- Email: [email protected]
Efficacy of coronary sinus cardioplegia in patients with complete coronary artery occlusions
Efficacy of Coronary Sinus Cardioplegia in Patients With Complete Coronary Artery Occlusions Philippe Menasch6, MD, PhD, Jean-Baptiste Subayi, MD, Line Veyssi6, MD, Olivier Le Dref, MD, Sylvie Chevret, PhD, and Armand Piwnica, MD Department of Cardiovascular Surgery, Hdpital Lariboisiere, and Department of Biostatistics, Hdpital Saint-Louis, Pans, France
Myocardial areas distal to complete coronary artery occlusions are poorly protected by antegrade cardioplegia. We assessed the effects of coronary sinus cardioplegia in 30 patients undergoing bypass operations and at high risk of cardioplegic maldistribution because of the following anatomical patterns of coronary artery disease: critical (250%) stenosis of the left main trunk with total occlusion of the right coronary artery (16 patients) or critical (270%) stenosis of the right coronary artery with total occlusion of the left anterior descending (11 patients) or circumflex artery (3 patients). After induction of arrest through the aorta, coronary sinus cardioplegia was given intermittently during the cross-clamp period at a flow rate of 100 mWmin. Intraoperatively, occluded arteries were consistently found to be filled with the retrogradely infused solution. One patient died early
postoperatively of low cardiac output and a second patient died later during his hospital stay, presumably of an arrhythmia. At autopsy, none of them had pathological evidence of inadequate myocardial protection. One patient sustained a myocardial infarction and 3 others required inotropes for more than 24 hours postoperatively. Postoperative values for right and left stroke volume indices were not significantly different from prebypass levels. Overall, these results are consistent with the occurrence of limited intraoperative ischemic damage and, by inference, suggest the efficacy of the coronary sinus route in preserving myocardial areas supplied by completely occluded coronary arteries and, hence, in jeopardy of inadequate cardioplegia delivery.
C
Material and Methods Patient Population
omplete occlusion of a coronary artery represents a challenge for adequate delivery of cold cardioplegia and hence for adequate preservation of the myocardial area subserved by the occluded vessel [l-31. In the past years, retrograde administration of cardioplegic solution through the coronary sinus has emerged as an attractive alternative to the various strategies of antegrade cardioplegic delivery [l, 41 because the coronary venous system represents an extensive and unobstructed network that is expected to behave as a very effective conduit for delivering core cooling and cardioplegic additives throughout the thickness of the myocardium. Stimulated by our favorable experience with the use of coronary sinus cardioplegia in valve operations [5] as well as by the consistent experimental demonstration that retrograde coronary sinus cardioplegia affords better protection to myocardial areas distal to coronary artery occlusions than do antegrade methods [6, 71, we undertook this observational study to assess the effects of the coronary sinus approach in a subset of patients undergoing coronary artery bypass grafting (CABG) and selected on the basis of a highly predictable risk of cardioplegic maldistribution because of multivessel disease involving at least the complete occlusion of a major coronary artery. Accepted for publication Nov 2, 1990. Address reprint requests to Dr Menaschk, Department of Cardiovascular Surgery, Hapita1 Lariboisihre, 2 rue Ambroise Pare, 75475 Paris Cedex, France.
0 1991 by The Society of Thoracic Surgeons
(Ann Thorac Surg 1991;51:418-23)
Thirty consecutive patients undergoing isolated CABG were entered into this prospective study, which was approved by the Institutional Review Board. There were 24 men and 6 women who ranged in age from 45 to 79 years (mean, 69 years). Almost one half of the study group (12 patients) was 70 years of age and older. Eighteen patients had unstable angina [8], among whom 4 were receiving intravenous nitroglycerin therapy at the time of operation. Nineteen patients (63%) had experienced remote myocardial infarctions. All preoperative coronary angiograms were reviewed by a single cardiologist (O.L.D.). Stenosis of the coronary arteries was estimated visually as the percentage of luminal cross-sectional area lost and usually was based on a composite of two projections. A serious lesion was defined as stenosis of 50% or more of the left main trunk and 70% or more for all other major coronary arteries. Based on these criteria, the patients could be classified into two groups. The first group consisted of 16 patients who had a serious stenosis of the left main trunk (9 patients) or its equivalent (7 patients) associated with a total occlusion of the right coronary artery. This occlusion involved the proximal segment of the right coronary artery in all patients but 1, in whom only the distal branches of the vessel were completely obstructed. Five of the 9 patients with left main stenosis had associated serious disease of both the left anterior descending and the circumflex 0003-4975/91/$3.50
MENASCHG ET AL CORONARY SINUS CARDIOPLEGIA
Ann Thorac Surg 1991;51:418-23
coronary arteries. A second group consisted of 14 patients who had a serious stenosis of the right coronary artery (proximal to the bifurcation in 12 patients and distal to the bifurcation in 2 patients) associated with a total occlusion of one major left-sided coronary artery. This occluded artery was the left anterior descending in 11 patients and the circumflex artery in the 3 remaining patients. All left anterior descending occlusions originated proximal to the first diagonal branch and, in 9 of the 11 patients, were associated with a serious stenosis of the circumflex artery involving the proximal segment of the vessel (before the first obtuse marginal) in 8 patients and its distal segment in the remaining patients. In the 3 patients who had circumflex occlusion, the arterial lesions were proximally located and, in all patients, were associated with a serious stenosis of the left anterior descending. Overall, triplevessel disease was present in 24 of our 30 patients (80%). For the whole group, the ejection fraction, calculated from a ventriculogram in the 30-degree right anterior oblique view, averaged 0.46 ? 0.13 (mean ? standard deviation).
Operative Technique Cardiopulmonary bypass was established between the ascending aorta and the two individually cannulated venae cavae, which were snared shortly after the onset of bypass. The left ventricle was consistently vented through the right superior pulmonary vein. After application of the aortic cross-clamp, 1 L of crystalloid cardioplegic solution at 4°C was given through the aortic root. Supplemental doses (500 mL) of cardioplegia were subsequently delivered retrogradely into the coronary sinus approximately every 25 minutes during the period of global ischemia, yielding an average of two retrograde cardioplegic infusions per patient. The technique of coronary sinus cardioplegia has already been extensively described (51. Briefly, a short (2- to 3-cm) atriotomy is made on the anterior free aspect of the right atrium, parallel to the atrioventricular sulcus. The coronary sinus is cannulated with a specific retroperfusion catheter that features a manually inflatable pearshaped balloon. Care is taken to keep the upper part of the inflated balloon seated around the intraatrial rim of the coronary ostium. The flow rate of the cardioplegia delivery system is fixed at approximately 100 mL/min, which prevents distal perfusion pressure from exceeding the safety limit of 40 mm Hg. The catheter is kept secured in the coronary sinus by means of balloon inflation throughout the arrest period. Ridges on the surface of the balloon further enhance self-maintenance of the device in the proper position. In addition to multidose cardioplegia, myocardial protection is provided by systemic cooling (25°C)and by topical hypothermia using ice slush. All proximal and distal anastomoses were constructed during a single period of aortic occlusion. On average, every patient received three bypass grafts; 4 had an additional extensive endarterectomy of the left anterior descending artery, which was patched with graft implantation. The left internal mammary artery was used in 22 patients (73%).After the last anastomosis was completed, 1 L of a modified crystalloid reperfusate at 28°C was
419
Table 1. Composition of Cardioplegic Solution and Reperfusion Solution Variable
K+ (mEq/L) Na+ (mEq/L) Mf (mEq/L) Ca2+ (mEq/L) C1- (mEq/L) Mannitol (g/L) Glutamic acid (g/L)
PH Osmolarity (mosm/L)
Cardioplegic Solution
Reperfusion
12 100 26 0.5
15 100
Solution
...
100
20 2,942 7.40 (at 20°C) 370
2.5 100 24.70 2,942 7.70 (at 28°C) 370
infused into the aortic root immediately before aortic unclamping [9]. The 5-minute period required for this last infusion was used for closing the right atriotomy with a continuous 5-0 polypropylene suture after catheter withdrawal. Because of the cardioplegic nature of the reperfusion solution, no attempts to defibrillate the heart were made until approximately 5 minutes had elapsed after removal of the aortic cross-clamp (during this time period, the heart is either asystolic or weakly fibrillating). The composition of the cardioplegic and reperfusion solutions is detailed in Table 1. The mean aortic cross-clamp time was 93 minutes (range, 66 to 135 minutes).
Assessment of Results The postoperative records of the patients were reviewed for assessment of the following clinical endpoints. Operative mortality was defined as deaths occurring during the postoperative hospital stay or within 30 days of the operation. A perioperative myocardial infarction was diagnosed by the appearance of new Q waves on the electrocardiogram and postoperative low cardiac output was defined as more than 5 pg kg-' min-' of inotropic support or intraaortic balloon support for more than 24 hours or more after bypass. Also recorded were the incidence of atrial and ventricular arrhythmias requiring medication, the time to extubation, and the cause of prolonged ventilatory support (>24 hours after bypass). In addition, hemodynamic measurements were obtained before initiation of cardiopulmonary bypass and subsequently at 1, 6, 12, and 24 hours postoperatively using a radial arterial line and a pulmonary artery balloontipped catheter (VIP 93A-831H, Baxter Healthcare Corporation, Santa Ana, CA). Measurements recorded included systemic, right atrial, pulmonary artery, and pulmonary capillary wedge pressures and cardiac output (thermodilution technique). Using standard formulas [9], values for cardiac index and right ventricular and left ventricular stroke work indices were calculated at identical intervals. Comparisons between postoperative and prebypass values were made using the nonparametric Wilcoxon and Kruskal-Wallis tests. A p value of less than 0.05 was considered significant.
-
-
420
MENASCHG ET AL CORONARY SINUS CARDIOPLEGIA
Ann Thorac Surg 1991;51:41%?3
Table 2. Postoperative Recovery Profile for Right and Left Ventricular Stroke Volume Indices" Variable EP(mmHg) PCWP(mmHg) PAP(mm Hg) AoP (mm Hg)
Before Bypass
After Bypass (h) 1
9 2 3 7 2 3 11 + - 4 11 2 4 18 f 4 18 +- 5
6
12
24
9 2 3 9 2 2 1023 11 2 5 1 2 2 4 1 2 2 3 20 ? 5 19 2 4 21 2 5b
85 f 16 83 2 15 74 f 9b 75 +- 9' 84 2 12 f3 4+-2 5 ? 4 5 f 2 6 2 3b 2 16 26 2 8 27 f 6 30 2 8 39 f 14
RVSWI(g*m/m2) 4 LVSWI (g m/m2) 33
Results are expressed as mean 2 standard deviation. p < 0.01 versus the corresponding prebypass value. p < 0.02 versus the corresponding prebypass value. = mean aortic pressure; LVSWl = left ventricular stroke work index; = mean pulmonary arte pressure; PCWP = pulmonary capillary wedge pressure; mean right atrial pressure; RVSWl = right ventricular stroke work index.
a
d=
Results There were two postoperative deaths. One patient died early in the intensive care unit of a low output syndrome that did not respond to inotropic drugs and intraaortic balloon pumping; multiple organ failure of rapid onset precluded an attempt at emergency heart transplantation. The second patient died suddenly the day before discharge, 1 week after operation. In spite of a massively dilated cardiomyopathy, which accounted for a low preoperative ejection fraction (0.25), he had had an uneventful postoperative course and presumably died of an arrhythmia. Interestingly, in neither of these 2 patients did the postmortem examination of the heart reveal left ventricular subendocardial hemorrhagic necrosis suggestive of inadequate intraoperative myocardial protection. One patient sustained a nonlethal perioperative myocardial infarction and 3 patients (in addition to the 1 who died of a low cardiac output syndrome) required inotropic drugs for more than 24 hours postoperatively (1 of these patients was weaned from inotropic support by postoperative day 2 and the 2 others, by postoperative day 3). Except for the patient who died of low cardiac output, no patient required intraaortic balloon pumping postoperatively. Atrial and ventricular arrhythmias requiring drug therapy occurred in 10 and 3 patients, respectively. Twenty-seven of the 30 patients were extubated within 24 hours of operation. One had to be reintubated because of hemodynamic deterioration that ultimately led to death (see previous paragraph). This was the only patient in whom the cause of prolonged intubation was cardiacrelated, as in the 3 patients who could not be extubated by postoperative day 1, the cause for prolonged ventilatory support was consistently pulmonary-related. The postoperative hemodynamic profile of our patient population is depicted in Table 2. At none of the study points were the postoperative values of right and left ventricular stroke work indices significantly depressed, as compared with preoperative baseline levels. This finding deserves some consideration in view of the concomitant
observation that with very few exceptions (see Table 2), the postoperative loading conditions of the two ventricles were virtually unchanged from the preoperative ones. No complications related to coronary sinus cannulation or to retroperfusion were observed in the course of the study.
Comment
Rationale for the Use of Coronary Sinus Cardioplegia in the Presence of Coronary Artery Occlusive Disease It has been demonstrated for several years that coronary artery occlusions limit the delivery of cardioplegic solution to the myocardial area distal to those occlusions [l,21. The resulting maldistribution of cooling and cardioplegia is now a well-identified cause of inadequate myocardial preservation [2, 31. To overcome this problem, several strategies of antegrade cardioplegia delivery have been developed [l,41 but their common limitation is to ensure delivery of cardioplegic flow beyond occlusions (or stenoses) only after the distal anastomoses have been completed. Further, none of these techniques avoids delayed protection of jeopardized myocardium that only becomes revascularized at the end of the cross-clamp period by means of an internal mammary graft, nor are they effective in preserving still-functional areas supplied by diseased but ungraftable vessels [4]. Additional topical cooling is undoubtedly helpful for reducing regional thermic gradients but its effectiveness is limited whenever the heart has to be lifted for completion of a right or circumflex distal anastomosis. In this context, retrograde administration of cardioplegic solution through the coronary sinus has emerged as an appealing alternative because the coronary venous system represents an extensive and atherosclerosis-free network and appears therefore well suited for the delivery of core cooling and cardioplegic agents throughout the myocardium. This assumption is primarily based on anatomical studies that have shown that a retrogradely perfused solution can distribute at the capillary level in a left ventricular area supplied by a totally occluded coronary artery [lo, 111. It is therefore not unexpected that several experimental studies [6, 71 have demonstrated that in the presence of a coronary artery obstruction, retrograde cardioplegia afforded more homogeneous distribution of cardioplegic solution and better myocardial cooling distal to the obstruction as compared with antegrade methods, including high-pressure, high-viscosity blood cardioplegia, and that these patterns of improved distribution correlated with an improved recovery of postischemic function. In contrast, the few clinical studies (12-151 that have addressed this issue have failed to demonstrate an unequivocal superiority of coronary sinus cardioplegia over the traditional techniques of antegrade cardioplegic administration. One possible explanation for these discrepant results lies in the acknowledgment by the authors of these clinical trials that their study patients were rather low-risk operative candidates, in particular with regard to the anatomical patterns of coronary artery disease. This
MENASCHE ET AL CORONARY SINUS CARDIOPLEGIA
Ann Thorac Surg 1991:51:41623
consideration is important because thermovision studies performed during CABG in humans have shown that coronary artery stenoses up to 90% did not necessarily preclude adequate cooling of jeopardized myocardium by antegradely infused cardioplegia [16]. It is therefore conceivable that the coronary artery lesions of the patients enrolled in the aforementioned studies were not the most appropriate for the potential benefits of coronary sinus cardioplegia to be readily apparent. This assumption is supported by the observation that all the experimental studies that have documented the benefits of retrograde cardioplegia have dealt with animal models of complete coronary artery occlusion (as opposed to stenosis). Consequently, this specific anatomical pattern of coronary artery disease was selected as the basic inclusion criterion for the present study.
Methodological Issues It is readily apparent that the major limitation of this study is that it does not include a control group in whom cardioplegia would have been given exclusively in an antegrade manner. The reason for this intentional flaw is the following: Our practice is to open the coronary artery to be grafted at the end of each cardioplegic infusion. In the past, we had noted that in patients with proximal total occlusion of the coronary arteries, there was no (or only minimal) cardioplegic fluid coming out of the arteriotomy when cardioplegia was given through the aortic root, thereby suggesting that reliance upon collateral circulation to deliver cardioplegic flow beyond occlusions was extremely hazardous. As we were gaining more experience with coronary sinus cardioplegia in valve operations [5], we started to expand its use to those coronary bypass patients who had total occlusion of a major coronary vessel. That such an approach was likely to be beneficial was then marshalled from the observation that the opened coronary artery was consistently found to be filled with retrogradely infused solution, thereby suggesting that the myocardial muscle distal to the occlusion was effectively reached by the cardioplegic retroperfusate. Comparison of myocardial temperatures between antegrade and retrograde cardioplegia was found to be more difficult to interpret because of the possibility for eventual between-group differences to be biased by our vigorous use of topical cooling. This issue, however, has been addressed by Shapira and associates [16]: Using temperature mapping by thermographic analysis during clinical CABG, these authors have demonstrated more uniform cooling in inflow-restricted regions with retrograde cardioplegia compared with aortic root cardioplegia (it is noteworthy that in this study, thermograms were taken before the application of topical hypothermia so that they truly reflect the myocardial cooling effects specifically exerted by each route of cardioplegia delivery). Therefore, when we selected totally occlusive disease as a prerequisite for patients to be included in the present study, the prediction that this anatomical pattern would prohibit adequate delivery of cardioplegia by the antegrade methods led us not to study such a control group. A similar line of reasoning has actually been followed by
421
Shapira and associates [17] in their study of coronary sinus cardioplegia in a patient population close to ours with respect to the pathological features of coronary arterial lesions. We do not think, however, that the observational nature of the present study brings into question some of the conclusions that can be drawn from its results.
Analysis of Results A first observation relates to the safety of the technique. No complications related to retrograde coronary sinus perfusion were observed, an observation that is shared by the other authors who have used this approach during CABG [12, 14, 151. The main technical guidelines that guarantee the safety of coronary sinus cardioplegia have been previously detailed [5] and, in brief, include positioning of the balloon at the ostium of the coronary sinus, avoidance of balloon overinflation, and maintenance of perfusion pressure below 40 mm Hg. As far as efficacy is concerned, both the clinical course and the hemodynamic recovery patterns of our patients are consistent with an overall adequate myocardial preservation during ischemic arrest. Our raw data with regard to mortality, myocardial infarction, and postoperative cardiac failure compare favorably with those reported in the literature if one keeps in mind that coronary artery occlusion, per se, is itself a predictor of increased early mortality [18], and that many of our patients cumulated additional risk factors such as advanced age, unstable angina, and impaired left ventricular function [19]. One can argue that stroke work indices that were derived from our hemodynamic measurements may not accurately reflect the intrinsic state of ventricular contractility because of their load-dependence. However, in this study, loading conditions were kept within a fairly narrow range throughout the perioperative period (see Table 2), which gives some credit to our use of these indices for comparing ventricular function preoperatively and postoperatively. We acknowledge that this assessment is limited to global ventricular performance and, as such, does not allow us to make conclusions regarding preservation of regional myocardium beyond coronary artery occlusions. Nevertheless, it is intuitively appealing to assume that the consistent occurrence of a cardioplegic flow into the distal runoff of these occluded arteries had beneficial effects on the preservation of the corresponding myocardial areas. However, it is evident that, by virtue of its design, this observational study does not allow us to conclude that the protection afforded by combined antegrade and retrograde cardioplegia is superior to what would have been obtained with the use of antegrade cardioplegia alone. Nor is our intention to persuade those content with antegrade techniques of cardioplegia delivery in patients with total coronary artery occlusions that they should switch to a combined approach. Rather, for those who are not fully satisfied with these antegrade techniques and therefore look for alternative strategies, we have attempted to document the results currently achievable when aortic root arrest is combined with intermittent
422
MENASCHE ET AL CORONARY SINUS CARDIOPLEGIA
retrograde delivery of cardioplegic solution through the coronary sinus.
Coronary Sinus Cardioplegia in Clinical Practice: Persisting Problems and Potential Solutions Like any technique, the use of coronary sinus cardioplegia raises some concerns, most of which can be easily addressed. Delay in the onset of arrest can be avoided by delivering the initial dose of cardioplegia through the aortic root [20], although the true usefulness of this antegrade cardioplegic induction can be questioned in view of the excellent clinical results reported by some groups with the exclusive use of the coronary sinus route in both valve [5] and CABG [12, 14, 171 procedures. Inadequate protection of the right ventricle does not seem to be a clinically relevant concern provided that the inflated balloon is kept at the ostium of the sinus and therefore does not impede delivery of cardioplegic flow into the terminal branches of the sinus [5]. Using radionuclide techniques, we [21] have previously shown that coronary sinus cardioplegia did not impair right ventricular function after aortic valve replacement. In the setting of bypass operations, this observation has been confirmed by Guiraudon [12], Fiore [14], and their co-workers and is further supported by the present findings that the postoperative values of right ventricular stroke work index yielded by our patients were virtually unchanged from prebypass values at fairly comparable levels of preload (reflected by mean right atrial pressures) and afterload (reflected by mean pulmonary artery pressures; see Table 2). It is likely that the discrepancy between the efficacy of coronary sinus cardioplegia in preserving right ventricular function clinically and the experimental observations made in dogs that the right ventricle receives relatively little cardioplegic solution from the coronary sinus route [lo, 11, 221 is largely due to species differences in the right ventricular venous drainage patterns [lo, 211. In clinical practice, protection of the inside of the right ventricular cavity is also expected to result from the cooling effect exerted by that fraction of the retroperfusate which shunts the capillary bed and directly escapes into the right heart chambers [lo]. It also sounds reasonable to speculate that the routine use of topical cooling further contributes to right ventricular preservation [14, 171. Cannulation of the coronary sinus under direct vision, with its attendant necessity for bicaval cannulation and snares, can be of concern for those who routinely use single venous cannulation. This problem can now be solved by the use of stylet-guided catheters, which are advanced blindly into the coronary sinus through a stab wound in the right atrial wall. In conclusion, the results obtained in a patient population at high risk of cardioplegic maldistribution suggest the occurrence of limited intraoperative ischemic damage and, by inference, the ability of a strategy of combined cardioplegia delivery (ie, induction of arrest through the aorta and subsequent maintenance of cardioplegia through the coronary sinus) to adequately preserve myocardial areas distal to complete chronic coronary artery occlusions. Additional studies are warranted to better
Ann Thorac Surg 1991;51:418-23
define the indication for use of this combined approach in the other subsets of patients who are also at high risk of inhomogeneous cardioplegic delivery, such as patients with acute coronary artery occlusion after failed balloon angioplasty or patients undergoing redo bypass operations [23].
References 1. Becker H, Vinten-Johansen J. Buckberg GD, Follette DM, Robertson JM. Critical importance of ensuring cardioplegic delivery with coronary stenoses. J Thorac Cardiovasc Surg 1981;81:507-15. 2. Hilton CJ, Teubl W, Acker M, et al. Inadequate cardioplegic protection with obstructed coronary arteries. Ann Thorac Surg 1979;28:323-34. 3. Dorsey LM, Colgan TK, Silverstein JI, Hatcher CR, Guyton RA. Alterations in regional myocardial function after heterogeneous cardioplegia. J Thorac Cardiovasc Surg 1983;86: 70-9. 4. Buckberg GD. Antegrade cardioplegia, retrograde cardioplegia, or both? Ann Thorac Surg 1988;45:589-90. 5. Menasche P, Subayi J-B, Piwnica A. Retrograde coronary sinus cardioplegia for aortic valve operations: a clinical report on 500 patients. Ann Thorac Surg 1990;49:556-64. 6. Chitwood WR. Myocardial protection by retrograde cardioplegia: coronary sinus and right atrial methods. Cardiac Surg: State Art Rev 1988;2:197-218. 7. Partington MT, Acar C, Buckberg GD, et al. Studies of retrograde cardioplegia. 11. Advantages of antegradehetrograde cardioplegia to optimize distribution in jeopardized myocardium. J Thorac Cardiovasc Surg 1989;97:613-22. 8. Scott SM, Luchi RJ, Deupree RH. Veterans Administration cooperative study for treatment of patients with unstable angina. Results in patients with abnormal left ventricular function. Circulation 1988;78(Suppl1):113-21. 9. Menasche P, Dunica S , Kural S , et al. An asanguineous reperfusion solution. An effective adjunct to cardioplegic protection in high-risk valve operations. J Thorac Cardiovasc Surg 1984;88:278-86. 10. Shiki K, Masuda M, Yonenaga K, Asou T, Tokunaga K. Myocardial distribution of retrograde flow through the coronary sinus of the excised normal canine heart. Ann Thorac Surg 1986;41:265-71. 11. Partington MT, Acar C, Buckberg GD, et al. Studies of retrograde cardioplegia. I. Capillary blood flow distribution to myocardium supplied by open and occluded arteries. J Thorac Cardiovasc Surg 1989;97605-12. 12. Guiraudon GM, Campbell CS, McLellan DG, et al. Retrograde coronary sihus versus aortic root perfusion with cold cardioplegia: randomized study of levels of cardiac enzymes in 40 patients. Circulation 1986;74(Suppl 3):105-15. 13. Walter PJ, Flameng W, Kind1 R, Podzuweit T. Preservation of myocardial energy-rich phosphates by retrograde application of Bretschneider cardioplegia during aortocoronary bypass surgery. Eur J Cardiothorac Surg 1988;2:25-30. 14. Fiore AC, Naunheim KS, Kaiser GC, et al. Coronary sinus versus aortic root perfusion with blood cardioplegia in elective myocardial revascularization. Ann Thorac Surg 1989;47 684-8. 15. Bhayana JN, Kalmbach T, Booth McLFV, Mentzer RM, Schimert G. Combined antegradehetrograde cardioplegia for myocardial protection: A clinical trial. J Thorac Cardiovasc Surg 1989;98:956-60. 16. Shapira N, Lemole GM, Spagna PM, Bonner FJ, Fernandez J,
Ann Thorac Surg 1991;51:418-23
17. 18.
19. 20.
Morse D. Antegrade and retrograde infusion of cardioplegia: assessment by thermovision. Ann Thorac Surg 1987;43:92-7. Shapira N, Turner PE, Morse D, Lemole GM. Retrograde cardioplegia infusion during coronary bypass surgery. Early clinical experience. J Cardiovasc Surg 1989;30:675-81. Brymer JF, Hannah H 111, Pugh DM, Dunn M, Reis RL. Left ventricular function and coronary obstruction as predictors of survival following aorta-coronary bypass. J Thorac Cardiovasc Surg 1976;72:73-9. Christakis GT, Ivanov J, Weisel RD, Birnbaum PL, David TE, Salerno TA. The changing pattern of coronary artery bypass surgery. Circulation 1989;8O:(Suppl 1):151-61. Kalmbach T, Bhayana JN. Cardioplegia delivery by combined
MENASCHB ET AL CORONARY SINUS CARDIOPLEGIA
423
aortic root and coronary sinus perfusion. Ann Thorac Surg 1989;47316-7. 21. Menasche P, Kucharski K, Mundler 0, et al. Adequate preservation of right ventricular function after coronary sinus cardioplegia. A clinical study. Circulation 1989;8O(Suppl 3): 19-24. 22. Stirling MC, McClanahan TB, Schott RJ, et al. Distribution of cardioplegic solution infused antegradely and retrogradely in normal canine hearts. J Thorac Cardiovasc Surg 1989;98: 1066-76. 23. Snyder HE, Smithwick W 111, Wingard JT, Agnew RC. Retrograde coronary sinus perfusion. Ann Thorac Surg 1988; 46~389-90.
Myocardial areas distal to complete coronary artery occlusions are poorly protected by antegrade cardioplegia. We assessed the effects of coronary sinus cardioplegia in 30 patients undergoing bypass operations and at high risk of cardioplegic maldistribution because of the following anatomical patterns of coronary artery disease: critical (250%) stenosis of the left main trunk with total occlusion of the right coronary artery (16 patients) or critical (270%) stenosis of the right coronary artery with total occlusion of the left anterior descending (11 patients) or circumflex artery (3 patients). After induction of arrest through the aorta, coronary sinus cardioplegia was given intermittently during the cross-clamp period at a flow rate of 100 mWmin. Intraoperatively, occluded arteries were consistently found to be filled with the retrogradely infused solution. One patient died early
postoperatively of low cardiac output and a second patient died later during his hospital stay, presumably of an arrhythmia. At autopsy, none of them had pathological evidence of inadequate myocardial protection. One patient sustained a myocardial infarction and 3 others required inotropes for more than 24 hours postoperatively. Postoperative values for right and left stroke volume indices were not significantly different from prebypass levels. Overall, these results are consistent with the occurrence of limited intraoperative ischemic damage and, by inference, suggest the efficacy of the coronary sinus route in preserving myocardial areas supplied by completely occluded coronary arteries and, hence, in jeopardy of inadequate cardioplegia delivery.
C
Material and Methods Patient Population
omplete occlusion of a coronary artery represents a challenge for adequate delivery of cold cardioplegia and hence for adequate preservation of the myocardial area subserved by the occluded vessel [l-31. In the past years, retrograde administration of cardioplegic solution through the coronary sinus has emerged as an attractive alternative to the various strategies of antegrade cardioplegic delivery [l, 41 because the coronary venous system represents an extensive and unobstructed network that is expected to behave as a very effective conduit for delivering core cooling and cardioplegic additives throughout the thickness of the myocardium. Stimulated by our favorable experience with the use of coronary sinus cardioplegia in valve operations [5] as well as by the consistent experimental demonstration that retrograde coronary sinus cardioplegia affords better protection to myocardial areas distal to coronary artery occlusions than do antegrade methods [6, 71, we undertook this observational study to assess the effects of the coronary sinus approach in a subset of patients undergoing coronary artery bypass grafting (CABG) and selected on the basis of a highly predictable risk of cardioplegic maldistribution because of multivessel disease involving at least the complete occlusion of a major coronary artery. Accepted for publication Nov 2, 1990. Address reprint requests to Dr Menaschk, Department of Cardiovascular Surgery, Hapita1 Lariboisihre, 2 rue Ambroise Pare, 75475 Paris Cedex, France.
0 1991 by The Society of Thoracic Surgeons
(Ann Thorac Surg 1991;51:418-23)
Thirty consecutive patients undergoing isolated CABG were entered into this prospective study, which was approved by the Institutional Review Board. There were 24 men and 6 women who ranged in age from 45 to 79 years (mean, 69 years). Almost one half of the study group (12 patients) was 70 years of age and older. Eighteen patients had unstable angina [8], among whom 4 were receiving intravenous nitroglycerin therapy at the time of operation. Nineteen patients (63%) had experienced remote myocardial infarctions. All preoperative coronary angiograms were reviewed by a single cardiologist (O.L.D.). Stenosis of the coronary arteries was estimated visually as the percentage of luminal cross-sectional area lost and usually was based on a composite of two projections. A serious lesion was defined as stenosis of 50% or more of the left main trunk and 70% or more for all other major coronary arteries. Based on these criteria, the patients could be classified into two groups. The first group consisted of 16 patients who had a serious stenosis of the left main trunk (9 patients) or its equivalent (7 patients) associated with a total occlusion of the right coronary artery. This occlusion involved the proximal segment of the right coronary artery in all patients but 1, in whom only the distal branches of the vessel were completely obstructed. Five of the 9 patients with left main stenosis had associated serious disease of both the left anterior descending and the circumflex 0003-4975/91/$3.50
MENASCHG ET AL CORONARY SINUS CARDIOPLEGIA
Ann Thorac Surg 1991;51:418-23
coronary arteries. A second group consisted of 14 patients who had a serious stenosis of the right coronary artery (proximal to the bifurcation in 12 patients and distal to the bifurcation in 2 patients) associated with a total occlusion of one major left-sided coronary artery. This occluded artery was the left anterior descending in 11 patients and the circumflex artery in the 3 remaining patients. All left anterior descending occlusions originated proximal to the first diagonal branch and, in 9 of the 11 patients, were associated with a serious stenosis of the circumflex artery involving the proximal segment of the vessel (before the first obtuse marginal) in 8 patients and its distal segment in the remaining patients. In the 3 patients who had circumflex occlusion, the arterial lesions were proximally located and, in all patients, were associated with a serious stenosis of the left anterior descending. Overall, triplevessel disease was present in 24 of our 30 patients (80%). For the whole group, the ejection fraction, calculated from a ventriculogram in the 30-degree right anterior oblique view, averaged 0.46 ? 0.13 (mean ? standard deviation).
Operative Technique Cardiopulmonary bypass was established between the ascending aorta and the two individually cannulated venae cavae, which were snared shortly after the onset of bypass. The left ventricle was consistently vented through the right superior pulmonary vein. After application of the aortic cross-clamp, 1 L of crystalloid cardioplegic solution at 4°C was given through the aortic root. Supplemental doses (500 mL) of cardioplegia were subsequently delivered retrogradely into the coronary sinus approximately every 25 minutes during the period of global ischemia, yielding an average of two retrograde cardioplegic infusions per patient. The technique of coronary sinus cardioplegia has already been extensively described (51. Briefly, a short (2- to 3-cm) atriotomy is made on the anterior free aspect of the right atrium, parallel to the atrioventricular sulcus. The coronary sinus is cannulated with a specific retroperfusion catheter that features a manually inflatable pearshaped balloon. Care is taken to keep the upper part of the inflated balloon seated around the intraatrial rim of the coronary ostium. The flow rate of the cardioplegia delivery system is fixed at approximately 100 mL/min, which prevents distal perfusion pressure from exceeding the safety limit of 40 mm Hg. The catheter is kept secured in the coronary sinus by means of balloon inflation throughout the arrest period. Ridges on the surface of the balloon further enhance self-maintenance of the device in the proper position. In addition to multidose cardioplegia, myocardial protection is provided by systemic cooling (25°C)and by topical hypothermia using ice slush. All proximal and distal anastomoses were constructed during a single period of aortic occlusion. On average, every patient received three bypass grafts; 4 had an additional extensive endarterectomy of the left anterior descending artery, which was patched with graft implantation. The left internal mammary artery was used in 22 patients (73%).After the last anastomosis was completed, 1 L of a modified crystalloid reperfusate at 28°C was
419
Table 1. Composition of Cardioplegic Solution and Reperfusion Solution Variable
K+ (mEq/L) Na+ (mEq/L) Mf (mEq/L) Ca2+ (mEq/L) C1- (mEq/L) Mannitol (g/L) Glutamic acid (g/L)
PH Osmolarity (mosm/L)
Cardioplegic Solution
Reperfusion
12 100 26 0.5
15 100
Solution
...
100
20 2,942 7.40 (at 20°C) 370
2.5 100 24.70 2,942 7.70 (at 28°C) 370
infused into the aortic root immediately before aortic unclamping [9]. The 5-minute period required for this last infusion was used for closing the right atriotomy with a continuous 5-0 polypropylene suture after catheter withdrawal. Because of the cardioplegic nature of the reperfusion solution, no attempts to defibrillate the heart were made until approximately 5 minutes had elapsed after removal of the aortic cross-clamp (during this time period, the heart is either asystolic or weakly fibrillating). The composition of the cardioplegic and reperfusion solutions is detailed in Table 1. The mean aortic cross-clamp time was 93 minutes (range, 66 to 135 minutes).
Assessment of Results The postoperative records of the patients were reviewed for assessment of the following clinical endpoints. Operative mortality was defined as deaths occurring during the postoperative hospital stay or within 30 days of the operation. A perioperative myocardial infarction was diagnosed by the appearance of new Q waves on the electrocardiogram and postoperative low cardiac output was defined as more than 5 pg kg-' min-' of inotropic support or intraaortic balloon support for more than 24 hours or more after bypass. Also recorded were the incidence of atrial and ventricular arrhythmias requiring medication, the time to extubation, and the cause of prolonged ventilatory support (>24 hours after bypass). In addition, hemodynamic measurements were obtained before initiation of cardiopulmonary bypass and subsequently at 1, 6, 12, and 24 hours postoperatively using a radial arterial line and a pulmonary artery balloontipped catheter (VIP 93A-831H, Baxter Healthcare Corporation, Santa Ana, CA). Measurements recorded included systemic, right atrial, pulmonary artery, and pulmonary capillary wedge pressures and cardiac output (thermodilution technique). Using standard formulas [9], values for cardiac index and right ventricular and left ventricular stroke work indices were calculated at identical intervals. Comparisons between postoperative and prebypass values were made using the nonparametric Wilcoxon and Kruskal-Wallis tests. A p value of less than 0.05 was considered significant.
-
-
420
MENASCHG ET AL CORONARY SINUS CARDIOPLEGIA
Ann Thorac Surg 1991;51:41%?3
Table 2. Postoperative Recovery Profile for Right and Left Ventricular Stroke Volume Indices" Variable EP(mmHg) PCWP(mmHg) PAP(mm Hg) AoP (mm Hg)
Before Bypass
After Bypass (h) 1
9 2 3 7 2 3 11 + - 4 11 2 4 18 f 4 18 +- 5
6
12
24
9 2 3 9 2 2 1023 11 2 5 1 2 2 4 1 2 2 3 20 ? 5 19 2 4 21 2 5b
85 f 16 83 2 15 74 f 9b 75 +- 9' 84 2 12 f3 4+-2 5 ? 4 5 f 2 6 2 3b 2 16 26 2 8 27 f 6 30 2 8 39 f 14
RVSWI(g*m/m2) 4 LVSWI (g m/m2) 33
Results are expressed as mean 2 standard deviation. p < 0.01 versus the corresponding prebypass value. p < 0.02 versus the corresponding prebypass value. = mean aortic pressure; LVSWl = left ventricular stroke work index; = mean pulmonary arte pressure; PCWP = pulmonary capillary wedge pressure; mean right atrial pressure; RVSWl = right ventricular stroke work index.
a
d=
Results There were two postoperative deaths. One patient died early in the intensive care unit of a low output syndrome that did not respond to inotropic drugs and intraaortic balloon pumping; multiple organ failure of rapid onset precluded an attempt at emergency heart transplantation. The second patient died suddenly the day before discharge, 1 week after operation. In spite of a massively dilated cardiomyopathy, which accounted for a low preoperative ejection fraction (0.25), he had had an uneventful postoperative course and presumably died of an arrhythmia. Interestingly, in neither of these 2 patients did the postmortem examination of the heart reveal left ventricular subendocardial hemorrhagic necrosis suggestive of inadequate intraoperative myocardial protection. One patient sustained a nonlethal perioperative myocardial infarction and 3 patients (in addition to the 1 who died of a low cardiac output syndrome) required inotropic drugs for more than 24 hours postoperatively (1 of these patients was weaned from inotropic support by postoperative day 2 and the 2 others, by postoperative day 3). Except for the patient who died of low cardiac output, no patient required intraaortic balloon pumping postoperatively. Atrial and ventricular arrhythmias requiring drug therapy occurred in 10 and 3 patients, respectively. Twenty-seven of the 30 patients were extubated within 24 hours of operation. One had to be reintubated because of hemodynamic deterioration that ultimately led to death (see previous paragraph). This was the only patient in whom the cause of prolonged intubation was cardiacrelated, as in the 3 patients who could not be extubated by postoperative day 1, the cause for prolonged ventilatory support was consistently pulmonary-related. The postoperative hemodynamic profile of our patient population is depicted in Table 2. At none of the study points were the postoperative values of right and left ventricular stroke work indices significantly depressed, as compared with preoperative baseline levels. This finding deserves some consideration in view of the concomitant
observation that with very few exceptions (see Table 2), the postoperative loading conditions of the two ventricles were virtually unchanged from the preoperative ones. No complications related to coronary sinus cannulation or to retroperfusion were observed in the course of the study.
Comment
Rationale for the Use of Coronary Sinus Cardioplegia in the Presence of Coronary Artery Occlusive Disease It has been demonstrated for several years that coronary artery occlusions limit the delivery of cardioplegic solution to the myocardial area distal to those occlusions [l,21. The resulting maldistribution of cooling and cardioplegia is now a well-identified cause of inadequate myocardial preservation [2, 31. To overcome this problem, several strategies of antegrade cardioplegia delivery have been developed [l,41 but their common limitation is to ensure delivery of cardioplegic flow beyond occlusions (or stenoses) only after the distal anastomoses have been completed. Further, none of these techniques avoids delayed protection of jeopardized myocardium that only becomes revascularized at the end of the cross-clamp period by means of an internal mammary graft, nor are they effective in preserving still-functional areas supplied by diseased but ungraftable vessels [4]. Additional topical cooling is undoubtedly helpful for reducing regional thermic gradients but its effectiveness is limited whenever the heart has to be lifted for completion of a right or circumflex distal anastomosis. In this context, retrograde administration of cardioplegic solution through the coronary sinus has emerged as an appealing alternative because the coronary venous system represents an extensive and atherosclerosis-free network and appears therefore well suited for the delivery of core cooling and cardioplegic agents throughout the myocardium. This assumption is primarily based on anatomical studies that have shown that a retrogradely perfused solution can distribute at the capillary level in a left ventricular area supplied by a totally occluded coronary artery [lo, 111. It is therefore not unexpected that several experimental studies [6, 71 have demonstrated that in the presence of a coronary artery obstruction, retrograde cardioplegia afforded more homogeneous distribution of cardioplegic solution and better myocardial cooling distal to the obstruction as compared with antegrade methods, including high-pressure, high-viscosity blood cardioplegia, and that these patterns of improved distribution correlated with an improved recovery of postischemic function. In contrast, the few clinical studies (12-151 that have addressed this issue have failed to demonstrate an unequivocal superiority of coronary sinus cardioplegia over the traditional techniques of antegrade cardioplegic administration. One possible explanation for these discrepant results lies in the acknowledgment by the authors of these clinical trials that their study patients were rather low-risk operative candidates, in particular with regard to the anatomical patterns of coronary artery disease. This
MENASCHE ET AL CORONARY SINUS CARDIOPLEGIA
Ann Thorac Surg 1991:51:41623
consideration is important because thermovision studies performed during CABG in humans have shown that coronary artery stenoses up to 90% did not necessarily preclude adequate cooling of jeopardized myocardium by antegradely infused cardioplegia [16]. It is therefore conceivable that the coronary artery lesions of the patients enrolled in the aforementioned studies were not the most appropriate for the potential benefits of coronary sinus cardioplegia to be readily apparent. This assumption is supported by the observation that all the experimental studies that have documented the benefits of retrograde cardioplegia have dealt with animal models of complete coronary artery occlusion (as opposed to stenosis). Consequently, this specific anatomical pattern of coronary artery disease was selected as the basic inclusion criterion for the present study.
Methodological Issues It is readily apparent that the major limitation of this study is that it does not include a control group in whom cardioplegia would have been given exclusively in an antegrade manner. The reason for this intentional flaw is the following: Our practice is to open the coronary artery to be grafted at the end of each cardioplegic infusion. In the past, we had noted that in patients with proximal total occlusion of the coronary arteries, there was no (or only minimal) cardioplegic fluid coming out of the arteriotomy when cardioplegia was given through the aortic root, thereby suggesting that reliance upon collateral circulation to deliver cardioplegic flow beyond occlusions was extremely hazardous. As we were gaining more experience with coronary sinus cardioplegia in valve operations [5], we started to expand its use to those coronary bypass patients who had total occlusion of a major coronary vessel. That such an approach was likely to be beneficial was then marshalled from the observation that the opened coronary artery was consistently found to be filled with retrogradely infused solution, thereby suggesting that the myocardial muscle distal to the occlusion was effectively reached by the cardioplegic retroperfusate. Comparison of myocardial temperatures between antegrade and retrograde cardioplegia was found to be more difficult to interpret because of the possibility for eventual between-group differences to be biased by our vigorous use of topical cooling. This issue, however, has been addressed by Shapira and associates [16]: Using temperature mapping by thermographic analysis during clinical CABG, these authors have demonstrated more uniform cooling in inflow-restricted regions with retrograde cardioplegia compared with aortic root cardioplegia (it is noteworthy that in this study, thermograms were taken before the application of topical hypothermia so that they truly reflect the myocardial cooling effects specifically exerted by each route of cardioplegia delivery). Therefore, when we selected totally occlusive disease as a prerequisite for patients to be included in the present study, the prediction that this anatomical pattern would prohibit adequate delivery of cardioplegia by the antegrade methods led us not to study such a control group. A similar line of reasoning has actually been followed by
421
Shapira and associates [17] in their study of coronary sinus cardioplegia in a patient population close to ours with respect to the pathological features of coronary arterial lesions. We do not think, however, that the observational nature of the present study brings into question some of the conclusions that can be drawn from its results.
Analysis of Results A first observation relates to the safety of the technique. No complications related to retrograde coronary sinus perfusion were observed, an observation that is shared by the other authors who have used this approach during CABG [12, 14, 151. The main technical guidelines that guarantee the safety of coronary sinus cardioplegia have been previously detailed [5] and, in brief, include positioning of the balloon at the ostium of the coronary sinus, avoidance of balloon overinflation, and maintenance of perfusion pressure below 40 mm Hg. As far as efficacy is concerned, both the clinical course and the hemodynamic recovery patterns of our patients are consistent with an overall adequate myocardial preservation during ischemic arrest. Our raw data with regard to mortality, myocardial infarction, and postoperative cardiac failure compare favorably with those reported in the literature if one keeps in mind that coronary artery occlusion, per se, is itself a predictor of increased early mortality [18], and that many of our patients cumulated additional risk factors such as advanced age, unstable angina, and impaired left ventricular function [19]. One can argue that stroke work indices that were derived from our hemodynamic measurements may not accurately reflect the intrinsic state of ventricular contractility because of their load-dependence. However, in this study, loading conditions were kept within a fairly narrow range throughout the perioperative period (see Table 2), which gives some credit to our use of these indices for comparing ventricular function preoperatively and postoperatively. We acknowledge that this assessment is limited to global ventricular performance and, as such, does not allow us to make conclusions regarding preservation of regional myocardium beyond coronary artery occlusions. Nevertheless, it is intuitively appealing to assume that the consistent occurrence of a cardioplegic flow into the distal runoff of these occluded arteries had beneficial effects on the preservation of the corresponding myocardial areas. However, it is evident that, by virtue of its design, this observational study does not allow us to conclude that the protection afforded by combined antegrade and retrograde cardioplegia is superior to what would have been obtained with the use of antegrade cardioplegia alone. Nor is our intention to persuade those content with antegrade techniques of cardioplegia delivery in patients with total coronary artery occlusions that they should switch to a combined approach. Rather, for those who are not fully satisfied with these antegrade techniques and therefore look for alternative strategies, we have attempted to document the results currently achievable when aortic root arrest is combined with intermittent
422
MENASCHE ET AL CORONARY SINUS CARDIOPLEGIA
retrograde delivery of cardioplegic solution through the coronary sinus.
Coronary Sinus Cardioplegia in Clinical Practice: Persisting Problems and Potential Solutions Like any technique, the use of coronary sinus cardioplegia raises some concerns, most of which can be easily addressed. Delay in the onset of arrest can be avoided by delivering the initial dose of cardioplegia through the aortic root [20], although the true usefulness of this antegrade cardioplegic induction can be questioned in view of the excellent clinical results reported by some groups with the exclusive use of the coronary sinus route in both valve [5] and CABG [12, 14, 171 procedures. Inadequate protection of the right ventricle does not seem to be a clinically relevant concern provided that the inflated balloon is kept at the ostium of the sinus and therefore does not impede delivery of cardioplegic flow into the terminal branches of the sinus [5]. Using radionuclide techniques, we [21] have previously shown that coronary sinus cardioplegia did not impair right ventricular function after aortic valve replacement. In the setting of bypass operations, this observation has been confirmed by Guiraudon [12], Fiore [14], and their co-workers and is further supported by the present findings that the postoperative values of right ventricular stroke work index yielded by our patients were virtually unchanged from prebypass values at fairly comparable levels of preload (reflected by mean right atrial pressures) and afterload (reflected by mean pulmonary artery pressures; see Table 2). It is likely that the discrepancy between the efficacy of coronary sinus cardioplegia in preserving right ventricular function clinically and the experimental observations made in dogs that the right ventricle receives relatively little cardioplegic solution from the coronary sinus route [lo, 11, 221 is largely due to species differences in the right ventricular venous drainage patterns [lo, 211. In clinical practice, protection of the inside of the right ventricular cavity is also expected to result from the cooling effect exerted by that fraction of the retroperfusate which shunts the capillary bed and directly escapes into the right heart chambers [lo]. It also sounds reasonable to speculate that the routine use of topical cooling further contributes to right ventricular preservation [14, 171. Cannulation of the coronary sinus under direct vision, with its attendant necessity for bicaval cannulation and snares, can be of concern for those who routinely use single venous cannulation. This problem can now be solved by the use of stylet-guided catheters, which are advanced blindly into the coronary sinus through a stab wound in the right atrial wall. In conclusion, the results obtained in a patient population at high risk of cardioplegic maldistribution suggest the occurrence of limited intraoperative ischemic damage and, by inference, the ability of a strategy of combined cardioplegia delivery (ie, induction of arrest through the aorta and subsequent maintenance of cardioplegia through the coronary sinus) to adequately preserve myocardial areas distal to complete chronic coronary artery occlusions. Additional studies are warranted to better
Ann Thorac Surg 1991;51:418-23
define the indication for use of this combined approach in the other subsets of patients who are also at high risk of inhomogeneous cardioplegic delivery, such as patients with acute coronary artery occlusion after failed balloon angioplasty or patients undergoing redo bypass operations [23].
References 1. Becker H, Vinten-Johansen J. Buckberg GD, Follette DM, Robertson JM. Critical importance of ensuring cardioplegic delivery with coronary stenoses. J Thorac Cardiovasc Surg 1981;81:507-15. 2. Hilton CJ, Teubl W, Acker M, et al. Inadequate cardioplegic protection with obstructed coronary arteries. Ann Thorac Surg 1979;28:323-34. 3. Dorsey LM, Colgan TK, Silverstein JI, Hatcher CR, Guyton RA. Alterations in regional myocardial function after heterogeneous cardioplegia. J Thorac Cardiovasc Surg 1983;86: 70-9. 4. Buckberg GD. Antegrade cardioplegia, retrograde cardioplegia, or both? Ann Thorac Surg 1988;45:589-90. 5. Menasche P, Subayi J-B, Piwnica A. Retrograde coronary sinus cardioplegia for aortic valve operations: a clinical report on 500 patients. Ann Thorac Surg 1990;49:556-64. 6. Chitwood WR. Myocardial protection by retrograde cardioplegia: coronary sinus and right atrial methods. Cardiac Surg: State Art Rev 1988;2:197-218. 7. Partington MT, Acar C, Buckberg GD, et al. Studies of retrograde cardioplegia. 11. Advantages of antegradehetrograde cardioplegia to optimize distribution in jeopardized myocardium. J Thorac Cardiovasc Surg 1989;97:613-22. 8. Scott SM, Luchi RJ, Deupree RH. Veterans Administration cooperative study for treatment of patients with unstable angina. Results in patients with abnormal left ventricular function. Circulation 1988;78(Suppl1):113-21. 9. Menasche P, Dunica S , Kural S , et al. An asanguineous reperfusion solution. An effective adjunct to cardioplegic protection in high-risk valve operations. J Thorac Cardiovasc Surg 1984;88:278-86. 10. Shiki K, Masuda M, Yonenaga K, Asou T, Tokunaga K. Myocardial distribution of retrograde flow through the coronary sinus of the excised normal canine heart. Ann Thorac Surg 1986;41:265-71. 11. Partington MT, Acar C, Buckberg GD, et al. Studies of retrograde cardioplegia. I. Capillary blood flow distribution to myocardium supplied by open and occluded arteries. J Thorac Cardiovasc Surg 1989;97605-12. 12. Guiraudon GM, Campbell CS, McLellan DG, et al. Retrograde coronary sihus versus aortic root perfusion with cold cardioplegia: randomized study of levels of cardiac enzymes in 40 patients. Circulation 1986;74(Suppl 3):105-15. 13. Walter PJ, Flameng W, Kind1 R, Podzuweit T. Preservation of myocardial energy-rich phosphates by retrograde application of Bretschneider cardioplegia during aortocoronary bypass surgery. Eur J Cardiothorac Surg 1988;2:25-30. 14. Fiore AC, Naunheim KS, Kaiser GC, et al. Coronary sinus versus aortic root perfusion with blood cardioplegia in elective myocardial revascularization. Ann Thorac Surg 1989;47 684-8. 15. Bhayana JN, Kalmbach T, Booth McLFV, Mentzer RM, Schimert G. Combined antegradehetrograde cardioplegia for myocardial protection: A clinical trial. J Thorac Cardiovasc Surg 1989;98:956-60. 16. Shapira N, Lemole GM, Spagna PM, Bonner FJ, Fernandez J,
Ann Thorac Surg 1991;51:418-23
17. 18.
19. 20.
Morse D. Antegrade and retrograde infusion of cardioplegia: assessment by thermovision. Ann Thorac Surg 1987;43:92-7. Shapira N, Turner PE, Morse D, Lemole GM. Retrograde cardioplegia infusion during coronary bypass surgery. Early clinical experience. J Cardiovasc Surg 1989;30:675-81. Brymer JF, Hannah H 111, Pugh DM, Dunn M, Reis RL. Left ventricular function and coronary obstruction as predictors of survival following aorta-coronary bypass. J Thorac Cardiovasc Surg 1976;72:73-9. Christakis GT, Ivanov J, Weisel RD, Birnbaum PL, David TE, Salerno TA. The changing pattern of coronary artery bypass surgery. Circulation 1989;8O:(Suppl 1):151-61. Kalmbach T, Bhayana JN. Cardioplegia delivery by combined
MENASCHB ET AL CORONARY SINUS CARDIOPLEGIA
423
aortic root and coronary sinus perfusion. Ann Thorac Surg 1989;47316-7. 21. Menasche P, Kucharski K, Mundler 0, et al. Adequate preservation of right ventricular function after coronary sinus cardioplegia. A clinical study. Circulation 1989;8O(Suppl 3): 19-24. 22. Stirling MC, McClanahan TB, Schott RJ, et al. Distribution of cardioplegic solution infused antegradely and retrogradely in normal canine hearts. J Thorac Cardiovasc Surg 1989;98: 1066-76. 23. Snyder HE, Smithwick W 111, Wingard JT, Agnew RC. Retrograde coronary sinus perfusion. Ann Thorac Surg 1988; 46~389-90.