M myocardial preservation [l], and may act by antagonizing

Transcription

1 Myocardial Preservation Related to Magnesium Content of Hyperkalemic Cardioplegic Solutions at 8 C Tommy R. Reynolds, MD, Gillian A. Geffin, MB, BS, James S. Titus, Dennis D. O Keefe, MD, and Willard M. Daggett, MD Department of Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts This study investigates whether the addition of magnesium to a hyperkalemic cardioplegic solution containing 0.1 mm ionized calcium improves myocardial preservation, and whether there is an optimal magnesium concentration in this solution. Isolated perfused rat hearts were arrested for two hours by this cardioplegic solution, which was fully oxygenated and infused at 8 C every 15 minutes to simulate clinical conditions. The cardioplegic solution contained either 0, 2, 4, 8, 16, or 32 mm magnesium. At end-arrest, the myocardial creatine phosphate concentration (nanomoles per milligram of dry weight) was20.7f 2.1, ,24.8 f 2.0, , 33.1 * 1.8, and 31.6 f 0.8, respectively, in hearts given cardioplegic solution containing these magnesium concentrations. Thus, the concentration of creatine phosphate was significantly higher at end-arrest when the cardioplegic solution contained 8,16, or 32 mm than 0 or 2 mm magnesium (p < 0.002) or 4 mm magnesium (p < 0.02), and highest with 16 mm magnesium. Also, creatine phosphate was more sensitive to the magnesium concentration of the cardioplegic solution than was endarrest adenosine triphosphate levels, which did not differ among the experimental groups. Aortic flow, expressed as a percentage of prearrest aortic flow, was 60.3 f 5.0, 70.2 & 5.5, 71.6 & 4.4, 71.8 & 4.8, 81.0 f 5.0, and 71.8 f 5.3, respectively. The addition of magnesium to the cardioplegic solution improved recovery of aortic flow (p < 0.05, 16 mm versus 0 mm magnesium). We conclude from these data that with deep myocardial hypothermia and at an ionized calcium concentration of 0.1 mm, the addition of magnesium, over a broad concentration range, improved preservation of myocardial creatine phosphate and, at a concentration of 16 mm, improved aortic flow. The optimal magnesium concentration in the cardioplegic solution was 16 mm. (Ann Tlzomc Surg 1989;47:907-13) agnesium in cardioplegic solutions can enhance M myocardial preservation [l], and may act by antagonizing calcium or by maintaining transcellular magnesium levels during ischemia [2-51. The present study investigates whether the addition of magnesium to a cardioplegic solution that contains 0.1 mm calcium would improve myocardial preservation and whether there is an optimal magnesium concentration in this cardioplegic soh tion. Our cardioplegic solution contained an ionized calcium concentration of 0.1 mm, comparable with the ionized calcium concentration of the dilute blood cardioplegic solution used in our clinical practice [6]. Also, the cardioplegic solution was infused at 8 C to simulate our cardiac surgical conditions. Hearse and co-workers (21 showed in the isolated rat heart perfused at warmer temperatures with St. Thomas s Hospital solution, which has a substantially higher calcium concentration, that the addition of 15 mm magnesium gave the best functional recovery. This group [7] has stressed the need to determine the optimal concentration of each additive to a given cardioplegic solution. With deep hypothermia, which we employ, the optimal concentration of magnesium may be Accepted for publication Jan 25, Address reprint requests to Dr Daggett, Department of Surgery, Massachusetts General Hospital, Boston, MA different from the concentration that provides the best preservation at warmer temperatures. Oxygenation (81 and multidose delivery [9] can further improve myocardial preservation with crystalloid cardioplegic solution, and represent additional factors not incorporated in the earlier study [2] that may alter the effect of magnesium. For this study, we chose the isolated working rat heart model (7, 101 and assessed myocardial preservation from the end-arrest creatine phosphate (CrP) and adenine nucleotide concentrations and from the recovery of left ventricular function. Material and Methods Hearts were obtained from male Sprague-Dawley rats weighing 250 to 425 g. Animals were anesthetized intraperitoneally with sodium pentobarbital (65 mg/kg) and given 100 units of heparin sodium intravenously. The heart was excised and placed initially in iced Krebs- Henseleit bicarbonate buffer (Table 1). The heart was then mounted on the perfusion apparatus [7] and perfused initially for five minutes through the aortic root at a pressure of 100 cm H,O by the method of Langendorff with the Krebs-Henseleit buffer, oxygenated with 95% oxygen and 5% carbon dioxide at 37 C in a reservoir above the heart. All solutions in this study were passed through a 5-pm-pore filter before administration by The Society of Thoracic Surgeons /89/$3.50

2 908 REYNOLDS ET AL Ann Thorac Surg 1989:47: Table 1. Composition of Solutions Variable Krebs-Henseleit Bicarbonate Buffer Base Cardioplegic Solution Na+ (meq/l) K' (meq/l) Ca" (meq/l) 2.4 a... Mg2+ (meq/l) 2.4 b... CI - (meq/l) HCO, (meq/l) H,PO,- (meq/l) so:- (meq/l) Glucose (mm) Mannitol (mm) a There was 0.1 mm ionized calcium. To 1 liter of the base cardioplegic solution, 0 to 16 ml of a magnesium sulfate solution (2.03 mmovml) was added to give appropriate concentrations for the six groups, ie, 0, 2, 4, 8, 16, and 32 mm. During the period of Langendorff perfusion, the left atrium was cannulated through a pulmonary vein. A working heart preparation was then established by perfusing the left atrium with the warm oxygenated Krebs- Henseleit buffer at a pressure of 16 cm H,O and allowing the left ventricle to eject against a pressure of 95 cm H,O into a recirculating aortic column [2, 71. Aortic flow was measured by a flowmeter in the aortic column, previously calibrated by timed volumetric collection. Coronary flow was measured by timed volumetric collection of fluid draining from the heart. Heart rate was obtained from strip-chart recordings of the aortic pressure or the electrocardiographic tracings. Aortic flow, coronary flow, and heart rate were measured at five-minute intervals during the 20-minute prearrest stabilization period, the final measurements being taken as the prearrest control values. Hearts with an aortic flow less than 40 ml/min, coronary flow greater than 25 ml/min, or a heart rate less than 240 beats per minute, findings indicating possible damage during preparation, were rejected from the study. At the end of the control period, the heart was rendered globally ischemic by clamping the left atrial and aortic catheters. Arrest was induced by the infusion of 15 ml of a cold, oxygenated, hyperkalemic crystalloid cardioplegic solution (see Table 1) at a constant pressure of 65 cm H,O through a sidearm on the aortic cannula. The cardioplegic solution, in a water-jacketed reservoir supplying this sidearm, was continuously oxygenated at 4 C with 98% oxygen and 2% carbon dioxide. This resulted in an average ph of 7.41, an average carbon dioxide tension of 39.0 mm Hg, and an average oxygen tension greater than 800 mm Hg measured at 37 C in the cardioplegic solution, with 800 mm Hg being the upper limit of the measurement (Corning model 170 phhlood gas analyzer; Ciba- Corning, Medfield, MA). After initial arrest, 10 ml of cardioplegic solution was infused by the same procedure every 15 minutes for a two-hour period. During arrest, the left ventricle was vented across the mitral valve by opening a fluid-filled catheter connected to the sidearm of the left atrial catheter to air, with the level of the open end adjusted so that no air entered the heart. The heart was reperfused with Krebs-Henseleit buffer at 37 C in the Langendorff mode for 15 minutes. Coronary flow, temperature, and heart rate was recorded, and then the working mode was resumed for a further 30 minutes. Aortic flow, coronary flow, heart rate, and myocardial temperature were recorded at 5, 10, 15, and 30 minutes in the working mode. The reservoirs and tubing of the perfusion apparatus and a chamber surrounding the heart were water jacketed for temperature control. Water at an appropriate temperature was supplied by pump circulators (Haake Buchler Instruments, Saddle Brook, NJ) with thermostats. Myocardial temperature, measured by a thermocouple probe placed in the right ventricle, was maintained close to 37 C during the control and reperfusion periods and at 8 C during arrest. Because function was often very temperature sensitive, care was taken to hold myocardial temperature during reperfusion as close as possible to the prearrest temperature, usually within 0.2"C. Prior to administration of cardioplegic solution, at endarrest, and at end-reperfusion, parallel groups of hearts were assayed for myocardial CrP and adenine nucleotides. Approximately 150 mg of ventricular apex tissue was excised and quickly frozen by compression between metal paddles precooled to -70 C. The frozen tissue was divided into three portions. Each portion was assayed for CrP, adenosine triphosphate (ATP), adenosine diphosphate, and adenosine monophosphate by high-pressure liquid chromatography as described previously [ll, 121. Heart water was determined by obtaining two samples from the ventricle remaining, desiccating them at 80 C, and averaging the results. The results of CrP and adenine nucleotide determinations were averaged and expressed as nanomoles per milligram of dry weight, using the heart water determined for each individual heart. Experimental groups were differentiated by the magnesium content of the cardioplegic solution. Calcium and magnesium were added to the base cardioplegic solution in the following manner: Approximately 0.16 ml of a 10% solution of calcium chloride was added to a liter of the base cardioplegic solution, and the resulting ionized calcium concentration was determined (Nova 2 calcium ion selective electrode; Nova Biochemical, Newton, MA). The ionized calcium for all cardioplegic solutions was t mm. A volume of magnesium sulfate solution (2.03 mmoll ml) was then added to the cardioplegic solution as appropriate to the experimental group (see Table 1). No additional osmotic agent was added to compensate for the differences in cardioplegic solution osmolarity or in the dilution of other cardioplegic solution constituents (1.6% at most) among the experimental groups, as none would be added to the solution subsequently formulated for clinical use. The sequence of experiments was determined by a blocked randomized design for pairs of hearts at a time. Groups differing by magnesium concentration in the cardioplegic solution were compared by one-way analysis of variance followed by t tests or Dunnett's test using

3 Ann Thorac Surg 1989; REYNOLDS ET AL 909 Table 2. Hemodynamic Variables Reperfusion (min)c Group (mm Mg2+) n Variable Prearrestb Aortic flow (ml/min) t f Coronary flow (ml/min) 19? 1 16 t f 1 17? 1 Heart rate (beats per minute) 296 f t Cardiac output (ml/min) ? 3 56 t Aortic flow (ml/min) 61 f 2 43 t ? 4 43? 5 Coronary flow (ml/min) 20? 1 16 f t 1 16? 1 Heart rate (beats per minute) t f f 16 Cardiac output (ml/min) t 5 62 f f 6 8 Aortic flow (ml/min) t f 4 Coronary flow (ml/min) t f 3 Heart rate (beats per minute) Cardiac output (ml/min) t Aortic flow (ml/min) ? 3 46 f 4 Coronary flow (ml/min) 18? 1 17 f Heart rate (beats per minute) 309 & Cardiac output (ml/min) 79? 4 61 t Aortic flow (ml/min) ? Coronary flow (ml/min) Heart rate (beats per minute) 304? t t t Cardiac output (ml/min) ? 4 63 f ? Aortic flow (ml/min) Coronary flow (ml/min) Heart rate (beats per minute) 304? t t t ? 10 Cardiac output (ml/min) t 2 * Data are shown as the mean? the standard error of the mean. working mode was preceded by 15 minutes in the Langendorff mode. Prearrest values were obtained at 20 minutes in the working mode. The BMDP statistical software program P7D [13]. Differences were considered significant at a p value of less than Data are expressed as the mean * the standard error of the mean. All rats received humane care in compliance with the Guiding Principles in the Care and Use of Animals approved by the Council of the American Physiological Society (revised 1980) and the Guide for the Care and Use of Laboratory Animals (NIH publication No , revised 1985). Results Left Ventricular Function Table 2 shows aortic flow, coronary flow, heart rate, and cardiac output recorded before arrest and during reperfusion. Figure 1 displays recovery of aortic flow expressed as a percentage of prearrest aortic flow. Percent recoveries were calculated for each individual animal to obtain the mean. Percent recovery of aortic flow at five minutes in the working mode in the 16 mm magnesium group (81%? 5%) was significantly higher than in the 0 mm group (60% * 5%) by Dunnett s test, although the tail probabil w a m ae do Mg* CONCENTRAllON (mm) Fig I. Recovery of aortic pow during reperfusion, expressed as a percentage of prearrest aortic pow for each individual animal, in the working inode (11 = 8 in each group). Hearts were previously subjected to two hours of cardioplegic arrest and 15 minutes of reperfusion in the Laiigeiidorff mode. Error bars are t the standard error of the iirean.

4 910 REYNOLDS ET AL Ann Thorac Surg 1989; Fig 2. Recovery of cardiac output during reperfusion, expressed as a percentage of prearrest cardiac output for each individual animal, in the working mode (n = 8 in each group). Hearts were previously subjected to two hours of cardioplegic arrest and 15 minutes of reperfusion in the Langendorff mode. Error bars are c the standard error of the mean. ity of the F statistic was greater than Although not significant, this difference persisted throughout the recovery period. There were no other significant differences among the groups for aortic flow. Because coronary flow was a substantial portion of cardiac output, when it was added to aortic flow to calculate percent recovery of cardiac output at five minutes in the working mode, the difference in cardiac output between the 16 mm magnesium group and the 0 mm group was similar to the difference in aortic flow, but no longer significant (Fig 2). Again, there were no other significant differences among the groups for cardiac output. Table 3. Myocardial Concentrations of Creatine Pliospltate and Adenosine Triphosphat8," Group 2-Hour Arrest 45-Minute Reperfusion' (mm Mg2+) n CrP ATP n CrP ATP k 2.1" 10.1 t ? ? C 1.7d 12.8? ? C ? 2.0' t ? ? ? ? k ? t t ? t t ? C 0.6 Data are shown as the mean t the standard error of the mean, and are measured in nanomoles per milligram of dry weight. ' The prearrest values, measured in 8 hearts, were as follows: CrP = 8.6 k 1.2; ATP = 9.3 f 0.8. This includes 15 minutes of reperfusion in the Langendorff mode followed by 30 minutes in the working mode. At end-arrest, CrP levels in the 0 and 2 mm Mg2+ groups were significantly less than in the 8, 16, and 32 Mg2+ groups (p < 0.002). At end-arrest, the CrP level in the 4 mm Mg2+ group was significantly less than in the 8 mm (p < 0.02), 16 mm (17 < 0.002), and 32 mm (y < 0.01) Mg2' groups. ATP = adenosine triphosphate; CrP = creatine phosphate. Fig 3. Myocardial creatine phosphate levels at the end of two hours of cardioplegic arrest (11 = 8 or 9 in each group). Error bars are? the standard error of the mean. High-Energy Phosphates Table 3 shows myocardial concentrations of CrP and ATP. The CrP concentration was highest at end-arrest in the 16 mm magnesium group (Fig 3). The CrP levels were paralleled by the recovery of aortic flow and cardiac output in the working mode. After two hours of arrest, the CfP concentration was higher in all groups than in the prearrest control group. After reperfusion, CrP returned to values close to the prearrest value. Levels of ATP, adenosine diphosphate, and adenosine monophosphate showed no significant differences among the groups at end-arrest or end-reperfusion. All solutions preserved ATP at or above the prearrest concentration for the two-hour arrest period, but ATP levels decreased somewhat during reperfusion. Heart Water Table 4 shows percent heart water at prearrest, endarrest, and end-reperfusion. At end-arrest, there was Table 4. Heart Water"," 2-Hour Arrest 45-Minute Reperfusion' Group Heart Water Heart Water (rnmmg") n (%) t 0.2" ? ? 0.2" t t 0.4' ? ? ? ? C ? 0.2 Data are shown as the mean 2 the standard error of the mean. The prearrest value, measured in 8 hearts, was 82.4%? 0.3%. This includes 15 minutes of reperfusion in the Langendorff mode followed by 30 minutes in the working mode. At end-arrest, percent heart water in the 0 and 2 mm Mg2+ groups was significantly less than in the 8, 16, and 32 mm Mg2+ groups ( 1) < ). At end-arrest, percent heart water in the 4 mm Mg2+ groups was significantly less than in the 8, 16, and 32 mm Mg2+ groups (p < 0.004).

5 Ann Thorac Surg 1989;47:9O7-13 REYNOLDS ET AL 911 OPTIMAL CARDIOPLEGJC MAGNESIUM CONCENTRATION significantly higher percent heart water with increasing magnesium levels in the cardioplegic solution. Heart water generally increased during reperfusion, but there were no significant differences among the groups by the end of reperfusion. Comment The purpose of the study was to find the optimal magnesium concentration for myocardial preservation in a cardioplegic solution containing 0.1 mm ionized calcium when the solution was used to produce hypothermic arrest at 8 C. We found that the CrP level at end-arrest in hearts protected by the cardioplegic solution containing 16 mm magnesium was higher than with any other magnesium concentration, some 50% higher than in the absence of magnesium and significantly higher than CrP in hearts protected by the cardioplegic solution with 4 mm magnesium or less. Although significance was borderline, a concentration of 16 mm magnesium also provided the best percent recovery of aortic flow and cardiac output during early reperfusion, and maintained the highest recovery of both these variables throughout the entire reperfusion period. Intracellular magnesium exists largely as a complex with adenine nucleotides and has a concentration of mmol/kg cell water in rat myocardium [14]. The cytosolic concentration of the free magnesium ion is much lower, nearer 1 mm, which is similar to the extracellular magnesium concentration of 0.5 to 1.5 mm [15]. The exchange of magnesium between intracellular and extracellular compartments is slow, with a half-time in the rat ventricle of approximately three hours [16]. A loss of magnesium during ischemia may restrict cellular function and the resumption of normal metabolic and contractile function. Magnesium in cardioplegic solutions may provide tissue preservation by preventing ischemic cellular loss of magnesium and also may combat some deleterious consequences of potassium arrest, particularly those related to increased calcium influx [3, 41. Magnesium has interactions with calcium, and its beneficial effects as a cardioplegic additive may be dependent on antagonism of calcium by magnesium [2, 41. Magnesium antagonizes the effects of calcium at the sarcolemma and mitochondria, and reduces calcium influx during ischemia. As shown by Boggs and associates [17] in the rat heart, a hyperkalemic cardioplegic solution that was acalcemic proved superior to the same solution with the addition of even a small amount of calcium (0.25 mm ionized calcium). The addition of calcium to cardioplegic solution may decrease myocardial preservation in part by inducing tension development during arrest and thereby increasing energy consumption (2, 181. Increases in left ventricular diastolic pressure during cardioplegia infusions have been observed recently in our laboratory in the isolated isovolumic rat heart and are associated with decreased myocardial high-energy phosphates [19]. When energy production was limited by hypoxia or by the absence of substrate, repeated infusions of cardioplegic solution containing calcium resulted in contracture. As contracture is a state of energy depletion, these observations suggest that the cardioplegic solution infusions and associated pressure development deplete myocardial energy reserves [ 191. Diastolic pressure during cardioplegia infusion was greatly diminished by the addition of magnesium to the calcium-containing cardioplegic solution [19]. We [20] recently confirmed that the addition of 0.1 or 1.2 mm calcium to a cardioplegic solution increases the magnitude of diastolic pressure development during infusion of the cardioplegic solution in a dose-dependent manner. At either of these calcium concentrations, endarrest myocardial high-energy phosphates were depleted and functional recovery was diminished. The addition of 16 mm magnesium to either of these calcium-containing solutions diminished the pressure development during cardioplegia, prevented high-energy phosphate depletion, and enhanced subsequent recovery of left ventricular developed pressure [20]. This previous study [20] and the present study differ in certain respects. In the previous study but not the present study, the left ventricle contained an isovolumic balloon to investigate diastolic tone, which was recorded as pressure development, during arrest; probably in part as a result of this pressure development, the cardioplegic solution containing 0.1 mm calcium but no magnesium caused end-arrest ATP depletion. The same solution did not cause ATP depletion in the present study, in which the ventricle could empty across the mitral valve during arrest and so, presumably, developed little or no pressure; instead it manifested contractile activity as shortening with less energy consumption. In this respect and also in the use of the ejecting rather than isovolumic model before arrest and during reperfusion, the present study is closer to the clinical situation. However, the isolated perfused heart is not stable for a sufficient period for study of long-term effects of cardioplegia. Pernot and colleagues [ 181 evaluated the substitution of magnesium for hyperkalemia in a cardioplegic solution containing 0.25 mm calcium, and found that this substitution increased end-arrest myocardial CrP concentrations and significantly improved early recovery of ventricular function. In general accord with those researchers, our present study showed that the addition of magnesium to a hyperkalemic calcium-containing cardioplegic solution increased CrP concentrations at end-arrest and increased recovery of ventricular function during reperfusion. These beneficial effects are probably due to an antagonism between magnesium and calcium. Hearse and co-workers (21 demonstrated enhanced myocardial preservation with a cardioplegic solution containing magnesium at temperatures of 28" and 37 C. They found that the optimal magnesium concentration was 15 mm in St. Thomas's Hospital cardioplegic solution, which has a calcium concentration of 1.2 mm and a potassium concentration of 16 mm. In our study, although differences in functional recovery among the groups were small, the optimal concentration of magnesium was 16 mm for maximal end-arrest CrP levels. However, the end-arrest CrP level was not significantly different when

6 912 REYNOLDS ET AL Ann Thorac Surg 1989;47: the cardioplegic solution contained 8 or 32 mm magnesium. Experimental conditions and cardioplegic solution formulations differ between our study and the St. Thomas s Hospital study, but in both studies, the optimal magnesium concentrations were similar despite an approximately tenfold difference in calcium concentrations. There was, therefore, no correlation between the optimal magnesium concentration and the calcium concentration. In our study, the CrP concentration at end-arrest was substantially increased above its prearrest value in all groups, but more so at the higher cardioplegic solution magnesium concentrations, thus indicating that CrP production exceeded consumption in the cold arrested hearts. A high myocardial content of CrP at end-arrest appears to correlate with enhanced functional recovery during reperfusion in this study. At the end of reperfusion, CrP had returned to prearrest levels in all groups. End-arrest ATP was at or higher than the prearrest level in all groups. Although these data provide no information on high-energy phosphate compound flux or distribution between subcellular compartments, they suggest good preservation of these compounds. However, the reason for the uniform decline in ATP levels during reperfusion, and whether this decline reflects some failure of protection during arrest or reflects injury during reperfusion, is unknown. The measurement of heart water does not distinguish between water in intracellular, interstitial, and intravascular tissue compartments. Heart water at end-arrest was higher in hearts given cardioplegic solutions with 8 mm magnesium or more. This may relate to the lower resting tone during arrest induced by magnesium in the cardioplegic solution [20]. Increased heart water was not related to impaired functional recovery during reperfusion, however, as functional recovery was best at a magnesium concentration of 16 mm. By end-reperfusion, there were no differences in heart water among the groups. All groups of hearts in this study were well protected by deep hypothermia and a multidose oxygenated hyperkalemic cardioplegic solution. In all groups, cardiac output recovered to 75% or more of its prearrest value by 15 minutes of reperfusion. Although it is difficult to establish differences between groups of hearts that were all well protected, the observations in this study together with previous observations by us [19, 201 and others [2, 4, 181 provide a rationale for including magnesium in calciumcontaining hyperkalemic cardioplegic solutions. The present data suggest that 16 mm is in the optimal concentration range for magnesium in our solution. In conclusion, we showed that hearts preserved by a hyperkalemic cardioplegic solution containing a small amount of calcium without the addition of magnesium had the lowest CrP concentrations at end-arrest and poorest percent recovery of aortic flow and cardiac output throughout the reperfusion period. When magnesium was included in this cardioplegic solution, CrP levels at end-arrest increased substantially and ventricular recovery improved modestly. Both of these variables were optimal at a concentration of 16 mm magnesium. These observations support the use of 16 mm magnesium in our cardioplegic solution. This work was supported in part by Grant HL from the National Institutes of Health. We gratefully acknowledge the high-pressure liquid chromatographic analyses by Alvin G. Denenberg, James E. Vath, and Henry Moise, other chemical analyses by Carmelo Bondi, and preparation of the manuscript by Cindy L. Getherall, Jill S. Jarvis, and Emily C. Burton. References 1. Hearse DJ, Stewart DA, Braimbridge MV. Cellular protection during myocardial ischemia. The development and characterization of a procedure for the induction of reversible ischemic arrest. Circulation 1976;54:19> Hearse DJ, Stewart DA, Braimbridge MV. Myocardial protection during ischemic cardiac arrest. The importance of magnesium in cardioplegic infusates. J Thorac Cardiovasc Surg 1978;74: Shen AC, Jennings RB. Myocardial calcium and magnesium in acute ischemic injury. Am J Pathol 1972;67: Hearse DJ, Braimbridge MV, Jynge P. Components of cardioplegic solutions. In: Protection of the ischemic myocardium: cardioplegia. New York: Raven Press, 1981: Iseri LT, French JH. Magnesium: nature s physiologic calcium blocker. Am Heart J 1984;108: Daggett WM Jr, Randolph JD, Jacobs M, et al. The superiority of cold oxygenated dilute blood cardioplegia. Ann Thorac Surg 1987;43: 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: Bodenhamer RM, DeBoer LWV, Geffin GA, et al. Enhanced myocardial protection during ischemic arrest: oxygenation of a crystalloid cardioplegic solution. J Thorac Cardiovasc Surg 1983;85: Engelman RM, Auvil J, O Donoghue MJ, Levitsky S. The significance of multidose cardioplegia and hypothermia in myocardial preservation during ischemic arrest. J Thorac Cardiovasc Surg 1978;75: Neely JR, Rovetto MJ. Techniques for perfusing isolated rat hearts. In: Hardman JG, OMalley BW, eds. Methods in enzymology. Vol39. New York: Academic Press, 1975: Randolph JD, Toal KW, Geffin GA, et al. Improved myocardial preservation with oxygenated cardioplegic solutions as reflected by on-line monitoring of intramyocardial ph during arrest. J Vasc Surg 1986;3: DeBoer LWV, lngwall JS, Kloner RA, Braunwald E. Prolonged derangements of canine myocardial purine metabolism after a brief coronary artery occlusion not associated with anatomic evidence of necrosis. Proc Natl Acad Sci USA 1980; Dixon WJ. BMDP statistical software manual. Berkeley, CA: University of California Press, 1985: Polimeni PI, Page E. Magnesium in heart muscle. Circ Res 1973;33 : Portzehl H, Zaoralek P, Gaudin J. The activation by CaZ+ of the ATPase of extracted muscle fibrils with variation of ionic strength, ph and concentration of Mg ATP. Biochim Biophys Acta 1969; 189: Page E, Polimeni PI. Magnesium exchange in the rat ventricle. J Physiol 1972;224:

7 Ann Thorac Surg 1989; REYNOLDS ET AL Boggs BR, Torchiana DF, Geffin GA, et al. Optimal myocar- 19. Torchiana DF, Love TR, Hendren WG, et al. Calciumdial preservation with an acalcemic crystalloid cardioplegic induced ventricular contraction during cardioplegic arrest. J solution. J Thorac Cardiovasc Surg 1987;93: Thorac Cardiovasc Surg 1987;94: Pernot AC, Ingwall JS, Menasche P, et al. Evaluation of 20. Geffin GA, Love TR, Hendren WG, et al. The effects of high-energy phosphate metabolism during cardioplegic ar- calcium and magnesium in hyperkalemic cardioplegic solurest and reperfusion: a phosphorus-31 nuclear magnetic tions on myocardial preservation. J Thorac Cardiovasc Surg resonance study. Circulation 1983;67: ;98: REVIEW OF RECENT BOOKS Intravascular and Intracardiac Interventional Catheter Therapy. Techniques and Instrumentation Edited by T. Bonze1 and P. W. Serruys Vienna, Springer-Verlag, pp, illustrated, $35.50 Reviewed by Timothy 1. Gardner, M D Developments in the field of interventional catheter therapy, a field that includes coronary and peripheral vascular angioplasty, cardiac valvuloplasty, and coronary thrombolytic interventions, have obvious implications for cardiovascular surgeons. Clinical experience with such techniques, in particular with coronary angioplasty, has grown at an exponential rate over the last several years, as improved instrumentation and technical refinements have fostered an increasingly bold and aggressive approach in the catheterization laboratories. The book under review, which is actually a special supplement of the Zeitschrift fur Kardiologie, the German journal of cardiology, includes papers presented at a symposium held in 1987 for the intended purpose of defining the state of the art of interventional catheter therapy. Many papers in the volume report on experiences at European centers, although the extensive coronary angioplasty experience at Emory is reviewed by Spencer B. King, and there is also an excellent review of the role of platelet-endothelial wall interactions and antithrombotic therapy for coronary angioplasty presented by Munson and Fuster from the Mt. Sinai School of Medicine in New York. Although the individual papers in this volume vary considerably in importance or relevance for a current understanding of interventional catheter therapy, the time devoted to a review of this material will provide the reader with a moderately useful background. Dr King's review of the Emory University Hospital experience with 7,000 coronary angioplasty procedures performed between 1980 and 1987 is helpful in providing what must be a "best case" report of coronary angioplasty. What is not included in Dr King's report is information about the role of angioplasty in multivessel coronary disease, data that, in fact, may not have been available in 1987 but are clearly of particular interest to those of us who must deal today with the complications of high-risk angioplasty procedures. The papers describing the use of percutaneous aortic and mitral valve dilations are similarly limited, in that the reports have a preliminary character to them. Although these techniques should be considered for the management of some patients with valvular stenoses, the reports in this volume, which were prepared in 1987, do not include information that is really needed today; ie, for whom should these procedures be reserved, and how many of these patients remain improved by catheter treatment at 6, 12, or 24 months later? Other papers in the volume describe specific technical innovations and developments and are of some usefulness to the reader as background information. What is missing from the volume is an editorial perspective, which would be helpful in providing the noncardiologist or nonradiologist with an overview with which to evaluate the importance and relevance of the various reports. If discussions of the various papers, which presumably occurred at the symposium, were included with the papers, the reader would, perhaps, be better informed. The absence of this type of editorial or "peer" review and the additional limitation that reports generated in this rapidly developing field 2 years ago may no longer be relevant to current practice reduce the potential usefulness of this volume for cardiac surgeons, despite our substantial interest in the field of catheter therapy. Baltimore, M D