Ann Thorac Surg 2000;69:551-555
© 2000 The Society of Thoracic Surgeons
Original Articles
A controlled trial of substrate-enhanced, warm reperfusion ("hot shot") versus simple reperfusion
Rhian Edwards, FRCAa,
Tom Treasure, FRCSa,
Mojgen Hossein-Nia, PhDb,
Andrew Murday, FRCSa,
George H. Kantidakis, MDa,
David W. Holt, FRCPathb
a Cardiothoracic Unit, St. George’s Hospital, London, England, United Kingdom
b Analytical Unit, Department of Cardiological Sciences, St. George’s Hospital Medical School, London, England, United Kingdom
Address reprint requests to Dr Treasure, Cardiothoracic Unit, St. George’s Hospital, Blackshaw Rd, Tooting, London SW17 0QT England
e-mail: treasure{at}talfourd.u-net.com
 |
Abstract
|
|---|
Background. Modified reperfusion after aortic cross-clamping is claimed to reduce myocardial injury, thus improving postoperative myocardial performance.
Methods. We measured perioperative release of creatine kinase-MB and troponin-T in 40 patients undergoing valve replacement (combined with coronary grafts in 12 cases) to determine whether infusion of a modified reperfusate before cross-clamp removal reduced myocardial injury. Patients were randomly allocated to one of two groups with minimization for age, surgeon, operation, and ventricular function. The control group received unmodified reperfusion, while the study group received a normothermic reperfusate, enhanced with glutamate and aspartate, for 5 minutes before removal of the cross-clamp. Serial determinations of troponin-T, creatine kinase-MB isoforms, and total creatine kinase-MB activity were made up to 5 days postoperatively. Requirements for inotropic support and evidence of myocardial infarction were documented.
Results. Creatine kinase-MB activity, creatine kinase-MB isoforms, and troponin-T were not significantly different between the two groups. There were no differences in the incidence of postoperative myocardial infarction or in inotrope requirement.
Conclusions. Our study did not demonstrate any advantage in using modified reperfusion in this group of patients.
|
Introduction
|
|---|
It is believed that the process of reperfusion may set in motion a train of events that result in myocardial dysfunction, arrhythmias, microvascular damage, and possibly acceleration of cell death [1]. The Journal of Thoracic and Cardiovascular Surgery has devoted two supplements entirely to the matters of optimal reperfusion [2, 3]. We shared the interest of others in the concept of reperfusion of the myocardium with blood, enhanced with substrates of an ideal composition to replenish essential metabolites, in warm oxygenated blood, while the heart remains nonworking, arrested in diastole. This approach is known as a "hot shot." The amino acids aspartate and glutamate enter the mitochondria on reperfusion to fuel the tricarboxylic acid cycle and hasten replenishment of high-energy phosphates [4]. The theoretical basis is compelling, but the evidence was largely based on experiments in dogs [2, 5, 6] and pigs [3, 4], and although there were clinical trials showing difference [7], much of the clinical discussion relies on remembered experience and historical controls. Nevertheless, the arguments for such a strategy have had such an appeal that the "hot shot" has been adopted as a routine clinical practice by many surgeons.
Enzyme release has been used to detect myocardial damage during cardiac surgery. However, until recently, the measurement of total creatine kinase (CK-MB) was relatively insensitive, and the release of enzyme from other sources limited its value in comparing techniques of myocardial preservation. Measurements of CK-MB isoforms (MB2, the tissue isoform; and MB1, the plasma modified isoform) have been used as sensitive markers of ischemic myocardial damage as well as for early diagnosis of acute myocardial infarction [8–10], which allow detection of changes at activities within the normal range of CK-MB and are therefore extremely sensitive indices of myocardial damage. The measurement of troponin-T concentration has also been shown to be a useful marker of myocardial cell injury in a number of clinical settings, including the diagnosis of ischemic myocardial damage [11, 12].
We set out to study this reperfusion strategy in our own practice. Our coronary operations at that time were performed warm (nadir temperature 32°C) using only brief periods (5 to 10 minutes) of ischemia when required to perform the distal anastomosis [13, 14]. For valve surgery, we used cold blood cardioplegia, and it is in that group that we have investigated the benefits of modifying reperfusion rather than simply removing the cross-clamp. We have endeavored to recruit as many of the more complex cases, double valve operations, and those having combined valve and coronary surgery to look for a difference in more vulnerable cases with longer ischemic times.
|
Material and methods
|
|---|
The study was approved by the district ethical committee and written informed consent was obtained from all patients. Forty patients undergoing aortic or mitral valve replacement or both, with or without coronary revascularization, between January and August 1993, were allocated to one of two groups by minimization. Minimization is a process of double-blind computer-based allocation that ensures that likely confounding variables are equally distributed between the two groups [15, 16]. These included urgent cases but not unstable angina or catheter laboratory emergencies or catastrophes. Patients were randomized for age (under or over 65 years), operating surgeon (T.T. or A.M.), nature of operation (aortic valve replacement, mitral valve replacement, or double valve replacement + coronary artery bypass grafts), and left ventricular function (normal, impaired, or poor). Left ventricular function was not rigorously defined but is used in the process of minimization to avoid important discrepancies between groups at allocation. The anesthetic technique was standardized for all patients. After systemic heparinization (3 mg/kg) and cannulation of the heart, cardiopulmonary bypass was instituted with a D703 hollow-fiber oxygenator and a roller pump (D703; Shiley Dideco, Modena, Italy). Pump prime consisted of 2 L of Hartmann’s solution. Blood cardioplegic solution in both groups contained 1 L of blood and 20 mL of cardioplegic solution (St. Thomas’ No. 1): magnesium chloride 3.253 g, potassium chloride 1.193 g, procaine 272.8 mg, and disodium edetate 0.01%. This was infused antegradely, initially into the aortic root, supplemented by direct coronary infusion in the aortic valve cases and into vein grafts where appropriate, at a temperature of 4°C via a Gish Medical Cardioplegia System (Irvine, CA). Apical vents were used routinely by one surgeon (A.M.) but not the other (T.T.). Cardioplegia was repeated at 30-minute intervals or as required to ensure cardiac quiescence. The nasopharyngeal temperature of the patient was lowered to 28°C, and systemic blood flow was maintained at 2.4 L/m2/min.
The procedure outlined above was identical for both groups until removal of the aortic cross-clamp. In the control group, designated "A," the aortic cross-clamp was simply removed and reperfusion of the coronary arteries was unmodified, with a perfusion pressure not greater than 75 mm Hg. In the study group, designated "B," a modified reperfusing solution made exactly as specified by Buckberg [17] was given. The buffered solution of electrolytes and substrates was provided by Edward Holt (Senior Pharmacist, Huddersfield Royal Infirmary), who has prepared the induction, maintenance, and reperfusion solutions according to Buckberg’s [17] precise specifications for a group of surgeons who use these techniques consistently. One liter of the modified oxygenated blood was infused into the coronary arteries at a temperature of 37°C over 5 minutes at a pressure not greater than 75 mm Hg. We measured the pH, calcium, potassium, glucose, and osmolarity of the "hot shot" reperfusate as delivered (K+ 20 mmol/L, Ca2+ 0.5–0.8 mmol/L, pH 7.7, osmolarity 320–360 mOsm, aspartate 20.5 g/L, glutamate 22.4 g/L, glucose 73.75 g/L). The aortic clamp was then removed and cardiopulmonary bypass discontinued when the myocardium was able to support the circulation. Throughout the reperfusion stages, by whichever technique or surgeon, left ventricular distension was actively avoided. Resuscitation drugs and support were administered as clinically indicated, and any requirement for inotropic or mechanical support was noted. Hemodynamic measurements made in both groups did not show any difference between them. A 5-day electrocardiogram (ECG) was analyzed for evidence of postoperative myocardial infarction by one of the workers (R.E.) who was blinded to the patient group. Arterial or peripheral venous (ie, mixed blood) samples for CK-MB activity and isoforms and troponin-T analysis were collected: (a) before induction of anesthesia; (b) after induction; (c) immediately after removal of the aortic cross-clamp; (d) at 30, 60, 90, and 120 minutes; (e) at 4, 6, 12, 24, 48, and 72 hours; (f) after 5 days. Blood samples for CK-MB activity and isoforms were collected into tubes containing disodium-ethyl-diamine tetra-acetic acid (EDTA) and were spotted at a final concentration of 15 mmol/L. This concentration of EDTA inhibits carboxypeptidase-N-mediated isoform conversion after sample collection [18]. The blood samples were then centrifuged for 10 minutes at 2,000 rpm, and the plasma was removed for storage at -20°C until enzyme analysis. Blood samples for troponin-T concentration were collected in untreated tubes and centrifuged for 10 minutes at 2,000 rpm, and serum was removed for storage at -20°C until analysis.
Analytical methods
Total CK-MB activity was measured immunochemically using Roche’s Isomune kit (Roche Diagnostic Systems, Nutley, NJ) [19]. The MB2 and MB1 isoforms were separated by high-resolution agarose gel electrophoresis and the percentages quantified by densitometric scanning (REP; Helena Laboratories, UK Ltd., Tyne and Wear, UK) [9, 10]. The MB2/MB1 ratio was calculated by dividing the percentage of MB2 in the fractions, derived from densitometry, by that of MB1. Total CK-MB2 activity was calculated by multiplying the percentage of MB2 in the fractions by total CK-MB activity as measured by Roche’s Isomune kit. The troponin-T concentration was measured using Enzymun-Test System (Boehringer Mannheim, Lewes, UK) on an ES-300 automated analyzer.
Time-activity curves for each assay were constructed for each patient, peak levels were noted, and the area under the curve (AUC) was calculated as a means of expressing total substance release. The analysis of the difference in the peak enzyme release and area AUC between the two groups was performed using a Mann- Whitney U test. The significance level for difference for all tests was p less than 0.05.
Definition of clinical end points
"Death" was defined as failure to leave hospital alive, for whatever reason. Usually, once the patient has been at 37°C for 20 minutes per hour of bypass time, we recommence intermittent positive pressure ventilation (IPPV), occlude the venous line, and discontinue perfusion within a minute or so. Failure to achieve this at the first attempt was defined as "delay in coming off bypass." This was not a blinded end point. Inotropes were given according to clinical judgment. Ventricular arrhythmias were noted by the research fellow (R.E.), as observed from the monitor in theater.
|
Results
|
|---|
Forty patients were studied as planned, 20 in each group. There were no preoperative differences between the groups with regard to the factors selected for minimization, that is age, left ventricular function, operating surgeon, or in the mix of operations performed (Table 1). There were no intraoperative differences between the groups with regard to aortic cross-clamp (median 47 minutes, range 30 to 88 minutes), cardiopulmonary bypass time (median 78 minutes, range 49 to 180 minutes), and hemodynamic measurements.
The possible adverse outcomes are given in Table 2, which gives the incidence of death, use of intraaortic balloon pump (IABP), difficulty in weaning from bypass, need for inotropic support, stay in intensive care of more than 24 hours, or use of renal support. As they stand, apart from delay in coming off pump, these results are well within the limits of the fourfold table test [20]; however, a study of 40 patients has insufficient power to exclude any differences in dichotomous events unless there is a very large discrepancy in outcome.
In Figure 1, the data are displayed for the area under the total MB release curve for each of the 40 patients. The characteristic log-normal distribution can be seen. We use the box and whisker plots [21] in subsequent figures to summarize the AUC data for MB2 isoform and troponin-T (Figs 2, 3) and for peak MB, MB2, MB2/MB1 ratio, and troponin-T (Figs 4–7). There was no significant difference between the groups with regard to total CK-MB, the MB2 isoform, or troponin-T release, whether expressed as peak or AUC.
View larger version (8K):
[in this window]
[in a new window]
|
Fig 1. The data for the area under the total MB release curve for each of the 40 patients. The characteristic log-normal distribution can be seen. (A) Control group; (B) modified reperfusion group with a "hot shot." The vertical axis uses a log scale.
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Fig 3. AUC for troponin T release. (A) Control group; (B) modified reperfusion group with a "hot shot."
|
|
View larger version (10K):
[in this window]
[in a new window]
|
Fig 4. Peak MB, displayed as box and whisker plots (median, interquartile range, and extreme range). (A) Control group; (B) modified reperfusion group with a "hot shot."
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Fig 5. Peak MB2 displayed as box and whisker plots (median, interquartile range, and extreme range. (A) Control group; (B) modified reperfusion group with a "hot shot."
|
|
View larger version (10K):
[in this window]
[in a new window]
|
Fig 6. Peak MB2/MB1 ratio displayed as box and whisker plots (median, interquartile range, and extreme range). (A) Control group; (B) modified reperfusion group with a "hot shot."
|
|
View larger version (10K):
[in this window]
[in a new window]
|
Fig 7. Peak troponin T displayed as box and whisker plots (median, interquartile range, and extreme range). (A) Control group; (B) modified reperfusion group with a "hot shot."
|
|
|
Comment
|
|---|
This is a pragmatic study [22] in which we have included a heterogeneous group of patients to see if there is a systematic benefit attributable to the "hot shot." We have failed to show any difference in the postoperative release of the MB2 isoform of CK, total CK-MB activity, or tropinin-T between modified ("hot shot") and unmodified reperfusion. There is no hint of a trend in favor of the "hot shot" as used in this study. Furthermore, the distribution of the data points indicate that however large the trial, it is unlikely that any difference would emerge. We have chosen to do a small, detailed study of biochemical markers of myocardial damage. These are, of course, surrogate end points, but to reject the data we present, we would have to postulate that the beneficial effects of the "hot shot" are independent of the well-established indices of myocyte injury, which we have measured.
Clinical trials of such a strategy are beset with difficulties. In the most vulnerable cases, who may well be the ones who would benefit most, it is not easy to randomize to a trial and it is then difficult to adhere to a protocol. If clinical events such as death are used, the trial has to be of considerable size to achieve significance and to avoid a beta error. For example, to show a statistically significant reduction from 5% mortality to 3% mortality would require over 4,000 cases. The more elective and safe the cases (and therefore the easier to recruit into trials), the lower the incidence of discrete events and therefore the larger the trial must be. The same argument applies if we take myocardial infarction as an outcome, but there is an additional problem. Anatomically discrete myocardial infarction, as judged by new Q waves, may be due to disease and technical failure, and then will not be preventable by this strategy. Global left ventricular injury, which is more likely to be ameliorated by a global strategy, is more difficult to define. Soft end points such as use of a balloon pump or inotropes, however the criteria are set, and length of stay in intensive care, are in the hands of the observers, and bias cannot be excluded.
It could be argued that to show benefit, one needs to adopt "the whole package" with warm induction, perhaps use of warm perfusion, retrograde and antegrade perfusion, and then the replenishing solution. Indeed, each element might have to be evaluated in a series of extremely rigorous clinical studies, just as they have been systematically studied in an elaborate series of laboratory studies [2, 3]. It could be that our patients are not at sufficiently high risk to show a benefit from modification of reperfusion. It is true that in spite of our intention to study higher risk cases, more than half of the cases had good left ventricular function and more than half had a single valve operation. Furthermore, the median bypass was not long, at 78 minutes for the whole series, and ischemic times ranged from 30 to 88 minutes (median 47 minutes).
In practice, we found that the additional manipulations added complexity to the operation, and in particular, the flaccid nature of the heart for several minutes after modified reperfusion necessitated special attention to the risk of increasing wall tension. Finally, the reperfusion solution adds a definite cost, equivalent to several hundred dollars, to each case. We cannot agree with Teoh and associates [7], who concluded: "Because no adverse effects of warm blood cardioplegia were encountered, we would recommend a ‘hot shot’ whenever cold blood cardioplegia is employed for myocardial protection." We would need to establish benefit in clinical practice to adopt this technique as a routine, which this trial has failed to do.
|
Acknowledgments
|
|---|
This study was supported by a grant from the British Heart Foundation (PG92077).
|
References
|
|---|
-
Miura T. Does reperfusion induce myocardial necrosis?. Circulation 1990;82:1070-1072.[Free Full Text]
-
Buckberg G.D. Studies of controlled reperfusion after ischaemia. I. When is cardiac muscle damaged irreversibly?. J Thorac Cardiovasc Surg 1986;92(Suppl 3 Pt 2):483-487.[Medline]
-
Buckberg G.D. Studies of hypoxemic/reoxygenation injury. I. Linkage between cardiac function and oxidant damage. J Thorac Cardiovasc Surg 1995;110(Suppl 4 Pt 2):1164-1170.
-
Engelman R.M., Rousou J.A., Flack J.E., Iyengar J., Kimura Y., Das D.K. Reduction of infarct size by systemic aminoacid supplementation during reperfusion. J Thorac Cardiovasc Surg 1991;101:855-859.[Abstract]
-
Follette D.M., Fey K., Buckberg G.D., et al. Reducing postischemic damage by temporary modification of reperfusate calcium, potassium, pH, and osmolarity. J Thorac Cardiovasc Surg 1981;82:221-238.[Abstract]
-
Lazar H.L., Buckberg G.D., Manganaro A.J., et al. Reversal of ischemic damage with secondary blood cardioplegia. J Thorac Cardiovasc Surg 1979;78:688-697.[Abstract]
-
Teoh K.H., Christakis G.T., Weisel R.D., et al. Accelerated myocardial metabolic recovery with terminal warm blood cardioplegia. J Thorac Cardiovasc Surg 1986;91:888-895.[Abstract]
-
Hossein-Nia M., Kallis P., Brown P.A., et al. Creatine kinase MB isoforms. Clin Chem 1994;40:1265-1271.[Abstract/Free Full Text]
-
Puleo P.R., Guadagno P.A., Roberts R., Perryman M.B. Sensitive, rapid assay of subforms of creatine-kinase MB. Clin Chem 1989;35:1452-1455.[Abstract/Free Full Text]
-
Puleo P.R., Guadagno P.A., Roberts R., et al. Early diagnosis of acute myocardial infarction based on assay for subforms of creatine-kinase MB. Circulation 1990;82:759-764.[Abstract/Free Full Text]
-
Katus H.A., Remppis A., Neumann F.J., et al. Diagnostic efficiency of tropinin-T measurements in acute myocardial infarction. Circulation 1991;83:902-912.[Abstract/Free Full Text]
-
Gerhardt W., Ljungdahl L., Herbert A.K. Troponin-T and CK MB (mass) in early diagnosis of ischemic myocardial injury. The Helsingborg Study, 1992. Clin Biochem 1993;26:231-240.[Medline]
-
Anderson J.R., Hossein-Nia M., Kallis P., et al. Comparison of two strategies of myocardial management during coronary artery operations. Ann Thorac Surg 1994;58:768-773.[Abstract]
-
Liu Z., Valencia O., Treasure T., Murday A.J. Cold blood cardioplegia or intermittent cross-clamping in coronary artery bypass grafting?. Ann Thorac Surg 1998;66:462-465.[Abstract/Free Full Text]
-
Treasure T., MacRae K. Minimization. Br Med J 1998;317:362-363.[Free Full Text]
-
Pocock S.J. Clinical trials—a practical approach. Chichester: Wiley, 1989:84-86.
-
Buckberg G.D. Antegrade/retrograde blood cardioplegia to ensure cardioplegic distribution. J Card Surg 1989;4:216-238.[Medline]
-
Davies J., Reynolds T., Penney M.D. Creatine kinase isoforms. Ann Clin Biochem 1992;29:202-205.
-
Wicks R.W., Usategui-Gomez M., Miller M., Warshaw M. Immunochemical determination of CKMB isoenzyme in human serum. II. An enzymic approach. Clin Chem 1982;28:54-58.[Abstract/Free Full Text]
-
Diem K., Lentner C. Documenta Geigy. Macclesfield: Geigy Pharmaceuticals, 1970.
-
Wilson A.P.R. Data presentation. Lancet 1993;341:282.[Medline]
-
Roland M., Torgerson D.J. What are pragmatic trials?. Br Med J 1998;316:285.[Free Full Text]
Accepted for publication September 8, 1999.
This article has been cited by other articles:
|
|
|
|
 
X. Y. Jin and R. Pillai
Invited Commentary
Ann. Thorac. Surg.,
February 1, 2008;
85(2):
458 - 459.
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. Rergkliang, A. Chetpaophan, V. Chittithavorn, P. Vasinanukorn, and V. Chowchuvech
Terminal Warm Blood Cardioplegia in Mitral Valve Replacement: Prospective Study
Asian Cardiovasc Thorac Ann,
April 1, 2006;
14(2):
134 - 138.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
