis by mitigating the degree of leukosequestration. In eight adult mongrel dogs, both gracilis muscles were weighed and then subjected to 6 hours of normothermic ischemia followed by 4 8 hours of normothermic reperfnsion. However, one randomly selected side (experimental side) was infused with oxygen (02) Fluosol-DA 20% (4.4 4- 0.2 mL O2/100 m L ) intra-arterially at 12 mL/min for 4 0 minutes immediately after ischemia. Muscle biopsy specimens were obtained before ischemia and after 1 hour and 4 8 hours of reperfnsion to estimate myeloperoxidase (MPO) activity, a m a r k e r of neutrophil infiltration. After 4 8 hours, both gracilis muscles were harvested and weighed in all animals. Muscle necrosis was measured by serial transections, nitroblue tetrazolium staining, and computerized planimetry. The transmuscular oxygen tension (pO2) of the gracilis muscle on the experimental side increased from 2 to 4 nun Hg during iscbemia to 315 4- 5 0 nun Hg during 02 Fluosol-DA 20% infusion. The percentage of muscle necrosis on the control side was 4 8 . 0 8 % 4- 8.46%, compared with 2 7 . 6 2 % 4- 6 . 9 6 % on the experimental side (p < 0 . 0 0 1 ). MPO activity was significantly higher at 4 8 hours of reperfnsion compared with pre-ischemic and 1-hour r e p e d u s i o n values ( 5 . 4 6 4- 1.52 U/mg tissue protein versus 0 . 0 6 4- 0.01 U/mg tissue protein and 0 . 1 6 4- 0 . 0 6 U/mg tissue protein, respectively, in the control group; 1.78 4- 0.60 U/mg tissue protein versus 0 . 1 6 4- 0.08 U/mg tissue prorein and 0.27 4- 0 . 1 0 U/mg tissue protein, respectively, in the experimental group, p < 0 . 0 5 ) . HowFrom the Divisionof Vascular Surgery,MaimonidesMedicalCenter, SUNY Health ScienceCenterof Brooklyn,Brooklyn,New York.This work was supportedby a grant from the MaimonidesMedical Center Researchand DevelopmentFoundation. *This work has been designated the 1992Peter B. SamuelsPrize Essayby a resident.Dr. Mohan is the recipientof the award. Requests for reprints should be addressed to EnricoAscer, MD, MaimonidesMedicalCenter,DivisionofVascularSurgery,4802Tenth Avenue,Brooklyn,New York 11219. Presentedat the 20th Annual Meeting of the Societyfor Clinical VascularSurgery,Orlando, Florida,March 25-29, 1992. 194
ever, MPO activity at 48 hours of reperfusion in the experimental group was significantly lower than in the control group ( p < 0 . 0 5 ) . There was no difference in the percentage of weight gain between the control and the experimental groups ( 3 8 . 3 1 % 49.36% and 2 8 . 3 4 % 4- 7.35%, respectively, p > 0 . 0 5 ) . These data show that perfluorocarbons minimize the extent of skeletal muscle necrosis in this canine model. Based on our data on MPO activity, we believe that the protective effect of perfluorocarbons is in part due to the decreased leukosequestration in the muscle during the periods of ischemia and reperfusion. The potential for clinical application of this new treatment modality is very attractive, since high mortality and morbidity rates continue to be reported in patients with acute limb ischemia. h e management of acute limb ischemia continT to be a challenging problem for the vascular surgeon. Advances in surgical technology and techniques have enabled surgeons to restore blood flow to the ischemic limb in most circumstances. Additionally, pharmacologic manipulation of the cellular response and understanding of the role of biochemical mediators have contributed to new approaches in the treatment of the acute ischemic limb. However, despite all recent advances in the management of acute vascular occlusion, the optimal treatment of reperfusion injury is still not defined. The pathophysiologic derangements that occur after restoration of blood flow to an ischemic limb, known as the reperfusion syndrome, continue to be responsible for the very high mortality rate associated with acute vascular occlusion [1,2]. The beneficial effects of reperfusion after prolonged skeletal muscle ischemia are negatively counterbalanced by the detrimental effects of the "no-reflow" phenomenon, which further impairs perfusion, and by the activation of cellular and biochemical mediators caused by the reperfusion itself [3]. Microcirculatory perfusion failure, the hallmark of the no-reflow phenomenon, is the result of increased blood viscosity [4,5], which causes stasis and intravascular clotting and narrowing of the arteriolar and capillary lumina [6-9]. Neutrophil activation, aggregation, and infiltration into the ischemic skeletal muscle play a dominant role in causing cellular damage: activated neutrophils produce oxygen free radicals through the N A D P H (reduced nicotinamide-adenine dinucleotide phosphate) oxidase enzyme system [10-13]. The CD1 !/CD18 receptors on the surface of neutrophils appear to play a determinant role in promoting adherence of
THE AMERICAN JOURNAL OF SURGERY VOLUME 164 SEPTEMBER 1992
PERFLUOROCARBONS IN SKELETAL MUSCLE ISCHEMIA
neutrophils to the endothelial surface. Adherence of neutrophils is a prerequisite for leukosequestration into the ischemic muscle, as evidenced by the fact that in vivo administration of monoclonal antibodies directed against the CD11 receptors significantly reduces neutrophil accumulation [14]. Since oxygenated perfluorocarbon emulsion (oxygen [02] Fhosol-DA 20%) can mitigate the effects of repotfusion after ischemia in a feline model, possibly by improving blood rhcology and preventing neutrophil adherence by nonchemical inhibition of their surface receptors [15,16], we designed a study to investigate whether 02 Fhosol-DA 20% can minimize the extent of skeletal muscle necrosis in a canine model of ischemia-reperfusion injury. MATERIALS AND METHODS Eight mongrel dogs, weighing 25 to 28 kg, were used for this study. All the animals were premedicated with intramuscular acepromazine (5 mg/kg) and then anesthetized with intravenous pentobarbital (30 mg/kg). After endotracheal intubation, they were ventilated with room air at a tidal volume of 15 mL/kg. The respiratory rate was adjusted to maintain carbon dioxide (CO2) tension between 35 and 45 mm Hg. Supplemental doses of intravenous pentobarbital were given as needed to maintain effective general anesthesia. A rectal probe was inserted to monitor the temperature, which was kept between 37~ and 39~ with a heating pad. Maintenance intravenous fluid (lactated Ringer's solution) was infused at the rate of 5 mL/kg/h, and any blood loss was replaced with the same solution in a 3:1 ratio. The bilateral isolated gracilis muscle preparation, as previously described, was used in this study [17]. Gracilis muscles were completely isolated on their major vascular pedicle by dividing the minor vascular pedicle and the collaterals arising from the underlying adductor magnus muscle. In addition, the gracilis nerve and the medial and lateral tendinous insertions were transected to eliminate all collateral vessels. All the branches of the gracilis vessels, as well as the parent femoral vessels, were divided. After complete vascular isolation, the muscles were weighed with a suspension spring balance. Original resting muscle length was restored by suturing the tendons back to their original position. The gracilis muscle on each side underwent 6 hours of complete normothermic ischemia by occlusion of the vascular pedicle with a microvascular clamp. Effective ischemia was documented by monitoring the transmuscular oxygen tension (pOe) with a Kontron transcutaneous pOE electrode (#530, W & W Electronics AG Basel, Munchenstein, Switzerland); ischemia was confirmed by a decrease of the transmuscular pOe from 75 to 85 nun Hg, at baseline, to 2 to 4 mm Hg after the application of the microvascular clamp. Preparation of oxygenated Fluosol-DA 20%: Fhosol-DA 20% (intravascular perfluorochemical emulsion, The Green Cross Corporation, Osaka, Japan) was prepared and oxygenated in the following manner. Fluosol consists of three separate parts, which were mixed before use. The main constituents are as follows: (1) the
Fhosol emulsion (porfluorodecalin, perfluorotri-n,propylamine, and poloxamer 188); (2) solution 1 (sodium chloride and potassium chloride); and (3) solution 2 (sodium chloride, dextrose, magnesium chloride, and calcium chloride). The additive solutions serve to adjust the pH, ionic strength, and osmotic pressure in the final 20% emulsion and were added separately and sequentially before administration. The Fhosol emulsion supplied as a 400-mL volume in a 500-mL flexible plastic container was stored at - 2 0 ~ before use. The additive solutions were stored at room temperature. The frozen Fluosol emulsion was placed in a protective plastic bag and thawed in a warming water bath (37~ The bag was gently agitated during the thawing period. After the emulsion had thawed, it was removed from the water bath. Solutions 1 and 2 were added separately and sequentially to the thawed Fluosol emulsion under aseptic conditions. The contents of the bag were mixed thoroughly but gently by inversion. The final preparation resulted in a 500-mL volume in the bag. With the use of the aseptic technique, a 14-gauge needle was inserted into the puncture zone of the emulsion bag's port. An oxygenation catheter was gently advanced through the needle into the bag until the distal tip of the catheter was positioned near the bottom of the bag. The bag of Fhosol was vented to rclicvc excess pressure during the oxygenation procedure by inserting a 18-gauge sterile needle into one of the ports of the bag. A 0.2-#m filter was connected to the hub of the vent needle. The oxygenation catheter was connected to a 100% oxygen source tank via a 0.2-#m filter. The gas was allowed to bubble through the Fluosol at a rate of 2 L/min for 15 minutes. The oxygen content of the O2 Fluosol-DA 20% was measured with a Lex-OE-COnapparatus (Lexington Instruments Corporation, Waltham, MA). This was measured to be 4.4 4- 0.2 mL O2/100 mL. Experimental protoeoh One randomly chosen side in each animal was selected as the experimental side, and the contralateral side served as the control. Each side underwent 6 hours of normothermic ischemia followed by normothermic reperfusion for 48 hours. However, the gracilis muscle on the experimental side was perfused with 02 Fhosol-DA 20% intra-arterially at a rate of 12 mL/min for 40 minutes immediately following the ischemic period before normothermic blood reperfusion was begun. Muscle biopsy specimens (I00 to 200 rag) were obtained before ischemia and after 1 hour and 48 hours of reperfusion on both the control and experimental sides to measure myeloperoxidase (MPO) activity, a marker of neutrophil infiltration. The biopsy specimens were immediately frozen in liquid nitrogen and stored at -80~ All muscles were harvested at the end of the experiment and weighed, and the percentage of muscle necrosis was calculated. MPO assay: Skeletal muscle biopsy specimens of approximately I00 to 200 mg were homogenized in 2 mL of an ice-cold solution of I00 mM sodium chloride, 20 mM NaPO4, and 15 mM sodium-EDTA (cthylenediaminetetraacetic acid) (pH 4.7) using a tissue homogenizer and centrifuged at 20,000g for 15 minutes at 4 ~ The supernatant, which contained the water-soluble heine proteins
THE AMERICAN JOURNAL OF SURGERY
VOLUME 164 SEPTEMBER 1992
195
MOHAN ET AL
TABLE 1
Percentage of Muscle Necrosis and Muscle Weight Gain Control Side Muscle necrosis Muscle weigh~ gain
(%)
48.08 --- 8.46 38.31 +- 9.36
Experimental Side
(%)
27.62 -- 6.96* 28.34 ~ 7.35
*Significantly lower than control side by paired t-test (p < 0.001).
T A B L E II
Myeloperoxidase Activity (Units/mg Tissue Protein)
Preischemia 1-hour reperfusion 48-hour repen'usion*
Control Side
Experimental Side
0.06 __. 0.01 0.16 -~ 0.06 5.46 ~ 1.52
0.16 +- 0.08 0.27 _ 0.10 1.78 + 0 . 6 0 T
multiplied by the weight of the slice to estimate the weight of necrotic tissue. The sum of the necrotic tissue for each of the six slices was defined as the percentage of muscle necrosis for each muscle. All animals received humane care and treatment in compliance with the Guidefor the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication No. 85-23, revised 1985). Statistical analysis: Results are expressed as mean 4standard error of the mean (SEM). The differences in muscle necrosis and MPO activity at various time intervals between the two groups were compared for statistical significance using a paired t-test. The values of the MPO activity at various time intervals within each group were analyzed with analysis of variance.
RESULTS Percentage of muscle necrosis (Table I): The transmuscular pC2 of the gracilis muscle on the experimental side increased from 2 to 4 mm Hg during the ischemic period to 315 4- 50 mm Hg during 02 Fluosol-DA 20% infusion. The percentage of muscle necrosis, as deterhemoglobin and myoglobin, was decanted, and the pellet mined by NBT staining, was 48.08% 4- 8.46% in control was resuspended in ice-cold 50-mM potassium phosphate muscles, as compared with 27.62% 4- 6.96% (p <0.001) buffer containing 0.5% hexadecyltrimethylammonium in muscles perfused with intra-arterial O2 Huosol-DA bromide, pH 5.4. This suspension was rehomogenized 20% (experimental group). MPO activity: As shown in Table IL MPO activity in and sonieated in an ice bath for 10 seconds, freeze-thawed three times, and centrifuged at 20,000g for 15 minutes at the control group increased significantly from a pro-is4~ The supernatant was collected and assayed for total chemic level of 0.06 4- 0.01 U/mg tissue protein to 5.46 41.52 U/mg tissue protein at 48 hours of reperfusion (p protein content and MPO activity. Enzyme activity was measured by monitoring the <0.05). However, there was no significant change at 1 hour of reperfusion (0.16 4- 0.06 U/mg tissue protein) H202-dependent oxidation of 3,3',5,5'-tetramethylbenzidine (TMB) [18]. A spectrophotometer (Spectronic 601, compared with the pre-ischemic value (0.06 4- 0.01 Milton Roy Co., Rochester, NY) was used to monitor the U/mg tissue protein). A similar trend was seen in the generation of oxidized TMB at 655 nm. The reaction experimental group (0.16 4- 0.08 U/mg tissue protein at mixture contained 40 t~L of supernatant, 1.6 mM of pre-ischemia compared with 0.27 4- 0.10 U/mg tissue TMB, 0.3 mM of H202, 80 mM of sodium phosphate protein at 1 hour of reperfusion and 1.78 4- 0.60 U/mg buffer (pH 5.4), 8% N,N-dimethylformamide, and 40% tissue protein at 48 hours of reperfusion). Nevertheless, Dulbecco's phosphate-buffered saline in a total volume of the rise at 48 hours of reperfusion was significantly lower 1 mL. Activity was expressed as the initial velocity of in the experimental group compared with the control absorbance increase at 655 nm/min/mg of tissue protein. group (p <0.05). Graeilis muscle weight: The gracilis muscle weight One unit of enzyme activity was defined as the amount of MPO that produced an absorbance change of 1.00D/ increased significantly in both the control and experimental groups. In control muscles, it rose from a baseline min/mg of tissue protein during incubation at 37~ Protein content was measured with the Coomassie blue value of 94 4- 3 g to 130 4- 9 g (p <0.05). A similar trend method of Bradford [19]. This assay measures the was seen on the Huosol-treated side, where it increased change in absorbance at 595 nm and was done with a from 94 4- 4 g at baseline to 120 4- 6 g at the end of Gilford Response II UV/VIS spectrophotometer (CIBA reperfusion (p <0.05). However, the percentage inComing Diagnostics Corp., Gilford Systems, Oberlin, creases in the control and experimental groups were not significantly different (38.31% 4- 9.36% versus 28.34% 4OH). Muscle necrosis measurement: After removal, mus- 7.35%, p >0.05). cles were sectioned perpendicularly to their long axis into six equal slices. The slices were weighed and incubated in COMMENTS Despite all the treatment modalities available and a 0.5% nitroblue tetrazolium (NBT) solution for 20 minutes at 250C [20]. Areas of viable muscle were stained better understanding of the pathophysiologic mechablue, and necrotic areas remained unstained. The total nisms of reperfusion injury, acute skeletal muscle isarea at risk and the area of necrosis were determined on chemia continues to be a significant cause of mortality both sides of each slice by computerized planimetry, and and morbidity [1,2]. The optimal management to prevent the means of each were calculated [2I]. The ratio of reperfusion injury continues to be a challenge for surmean nonviable area to total area, in each slice, was geom. *Significantly higher than preischemia and 1-hour reperfusion in both groups oy analysis of variance (p < 0.05). TSignificantly lower than control side by paired t-test (p < 0.05).
196
THE AMERICAN JOURNAL OF SURGERY
VOLUME 164 SEPTEMBER 1992
PERFLUOROCARBONS IN SKELETALMUSCLEISCHEMIA
There are two major factors that seem to play a role in the pathogenesis of reperfusion syndrome: the no-reflow phenomenon, which causes the myocytes to progress from a reversibly injured state to an irreversible one, and direct cellular damage that follows the restoration of blood flow. Some of the pathogenetic factors of the no-reflow phenomenon include fibrin deposition and thrombus formation in the microcirculation, cellular edema occurring during ischemia, capillary collapse, and endothelial cell damage and dysfunction [3,22-24]. Alternatively, cellular injury is mediated by production of oxygen free radicals predominantly by the activated neutrophils [10-13]. Therefore, any attempt at mitigating ischemic injury should address both the effects of the no-reflow phenomenon and the injury related to reperfusion itself [3]. Impairment of the microcirculation after acute ischemia may be a determinant factor in the pathogenesis of infarction. Microcirculatory failure occurs secondarily to stasis and intravascular clotting due to increased blood viscosity [4,5] and capillary narrowing [6-9]. Blood viscosity relates directly to a concentration of plasma protein and of formed elements of the blood [25]. The concentration of these osmotically active substances is increased during ischemia by the shift of inorganic ions, namely, sodium, from the extracellular to the intracellular compartment [26,27]. Because ischemia paralyzes the sodium-potassium (Na +/K +) pump, the active transcellular transport of electrolytes is impaired [28]; therefore, water migrates from the intravascular to the intracellular compartment, causing increased blood viscosity and increased tissue pressure. The increased blood viscosity in conjunction with a spastic response of the microcirculation may be responsible for progressive and irreversible ischemia. Perfluorochemical emulsion (Fluosol-DA 20%), an acellular blood substitute, can exert a protective effect on the ischemic myocardium by improving myocardial blood flow and can actually reduce infarct size [29-32]. Some of the beneficial effects appear to be due to inhibition of neutrophil chemotaxis and lysosomal enzyme release [31]. Intracoronary perfusion of oxygenated perfluorochemical has also been shown to reduce the no-reflow injury in myocardial ischemia models, in addition to causing relative preservation of endothelial cell integrity [33]. Fluosol-DA 20% significantly decreased the percentage of gracilis muscle necrosis after 6 hours of normothermic ischemia followed by reperfusion. Although neutrophil infiltration is known to continue for 48 hours after ischemia [34,35], we found that a limited period of Fluosol-DA 20% infusion (40 minutes) in the ischemic muscles, before blood reperfusion, was effective in reducing the extent of muscle necrosis. Obviously, it would b e important to evaluate whether a prolonged infusion of perfluorocarbon could reduce the extent of muscle necrosis even further. Of interest, although oxygenated Fluosol-DA 20% increased the oxygen tension of the gracilis muscle to higher values (315 q- 50 mm Hg) than are obtained during reperfusion with blood, this did not affect
its capability to minimize muscle necrosis. Hence, one can infer that an hyperoxic environment, in the absence of cellular elements, does not play a pathogenetic role in muscle necrosis. Cytotoxic oxygen free radicals are the product of activated neutrophils exposed to oxygen during reperfusion. This study also confirms the predominant role played by activated neutrophils in the production of oxygen free radicals. Since there was no significant difference in the percentage weight gain in experimental and control muscles, we believe that perfluorocarbon infusion probably does not affect vascular permeability after ischemia-reperfusion. However, because weight gain is not as sensitive a methodology as measurement of radioactive-labeled albumin leakage, we cannot draw scientific conclusions regarding the effect of perfluorocarbon on vascular permeability. The beneficial effects of oxygenated perfluorocarbon emulsion infusion seem to be neutrophil mediated. Since the MPO activity at 48 hours of reperfusion was significantly lower than that of the control side (1.78 4- 0.6 U/ mg tissue protein versus 5.46 4- 1.52 U/mg tissue protein), it is possible that there is a reduction in neutrophil sequestration into the ischemic muscle due to their action of inhibiting neutrophil surface receptors (CD11/CD18 molecules). The surface receptors of neutrophils (CD11/CD18 molecules) are necessary for their adherence to the endothelium, which, in turn, appears to be a prerequisite before their migration into the ischemic skeletal muscle. Another aspect that requires investigation is the effect of perfluorocarbon emulsion on cytokine production after skeletal muscle ischemia and reperfusion. We have previously shown that there is increased production of interleukin-1 when skeletal muscle is subjected to prolonged ischemia followed by reperfusion [36]. Since interleukin-1 is known to be an important mediator that causes expression of CD 11/ CD 18 surface receptors of neutrophils, we believe that it is important to study the effect of perfluorocarbons on interleukin-1 release in ischemia-reperfusion. The inhibitory effect of perfluorocarbons on the surface receptors of neutrophils may be mediated indirectly via its effect on interleukin-1 production. Based on the results of this study, we conclude that intra-arterial infusion of oxygenated perfluorocarbon emulsion at the time of reperfusion appears to be a promising novel approach to decrease the extent of muscle necrosis after prolonged skeletal muscle ischemia followed by reperfusion. This approach is attractive because it is applicable to the clinical situation in which patients are first seen after the onset of ischemia and, therefore, cannot be pretreated. In the clinical arena, the only therapeutic modalities applicable to patients are those that can be instituted at the time of restoration of blood flow. REFERENCES 1. Haimovici H. Metabolic complications of acute arterial occlusions. J Cardiovasc Surg (Torino) 1979; 20: 349-57.
2. HaimoviciH. Arterial embolismwith acute massiveischemic myopathy and myoglobinuria. Surgery 1960; 116: 414-21.
THE AMERICAN JOURNAL OF SURGERY VOLUME 164 SEPTEMBER 1992 197
MOHAN ET AL
3. Quinones-Baldrich WJ. The role of fibrinolysis during reperfusion of isehemic skeletal muscle. Microeirc Endothelinm Lymphatics 1989; 5: 299-314. 4. Little JR. Implanted device for middle cerebral artery occlusion in conscious cats. Stroke 1977; 8: 258-60. 5. Boschenstein FK, Reilly JA, Yahr MD, Correil JW. Effect of low molecular weight dextran on cortical blood flow. Arch Neurol 1966; 14: 288-93. 6. Little JR. Modification of acute isehemia by treatment with mannitol. Stroke 1978; 9: 4-9. 7. Ames A III, Wright RL, Kowada M, Thurston JM, Majno G. Cerebral isohemia. The no-reflow phenomenon. Am J Pathol 1968; 52: 437-53. 8. Little JR. Microvascular alterations and edema in focal cerebral isohemia. In: Pappius H, Feindel W, editors. Dynamics of brain edema. New York: Springer-Verlag, 1976: 236-43. 9. Chiang J, Kowada M, Ames A III, Wright RL, Majno G. Cerebral isehemia III. Vascular changes. Am J Pathol 1968; 52: 455-76. 10. Rornson J, Hook BG, Kunkel SL, Abrams GD, Sehork MA, Luchessi BR. Reduction of the extent of ischemic myocardialinjury by ncutrophil depletion in the dog. Circulation 1983; 67:1016-23. 11. Engler R, Dahlgren M, Morris D, Pcterson M, Schmid-Schonbein G. Role of leukoeytes in response to acute myocardial isehemia and reflow in dogs. Am J Physiol 1986; 251: H314-22. 12. MeCord JM. Oxygen-derived free radicals in post-ischemic tissue injury. N Engl J Med 1985; 312: 159-63. 13. Noveili GP, Livi P, Ghinussi L, Lisi L, Brunelleschi S, Fantozzi R. Superoxide generation by granuloeytes during superior mesenteric artery occlusion shock in rabbits. In: Novelli GP, Ursini F, editors. Oxygen radicals in shock. International workshop. Basel: Karger, 1986: 587-93. 14. Simpson P J, Todd RF, Fantone JC, et al. Reduction of experimental canine myocardial reperfusion injury by a monoclonal antibody (anti-Mol, anti-CD 1lb) that inhibits leukocyte adhesion. J Clin Invest 1988; 81: 624-9. 15. Osterholm JL, Alderman JB, Triolo AJ, D'Amore BR, Williams HD, Frazer G. Severe cerebral isehemia treatment by ventriculosubarachnoid perfusion with an oxygenated fluorocarbon emulsion. Neurosurgery 1983; 13: 381-7. 16. Peerless SJ, Ishikawa R, Hunter IG, Peerless MJ. Protective effect of Fluosol-DA in acute cerebral isehemia. Stroke 1981; 12: 558-63. 17. Kuzon WM, Walker PM, Mickle DAG, Harris KA, Pynn BR, McKee NH. An isolated skeletal muscle model suitable for acute ischemia studies. J Surg Res 1986; 41: 24-32. 18. Suzuki K, Ota H, Sasagawa S, Sakatani T, Fujikura T. Assay method for myeloperoxidase in human polymorphonuclear leukocytes. Anal Biochem 1982; 132: 345-52. 19. Bradford M. A rapid and sensitive method for the quantitation
198
of microgram quantities of protein utilizing the principle of proteindye binding. Anal Biochem 1976; 72: 248-54. 20. Novikoff AB, Shin W, Drucker JJ. Staining results with two tetrazolium salts. J Biophys Biochem Cytol 1961; 9: 47-61. 21. Labbe R, Lindsay T, Walker PM. The extent and distribution of skeletal muscle necrosis after graded periods of complete ischemia. J Vase Surg 1987; 6: 152-7. 22. Chait LA, May JW Jr, O'Brian BM, Hurley JV. The effects of the perfusion of various solutions on the no reflow phenomenon in experimental free flaps. Hast Reconstr Surg 1978; 61: 421-30. 23. Leaf A. Cell swelling. A factor in isehemic tissue injury. Circulation 1973; 48: 455-8. 24. Stroek PE, Majno G. Mierovaseular changes in experimental tourniquet isehemia. Surg Gynecol Obstet 1969; 129: 1213-23. 25. Wells RE. Rheology of blood in the microvaseulature. N Engl J Med 1964; 270: 832-9. 26. Edstrom RFS, Essex HE. Swelling of the brain induced by anoxia. Neurology (Minneapolis) 1956; 6:118-24. 2 7 . Van Harreveld A. Changes in volume of cortical neuronal elements during asphyxiation. Am J Physiol 1957; 191: 233-42. 28. Leaf A. On the mechanism of fluid exchange of tissues in vitro. Biochem J 1957; 62: 241-8. 29. Forman MB, Bingham SE, Kopelman HA, et al. Reduction of infarct size with intracoronary perfluorochemical in a canine preparation of reperfusion. Circulation 1985; 71: 1060-8. 30. Forman MB, Puett DW, Bingham SE, et al. Preservation of endothelial cell structure and function by intracoronary perfluorochemical in a canine preparation of reperfusion. Circulation 1987; 76: 469-79. 31. Bajaj AK, Cobb MA, Virmani R, Gay JC, Light RT, Forman MB. Limitation of myocardial reperfusion injury by intravenous perfluorochemicals; role of ncutrophil activation. Circulation 1989; 79: 645-56. 32. Forman MB, Puett DW, Wilson BH, et al. Beneficial longterm effect of intracoronary perfluorochemieal on infarct size and ventricular function in a canine reperfusion model. J Am Coil Cardiol 1987; 9: 1082-90. 33. Kolodgie FD, Virmani R, Farb A. Limitation of no reflow injury by blood-free reperfusion with oxygenated perfluoroehemical (Fluosol-DA 20%). J Am Coil Cardiol 1991; 18: 215-23. 34. Rubin BB, Smith A, Liauw S, Isenman D, Romasehin AD, Walker PM. Complement activation and white cell sequestration in postisehemic skeletal muscle. Am J Physiol 1990; 259: H525-31. 35. Mohan C, Ascer E, Marini CP, Maughan R, Gennaro M. Does iloprost mediate thromboxane release and polymorphonuclear leukocyte sequestration in ischemic skeletal muscle? J Cardiovase Surg 1992 (in press). 36. Aseer E, Mohan C, Gennaro M, Cupo S. Interleukin-1 and thromboxane release after skeletal muscle ischemia and reperfusion. Ann Vase Surg 1992; 6: 69-73.
THE AMERICAN JOURNAL OF SURGERY VOLUME 164 SEPTEMBER 1992