Calcium Content of St. Thomas I1 Cardioplegic Solution Damages Ischemic Immature Myocardium

Transcription

1 Calcium Content of St. Thomas I1 Cardioplegic Solution Damages Ischemic Immature Myocardium E. Jack Baker IV, MD, Gordon N. Olinger, MD, and John E. Baker, PhD Department of Cardiothoracic Surgery, Medical College of Wisconsin, Milwaukee, Wisconsin Clinical application of hypothermic pharmacologic cardioplegia in pediatric cardiac surgery is less than satisfactory, despite its well known benefits in adults. Protection of the ischemic immature rabbit heart with hypothermia alone is better than with hypothermic St. Thomas I1 cardioplegic solution. Control of cellular calcium is a critical component of cardioplegic protection. We determined whether the existing calcium content of St. Thomas I1 solution (1.2 mmol/l) is responsible for suboptimal protection of the ischemic immature rabbit heart. Modified hypothermic St. Thomas 11 solutions (calcium content, to 2.4 mmol/l) were compared with hypothermic Krebs bicarbonate buffer in protecting ischemic immature (7- to 1-day-old) hearts. Hearts (n = 6 per group) underwent aerobic working perfusion with Krebs buffer, and cardiac function was measured. The hearts were then arrested with a 3-minute infusion of either cold (14 C) Krebs buffer (1.8 mmol calcium/l) as hypothermia alone or cold St. Thomas I1 solution before 6 hours of hypothermic (14OC) global ischemia. Hearts were reperfused, and postischemic enzyme leakage and recovery of function were measured. A bell-shaped doseresponse profile for calcium was observed for recovery of aortic flow but not for creatine kinase leakage, with improved protection at lower calcium concentrations. Optimal myocardial protection occurred at a calcium content of.3 mmol/l, which was better than with hypothermia alone and standard St. Thomas I1 solution. We conclude that the existing calcium content of St. Thomas I1 solution is responsible, in part, for its damaging effect on the ischemic immature rabbit heart. (Ann Thorac Surg 1991;52:993-9) linical application of hypothermic pharmacologic car- C dioplegia in pediatric cardiac surgery is less than satisfactory [l], despite its well-known benefits in adults [2, 31. Previously, we have shown that protection of the ischemic immature rabbit heart with hypothermic Krebs bicarbonate buffer is better than with hypothermic, St. Thomas I1 cardioplegic solution [4, 51. The effect is age-dependent [4] and species-dependent [5]. It is also dependent on oxygen availability in the preischemic perfusate [6]. We have hypothesized that the concentration of one or more of the components of St. Thomas I1 solution is not optimal for the protection of the ischemic immature heart and results in the damaging effect of this cardioplegic solution. Strict control of cellular calcium during ischemia is critical for optimal cardioplegic protection of the myocardium. This control is achieved by manipulating the concentration of the individual components of the cardioplegic solution. The formulation of St. Thomas I1 cardioplegic solution is based on the normal ionic composition of the extracellular fluid, with calcium and sodium concentrations, the main determinants of cellular calcium exchange, present in modestly reduced levels. The study that defined the optimal calcium content of this solution as 1.2 Presented at the Myocardial Preservation Symposium, Oxford, England, Aug 12-15, 199. Address reprint requests to Dr John E. Baker, Department of Cardiothoracic Surgery, Medical College of Wisconsin, 871 Watertown Plank Rd, Milwaukee, WI mmol/l was performed in the mature rat heart subjected to normothermic ischemia [7]. In our study, modified hypothermic St. Thomas I1 solutions, in which the calcium content varied from to 2.4 mmol/l, were compared with hypothermic Krebs buffer in protecting ischemic immature rabbit hearts. The objective of this study was to determine whether the existing calcium content of St. Thomas I1 solution is optimal to protect the ischemic immature heart. Material and Methods Experimental Animals Immature New Zealand white rabbits (n = 7; age, 7 to 1 days) were obtained from a commercial breeder. To avoid dehydration, we kept the immature animals with their mothers until just before induction of anesthesia. Data for body weight, heart weight, and preischemic cardiac function are shown in Table 1. Heparin (15 IUkg) was administered intraperitoneally before anesthesia, which was induced and maintained with halothane (4.% and 1.5%, respectively). The chest was opened and the pericardium removed. The aorta and inferior vena cava were isolated, and the heart was rapidly excised and placed in cold Krebs buffer [8]. The time taken from opening of the chest to excision of the heart was less than 1 minute. Animals used in this study received humane care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the Guide for the Care and Use of Labo by The Society of Thoracic Surgeons /91/$3.5

2 994 MYOCARDIAL PRESERVATION BAKER ET AL Ann Thorac Surg 1991;52:993-9 Table 1. Baseline Measurements in Immature Rabbits" Variable Value Body weight (kg).13 t.1 Heart wet weight (9).7 t.2 Mean aortic pressure (mm Hg) 42 3 Heart rate (beats/min) 294 * 11 Aortic flow (ml/min) Coronary flow (ml/min) 8*1 a Data are mean 2 standard deviation, pooled from all 7 animals used in the study. There were no differences in preischemic control values between any of the individual experimental groups studied. With the exception of body weight and mean aortic pressure, all data are derived from isolated hearts. ratory Animals" published by the National Institutes of Health (NIH publication No , revised 1985). Perfusion System The standard isolated, working heart preparation (Biomedix, Elm Grove, WI), described initially by Neely and colleagues [9], was modified to provide a physiologic workload and temperature suitable for immature rabbits [4]. The cannulas for the aorta and left atrium of immature hearts had an inner diameter of 1.8 mm and an outer diameter of 2. mm. The outer diameter of the aortic cannula matched the inner diameter of the ascending aorta for the immature rabbit. The preparation has been described in detail previously [4]. Essentially, this is a left heart preparation in which oxygenated Krebs bicarbonate buffer supplemented with glucose enters the cannulated left atrium at a pressure of 6 mm Hg. The perfusate then passes to the left ventricle, from which it is spontaneously ejected through the cannulated aorta against a compliant hydrostatic head with a mean pressure of 42 mm Hg. These pressures are physiologic for immature rabbit hearts. Hearts in each test group in which the control hemodynamic performance deviated by one standard deviation or more from the mean values shown in Table 1 were excluded from the study. This exclusion criteria resulted in rejection of approximately 15% of immature hearts. The hearts and perfusates were kept in temperature-controlled chambers to maintain myocardial temperature at the desired level. During ischemia the heart was immersed in the ungassed, preischemic coronary perfusate, either Krebs buffer or St. Thomas' I1 solution. Perfusion Media The standard perfusion medium was Krebs bicarbonate buffer [8] (ph 7.41 at 39 C and 7.25 at 14 C when gassed with 95%, and 5% CO,) in which the calcium content was reduced to 1.8 mmol/l. Glucose (11.1 mmoyl) was added to this solution. The cardioplegic solution was either St. Thomas' I1 solution or a modification thereof in which the calcium content varied between and 2.4 mmoyl [3]. Compositions of the individual perfusates are shown in Table 2. Glucose was excluded from all perfusates during the hypothermic Langendorff perfusion of the coronary network immediately before and throughout ischemia. The oxygen content of Krebs buffer during preischemic perfusion was reduced to.93 ml/dl from its standard value of 2.3 mwdl by gassing the perfusate with a mixture of 25%,:5% C,:7% N,. This adjustment equalized the oxygen content of the two preischemic coronary perfusates [6]. The calcium content of St. Thomas' I1 solution (normally 1.2 mmoyl) was modified from to 2.4 mmol/l. During the preparation of all calciumcontaining solutions, precautions were taken to prevent the precipitation of calcium by gassing the solutions with 5% CO, [4]. Oxygen content of the perfusates was measured using a Lex-,-Con fuel cell analyzer (Lexington Instruments, Waltham, MA) calibrated against room air at room temperature. All measurements of ph were made using a Data Tech 14 ph meter (Data Tech Systems Ltd, Victoria, British Columbia, Canada) against known ph standards at 14 C. The total calcium content of the coronary perfusates was measured using an IL-357 atomic absorption spectrophotometer (Instrumentation Laboratories, Inc, Lexington, MA). Before use, all perfusates were filtered through cellulose acetate membranes having a pore size of 5. pm. Measurement of Indices of Function Aortic flow was defined as the ability of the heart to eject perfusate against the hydrostatic head, with flow rate measured by an in-line analog flow meter (Biomedix). Coronary flow was defined by timed collections of the effluent from the right heart into a graduated cylinder. Electrical activity and heart rate were measured using bipolar electrocardiographic leads attached to the aortic and atrial cannulas. Creatine Kinase Determination The entire coronary effluent during the 15-minute Langendorff reperfusion period was collected in a graduated cylinder surrounded by ice. After the volume was re- Table 2. Composition of the Coronary Perfusates Krebs-Henseleit St. Thomas' Hospital Bicarbonate Cardioplegic Component Buffer (mmol/l) Solution I1 (mmovl) NaCl NaHCO, KCI KH,PO, MgSO, * 7 H2 MgCI,.6 H,O CaCI, * 2 H,O PH", content (ml/dl)b I.1.93 I t.1.92 t.1 a The ph of Krebs-Henseleit bicarbonate buffer was 7.41 at 39 C and 7.25 at 14 C when gassed with 5% CO,. The ph of the cardioplegic solution was manually titrated to 7.8 at 14 C using HC1. The oxygen content of the preischemic coronary perfusates at 14 C was determined at room temperature. Data on ph and oxygen content represent the mean * standard deviation of eight observations.

3 Ann Thorac Surg 1991;52:993-9 MYOCARDIAL PRESERVATION BAKER ET AL P u. 6 & W > 8 4 a: I- z W 8 2 a 1 I I i eb CALCIUM CONTENT IN CORONARY PERFUSATE (mmoles/liter) 1 i Fig 1. Relationship between recovery of postischemic aortic pow and calcium content of the coronary perfusate. Closed histograms represent data from hearts protected by hypothermia plus cardioplegia. The open histogram represents data from hearts protected by hypothermia alone. Data shown represent the mean? standard deviation from six hearts per group. (a = p <.5, hypothermia plus cardioplegia: calcium content = I.2 mmolll versus modified calcium content; b = p <.5, hypothermia alone versus hypothermia plus cavdioplegia.) corded and the contents were mixed, an aliquot of the coronary effluent was taken for the determination of total creatine kinase (adenosine triphosphate: creatine N-phosphotransferase) activity [lo]. Perfusion Sequence The following experiments were performed in a blinded and random manner in ten groups of six hearts each to test the null hypothesis that hypothermic cardioplegic protection of the ischemic immature rabbit heart is unaffected by the calcium content of the cardioplegic solution. Immediately after aortic cannulation, the hearts were perfused in the Langendorff mode [ll] for a 15-minute stabilization period during which the left atrium was cannulated. The hearts were then converted to the working [9] mode for 3 minutes, and control values for aortic and coronary flow, heart rate, and electrical activity were recorded under steady-state conditions. After perfusion in the working mode, hearts were arrested with a 3-minute infusion of the coronary network with either cold (14 C) Krebs buffer (as the hypothermia alone group) or modified St. Thomas' I1 solution (as the hypothermia plus cardioplegia group). Hearts were then subjected to hypothermic (14 C) global ischemia for 6 hours with the coronary network reinfused with the preischemic perfusate for 3 minutes every 3 minutes throughout the ischemic period. This degree of hypothermia is achieved clinically during cardioplegic arrest of the heart. The coronary perfusion was carried out with the perfusion pressure fixed at 6% of the mean aortic pressure for the immature rabbit heart. This reduced pressure was selected to minimize the development of iatrogenic myocardial edema while maintaining adequate volume of perfusion and cooling of the heart. After the ischemic period, hearts were reperfused for 15 minutes in the Langendorff mode, and the entire coronary effluent was collected for determination of creatine kinase activity. Hearts were then converted to the working mode for 3 minutes, during which time the various indices of cardiac function were again measured under steady-state conditions. Each heart thus served as its own control. Expression of Results Recovery of aortic flow was expressed as a percentage of its preischemic value. Creatine kinase leakage was expressed as international units released per 15 minutes per gram dry weight. Six hearts were used for each condition studied, and the results were expressed as the mean k standard deviation. Statistical analysis was performed using the Kruskal-Wallis nonparametric analysis of variance by ranks as a first step, and where this proved significant, the Mann-Whitney test was used as a second step to identify which groups were significantly different [12]. Significance was accepted at the p less than.5 level. Results Baseline Measurements Baseline data for immature rabbit hearts are shown in Table 1. The hemodynamic stability of the model was assessed by perfusing six hearts in the working mode for 12 minutes with nonrecirculating perfusate. No statistically significant changes in aortic flow or heart rate from the values reported in Table 1 occurred until after 85 minutes of perfusion in these immature hearts. In the study described that tests our hypothesis, hearts were perfused in the working mode for a maximum of 6 minutes, well within the stability limits of the preparation. Calcium and Protection of the Immature Heart Figure 1 illustrates several findings with respect to the calcium concentration of St. Thomas' I1 cardioplegic solu-

4 996 MYOCARDIAL PRESERVATION BAKER ET AL Ann Thorac Surg 1991;52:99%9 Fig 2. Relationship between postischemic creatine kinase leakage and calcium content of the corona y perfusate. Closed histograms represent data from hearts protected by hypothermia plus cardioplegia. The open histogram represents data from hearts protected by hypothermia alone. Data shown represent the mean f standard deviation from six hearts per group. (a = p <.5, hypothermia plus cardioplegia: calcium content = 1.2 mmol/l versus modified calcium content; b = p <.5, hypothermia alone versus hypothermia plus cardioplegia.) 25 ~ ~ 5 ~ - I I I I I I I I CALCIUM CONTENT IN CORONARY PERFUSATE (mmoles/liter) T l tion and recovery of aortic flow in immature rabbit hearts subjected to hypothermic global ischemia. First, no significant difference in recovery of postischemic aortic flow was observed in immature hearts protected by hypothermia alone (56% * 5%) compared with the standard St. Thomas I1 solution (6% 2 6%). Second, a bell-shaped dose-response profile for calcium was observed. As the calcium concentration of St. Thomas I1 solution progressively decreased from its normal value of 1.2 mmol/l, the percent recovery of aortic flow increased. Optimal recovery of aortic flow occurred at a calcium concentration.3 mmol/l. Recovery of aortic flow in this group was 81% * 9%; this was significantly greater than with protection provided by either hypothermia alone (56% * 5%) or the standard St. Thomas I1 solution (6% & 6%). Third, in the calcium-free cardioplegia group (3 pmol calciudl), no immature hearts were able to generate postischemic aortic flow, although they were able to develop ventricular pressure and thus maintain the hydrostatic head in the working mode. These results for the calcium-free cardioplegia group were significantly different when compared with both the hypothermia alone group and the standard St. Thomas I1 solution group. All hearts returned spontaneously to a normal rhythm within 3 seconds of reperfusion. There were no significant changes in coronary flow or heart rate upon reperfusion for any calcium concentration investigated. Postischemic Enzyme Leakage Figure 2 shows that for hearts protected with hypothermia alone, creatine kinase leakage during the 15-minute Langendorff reperfusion period after ischemia was 13 I?r 18 IU/g dry wt. When hearts were protected by hypothermia plus the standard St. Thomas I1 solution, creatine kinase leakage was 18 & 37 IU/g dry wt; this was not significantly different from the hypothermia alone group. In hearts where the calcium content of the cardioplegic solution was modified from.1 to 2.4 mmovl, there was no difference in postischemic creatine kinase leakage between these calcium concentrations and the standard St. Thomas I1 solution. However, omission of calcium from the cardioplegic solution resulted in a massive postischemic creatine kinase leakage of 2,28? 43 IU/g dry wt. Thus, with the exception of the calcium-free cardioplegia group, there were no significant differences in postischemic enzyme leakage for the range of calcium concentrations studied. Comment Our study demonstrates that although calcium is an essential component of St. Thomas I1 cardioplegic solution, improved protection of immature hearts subjected to hypothermic global ischemia occurs at a calcium concentration of.3 mmovl. It is apparent that the existing calcium content of St. Thomas I1 solution is responsible, in part, for its damaging effect on the ischemic immature myocardium. Experimental Model Rabbits were selected for these studies because we [4-61 and others [13, 141 have shown that the ischemic immature heart of this species is damaged by the St. Thomas I1 cardioplegic solution. The isolated, working heart model selected for this study allows aspects of left heart function and metabolism to be effectively studied in the same preparation over a wide range of experimental conditions, with the heart subjected to a physiologic workload and temperature. In addition, the model has stable mechanical function, with the perfusate composition well characterized and easily exchanged within seconds. Through preliminary studies, we carefully defined the in vivo mean

5 Ann Thorac Surg 1991;52:99%9 MYOCARDIAL PRESERVATION BAKER ET AL 997 aortic pressure of the rabbits and then matched the hydrostatic pressure of the isolated hearts to the level appropriate for the immature rabbit [4]. The limitations of this type of perfusion system are well known. Aqueous solutions free of red cells, protein, and fatty acids are used to perfuse the coronary arteries, and noncoronary collateral flow is absent. Caution must therefore be applied when extrapolating results obtained using aqueous perfusates in this experimental model to the clinical setting. St. Thomas I1 solution was chosen as an example of a crystalloid cardioplegic solution that is well characterized and is in current clinical use in pediatric cardiac surgery. Hearts were immersed in a bathing media during the ischemia period to maintain the myocardial temperature at the desired degree of hypothermia and also to prevent desiccation of the heart. The medium used was either ungassed, substrate-free Krebs buffer for hearts protected by hypothermia alone or modified St. Thomas I1 solution for hearts protected by hypothermia plus cardioplegia. We have previously shown that the composition of the bathing media used to surround the heart does not influence the extent of postischemic functional recovery [4]. Therefore diffusion of oxygen or substrate into the thin-walled immature hearts to exert a potential metabolic effect cannot be invoked to explain differences in tolerance to ischemia in our study. We have also shown that recovery of postischemic function in the immature rabbit heart is dependent, in part, on the availability of oxygen in the preischemic coronary perfusate [6]. To control for this dependency in our study, the oxygen content of the preischemic Krebs buffer was reduced to.93 ml/dl from its normal value of 2.3 ml/dl to equal the oxygen content of the cardioplegic solution. This reduction in oxygen availability resulted in a decreased recovery of aortic flow in the hypothermia alone group in comparison with our previous studies [4, 51, where the oxygen content of the preischemic Krebs buffer was 2.3 ml/dl. Thus, protection with hypothermia alone in the current study was no better than with the standard hypothermic St. Thomas I1 solution. The prolonged period of ischemia selected for our protocol was necessary as we [4] have shown that the immature rabbit heart is considerably more tolerant of ischemia than the mature heart. As we wished to design an experiment that would reveal any beneficial or detrimental effects of varying the calcium content of the standard cardioplegic solution, the control group (1.2 mmol calcium/l) in such a study should ideally recover approximately 5%. Therefore, it was necessary to extend the ischemic period to 6 hours to produce this degree of injury. Calcium and Protection of the Immature Heart Although the use of cardioplegic solutions during adult cardiac operations has improved postischemic functional recovery and has reduced postoperative mortality [3], these same benefits have not extended as satisfactorily to the pediatric patient [l]. In support of this clinical observation, studies in the rabbit [P6, 13, 141, pig [15], dog [16], and guinea pig [17] have demonstrated that hypothermic potassium cardioplegia inadequately protects the ischemic immature myocardium. Myocardial ischemia results in an adverse redistribution of electrolytes, especially calcium, between the extracellular and intracellular space, as well as within the myocyte organelles. Because this process can result in cell injury and poor recovery of postischemic function, strict control of cellular calcium during ischemia has been shown to be a critical factor in cardioplegic protection [3]. Control of calcium and other cellular electrolytes is achieved by manipulating the concentration of the components of the cardioplegic solution. Historically, development of St. Thomas I1 solution was based on the normal ionic composition of the extracellular fluid, with calcium and sodium concentrations modestly reduced [3]. The calcium content of the St. Thomas I1 solution was set at 1.2 mmovl, a value determined from studies performed in the mature rat heart subjected to normothermic ischemia [7]. These studies further demonstrated that the mature rat heart is extremely sensitive to small changes in extracellular calcium, such that a 1% to 2% change from the optimal calcium concentration could result in as much as a 5% loss in protection, as determined by both functional and enzymatic criteria. Our study demonstrates that the standard St. Thomas I1 solution damages the ischemic immature myocardium. Optimal myocardial protection in our study occurred when the calcium concentration was reduced to.3 mmovl from its standard value of 1.2 mmol/l. This improved level of protection of the ischemic immature heart over that provided by hypothermia alone and the standard St. Thomas I1 solution was also manifest over a narrow range of calcium concentrations. In a separate study of the dose-response characteristics of calcium for the protection of the ischemic immature rat heart with St. Thomas I1 solution [18], the optimal calcium content of.3 mmol/l, as defined in our study, was not investigated. Another investigation of cardioplegic protection in the ischemic immature rabbit heart [19] evaluated only extremes of calcium concentrations. It is clearly appropriate that dose-response studies of components of cardioplegic solutions be performed using a comprehensive range of concentrations of the individual component under investiga tion. Moreover, our finding is in contrast to the study of Konishi and Apstein [ZO], who showed that in the isolated, blood-perfused, 4- to 6-day-old rabbit heart, recovery of postischemic ventricular function after protection with hypothermia plus St. Thomas I1 cardioplegia was superior to that provided by hypothermia alone. However, in their study, lidocaine hydrochloride, which is normally excluded from St. Thomas I1 solution, was included at a concentration of.9 mmovl, and the sodium content of the solution was reduced from its normal value of 12 mmol/l to 11 mmol/l. These alterations in the composition of the components of the cardioplegic solution, in conjunction with the different experimental model and the younger age group, may be responsible for the discrepancy between our two studies. Effective com-

6 998 MYOCARDIAL PRESERVATION BAKER ET AL Ann Thorac Surg 1Y91;52:9Y%Y parisons between studies can be made only if these variables are standardized between laboratories before investigation. The reason for the shift in the optimal calcium content of the cardioplegic solution from a physiologic value of 1.2 mmol/l to a relatively nonphysiologic value of.3 mmol/l is unknown. However, Robinson and Harwood [21] also have shown that lowering the calcium content in St. Thomas I1 cardioplegic solution improves myocardial protection in the adult rat heart subjected to hypothermic ischemia. Although their study was performed in hearts from a different species and age group, one common feature of these two studies is that hearts were subjected to hypothermic global ischemia. Hypothermia affects a number of membrane processes, including transition of sarcolemmal membranes from a fluid to a gel state and decreasing calcium influx through the slow calcium channel. These hypothermia-induced changes may result in stabilization of the sarcolemmal membrane, which prevents adverse transmembrane electrolyte movements, especially calcium during ischemia. Under these conditions, the presence of physiologic levels of extracellular calcium may be unnecessary for cell integrity, and even injurious to the myocardial cell, resulting in damaging processes, such as calcium-induced energy depletion, which may in turn impair functional recovery of the heart. We propose that the optimal calcium content of St. Thomas I1 cardioplegic solution is temperature-dependent, resulting in the shift of the optimal calcium content under clinically relevant conditions of hypothermia to a value of.3 mmovl. Calcium and Postischemic Enzyme Leakage Although our study demonstrates a relationship between recovery of postischemic function and the calcium concentration of the cardioplegic solution, no such relationship exists between postischemic creatine kinase leakage and calcium concentration. All groups of hearts, with the exception of the group infused with calcium-free St. Thomas I1 solution, exhibited similar postischemic creatine kinase leakage compared with the hypothermia alone group and the standard St. Thomas I1 solution group. The absence of a correlation between postischemic creatine kinase leakage and the calcium content of the cardioplegic solution stresses the need for multiple, independent markers to assess myocardial protection [3], as we have shown that the use of a single marker of tissue injury may be unreliable. Further investigation is necessary to determine the precise relationship between functional and enzymatic indices of tissue injury in the immature heart. In the calcium-free cardioplegia group, massive creatine kinase leakage into the coronary effluent was observed during post-ischemic reperfusion. This may be attributed to the induction of a calcium paradox, as the calcium content in this group was low enough (3 pmol/l) to predispose hearts to this severe form of myocardial injury. However, the calcium paradox is characterized by severe contractile failure in addition to massive enzyme leakage. We observed that calcium-depleted hearts were still able to generate ventricular pressure but not aortic flow upon reperfusion in the working mode. This dissociation of functional recovery from enzyme leakage in the calcium-free cardioplegia group may reflect either the insensitivity of immature hearts to the calcium paradox or the presence of serious heterogeneity of injury, in which some cells are susceptible to the calcium paradox whereas others are not. This may result from a mixture of functional cells, which retain their ability to contract and have also retained their enzymes, and nonfunctional cells, which have been damaged by the induction of the calcium paradox where substantial enzyme leakage has occurred. In conclusion, this study demonstrates that although calcium is a necessary component of St. Thomas I1 solution, the existing concentration of 1.2 mmol/l affords suboptimal protection to ischemic immature hearts during conditions of hypothermic global ischemia, therefore resulting in postischemic myocardial damage. Maximal recovery occurs when the calcium concentration of the cardioplegic solution is reduced to.3 mmovl. Further evaluation of the components of this and other cardioplegic solutions should be undertaken to expand the database necessary to develop a cardioplegic solution specific for the needs of the pediatric patient. This work was supported by grants from the National Institutes of Health (HL-433) and Ronald McDonald Children s Charities. We thank Bob Byrnes, PhD, for assistance in determining the calcium content of the coronary perfusates. The excellent secretarial assistance of Mary Lynne Koenig is gratefully acknowledged by J.E.B. References 1. Bull C, Cooper J, Stark J. Cardioplegic protection of the child s heart. J Thorac Cardiovasc Surg 1984;88: Engelman RM, Levitsky S. A textbook of clinical cardioplegia. 1st ed. Mount Kisco, NY: Futura, Hearse DJ, Braimbridge MV, Jynge P. Protection of the ischemic myocardium: cardioplegia. 1st ed. New York: Raven, Baker JE, Boerboom LE, Olinger GN. Age-related changes in the ability of hypothermia and cardioplegia to protect ischemic rabbit myocardium. J Thorac Cardiovasc Surg 1988;96: Baker JE, Boerboom LE, Olinger GN. Is protection of ischemic neonatal myocardium by cardioplegia species dependent? J Thorac Cardiovasc Surg 199;99: Baker JE, Boerboom LE, Olinger GN. Cardioplegia-induced damage to ischemic immature heart is independent of oxygen availability. Ann Thorac Surg 199;5: Yamamoto F, Braimbridge MV, Hearse DJ. Calcium and cardioplegia. The optimal calcium content for the St. Thomas Hospital cardioplegic solution. J Thorac Cardiovasc Surg 1984;87: Krebs HA, Henseleit K. Untersuchungen uber die Harstoffbidung im Tierkorper. Hoppe-Seyler s Z Physiol Chem 1932; 21:3> Neely JR, Liebermeister H, Battersby DJ, et al. Effect of pressure development on oxygen consumption by the isolated rat heart. Am J Physiol 1967;212: Horder M, Elser RC, Gerhardt W, et al. IFCC methods for the measurement of catalytic concentration of enzymes. Part 7.

7 Ann Thorac Surg 1991: MYOCARDIAL PRESERVATION BAKER ET AL 999 IFCC method for the creatine kinase (ATP: creatine N-phosphotransferase, E.C ). J Int Fed Clin Chem 1989;l: Langendorff. Untersuchungen am uberlebenden Saugertierherzen. Pflugers Arch Ges Physiol 1895;61: SAS Institute Inc. SAS/STAT User s Guide, Release 6.3 Edition. North Carolina: SAS Institute Inc, 1988: Magovern JA, Pae WE, Miller CA, et al. The immature and mature myocardium. Responses to multidose crystalloid cardioplegia. J Thorac Cardiovasc Surg 1988;95: Kempsford RD, Hearse DJ. Protection of the immature heart. Temperature-dependent beneficial or detrimental effects of multidose crystalloid cardioplegia in the neonatal rabbit heart. J Thorac Cardiovasc Surg 199;99: Corno AF, Bethencourt DM, Laks H, et al. Myocardial protection in the neonatal heart. A comparison of topical hypothermia and crystalloid and blood cardioplegic solution. J Thorac Cardiovasc Surg 1987;93: Baxter TB, Campbell DN, Clarke DR, et al. Evidence against hypothermic cardioplegic arrest in pediatric heart surgery [Abstract]. International Symposium on Myocardial Protection. Boston University School of Medicine 1985: Wantanabe H, Yokosama T, Eguchi S, et al. Functional and metabolic protection of the neonatal myocardium from ischemia: insufficient protection by cardioplegia. J Thorac Cardiovasc Surg 1988;97: Riva E, Hearse DJ. Dose-dependency of the neonatal heart to calcium with different durations of ischemia [Abstract]. Circulation 1988;78(Suppl 4): Diaco M, DiSesa VJ, Sun SC, et al. Cardioplegia for the immature myocardium. A comparative study in the neonatal rabbit. J Thorac Cardiovasc Surg 199;1:91C Konishi T, Apstein CS. Comparison of three cardioplegic solutions during hypothermic ischemic arrest in neonatal blood-perfused rabbit hearts. J Thorac Cardiovasc Surg 1989; 98: Robinson LA, Harwood DL. Lower calcium improves protection with St. Thomas Hospital cardioplegic solution during hypothermic ischemia [Abstract]. J Mol Cell Cardiol 1988; 2O(Suppl 5):82.