| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;107:265-270
© 1994 Mosby, Inc.
CARDIOPULMONARY BYPASS, MYOCARDIAL MANAGEMENT, AND SUPPORT TECHNIQUES |
Bordeaux, France, and Birmingham, Ala.
Received for publication Dec. 2, 1992. Accepted for publication May 24, 1993. Address for reprints: Francis Fontan, MD, Clinique Saint-Augustin 114 Ave. D'Ares, 33074 Bordeaux, France.
Abstract
Among 60 patients randomly assigned to a reperfusion pressure of 50 mm Hg or one of 75 mm Hg (30 mm Hg during the first 2 minutes in both groups) during initially hyperkalemic, controlled aortic root reperfusion after coronary artery bypass grafting, no in-hospital deaths occurred, no patient received an intraaortic balloon pump, no patient had new Q waves, and creatine kinase MB release was similar in the two groups. Median interval between the beginning of reperfusion and the return of rhythmic cardiac contractions in the 50 mm Hg group was 7 minutes; in the 75 mm Hg group, it was 5 minutes (P = 0.1). The initial reactive hyperemic response was greater in the 75 mm Hg group. There were no believable differences (P < 0.1) between the two groups in postoperative cardiac output, left and right atrial pressure, arterial blood pressure, and prevalence of catecholamine administration. (J THORAC CARDIOVASC SURG 1994;107:265-70)
The most appropriate reperfusion pressure after global myocardial ischemia has been debated. This randomized clinical trial was performed to examine the outcome differences, if any, between reperfusion pressures of 50 and 75 mm Hg during initially hyperkalemic controlled aortic root reperfusion.
PATIENTS AND METHODS
Patients and randomization.
Sixty patients scheduled for primary coronary artery bypass grafting operation with antegrade, initially warm and then cold, hyperkalemic, enriched blood cardioplegia and warm, initially hyperkalemic, enriched controlled blood aortic root reperfusion were randomly assigned to a coronary reperfusion pressure of 50 mm Hg ("group 50") or one of 75 mm Hg ("group 75"). The patients were operated on by the senior author (F. F.) between March 6, 1991, and September 27, 1991. The random assignment of all 60 patients was done with a computer pseudorandom-number generator before the beginning of the study.
Surgical techniques.
Patients were operated on through a median sternotomy. The extracorporeal circulation was established by insertion of an arterial cannula into the ascending aorta and of a single venous cannula through the right atrial appendage. A membrane oxygenator and pulsatile arterial inflow were used in all patients. The temperature of the perfusate during cardiopulmonary bypass (CPB) was 33° C, except that the perfusion was kept normothermic until after the completion of proximal anastomoses, the first step in the operation performed before aortic crossclamping. The left ventricle was vented by a catheter (silicone rubber catheter DLP-31481; DLP, Inc., Grand Rapids, Mich.) introduced through the right superior pulmonary vein and advanced to the apex of the left ventricle. The left atrial pressure was reduced to zero or slightly negative pressure by adjustment of the amount of suction applied.
Cardioplegia and initially hyperkalemic reperfusion.
Cold hyperkalemic, enriched blood was used as the cardioplegic agent (Appendix 
Table I). The cold cardioplegic and reperfusion solutions were delivered through a catheter (DLP 24009, which has a side arm for monitoring pressure) in the aortic root, which was isolated from the systemic circulation by an aortic crossclamp distal to the aortic root catheter. 3 The initial dose of cold cardioplegic solution was 450 ml/m2 body surface area and was given at an aortic root pressure of 50 mm Hg. A second dose of 75 ml/m2 (Appendix Table I) was administered after approximately 30 minutes. Because all proximal venous anastomoses were performed before aortic crossclamping, the second dose of cardioplegic solution perfused coronary arteries to which vein grafts had already been attached.
|
In all patients during the first 2 minutes of reperfusion, the reperfusion pressure was 30 mm Hg. This practice has been followed throughout our clinical experience with controlled reperfusion, 3, 4 because of early experimental work (Shangyi J, Hongxi S, Gongshong L. Unpublished data, 1987) 5, 6; random assignment of patients to low and high initial reperfusion pressures did not seem justifiable. After 30 minutes, the aortic root and therefore the coronary perfusion pressures were 50 mm Hg in the patients randomly assigned to "group 50" and 75 mm Hg in those randomly assigned to "group 75." The pressure value of 75 mm Hg was chosen because it is the usual diastolic blood pressure in adults.
Measurements.
The rate of coronary blood flow was the rate of controlled aortic root reperfusion, as continuously measured by the calibrated flowmeter of the roller pumps, which were adjusted to be just occlusive. The aortic valve was competent in all patients. The aortic root (coronary perfusion) pressure was continuously measured through the side arm of the aortic root catheter. Coronary resistance was computed by dividing the aortic root pressure by coronary blood flow, and is expressed as millimeters of mercury per milliliter per minute.
Cardiac indexes (L · min-1 · m-2) were measured by thermodilution 20 to 30 minutes after the end of CPB, 3 to 4 hours after return of the patient to the intensive care unit, and the next morning. The methods for this and for the other measurements are the same as in a previous study from Hopital Cardiologique du Haut-Leveque. 4
RESULTS
Results of random assignment.
The two groups that resulted from the random assignment were similar with regard to patient characteristics (Appendix Table II) and surgical variables. One or more arterial grafts were used in 30 (100%) of the patients in "group 50" and in 28 (93%) of those in "group 75" (P[
2] = 0.15); the left internal mammary (internal thoracic) artery was used in 28 "group 50" patients (93%) and 26 "group 75" patients (87%, (P[2] = 0.4) and the right internal mammary artery was used in six "group 50" (20%) and eight "group 75" patients (27%) (P[2] = 0.5). Vein graftswere used in 27 "group 50" patients (90%) and 28 "group 75" patients (93%).
|
2] = 0.6). Carbondioxide tensions were 22 ± 2.3 mm Hg in "group 50" and 20 ± 1.1 mm Hg in "group 75" (P[2] = 0.3), oxygentensions were 139 ± 6.2 and 141 ± 5.6 mm Hg (P[2] = 0.8), hemoglobinconcentrations were 7.7 ± 0.23 and 7.7 ± 0.20 gm · dl-1(P[2] = 0.9), sodiumconcentrations were 120 ± 1.4 and 118 ± 1.3 mmol · L-1(P[2] = 0.3), potassiumconcentrations were 11.8 ± 1.19 and 12.2 ± 1.22 mmol · L-1(P[2] = 0.8), ionizedcalcium concentrations were 0.95 ± 0.026 and 0.94 ± 0.019 mmol · L-1(P[2] = 0.7), and glucoselevels were 55 ± 2.7 and 54 ± 2.4 mmol · L-1(P[2] = 0.6). The median durations of the single period of global myocardial ischemia were 43 in "group 50" and 42 minutes in "group 75"; the 10th, 25th, 75th, and 90th percentiles were 27, 34, 51, and 58 minutes in "group 50" and 28, 34, 54, and 60 minutes in "group 75" (P[Wilcoxon] = 0.8).
Deaths and cardiac events.
No in-hospital deaths occurred in either group. No patient received an intraaortic balloon pump. Catecholamines were administered to one patient in "group 50" and two in "group 75" (P[2] = 0.6). Cardiac index was slightly higher in "group 75" in the operating room 30 minutes after repair and slightly higher in "group 50" the next morning, but all differences could have resulted from chance alone (Table I).
|
At the same three times, values in the two groups were similar (P[t test] > 0.1) for arterial pH (except in the operating room, when it was 7.41 ± 0.010 in "group 75" and 7.43 ± 0.012 in "group 50," P[t test] = 0.13), arterial carbon dioxide tension (except in the operating room, when it was 41 ± 1.2 mm Hg in "group 75" and 37 ± 0.9 in "group 50," P[t test] = 0.006), arterial oxygen tension, hematocrit value, and mixed venous (right atrial) oxygen saturation and tension.
Creatine kinase MB isoenzyme release was similar in the two groups, as evidenced by similarity in the serum levels (Fig. 1). No patients (0%, 70% confidence limits [CL] 0% to 3%) in either group had new Q waves.
|
2] = 0.6).
Coronary flow and resistance.
The 10th, 25th, 50th, 75th, and 90th percentiles of the duration of the hyperkalemic phase of reperfusion were 1.5, 2, 2.5, 3, and 4 minutes, respectively, in "group 50," and 2, 2, 2.5, 2.5, and 3 minutes, respectively, in "group 75" (P[Wilcoxon] = 0.8). During the first 2 minutes of reperfusion, coronary flow was the same (P[Wilcoxon] = 0.8) in both groups (see later Fig. 3, A).
|
|
minutes of reperfusion, but thereafter resistance became progressively higher in "group 50" and increased only mildly in "group 75" (Fig. 3, B). Because some hearts began to beat at about 5 minutes after the start of controlled aortic root reperfusion, the pressure-flow relationships during reperfusion in nonbeating (electromechanically quiescent) hearts and beating hearts could be examined separately. In "group 75" within 1.5 minutes of the start of beating, flow tended to be higher in the electromechanically quiescent hearts than in the beating hearts, but these differences could have been caused by chance alone (Fig. 4). The opposite relationship pertained in "group 50," with flow being lower in the quiescent hearts (Fig. 4).
|
Critique of the study.
The preoperative patient and surgical characteristics in the two groups were well matched by the randomization process. The patients in this study were in general patients at good risk who were undergoing elective coronary artery bypass grafting, although eight patients had unstable angina. Hospital mortality rates in general are low for such patients, and there were no deaths in this group of 60 patients or in the group of 160 patients in the previous randomized trial (0% of 220 patients, 70% CL 0% to 0.9%). 2 The study therefore does not test the efficacy of the two protocols in a group of seriously ill patients, or in a group in which many of the patients had ventricular hypertrophy or chronic heart failure.
The power of the study was inadequate for detect ing small differences in the outcome events. In this study, however, small differences would be of little clin ical importance, so this is not considered a serious flaw.
Previous studies.
Most previous studies in animal models have supported the hypothesis that myocardial edema and damage from reperfusion after global myocardial ischemia, particularly in hearts that are hypertrophied, are minimized when the reperfusion pressure is kept low (<50 mm Hg) during the first few minutes of reperfusion. 5-9 The hypothesis is that the coronary endothelial cells are particularly vulnerable to pressure damage during that period. 6 We accepted that hypothesis and considered testing it in this clinical trial to be inappropriate.
Studies in animal models have given conflicting results as to optimal reperfusion pressure after the first few minutes of reperfusion. Few previous studies have used a low reperfusion pressure initially. The use of this initial low pressure in this study may have minimized differences that would otherwise have appeared with the two different reperfusion pressures. Okamoto and colleagues 5 suggested from their studies that reperfusion pressures greater than 50 mm Hg produce important myocardial edema. Other studies in animal models have shown that low reperfusion pressures (<75 to 100 mm Hg) inadequately reperfuse the endocardial layer of the myocardium and inhomogeneously reperfuse much of the rest of the myocardium, particularly in hypertrophied ventricles. 10,11Rabinov and colleagues 11 found lesser initial reactive hyperemia with low reperfusion pressure after global myocardial ischemia, just as in this study in human beings. The relationship between coronary perfusion pressure and flow appears to be continuous but nonlinear. 10
Inferences from the study.
In the circumstances of this study, reperfusion pressures of 50 and 75 mm Hg after the first 2 minutes are safe and effective. A believable and potentially important difference between the two reperfusion pressure protocols is the shorter interval between the beginning of controlled aortic root reperfusion and the resumption of cardiac contractions in "group 75" (see Fig. 2).
Cold blood cardioplegia and initially hyperkalemic controlled aortic root reperfusion are simple to use, provide excellent surgical conditions, and in patients such of the type entered into this study and the previous study result in minimal myocardial damage and excellent cardiac performance postoperatively. 4
Implications regarding the fundamental phenomena of reperfusion.
The earlier resumption of rhythmic ventricular contractions in "group 75" (whether measured from the beginning of reperfusion or from the termination of the hyperkalemic phase) leads to the speculation that myocytes and specialized cells for conduction are more generously and homogeneously provided with oxygen and substrate and rid of metabolites with the higher perfusion pressure than with one of 50 mm Hg. This speculation is supported by the higher coronary blood flow and lower coronary resistance after the first 2 minutes in "group 75" than in "group 50" (Fig. 3). The advantages of a higher reperfusion pressure may be even greater in hypertrophied hearts and failing hearts.
No data directly relating to myocardial edema were obtained. However, the similar postoperative cardiac outputs and left atrial pressures in the two groups lead to the speculation that left ventricular distensibility, and thus myocardial water content, were similar in the two groups.
The similarity of the coronary resistance in the two groups between 2 and 4 minutes after the start of reperfusion (Fig. 3, B). Invites the speculation that autoregulation of flow by variable resistance (such as occurs normally) generally does not occur during this early phase of reperfusion after hypothermic global myocardial ischemia of about 50 minutes. The differences in coronary resistance between "group 50" and "group 75" after this time would thus result from the return of coronary autoregulation, at least to some degree. One may speculate that the considerable rise in coronary resistance seen after about 4 minutes in "group 50" may be disadvantageous to oxygen delivery and result in additional myocardial damage.
Footnotes
From Hopital Cardiologique du Haut-Leveque, University of Bordeaux, Bordeaux, France,a and the Division of Cardiothoracic Surgery, Department of Surgery, The University of Alabama Medical Center, Birmingham, Ala.b 
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
