Retrograde continuous warm blood cardioplegia: Maintenance of myocardial homeostasis in humans

Retrograde Continuous Warm Blood Cardioplegia: Maintenance of Myocardial Homeostasis in Humans Steven R. Gundry, MD, N a n Wang, MD, David Bannon, MD, Robert E. Vigesaa, MD, Clifford Eke, MD, Simon Pain, MS, and Leongrd L. Bailey, MD Division of Cardiothoracic Surgery, Department of Surgery, Lorna Linda University School of Medicine, Lorna Linda, California

Recent clinical reports have suggested that continuous delivery of oxygenated warm blood cardioplegia through the coronary veins (retrograde cardioplegia) produces good myocardial preservation during aortic crossclamping. No data exist, however, about actual myocardial metabolismhomeostasis during retrograde warm blood cardioplegia. We studied 100 consecutive patients undergoing coronary artery bypass grafting, aortic valve replacement, or both who received retrograde continuous warm blood cardioplegia (4:l dilution) during aortic cross-clamping for 54 to 174 minutes. We measured pH, oxygen tension, carbon dioxide tension, HCO,, base excess, and oxygen content of the inflow cardioplegia and the blood egressing from coronary arteries during each arteriotomy for bypass grafting (arteries act as postcapillary veins with retrograde cardioplegia) or the left and right coronary orifices during aortic valve replacement. We also measured these variables from the coronary sinus effluent 1 minute after release of the aortic cross-clamp. Retrograde cardioplegia flow ranged from 50 to 250 mL/min (mean flow, 150 mL./min). All patients were maintained at normothermia during bypass. A total

of 460 samples were analyzed (4.6 per patient). Neither the duration of aortic cross-clamping nor the artery sampled affected myocardial blood gases. The pH dropped from 7.41 f 0.05 for the inflow cardioplegia to 7.32 f 0.1 when sampled from coronary arteries, and the oxygen tension fell from 181 2 25 to 28 f 5 mm Hg, respectively. Carbon dioxide tension rose from 31.0 & 4.1 to 41.4 f 9.8 mm Hg. Coronary sinus blood gases 1 minute after cross-clamp removal showed no acidosis or oxygen debt. These results indicate that (1) myocardial homeostasis is preserved by retrograde continuous warm blood cardioplegia during normothermic aortic crossclamping for up to 3 hours in humans; (2) normal myocardial metabolism and oxygFn uptake can be maintained solely by retrograde cardioplegia in the arrested heart, eliminating oxygen debt and ischemia; and (3) unlike experimental models, the arrested human heart is very active metabolically. Further application of this new clinical technique appears warranted, provided appropriate monitoring is used.

T

cently, owing largely to improved delivery systems and the recognition of the consequences of myocardial ischemia such as reperfusion injury, retrograde continuous warm perfusion of the heart has been reintroduced into the clinical arena [5, 61. Despite several reports on the successful human application of this technique, little experimental or clinical work is available that describes the effectiveness of this technique in preventing myocardial ischemia. To understand how continuous retrograde perfusion of oxygenated blood cardioplegia might preserve myocardial viability, the interrelationship of the coronary venous and arterial anatomy must be considered (Fig 1).The coronary venous system not only connects with myocardial capillaries but also interconnects with other coronary veins by way of veno-venous collaterals and with the ventricular and atrial cavities directly through the thebesian vessels. So rich are these venous connections that only 25% of blood retroinfused into the coronary sinus traverses the capillaries, the remainder exiting the myocardium through the venous channels without providing nutrient

he ability of surgeons to preserve myocardial viability during periods of aortic cross-clamping has existed since the earliest days of clinical open-heart operations in the 1950s. Prompted by the laboratory successes of Blanco and associates [l],Lillehei and his colleagues [2] successfully operated on the aortic valve using continuous retrograde perfusion of the beating, warm human heart through a catheter placed in the coronary sinus. Even in 1954, the idea of perfusing oxygenated warm blood into the myocardium through the coronary veins to preserve the heart was not new, having been described as early as 1898 by Pratt [3]. By 1959, however, despite 4 years of successful clinical application, retrograde continuous warm perfusion of the heart was largely abandoned in favor of antegrade intermittent hypothermic perfusion or cardioplegia [4]. RePresented at the Twenty-eighth Annual Meeting of The Society of Thoracic Surgeons, Orlando, FL, Feb 55, 1992. Address reprint requests to Dr Gundry, Division of Cardiothoraac Surgery, Department of Surgery, Lorna Linda University Medical Center, 11234 Anderson St, Rrn 25628. Lorna Linda, CA 92354.

0 1993 by The Society of Thoracic Surgeons

(Ann Thorac Surg 1993;55:35863)

0003-4975/93/$6.00

GUNDRY E-r AL RETROGRADE WARM CARDIOI'LEGIA

Ann Thorac Surg 1993;55:358-63

Coronary Arteries

Venules

- -

Coronary Sinus

- -

Right Atrium

Thebesian

Left Ventricle

Right Ventricle

Fig 1 . Coronary venous and arterial drainage patterns. Dark arrows show retroperfused blood flow patterns.

blood flow [7]. The retroperfused blood cardioplegia that does traverse the capillaries, in general, uses the coronary arteries to egress from the myocardium. Thus, in continuous retrograde cardioplegia, sampling the coronary artery effluent would be expected to give a most reliable measure of myocardial oxygen extraction, acid-base status, and homeostasis. The testing of this hypothesis in 100 consecutive patients undergoing myocardial revascularization, valve replacement, or both forms the basis of this report.

Material and Methods The study comprises all adult patients having myocardial revascularization, valve replacement or repair, or a combination of these from February 1991 to the present on the service of the primary author (S.R.G.). Myocardial protection was provided by continuous retrograde infusion of warm (37°C)blood cardioplegia (4:l dilution) administered through a flexible catheter with a manually inflated distal balloon and with constant monitoring of coronary sinus pressure (Gundry RSCP catheter; DLP, Inc, Grand Rapids, MI). The technique of insertion of this catheter has been previously described [8]. Essentially, it entails placing the catheter into the coronary sinus through a pursestring suture in the high right atrium and directing the catheter using a flexible stylet. Induction of myocardial arrest was accomplished by the infusion of warm blood cardioplegia (4:l dilution) at 250 to 300 mL/min using a final KC1 concentration of 27 mEq/L into the coronary sinus RSCP catheter until arrest was achieved and then switching to continuous infusion of a 4:l blood cardioplegia mixture with a KCI concentration of 9 mEq/L at 110 to 150 mUmin. Whenever a coronary arterial blood gas analysis showed a pH of less than 7.30, cardioplegia flow was increased by 50 mL/min up to 300 mL/min. Assessment of myocardial perfusion was made visually by noting engorged red veins and blue arteries and by noting the issuance of dark blood from arteriotomies or coronary orifices; also, a low-pressure alarm was placed on the coronary sinus pressure line (DLP, Inc) and set to sound when coronary sinus pressure was less than 15 mm Hg.

359

Samples for blood gas analysis were taken from the inflow cardioplegia line and from each coronary arteriotomy at the time of bypass graft placrmcnl, and from the coronary artery ostia in the case of aortic valve replacement. They were repeated at least every half hour. At the termination of aortic cross-clamping, cardioplegia infusion was stopped and the balloon deflated. One minute later, a final blood gas sample was taken from the coronary sinus through the retrograde catheter, which was still in place, and compared with a blood gas sample taken from the arterial line from the bypass pump. All blood gas samples were analyzed for pH, oxygen tension (Po2), carbon dioxide tension (PCO~), HCO,, base excess, and oxygen content. Comparison was made between the inflow cardioplegia and the blood egressing from the coronary arteries as a measure of myocardial metabolism and between the inflow cardioplegia and the coronary sinus blood after release of the cross-clamp as a measure of oxygen debt. All patients were maintained at 37°C during sampling.

Results The results of the blood gas analyses for the first 100 patients studied are shown in Table 1. A total of 460 samples were analyzed (4.6 t 1 samples per patient). The inflow cardioplegia reflected a 4:l mixture of perfusion blood and Roe's solution, which supposedly had been buffered to a pH of 7.45 before delivery to the operating room but was consistently slightly acidotic on delivery (base excess -3.7 2 2.6 and pH 7.41 _" 0.05). Blood gases from the coronary artery effluent, representing blood that had traversed the myocardial capillaries, had a mean pH of 7.32 t 0.1 with a base excess of -4.7 2 2.6. Postcapillary pH ranged from a low of 7.20 to a high of 7.55. Coronary sinus blood pH after cross-clamp release was 7.38 f 0.1 with a base excess of -0.4 k 3.0 versus a pH of 7.42 -+ 0.08in the arterial line. The Po, of the inflow cardioplegia was 181 -t 25 mm Hg, which fell to 28 _" 5 mm Hg when measured in the coronary artery effluent. This value rose to 37 k 6 mm Hg when measured in the coronary sinus after

Table 1. Summary of Blood Gas Analyses" Variable

PH

Po2 (mm Hg) Pco, (mm Hg) HCO, Base excess

0, content

Cardioplegia

Coronary Artery Effluent

Coronary Sinus 1 Minute After ACC Release

7.41 181

+ 0.05 + 25

7.32 -+ 0.1 28 -t 5

7.38 2 0.1 37 + 6

31.0

+ 4.1

41.4

9.8

40.5 2 4.5

20.5 -t 1.8 -4.7 -t 2.6 4 2 2

23.1 2 1.9 -0.4 2 3.0

19.3 -t 1.5 -3.7 ? 2.6 7-t2

2

5 + 2 ~

Data are shown as the mean

f

the standard error of the mean.

ACC = aortic cross-clamp; 0, = oxygen; Po, = oxygen tension. tension;

Pcoz = carbon dioxide

360

GUNDRY E T AL RETROGRADE WARM CARDIOPLEGIA

cross-clamp release versus 180 -+ 21 mm Hg systemically. The Pco, of the inflow cardioplegia was 31.0 f 4.1 mm Hg and rose to 41.4 -1- 9.8 mm Hg in the coronary artery samples, similar to the value of 40.5 & 4.5 mm Hg measured in the coronary sinus later. Little change in bicarbonate ion was detected, which increased from 19.3 f 1.5 (inflow) to 20.5 2 1.8 (postcapillary outflow) and measured 23.1 f 1.9 in the coronary sinus after unclamping. Oxygen content of the inflow cardioplegia was 7 f 2, which decreased to 4 2 2 in the coronary artery effluent, a value similar to that for the coronary sinus blood. Cross-clamp times ranged from 54 to 174 minutes. No difference in blood gases could be detected based on duration of cross-clamping. Nor could a relationship be found in myocardial blood gases based on the region of the heart served by a coronary artery, although the right coronary and posterior descending coronary arteries had slightly lower pHs and Po,s than other regions. In several hearts, notably those with hypertrophy, postcapillary blood pH appeared to correlate directly with cardioplegia flow. As an example, take the case of a 76-year-old man with left ventricular hypertrophy who was undergoing a bypass procedure. The initial blood gas analysis of the inflow cardioplegia showed a pH of 7.40, Pco, of 31 mm Hg, Po, of 176 mm Hg, base excess of -4.5, and oxygen content of 7.3. At a flow of 125mL/min, the initial left anterior descending coronary artery blood had a pH of 7.20, Pco, of 53 mm Hg, Po, of 45 mm Hg, base excess of -6.7, and oxygen content of 4.9. When the inflow cardioplegia rate was increased to 180 mL/min, the subsequent blood gas values in that artery changed to a pH of 7.36, Pco, of 33 mm Hg, Po, of 35 mm Hg, base excess of -5.6, and oxygen content of 4.6, thus indicating normalization of pH and Pco, but a still high oxygen extraction. When the pH of the coronary arterial blood gas fell to less than 2.30, flows were increased up to 300 mL/min until a pH of greater than 7.30 was obtained. Increased flows were required in 48 of the 100 patients.

Comment Retrograde continuous warm (or normothermic) cardioplegia for myocardial protection during aortic crossclamping differs little from that used by Lillehei and co-workers (21 nearly 40 years ago with the exception that potassium is now added to the perfusate to arrest the heart. This addition was made secondary to the experimental findings that most of the energy savings afforded the heart by cardioplegia come from mechanical arrest and that little additional benefit is gained by cooling [9]. Extrapolating from these experimental findings, Salern0 and co-workers [6] and Lichtenstein and associates [lo] clinically showed that good myocardial protection could be achieved by continuous perfusion of the heart with potassium blood cardioplegia in either an antegrade or retrograde fashion. The former, because of its complicated delivery system and the need to shut off cardioplegia or snare coronary arteries to construct anastomoses, was and remains clinically difficult to apply. However,

Ann Thorac Surg 1993;55:356-63

since our introduction in 1987 of transatrial insertion of coronary sinus catheters, facile retrograde delivery of cardioplegia makes retrograde warm cardioplegia an everyday clinical reality [8]. Whether retrograde continuous warm cardioplegia allows normal myocardial homeostasis and prevents myocardial ischemia in humans was not known heretofore. The degree of metabolic activity of the warm, arrested human heart was also unknown, despite conjecture based on animal experiments. Last, the effect of using retrograde warm cardioplegia solely without an initial antegrade dose was unknown. The results of our study clearly indicate that the human heart is fully capable of using blood perfused retrogradely through the coronary veins to maintain myocardial homeostasis. This, in itself, is not surprising, as retrograde perfusion of oxygenated blood without cardioplegia was shown by Lillehei and colleagues [Z] in 1956 to maintain myocardial viability during short periods of aortic crossclamping. Hammond and associates [ll]demonstrated in dogs similar myocardial "venous" Po+ to those of our patients during 1 hour of retrograde coronary sinus perfusion at 37°C in the beating heart, but they noted better functional recovery, particularly of the right ventricle, if inflow perfusion temperature was lowered to 26°C. Our study demonstrates that blood gas measurements of blood egressing from the heart through the coronary arteries closely approximates normal coronary sinus blood gas values. What is perhaps surprising is the high degree of oxygen extraction and therefore metabolism of the arrested human heart. This is contrary to reports of myocardial metabolism in animals by Buckberg [12], Bernard [9], and their co-workers that suggest that the normothermic, arrested heart requires about 90% less oxygen than the beating, working heart. Because coronary blood flow is approximately 4% of total cardiac output, normal coronary blood flow usually is 200 to 300 mL/min. Thus, a warm, arrested heart should require only 10% of this flow or 20 to 30 mWmin to maintain homeostasis. As only 25% of retrograde perfusion through the coronary veins traverses the myocardial capillary bed, it would be proper to suppose that four times as much flow would be needed retrogradely to maintain cellular needs, or 80 to 120 mL/min. Although in many patients this does appear to be the case, our study uncovered many patients, particularly those with left ventricular hypertrophy, who required 250 to 300 mWmin to maintain homeostasis and to prevent myocardial ischemia. Our study suggests that the degree of metabolic activity of some human hearts, unlike their experimental animal counterparts, may be quite high and must be met. This indicates that retrograde continuous warm cardioplegic protection of the myocardium is not guaranteed, but rather its effectiveness must be monitored. In this study, direct sampling of postcapillary blood gases appeared to be a simple and effective method of accomplishing this goal. Other simple methods, however, such as continuously monitoring myocardial Po,, usually cannot detect adequate myocardial delivery of substrate but merely detect the presence or absence of infused cardioplegia, in

GUNDRY ET AL RETROGRADE WARM CARDIOPLEGIA

Ann Thorac Surg 1993;55358-63

the same way as the low-pressure alarm on the coronary sinus transducer in our study. Last, our study demonstrates that myocardial energy needs can be met entirely by retrograde delivery of warm cardioplegia without the need for antegrade cardioplegia. As we [S] have previously shown with retrograde intermittent cold cardioplegia, antegrade cardioplegia imposes an unnecessary step and complicates an otherwise simple procedure. Moreover, it stretches the imagination a bit to believe that in a 2-hour cross-clamp period, a 2- to 3-minute dose of antegrade cardioplegia would have much influence on myocardial protection. Also, in redo operations, antegrade cardioplegia may impose dangers of its own in embolizing atheromatous debris from old vein grafts [13]. In conclusion, based on blood gas analysis of inflow and outflow cardioplegia samples in 100 patients, we have shown that retrograde continuous warm cardioplegia is capable of maintaining myocardial homeostasis and preventing myocardial ischemia in humans for up to 3 hours of aortic cross-clamping. Further, we have demonstrated variability in the metabolic needs of the warm, arrested human heart that sometimes seem at odds with and are greater in amount than experimental literature might suggest. During retrograde warm cardioplegia, the surgeon must ensure that the delivery of cardioplegia blood flow and volumes are sufficient to meet myocardial energy needs. In this study, this was easily achieved by sampling blood gases from arteriotomies. Finally, this study establishes that retrograde warm cardioplegia does not need an antecedent antegrade dose of cardioplegia to provide myocardial protection. Much basic clinical research must be completed to understand the nature of retrograde delivery of cardioplegia to the capillary level and how to consistently measure the adequacy of this delivery. Although this study demonstrates the effectiveness of the technique of retrograde warm cardioplegia, it also imposes on the surgeon the necessity to monitor and confirm the adequacy of myocardial protection. At this stage in its development, we recommend continuous monitoring of coronary sinus

361

pressure, a standby low-pressure alarm, and intermittent sampling of coronary artery effluent for pH, Pco,, and Po,, with adjustment of retrograde flows accordingly.

References 1. Blanco G, Adam A, Fernandez A. A direct experimental approach to the aortic valve. 11. Acute retroperfusion of the coronary sinus. J Thorac Surg 1956;32171-7. 2. Lillehei CW, DeWall RA, Gott VL, Varco RL. The direct vision correction of calcific aortic stenosis by means of a pump-oxygenator and retrograde coronary sinus perfusion. Dis Chest 1956;30123-32. 3. Pratt FH. The nutrition of the heart through the vessels of Thebesius and the coronary veins. Am J Physiol 1898;1:86103. 4. Shumway NE. Forward versus retrograde coronary perfusion for direct vision surgery of acquired aortic valvular disease. J Thorac Cardiovasc Surg 1959;38:7%30. 5. Lichtenstein SV, Dalati HEL, Panos A, Slutsky AS. Long cross-clamp time with warm heart surgery [Letter]. Lancet 1989;l:1443. 6. Salerno TA, Houck JP, Barrozo CAM, et al. Retrograde continuous warm blood cardioplegia: a new concept in myocardial protection. Ann Thorac Surg 1991;51:245-7. 7. Lolley DM, Hewitt RL. Myocardial distribution of asanguineous solutions retroperfused under low pressure through the coronary sinus. J Cardiovasc Surg (Torino) 1980;21:287-93. 8. Gundry SR, Sequiera A, Razzouk AM, McLaughlin IS, Bailey LL. Facile retrograde cardioplegia: transatrial cannulation of the coronary sinus. Ann Thorac Surg 1990;50:882-7. 9. Bernard WF, Schwartz HF, Malick NP. Selective hypothermic cardiac arrest in normothermic animals. Ann Surg 1961; 153:43-51. 10. Lichtenstein SV, Ashe KA, Dalate HEL, Cusimano RJ, Panos A, Slutsky AS. Warm heart surgery. J Thorac Cardiovasc Surg 1991;101:269-74. 11. Hammond GL, Davis AL, Austen WG. Retrograde coronary sinus perfusion: a method of myocardial protection in the dog during left coronary artery occlusion. Ann Surg 1967;166: 39-47. 12. Buckberg GD, Brazier JR, Nelson RL, Goldstein SM, McConnell DH, Cooper N. Studies of the effects of hypothermia on regional blood flow and metabolism during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1977;73:87-94. 13. Gundry SR, Razzouk AJ, Vigesaa RE, Wang N, Bailey LL. Optimal delivery of cardioplegic solutions for "redo" operations. J Thorac Cardiovasc Surg (in press).

DISCUSSION D R SIDNEY LEVITSKY (Boston, MA): I have a question that I think concerns many of us, and that is the problem of right heart versus left heart preservation during retrograde continuous warm blood cardioplegia. There is concern that much of the right heart coronary flow drains out through the thebesian veins, thus avoiding the coronary sinus. At New England Deaconess Hospital, we started using retrograde continuous warm blood cardioplegia in April 1991. Not infrequently, we have observed the absence of blood flow in the posterior descending coronary artery after arteriotomy in patients with a total occlusion of the right coronary artery. Nevertheless, these patients have a benign postoperative course with no evidence of right heart failure. Have you noticed any differences in oxygenation or pH, for example, between the right posterior descending or right marginal coronary vessels as opposed to the left-sided vessels?

D R G U N D R Y That is an excellent question and one that concerned us a great deal as we embarked on warm cardioplegia. We did notice a difference in samples taken from the right coronary artery or the posterior descending coronary artery in that they invariably had a lower pH and lower oxygen content. The oxygen tensions were often in the range of 10 to 15 mm Hg. However, when the data for the 100 patients were analyzed, none of these figures reached statistical significance compared with the other areas sampled in the heart. Using transesophageal echocardiography, we have looked at the function of almost all of these hearts, and much to our surprise, even with "lousy" right coronary artery blood gases, the right ventricular function of these hearts has been superb, better than I would have expected. But, yes, there seems to be a lower pH in those vessels and much more avid oxygen extraction. The flow is not as good to the right side.

362

GUNDRY ET AL RETROGRADE WARM CARDIOPLEGIA

DR GEORGE E. CIMOCHOWSKI (Wikes-Barre, PA): That was an elegant presentation, Dr Gundry. My colleagues and I also have been interested in coronary effluents, and in the past 6 months, we have studied a small series of patients. We frequently turn off the retrograde flow for a short time, about 5 to 10 minutes, and did so in each of the study patients. We never turn it off in a patient with a hypertrophied ventricle. We studied coronary effluents from the left anterior descending coronary artery. These were always measured after the last anastomosis, which in each case was the internal mammary artery. Initially we were timid; we stopped at 5 to 7 minutes and measured the pH and the Po,. The pH was quite good, 7.37. After we got a little more bold, we stopped it for 10 minutes, and our mean pH was 7.33, but there certainly was a range-7.11 to 7.52. We think that it is "probably" safe to stop retrograde flow for a short time. In our experience and probably in yours, the hypertrophied left ventricle with coronary artery disease is a flag. I say that because there is a lot of variation in blood flow retrograde in the coronary sinus system. It is not simply a matter of giving cardioplegia in the coronary sinus and seeing it come out the coronary artery. An example is our recent case of an 85-year-old woman with coronary artery disease who was having a double-valve operation. During the mitral valve replacement, we noticed clinically there was good flow coming from the aorta. However, when we did the aortic valve replacement, we did not see good flow from the coronary ostia. We even gave antegrade cardioplegia and used a catheter with a manual inflating device. At 5 mL, we did not increase flow from the coronary ostia, and the patient died of low cardiac output. We subsequently took the heart to the autopsy suite, and at 6 mL in the balloon with the manual inflating device, there was free flow back and forth into the right atrium. With the self-inflating balloon, there was massive regurgitation. Have you seen similar patients in whom the coronary sinus is so large that our equipment does not seem to cause the flow to go only retrograde? DR GUNDRY Yes, indeed. We have been impressed with the ability of the coronary sinus of a warm heart to expand. Unlike a cold heart, in which the coronary sinus and its surrounding fat tissue become quite rigid, during warm cardioplegia, the coronary sinus is really free to expand much like many other thin-walled veins. In fact, during valve replacement on several large hypertrophied hearts, we have used as much as 10 mL of air in an operator-inflated balloon to seal the coronary sinus. This is particularly important if you use transesophageal echocardiography, in that the anesthesiologist can put a Doppler probe across the coronary sinus proximal to the balloon. He or she can tell you how much air to inject into the balloon to seal the coronary sinus by looking at blood leaking around the balloon. It is often very dramatic. Your point about hypertrophied hearts is extremely well taken. We have recently submitted an abstract to the Western Thoracic Surgical Association issuing a warning about hypertrophied hearts and some redo operations. This is a very dangerous area in retrograde continuous warm cardioplegia. We have got to be very careful and know a lot more about it before we can endorse it wholeheartedly. DR CIMOCHOWSKI: We have done studies of the amount of blood flow coming out of the aorta retrogradely. As you raise the heart for the various maneuvers, by the time you get to the circumflex, you may have a 50% to 75% reduction in blood flow coming out the aorta. So all the answers are not in yet.

Ann Thorac Surg 1993;55:3W

DR GUNDRY: That is correct. Even holding the heart to press on a circumflex area will dramatidy change what comes out of that artery. Your assistant has to learn how to hold the heart in a slightly different way so as not to compress those veins. DR PAUL T. FRANTZ (Roanoke, VA): I enjoyed your presentation very much. Based on our experience, the heart can tolerate intermittent ischemia during aerobic warm arrest. This experience involves 457 patients, 361 of whom had coronary bypass operations. Our early technique was similar to that used in Toronto. In normothermic patients, we used intermittent antegrade and continuous retrograde cardioplegia while performing proximal and distal grafts with the aortic cross-clamp on. Because of technical problems associated with retrograde catheters, we began using an intermittent antegrade technique. The appearance of multiorgan failure mentioned earlier led us to switch from normothermic to hypothermic systemic perfusion. One of our current techniques is to use intermittent antegrade cardioplegia, turning the solution off for 10 to sometimes 15 minutes. The proximal anastomosis is done after each distal graft. In some patients we use intermittent antegrade and intermittent retrograde cardioplegia, and in these situations, all the proximal grafts are done after the distal grafts have been completed. There were 167 patients in whom we used the antegrade and retrograde technique. There were nine deaths and three myocardial infarctions in this group. Three patients required the balloon pump. The deaths were due to multiorgan failure (l),isolated ventricular fibrillation (2), preoperative myocardial infarction (l), sepsis (l), cerebrovascularaccident (Z), and hypertrophic cardiomyopathy (2). There was one preoperative massive myocardial infarction. An intermittent antegrade technique was used in 194 patients. In this group there were only one death and three myocardial infarctions. Four patients required the balloon pump. In our experience, the heart tolerates periods of interruption in cardioplegic flow. Based on your experience with the continuous retrograde technique, how do you explain this observation? DR GUNDRY I appreciate your comments and those of the last discussants about technique. As we are all practicing cardiac surgeons, we know that the heart can tolerate imperfection. We practice it nearly every day, unfortunately. We also know that a heart operated on 15 years ago, for example, that of a 45-year-old man undergoing placement of a couple of bypass grafts, could tolerate a cross-clamp across the aorta every 15 minutes, and the patient could come off bypass extremely easily. Unfortunately, I do not see those patients anymore; the cardiologists see them. We are talking about applying a technique of preservation to a completely different stratum of patients, patients who cannot get a heart transplant because there are no donors; these patients have an ejection fraction of 0.15. As Dr Swain said a few minutes ago, it is one thing to take an ejection fraction of 0.60 and lose a third of it; the patient would leave the operating room smiling. When you take an ejection fraction of 0.20 and lose a third of it, the results include a left ventricular assist device, balloon pump, and death. So why tolerate any myocardial ischemia in a heart that if it loses three myocytes the patient may not leave the operating room alive?Hearts can tolerate ischemia; we prove that every day. However, we are trying to develop a technique where there is absolutely no ischemia. We are not quite there yet, but we are pretty close. DR ROBERT A. GUYTON (Atlanta, GA): I have first a rebuttal to Dr Frantz. I think it is a dangerous and inappropriate leap of

Ann Thorac Surg 1993;5535643

faith to go from continuous warm cardioplegia to intermittent warm cardioplegia. I believe that is a backward step. There is abundant theoretical and experimental evidence that after 2 or 3 minutes of ischemia, myocardial damage begins. This is seen in nuclear magnetic resonance machines, in myocardial creatine kinase depletion, and in adenosine triphosphate depletion; repeated stunning of the heart causes metabolic damage. Because you can get away with something in the operating room does not mean that this is what is best for the heart in all situations, and I recommend that we not adopt intermittent warm cardioplegic techniques. Second, Dr Gundry, I question and perhaps reject your hypothesis that sampling the coronary artery samples the heart because a hypoperfused region contributes very little to the sampled blood. If most of the blood is going to myocardium that is perfused, then that is what you are going to be sampling in the coronary artery. You do not sample the maybe 15% of the heart that is getting 2% of the cardioplegia because that 2% makes u p such a small volume of what you are sampling. I think that to look at metabolism in a global way, you must sample globally and collect in a controlled situation the entire effluent. DR GUNDRY: Thank you very much. I strongly echo your statements about the intermittent warm technique. As for sampling, what we tried to d o was look at a regional sample as represented by a coronary artery and to compare that with the effluent from either a left coronary artery ostium or right coronary artery ostium. We could not find any difference between a regional area and a more global area. So I think an individual coronary artery sample can tell you what is happening in that region of interest at that point. A more global sample, such as that from a left main artery, can tell you more about what is happening to the heart as a whole. Again, these are hypotheses. We hoped we proved or suggested we proved the hypothesis, but I take your point. DR SAMUEL V. LICHTENSTEIN (Toronto, Ont, Canada): I congratulate you on an excellent study. It certainly puts some of our own concerns to rest. When Dr James Abel first introduced retrograde continuous warm blood cardioplegia in Toronto and the rest of us later adopted it, we were very concerned about the distribution of nutritive flow to the right ventricle and septum. The theory of warm heart surgery is based on the fact that electromechanical arrest reduces myocardial oxygen requirements to nearly minimal levels with little further benefit attributable to hypothermia. You have referred to the warm arrested

GUNDRY ET AL RETROGRADE WARM CARDIOPLEGIA

363

heart as metabolically active. My question is, have you looked at retrograde continuous cold cardioplegia and compared its oxygen consumption with that of retrograde continuous warm cardioplegia? DR GUNDRY Anecdotally we have been able to do this in a very small number of patients. Because the cardioplegia is so cold, there is an extremely small drop in oxygen saturation or oxygen tension when studied across that capillary area. Whether that is because the heart is not metabolically active or much less metabolically active or because hemoglobin simply would not give up oxygen because of the cold, we cannot prove in a human model; we are trying to think of a way to d o it. Nevertheless, if you look at the basis of calculation and try to look at myocardial mass and oxygen consumption per mass, experimental studies suggest that myocardial metabolism of a warm, arrested heart drops to about 10% of that of the normal working heart. If you calculate even a 25% capillary crossing of retrograde flow, that is a nutrient blood flow of 25%; in most hearts the calculations work out well; about 125 to 150 mL of flow should meet metabolic demands. But in about half of the hearts, we had to go to higher flows, contradicting what calculations from animals tell us should have been adequate to meet the heart’s demands. These flows are with the balloons inflated. So there is a lot we do not know. DR CONSTANTINE E. ANAGNOSTOPOULOS (New York, NY): My colleagues and 1 abandoned the use of isolated experimental retrograde perfusion some years ago because we observed myocardial edema. I have not heard about any methods you use at present to assess whether you have myocardial edema with a higher flow, for example, echocardiography. Furthermore, I understand that you have tried to use your samples as regional samples, but perhaps if you used pulmonary artery vents and tied the superior and inferior venae cavae, you would be able to have better representation of a global specimen. DR GUNDRY Your points and contributions are well taken. We have looked at intermittent retrograde cardioplegia and low-flow retrograde cardioplegia in an experimental model in edema, and like the Emory group, we could find no increase in myocardial edema. Apparently, based on their report earlier today, they also could not confirm myocardial edema with their continuous high-flow retrograde method. It does not seem to be a clinical concern.