The Effect of Acute Coronary Artery Occlusion during Cardioplegic Arrest

Transcription

1 The Effect of Acute Coronary Artery Occlusion during Cardioplegic Arrest and Reperfusion on Myocardial Preservation John H. Rousou, M.D., Richard M. Engelman, M.D., William A. Dobbs, Ph.D., and Mooideen K. Meeran, M.D., Ph.D. ABSTRACT A study was undertaken to evaluate the effect of acute occlusion of a coronary artery during cardioplegic arrest on myocardial preservation and to elucidate the influence of reestablishment of flow versus continued occlusion during the phase of myocardial reperfusion. Coronary occlusion was simulated, and myocardial viability was determined by measuring tissue levels of adenosine triphosphate (ATP) and creatine phosphate (CP) in biopsies of the posterior left ventricular wall. Eighteen pigs were divided into three equal groups consisting of animals with (1) patent right coronary arteries during arrest, (2) occluded right coronary arteries during arrest and patent during reperfusion, and (3) occluded right coronary arteries during arrest. The results of ATP and CP measurements showed that while poorer protection was afforded during two-hour arrest when the coronary artery was occluded, the risk of damage was much greater during reperfusion. Failure to restore adequate blood flow by retention of occlusion caused a concurrent decrease in ATP and CP levels below prescribed limits of myocardial tolerance. When occlusion occurs in the clinical setting, impeding cardioplegia, the importance of revascularization is emphasized. Hypothermic potassium cardioplegia, when properly administered, safely protects the myocardium for up to two hours of ischemic arrest [l]. Potassium depolarization in conjunction with applied hypothermia minimizes degradation of high-energy phosphate compounds by sharply reducing aerobic glycolysis From the Departments of Surgery, University of Connecticut Health Center, Farmington, CT, and the Baystate Medical Center, Springfield, MA. Accepted for publication June 24, Address reprint requests to Dr. Rousou, Baystate Medical Center, 759 Chestnut St, Springfield, MA and decreasing the myocardial demand for oxygen [2, 31. Although sustained arrest are well tolerated by the myocardium under optimal conditions, the effect of acute coronary occlusion on preservation has not been well characterized. The pig heart was chosen as a model experimental system because of its lack of collateral circulation and its similarities to the human heart in response to acute ischemia [4, 51. A standard protocol was used to administer cold potassium cardioplegia during two hours of ischemic arrest [l]. A ligature placed near the origin of the right coronary artery was used to simulate the effect of an acute occlusion. The effect of vessel occlusion on high-energy phosphate levels during cardioplegic arrest was measured, and the results were compared with those from control animals with patent coronary arteries. This provides a comparable situation to that of cardioplegic arrest in the clinical setting when cardioplegia is not administered adequately into a coronary artery during arrest or when a distal coronary artery is inadvertently occluded during coronary sevascularization. Material and Methods Eighteen pigs weighing 17 to 24 kg underwent cardiopulmonary bypass using a perfusion circuit previously described [l]. In this preparation, the heart was isolated in its own perfusion circuit in situ, without perfusion of the rest of the animal (heart immersed in cold saline solution at 4 C). Following control studies in a normal beating heart, the animals underwent two hours of hypothermic cardioplegic arrest consisting of topical hypothermia and injection of a 50 ml bolus of cold (4 C) potassium cardioplegic solution to reach a myocardial temperature of 10" to 15 C. The composition of the solution was by The Society of Thoracic Surgeons

2 386 The Annals of Thoracic Surgery Vol 33 No 4 April 1982 as follows: 6% hydroxyethyl starch, 1 gmll of dextrose, 87 meqll of sodium chloride, 25 meq/l of sodium bicarbonate, and 35 meq/l of potassium chloride. Intramyocardial temperature was measured continuously in the septum but not in other areas of the myocardium. The osmolality of the solution was 350 mosm and the ph, 7.4 to 7.8. Fifty milliliters of the solution was administered into the aortic root, at 75 torr pressure, immediately following cessation of perfusion and every 15 minutes during arrest. During 30 minutes of reperfusion, myocardial temperature was promptly brought to 37 C by placing the heart in warm saline solution and using a normothermic perfusate. During this period the heart was defibrillated and a calibrated roller pump was used to maintain perfusion pressure at 75 torr. The animals were divided into three groups of 6 animals each. In control animals, patent coronary arteries permitted uninterrupted flow of perfusate during and after arrest. In the two experimental groups, the right coronary artery was occluded on the start of aortic crossclamping but prior to the administration of cardioplegia, by means of a ligature placed near the origin of the vessel in order to mimic the effect of acute coronary occlusion. The two experimental groups differed in that the ligature was maintained in one group during the arrest interval only and then was removed during reperfusion, while in the other group the vessel remained occluded throughout arrest. Metabolic Studies Each animal had biopsy samples 6 mm in diameter taken from the posterior wall of the left ventricle prior to arrest, at one hour and two hours of arrest, and after 30 minutes of reperfusion. Samples were removed by means of a modified dental drill. The tissue was immediately cooled in liquid nitrogen. No attempt was made to close the biopsy sites. The frozen tissue-biopsy was weighed on a torsion balance at 20 C and then pulverized in a cooled mortar with frozen 0.6 N perchloric acid. The resultant powder was thawed in an iced acetone bath at O'C, and a perchloric acid extract was obtained by centrifugation. The extract was neutralized and brought to 10 ml. Adenosine triphosphate (ATP) and creatine phosphate (CP) levels were determined by the colorimetric method of Lamprecht and Trautschold [6]. Statistical Methods Statistical analysis was performed using the two-sample Student's t test. Significance of differences was established when the p value was less than or equal to Results are presented for each treatment group as the mean plus or minus the standard error of the mean. Results Adenosine Triphosphate The levels of ATP measured in myocardial tissue of the three animal groups are listed in the Table and are represented graphically in Figure 1. They were found to be directly related to the presence of arterial occlusion. In control animals, ATP levels during arrest increased slightly to 109% of prearrest levels (from 4.04 to a mean value of 4.46 pmolelgm wet weight), supporting observations of a previous study [l]. After two hours of arrest, the control ATP was significantly higher (p < 0.01) than that obtained from either occluded group (4.46 f 0.31 versus 3.14 f 0.09 and 3.16 f 0.27 pmolelgm wet weight). With coronary arterial occlusion during arrest, ATP levels in the two experimental groups decreased gradually to final measurements of 89.4 and 82.2% of prearrest levels (from 3.56 to 3.14 and from 3.84 to 3.16 pmolelgm wet weight, respectively). The response of ATP to reperfusion, regardless of arterial occlusion, was to decrease from the levels measured before arrest. The difference between the measurements obtained before arrest and after reperfusion were significant in all three groups (p < 0.05), but the most dramatic difference was noted with the persistence of occlusion during the reperfusion period. The level of ATP was significantly lower (p < 0.01) in the occluded group than in either of the other two groups (see Fig 1). The mean level after reperfusion was highest in control animals at 85.7% of the prearrest level (3.12 pmolelgm wet weight). Removal of the ligature in one of the experimental groups, permitting reperfusion, allowed ATP to decrease only to

3 387 Rousou et al: Coronary Artery Occlusion and Myocardial Preservation Levels of Adenosine Triphosphate and Creatine Phosphate in Biopsies of Posterior Left Ventricular Wall Arrest Prearrest Reperfusion Groups Control 1 Hour 2 Hours 30 Minutes ADENOSINE TRIPHOSPHATE RCA patent during arrest 4.04 f f f f 0.22 RCA occluded during arrest, 3.56 f f f f 0.16 patent during reperfusion RCA occluded during arrest 3.84 * f f * 0.27 CREATINE PHOSPHATE RCA patent during arrest 7.88 f f f f 0.66 RCA occluded during arrest, 7.45 f f f * 0.45 RCA occluded during arrest patent during reperfusion 7.47 f f f f 0.36 adata shown as mean k standard error of the mean and measured in pmole/gm wet weight. RCA = right coronary artery 77.4% of prearrest levels (2.75 pmolelgm wet weight) and not differ significantly from control. Retention of the ligature effected a significantly greater (p < 0.01) reduction to 43% of prearrest levels (1.70 pmolelgm wet weight). Creatine Phospate Comparative CP levels during arrest are listed in the Table and are graphed in Figure 2. Regardless of arterial occlusion, all animals exhibited significant (io < 0.01) decreases in tissue levels of CP from prearrest values to one- and two-hour time points. After one hour of arrest, levels were reduced to 60% of prearrest in control animals (from 7.88 to 4.67 pmolelgm wet weight). In the two groups with occlusion, CP levels fell to 32.6% (from 7.45 to 2.32 pmolelgm wet weight and to 46.4% (from 7.47 to 3.53 pmolelgm wet weight) after one hour. Measurements taken after two hours of arrest showed an additional 6% decrease (from 4.67 to 4.06 pmolelgm wet weight) in control animals, while significantly greater reductions (io < 0.05) were observed when coronary arteries were occluded. Final mean percentage levels of CP after two hours of arrest were 17.8% of control in one experimental group (from 7.45 to 1.28 pmolelgm wet weight) and 29.7% in the other (from 7.47 to 2.04 pmolelgm wet weight). The difference between the control group and both occluded groups is significant (p < 0.01) at two hours of arrest. After 30 minutes of reperfusion, CP levels returned to the range of prearrest levels only in the absence of arterial occlusion. Animals in which the coronary arterial ligature was retained had significantly (p < 0.01) depressed levels of CP compared with the groups with patent arteries. Values in control animals approximated prearrest levels (99.3%), as did those in animals in which occlusion was present only during the arrest period (94.4%). In contrast, arterial occlusion during reperfusion kept CP levels well below normal at 51.9% of prearrest levels. Comment Myocardial viability following sustained ischemic arrest has been directly correlated with levels of high-energy phosphates utilized in aerobic and anaerobic metabolism, namely All and CP [7-91. The ability of the heart to recover after an ischemic insult will depend not only on the preservation

4 388 The Annals of Thoracic Surgery Vol 33 No 4 April 1982 T 0 I I I I \ \j Preorrest 60 min 120 min 30 min Control Reperfusion ARREST Fig 1. Effect of occlusion of the right coronary artery (RCA) on the adenosine triphosphate (ATP) level during and following two-hour cardioplegic arrest and 30-minute reperfusion. Potent RCA during Arrest - Potent RCA during Reperfusion Patent RCA during Reperfusion - -A Occluded RCA during Reperfusion T T 0 I I I I Prearresl 60 min I20 min 30 rnin Control Reperfusion ARREST Fig 2. Effect of occlusion of the right coronary artery (RCA) on the creatine phosphate (CP) level during and after cardioplegic arrest. - Pafent Potent RCA during Arrest - ' Patent RCA during Reperfusion RCA during Reperfurion --A Occluded RCA during Reperfuslon

5 389 Rousou et al: Coronary Artery Occlusion and Myocardial Preservation during ischemia, but also the regeneration of high-energy phosphates in the reperfused myocardium. A drop in the ATP content to less than 35 to 40% of preischemic levels will cause irreversible damage synonymous with myocardial necrosis [3, 7, 103. The effectiveness of a cardioplegic agent in protecting the myocardium during ischemia, therefore, can be measured by its ability to preserve the high-energy stores. To achieve this the cardioplegic solution has to gain access to the myocardial cell through the coronary capillary bed, either through its primary feeding artery or through collateral vessels in case of coronary stenosis. Use of hypothermic potassium cardioplegia under optimal conditions with nonobstructed coronary flow minimizes degradation of highenergy phosphates and affords myocardial preservation for up to two hours of ischemic arrest [l]. The presence of acute coronary occlusion, however, can hamper the effectiveness of the cardioplegic agent in protecting the myocardium by limiting adequate contact of the agent with myocardial tissue in the affected region. This has been clearly demonstrated by Hilton and associates in a study in dogs [ll]. They demonstrated metabolic as well as hemodynamic evidence of suboptimal myocardial protection afforded by the cardioplegic agent in regions of coronary occlusion. In our experiment, the impact of acute coronary occlusion was measured on myocardial high-energy phosphates present during the ischemic interval and during the period of reperfusion in a setting simulating inadequate myocardial protection and persistent acute coronary artery occlusion. The acute coronary artery occlusion, especially in areas of inadequate collateral flow, limits the access of cardioplegic solution to the myocardium during the ischemic interval and if not corrected or bypassed, also prevents the restoration of adequate coronary blood flow immediately following the arrest interval. It has already been demonstrated that prevention of adequate exposure of the myocardium to the cold cardioplegic solution leads to increased degradation of high-energy phosphates and poor preservation during ischemic arrest [ll]. We carried the same line of thought one step further into the reperfusion period, to elucidate the consequences of acute coronary occlusion with incomplete revascularization. We recognize and emphasize that this model has some dissimilarities to the clinical setting in the operating room, mainly in regard to the absence of collateral vessels in the pig myocardium. The presence of collaterals in the human heart, however, if indeed present in the underperfused area of myocardium, does not ensure safety, as they may not function in an acute OCclusion in contradistinction to a slow and chronic narrowing of the artery. Collateral protection is, of course, a variable even in human hearts, and this experiment emphasizes one end of this spectrum-the extreme-to bring out the point to be made. In the clinical setting, acute coronary occlusion could occur in a number of instances with resulting myocardial damage similar to what was found in this experiment. In aortic valve replacement, for example, failure to cannulate one coronary artery in the left system because of a short or nonexistent left main coronary artery would result in inadequate myocardial protection much like the second group in which the ligature was present only during the arrest period. On the other hand, a persistent coronary embolus, endothelial damage, or other acute and persistent mechanical coronary occlusion during arrest would result in the very poor myocardial protection seen in the third experimental group in which the ligature was left on through reperfusion. Previous work in our laboratory demonstrated that reperfusion through the normal coronary bed returns CP to normal preischemic levels (100%) and ATP to near normal levels (89.4%), much like the control group in this study (99.3 and 85.7% for CP and ATP, respectively). However, the inability of hypoperfused myocardium to regenerate its energy stores to normal or near normal levels during reperfusion is clearly demonstrated. Both groups with a nonoccluded right coronary artery during reperfusion regenerated the ATP and CP to near normal levels (85.7% and 77.4%, ATP, and 99.3% and 94.4%, CP) while the group with an

6 390 The Annals of Thoracic Surgery Vol 33 No 4 April 1982 occluded right coronary artery showed significantly lower ATP and CP levels (43.7% and 51.9'/0, respectively). These data support our main contention that unrecognized and acute unrelieved coronary occlusion will compromise myocardial protection not only during the ischemic period, by barring access of the cardioplegic agent to myocardial cells, but more importantly during reperfusion by inhibiting adequate highenergy phosphate regeneration. Therefore, in the clinical setting, optimal myocardial protection depends on two major factors. The first is the access of the cardioplegic agent to the myocardial cell in order to minimize the effects of ischemia, and the second is the completeness of reperfusion to ensure reversal of the existing effects of ischemia, the latter being obviously more critical. The surgeon must adopt techniques and means to achieve these two ends, namely, optimal preservation during ischemia and optimal recovery during reperfusion. For example, during aortic valve replacement in a patient with a short left main coronary artery, it is important to cannulate both the left anterior descending and circumflex coronary arteries individually to avoid a situation analogous to the second group in this experiment. Air emboli must be avoided to prevent acute occlusion of coronary arteries and insufficient myocardial preservation or reperfusion or both. Measurement of flow in coronary grafts should be performed, as it will detect obstructive anastomoses that will prevent adequate reperfusion. Probing low-flow grafts sometimes helps to identify technical errors in the anastomoses or to disperse an obstructive air embolus. Finally, we prefer to perform the proximal anastomoses first and then, during the arrest period, to revascularize the warmest areas of the myocardium first so that cardioplegic solution administered between anastomoses can reach these areas of the heart early. These and perhaps other techniques constitute the fine points of myocardial revascularization that will affect preservation of function for immediate and long-term cardiac performance. Supported by Grant 1 R01 HL A1 from the National Institutes of Health and Grant from the American Heart Association. Assistance in the preparation of this manuscript was provided by Mrs. Jamie L. Banks. References 1. Engelman RM, Rousou JH, Longo F, et al: The time course of myocardial high-energy phosphate degradation during potassium cardioplegic arrest. Surgery 86:138, Levitsky S, Wright RN, Rao KS, et al: Does intermittent coronary perfusion offer greater myocardial protection than continuous aortic cross-clamping? Surgery 82:5159, Lochner W, Arnold G, Muller-Rucholtz ER: Metabolism of the artificially arrested heart and of the gas perfused heart. Am J Cardiol 22:299, Bustad LK (ed): Swine in Biochemical Research: Proceedings of a Symposium at the Pacific Northwest Laboratories, Richland, WA, July 19-22, Richland, WA, Batelle Memorial Institute, Engelman RM, Auvil J, O'Donoghue MN, et al: The significance of normothermic anoxic arrest and ventricular fibrillation on the coronary blood flow distribution of the pig. J Thorac Cardiovasc Surg , Lamprecht W, Trautschold J: Determination with hexokinase and glucose-6-phosphate dehydrogenase. In Bergmeyer U (ed): Methods of Enzymatic Analysis. New York, Academic, 1965, pp 543, Hearse DJ, Stewart DA. Chain EB: Recovery from cardiac bypass and elective cardiac arrest: the metabolic consequences of various cardioplegic procedures in the isolated rat heart. Circ Res 35:448, Rao KS, Levitsky S, Holland C, et al: Does potassium chloride-induced cardiac arrest protect the myocardium during aortic cross-clamping? Surg Forum 27:262, Wollenberg A, Krause EG: Metabolic control characteristics of the acutely ischemic myocardium. Am J Cardiol 22:349, Lungsgaahd-Hansen P: Surgical aspects of cardiac metabolism (collective review). Surg Gynecol Obstet 122:1095, Hilton CJ, Teubl W, Acker M, et al: Inadequate cardioplegic protection with obstructed coronary arteries. Ann Thorac Surg 28:323, 1979