Ann Thorac Surg 1998;66:466-470
© 1998 The Society of Thoracic Surgeons
Original articles: cardiovascular
Effects of cardioplegic solutions on the vasoreactivity of the internal mammary artery
Catherine I. Chardigny, MDa,
Victor A. Jebara, MDa,
Tony J. Verbeuren, PhDb,
Alain F. Carpentier, MD, PhDa,
Jean-Noël Fabiani, MDa
a Department of Cardiovascular Surgery, Broussais Hospital, Paris, France
b Servier Research Institute, Suresnes, France
Accepted for publication March 21, 1998.
Address reprint requests to Dr Fabiani, Département de Chirurgie Cardio-vasculaire, Hôpital Broussais, 96, rue Didot, 75014 Paris, France
 |
Abstract
|
|---|
Background. During free internal mammary artery grafting, cardioplegia administration can be performed through the internal mammary artery. The present study examined whether cardioplegic solutions produce arterial graft constriction and functional endothelial damage.
Methods. Forty internal mammary artery segments from 10 patients were incubated in Krebs solution (n = 10), University of Wisconsin solution (n = 10), Broussais Hospital solution (n = 10), or blood cardioplegia (n = 10).
Results. There was a significant difference in sensitivity to norepinephrine between segments in Krebs solution and those in University of Wisconsin solution or Broussais Hospital solution but not segments in blood cardioplegia. There was a significant difference in relaxation to acetylcholine between segments in Krebs solution and those in the three other cardioplegic solutions and between those in blood cardioplegia and segments in University of Wisconsin solution or Broussais Hospital solution. There was no significant difference in relaxation to sodium nitroprusside between segments in any of the solutions.
Conclusions. These experiments suggest that storage in the different cardioplegic solutions studied does not preserve the initial vasoreactivity of the internal mammary artery. However, blood cardioplegia appears to be less deleterious in regard to endothelial and myogenic vascular function.
|
Introduction
|
|---|
The internal mammary artery is widely used as a coronary artery bypass conduit. In some cases, and especially with the increasing interest in total arterial revascularization, bilateral internal mammary arteries are used in coronary artery bypass grafting, with or without saphenous vein conduits. Some authors [1] prefer primarily to use a second internal mammary artery graft for younger patients with good ventricular function.
Encouraged by favorable reports on the performance of the internal mammary artery, some groups [2] routinely use a free right internal mammary artery for the right coronary distribution. Therefore, a free internal mammary artery, like other free arterial bypass grafts, is subject to the effects of cardioplegia during cardiac bypass operations. Early vasospasm of internal mammary artery grafts has been reported [3]. It can be influenced by the method of harvesting, arterial wall damage from clamping [4], reactivity to circulating vasoconstrictors such as inotropic agents and endogenous catecholamines [5], and thromboxane A2 release [6] during cardiopulmonary bypass. Moreover, it has been demonstrated that cardioplegic solutions cause marked arterial endothelial damage [7]. The present study was designed to examine whether cardioplegic solutions and their components produce arterial graft constriction and functional endothelial damage.
|
Material and methods
|
|---|
Institutional approval was obtained for the study protocol
Sampling of vessels
Left internal mammary arteries were obtained intraoperatively from 10 patients undergoing coronary artery bypass grafting. After sternotomy, the left internal mammary artery was mobilized from the chest wall. A Favaloro or similar retractor was used for internal mammary artery pedicle dissection. Low-voltage coagulation was used to avoid thermal damage to the artery. After the internal mammary artery was mobilized from its origin to the distal end close to the bifurcation, the pedicle was wrapped with a papaverine hydrochloride–soaked gauze sponge. No bulldog clamp was applied. The artery was left with blood circulating inside until the pericardium was opened and the length required to reach the target vessel was obtained. After heparinization, the internal mammary artery was divided distally, and free flow was examined. The conduit was then cut, and the extra length was taken for experimental work.
Arterial segments
Vessel segments obtained for this study were immediately placed in a container with oxygenated physiologic salt solution (Krebs), maintained at 4°C, rinsed, freed from clotted blood, and transferred to the laboratory. The mean time between harvesting and experimentation was 0.5 to 1 hour. The vessel was opened and pinned in a dish coated with silicone rubber, and the surrounding adipose tissue was removed under magnification. Each vessel was cut into four 3-mm segments, which were then suspended on wires in organ baths. The Krebs solution had the following composition (millimolar): NaCl, 118.3; KCl, 4.7; NaHCO3, 25; MgSO4, 1.2; KH2PO4, 1.2; CaCl2, 2.5; glucose, 11. The solution was aerated with a gas mixture of 95% oxygen and 5% carbon dioxide at 37°C.
Organ-bath technique
Artery ring segments were mounted on two stainless steel wires in a 20-mL water-jacketed glass organ bath (Fig 1). The full description and the technical details of this technique have been published [8].
View larger version (18K):
[in this window]
[in a new window]
|
Fig 1. Organ bath apparatus with strain gauge transducer. (a = 20-mL water-jacketed glass organ bath; b = arterial ring segment mounted on two stainless steel wires; c = lower wire fixed to micrometer; d = upper wire attached to force transducer.)
|
|
Cardioplegic solutions
University of Wisconsin solution was chosen as the prototype of a solution with a high potassium concentration (120 mmol/L). Because we usually use it in our institution, Broussais Hospital solution was selected as an alternative solution containing a lower potassium concentration (19.9 mmol/L). The third solution was blood cardioplegia (KCl, 18 mmol/L; hematocrit, 15% to 20%). The effects of each solution were studied in comparison with standard Krebs solution. The actions of the cardioplegic solutions were recorded under normothermic (37°C) conditions to study the effects of cardioplegia alone, without adding the consequences of hypothermia, on vasoreactivity.
The drugs used were potassium chloride (Sigma Chimie, St-Quentin Fallavier, France), norepinephrine (arterenol hydrochloride, Sigma Chimie), acetylcholine (acetylcholine chloride, Sigma Chimie), and sodium nitroprusside (sodium nitroferricyanide, Sigma Chimie). Solutions were prepared daily for each experiment and then discarded after the experiment. Initial solubilization and dilutions were made with distilled water.
Protocol
Contraction
In Krebs solution, each arterial segment was contracted using potassium chloride (100 mmol/L) to obtain maximum smooth-muscle contraction (100%), rinsed, and allowed to equilibrate after contraction. Contraction-response curves were obtained by adding increasing concentrations of norepinephrine to the organ-bath fluid in 0.5-log-unit steps. Each step was made when previous maximum contraction was achieved and had stabilized. At the end of each test, injection of potassium chloride was applied to obtain a reference contraction (the results were expressed as percent potassium chloride contraction). Mean curves were obtained with ten segments from the 10 patients (Fig 2A).
View larger version (24K):
[in this window]
[in a new window]
|
Fig 2. (A) Drug protocol in Krebs solution and (B) protocol for incubation in cardioplegic solutions. (ACh = acetylcholine; KCl = potassium chloride; NE = norepinephrine; NPS = sodium nitroprusside.)
|
|
Endothelium-dependent relaxation
In Krebs solution and after contraction, segments were rinsed until they stabilized at resting length. Information was obtained from the contraction-response curve for each constrictor agent to produce a submaximal contraction before the relaxation test. The norepinephrine concentration used was 10-7 mol/L. When a stable level of contraction was obtained, acetylcholine was applied at a concentration of 10-6 mol/L. Presence of endothelium was assessed by the functional relaxation response to acetylcholine [9].
Endothelium-independent relaxation
Segments were rinsed and equilibrated at resting tension. After submaximal precontraction as previously described, the last step was the measure of maximum smooth-muscle relaxation using sodium nitroprusside at a concentration of 10-5 mol/L.
Incubation in different solutions
The four segments of internal mammary artery from each patient were used as follows: one segment was incubated for 1 hour in Krebs solution (control), one in University of Wisconsin solution, one in Broussais Hospital solution, and one in blood cardioplegia. The segments were then rinsed many times in Krebs solution for 1 hour, after which they were subjected to the same protocol as just described (Fig 2B).
Data analysis
Forces generated by the vessels were digitized by a personal computer using the Moise 3 software package (EMKA Technologies, Paris, France). Several variables were studied to characterize arterial segment reactivity. (1) Contraction force was normalized to potassium chloride (10-1 mol/L) and expressed as percent potassium chloride contraction obtained after each dose-response curve with norepinephrine. (2) Sensitivity was measured by effective drug concentration producing 50% of maximum contraction. Fifty percent of maximum contraction was calculated from contraction-response curves generated by a sigmoid curve model for every ring segment. The variable 50% of maximum contraction was determined by nonlinear curve fitting regression with the simplex algorithm of Caceci and Cacheris [10] using the equation of Michaelis and Menten [11]: E = Emax x Cn)/(ECn + Cn), where E = contraction, Emax = maximum contraction, C = concentration, EC = 50% of maximum contraction, and n = Hill’s coefficient. (3) Endothelium-dependent relaxation to acetylcholine was expressed as percent contraction to the constrictor agent used. (4) Maximum smooth-muscle relaxation to sodium nitroprusside was expressed as percent contraction to the constrictor agent used.
The data were reported as the mean ± the standard error of the mean. They were evaluated for significance by applying analysis of variance with the Student-Newman-Keuls test for unpaired observations. A probability value of less than 0.05 was considered significant.
|
Results
|
|---|
Before cardioplegia
In Krebs solution, the maximum contraction force to potassium chloride was equal to 4.6 ± 0.76 g for the 40 segments studied; there was no significant difference between the 10 patients. Sensitivity to norepinephrine was not significantly different between the 40 segments from the 10 patients (Fig 3). Relaxation in response to acetylcholine was obtained in all of the internal mammary artery segments used for this study. There was no significant difference in the percent relaxation between the 40 segments from the 10 patients (range, 56% to 100%). There was no significant difference in the percent relaxation to sodium nitroprusside between the 40 segments (range, 85% to 120%).
View larger version (15K):
[in this window]
[in a new window]
|
Fig 3. Dose-response curves to norepinephrine in Krebs solution before incubation in any cardioplegic solution. Sensitivity to norepinephrine was not significantly different between the 40 segments from the 10 patients. For clarity, standard errors of the mean for each point on each curve are not shown. (BC = blood cardioplegia; BH = Broussais Hospital solution; KCl = potassium chloride; NE = norepinephrine; UW = University of Wisconsin solution.)
|
|
During cardioplegia
In the ten segments (from 10 patients) incubated in University of Wisconsin solution, a spontaneous contraction was observed for the first 5 minutes of incubation (2.11 ± 0.79 g), followed by a spontaneous decrease to resting tension within a mean of 5 minutes (Fig 4). No change from resting tension was observed in segments incubated in Krebs solution, Broussais Hospital solution, or blood cardioplegia.
View larger version (9K):
[in this window]
[in a new window]
|
Fig 4. Evolution of resting tension in University of Wisconsin solution. In ten segments (one from each patient) incubated in University of Wisconsin solution, a spontaneous contraction was observed for the first 5 minutes of incubation (2.11 ± 0.79 g), followed by a spontaneous decrease to resting tension within a mean of 5 minutes.
|
|
After cardioplegia
There was a significant difference in the contraction force to potassium chloride between segments incubated in Krebs solution (4.5 ± 0.3 g) and those incubated in University of Wisconsin solution (0.6 ± 0.2 g), Broussais Hospital solution (1.1 ± 0.6 g), or blood cardioplegia (0.9 ± 0.3 g). There was no significant difference in the contraction to potassium chloride between segments incubated in each of the three studied cardioplegic solutions.
There was a significant difference in sensitivity to norepinephrine between segments incubated in Krebs solution (3.1 x 10-8 ± 0.8 x 10-8 mol/L) and segments incubated in University of Wisconsin solution (1.2 x 10-7 ± 0.3 x 10-7 mol/L) (p = 0.009) or Broussais Hospital solution (1.6 x 10-7 ± 0.4 x 10-7 mol/L) (p = 0.005), but not segments incubated in blood cardioplegia (3.2 x 10-8 ± 0.8 x 10-8 mol/L). There was no significant difference in sensitivity to norepinephrine between segments incubated in University of Wisconsin solution and segments incubated in Broussais Hospital solution (Fig 5).
View larger version (15K):
[in this window]
[in a new window]
|
Fig 5. Dose-response curves to norepinephrine (NE) in Krebs solution after incubation in cardioplegic solutions. There was a significant difference in sensitivity to norepinephrine between segments in Krebs solution and segments in University of Wisconsin (UW) or Broussais Hospital (BH) solutions, but not segments incubated in blood cardioplegia (BC). For clarity, standard errors of the mean for each point on each curve are not shown. (KCl = potassium chloride.)
|
|
There was a significant difference in response and percent relaxation to acetylcholine between segments incubated in Krebs solution and those incubated in each of the three cardioplegic solutions. There was a significant difference in response and in percent relaxation to acetylcholine between segments incubated in blood cardioplegia and those incubated in University of Wisconsin solution or Broussais Hospital solution (Fig 6).
View larger version (19K):
[in this window]
[in a new window]
|
Fig 6. Endothelium-dependent relaxation to acetylcholine in Krebs solution after incubation in cardioplegic solutions. There was a significant difference in relaxation to acetylcholine between segments incubated in Krebs solution and those incubated in the three cardioplegic solutions and between segments incubated in blood cardioplegia (BC) and those incubated in University of Wisconsin solution (UW) or Broussais Hospital solution (BH). (NE = norepinephrine.)
|
|
There was no significant difference in percent relaxation to sodium nitroprusside between segments incubated in Krebs solution and each of the other solutions (Fig 7).
View larger version (19K):
[in this window]
[in a new window]
|
Fig 7. Endothelium-independent relaxation to sodium nitroprusside in Krebs solution after incubation in cardioplegic solutions. There was no significant difference (p = NS) in percent relaxation to sodium nitroprusside between segments incubated in Krebs solution and in each cardioplegic solution. (BC = blood cardioplegia; BH = Broussais Hospital solution; NE = norepinephrine; UW = University of Wisconsin solution.)
|
|
|
Comment
|
|---|
Like saphenous vein grafts, free internal mammary artery grafts, though they are rarely used, can be exposed to a cardioplegic solution. Vascular reactivity, particularly endothelium-dependent relaxation, is impaired after exposure to a hyperkalemic crystalloid solution. The addition of blood to the cardioplegic solution significantly preserves endothelium-dependent responses. In this study, the actions of the cardioplegic solutions were recorded under normothermic (37°C) conditions to study the effects of cardioplegia alone, without adding the consequences of hypothermia, on vasoreactivity.
University of Wisconsin solution has proved to be a superior agent for organ preservation [12]. Furthermore, because of the clear superiority of this solution in myocardial protection for transplantation, it has been suggested that University of Wisconsin solution may have a role in myocardial protection for purposes other than transplantation. In particular, the solution could be applied for the protection of the immature myocardium, which remains suboptimal with current techniques [13, 14].
However, increased recognition of the importance of endothelial physiology in the regulation of coronary blood flow [15, 16], platelet aggregation [17], and prevention of atherosclerosis has raised concern regarding potential endothelial injury from the high potassium concentration in University of Wisconsin solution. Loss of vasodilative reserve has been demonstrated after arrest with and storage in high-potassium cardioplegic solutions [18, 19], including University of Wisconsin solution [20]. It is known that high potassium concentrations as frequently used in clinical cardioplegic solutions cause the depolarization of both myocardium and vascular tissue. It is possible that prolonged exposure of endothelial cells to elevated levels of potassium can alter membrane transport or membrane potential. Endothelial cell hyperpolarization is thought to be important for the release of endothelium-derived nitric oxide.
In this study, all segments fully relaxed with sodium nitroprusside, an endothelium-independent endothelial vasodilator. This suggests that the ability of the smooth muscle cell to relax with nitric oxide is preserved with cardioplegia. This establishes that the loss of vasodilative response to acetylcholine is due to endothelial damage and not vascular smooth-muscle damage.
It is well known that a cardioplegic solution in an endothelial cell culture is responsible for cellular lesions leading to the death of 30% of cells after 2 hours. Several reasons for this have been given: potassium chloride cytotoxicity, fast acidification of the environment of the cells and production of oxygen-derived free radicals [7].
It has been established that myocardial protection with blood cardioplegia produces better results than other solutions. There are numerous theoretical advantages, particularly a less deleterious effect on endothelial cells. Blood is an inhibitor of oxygen-derived free radicals, which can be released during reperfusion immediately after myocardial or vascular ischemia. Also, the presence of blood prevents detrimental endothelial effects caused by osmotic and rheologic factors. Cullen and coauthors [21] attempted to measure in vitro changes in levels of endothelium-derived relaxing factor in response to incubation of pulmonary endothelial cells in University of Wisconsin solution. Surprisingly, release of endothelium-derived relaxing factor in response to bradykinin was maintained in cells incubated in University of Wisconsin solution (potassium concentration, 140 mEq/L) or other crystalloid solutions with potassium concentrations of 80 to 107 mEq/L but was impaired after incubation in a blood cardioplegic solution with a potassium concentration of only 6.8 mEq/L. The authors attributed the loss in release of endothelium-derived relaxing factor in the blood cardioplegia–incubated cells to the acidity of the solution, not to the potassium concentration.
The duration of endothelial dysfunction after high-potassium cardioplegia is unknown. Shimokawa and associates [17] found that 4 weeks after endothelial injury, regenerated endothelium, although structurally normal, has a blunted response to vasodilative stimuli. Cartier and colleagues [22] showed incomplete recovery of endothelium-dependent vasodilation and increased vasoconstrictive potential 8 weeks after endothelial injury, despite the presence of histologically intact endothelial cells.
The importance of impaired endothelium-dependent vasodilation is uncertain. The increased predisposition of the endothelium toward spasm and inability to release nitric oxide results in platelet deposition [23] and an increased propensity toward thrombosis. Treatment with arachidonic acid inhibitors such as indomethacin or with calcium-channel blockers has not been effective in modifying the early loss of vasodilative reserve [24]. Vascular endothelial preservation and enhancement of coronary flow after cold cardioplegia have been correlated to myocardial functional recovery. Because increased coronary perfusion in the early postoperative period may reduce early morbidity after ischemic cardioplegia and subsequent reperfusion, improved methods of endothelial preservation such as the use of blood-containing cardioplegic solutions may result in improved cardiac function after surgical procedures [25].
|
Acknowledgments
|
|---|
We thank Dr Jean-Jacques Descombes, Mrs Yvette Menant, and Véronique Barou for their contribution to this work.
|
References
|
|---|
- He G.-W., Ryan W.H., Acuff T.E., et al. Risk factors for operative mortality and sternal wound infection in bilateral internal mammary artery grafting. J Thorac Cardiovasc Surg 1994;107:196-202.[Abstract/Free Full Text]
- Loop F.D., Lytle B.W., Cosgrove D.M., Golding L.A.R., Taylor P.C., Stewart R.W. Free (aorta-coronary) internal mammary artery graft: Later results. J Thorac Cardiovasc Surg 1986;92:248-258.
- Sarabu M.R., McClung J.A., Fass A., Reed G.E. Early postoperative spasm in left internal mammary artery bypass grafts. Ann Thorac Surg 1987;44:199-200.[Abstract]
- Fonger J.D., Yang X.M., Cohen R.A., Haudenschild C.C., Shemin R.J. Impaired relaxation of the human internal mammary artery after temporary clamping. J Thorac Cardiovasc Surg 1992;104:966-971.[Abstract]
- Reves J.G., Karp R.B., Buttner E.E., et al. Neuronal and adrenomedullary catecholamine release in response to cardiopulmonary bypass in man. Circulation 1982;66:49-55.[Abstract/Free Full Text]
- Ylikorkala O., Saarella E., Viinikka L. Increased prostacyclin and thromboxane production in man during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1981;82:245-247.[Abstract]
- Follette D.M., Buckberg G.D., Mulder D.G., Fonkalsrud E.W. Deleterious effects of crystalloid hyperkalemic cardioplegic solutions on arterial endothelial cells. Surg Forum 1980;31:253-255.
- Chardigny C., Jebara V.A., Acar C., et al. Vasoreactivity of the radial artery: comparison with the internal mammary and gastroepiploic arteries with implications in coronary artery surgery. Circulation 1993;88(Suppl 2):115-127.
- Furchgott R.F., Zawadzki J.V. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980;288:373-376.[Medline]
- Caceci M.S., Cacheris W.P. Fitting curves to data: the Simplex algorithm is the answer. Byte 1984:340-362.
- Michaelis L., Menten M.L. Die kinetik der invertinwirkung. Biochem Zeitschrifft 1913;49:333-369.
- Ko W., Zelano J.A., Lazzaro R., et al. Superiority of the University of Wisconsin solution over simple crystalloid for extended heart preservation: a study of the left ventricular pressure–volume relationship. J Thorac Cardiovasc Surg 1992;103:580-592.
- Bull C., Cooper J., Stark J. Cardioplegic protection of the child’s heart. J Thorac Cardiovasc Surg 1984;88:287-293.[Abstract]
- Avkiran M., Hearse D.J. Protection of the neonatal myocardium during global ischemia: Is crystalloid cardioplegia effective in the immature myocardium?. J Thorac Cardiovasc Surg 1989;97:220-228.[Abstract]
- Kelm M., Shrader J. Control of coronary vascular tone by nitric oxide. Circ Res 1990;66:1561-1575.[Abstract/Free Full Text]
- Parent R., Paré R., Lavallé M. Contribution of nitric oxide to dilation of resistance coronary vessels in conscious dogs. Am Physiol Soc 1992;262:H10-H16.
- Shimokawa H., Vanhoutte P.M. Impaired endothelium-dependent relaxation to aggregating platelets and related vasoactive substances in porcine coronary arteries in hypercholesterolemia and atherosclerosis. Circ Res 1989;64:900-914.[Abstract/Free Full Text]
- Carpentier S., Murawsky M., Carpentier A. Cytotoxicity of cardioplegic solutions: evaluation by tissue cultures. Circulation 1991;64(Suppl 2):90-95.
- Cartier R., Hollmann C., Dagenais F., Buluran J., Pellerin M., Leclerc Y. Effects of University of Wisconsin solution on endothelium-dependent coronary artery relaxation in the rat. Ann Thorac Surg 1993;55:50-56.[Abstract]
- Jeevanadam V., Barr M.L., Auteri J.S., et al. University of Wisconsin solution versus crystalloid cardioplegia for human donor heart preservation: a randomized blinded prospective clinical trial. J Thorac Cardiovasc Surg 1992;103:194-199.[Abstract]
- Cullen S., Haworth S.G., Warren J.D. Donor organ preservation fluids differ in their effect on endothelial cell function. J Heart Lung Transplant 1991;10:999-1003.[Medline]
- Cartier R., Pearson P., Pyng J.L., Schaff H. Time course and extent of recovery of endothelium-dependent contractions and relaxations after direct arterial injury. J Thorac Cardiovasc Surg 1991;102:371-377.[Abstract]
- Azuma H., Ishikawa M., Sekizaki S. Endothelium-dependent inhibition of platelet aggregation. Br J Pharmacol 1986;88:411-415.[Medline]
- Mehta J.L., Nichols W.W., Donnelly W.H., Lawson D.L., Saldeen T.G.P. Impaired canine coronary vasodilator response to acetylcholine and bradykinin after occlusion-reperfusion. Circ Res 1989;64:43-54.[Abstract/Free Full Text]
- Amrani M., Shirvani R., Allen N.J., Ledingham S., Yacoub M.H. Enhancement of low coronary reflow improves postischemic myocardial function. J Thorac Cardiovasc Surg 1992;104:1375-1382.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
P. J. Pearson and H. V. Schaff
Effects of cardioplegic solutions on vasoreactivity of the internal mammary artery
Ann. Thorac. Surg.,
August 1, 1998;
66(2):
311 - 312.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
