Pentoxifylline Reduces Coronary Leukocyte Accumulation Early in Reperfusion After Cold Ischemia

Transcription

1 Pentoxifylline Reduces Coronary Leukocyte Accumulation Early in Reperfusion After Cold Ischemia Stephen C. Gale, MD, Jason Y. Hokama, PhD, Leslie S. Ritter, PhD, Grace D. Gorman, BS, Jack G. Copeland, MD, and Paul F. McDonagh, PhD Section of Cardiovascular and Thoracic Surgery and the Sarver Heart Center, University of Arizona Health Sciences Center, Tucson, Arizona Background. Ischemia/reperfusion injury can complicate recovery in cardiac operations. Ischemia induces endothelial dysfunction, which may contribute to leukocyte accumulation during reperfusion. Leukocytemediated injury may then occur. Using intravital microscopy we previously reported increased leukocyte retention in coronary capillaries and venules during early reperfusion during warm ischemia/reperfusion. In this study we investigated whether cold cardioplegic protection would limit leukocyte sequestration in coronary microvessels early in reperfusion. Pentoxifylline (PTX) has antiinflammatory effects and may limit endothelial dysfunction during ischemia/reperfusion. The effect of cardioplegia modification with PTX was also examined. Methods. Isolated rat hearts were subjected to 90 minutes of 4 C ischemia after arrest with cardioplegia. Hearts were reperfused with diluted whole blood containing fluorescent-labeled leukocytes. Leukocyte retention in coronary microvessels was observed with intravital microscopy. Three groups were studied, nonischemic control, cold ischemia, and PTX-modified cold ischemia. Results. In coronary capillaries, leukocyte trapping was nearly doubled in unmodified cold ischemia versus control. PTX modification significantly reduced leukocyte accumulation. In coronary venules, greater leukocyte adhesion was observed in unmodified cold ischemia compared to nonischemic controls. PTX modification significantly reduced leukocyte adhesion. Conclusions. Cold cardioplegia did not prevent leukocyte retention in the coronary microcirculation early in reperfusion. PTX modification of cardioplegia significantly reduced leukocyte sequestration in coronary capillaries and venules. Preserving endothelial function during ischemia may limit leukocyte accumulation and ischemia/reperfusion injury after cardiac operation. (Ann Thorac Surg 2001;71: ) 2001 by The Society of Thoracic Surgeons The pathogenesis of ischemia-reperfusion (I/R) injury is multifactorial and both leukocyte-dependent [1] and leukocyte-independent [2] mechanisms are operative. During reperfusion leukocytes, particularly phagocytes, accumulate in tissues and become activated causing tissue damage through the release of inflammatory mediators [1]. Evidence suggests that changes occur in the endothelium and microvasculature [3] during ischemia that lead to early leukocyte accumulation during reperfusion, which may be responsible for an early leukocyte-mediated reperfusion injury [4]. In I/R, leukocyte accumulation is usually described as a later event involving adhesion molecule synthesis and occurring mostly in postcapillary venules [5]. However, an early rapid event also occurs in the heart. Studies by Pearl [6, 7] and Galinanes [8] and their colleagues point to an early leukocyte-mediated injury in the first minutes of reperfusion. Furthermore, early leukocyte accumulation Accepted for publication Nov 21, Address reprint requests to Dr McDonagh, Department of Cardiothoracic Surgery, University of Arizona Health Sciences Center, PO Box , 1501 N Campbell, Tucson, Arizona 85724; pmcdonag@ u.arizona.edu. in coronary capillaries during I/R is also well described [9 11]. Despite considerable investigative effort, many details of when, where, and under what conditions leukocytes accumulate in the coronary microcirculation during reperfusion remain unclear. Understanding the timing and location of leukocyte accumulation during reperfusion is likely an important key to finding interventions aimed at limiting the phagocyte contribution to I/R injury. To investigate the timing and location of leukocyte accumulation during myocardial I/R, we developed an intravital fluorescence microscopy model to visualize directly the coronary microcirculation during reperfusion. In previously studies of normothermic ischemia, we reported significant leukocyte accumulation early in reperfusion, particularly in coronary capillaries [9, 10]. These findings led us to consider whether the early leukocyte accumulation we observed might be prevented by cardiac preservation during ischemia with hypothermia and cardioplegia. We report that cardiac preservation during ischemia with cardioplegia and hypothermia did not prevent early leukocyte accumulation in the first minutes of reperfusion. Despite preservation, leukocyte 2001 by The Society of Thoracic Surgeons /01/$20.00 Published by Elsevier Science Inc PII S (01)

2 1306 GALE ET AL Ann Thorac Surg PTX REDUCES CORONARY LEUKOCYTE 2001;71: accumulation was increased in both coronary capillaries and venules compared to nonischemic controls. Furthermore, by modifying the Plegisol preservation solution with pentoxifylline (PTX), a methyl xanthine phosphodiesterase inhibitor, we observed significant attenuation of leukocyte accumulation during reperfusion in both capillaries and venules to the level of nonischemic controls. Material and Methods Rat Heart Isolation All experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication 85-33, revised 1985). The isolated rat heart preparation used was described previously [9, 10]. Briefly, male Sprague-Dawley rats weighing 450 to 525 g were anesthetized with sodium pentobarbital (50 mg/kg). Tracheostomy and intubation were performed and the animal was respirated at 55 breaths/min (model 683, Harvard Apparatus, Hollister, MA). A sternotomy was performed to expose the heart. The great vessels were isolated and controlled with loose silk ligatures. After heparinization (150 U), the carotid artery was ligated distally and a catheter (No. 20 Jelco) was inserted into the proximal carotid and advanced into the ascending aortic arch for perfusate injection. The catheter was secured with silk sutures. The distal aortic arch was ligated and the right atrium was cut to provide drainage as 20 ml of cold Plegisol (Abbott Laboratories, Abbott Park, IL) cardioplegia was infused into the coronary vessels. A perfusion pressure of 80 mm Hg was maintained. Ice chips were used to maintain hypothermia. After complete arrest the heart was carefully removed. The hearts in the nonischemic control group were immediately placed on the heated (37 C) Lucite stage for intravital microscopy. The hearts in the cold ischemic groups were wrapped in phosphate-buffered saline-soaked gauze and Saran- Wrap and refrigerated at 4 C for 90 minutes. After the ischemic period hearts in the ischemic groups were placed on the heated Lucite stage for microscopy. Preparation of Coronary Perfusate: Labeled Diluted Whole Blood The procedure for labeling the leukocytes in diluted whole blood was described previously [9, 10]. Briefly, approximately 16 ml of freshly drawn heparinized blood was obtained by cardiac puncture from each of two anesthetized donor rats (Sprague-Dawley, 450 to 550 g). Specimens with evidence of clot were discarded. The blood was centrifuged at 546 g for 15 minutes. Red cells and plasma were collected, combined, and stored together on ice. The remaining white cell pellet was resuspended in 2.5 ml of filtered acridine orange (0.01 mg of acridine orange/ml of phosphate-buffered saline) and incubated at room temperature for 15 minutes. The fluorochrome acridine orange labels nucleated cells. This concentration of acridine orange does not affect white cell function [9, 10]. The labeled white blood cell fraction was washed twice with an albumin/pbs solution to remove unbound fluorochrome. The blood was reconstituted with the plasma and red cells and diluted 1:1 with modified Krebs solution. This solution was referred to as labeled, diluted whole blood (DWB*). Blood counts were obtained to assure a 1:1 dilution (hematocrit, 18% to 21%). Blood gas analysis of the DWB* was performed after dilution. The blood was oxygenated and buffered with 8.4% NaHCO 3 to obtain a ph of 7.35 to 7.45, partial pressure of carbon dioxide 35 to 45 mm Hg, partial pressure of oxygen greater than 60 mm Hg, and arterial oxygen percent oxygenation more than 90%. Preparation of PTX-Modified Cardioplegia For the pentoxifylline treatment group, PTX (Sigma Chemical Company, St. Louis, MO) was added to Plegisol and dissolved to achieve a 5 mmol/l PTX solution. The modified cardioplegia solution was filtered to remove any debris. This solution was used to arrest hearts in the PTX group exactly as performed in the unmodified Plegisol group. Cardiac Reperfusion and Data Collection During reperfusion hearts were placed on the microscope stage and oriented such that the great coronary vein was facing upward. Hearts were reperfused for 16 to 20 minutes with DWB* at a constant rate of 3 ml/min through a syringe pump (model 11, Harvard Apparatus). The epicardial microcirculation was illuminated and viewed with an epifluorescence microscope (Zeiss). The microscope image was viewed continuously with a lowlight-level silicon-intensified target video camera (Hamamatsu C2400) and was displayed on a video monitor (Panasonic, WV5410). The microscopic field was 350 by 270 m, with a specimen to monitor magnification of 780. Images were recorded on a half-inch SuperVHS videotape recorder (Mitsubishi, U82). During reperfusion, four to six coronary capillary fields and four to six coronary venules (20 to 100 m in diameter) were selected at random, brought into focus, and recorded for at least 30 seconds. To measure capillary leukocyte accumulation on video playback, fluorescent leukocytes in the plane of focus, which remained stationary in a capillary for at least 30 seconds, were considered trapped [9, 10]. A capillary field was defined as a full video field containing only capillaries and devoid of larger vessels. Leukocytes were counted and expressed as leukocytes per capillary field. To measure venular leukocyte adhesion, leukocytes in the plane of focus on the top and bottom margins of the vessel, which remained stationary for at least 30 seconds, were counted as adhered [9, 10]. Unbranched venule segments were selected. The length and width of venule segments were measured and recorded. Data was normalized and expressed as adhered leukocytes/100 m venule length [9, 10]. Experimental Protocol Three groups were examined in this study. For all groups, preparation of the perfusate (DWB*) and the rat heart

3 Ann Thorac Surg GALE ET AL 2001;71: PTX REDUCES CORONARY LEUKOCYTE 1307 Table 1. Study Group Demographics Variable Group I Control Group II Cold Ischemia Group III Cold Ischemia PTX Treated Rat wt. (gr) Heart wt. (gr) Perfusate volume (ml) Total WBC delivered (millions of WBC) Ischemia time (min) No statistical difference between groups for parameters. WBC white blood cells. isolation were performed identically. There were six isolated rat heart preparations per group. Group I was a control nonischemic group. After isolation, rat hearts were perfused immediately and data was collected. Group II was a cold ischemic group for which hearts were protected with Plegisol at 4 C for 90 minutes before reperfusion. Group III was a cold ischemic PTX group. These hearts received modified cardioplegia (Plegisol with 5 mmol/l PTX). Statistical Analysis Data was tabulated on computer spreadsheets (Microsoft Excel 97; Microsoft, Redmond, WA). Summary data was expressed as means standard error of the mean. Comparison among groups was performed by analysis of variance. A post hoc test was performed when appropriate using SPSS statistical software (SPSS, Chicago, IL). A probability value of less than 0.05 was considered statistically significant. Results Table 1 summarizes body weight, heart weight, volume of perfusate delivered, number of white blood cells delivered, and ischemia time for the three groups. These parameters were similar for all three groups. Leukocyte Accumulation in Coronary Capillaries Leukocytes were observed to trap and plug in coronary capillaries early in reperfusion despite cardioplegic protection during ischemia. Figure 1 provides representative images of capillary fields (left) and venules (right) in a nonischemic control, cold ischemia, and cold ischemia/ PTX heart during reperfusion with DWB*. Few leukocytes were retained in the capillaries of nonischemic control hearts (Fig 1A). In contrast, a significantly greater leukocyte accumulation occurred in the capillaries of hearts subjected to 90 minutes of cold ischemia before reperfusion (Fig 1C). Leukocyte retention was significantly reduced (Fig 1E) in hearts receiving PTX-modified cardioplegia before cold ischemia. Figure 2 summarizes the measurements of leukocyte accumulation in coronary capillaries. Compared to nonischemic control hearts, 90 minutes of cold ischemia led to a significant increase in capillary leukocyte accumulation early during reperfusion (group I: versus group II: leukocytes/capillary field, p 0.05). Adding PTX to the cardioplegia solution decreased leukocyte accumulation to the level observed in nonischemic control hearts (group II: versus group III: leukocytes/capillary field, p 0.05). Leukocyte Accumulation in Coronary Venules Leukocyte adhesion to the walls of postcapillary venules was also observed early in reperfusion after cold ischemia. The images on the right in Figure 1 illustrate leukocyte adhesion to postcapillary venules during DWB* reperfusion. Leukocytes were clearly seen flowing through coronary venules during reperfusion. Adherent leukocytes were observed as either distinct rounded cells marginated on the vessel or as flattened cells adhered to the venular wall. Few adherent leukocytes were observed in venules of nonischemic control hearts (Fig 1B). A significant increase in leukocyte adhesion was observed in the venules of hearts subjected to 90 minutes of cold ischemia (Fig 1D). The use of PTX-modified cardioplegia significantly reduced leukocyte adhesion to coronary venules (Fig 1F). Figure 3 summarizes the measurements of leukocyte adhesion to postcapillary venules. Compared to nonischemic controls, 90 minutes of cold ischemia significantly increased leukocyte adhesion (group I: versus group II: leukocytes/100 m venule, p 0.05). Preservation with the PTX-modified cardioplegia solution significantly reduced the number of adherent leukocytes observed in coronary venules during reperfusion (group II: versus group III: leukocytes/100 m venule, p 0.05). Comment This study directly visualizes coronary microvascular perfusion in the early moments of reperfusion after cold ischemia. Previously, using intravital microscopy, we reported that after warm ischemia, leukocyte retention in the coronary microcirculation, particularly in capillaries, was elevated early in reperfusion [9]. We postulated that because ischemic damage, including microvascular damage [12], is limited by hypothermia and by cardioplegia protection, significant elevations in leukocyte retention in the coronary microcirculation would not be observed early in the reperfusion of cold cardioplegia-protected

4 1308 GALE ET AL Ann Thorac Surg PTX REDUCES CORONARY LEUKOCYTE 2001;71: Fig 1. Representative images during reperfusion. (A) Nonischemic capillary field with no leukocyte trapping. (B) Nonischemic coronary venule without adherent leukocytes. (C) Capillary field after 90 minutes of cold ischemia with many trapped leukocytes (arrows) and occluded capillaries. (D) Coronary venule after 90 minutes of cold ischemia lined with adherent leukocytes (arrows). (E) Capillary field after 90 minutes of pentoxifylline exposure during cold ischemia with occasional leukocytes but patent capillaries. (F) Coronary venule after 90 minutes of pentoxifylline exposure during cold ischemia with occasional adherent leukocytes. hearts. In the current study we demonstrated that leukocyte accumulation does occur in the coronary microcirculation very early in reperfusion despite protection with cardioplegia and hypothermia. Significantly increased leukocyte accumulation was observed both in coronary capillaries and postcapillary venules compared to nonischemic controls. As the leukocyte-containing reperfusate was identical in all groups, these findings provide

5 Ann Thorac Surg GALE ET AL 2001;71: PTX REDUCES CORONARY LEUKOCYTE 1309 Fig 2. Leukocyte accumulation in coronary capillaries. Summary results are given as the mean number of leukocytes per capillary field plus standard error of the mean (n 6 for each group); p Cold cardioplegia did not prevent leukocyte accumulation in capillaries early in reperfusion. Pentoxifylline (PTX) modified cardioplegia significantly attenuated leukocyte trapping in coronary capillaries. evidence that changes occur in the heart during cold cardioplegia-protected ischemia, which result in rapid leukocyte recruitment and stasis early during reperfusion. In addition, we observed that modifying the cardioplegia solution with PTX prevented the increased leukocyte accumulation seen in the coronary capillaries and Fig 3. Leukocyte adhesion in coronary venules. Summary results are given as the mean number of leukocytes per 100- m venule length plus standard error of the mean (n 6 for each group); p Cold cardioplegia did not prevent leukocyte adhesion to coronary venules early in reperfusion. Pentoxifylline (PTX) modified cardioplegia significantly attenuated leukocyte adhesion to coronary venules. venules during reperfusion. Having previously documented that treatment of the blood with PTX prevents leukocyte accumulation during reperfusion [10], we now demonstrate that treatment of the heart, by PTX enhancement of cardiac preservation, is also preventative, although likely by different mechanisms. Although studies suggest that leukocyte-mediated injury is not immediate during reperfusion, other researchers demonstrate that early leukocyte sequestration is an important contributor to a rapid reperfusion injury after ischemia. Early leukocyte-mediated depression of ventricular function [13] and loss of endothelial-dependent vasodilation [14] have been reported. In a rat cardiac transplantation model, Galinanes and colleagues [8] reported improved ventricular functional recovery with leukocyte depletion during early reperfusion. In addition, Pearl and associates [6, 7] demonstrated in human cardiac transplant patients that leukocyte depletion during the initial 10 minutes of reperfusion decreased ultrastructural and biochemical evidence of reperfusion injury. These studies, in cardioplegia-preserved organs, indicate that leukocyte-mediated injury occurs very early in reperfusion with structural and functional consequences and suggest that inhibiting early leukocyte accumulation may be beneficial in preventing I/R injury after cold ischemia. This study documents, by direct observation, the occurrence of early leukocyte accumulation and demonstrates its inhibition with PTXenhanced cardiac preservation. Leukocyte Accumulation in Capillaries After Unmodified Cold Ischemia For the cold cardioplegia group, the leukocyte accumulation in coronary capillaries was nearly double that observed in nonischemic controls. Because the perfusion rate (3 ml/min) and perfusate composition were the same for all groups, these findings were likely from microvascular changes that occurred during cold ischemia. The physical trapping of leukocytes in coronary capillaries during reperfusion after normothermic ischemia has been described histologically [11, 15] and by direct visualization with intravital microscopy [9, 10]. Leukocyte trapping in capillaries completely obstructs flow and likely contributes to the no reflow phenomenon [16, 17]. Despite cardioplegic protection, damage to the coronary microvasculature and endothelial dysfunction occur during ischemia [3, 18, 19] leading to endothelial cell swelling [18] and the development of interstitial edema [17, 18]. The accumulation of both intracellular and extracellular fluid may reduce the diameter of the coronary capillary lumen. A large discrepancy exists between phagocyte diameter (12 m) and coronary capillary lumen diameter (4.5 m) requiring significant white cell deformation for passage [16, 20]. Further narrowing in the capillary lumen diameter may lead to mechanical trapping of leukocytes in capillaries [17, 21].

6 1310 GALE ET AL Ann Thorac Surg PTX REDUCES CORONARY LEUKOCYTE 2001;71: Effect of PTX-Modified Cardioplegia on Leukocyte Trapping in Capillaries In preserving hearts with PTX-modified cardioplegia, we observed a significant reduction in leukocyte capillary trapping during reperfusion to the level of nonischemic control hearts. PTX has several effects on endothelial cells that may reduce microvascular dysfunction, which may be operative. As a phosphodiesterase inhibitor, PTX maintains intracellular cyclic camp (camp) [22]. Studies by Pinsky [23] and Yan and their colleagues [24] reported maintained endothelial cell barrier function during cardiac preservation through the maintenance of intracellular camp with phosphodiesterase inhibitors or through the use of camp analogs. PTX is also an effective hydroxyl radical scavenger [25] and may reduce endothelial cell damage indirectly through that mechanism. Limiting endothelial cell damage and swelling would attenuate tissue edema preventing leukocyte trapping. Leukocyte Adhesion to Postcapillary Venules After Unmodified Cold Ischemia The mechanisms underlying the leukocyte adhesion to venules are likely different from those responsible for leukocyte trapping in capillaries. Endothelial cell swelling and extravascular compression are not likely to cause leukocyte retention in these larger microvessels. In addition to the firmly adherent leukocytes observed during reperfusion, leukocytes were also directly observed rolling on the endothelial surface before rejoining the bloodstream. This leukocyte rolling and firm adhesion to coronary venules likely represents upregulation of endothelial adhesion molecules. The observation of leukocyte rolling strongly supports the assertion by Pinsky and colleagues [4] that, early in reperfusion, P- selectin is already expressed on endothelial cell surfaces secondary changes occurring during cold ischemia. In addition, Shreeniwas and associates [26] reported, in cultured endothelial cells, that ischemia alone may upregulate intercellular adhesion molecule-1 (ICAM-I) mrna, before reperfusion through an autocrine mechanism. These reports suggest that ischemia alone is a sufficient stimulus to activate endothelial cells and prime them for leukocyte adhesion early in reperfusion. Our data further support this hypothesis and document its occurrence after cardioplegia-protected ischemia. Effect of PTX-Modified Cardioplegia on Leukocyte Adhesion to Postcapillary Venules Pentoxifylline-modified cardioplegia reduced leukocyte adhesion to postcapillary venules during reperfusion to the level of nonischemic control hearts. As a phosphodiesterase inhibitor, PTX may inhibit leukocyte adhesion to coronary venules in this setting. The decrease in camp that occurs during ischemia results from increased phosphodiesterase activity causing camp breakdown, not because of decreased camp production. Indeed, Franzini and colleagues [27] concluded that this mechanism of camp depletion is responsible for oxygen radicalmediated leukocyte adhesion to stimulated endothelial cells and that adhesion is inhibited by PTX. In addition, PTX is an effective hydroxyl radical scavenger [25], potentially preventing endothelial damage by reactive oxygen species. Through these mechanisms, PTX may decrease adhesion molecule expression limiting early leukocyte adhesion coronary venules during reperfusion. In conclusion, in the present study we directly observed that (1) leukocytes accumulate in the coronary microcirculation during reperfusion despite cold cardioplegia protection during ischemia; (2) after cold ischemia, leukocyte accumulation is immediate with cells adhering to coronary venules and trapping in coronary capillaries; and (3) adding PTX to cardioplegia prevents leukocyte accumulation in coronary microvessels early in reperfusion after cold ischemia. These findings support studies by other investigators demonstrating vascular endothelial damage and dysfunction early in reperfusion despite cold cardioplegiaprotected ischemia [3, 18, 19]. Despite the protection provided to the myocardium, the endothelium remains vulnerable to ischemia. The use of preservation solutions that limit endothelial damage during ischemia, by providing additional substrates or preventing substrate breakdown, may attenuate leukocyte accumulation in coronary microvessels during reperfusion and limit the leukocyte component of I/R injury. Finally, this intravital microscopy model of cold I/R may be useful in evaluating preservation solutions, pharmacologic agents, and substrates to investigate their ability to limit leukocyte accumulation during early reperfusion after cardioplegic protection or longer term organ preservation. This study was supported by NIH HLB Dr Gale was supported by a Surgery Resident Fellowship from the Department of Surgery, AHSC. References 1. Siminiak T, Ozawa T. Neutrophil mediated myocardial injury. Int J Biochem 1993;25: Boyle E, Pohlman T, Cornejo C, Verrier E. Endothelial cell injury in cardiovascular surgery: ischemia reperfusion. Ann Thor Surg 1996;62: Pathi V, McPhaden A, Morrison J. The effects of cardioplegic arrest and reperfusion on the microvasculature of the heart. Eur J Cardiothor Surg 1997;11: Pinsky DJ, Naka Y, Liao H, et al. Hypoxia-induced exocytosis of endothelial cell Weibel-Palade bodies: a mechanism for rapid neutrophil recruitment after cardiac preservation. J Clin Invest 1996;97: Jordan J, Zhao Z, Vinten-Johansen J. The role of neutrophils in myocardial ischemia reperfusion injury. Cardiovasc Res 1999;43: Pearl JM, Drinkwater DC, Laks H, Stein DG, Capouya ER, Bhuta S. Leukocyte-depleted reperfusion of transplanted human hearts prevents ultrastructural evidence of reperfusion injury. J Surg Res 1992;52: Pearl JM, Drinkwater DC, Laks H, Capouya ER, Gates RN. Leukocyte-depleted reperfusion of transplanted human hearts: a randomized, double-blind clinical trial. J Heart Lung Transplant 1992;11: Galinanes M, Lawson CS, Ferrari R, Limb GA, Derias NW, Hearse DJ. Early and late effects of leukopenic reperfusion on the recovery of cardiac contractile function: studies in the

7 Ann Thorac Surg GALE ET AL 2001;71: PTX REDUCES CORONARY LEUKOCYTE 1311 transplanted and isolated blood-perfused rat heart. Circulation 1993;88: Ritter LS, Wilson DS, Williams SK, Copeland JG, McDonagh PF. Early in reperfusion following myocardial ischemia, leukocyte activation is necessary for venular adhesion but not capillary retention. Microcirculation 1995;2: Ritter LS, Wilson DS, Copeland JG, Williams SK, McDonagh PF. Pentoxifylline reduces leukocyte retention in the coronary microcirculation early in reperfusion following ischemia. Int J Microcirc 1996;16: Engler RL, Schmid-Schonbein GW, Pavelec RS. Leukocyte capillary plugging in myocardial ischemia and reperfusion in the dog. Am J Pathol 1983;111: McDonagh PF, Laks H. Use of cold blood cardioplegia to protect against coronary microcirculatory injury to ischemia and reperfusion. J Thorac Cardiovasc Surg 1982;84: McDonagh PF, Reynolds JM. Brief perfusion with diluted whole blood after global myocardial ischaemia increases reperfusion injury. Cardiovasc Res 1994;28: Tsao PS, Ma X, Lefer AM. Activated neutrophils aggravate endothelial dysfunction after reperfusion of the ischemic feline myocardium. Am Heart J 1992;123: Hale SL, Kloner RA. Time course of infiltration and distribution of neutrophils following coronary artery reperfusion in the rat. Coronary Artery Dis 1991;2: Schmid-Schönbein GW. The damaging potential of leukocyte activation in the microcirculation. Angiology J Vasc Dis 1993;44: Manciet LH, Poole DC, McDonagh PF, Copeland JG, Mathieu-Costello O. Microvascular compression during myocardial ischemia: mechanistic basis for no-reflow phenomenon. Am J Physiol 1994;266:H Lindal S, Sorlie D, Jorgensen L. Endothelial cells of the cardiac microvasculature during and after cold cardioplegic ischemia. Comparison of endothelial and myocyte damage. Scand J Thorac Cardiovasc Surg 1988;22: Mattila S, Harjula A, Mattila I, Myllarniemi H, Harkonen M. Coronary endothelium and cardioplegic solutions. Ann Chir Gyn 1987;76: Chien S. Effects of inflammatory agents on leukocyte rheology and microcirculation. Prog Appl Microcirc 1987;12: Forman M, Puett D, Virimani R. Endothelial and myocardial injury during ischemia and reperfusion: pathogenesis and therapeutic implications. J Am Col Card 1989;13: Beavo J, Reifsnyder D. Primary sequence of cyclic nucleotide phosphodiesterase enzyme isozymes and the design of selective inhibitors. Trends Pharm Sci 1990;11: Pinsky D, Oz M, Liao H, et al. Restoration of the camp second messenger pathway enhances cardiac preservation for transplantation in a heterotopic rat model. J Clin Invest 1993;92: Yan SF, Ogawa S, Stern DM, Pinsky DJ. Hypoxia-induced modulation of endothelial cell properties: regulation of barrier function and expression of interleukin-6. Kidney Intern 1997;51: Pasquier C, Franzini EAZ, Kaouadji M, Hakim J. Gamma pulse radiolysis study of pentoxifylline, a methylxanthine. Int J Rad Biol 1991;60: Shreeniwas R, Koga S, Karakurum M, Pinsky D, Kaiser E, et al. Hypoxia-mediated induction of endothelial cell interleukin-1a. J Clin Invest 1992;90: Franzini E, Sellak H, Babin-Chevaye C, Hakim J, Pasquier C. Effects of pentoxifylline on the adherence of polymorphonuclear neutrophils to oxidant-stimulated human endothelial cells: involvement of camp. J Cardiovasc Pharm 1995;(Suppl 2):S92 5.