Ann Thorac Surg 1999;68:1295-1301
© 1999 The Society of Thoracic Surgeons
Original Articles: Cardiovascular
Risk of low calcium and high magnesium in continuous warm hyperkalemic cardioplegia
Yoshinobu Nakamura, MDa,
Hiroaki Kuroda, MDa,
Naoaki Takemoto, MDa,
Shigetsugu Ohgi, MDa,
Tohru Mori, MDa
a Second Department of Surgery, Tottori University Faculty of Medicine, Tottori, Japan
Address reprint requests to Dr Ohgi, Second Department of Surgery, Tottori University Faculty of Medicine, 36-1 Nishi-machi, Yonago, Tottori 683-8504, Japan
 |
Abstract
|
|---|
Background. The recent introduction of operations on a warm heart has prompted clinical reports on the usefulness of continuous blood cardioplegia, but no in-depth basic evaluation of continuous cardioplegia has been done. The cardioprotective effects of magnesium (Mg) and calcium (Ca) in continuous warm hyperkalemic crystalloid cardioplegic solutions were investigated in an isolated rat heart model.
Methods. Isolated rat hearts were arrested for 180 minutes at 37°C with a continuous warm hyperkalemic (20 mmol/L) modified Krebs-Henseleit bicarbonate buffer solution containing 1.2, 8.0, or 16.0 mmol/L of Mg and 0.1 to 2.5 mmol/L of Ca in different concentrations. Recovery of cardiac function and tissue damage were estimated.
Results. For each Mg concentration, the percentage recovery of aortic flow generated dose-response curves depending on Ca concentration. However, as Mg concentration increased, the recovery of aortic flow decreased in the groups with 0.5 mmol/L of Ca or less.
Conclusions. In continuous warm cardioplegia the combination of low Ca and high Mg concentration caused severe cardiac injury, and normal Ca concentration avoids cardiac injury regardless of Mg concentrations.
|
Introduction
|
|---|
Recently, continuous warm blood cardioplegia (CWBCP) has been used in cardiac surgical procedures [1]. The ionic composition used in CWBCP is the same as that used for intermittent cardioplegia, but the electrolyte composition in such perfusates has rarely been analyzed, probably because it provides satisfactory cardiac arrest without causing ischemic reperfusion disorders. Engelman and associates [2] compared the myocardial protective effects of different methods of administration of myocardial protective fluid, and they found greater restoration of myocardial function in the repeated infusion group than in the continuous perfusion group. By contrast, Lichtenstein and coworkers [3] reported that satisfactory myocardial protection could be achieved through continuous blood cardioplegic perfusion in clinical use. As such, it has not yet been determined which method of administration, continuous coronary perfusion or repeated infusion, is better. This discrepancy between findings results from the lack of sufficient study of electrolyte compositions in fluids for continuous coronary perfusion cardiac arrest, whereas electrolyte compositions in ischemic heart-arresting fluids have been analyzed extensively [4, 5]. The present study investigated electrolyte compositions in high-potassium myocardial protective fluids continuously perfused at 37°C, focusing on the relationship between myocardial protective effect and magnesium (Mg) and calcium (Ca) concentrations.
|
Material and methods
|
|---|
Hearts
Adult male Wistar rats weighing 250 to 350 g were used in this study. The experimental rats 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. 85-23, revised 1985). At least 6 rats were used in each experimental group.
Experimental model
The preparation was a modification of the isolated working rat heart model described by Neely and associates [6]. All experimental hearts were excised in the beating state without ischemia as rapidly as possible by retrograde cannulation and perfusion of the aorta. Langendorff perfusion was then done at a constant pressure of 75 cm H2O. The left atrium was cannulated and the preload was set at 20 cm H2O. The oxygenated perfusate entered the left atrium, passed into the left ventricle, and was then ejected from the left ventricle against an afterload pressure of 100 cm H2O. The cardioplegic solution was infused in a retrograde manner through a side arm on the aortic cannula at a pressure of 60 cm H2O. The entire system was water-jacketed at 37°C.
Perfusate and cardioplegic solution
The standard perfusate was Krebs-Henseleit bicarbonate (KHB) buffer solution (Table 1). The continuous warm hyperkalemic crystalloid cardioplegic solution (CWCP) had the same basic composition as the KHB buffer, except that the potassium chloride concentration was 18.8 mmol/L, and the Ca and Mg concentrations were varied as listed Table 1. Three groups contained either 1.2, 8.0, or 16.0 mmol/L of Mg. Each group was further subdivided into subgroups each with one of various Ca concentrations (Table 1). All perfusates were exposed to a gaseous mixture of 95% O2 and 5% CO2 at 37°C. The pH was maintained at 7.4 ± 0.1. The osmolarity in all cardioplegic solutions was adjusted with sucrose to 390 mOsm/L [7]. All perfusates were filtered through cellulose-acetate filters with a pore size of 3.0 µm before use.
Experimental time course
As reported by Weng and colleagues [8], myocardial edema occurred more frequently as colloid osmotic pressure decreased. Thus, damage to the myocardium by continuous perfusion of crystalloid fluid has been observed. We also observed in our preliminary study that the cardiac output decreased to about 80% of the initial level when empty beating was continued for 240 minutes under continuous perfusion of KHB fluid (Fig 1). In consideration of myocardial damage by continuous perfusion of crystalloid fluid, we set the safety limit of continuous perfusion at 240 minutes for the present study.
View larger version (14K):
[in this window]
[in a new window]
|
Fig 1. Damage to the myocardium by continuous perfusion of crystalloid fluid (Krebs-Henseleit bicarbonate solution). Each point represents the mean value of 6 hearts. Bars indicate standard error of the mean.
|
|
After anesthesia with diethyl ether and infusion of 200 IU of sodium heparin into the right femoral vein, each rat heart was removed and placed in ice-cold KHB solution until contraction ceased (Fig 2). The aorta was then immediately connected to the aortic cannula, and Langendorff perfusion was initiated and continued for 5 minutes with oxygenated KHB solution. During this 5-minute period, left atrial cannulation was completed. Next, the working mode was initiated and continued for 15 minutes. Baseline values for the cardiac function indices were recorded during the last 1 minute of this working period. Then the hearts were randomly assigned to two groups. The hearts used as the control group were subjected to extended Langendorff perfusion for 180 minutes. Empty beating hearts were selected for use as the control group to observe the influence on the heart of a crystalloid perfusion with an oxygenated KHB solution. The remaining hearts were arrested for 180 minutes by continuous perfusion with an oxygenated CWCP containing various Ca and Mg concentrations at 37°C. During this period of arrest, coronary flow was assessed every 60 minutes to measure the change in coronary compliance. After this 180-minute period, small, right ventricular muscle specimens were obtained to estimate damage. Thereafter, each heart was perfused in the Langendorff mode with a standard KHB buffer for 15 minutes. During that time, coronary effluent was collected into an ice-cold beaker, and creatine kinase leakage (IU per 15 min/g dry weight) was analyzed. Subsequently the heart was converted to working mode again. After a 20-minute period, hemodynamic measurements indicating postarrest cardiac function were measured. Recovery was expressed as a percentage of the initial functional performance.
View larger version (20K):
[in this window]
[in a new window]
|
Fig 2. Experimental time-course. In this protocol, the temperature of the heart was maintained at 37°C. (L = Langendorff mode; W = working mode; CK = creatine kinase.)
|
|
Myocardial function assessment
During the prearrest working control period and the postarrest working recovery period, the following variables were measured: aortic flow (mL/min), coronary flow (mL/min), heart rate (beats/per min), and aortic peak systolic pressure (mm Hg). The absolute recovery values for different variables in individual hearts were compared and expressed as percentages of values obtained during the prearrest baseline period. During the cardiac arrest period, change in coronary compliance was measured. The standardized coronary flow, coronary flow (mL/min) per dry weight (g dry weight), was used as the coronary compliance index.
Biochemical measurement
Creatine kinase release into the coronary venous effluent was determined in 15-minute intervals by a Shimazu TCC-240A spectrophotometer (Shimazu, Kyoto, Japan) with a commercially prepared kit (Wako Pure Chemical Industries, Osaka, Japan). Myocardial water content was calculated as [1 - (dry weight/wet weight)] x 100 (%).
Cardiac ultrastructual examination
Small myocardial biopsy specimens obtained after 180 minutes of cardiac arrest were cut into slices of approximately 1 mm3, which were prefixed with 2% glutaraldehyde and finally fixed with 1% osmium acid. They were then dehydrated and embedded in epoxy resin. Ultrathin slices were double-stained with uranyl acetate and lead hydroxide and examined with an electron microscope (model HU-12A; Hitachi, Tokyo, Japan).
Statistical analysis
Results were expressed as mean ± standard error of the mean. Data were analyzed by one-way analysis of variance followed by Dunnett’s t test at the 0.05 significance level. Analysis of variance was used to examine whether changes in the ionic concentration affected cardiac functional recovery, water content, and creatine kinase leakage. Dunnett’s t test was used for between-group comparisons.
|
Results
|
|---|
The prearrest (baseline) cardiac function values were as follows: aortic flow, 67.3 ± 0.5 mL/minute; coronary flow, 18.6 ± 0.4 mL/minute; heart rate, 312.8 ± 2.9 beats per minute; and peak systolic pressure, 107.4 ± 1.2 mm Hg. These values did not differ significantly between groups. The postarrest cardiac function values are shown in Table 2.
Recovery of aortic flow
The percentage recovery of aortic flow depended on Ca concentration in the continuous warm hyperkalemic perfusate and generated dose-response curves for each Mg concentration (p < 0.01). For each Mg concentration, the maximum percentage recovery of aortic flow was detected at 1.0 mmol/L of Ca. The differences in percentage recovery of aortic flow among Mg 1.2 + Ca 1.0 mmol/L, Mg 8.0 + Ca 1.0, and Mg 16.0 + Ca 1.0 mmol/L cardioplegic solutions, which had the maximum percentage recovery of aortic flow at the respective Mg concentrations, were not statistically significant. Hearts that were arrested for 180 minutes, with perfusates with 0.1 mmol/L of Ca, became pale and contracted, exhibiting the so-called calcium paradox. In the 16 mmol/L of Mg groups, calcium paradox was observed even with 0.3 mmol/L of Ca in the cardioplegic solution. The percentage recovery of aortic flow decreased in hearts with 0.5 mmol/L or less Ca as Mg concentration increased (Table 2, Fig 3).
View larger version (39K):
[in this window]
[in a new window]
|
Fig 3. Effects of the concentration of calcium in the continuous cardioplegic solutions on myocardial protection. Dose-response curves for the relationship between recovery of aortic flow and concentration of calcium at the respective concentrations of magnesium. Columns represent mean values of 6 hearts, and bars indicate standard error of the mean.
|
|
Recovery of other cardiac function variables
When the recoveries of coronary flow, aortic peak systolic pressure, and heart rate among the Mg 1.2 + Ca 1.0 mmol/L, Mg 8.0 + Ca 1.0 mmol/L, and Mg 16.0 + Ca 1.0 mmol/L cardioplegic solutions, which had the maximum percentage recovery of aortic flow at the respective Mg concentrations, were compared, there were no significant differences among these solutions (Table 2). However, the solutions containing Mg 1.2 + Ca 0.1 mmol/L, Mg 8.0 + Ca 0.1 mmol/L, Mg 16.0 + Ca 0.1 mmol/L, and Mg 16.0 + Ca 0.3 mmol/L had the most deleterious effect on the heart, and with them the heart beat did not recover after cardiac arrest.
Creatine kinase leakage
Differences in creatine kinase leakage among Mg 1.2 + Ca 1.0 mmol/L, Mg 8.0 + Ca 1.0, and Mg 16.0 + Ca 1.0 mmol/L cardioplegic solutions, which had the maximum percentage recovery of aortic flow at the respective Mg concentrations, were not statistically significant. However, creatine kinase leakages tended to be lower in solutions that had good recovery of aortic flow. Creatine kinase leakage in Mg 1.2 + Ca 0.1 mmol/L, Mg 8.0 + Ca 0.1 mmol/L, Mg 16.0 + Ca 0.1 mmol/L, and Mg 16.0 + Ca 0.3 mmol/L solutions were 100 IU per 15 minutes per gram dry weight or higher, significantly higher than for the other solutions (Table 2).
Myocardial water content
The myocardial water content was significantly higher in the Mg 1.2 + Ca 0.1 mmol/L, Mg 8.0 + Ca 0.1 mmol/L, Mg 16.0 + Ca 0.1 mmol/L, and Mg 16.0 + Ca 0.3 mmol/L groups than in the other groups for each Mg concentration (Table 2).
Standardized coronary flow
For each Mg concentration, as perfusion time increased, the standardized coronary flow decreased. After 180 minutes of arrest, the highest standardized coronary flow was 42.9 ± 3.8 (mL/min · g dry weight), with Mg 16.0 + Ca 1.0 mmol/L solution (Fig 4).
View larger version (55K):
[in this window]
[in a new window]
|
Fig 4. The change in coronary compliance (the standardized coronary flow) during the cardiac arrest period. Standardized coronary flow, coronary flow (mL/min)/dry weight (g) was used as the coronary compliance index.
|
|
Cardiac ultrastructual examination
Electron microscopy (Mg 1.2 + Ca 1.0 mmol/L) showed that the sarcolemma was intact with depressions at the Z band regions. Mitochondria were found superficially located beneath the intact sarcolemma and in rows between the individual myofibrils (Fig 5A). In Figure 5B (Mg 1.2 + Ca 0.1 mmol/L) severe tissue damage with evidence of mitochondrial swelling, indistinct damaged cristae, and fluid-filled spaces in the myofibrils can be seen.
View larger version (107K):
[in this window]
[in a new window]
|
Fig 5. (A) Electron micrograph of tissue obtained from a heart group treated with Mg 1.2 + Ca 1.0 mmol/L. The sarcolemma was intact (original magnification x3,000). (IM = intact mitochondria; MF = myofibrils. (B) Electron micrograph of tissue obtained from a heart group treated with Mg 1.2 + Ca 0.1 mmol/L. Severe tissue damage was observed (original magnification x3,000). (DM = damaged swelling mitochondria; white arrow = damaged crista; black arrow = fluid-filled spaces.
|
|
|
Comment
|
|---|
The present study found the optimal Ca concentration in CWCP. For each Mg concentration, the maximum percentage recovery of aortic flow was detected at 1.0 mmol/L Ca. However, peak percentage recovery of aortic flow was not influenced by Mg concentration at the Ca concentration of 1.0 mmol/L. In ischemic myocardium, Takemoto and colleagues [9] found that the recovery of aortic flow was Ca dependent at each concentration of Mg, and was Mg dependent at each concentration of Ca. In other words, the optimal Ca concentration increased with increases in the Mg concentration and decreased with decreases in the Mg concentration. In their explanation of these results, they inferred that the failure of intracellular Ca homeostasis (ie, Ca overload) during ischemia and reperfusion was one of the main causes of myocardial injury and that the antagonism between Mg and Ca played a role in this phenomenon. Magnesium inhibits the influx of Ca through the Ca channels and reduces intracellular Ca overload. In fact, Mg is generally considered to be critical for the maintenance of energy metabolism in cells, and antagonism to Ca is considered to be a physiologic characteristic of Mg, as reported by Shine and Douglas [10]. In contrast to those findings for ischemic myocardium, the optimal Ca concentration was not affected by Mg concentration in CWCP in our experiments. There are several probable reasons for this discrepancy. First, Mg serves as the weakest slow channel blocker among bivalent cations, as reported by Touraki and Beis [11]; second, unlike in the ischemic heart, each channel can maintain its function in CWCP because of retention of the cell membrane structure; and third, slow channels are closed at a potassium concentration of 20 mmol/L.
The present study also found the calcium paradox described by Zimmerman and associates [12], namely, that the reduction of Ca concentration induces severe myocardial injury. According to our findings, the risk of the calcium paradox was increased at higher Mg concentrations. In a previous study, it was assumed that the critical Ca concentration at which the calcium paradox occurs is 50 µmol/L [13]. However, in the present study, the calcium paradox occurred even at Ca concentrations as high as 0.3 mmol/L in the 16.0 mmol/L of Mg group. The occurrence of the calcium paradox is influenced by many factors, including pH, volume, temperature, perfusion volume, concentrations of other ionic components, and coincident ischemia [14]. Therefore, the calcium paradox is to be expected in prolonged hyperkalemic perfusion. In a preliminary experiment, the calcium paradox occurred at a Ca concentration of 0.1 mmol/L even with a perfusion time of only 60 minutes. The mechanism of the calcium paradox has recently been explained by the observation that calcium-free perfusion renders the membrane highly permeable to sodium through Ca channels. Extracellular Ca affects the physiologic properties of double lipid layers in the cell membrane, and a low Ca concentration increases membrane permeability and decreases the selectivity of the Ca channel. Consequently, reexposure to Ca-containing solutions results in a massive influx of Ca by sodium-calcium exchangers, with a loss of membrane structure and function [15, 16]. When the calcium paradox occurred in this study, electron microscopy showed severe tissue damage with mitochondrial swelling and fluid-filled spaces in the myofibrils, and myocardial water content was at its highest levels. This increase in myocardial water content is consistent with sodium accumulation. Our findings are consistent with this hypothesis. Accordingly, we were compelled to question why the risk of the calcium paradox was higher with increases in Mg concentration in CWCP. The answer is probably that the total volume of coronary perfusion in CWCP was increased by the addition of Mg, as a result of the inhibitory effect of Mg on the increase in coronary vascular resistance (Fig 4), resulting in the calcium paradox, even at the same Ca concentration as that causing no calcium paradox at lower Mg concentrations.
Continuous warm blood cardioplegia has been an interesting technique used in cardiac surgical procedures. The ionic composition used in CWBCP is the same as that used for intermittent cardioplegia. The Ca concentration in solutions used for CWBCP is reduced by the use of Ca-chelating agents such as citrate-phosphate-dextrose. At our hospital, cardiac arrest is attained by continuous perfusion of Fremes’ fluid. From April 1995 through May 1998, 62 patients had cardiac operations with the CWBCP, and in each case, the Ca concentration in the blood cardioplegic solution was measured after 5 and 60 minutes of continuous perfusion. According to these measurements, the mean Ca concentration in the perfusate was 0.83 ± 0.11 mmol/L after 5 minutes and decreased to 0.67 ± 0.12 mmol/L after 60 minutes. Chambers and colleagues [17] also reported that the blood Ca concentration decreases during cardiopulmonary bypass. The potential for the calcium paradox in the clinical setting of cardioplegic arrest during cardiopulmonary bypass has been a matter of controversy. Rebeyka and associates [18] demonstrated that in in vivo hearts treated with multidose, moderately-hypothermic, Ca-free cardioplegic solution necrosis developed in one third of the ventricular mass, and there was comparable depression in their functional recovery after prolonged ischemia and reperfusion. On that basis, they suggested that the risk of the calcium paradox should not be ignored entirely in the clinical setting. However, their conclusions do not apply to our model because their multidose method of cardiac protection included an ischemic period whereas our continuous perfusion method did not. When CWCP is used, a blood-based solution of a type not used in the present study is used frequently. Blood obtained from the pump oxygenator system has a molecular weight of approximately 1.0 mmol/L, and collateral blood flow during cardiac arrest maintained Ca levels in the myocardium regardless of the use of a continuous cardioplegic solution with a low Ca concentration. Furthermore, the addition of both Mg and Ca to serum proteins within blood drastically influenced the relative values of ionized Mg and Ca in CWBCP. Thus it would be difficult to maintain Ca levels of less than 0.5 mmol/L if such a solution were used. Despite these limitations in the clinical relevance of this experimental study, the results indicate that recovery of cardiac function could be impaired by prolonged perfusion with CWCP when Ca concentration is low. In addition, when the extracellular Ca level was excessively high, the recovery of cardiac function was also impaired. Therefore, the maintenance of membrane Ca gradient within the normal range is critical to satisfactory recovery of cardiac function when CWCP is used.
|
References
|
|---|
-
Salerno T.A., Houck J.P., Barrozo C.A.M., et al. Retrograde continuous warm blood cardioplegia. Ann Thorac Surg 1991;51:245-247.[Abstract/Free Full Text]
-
Engelman R.M., Auvil J., O’Donoghue M.J., Levitsky S. The significance of multidose cardioplegia and hypothermia in myocardial preservation during ischemic arrest. J Thorac Cardiovasc Surg 1978;75:555-563.[Abstract]
-
Lichtenstein S.V., Ashe K.A., Dalati H.E., Cusimano R.J., Panos A., Slutsky A.S. Warm heart surgery. J Thorac Cardiovasc Surg 1991;101:269-274.[Abstract]
-
Hearse D.J., Stewart D.A., Braimbridge M.V. Myocardial protection during ischemic cardiac arrest. Possible deleterious effects of glucose and mannitol in coronary infusates. J Thorac Cardiovasc Surg 1978;76:16-23.[Abstract]
-
Yamamoto F., Braimbridge M.V., Hearse D.J. Calcium and cardioplegia. J Thorac Cardiovasc Surg 1984;87:908-912.[Abstract]
-
Neely J.R., Liebermeister H., Battersby E.J., Morgan H.E. Effect of pressure development on oxygen consumption by isolated rat heart. Am J Physiol 1967;212:804-814.[Free Full Text]
-
Hopkinson D.N., Odom N.J., Bridgewater B.J., Hooper T.I. Comparison of saccharides as osmotic impermeants during hy-pothermic lung graft preservation. Transplantation 1996;61:1667-1671.[Medline]
-
Weng Z.C., Nicolosi A.C., Detwiler P.W., et al. Effects of crystalloid, blood, and University of Wisconsin perfusates on weight, water content, and left ventricular compliance in an edema-prone, isolated porcine heart model. J Thorac Cardiovasc Surg 1992;103:504-513.[Abstract]
-
Takemoto N., Kuroda H., Hamasaki T., Hara Y., Ishiguro S., Mori T. Effect of magnesium and calcium on myocardial protection by cardioplegic solutions. Ann Thorac Surg 1994;57:177-182.[Abstract/Free Full Text]
-
Shine K.I., Douglas A.M. Magnesium effects on ionic exchange and mechanical function in rat ventricle. Am J Physiol 1974;227:317-324.[Free Full Text]
-
Touraki M., Beis I. Protective effects of magnesium, cobalt, nickel, and barium against a calcium paradox in the isolated frog heart. J Exp Zool 1991;259:287-293.[Medline]
-
Zimmerman A.N.E., Daems W., Hulsmann W., Snijder J., Wisse E., Durrer D. Morphological changes of heart muscle caused by successive perfusion with calcium-free and calcium-containing solutions (calcium paradox). Cardiovasc Res 1967;1:201-209.[Abstract/Free Full Text]
-
Frank J.S., Rich T.L., Beydler S., Kreman M. Calcium depletion in rabbit myocardium. Circ Res 1982;51:117-130.[Abstract/Free Full Text]
-
Frank J.S., Langer G.A., Nudd L.M., Seraydarian K. The myocardial cell surface, its histochemistry, and the effect of sialic acid and calcium removal on its structure and cellular ionic exchange. Circ Res 1977;41:702-714.[Free Full Text]
-
Hess P., Tsien R.W. Mechanism of ion permeation through calcium channels. Nature 1984;309:453-456.[Medline]
-
Hohl C.M., Alschuld R.A., Brierley G.P. Effects of calcium on the permeability of isolated adult rat heart cells to sodium. Arch Biochem Biophys 1983;221:197-205.[Medline]
-
Chambers D.J., Dunham J., Braimbridge M.V., Slavin B., Quiney J., Chayen J. The effect of ionized calcium, pH, and temperature on bioactive parathyroid hormone during and after open-heart operations. Ann Thorac Surg 1983;36:306-313.[Abstract/Free Full Text]
-
Rebeyka I.M., Robert A., Axford-Gatley R.A., Bush B.G. Calcium paradox in an in vivo model of multidose cardioplegia and moderate hypothermia. J Thorac Cardiovasc Surg 1990;99:475-483.[Abstract]
Accepted for publication April 3, 1999.
This article has been cited by other articles:
|
|
|
|
 
M. Ruel, T. A. Khan, P. Voisine, C. Bianchi, and F. W. Sellke
Vasomotor dysfunction after cardiac surgery
Eur J Cardiothorac Surg,
November 1, 2004;
26(5):
1002 - 1014.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
