Mechanisms responsible for cell volume regulation during hyperkalemic cardioplegic arrest

Mechanisms Responsible for Cell Volume Regulation During Hyperkalemic Cardioplegic Arrest Xiwu Sun, MD, PhD, Christopher T. Ducko, MD, Eric M. Hoenicke, MD, Karen Reigle, and Ralph J. Damiano, Jr, MD Division of Cardiothoracic and Vascular Surgery, The Milton S. Hershey Medical Center, Penn State University, Hershey, Pennsylvania

Background. Cardioplegia has been shown to induce significant cell swelling. This study tested the hypothesis that (1) the [Kⴙ][Clⴚ] product of the cardioplegia solution is the main determinant of myocyte swelling, and (2) reperfusion myocyte shrinkage results from a rectifying Clⴚ conductance. Methods. Rabbit ventricular myocytes were superfused with 37°C Krebs-Henseleit solution for 10 minutes. Then cells underwent 20 minutes of superfusion with standard St. Thomas’ solution ([Kⴙ][Clⴚ] product ⴝ 2566 mmol/L2) and two solutions with lower [Kⴙ][Clⴚ] product (1500 and 700 mmol/L2) at 9°C. Cells were then resuperfused with 37°C Krebs-Henseleit solution for 30 minutes. Cell volume was measured by videomicroscopy.

Results. Cells superfused with St. Thomas’ having [Kⴙ][Clⴚ] products of 2,566, 1,500, and 700 mmol/L2 swelled by 9.18% ⴞ 3.57%, 5.51% ⴞ 1.08%, and 1.49% ⴞ 1.56%, respectively. Reexposure to Krebs-Henseleit solution caused these cells to shrink by 5.79% ⴞ 1.41%, 8.72% ⴞ 3.68%, and 13.46% ⴞ 5.60%, respectively. This shrinkage was blocked by Clⴚ channel blockers given at the onset of superfusion. Conclusions. Lowering the [Kⴙ][Clⴚ] product of St. Thomas’ solution attenuated myocyte edema. Myocyte shrinkage during reexposure to Krebs-Henseleit solution resulted from the volume-activated Clⴚ channel. (Ann Thorac Surg 2000;70:633– 8) © 2000 by The Society of Thoracic Surgeons

H

mic cardioplegia for short periods of time (20 minutes) exhibit significant cell swelling in the absence of ischemia [10, 12, 13]. The purpose of this study was to elucidate the mechanisms responsible for cell volume regulation during hypothermic cardioplegic administration and reperfusion. During hypothermic perfusion, the active transport pumps normally responsible for maintaining cell volume regulation are inhibited. It has been hypothesized that myocyte cell volume is governed under these condition by the passive flux of ions across the cell membrane. Because the permeability of the membrane to K⫹ and Cl⫺ ions is high, the [K⫹][Cl⫺] product of the cardioplegia solution has been proposed to govern the amount of cell swelling according to a Donnan equilibrium [12]. Our preliminary data indicated that a reduction in the chloride concentration of traditional potassium cardioplegia ameliorated cell swelling. The first aim of this study was to test the hypothesis that the amount of cell swelling that occurs during hypothermic hyperkalemic cardioplegic arrest is related to differences in the [K⫹][Cl⫺] product of the cardioplegia solution. In recent years, it has been demonstrated that volumeregulated chloride channels are ubiquitous in mammalian cells [14]. Several studies using the patch clamp technique have shown that cell swelling can result in the activation of outward rectifying chloride channels in an attempt to maintain cell volume [15–17]. Previously, we

ypothermic hyperkalemic cardioplegia has been the foundation of most myocardial protection strategies since its introduction more than 25 years ago. It has been responsible for many of the remarkable advances in cardiac surgery [1]. Despite excellent results, these solutions are far from perfect. This is perhaps best illustrated by the innumerable modifications in both their chemical composition and method of delivery during the past two decades. It is our basic hypothesis that hypothermic hyperkalemic cardioplegia itself can result in significant physiologic derangements in the absence of ischemia. Under these conditions, metabolic intermediates and mitochondrial respiration are disturbed [2, 3], left ventricular function is depressed [2, 4, 5], rhythm and conduction abnormalities are prevalent [6, 7], and intracellular calcium overload occurs [8, 9]. Myocardial edema and cell swelling have been observed with both crystalloid and blood cardioplegia [4, 5, 10]. Cellular edema has been believed to contribute to abnormal left ventricular function [4 –7], slowed conduction and arrthymogenesis [7], and decreased coronary flow upon reperfusion [5, 11]. During hypothermic perfusion, our laboratory and others have shown that myocytes exposed to hyperkaleAccepted for publication Feb 23, 2000. Address reprint requests to Dr Damiano, Division of Cardiothoracic Surgery, The Washington University School of Medicine, Queeny Tower, Suite 3108, One Barnes Jewish Hospital Plaza, St. Louis, MO 63110-1013; e-mail: [email protected].

© 2000 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

0003-4975/00/$20.00 PII S0003-4975(00)01403-X

634

SUN ET AL SINGLE CELL VOLUME AND CARDIOPLEGIA

have observed significant cell shrinkage after exposure to hypokalemic cardioplegia [13]. We hypothesized that the Cl⫺ conductance triggered in response to cardioplegiainduced cell swelling resulted in the observed cell shrinkage. Therefore, the second aim of this study was to test the hypothesis that cell shrinkage can be prevented by blockade of the volume-sensitive Cl⫺ channel.

Material and Methods Adult New Zealand white rabbits of either sex, weighing 2.5 to 3.5 kg, were used in the study. All animals received humane care in AAALAC, USDA registered (#23-R-02) facilities in compliance with the “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH publication No. 85-23, revised 1985).

Isolation of Ventricular Cardiomyocytes Left ventricular cardiomyocytes were obtained after enzymatic dissociation. Rabbits were anesthetized using zylazine (23.3 mg/kg), acepromazine maleate (1.3 mg/kg), and ketamine (83.3 mg/kg). After intravenous heparinization (400 U/kg), the chest was opened and the heart quickly removed and mounted, through the ascending aorta, on the Krebs-Henseleit (K-H) perfusion apparatus. Blood was flushed from the coronary vasculature with Krebs solution, containing in millimoles: NaCl 118, KCl 4.7, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25, glucose 11, and plasma level of 20 amino acids [18]. The pH of the solution was 7.35 to 7.45, and was saturated with mixed gas 95% O2/5% CO2 at 37°C. The buffer passed through the heart a single time and was discarded. After 10 to 15 minutes of perfusion at a pressure of 90 mm Hg, the heart assumed a regular rate of contraction, recovering from the period of anoxia associated with removal from the rabbit. The perfusate was then changed to an oxygenated nominally calcium-free bicarbonate buffer containing in millimoles: NaCl 120, KCl 2.7, KH2PO4 1.4, MgSO4 1.4, NaHCO3 28, glucose 15, taurine 60, and creatine 20 for 6 minutes. The heart ceased to contract within 15 seconds in this buffer and remained in diastolic arrest until the end of the perfusion period. Collagenase type II (0.5 mg/mL; Worthington Biomedical Corporation, Freehold, NJ), hyaluronidase (0.3 mg/mL, Sigma, St. Louis, MO), bovine serum albumin 0.1%, and CaCl2 (final concentration 50 ␮mol/L) were then added to the calcium-free bicarbonate buffer, and the buffer was recirculated through the heart for an additional 15 minutes. After the enzymatic digestion, the atrium, right ventricle, and septum were discarded, and the left ventricular free wall was sliced vertically toward the apex and cut into 1to 3-mm pieces. The minced tissue was placed in the collagenase-containing perfusate, which had been supplemented with 0.5% bovine serum albumin and gassed with 95% O2/5% CO2, and the mixture digested in 37°C shaking water bath for 10 minutes. The digestion mixture

Ann Thorac Surg 2000;70:633– 8

was triturated 5 to 10 times using a 10-mL pipette. After digestion, large debris was manually removed and the cells were centrifuged for 1 minute at 100 g. The cells were washed two times with a calcium-free bicarbonate buffer, and then washed another three times with a mixture of K-H solution and the calcium-free bicarbonate buffer solution, increasing the CaCl2 concentration sequentially to 1.25 mmol/L by increasing the proportion of Krebs in the mixture. Finally, the cells were stored in cell culture medium (M199 with 100 IU/mL streptomycin, 100 ␮g/mL penicillin, pH 7.4) saturated with 95% O2 and 5% CO2 at 37°C for up to 2 to 4 hours until used for imaging. The suspensions of cardiomyocytes were more than 50% viable as estimated by their elongated, rodshaped morphology.

Imaging The isolated myocytes were placed in a laminin-coated custom-made chamber, constructed from a cover slip and Lucite sidewalls. The myocytes were allowed to settle for 10 minutes and the chamber was perfused with the K-H solution at 37°C. The temperature of the perfusate was monitored during the experiment with a thermistor (model Bat 8; Bailey Instruments, Saddle Brook, NJ). The chamber was placed on an inverted microscope (Leitz, Wetzlar, Germany) equipped with Hoffman modulation optics. The cell images were displayed on a video monitor with 800-line resolution (Hitachi VM-1220; Hitachi, Tokyo, Japan) by a high-resolution television camera (Hitachi HP-101A). The total magnification of the videooptical system was X1,912 with a 40x objective. Cells were inspected for viability, clear striations, sharp borders, and smooth surface. Cell volumes were determined by the method described by Drewnoska and colleagues [12]. The myocyte images were captured using custom software and a video frame grabber (Targa 16/32, Truevision, Indianapolis, IN) by a Pentium 120-Mhz personal computer. The resolution of the digitization was 0.24 ␮m/ pixel. The cell borders were traced using JAVA image analysis software (JAVA; Jandel Scientific Corp, Corte Madera, CA). Contrast enhancement, image magnification, and an edge tracing algorithm identified the borders of the cell with the assistance of the operator. Cell volume was determined using a custom ASYST program (Keithley Asyst, Rochester, NY). Assuming that the changes in cell width and thickness were proportional, relative cell volume was determined as: Volume (test)/Volume (control) ⫽ [Area (test) ⫻ Width (test)/[Area (control) ⫻ Width (control)]. The calculations of volume made the assumption that the cells are brick-shaped with equal thickness and width. If the cross-section was instead cylindrical, the absolute cell volumes were overestimated by a factor of 4/␲, 1.27. To avoid this uncertainty in the remaining results, cell volumes were expressed relative to control. On the basis of repeated measurements of single images and measurements of multiple images of a cell, the estimates of cell volume have been shown to be reproducible to less than 1% [19].

Ann Thorac Surg 2000;70:633– 8

SUN ET AL SINGLE CELL VOLUME AND CARDIOPLEGIA

635

Superfusion Procedures The superfusates were placed in temperature-controlled water jackets before infusion and bubbled with 95% O2 and 5% CO2. All solutions were delivered at 4 to 5 mL/min. Myocytes were superfused with the control K-H solution for 10 minutes at 37°C to establish baseline cell volume, followed by a 20-minute superfusion with either the study solutions or the control K-H solution at 9°C. After 20 minutes of 9°C superfusion, the myocytes were resuperfused with 37°C K-H. Ten myocytes, each from a different rabbit, were studied in each group. The study solutions included A: standard St. Thomas’ solution consisted of (in millimoles per liter): NaCl 110, NaHCO3 10, KCl 16, MgCl2 16, CaCl2 1.2, [K⫹][Cl⫺] ⫽ 2560 mmol/L2; B: low Cl⫺ (43.75 mmol/L) St. Thomas’ solution prepared by equimolar replacement of NaCl with Namethanesulfonate, [K⫹][Cl⫺] ⫽ 700 mmol/L2; and C: mid Cl⫺ (93.75 mmol/L) St. Thomas’ solution prepared by partial replacement of NaCl with Na-methanesulfonate, [K⫹] ⫻ [Cl⫺] ⫽ 1500 mmol/L2. Myocyte images were obtained every 5 minutes throughout the study. Methanesulfonate, a monovalent anion, has been used as a chloride surrogate to probe the role of anions in arrhythmogenesis and contractile function. Compared to other anions, such as nitrate, iodide, bromide, and chloride, methanesulfonate is impermeable and has a compatible arrhythymogenesis with chloride [20]. The present study used its low membrane permeability to test our hypothesis that the chloride concentration of the cardioplegic solution is a major determinant of cell swelling. In another group of cell studies, after superfusion with standard St. Thomas’, Cl⫺ channel blockers (glyburide and clofibric acid) were applied during reperfusion to test our hypothesis that Cl⫺ outward flow was involved in the cell shrinkage process. Glyburide has been known to be a potent inhibitor for the ATP-sensitive K⫹ channels. It also has been reported that at a higher concentration (500 ␮mol/L), glyburide totally blocked the Cl⫺ current [15]. Clofibric acid has also been reported as a selective blocker of the cystic fibrosis transmembrane conductance regulator Cl⫺ channel, which is closely related to the outwardly rectifying Cl⫺ channels [17, 21].

Statistics Data are expressed as mean ⫾ standard error. Analysis of variance was used for comparison among the different groups (SigmaStat; Jandel Scientific, San Rafael, CA). Comparisons of treatment to control were done using the paired t test. Statistical significance was defined as p value less than 0.05.

Results ⫹

⫺

The Effects of Different [K ][Cl ] Products on Cell Width and Length The width and length of rabbit ventricular myocytes were measured after the cells were equilibrated at 37°C in a physiologic solution for 10 minutes, during superfusion

Fig 1. Relative cell volume changes of rabbit ventricular myocytes during 20 minutes in 9°C St. Thomas’ solution and 9°C KrebsHenseleit (K-H) solution perfusion, and reperfusion in 37°C KrebsHenseleit solution. Volumes are normalized by the initial volume in 37°C Krebs-Henseleit solution (the first three circles). Myocyte swelled in 9°C St. Thomas’ solution and shrank to less than their initial volume on reperfusion in Krebs-Henseleit solution (n ⫽ 8). *p ⬍ 0.05, **p ⬍ 0.01, respectively, versus the initial volume. Hypothermic and reperfusion in Krebs-Henseleit solution did not affect the cell volume (n ⫽ 10).

with the different study solutions, and after reperfusion. There were no differences in cell width and length among the groups at baseline. Both cell width and length increased after perfusion with the study solutions, and then shrank on reperfusion with 37°C K-H solution.

The Effects of St. Thomas’ Cardioplegia on Cell Volume ([K⫹] ⫻ [Cl⫺] ⫽ 2566 mmol/L2) Rabbit ventricular myocytes quickly swelled after exposure to hypothermic St. Thomas’ solution (Fig 1). Cell swelling was found to be significant at 1 minute with an increase in cell volume of 5.25% ⫾ 0.44% ( p ⬍ 0.001). Relative cell volume increased by 9.18% ⫾ 3.57% ( p ⬍ 0.05) and 7.05% ⫾ 1.38% ( p ⬍ 0.001) at 5 and 15 minutes of perfusion, respectively. Myocytes significantly shrank after 10 minutes of reperfusion, and remained so throughout reperfusion. The relative cell volume at 10, 20, and 30 minutes of reperfusion were ⫺2.45% ⫾ 0.98%, ⫺4.71% ⫾ 1.19%, and ⫺5.79% ⫾ 1.41%, respectively. To assure that the volume changes observed were due to the exposure of the myocytes to St. Thomas’ solution, and not to hypothermia alone, a group of rabbit ventricular myocytes (control group, n ⫽ 10) were exposed to hypothermic K-H solution. There were no significant changes in cell volume in this group (Fig 1).

The Effects of Low Cl⫺ St. Thomas’ on Cell Volume On exposure to the mid Cl⫺ St. Thomas’ cardioplegia, the cells also swelled, but to a lesser extent than with standard St. Thomas’ solution (Fig 2A). When the Cl⫺ concentration was lowered to that of plasma ([K⫹][Cl⫺] product 700 mmol/L2), rabbit ventricular myocytes

636

SUN ET AL SINGLE CELL VOLUME AND CARDIOPLEGIA

Ann Thorac Surg 2000;70:633– 8

Fig 2. Relative cell volume change of rabbit ventricular myocytes during 20 minutes in 9°C low Cl⫺ St. Thomas’ solution perfusion and reperfusion with 37°C Krebs-Henseleit (K-H) solution. Volumes are normalized by the initial volume perfused in 37°C Krebs-Henseleit solution. (A) Ventricular myocytes swelled during 9°C mid-Cl⫺ St. Thomas’ ([K⫹][Cl⫺] ⫽ 1500 mmol/L2) perfusion, and shrank to less than the initial volume during reperfusion with 37°C KrebsHenseleit solution (n ⫽ 8). (B) Rabbit ventricular myocytes swelling and shrinkage during 9°C low Cl⫺ St. Thomas’ ([K⫹][Cl⫺] ⫽ 700 mmol/L2) perfusion and reperfusion with 37°C Krebs-Henseleit solution (n ⫽ 8). *p ⬍ 0.05, **p ⬍ 0.01 versus the initial volume. (C) The combined graph shows the relative swelling and shrinkage during hypothermal perfusion with different [K⫹][Cl⫺] product solutions and reperfusion with 37°C Krebs-Henseleit solution. *p ⬍ 0.05 versus low Cl⫺ St. Thomas’ ([K⫹][Cl⫺] ⫽ 700 mmol/L2) solution. 4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™

was the same in the different study solutions, the shrinkage may have been related to [Cl⫺]. Recent studies have shown a volume-regulating Cl⫺ channel on myocytes [16]. Therefore, we tested the hypothesis that the Cl⫺ channel may have played a role in cell shrinkage. A group of cells were superfused with hypothermic St. Thomas’ for 20 minutes as described above, and a Cl⫺ channel blocker (glyburide, 500 ␮mol/L) was added to the reperfusion medium. The shrinkage of the cells was prevented by glyburide (Fig 3A). To confirm our finding, we also examined clofibric acid (200 ␮mol/L), a more specific Cl⫺ channel blocker, added to the reperfusion medium. Again, the cell shrinkage was prevented by Cl⫺ channel blockade (Fig 3B). These results indicated that Cl⫺ efflux through the membrane Cl⫺ channel was at least partially responsible for this phenomenon.

Comment [K⫹][Cl⫺] Product and Myocyte Edema

showed little edema when exposed to St. Thomas’ solution (Fig 2B). Upon reperfusion, the rabbit ventricular myocytes shrank significantly (Fig 2B,C), and remained so throughout the reperfusion period.

The Effects of Cl⫺ Channel Blockade on Cell Shrinkage The magnitude of cell shrinkage varied with the [K⫹][Cl⫺] product of the cardioplegia. Because the [K⫹]

Although hypothermic hyperkalemic cardioplegia is intended to protect the myocardium during elective cardiac arrest [1], it has been shown that exposure to 9°C St. Thomas’ solution rapidly induces significant cellular edema, in the absence of ischemia, in rabbit ventricular myocytes [12], as well as rabbit and human atrial myocytes [13]. The [K⫹][Cl⫺] product of the solution has been thought to be critical to the development of cellular edema during hypothermia. Previous work in our laboratory has demonstrated that reducing the [K⫹][Cl⫺] product, by the substitution of the impermeant anion methanesulfonate for Cl⫺, to that of physiologic solution (700 mmol/L2), reduced or eliminated the myocyte edema [12, 13]. This study was designed to further examine this phenomenon by studying other [K⫹][Cl⫺] products. Our results demonstrated that the [K⫹][Cl⫺] product was related to cell swelling. This supports our hypothesis that the higher the [K⫹][Cl⫺] product in the cardioplegic solution, the greater the amount of cellular edema with hypothermic cardiac arrest. Although only the lowest [Cl⫺] group was significantly different than control, the intermediate group approached significance ( p ⫽ 0.054). Whereas a direct proportional relationship cannot be

Ann Thorac Surg 2000;70:633– 8

SUN ET AL SINGLE CELL VOLUME AND CARDIOPLEGIA

637

where the bracketed chemical symbols refer to the intracellular i and the extracellular o concentrations, R is the gas constant, T is the temperature (degrees Kelvin), and F is Faraday’s constant R. As a consequence of these relationships, increasing the [K⫹][Cl⫺] product of the extracellular solution should lead to an accumulation of ions within the cell, and water will follow osmotically [19]. Cell swelling has significant pathophysiologic sequelae. In an isolated Langendorff preparation, we have shown that cellular swelling plays an significant role in myocardial stunning after cardioplegic arrest [22]. The role of cellular swelling in ischemic injury has also been clearly established in the nonsurgical setting [23]. To test our hypothesis that the [Cl⫺] of the superfusate is critical to cell volume regulation, the impermeant anion methanesulfonate was substituted for chloride. This preserved the osmolarity of the solution, but lowered its tonicity. A weakness of the study is that methanesulfonate may have had a direct effect on preventing cell swelling. However, preliminary work in our laboratory with another impermeant anion, aspartate, has revealed similar results [22]. This suggests that the prevention of swelling is not significantly related to methanesulfonate, but is seen with any impermeant ion.

[K⫹][Cl⫺] Product and Cell Shrinkage

Fig 3. Relative cell volume change during 20 minutes in 9°C St. Thomas’ solution perfusion and reperfusion with 37°C Krebs-Henseleit (K-H) solution contain 500 ␮mol/L glyburide and 200 ␮mol/L clofibric acid respectively (n ⫽ 8 for each group). Volumes are normalized by the initial volume in 37°C Krebs-Henseleit solution (the first three filled circle). The cells quickly swelled during 9°C St. Thomas’ perfusion, shrinkage were prevented by both glyburide (A) and clofibric acid (B). *p ⬍ 0.05, **p ⬍ 0.01 versus the initial volume.

inferred from our data, the trend clearly supports our basic hypothesis. Unfortunately, the study is underpowered to pick up small differences between the groups. The reason for these volume changes can be explained by Donnan equilibrium theory. Under hypothermic conditions, the transport processes responsible for normothermic cell volume regulation are inhibited, and cell volume is mediated by the passive fluxes of ions. In the Donnan equilibrium system, the membrane potential (Em) and the Nernst equilibrium potentials for K⫹ and Cl⫺ are equal. Writing this relationship and simplifying gives:

Em ⫽ ⫺

RT 关K⫹兴i RT 关Cl⫺兴o ln ⫹ ⫽ ⫺ ln F 关K 兴o F 关Cl⫺兴i

[K⫹]o ⫻ [Cl⫺]o ⫽ [K⫹]i ⫻ [Cl⫺]i

Cell shrinkage was found during the normothermic reperfusion period in each group. Although the shrinkage was not statistically different between groups at most time points, there was a trend to suggest that the lower the [K⫹][Cl⫺] product of the cardioplegic solution, the greater the degree of cell shrinkage. Cell volume regulation during normothermic conditions is more complex than a Donnan equilibrium, and likely is the result of an interaction between multiple ion transport systems and second messengers. Under normothermic conditions, the Na⫹/K⫹/2Cl⫺ and Na⫹/Cl⫺ cotransport systems [19, 22, 24] have been implicated in volume regulation. Previous studies in both our laboratory and by others have demonstrated that cellular shrinkage during reperfusion is related to Na⫹ efflux [12]. However, this did not explain the apparent relationship between cell shrinkage and the [K⫹][Cl⫺] product of the cardioplegic solution. Recently, a volume-regulated Cl⫺ channel has been identified as an outward rectifying Cl⫺ current [15–18]. Activation of a chloride conductance during cell swelling has been suggested to be an important volume regulatory process [16]. We tested the hypothesis that Cl⫺ efflux through the Cl⫺ channel was involved in reperfusion cell shrinkage. Both glyburide and clofibric acid completely inhibit the volume-regulated Cl⫺ channel activity [15, 16]. In the present study, Cl⫺ channel blockers added to the reperfusion solution-blocked the cellular shrinkage, indicating that chloride efflux through channels contributed to the shrinkage. However, it is worth noting that glyburide is also a potent inhibitor for the ATP-sensitive K⫹ channel. Although this drug totally blocks Cl⫺ conductance, it is possible that blockade of the ATP-sensitive K⫹ channel may effect cell volume. However, a previous study in our laboratory showed no effects of ATP-

638

SUN ET AL SINGLE CELL VOLUME AND CARDIOPLEGIA

sensitive K⫹ channel openers on cell volume [25]. The precise mechanisms involved in cellular shrinkage after cardioplegia infusion will require further studies to delineate the respective contributions of the Cl⫺ channel, the Na⫹/K⫹ pump, and other ion transport systems.

Model Limitation The isolated myocyte preparation has been well accepted in the literature for its utility in providing insights into the mechanisms of cardioplegic injury at the cellular level [10, 12]. Studying isolated myocytes has the advantage of allowing repeated measurements of cell volume, and the volume response to differing solutions, in the absence of ischemia. However, isolated myocytes in a flowing solution do not reflect the complex geometry of the arrested heart wherein extracellular volume is limited. Care must be taken when extrapolating these data to the clinical situation. Also in the isolated myocyte model, the role of vascular, neural, and interstitial elements are ignored. For instance, the high K⫹ in the cardioplegic solution may directly damage vascular endothelium [22]. Using an isolated myocyte, the present study demonstrated that the [K⫹][Cl⫺] product of the cardioplegia is the critical determinant of cellular edema. Decreasing the [K⫹][Cl⫺] product of St. Thomas’ solution by substituting the Cl⫺ with an impermeant anion can prevent myocyte edema. This study also showed that significant cell shrinkage occurred during reperfusion, and that the shrinkage was due to Cl⫺ efflux through the volumeregulated [Cl⫺] channel. Further investigation is needed at the whole heart level to evaluate the physiologic significance of cellular edema/shrinkage on ventricular mechanics. This study was supported by NIH grant HL-51032 (Ralph J. Damiano, Jr) and NIH NRSA grant HL-09925 (Christopher T. Ducko, Ralph J. Damiano, Jr).

References 1. Hearse DJ, Baimbridge MV, Jynge P. Protection of ischemic myocardium: cardioplegia. New York: Raven Press, 1981. 2. Fremes SE, Weisel RD, Mickle DAG, et al. Myocardial metabolism and ventricular function following cold potassium cardioplegia. J Thorac Cardiovasc Surg 1985;89:531– 46. 3. Engelman RM, Rousou JH, Lemeshow S, Dobbs WA. The metabolic consequences of blood and crystalloid cardioplegia. Circulation 1981;64(Suppl 2):67–74. 4. Hsu DT, Weng ZC, Nicolosi AC, Detweiler PW, Sciacca R, Spotnitz HM. Quantitative effects of myocardial edema on the left ventricular pressure-volume relationship. Influence of cardioplegia osmolarity over two hours of ischemic arrest. J Thorac Cardiovasc Surg 1993;106:651–7. 5. Wang ZC, Nicolosi AC, Detwiler PW, 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;103:504–13. 6. Cohen NM, Allen CA, Belz MK, Nixon TE, Wise RM,

Ann Thorac Surg 2000;70:633– 8

7.

8.

9. 10.

11. 12.

13.

14. 15. 16. 17. 18. 19. 20. 21. 22.

23.

24. 25.

Damiano RJ Jr. Electrophysiological consequences of hypothermic hyperkalemic elective cardiac arrest. J Card Surg 1993;8:156– 60. Cohen NM, Allen CA, Hsia PW, Nixon TE, Wise RM, Damiano RJ Jr. Electrophysiological consequences of hyperkalemic cardioplegia during surgical ischemia. Ann Thorac Surg 1994;57:1076– 83. Jovanovic A, Alekseev AE, Lopez JR, Shen WK, Terzic A. Adenosine prevents hyperkalemia induced calcium loading in cardiac cell: relevance for cardioplegia. Ann Thorac Surg 1997;63:153– 61. Lopez JR, Ghanbari RA, Terzic A. KATP-channel opener protects cardiomyocytes from Ca2⫹ waves: a laser confocal microscopy study. Am J Physiol 1996;270:H1354 –H89. Handy JR Jr, Dorman BH, Cavallo MJ, et al. Direct effects of oxygenated crystalloid or blood cardioplegia on isolated myocyte contractile function. J Thorac Cardiovasc Surg 1996; 112:1064–72. Rubboli A, Sobotka PA, Euler DE. Effect of acute edema on left ventricular function and coronary vascular resistance in the isolated rat heart. Am J Physiol 1994;267:H1054 – 61. Drewnowska K, Clemo HF, Baumgarten CM. Prevention of myocardial intracellular edema induced by St. Thomas’ Hospital cardioplegic solution. J Mol Cell Cardiol 1991;23: 1215–21. Shaffer RF, Baumgarten CM, Damiano RJ Jr. Prevention of cellular edema directly caused by hypothermic cardioplegia: studies in isolated human and rabbit atrial myocytes. J Thorac Cardiovasc Surg 1998;115:1189–95. Duan D, Winter C, Cowley S, Hume JR, Horowitz B. Molecular identification of a volume regulated chloride channel. Nature 1997;390:417–21. Sakaguchi M, Matsuura H, Ehara T. Swelling-induced Cl⫺ current in guinea-pig atrial myocytes: inhibition by glibenclamide. J Physiol 1997;505:41–52. Zhang JP, Lieberman M. Chloride conductance is activated by membrane distention of cultured chick heart cells. Cardiovasc Res 1996;32:168–79. Walsh KB, Wang CM. Effect of chloride blocker on the cardiac CFTR chloride and L-type calcium currents. Cardiovasc Res 1996;32:391–9. Cheung JY, Thompson IG, Bonventre JV. Effects of extracellular calcium removal and anoxia on isolated rat myocytes. Am J Physiol 1982;243:C184 –90. Drewnowska K, Baumgarten CM. Regulation of cellular volume in rabbit ventricular myocytes: bumetanide, chlorthiazide, and ouabain. Am J Physiol 1991;260:C122–31. Curtis MJ, Garlick PB, Ridley PD. Anion manipulation, a novel antiarrhythmic approach: mechanism of action. J Mol Cell Cardiol 1993;25:417–36. Schwiebert EM, Egan ME, Hwang TH, et al. CFTR regulates outwardly rectifying chloride channels through an autocrine mechanism involving ATP. Cell 1995;81:1063–73. Jayawant AM, Stephenson ER Jr, Baumgarten CM, Damiano RJ Jr. Prevention of cell swelling with low chloride St. Thomas’ solution improves post-ischemic myocardial recovery. J Thorac Cardiovasc Surg 1998;116:131– 8. Tranum-Jensen PT, Janse MJ, Fiolet JWT, Krieger WJG, D’Alnoncourt CN, Durrer D. Tissue osmolarity, cell swelling, and reperfusion in acute regional myocardial ischemia in the isolated porcine heart. Circ Res 1981;49:364– 81. Desilets M, Baumgarten CM. K, Na, Cl activity in ventricular myocytes isolated from rabbit heart. Am J Physiol 1986;251: C197–208. Stephenson ER, Ducko CT, Damiano RJ Jr. The myocyte edema associated with hyperkalemic cardioplegia does not occur during hyperpolarized arrest with pinacidil. Circulation 1998;98:I3605.