Effects of Cardioplegic Solution on Human Contractile Element Velocity

Transcription

1 ORIGINAL ARTICLES Effects of Cardioplegic Solution on Human Contractile Element Velocity Douglas M. Behrendt, M.D., Marvin M. Kirsh, M.D., Kenneth E. Jochim, Ph.D., and Herbert Sloan, M.D. ABSTRACT A technique for measuring the maximum contractile element velocity (Vp,) of the myocardium was developed, verified, and employed in patients to allow accurate intraoperative assessment of the adequacy of myocardial protection. Four groups of patients were studied. Ten patients had coronary artery bypass grafts (CABG) with cardioplegia; 13 had CABG with coronary perfusion, ventricular fibrillation at 28"C, and aortic clamping for distal anastamoses; 6 had aortic valve replacement (AVR) with cardioplegia; and 7 had AVR with coronary perfusion to the beating heart. For cardioplegia, a solution of 5% dextrose in 0.2% saline at 4 C with 25 meq of potassium chloride and 12.5 gm of mannitol was infused initially, followed by 500 ml every 30 minutes. Clinically all patients did well, and there were no deaths. Patients having CABG with intermittent coronary perfusion during ventricular fibrillation had significant (p < 0.01) depression of V, from 38.3 to 30.8 sec-' while V,, in patients having CABG with cardioplegia was unchanged. Patients having AVR with continuous coronary perfusion or with cardioplegia (average anoxia time, 70.4 minutes) had no significant change in V,,. We conclude that this cardioplegic solution provided adequate protection of myocardial function for up to 105 minutes of continuous aortic clamping in humans. The depression in V, observed following CABG with intermittent coronary perfusion is consistent with previous suggestions that this combination is detrimental because of maldistribution of coronary blood flow during ventricular fibrillation. One of the most fundamental problems in cardiac surgery is how best to protect the myocardium during open-heart procedures, especially when these involve aortic cross-clamping. From the Department of Surgery, Section of Thoracic Surgery, and the Department of Physiology, University of Michigan Medical Center, Ann Arbor, MI Presented at the Fourteenth Annual Meeting of The Society of Thoracic Surgeons, Jan 23-25, 1978, Orlando, FL. Address reprint requests to Dr. Behrendt, University Hospital, C7079 Outpatient Building, Ann Arbor, MI There are many opinions on this subject but few answers, and the literature is replete with techniques said to protect the myocardium, including coronary perfusion with or without ventricular fibrillation, hypothermia, and cardioplegic solutions. One reason for this confusion is the difficulty of precisely assessing the effects of any particular technique on myocardial function. The performance of the heart depends upon four factors: preload, afterload, heart rate, and the contractile state of the myocardium. It is contractility itself that must be evaluated if the efficacy of a proposed technique of "myocardial protection" is to be assessed. Unfortunately, the indices of cardiac function commonly used, such as cardiac output, ejection fraction, stroke work, and maximum rate of rise of left ventricular pressure (dpldt) are influenced by all of the above factors to a greater or lesser extent, and it is impossible to analyze contractile state specifically without controlling the other three. Particularly in clinical situations, controlling heart rate, preload, and afterload simultaneously so that contractility may be assessed is very difficult, and the necessary measurements are tedious. In the course of our investigations into the effects of ischemia on myocardial function, we have found that the measurement of contractile element velocity (V,,,,,) offers a tool for assessing contractility that can be utilized easily in clinical settings. In the present study we used it to assess the adequacy of myocardial protection afforded by either a simple cardioplegic solution or by coronary perfusion in patients undergoing aortic valve replacement (AVR) or coronary artery bypass grafts (CABG). Methods Assessment of MyocardiaI Contractility Sonnenblick 1151 in 1962 proposed that cardiac muscle behavior can be analyzed in the same by Douglas M. Behrendt

2 500 The Annals of Thoracic Surgery Vol 26 No 6 December 1978 manner as skeletal muscle. Using cat papillary muscles contracting isotonically in a bath of physiological solution, he developed forcevelocity curves, which related rate of muscle shortening to afterload. He showed that for each muscle a family of curves corresponding to different preloads could be obtained, and that all of these curves could be extrapolated back to a single point at zero afterload. He termed this point Vmax -the velocity of muscle fiber shortening at zero afterload-and stated that this was a unique characteristic of each muscle under a given set of conditions, reflecting its contractile state independent of preload. Subsequently, it has been shown that V, is probably not independent of preload in other preparations such as the intact human heart, thus limiting its usefulness as a contractility index [ll]. Also, it is difficult to calculate accurately. For these reasons, Mirsky and associates [8] and others have suggested using the quantity Vpm -the actual or physiological maximal observed contractile element velocity-as a contractility index. During the isometric period of contraction, wall tension is produced by shortening of the myocardial contractile elements. The velocity of contractile element shortening (Vc,) can be shown to be related to interventricular pressure by the equation Vcg, = dp/dt/kp where K is a constant for which various values have been determined in cat papillary muscles and P is left ventricular pressure. Mirsky and colleagues [81 found that a value of V, (i.e., the maximum value of Vce) greater than 4.5 sec- (assuming K = 28) indicates normal myocardial contractility; lower values reliably separate out patients known to have abnormal myocardial function by other criteria. Because V,, appeared to be a useful practical index of contractility from these previous studies, we constructed a system for measuring it easily in the operating room to help assess the efficacy of various methods of myocardial protection. A simple electronic circuit was designed to allow (dp/dt)/p to be displayed instantaneously on an oscilloscope screen for photographic recording [4]. The output of a carrier amplifier receiving a high-fidelity left ventricular pressure signal (P) is connected to a logarithmic amplifier. The output of this, now log P, is differentiated by a resistance-capacitance (RC) circuit whose output (E,,) then is: A E,, = - x RC d(lo /P) where A is the gain of the logarithmic amplifier, and R and C are the resistance and capacitance constants of the circuit. d(lo P) d(1og P) dp Since + - dp X- dt and 9 = lip then E,, = - A d(1og P) -- A x dp/dt.- RC dt RC P :. - v,, = - RC x E,, AK The voltage output of this circuit (E,) is connected to the vertical deflection plates while P is connected to the horizontal deflection plates of an oscilloscope. The resulting curve forms a loop with (dp/dt)/p as the ordinate and P as the abscissa (Figs 1, 2). It is a simple matter to photograph the oscilloscope trace and measure V,, directly, using an appropriate calibration factor [41. Thus, contractile element velocity can be easily determined from left ventricular pressure data alone in the operating room. The high-fidelity pressure signal required for differentiation was obtained from a transducer-tipped catheter (6F Millar) placed in the left ventricle either through an apical stab wound or across the mitral valve from the right superior pulmonary vein. We ignored the constant K in our calculations because its value is unknown in humans. A value of 28 has been assumed by some authors based on cat papillary muscle data. Validation of V,,, as Contractility Index In order to use V,, as a contractility index, it was first essential to determine the effects alterations in preload, afterload, or heart rate might have on it so that they might be controlled if necessary. Previous experimental work sug-

3 501 Behrendt et al: Cardioplegic Solution and Contractile Element Velocity MILLAR CATHETER I - dp/dt --. P Yi n Fig 1. Method of recording contractile element velocity on oscilloscope. (Log Amp = logarithmic amplifier; RC Diff = resistance and capacitance difference; (dp/dt)/p = left ventricular pressure data.) - h LVEDP Vpm 4 DEVELOPED PRESSURE 46.5 ' ).? 15.5 a 1 c U a' 0 Y TOTAL PRESSURE (mmhg) Fig 2. Contractile element velocity loop. The V,,,n is the maximum positive vertical deflection. (LVEDP = left ventricular end-diastolic pressure; (dp/dt)/p = left ventricular pressure data.) gested that large changes in preload or afterload may significantly effect V,, [2, 6, 7, 131. Increasing heart rate has been shown to increase V, [3, 131. Unfortunately, in these studies preload and afterload were allowed to vary together, preventing analysis of the effects of each alone. We assessed these effects under more controlled circumstances to verify the validity of these previous studies. Loops of (df/dt)/f were recorded during isotonic contractions in three standard cat papillary muscle preparations. F is the force developed by the muscle (analogous to P in the whole heart) and dfldt is the rate of change of force. Changes in afterload had no effect On v,,, whereas large preload or rate changes had a marked effect (Fig 3). Changing heart rate often affected V, considerably. Similar observations were made in the operating room in patients cannulated for cardiopulmonary bypass (see Fig 3). Preload was changed in 5 patients by infusion of 400 to 600 ml from the pump resulting in an increase of 0 to 5 mm Hg in left ventricular end-diastolic pressure (LVEDP) and a -10 to +SO/O change in V,,. Heart rate was increased by 20% in 3 patients and V,, increased substantially in 2. Afterload was altered in 2 patients without a change in preload by administration of Neo- Synephrine (phenylephrine hydrochloride) or nitroprusside, and V,, changed in neither patient. From these observations we concluded that heart rate and preload must be controlled within reasonable limits but that afterload can be ignored in using V,, as a contractility index. These observations essentially confirm the previous experimental studies already mentioned. Although V, has limitations just like other contractility indices, it appeared to be easier to record and less subject to error than most. Material With this background we studied four groups of patients undergoing cardiac operation. Group 1 consisted of 10 patients having CABG with cardioplegia. Distal anastomoses were completed during one continuous period of cross-clamping (average, 62.5 minutes; range, 22 to 107 minutes). Then the proximal anastomoses were constructed with partial occlu-

4 502 The Annals of Thoracic Surgery Vol 26 No 6 December 1978 i A Y PAPILLARY MUSCLE 31 i INTACT HEART mmhg PAPILLARY MUSCLE B i 60/min INTACT HEART C PAPILLARY MUSCLE INTACT HEART Fig 3. (A) Effects of changing afterload on V,,,,, in isolated rnuscle (left) and in the intact heart (right). Little effecf is observed. (B) Effects of changing preload on V,,,, in isolated muscle (left) and in the intact heart (right). Large increases in preload cause a reduction in V,,,,, in the papillary muscle. The effect in the intact heart is small. In this example, a 600-ml acute increase in blood uoluine caused a 5 mm Hg increase in LVEDP and a 10% reduction in V,,,,,. (C) Effects of changing heart rate on V,,,,, in isolated muscle (left) and in the intact heart (ri,qht). In both situations, increasing rate increases V,,,,, sion clamps after the heart had been rewarmed and defibrillated (when necessary). An average of 2.9 grafts per patient was performed. In Group 2 were 13 patients having CABG with coronary perfusion to the fibrillating heart at 28 C. Intermittent periods of aortic clamping were used for the distal anastomoses (average, 7 minutes; range, 2 to 12.5 minutes). The average number of grafts per patient was 2.7. In Group 3 were 6 patients having AVR with cardioplegia. Ischemic time averaged 67 minutes (range, 53 to 91 minutes). The fourth group had 7 patients undergoing AVR with continuous coronary perfusion to the beating heart at 32 C. The patients having cardioplegia did not differ from those having coronary perfusion in age, sex, New York Heart Association Functional Class, number of coronary vessels occluded, ejection fraction, LVEDP, or number of previous myocardial infarctions.

5 503 Behrendt et al: Cardioplegic Solution and Contractile Element Velocity Summary of Results in the Four Groups of Patientsa CORONARY BYPASS V,, (sec-i) LVEDP (mm Hg) No. of CABG Patients Preou. Postov. Preov. Postov. Coronary perfusion k f 7.7b 11.3 k k Cardioplegia k f 6.9 AORTIC VALVE REPLACEMENT Coronary perfusion k f k k 10.8 Cardiouleeia k f k k 10.6 "All values expressed as mean + standard deviation. hdifference between preoperative and postoperative values, p < 0.01 by paired t test. No other postoperative data are significantly different from preoperative data. CABG = coronary artery bypass graft; V,,, = maximum contractile element velocity; LVEDP = left ventricular end-diastolic pressure. In each patient V,, and LVEDP were recorded immediately before and after cardiopulmonary bypass with the heart cannulated, the body temperature between 35" and 37"C, and the heart rate held constant by atrial pacing. No catecholamines were given until these recordings were completed. The cardioplegic solution consisted of a 4 C solution of 5% dextrose in 0.2% saline with 25 meq of potassium chloride and 12.5 gm of mannitol added per liter. It had a ph of 6.8 and an osmolarity of 380 mosm. One liter was given initially over a 3-minute period either into the aortic root from a pressurized bag or through individual coronary perfusion cannulas when there was aortic valve insufficiency. Additional 500-ml infusions were given every 30 minutes. The myocardial temperature was reduced to 10" to 15 C during each infusion and then gradually rose to 20" to 25 C during the ensuing 30 minutes. The patient's body temperature was lowered to 28 C before aortic clamping and maintained at that level during perfusion. It was raised to 32 C just before the clamp was released. Approximately half of the patients required a single low-energy shock for defibrillation. Results Patients having CABG with cardioplegia had no statistically significant change in V, following operation (Table). One required 0.2 mg of Isuprel (isoproterenol hydrochloride) per minute for 12 hours after operation. There were no myocardial infarctions or deaths. Patients having CABG with coronary perfusion and intermittent clamping for distal anastamoses had a statistically significant (p < 0.01) depression in V,, from an average of 38.3 sec-i to 30.8 sec-i. There were no deaths or myocardial infarctions in this group. Three patients required 5 to 10 pg per minute of norepinephrine (with simultaneous phentolamine) for twelve hours. Neither patients having AVR with cardioplegia nor those having AVR with coronary perfusion had statistically significant changes in V,,. Two in each group required small doses of catecholamines briefly in the operating room. Two who had coronary perfusion required 5 to 15 pg per minute of norepinephrine (with simultaneous phentolamine) for 12 to 24 hours after operation. There were no deaths or myocardial infarctions in either group. Comment Protection of the myocardium during aortic clamping is clearly one of the important goals of cardiac surgeons. That myocardial protection may be often inadequate is suggested by an appreciable incidence of early low-output syndrome and late myocardial dysfunction evident upon follow-up catheterization. Coronary perfusion would seem to be a logi-

6 504 The Annals of Thoracic Surgery Vol 26 No 6 December 1978 cal method of protection. However, it is often inconvenient and the unrelaxed heart may be difficult to repair without trauma. Coronary osteal damage has occurred from perfusion cannulas. Most perfusion systems produce nonpulsatile flow, which may not be ideal. Sometimes satisfactory coronary perfusion cannot be achieved. Hypothermia by systemic and topical cooling (the so-called Shumway technique) is another clinically proved method, but sometimes it is cumbersome. Especially in hypertrophied hearts, cooling may be nonuniform, and unsuspected endocardia1 warming may occur unless myocardial temperature is carefully monitored. Furthermore, the heart fibrillates for a considerable period both during cooling and rewarming, and this has been shown to increase oxygen demands and to cause maldistribution to coronary flow in experimental animals [5]. Thus, a reexamination of potassium cardioplegia has been advocated. Virtually all controlled experimental data come from work with isolated perfused rat or dog hearts. Papillary muscles are unsuitable because they are not arrested by these solutions. Previous studies on humans have been limited to qualitative results such as survival, freedom from complications, and the need for postoperative support measures. From available data it would appear that solutions containing 15 to 30 meq of potassium per liter are both safe and effective in producing cardioplegia [9, 14, 161. Earlier methods (i.e., the Melrose technique) were detrimental probably because excessive concentrations of potassium or citrate or both were used. Potassium produces cardioplegia by reducing the transmembrane potential, thereby inhibiting myocardial excitation and conserving cellular energy stores during the arrest period. We added mannitol to our solution because there is considerable evidence that it protects the myocardium from the cell swelling that accompanies ischemia [ll]. Many other agents have been utilized, but there is little information about the ideal composition. In particular, the ideal ionic composition, ph, osmolarity, and temperature must be determined. Current work in many laboratories is directed at these problems. Nevertheless, the present data clearly show that immediate postoperative myocardial function is well preserved by the simple cardioplegic solution utilized in this study. They further suggest that this technique is superior to intermittent coronary perfusion to the fibrillating heart. It is interesting that this should appear to be the case even in hearts that are not hypertrophied. These conclusions have recently been confirmed by Adappa and associates [l J who found that cardioplegia resulted in a lower incidence of arrhythmias and of myocardial damage among their patients having coronary bypass than did intermittent aortic clamping. Similarly, Wakabayashi and associates [17] noted a progressive decline of myocardial contractility in rabbits as aortic cross-clamping and coronary reperfusion were repeated. Associated with the decrease in V,, among the patients having intermittent coronary perfusion was a slight increase in LVEDP. Of course, this might be a further reflection of diminished contractility, but, alternatively, it might represent decreased left ventricular compliance due to edema accumulation during ventricular fibrillation. If this were the case, the decrease in V, observed in this group might merely be due to the increased LVEDP and not to a change in contractility itself. This distinction cannot be made from the data obtained in this study. References 1. Adappa MG, Jacobson LB, Hetzer R, et al: Cold hyperkalemic cardiac arrest versus intermittent aortic cross-clamping and topical hypothermia for coronary bypass surgery. J Thorac Cardiovasc Surg 75:171, Cosyn J, Gutierrez-Miranda M, Reyns PH, et al: Superiority of developed over total pressure for heart contractility indices in dogs. Pfluegers Arch 362:165, Cove11 JW, Ross J, Sonnenblick EH, et al: Effects of increasing frequency of contraction on the force velocity relations of left ventricle. Cardiovasc Res 1:2, Grossman W, Brooks H, Meister S, et al: New technique for determining instantaneous myocardial contractility in man from ventricular pressure recordings. Circ Res 28:290, 1977

7 505 Behrendt et al: Cardioplegic Solution and Contractile Element Velocity 5. Hottenrott CE, Towers B, Kurkji HJ, et al: The hazard of ventricular fibrillation in hypertrcphied ventricles during cardiopulmonary bypass. J Thorac Cardiovasc Surg 66:742, Mahler F, Ross J, O Rourke RA, et al: Effects of changes in preload, afterload and inotropic state on ejection and isovolumic phase measures of contractility in the conscious dog. Am J Cardiol 35:626, Mehmel H, Krayenbuehl HP, Rutishauser W: Peak measured velocity of shortening in the canine left ventricle. J Appl Physiol 29:637, Mirsky I, Ellison RC, Hugenholtz PG: Assessment of myocardial contractility in man from ventricular pressure recordings. Clin Res 17:255, Molina JE, Feiber W, Sisk A, et al: Cardioplegia without fibrillation or defibrillation in cardiac surgery. Surgery 81:619, Nejad NS, Klein MD, Mirsky I, et al: Assessment of myocardial contractility from ventricular pressure recordings. Cardiovasc Res 5:15, Pollack GH: Maximum velocity as an index of contractility in cardiac muscle. Circ Res 26:111, Powell WJ, DiBona DR, Flores J, et al: The protective effect of hyperosmotic mannitol in myocardial ischemia and necrosis. Circulation 54:603, Quinones MA, Goasch WH, Alexander JK: Influence of acute changes in preload, afterload, contractile state and heart rate on ejection and isovolumic indices of myocardial contractility in man. Circulation 53:293, Roe BB, Hutchinson JC, Fishman NH, et al: Myocardial protection with cold, ischemic, potassium-induced cardioplegia. J Thorac Cardiovasc Surg 73:366, Sonnenblick EH: Force-velocity relations in mammalian heart muscle. Am J Physiol 202:931, Tyers GFO, Manley NJ, Williams EH, et al: Preliminary clinical experience with isotonic hypothermic potassium-induced arrest. J Thorac Cardiovasc Surg 74:674, Wakabayashi A, Mihranian M, Guilmette JE, et al: Functional evaluation of normothermic intermittent coronary perfusion: experimental study. J Thorac Cardiovasc Surg 75:414, 1978 Discussion DR. ROBERT w. M. FRATER (New York, NY): The necessity for controlled trials is very obvious, and what inhibits most of us from performing them is the difficulty of obtaining good data on myocardial contractile performance in or out of the operating room. We have used two simple methods to try to get these data. The first is using intraoperative echocardiograms to look at left ventricular measurement changes and rate of measurement changes. It is very easy to do, and we used it, in fact, to decide whether it would be reasonable to conduct a randomized study of cold hyperkalemic cardioplegia. We measured diastolic dimension changes, systolic ejection time, and rate of circumferential fiber shortening in 15 patients and found there was no statistical difference before and after performing bypass with the aid of hyperkalemic hypothermic cardioplegia. Subsequently we have used the technique in a randomized prospective study. The second technique is a simple old-fashioned Starling curve. We did a prospective randomized study of 47 patients. Hyperkalemic cardioplegia was used in not quite half of them and intermittent aortic clamping at 30 C with the heart otherwise beating for the remaining patients. Looking at stroke volume indices in relation to at least three left ventricular filling pressures before and after bypass, we found that 70% had the same or an improved Starling curve immediately after bypass; with the intermittent ischemia technique, only 36% showed the same or an improved curve. With the Starling curves we have not seen quite as dramatic a difference as did Dr. Behrendt s group using ventricular contractile velocity, since 3O0/o of the patients had, in fact, a depressed Starling curve after cardioplegia. This was nevertheless far fewer than the 64% with reduced Starling curves using aortic clamping at 30 C. These are two simple methods. Our success with them may encourage more surgeons to do these kinds of important control studies. DR. BRUNO J. MESSMER (Aachen, West Germany): We performed similar studies during the past few months and were primarily interested in resumption of myocardial function after cardiac arrest induced by cardioplegic solution on the basis of magnesium asparaginate. Left ventricular pressure curves were recorded before the arrest and during the time coming off the pump, using a tip manometer placed into the left ventricle. Later, (dp/dt)/p and V,,, were calculated. The first group studied consisted of 6 patients who underwent a coronary artery bypass procedure with an average aortic cross-clamping time of 72 minutes. Although both factors, (dp/dt)/p and V,,,, were decreased during the period immediately after crossclamping, the contractile status improved gradually reaching almost the initial level at the end of cardiopulmonary bypass. Similar behavior of contractility factors was found in patients undergoing mitral valve replacement. Here also (dp/dt)p and V,,, were decreased immediately after retraction of the left ventricular decompression cannula into the left atrium but thereafter steadily increased to levels that were not statistically different from the initial values.

8 506 The Annals of Thoracic Surgery Vol 26 No 6 December 1978 We, therefore, can only confirm Dr. Behrendt's conclusion that cardioplegic arrest provides adequate protection of myocardial function. Cardioplegic solutions may be of benefit, if not lifesaving, for patients who run into surgeons like me who are not fast workers and still dare to say without fear that normothermic anoxic arrest will be sufficient for myocardial protection. DR. DERWARD LEPLEY, JR. (Milwaukee, WI) Last April at the meeting of the American Association of Thoracic Surgery, several papers were presented on this subject of cardioplegia, and I raised a question at that time as to the temperature of the control group compared with that of the study group who received cardioplegic solution. For those of us who have struggled with hypothermic protection through the years and know how a small temperature change can alter metabolism a great deal, it seems important that we should have this information. In the paper by Dr. Behrendt and associates, the heart was controlled at 30 C and the solution injected into the experimental group was 4 C. I cannot determine what the heart solution was in the paper by Dr. Scott and associates, and in the presentation of Dr. Lolley and colleagues the patient was apparently normothermic and the study groups were given a solution at 4 C. I am sure cardioplegic solutions do help, but I think it is very hard to determine this when another factor which has so much influence on cardiac metabolism is not controlled also, and it is extremely difficult to come to a finite conclusion in research when two variables are present. DR. BEHRENDT: I would like to thank all the discussants for their support. What we have attempted to do is add some quantitation to the qualitative observations made by all of us who have used cardioplegic solutions, namely, that patients undergoing operation with these solutions really do very well indeed. It is reassuring that electron micrographic and histochemical work recently appearing in the literature show that cardioplegia preserves the heart very well. As Dr. Levitsky pointed out, the ideal solution remains to be defined. It is clear from our data, from the data that Dr. Scott presented today, and from data obtained by Dr. Levitsky and others that hypothermia is not the only important protective factor in these solutions. The composition of the cardioplegic solution does matter. Moderate hyperkalemia itself seems to be beneficial and adds to the protective effects of hypothermia. Much of the work that we will be seeing in the next few years regarding the ideal composition of these solutions will be coming from Langendorf or working rat-heart preparations. Those of you who have not used the rat heart preparation should realize that it is very tricky and very difficult to reproduce. Any data from it regarding the ideal components for cardioplegic solutions will necessarily have to be tested in humans before they can be accepted as valid. Dr. Lepley correctly points out that the temperature of the myocardium during these experiments or operations is an important variable which must be assessed in order to interpret the results. When a liter of cardioplegic solution at 4 C is administered to an adult human, the myocardial temperature falls to 10 to 15 C rather promptly, according to our measurements. Over the next half hour it gradually rises to room temperature-about 22 C. If cardioplegic solution is not reinfused periodically, the myocardial temperature will remain in the range of 20" to 30 C. Of course, the exact temperature will be unknown unless it is monitored. So in reporting the results of these experiments it is very important to specify the myocardial temperature during the ischemia period and to note how the cardioplegic administration was carried out. A word of caution, however, is in order. These are all acute studies, not chronic studies. There is really very little controlled data on what cardioplegia does to myocardial function in the long run.