C spread clinical use for operative myocardial protection.

Size: px

Start display at page:

Download "C spread clinical use for operative myocardial protection."

Oswin Newman
5 years ago
Views:

Role of Potassium Concentration in Cardioplegic Solutions in Mediating Endothelial Damage Pankaj S. Mankad, FRCS, Adrian H. Chester, BSc, and Magdi H.

1 Role of Potassium Concentration in Cardioplegic Solutions in Mediating Endothelial Damage Pankaj S. Mankad, FRCS, Adrian H. Chester, BSc, and Magdi H. Yacoub, FRCS National Heart and Lung Institute, London, and Harefield Hospital, Harefield, United Kingdom We studied the effect of potassium concentration in cardioplegic solutions on endothelial function by examining its influence on 5-hydroxytryptamine- (5-HT) and nitroglycerin-induced vasodilation in the isolated rat heart. Forty-eight rat hearts were perfused on a modified Langendorff preparation. After a baseline record of increase in coronary flow induced by M 5-HT and 10 pg/ml nitroglycerin, the hearts were perfused for 30 or 60 minutes with either St. Thomas solution or Bretschneider solution containing 20 mmol/l of potassium or for 30 minutes with either solution containing 30 mmol/l of potassium (n = 8 in each). Initially, 5-HT and nitroglycerin caused a 39.0% f 3.3% and 39.7% f 2.8% increase in coronary flow, respectively. After 30 or 60 minutes perfusion with St. Thomas solution containing 20 mmol/l of potassium, there was little change in the response to 5-HT or nitroglycerin (5-HT, 43.1% * 4.1%; nitroglycerin, 38% * 3.2%). Similarly, perfusion with Bretschneider solution (20 mmol/l K+) for 30 or 60 minutes did not alter the degree of vasodilation (5-HT, 39.2% f 2.9%; nitroglycerin, 38.0% f 3.3%). However, perfusion with St. Thomas solution containing 30 mmol/l of potassium for 30 minutes abolished the endothelial-dependent 5-HT-induced vasodilation (5-HT, -1.6% f 1.4%; nitroglycerin, 36.9% f 2.2%). Perfusion with Bretschneider solution (30 mmol/l K+) gave similar results (5-HT, -2.1% f 1.2%; nitroglycerin, 36.4% * 1.7%). We conclude that the concentration of potassium in cardioplegic solutions plays a critical role in causing functional endothelial damage. (Ann Thorac Surg 1991;51:89-93) rystalloid cardioplegic solutions have been in wide- C spread clinical use for operative myocardial protection. Though the principles of cardioplegia-induced myocardial protection are well established [l, 21, the optimum concentration of various agents and the nature of additives have not been standardized, thereby leading to a variety of crystalloid cardioplegic solutions in clinical use with varying concentrations of principle ingredients and with different additives. Currently the most commonly used cardioplegic agent is potassium in the concentration of 15 to 35 meq/l [3]. Potassium is a known vascular irritant and has been used to denude the vessels of endothelium [4]. The cytotoxicity of hyperkalemic crystalloid cardioplegic solutions to cultured endothelial cells has been demonstrated by Carpentier and associates [5] and Solberg and coworkers [6]. However, the effect of potassium concentration in cardioplegic solutions on endothelial function has not been studied. We [7] have previously shown that 5-hydroxytryptamine (5-HT) causes a dose-dependent increase in coronary flow in the isolated perfused rat heart that is dependent on an intact endothelium and that this model could be used to study the effect of various agents thought to have a deleterious effect on coronary vascular endothelium. In this study, we have used this model (isolated perfused rat heart) to investigate the effect of potassium Accepted for publication Sep 10, Address reprint requests to Mr Mankad, Harefield Hospital, Harefield, Uxbridge, Middlesex UB9 6JH, United Kingdom. concentration in two different cardioplegic solutions in causing functional damage to the coronary vascular endothelium Material and Methods Animals and Reagents Hearts were obtained from male AO/Olac rats (body weight, 300 to 350 g) that had been maintained on a standard laboratory diet. 5-hydroxytryptamine was obtained from Sigma Chemicals Ltd, St. Louis, MO; Nitroglycerin, from Dupont (UK) Ltd; and St. Thomas Hospital cardioplegic solution from Macarthy Medical, Romford, Essex, UK. Bretschneider solution was prepared by the Pharmacy at Harefield Hospital. All the animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication No , revised 1985). Perfusion Technique Rats were anesthetized with halothane. The chest was opened and 200 units of heparin was injected into the right ventricle. After about 60 seconds, the heart was excised and immersed in a beaker containing Krebs- Henseleit bicarbonate buffer [8] with the following composition : NaCI, ; KCI, 4.69; MgSO, 7H,O, 1.18; KH,PO,, 1.18; NaHCO,, 25.0; glucose, 11.1; and CaCI,, 2.5. The heart was then mounted onto a modified Langendorff preparation [9] that allowed accurate measurement of coronary flow at a constant perfusion pressure (100 cm H,O) from any one of four reservoirs by The Society of Thoracic Surgeons /91/$3.50

90 MANKADETAL Ann Thorac Surg 1991:51:89-93 Table 1. Composition of St. Thomas' Hospital Cardioplegic Solution" Potassium Comvonent 20 mmol/l 30 mmol/l Sodium chloride 147.0 147.

2 90 MANKADETAL Ann Thorac Surg 1991:51:89-93 Table 1. Composition of St. Thomas' Hospital Cardioplegic Solution" Potassium Comvonent 20 mmol/l 30 mmol/l Sodium chloride Potassium chloride Magnesium chloride Calcium chloride Procaine hydrochloride (mmovl) a ph = 5.5 to 7.5; osmolality = 300 to 320 mosm/l Hearts were perfused through the aorta with ultrafiltered (5-pm filter) Krebs-Henseleit buffer and gassed with 95% O2 plus 5% C02 (ph 7.4 at 37 C.). Drugs used in the experiments were dissolved in the buffer at the desired concentration and were gassed uniformly throughout the study. After the first 10-minute stabilization period, during which the coronary flow was measured continuously, the hearts were perfused with mmol/l 5-HT for 3 minutes. Coronary flow was recorded for the last 2 minutes of perfusion with 5-HT. The coronary circulation was washed with the drug-free buffer for the next 12 minutes to reestablish the basal coronary flow. The hearts subsequently were perfused with nitroglycerin (10 pg/ml) for 3 minutes and the coronary flow was measured for the last 2 minutes of perfusion. The mean increase in coronary flow after 5-HT and nitroglycerin perfusion was calculated and expressed as a percentage over the control value. After the establishment of baseline response of the coronary vasculature to 5-HT and nitroglycerin, the hearts were perfused continuously at 20 C for 30 or 60 minutes with either St. Thomas' hospital solution or Bretschneider solution containing 20 mmoyl of potassium or for 30 minutes with either solution containing 30 mmoyl of potassium. The composition of these solutions is shown in Tables 1 and 2. Coronary flow was measured for the total duration. Table 2. Composition of Bretschneider Cardioplegic Solution" Potassium Comvonent 20 mmovl 30 mmol/l Sodium chloride Potassium chloride Magnesium chloride Procaine hydrochloride (mmolll) Mannitol a ph = 6.0 to 7.5; osmolality = 320 to 340 mosmll. The hearts were then rewarmed by perfusing with Krebs-Henseleit buffer at 37 C for 15 minutes during which the basal coronary flow was reestablished. Perfusion with lop7 mmoyl 5-HT and 10 pg/ml of nitroglycerin was repeated for 3 minutes with a 12-minute washout period in between and the percentage change in coronary flow calculated as before. Pilot experiments were performed to find the optimum perfusion time with drug-free buffer required to achieve stabilization, to establish the perfusion time required for 5-HT and nitroglycerin to produce the maximal response, and to standardize the period of wash-out with the drug-free buffer required to reestablish the basal coronary flow after drug-induced changes in coronary flow. In additional studies, it was demonstrated that the hearts perfused with the buffer for a period of 2 hours maintained a similar response to 5-HT and nitroglycerin. Expression of Results Coronary flow was measured continuously. Values used for determining the changes in coronary flow were taken during the last 4 minutes before changing to perfusion with the drug-containing buffer and during the last 2 minutes of perfusion with 5-HT and nitroglycerin. Changes in coronary flow were expressed as a percentage of the control value. Eight hearts were used in each experimental protocol. All data were expressed as the mean & standard error of the mean. Statistical analysis was performed by unpaired Student's t test. Table 3. Percentage Increase in Coronary Flow Induced by 5-Hydroxytryptamine (lop7 mmolll) and Nitroglycerin (10 pg/ml) Before and After Continuous Infusion of St. Thomas' and Bretschneider Cardioplegic Solutions Containing 20 rnmolll of Potassium for 30 Minutes in the Isolated Rat Heart % Increase in Coronary Flow 5-Hydroxytryptamine Nitroglycerin Solution Before After Before After St. Thomas' 37.9? ? 3.9b 45.2? ? 3.6b Bretschneider ? 3.7b 36.8 t f 2.6b a n = 8. p = not significant.

3 Ann Thorac Surg 1991;51:89-93 MANKADETAL 91 Table 4. Percentage lncrease in Coronary Flow lnduced by 5Hydroxytryptamine (ZO-7 mmolll) and Nitroglycerin (10 pglml) Before and After Continuous lnfusion of St. Thomas' and Bretschneider Cardioplegic Solutions Containing 20 mmolll of Potassium for 60 Minutes in the Isolated Rat Heart" Solution 5-Hydroxytryptamine % Increase in Coronary Flow Nitroglycerin Before After Before After St. Thomas' b 38.6 f f 2.8b Bretschneider 44.1 f Zb 40.9 f f 4.0b a n = 8. p = not significant. Results Both 5-HT and nitroglycerin initially caused an increase in the coronary flow. After continuous infusion of the two solutions containing 20 mmoyl of potassium for 30 minutes, there was no significant difference in the percentage increase in the coronary flow induced by 5-HT or nitroglycerin (Table 3). Continuous infusion of St. Thomas' or Bretschneider solution containing 20 mmoyl of potassium for 60 minutes maintained the basal vasodilatory response of coronary vasculature to 5-HT and nitroglycerin (Table 4). In contrast, continuous infusion for 30 minutes of St. Thomas' solution containing 30 mmoyl of potassium completely abolished the 5-HT-induced increase in coronary flow (36.8% & 2.3% increase in flow before versus 1.6% & 1.4% decrease in flow after the infusion, p < 0.001); however, nitroglycerin-induced vasodilation was not affected (38.1% * 2.9% increase in flow before and 36.9% * 2.2% increase after the infusion, p = not significant) (Fig 1A). Continuous infusion of Bretschneider solution containing 30 mmoyl of potassium had a similar effect on the response to 5-HT and nitroglycerin (Fig 1B). 5-Hydroxytryptamine and nitroglycerin caused a 34.6% & 2.2% and 38.4% % basal increase in the flow, respectively. After the infusion, 5-HT caused a mild vasoconstriction (2.1% & 1.2% decrease in the flow, p < 0.001) but the vasodilaton induced by nitroglycerin was well maintained (36.4% f 1.7% increase in the flow, p = not significant). Comment In the present series of experiments, we have demonstrated that continuous infusion for 60 minutes of cardioplegc solutions containing 20 mmoyl of potassium is not detrimental to endothelial function. Our experimental protocol is only slightly at variance with the clinical practice, where after a single dose of cardioplegia, the heart is arrested and a second dose is administered after a variable period of time. During the period of cardioplegic arrest, the potassium is slowly washed out of the coronary circulation by noncoronary collateral flow [lo]. In this study, we have used the continuous infusion technique to avoid ischemic endothelial damage. Both the duration of the continuous infusion and the temperature of the solution were selected to approximate the clinical situation. Under these experimental conditions, we have shown that an increase of the potassium concentration from 20 mmovl to 30 mmolil without any alteration in the components or dosage of St. Thomas' or Bretschneider solution is highly detrimental to endothelial function. The conversion of a 5-HT-induced vasodilatory response to a slight vasoconstriction after high-potassium cardioplegia infusion was not due to any alteration in the vascular "1 - L x HT GTN 5-HT GTN A B Fig I. Percentage increase in coronary flow induced by 5-hydroxytryptamine (5-HT) (lo-' rnmolll) and nitroglycerin (GTN) (10 &ml) before (open bars) and after (hatched bars) the continuous infusion of St. Thomas' solution (A) or Bretschneider solution (B) containing 30 mmolll of potassium for 30 minutes in the isolated rat heart (n = 8 in each). (The negative value indicates reduction in the flow.) -10 1

4 92 MANKADETAL Ann Thorac Surg 1991;51:89-93 smooth muscle responsiveness as nitroglycerin-induced vasodilation remained unchanged. We [7] have previously shown that in the isolated rat heart, 5-HT causes coronary vasodilation by releasing endothelium-derived relaxing factor and that when the release of endotheliumderived relaxing factor is blocked or when the endothelium is damaged, the vasoconstrictive effect of 5-HT on the coronary vasculature is unmasked. Therefore, the apparent reduction in coronary flow after high-potassium cardioplegia is a reflection of endothelial damage. Our findings are in agreement with those of Saldanha and Hearse [ll]. Solberg and co-workers [6] studied the effect of cardioplegic solutions on cultured human endothelial cells. They concluded that the presence of procaine increased the injurious effect of the solutions. The two solutions used in this study differed widely with regard to their procaine concentration. However, our results indicate that in the isolated rat heart, high concentration of procaine per se is unlikely to cause endothelial dysfunction. Potassium is a known vascular irritant [12] and has been shown to cause endothelial damage of vein grafts in animal models [13, 141. Hoover and co-workers [13] exposed the canine jugular vein at low pressure to hyperkalemic lactated Ringer s solution (K+, 40 meq/l) for several minutes and then grafted into the external carotid artery. At 6 weeks, the grafts exposed to hyperkalemic solution showed intense adventitial fibrotic reaction under light microscopy suggesting a full-thickness graft injury as compared with the normal-looking controls. In the primate model of vein graft atherogenesis, Olinger and associates [14] have shown significantly higher cholesterol accumulation in cephalic vein grafts exposed to hyperkalemic solution (27 meq/l) when evaluated at 3 months after implantation. This phenomenon is most likely to be related to endothelial injury. Endothelial damage could have important clinical implications. Intact endothelium produces prostacyclin, antithrombin 111, plasminogen activator, heparan sulfate, and a,-macroglobulin [ 15-19]. This prevents platelet aggregation and fibrin deposition on the vessel wall. Platelet deposition on the damaged endothelium in the early postoperative period could lead to reperfusion injury [20, 211. Endothelial damage could also result in early graft occlusion or the long-term complication of graft atherosclerosis [22]. Because the action of a majority of vasoactive endogenous and exogenous agonists is strongly modulated by the basal and stimulated release of endothelium-derived relaxing factor, endothelial damage can have a profound effect on tissue perfusion by unmasking the receptormediated direct vasoconstrictive effect [23, 241 of these agents. This could have an adverse effect on myocardial function especially in the early postoperative period or could produce irreversible myocardial damage. Other investigators have provided evidence that hyperkalemic cardioplegic solutions damage coronary vascular endothelium [5, 25-27]. Carpentier and associates [5] evaluated the cytotoxicity of 12 different cardioplegic solutions using tissue culture technique. They found that the percentage of dead endothelial cells after 3 hours incubation in various cardioplegic solutions at 19 C ranged from 4.7% to 13.1% (4.5% dead cells in the control solution). Solberg and co-workers [6] studied the effects on cultured human endothelial cells of incubation with St. Thomas and Bretschneider cardioplegic solutions at 10 or 20 C by measuring release of chromium 51 from the prelabeled cells and by electron microscopy. They demonstrated that there was mild to moderate injury to the cells after 1 or 5 hours of incubation with either solution. Using morphological means, Harjula and associates [26] also demonstrated coronary endothelial damage after use of crystalloid cardioplegia. These studies provide evidence of the cytotoxic effect of potassium on the endothelium but differ from the clinical situation as either cultured cells were incubated with cardioplegic solutions [5, 6, 271 or ischemia time was prolonged without repeating the cardioplegia [26], thereby increasing the possibility of ischemia-induced endothelial damage. In summary, we have shown the critical role of potassium concentration in cardioplegc solutions in causing functional endothelial damage. Therefore, the potassium concentration in cardioplegic solutions should be optimized not only to protect the myocardium but also to prevent endothelial damage. References 1. Hearse DJ, Braimbridge MV, Jynge P. Protection of the ischemic myocardium. Cardioplegia. New York: Raven Press, Roberts AJ, ed. Myocardial protection in cardiac surgery. New York: Marcel Dekker, Kirklin JW, Barratt-Boyes BG. Myocardial protection during cardiac surgery with cardio-pulmonary bypass. In: Kirklin JW, Barratt-Boyes, BG, eds. Cardiac surgery. New York: Wiley, Griffiths TM, Edwards DH, Lewis MJ, et al. The nature of endothelium-derived relaxant factor. Nature 1984;308: Carpentier S, Murawsky M, Carpentier A. Cytotoxicity of cardioplegic solutions: evaluation by tissue culture. Circulation 1981;64(Suppl 2): Solberg S, Larsen T, Linda1 S, Prydz P, Jorgensen L, Sorlie D. The effects of two different crystalloid cardioplegic solutions on cultured human endothelial cells. J Cardiovasc Surg 1989;30: Mankad PS, Chester AH, Yacoub MH. 5-Hydroxytryptamine mediates endothelial dependent coronary vasodilation in the isolated rat heart by the release of nitric oxide. Cardiovasc Res (in press). 8. Krebs HA, Henseleit K. Untesuchungen uber die Hamstoffbilding im Tierkorper. Hppe-Seyler s Zeitshrift fur Physiologische Chemie 1932;210: Langendorff 0. Untersuchungen am uberlebender Saugertierherzen. Pfluger s Archiv Gesamte Physiol Menschen Tiere 1935;61: Krausse S, Hess ML. Characterization of cardiac sarcoplasmic reticulum dysfunction during short-term, normothermic, global ischemia. Circ Res 1984;55: Saldanha C, Hearse DJ. Coronary vascular responsiveness to 5-hydroxytryptamine before and after infusion of hyperkalemic crystalloid cardioplegic solution in the rat heart. Possible evidence of endothelial damage. J Thorac Cardiovasc Surg 1989;98: Gazitua R, Wilson K, Bistrian BN, Blackbum GL. Factors

Ann Thorac Surg 1991;51:8%93 MANKAD ET AL 93 determining peripheral vein tolerance to amino acid infusions. Arch Surg 1979;114:897-900. 13. Hoover EL, Pett SB, Eldor A, et al.

5 Ann Thorac Surg 1991;51:8%93 MANKAD ET AL 93 determining peripheral vein tolerance to amino acid infusions. Arch Surg 1979;114: Hoover EL, Pett SB, Eldor A, et al. The effects of crystalloid potassium cardioplegic solution on arterialized canine vein grafts. Circulation 1981;65(Suppl 2): Olinger GN, Boerboom LE, Bonchek LI, Hutchinson LD, Kissebah AH. Hyperkalemia in cardioplegic solutions causing increased cholesterol accumulation in vein grafts. J Thorac Cardiovasc Surg 1983;85: Weksler BB, Eldor A, Falcone D, et al. Prostaglandins and vascular endothelium. In: Herman AG, Vanhoutte PM, Denolin H, et al, eds. Cardiovascular pharmacology of the prostaglandins. New York: Raven, 1982: Chan V, Chan TK. Antithrombin 111 in fresh and cultured human endothelial cells: a natural anticoagulant from the vascular endothelium. Thromb Res 1979;15: Loskutoff DJ, Edgington TS. Synthesis of a fibrinolytic activator and inhibitor by endothelial cells. Proc Natl Acad Sci USA 1977; Becker CG, Harpel PC. Alfa,-macroglobulin on human vascular endothelium. J Exp Med 1976;144: Buonassisi V. Sulfated mucopolysaccharide synthesis and secretion in endothelial cell cultures. Exp Cell Res 1973;76: VanBenthuysen KM, McMurtry IF, Honvitz LD. Reperfusion after acute coronary occlusion in dogs impairs endotheliumdependent relaxation to acetylcholine and augments contractile reactivity in vitro. J Clin Invest 1987;79: Rosenbaum D, Levitsky 5, Silverman N, et al. Cardioplegia does not prevent reperfusion injury induced by intracoronary platelet deposition. Circulation 1983;68(Suppl 2): Born GVR. Platelets and blood vessels. J Cardiovasc Pharma- COI 1984;6: Vanhoutte I'M, Houston DS. Platelets, endothelium and vasospasm. Circulation 1985;72:72% Bassenge E, Busse R. Endothelial modulation of coronary tone. Frog Cardiovasc Dis 1988;30: Harjula ALJ, Mattila S, Jarvinen A, Myllarniemi H, Salmenpera M. Endothelial cell damage following crystalloid cardioplegic solution infusion. Arch Surg 1984;119: Harjula A, Mattila 5, Mattila I, et al. Coronary endothelial damage after crystalloid cardioplegia. J Cardiovasc Surg 1984;25: Solberg 5, Larsen T, Jorgensen L, Sorlie D. Injury to human endothelial cells in culture under conditions simulating the use of vein grafts for vascular surgery. Eur Surg Res 1984;16:

H centrations of 10 to 110 mmoyl have become standard

H centrations of 10 to 110 mmoyl have become standard Impairment of Vascular Endothelial Function by High-Potassium Storage Solutions Barry B. K. Chan, MD, Irving L. Kron, MD, Terry L. Flanagan, MPH, John A. Kern, MD, Charles E. Hobson, MD, and Curtis G.