- Email: [email protected]
Imidazole-buffered cardioplegic solution
J
THoRAc CARDIOVASC SURG
90:225-234, 1985
Imidazole-buffered cardioplegic solution Improved myocardial preservation during global ischemia Progressive acidosis is a constant rmding in global myocardial ischemia and is associated with reduced myocardial contractility after ischemia. The hypothesis tested in these experiments was that imidazole (pKa = 6.7 at 37° C), a commonly used buffer in physiology and microbiology, would provide superior buffering capacity when used in lieu of bicarbonate (pKa = 6.1 at 37° C) in a cardioplegic solution. Twenty-eight isolated, working rabbit hearts were perfused, and preischemic and postischemic determinants of performance were measured. The 30 minute interval of normothermic global ischemia was altered by the injection at 0 and 15 minutes of 2 m1/gm wet weight of a buffered cardioplegic solution. Control bearts received a bicarbonate-buffered cardioplegic solution and experimental hearts receiveda solution buffered with imidazole. In the imidazole-buffered group, there was a superior recovery of coronary flow, developed left ventricular pressure, peak rate of rise of left ventricular pressure, peak rate of relaxation, andstroke work indices (p < 0.05). Recovery of mechanical parameters was coincident with an improved acid-base status of the coronary sinus effluent at the end of ischemia. Coronary sinus effluents in the imidazole group had significantly higher pH values and lower partial pressures of carbon dioxide than coronary sinus effluents in the bicarbonate-buffered group (p < 0.001). The data suggest that improved buffering of the extracellular and possibly intracellular space during global ischemia with a nonbicarbonate buffer is beneficial and provides improved postischemic myocardial recovery.
John C. Vander Woude, M.D., Ignacio Y. Christlieb, M.D., Gregorio A. Sicard, M.D., and Richard E. Clark, M.D., St. Louis, Mo.
In the present era of open cardiac operations most surgeons use some form of chemical cardioplegia. A variety of cardioplegic solutions are in clinical use, employing different concentrations of constituents such as sodium, potassium, and buffering agents. Many of these solutions have never been tested under experimental conditions. I The differences in composition of current solutions are not surprising given the large gaps in our understanding of the mechanisms of cell death in global ischemia.' The heart requires a continuous supply of oxygen to generate energy for contraction and for maintenance of its cellular structure and function.' During the period of From the Divisions of Cardiothoracic and General Surgery, Department of Surgery, Washington University School of Medicine, St. Louis, Mo. Received for publication Aug. 30, 1984. Accepted for publication Oct. 17, 1984. Address for reprints: John C. Vander Woude, M.D., Department of Surgery, 910 Wohl Hospital, 4960 Audubon Ave., St. Louis, Mo. 63110.
aortic cross-clamping and cardioplegic arrest in cardiac operations, metabolic perturbations occur. Myocardial oxygen reserves are depleted within 8 seconds after the initiation of ischemia, precluding continued oxidative phosphorylation in the mitochondria.' Concentrations of lactic acid rise markedly and intracellular pH falls. Glycolysis, an inefficient, anaerobic pathway of adenosine triphosphate production, is rate limited because phosphofructokinase and glyceraldehyde 3-(P) dehydrogenase activities are inhibited by the resultant intracellular acidosis." 5 Bretschneider and associates" have shown that the small energy output of anaerobiosis can maintain prolonged myocardial viability if cellular glycolytic metabolism can continue unimpeded. Investigators have shown salutory effects on ischemic myocardium by interventions within the glycolytic pathway, either by providing substrate such as fructose diphosphate or by enhancing buffering capacity with HEPES or tricine buffers.> 7 The hypothesis tested in these experiments was that imidazole (pKa = 6.7 at 37° C), a buffer not dependent on bicarbonate--carbon dioxide flux, would provide
225
The Journal of Thoracic and Cardiovascular Surgery
226 Vander Woude et al.
100
AORTIC~ o 25
LVEDP
o
EKG
100
LV
o
dP/dt
WN/0M
+2500
-2500
Fig. 1. Representative hemodynamic tracings obtained in one experiment demonstrating aortic pressure, left ventricular end-diastolic pressure (LVEDPj. electrocardiographic tracing (EKG). left ventricular pressure (LV), and ± rate of pressure development (± dPjdt). superior buffering capacity during global myocardial ischemia when used in place of bicarbonate (pKa = 6.1 at 37 0 C) in a cardioplegic solution. The rationale for the substitution was twofold. Henderson" pointed out in 1908 that a weak acid exerts maximal buffering when its dissociation constant (pKa) is equal to or near the pH of neutrality. Second, the buffering capacities of TRIS, imidazole, and HEPES have been shown to be independent of the partial pressure of carbon dioxide (Pco.), Since the pKa of the carbonic acid-bicarbonate-carbon dioxide system is 6.10, in the absence of constant replenishment of bicarbonate or the removal of carbon dioxide, bicarbonate solutions do not function as effective buffers." IO
Materials and methods Perfusion technique. These studies were performed with 28 isolated working hearts obtained from 2.5 ± OJ kg New Zealand white rabbits. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" as formulated by the National Society for Medical Research. The animals were anesthetized with intravenous pentobarbital, 30 mg/kg, 5 minutes after systemic heparinization (2,000 units per animal). Hearts were extracted rapidly through a median sternotomy, placed in iced saline, and transferred within 1 minute to an isolated heart perfusion apparatus similar to that described by Neely and colleagues.I I Retrograde perfusion was established at 38 0 C at a pressure of 48 to 49 mm Hg by the method of Langendorff." The hearts were perfused with a modified Krebs-Henseleit solution augmented with 1% bovine serum albumin and insulin, 400 mU /L; perfusate was filtered with a 0.45 ~m Millipore filter prior to use. The concentrations of the perfusate constituents (mm/ L) were as follows: NaCl 118, KCl 4J, cso, . 2H 20 2.5, MgS04 • 7H 20 1.25, KH 2P04 1.2, NaHC03 25.0, and dextrose 15.0. The perfusate was gassed with a mixture of 95% oxygen and 5% carbon dioxide through scintered glass of 15 ~m porosity to reduce foaming caused by albumin. This produced a partial pressure of oxygen (Po 2) of 450 to 525 mm Hg, a Pco, of 32 to 42 mm Hg, and a pH of 7.40 to 7.48. The entire apparatus was maintained at 38 0 C by water jackets surrounding all reservoirs and tubing. During the period of retrograde washout, a cannula was placed in the pulmonary artery and all pulmonary veins and hilar structures were ligated. A ribbed, rigid cannula, 6 mm inner diameter, was inserted into the left atrial appendage to allow left atrial inflow for the working heart perfusions. Within the inflow cannula were two, soft 0.031 inch polyethylene catheters. One terminated at the end of the inflow cannula and was used to measure left atrial pressure. The other tube extended beyond the tip of the inflow cannula and was placed through the mitral valve into the midventricle for measurement of left ventricular pressure. This latter catheter allowed high-fidelity recordings of peak rate of rise of left ventricular pressure (+LVdP/dt), peak rate of relaxation (-LVdP /dt), and left ventricular enddiastolic pressure (LVEDP) without an apex puncture or use of an isovolumic balloon (Fig. 1). The sinus node was crushed and the hearts were paced at 214 beats/min by twin electrodes sutured on the right atrium with a Grass SD-9 stimulator. The stimulus voltage was 3.0 V with a duration of 10 msec. Cardiac and solution temperatures were measured with thermistors (Yellow
Volume 90 Number 2 August. 1985
Springs Instrument Co., Yellow Springs, Ohio). Retrograde perfusion and cannulation required 20 minutes after which the heart was converted to the working mode. Perfusate from the left atrial reservoir flowed into the left atrium, and left ventricular contraction pumped the perfusate through the aortic circuit. A Starling apparatus fashioned from 15 em of V2 inch Penrose drain tubing and a glass chamber served as a compliance-resistance apparatus and was placed 20 em above the heart. The size and shape of the aortic pressure curve measured just distal to the aortic valve were altered by adjusting the air pressures in the Starling chamber. Once adjusted, mean aortic pressures were maintained at 49.0 ± 0.5 mm Hg over the range of ventricular loadingconditions (1 to 4 mm Hg). The working circuit had a recirculating volume of 1,000 rnl with aortic flow returning to the reservoir by gravity and coronary flow by means of a rotary pump. Left atrial inflow was filtered by an inline 13 tLm Pioneer filter. During a 20 minute work period, with rate and mean aortic pressure constant, ventricular function was assessed by increasing the LVEDP over the range of 1 to 4 mm Hg. A 3 to 5 minute period was required to achieve stability of the preparation at each LVEDP. The following measurements were recorded for each heart at each time point: left ventricular developed pressure; +LVdP/ dt; - LVdP/ dt; aortic and coronary flow; and reservoir and coronary sinus pH, Pco., and Paz. Phasic and mean left ventricular and aortic pressures were measured with Sanborn 267 transducers and the LVEDP was measured with a Hewlett-Packard 1280 transducer. Aortic flow was measured with a Biotronex 3.5 mm cannulatingflow probe. Coronary flow was measured volumetrically during the recording interval. The left ventricular pressure signal was differentiated on a Biotronex 620622 differentiator-integrating unit. The pH, Paz, and Pco, of the perfusate and coronary sinus effluents were determined with an IL-313 analyzer. All pressure tracings, differential signals, and bipolar electrocardiogramswere recorded on a Sanborn 8-channel strip-chart recorder at a paper speed of 100 mm/sec, Ischemic heart model. After initial work parameters had been measured, the hearts were subjected to a 30 minute period of normothermic global ischemia. Global ischemia was altered by an injection of 10 rnl (2 ml/gm wet weight per injection) of a buffered cardioplegic solution at initiation of and 15 minutes after onset of ischemia. Control hearts (n = 15) received a bicarbonate-buffered cardioplegic solution and experimental hearts (n = 13) received an imidazole-buffered cardioplegic solution. Composition of the two cardioplegic solutions is shown in Table I. The cardioplegic solutions
Imidazole-buffered cardioplegic solution
227
Table I. Composition of the cardioplegic solutions Control
Experimental
Na+ (mmol/L) K+ (rnrnol/L) Cl (rnrnol/L) HCO , (mmol/L) Imidazole (mmol/L) Dextrose (mmol/L)
143
135
Mg?" (mmol/L)
1.3 7 10
1.3
366
366
Constituent
Mannitol (rnrnol/L) O.IN HCI (rnl/L) Osmolality (calculated)
27
27
148 25
163
15
25 15 30
were designed to be as similar to each other as possible in terms of ionic content, pH, partial pressures, and osmolality. The buffers used were bicarbonate for the control group and an equimolar concentration of imidazole for the experimental group. At the onset of ischemia, the left atrial and aortic lines were clamped and 10 rnl of cardioplegic solution was infused into an aortic root cannula through a 0.45 tLm filter at 45 to 50 mm Hg pressure. All hearts stopped mechanical and electrical activity within 10 seconds. At the end of the ischemic period, hearts were reperfused at 48 to 49 mm Hg retrograde pressure while the left ventricular cavity was vented via the left ventricular pressure measurement catheter. The initial 10 rnl aliquot of reperfusate was aspirated from the pulmonary artery cannula, immediately sealed from ambient air, and analyzed for acid-base status. After the retrograde reperfusion period, hearts were converted to the working mode with fresh perfusate and postischemic indices of performance were obtained. The mean aortic pressure was maintained at 49.0 ± 0.5 mm Hg, the heart rate at 214 beats/min, and the LVEDP was varied from 1 to 4 mm Hg to obtain several data points that permitted construction of ventricular function curves for comparison of preischemic and postischemic performance. Analysis of coronary sinus effluent. Myocardial oxygen consumption was determined during the left ventricular loading tests. Recognizing the limited oxygen availability of assanguineous perfusate, we assessed ventricular performance in the limited range of 1 to 4 mm Hg LVEDP to prevent oxygen availability from being a limiting factor. Accordingly, Po.s of coronary sinus effluent were uniformly 95 to 115 mm Hg at higher LVEDPs. The sealed sample of the initial 10 rnl of reperfusate was measured for pH, Pco., and Paz and served as an indicator of the acid-base status of the extracellular space at the onset of reperfusion. As imidazole has pharmacologic activity by mechanisms other than buffering, including blockade of
228
The Journal of Thoracic and Cardiovascular Surgery
Vander Woude et al.
Table II. Hemodynamic data comparison of bicarbonate-buffered versus imidazole-buffered hearts Preischemia
HCO J (n = 60)
Parameter CFI (rnl/rnin/grn)
MVO, (ml/rnin/gm) DLVP (mm Hg) +LVdP/dt (mm Hg . sec') -LVdP/dt (mm Hg . sec-') SWI (gm-rn/gm)
63.8 0.677 88.9 +1,767 -1,534 1.10
I
Imidazole (n = 52)
Postischemia
HCO J (n = 60)
I
Pre-post difference
Imidazole (n= 52)
61.4 56.3 ± 1.0 ± 1.3 63.2 ± 1.4 ± 0.030 0.654 0.663 ± 0.032 0.610 ± 0.018 86.7 ± 1.1 82.6 79.1 ± 0.9 ± 0.9 +1,576 ± 25 + 1,589 ± 28 +1,672 ± 35 ± 26 -1,452 ± 30 -1,467 ± 18 -1,544 0.65 ± 0.03 1.10 ± 0.05 0.93 ± 0.05
HCO J (n = 60)
I
Imidazole (n = 52)
±. 1.1 7.4 ± 0.9 2.3 ± 0.8 ± 0.026 0.061 ± 0.021 0.016 ± 0.017 ± 1.0 10.1 ± 0.7 4.0 ± 0.7 ± 26 +200 ± 19 +81 ± 18 ± 25 -88 ± 31 +77 ± 30 ± 0.05 0.46 ± 0.03 0.19 ± 0.02
p Value pre-post 0.05
NS 0.01 0.05 0.05 0.001
Legend: Mean arterial pressure = 49 ± 0.5 mm Hg. Heart rate = 214 beats/min. Mean left ventricular end-diastolic pressure = 2.5 mm Hg. CFI, Coronary flow index. MVO" Myocardial oxygen consumption. DLVP, Left ventricular developed pressure. +L VdP /dt, Peak rate of rise of left ventricular pressure. -LVdP/dt, Peak rate of relaxation of left ventricular pressure. SWI, Stroke work index.
thromboxane synthetase," it was important to confirm the absence of thromboxane in the asanguineous effluents of control experiments. Reperfusion effluents in five control experiments were analyzed by radioimmunoassay for the concentration of thromboxane B2, the stable metabolite of thromboxane A 2• Reperfusate lactate determinations were performed in the experiments from each group as we wished to correlate acid-base conditions with concentrations of lactate, the end product of anaerobic, glycolytic energy production. Analysis was performed by standard spectrophotometric methods (Sigma Chemical Co., St. Louis, Mo.). Myocardial water content and hemodynamic calculations. At the conclusion of each experiment, the heart was blotted dry, debrided of all atrial and hilar tissue, and dried at 80° C to a constant dry weight (Mettler Analytical Scale, Mettler Instrument Corp., Hightstown, N. J.). Percent water content was calculated as: 100 X (wet weight - dry weightj/wet weight. Stroke work and myocardial oxygen consumption (MV02) were calculated from the following indices: Cardiac output Stroke work = - - - - = - Rate X (Mean aortic pressure - LVEDP) X 0.0136
MVO,
= Coronary
flow X Perfusate Po, - Coronary sinus Po, X 0.0031 *
Stroke work, coronary flow, and oxygen consumptions for each heart were divided by the dry weight to provide an index of comparison between groups. Data analysis. All indices of performance were paired for individual experiments regarding conditions of preload, afterload, and rate. Performance data for the *Oxygen solubility constant of perfusate at 37° C.
preischemic and postischemic states were then collated. These data were analyzed by analysis of variance, looking specifically at the preischemic minus postischemic differences. Group means for lactate concentrations and water contents were compared by the Student's t test for unpaired data. Values are expressed as means ± standard error of the mean. Differences were considered significant if the p value was less than 0.05. Results Hemodynamics. Table II provides data for preischemic and postischemic performances of hearts subjected to bicarbonate and imidazole buffering as well as the pre-post differences. The data are as stated for coronary flow indices, oxygen consumption, developed pressure (left ventricular pressure - LVEDP), +LVdP/ dt, -LVdP / dt, and stroke work indices. Analysis of preischemic performance indices showed no significant differences between the bicarbonate- and imidazole-buffered experiments. Examination of the pre-post differences showed improved recovery of the imidazole group in all parameters except for oxygen consumption (p = 0.1180). Coronary flow index after ischemia was decreased less from the preischemic level in the imidazole group than in the bicarbonate group. Developed pressure fell less after the ischemic period in the imidazole group. Contractility, as reflected by + LVdP / dt, recovered to a greater degree in the imidazole group. Diastolic relaxation, as reflected by the - LVdP / dt, was better maintained in the imidazole group and was actually greater (more negative) than its preischemic control value. Most significantly, the stroke work index of the imidazole group after the ischemic period decreased by only 16% compared to a decrease of 41% in the bicarbonate group (p < 0.001). Fig. 2 provides these results in bar graph form.
Volume 90
Imidazole-buffered cardioplegic solution
Number 2 August, 1985
70
CFI ml/min/gm
0.700
50
MV02 ml/min/gm
90
SWI gm'M/gm
mmHg
120
(+)dP/dt rnmrlq-sec -I 1800
(-)dP/dt mmHg'sec- 1
1600
65
0600 •
DLVP
229
Pre-Ischemia (all)
o Past- Ischemia (HCO;) IfJI Post -Ischemia ( Irnid)
Fig. 2. Collated hemodynamic data for the preischemic and postischemic performances of each group are represented in graphic form. eFI, Coronary flow index. MV02, Myocardial oxygen consumption. DLVP, Left ventricular developed pressure. SWl, Stroke work index. (+ jdPjdt, Rate of pressure rise. (-)dP /dt, Rate of relaxation. Imid, Imidazole.
Table m. pll, Pco;
POlo
and lactate concentration of cardioplegic infusates and coronary sinus effluents Cardioplegic solution
pH
Bicarbonate Imidazole p Value
7.78 ± 0.02 7.79 ± 0.01
NS
I
Coronary sinus ejJluent
Peal
P02
Lactate
(mmHg)
(mmHg)
(mgfdl]
19.4 ± 0.9 4.4 ± 0.3 <0.001
I
I
163 ± 3 170 ± 5
NS
Analysis of coronary sinus effluent. Effluents from five bicarbonate-buffered experiments showed no evidence of thromboxane B2 activity. All effluents contained less than 1 pg of thromboxane B2 per milliliter. Table III provides the blood gas analysis for the two groups of experimentsas well as coronary sinus lactate concentrations, The pH and P0 2 of the cardioplegic solutions were similarin the two groups. The Pco, values were different as the solutions were prepared to have similar pH (adjustment with O.1N HCl), Neither solution was gassed with carbon dioxide. A significantly higher pH of the effluent was found in the imidazole-buffered group. Carbon dioxide concentration was much higher in the
pH 6.78 ± 0.02 7.02 ± 0.02 <0.001
I
reo, (mmHg) 119 ± 5 38 ± 2 <0.001
P02
I
(mmHg) 35 ± 3 34 ± 3
NS
I
Lactate (mg/dl) 31.2 ± 1.7 38.3 ± 1.6 <0.01
effluents from the bicarbonate-buffered group than in the effluents from the imidazole group. Coronary sinus P0 2 was similar in the two groups, Lactate levels in the effluents from imidazole-buffered hearts were significantly greater than in effluents from bicarbonatebuffered hearts. Water contents were not significantly different; 80.89% ± 0.30% for the imidazole group as opposed to 81.07% ± 0.31% for the bicarbonate group. Discussion This study demonstrates the beneficial effects of using an imidazole buffer in a cardioplegic solution instead of
The Journal of Thoracic and Cardiovascular Surgery
230 Vander Woude et al.
Glucose
I t
ATP
~ADP
Glucose-6-®
t
Fructose-6-® ,r:P"""HC-::O-:CS-=-P"""HO-::-CF=R""-U-'-C-T-""'O-K-1N-A-S-'EI • A~P
t
ADP
Fructose 1,6-di® Dihydroxyocetone [email protected]
t-.. Glyceroldehyde-3~ + (j:i)...
,...NAD
r'G'---LY-C-E-R-A-L-DE-H-Y-D-E-3--....,,0~
~~+'NADH DEHYDROGENASE 1,3 dip -glycerote •
t
ADP
~ATP (2) 3- P glycerote
t t
t
Phosphoenol pyruvo te NAD NADH '--'
ADP
~ATP(2) Loctote ....Pyruvote Fig. 3. Inhibition of glycolytic, anaerobic energy production occurs at the steps catalyzed by phosphofructokinase and glyceraldehyde 3-phosphate dehydrogenase and results from increases in NADH and hydrogen ion. AT?, Adenosine triphosphate. AD?, Adenosine diphosphate. NAD, Nicotinamide-adenine dinucleotide. NADH, NAD, reduced.
a bicarbonate buffer for preservation of myocardial performance after global ischemia. Imidazole-buffered hearts displayed superior return of cardiac function coincident with more normal pH and Pco, in the coronary sinus effluent at the termination of ischemia. During cardioplegic arrest of the myocardium, energy must be supplied by anaerobic glycolysis as oxidative phosphorylation is decreased by lack of oxygen." As concentrations of acidic end products rise, intracellular pH falls, resulting in the inhibition of the glycolytic enzymes phosphofructokinase and glyceraldehyde 3-(P) dehydrogenase (Fig. 3). Both enzymes are controlled by accumulations of metabolites that regulate their activity, most notably nicotinamide-adenine dinucleotide, reduced (NADH) and hydrogen ion. As adenosine triphosphate hydrolysis produces hydrogen ions, buffering by the bicarbonate system results in the generation of one carbon dioxide molecule for each hydrogen ion buffered. The increase in intracellular hydrogen ion also alters calcium ion availability at the level of the sarcoplasmic reticulum and calcium entry into the cell. Thus,
acidosis has been shown to alter both intermediary metabolism and calcium metabolism." In the ischemic myocardium, reducing equivalents (NADH) from glycolysis accumulate in the cytoplasm and must be oxidized for the glycolytic pathway to continue." The NADH-NAD ratio, as noted by Rovetto, Whitmer, and Neely" can increase up to 16-fold in severely ischemic cardiac muscle and thus contributes to reduced glycolytic rates. The fall of the intracellular pH is a consequence of an accumulation of acidic end products of anaerobic and aerobic metabolism and strongly interferes with the buffering capacity of the bicarbonate-carbon dioxide system. This is believed to be secondary to the high levels of intracellular carbon dioxide." Bicarbonate is unable to effectively buffer and oxidize NADH as it exerts its major buffering action in the extracellular space." An additional inadequacy of bicarbonate in a "closed" system (global ischemia) is that it is dependent upon a constant Pco, for its buffering capacity. 10 Ideally, one would wish to deliver a buffer that is not dependent upon narrow ranges of Pco, for buffering
Volume 90 Number 2
Imidazole-buffered cardioplegic solution 2 3 1
August, 1985
Imidazole (Glyoxaline)
pK 7.5
pK =6.7 at 37°C = 7.0 at 25°C
7.3
7.1
6.9
H
HC-N
II
HC
'N
'CH ~
H+
Fig. 4. Imidazole buffering of protons occurs at the siteof its
electronegative nitrogen.
capacity and one that is uncharged to a significant degree, to allow passage into the intracellular cytosolic space. Tris (hydroxymethyl) aminomethane (TRIS, pKa = 7.8 at 37 0 C) has been utilized in ischemic preparations. It is 30% nonionized at pH 7.4. Effron and associates" concluded TRIS was beneficial in the ischemiccat heart by an intracellular buffering mechanism. TRIS, however, affects tissue in a manner independent of its ability to accept hydrogen ions and has been thought to interfere with the binding of calcium at the cell membrane." Liedtke and co-workers" were able to improve postischemic function and show acceleration of glycolysis in the ischemic swine heart by use of TRIS and pyruvate. The use of histidine in concentrations of 150 to 200 mmol/L has been shown by Bretschneider and colleagues" to provide extracellular buffer capacity to stimulate anaerobic energy production. The imidazole-buffered cardioplegic solution used in these experiments departs from the Bretschneider solution in that the present solution is an extracellular type, high-sodium solution with buffering imparted. by imidazole, 25 mmol/L. The buffered Bretschneider solution contains sodium, 15 mmolfL, with the majority of the osmotic space occupied by histidine, 180 mmolfL. Imidazole (pKa = 7.0 at 25 0 C) (Fig. 4) has been
......
••••.••• Phosphate
6.7
6.3
.,
...-.
.... .... ......
.
.,.... Bicarbonate
6.1
o
.............. .
10
...... . -.. ......
20
.......... ....... 30
40°C
Fig. 5. The dissociation curve of imidazole parallels that of water as temperatures trend toward hypothermia. Other COmmon buffers depart from this physiological ideal.
used as a buffering agent for many years in laboratory medicine and microbiology." Imidazole is highly soluble in water, readily available, and stable in dry form. It is not a foreign compound, but is prevalent in nature as the proton-accepting component of histidine. The imidazole moiety of histidine exists in sufficient concentration in blood (30 mmolfL in human plasma) to dominate other buffer systems." Eighty percent of the buffering capacity of blood is the result of its presence. The importance of the imidazole group to physiological systems is that it is the only buffering agent that provides alkalinity of blood relative to the neutral pH of water as temperature decreases (Fig. 5) Imidazole, like its parent histidine, may exert biological effects over and above its buffering action. In addition to acting as a proton donor/acceptor, it has been thought to influence some enzyme-substrate binding reactions and to trigger the release of calcium
2 3 2 Vander Woude et al.
from the sarcoplasmic reticulum (special conditions, pH = 9.0 at 20° C).24 Comparison of imidazole to two commonly used buffers in cardioplegic solutions, bicarbonate and TRIS, shows that there are theoretical advantages to imidazole. Bicarbonate, if infused at a pH of 7.8 at 3r C in a cardioplegic solution, would exist predominantly in the HC0 3 state by the Henderson-Hasselbach relationship. For it to act effectively as a buffer, the Pco, must remain stable. Pco, rises in the closed system of cardioplegic arrest. If TRIS (pKa 7.8 at 37° C) is infused at pH 7.8, the active nitrogen is 50% protonated and 50% nonprotonated. Thus, a significant amount is available to cross the membrane to the intracellular space for buffering during cellular ischemia." Imidazole (pKa 6.7 at 37° C), if infused at pH 7.8, exists primarily in the nonprotonated form to satisfy the HendersonHasselbach relationship. Theoretically, it would be better suited for membrane passage and intracellular buffering. All indices of performance in the postischemic state of the imidazole-buffered group were significantly better than those in the bicarbonate-buffered group except for myocardial oxygen consumption. Slightly increased postischemic +LVdP jdts, reflecting better contractility, and significantly greater - LVdP j dts, reflecting improved diastolic relaxation, were seen. Concomitantly, coronary flows were improved after ischemia. Though graphic portrayal (Fig. 2) of the raw data for +LVdPj dt would indicate no statistical difference, the analysis of variance of preischemic minus postischemic differences did indicate significance. As the effect of the buffer is dependent on both preischemic and postischemic values, the lack of a postischemic difference between groups in no way precludes the effects as described. Most importantly, recovery of stroke work indices was significantly better in the hearts given imidazole-buffered cardioplegic solution. In the absence of information about ventricular volumes, it is not known whether the enhanced performances resulted from greater fiber shortening from the same end-diastolic volume (increased contractility), from increased end-diastolic volumes due to improved compliance, or both." As myocardial and coronary venous Pcos are similar at the moment of release of coronary occlusion, as shown by the work of Case, Felix, and Wachter," we chose to examine coronary sinus effluents at the release of the cross-clamp. Volumes of effluent and times of collection were similar between the two groups, so that sampling time variation was not a cause for the differences. Analysis of coronary sinus effluents supports our
The Journal of Thoracic and Cardiovascular Surgery
conclusions about the mechanism by which imidazole provides improved buffering capacity and therefore improved myocardial protection compared to bicarbonate. The Pco, of the imidazole-buffered cardioplegic solution was initially lower than that of the bicarbonatebuffered solution owing to the affinity of imidazole for the proton of the carbonic acid intermediate. As noted in Table I, three times the volume of O.lN HCl was required for pH adjustment in the imidazole solution to bring it equal to the bicarbonate solution. The absolute value of the coronary sinus Pco, was markedly higher for the bicarbonate-buffered experiments, concordant with the lower pH. The possibility that imidazole inhibited formation of thromboxane and led to improved coronary flows and postischemic recoveries in the imidazole group was considered unlikely, because thromboxane was not detected in the bicarbonate-buffered group by a highly sensitive technique. Thromboxane synthesis has not been demonstrated in the coronary artery"; rather the majority of prostaglandin released from the isolated heart is in the form of prostacyclin, a vascular smooth muscle relaxant. 28 The finding of greater concentrations of lactate in effluents of imidazole-buffered hearts would, at first, appear contrary to the concept of an improved acid-base status of the intracellular space by imidazole. Under conditions of ischemia, anaerobic glycolysis utilizes myocardial glucose and glycogen and results in the accumulation of lactate. The production of lactate from glucose results in no excess hydrogen ions, but the hydrolysis of adenosine triphosphate does generate hydrogen ions and results in intracellular acidosis.29 With ischemia, control of the glycolytic rate shifts to later reactions in the pathway, notably that catalyzed by glylceraldehyde 3-phosphate dehydrogenase. This enzyme is regulated by product inhibition, largely NADH accumulation.v-" If a lower or more favorably NADHjNAD+ ratio can be achieved by augmented buffering capacity, glycolysis may continue on to its anaerobic product, lactate. This would be in agreement with the work of Markov and associates," who used fructose 1,6, diphosphate as a substrate-enhancing agent in a dog model of myocardial ischemia. Markov's group observed improved mechanical recovery, regression of electrocardiographic ischemic changes, and significantly elevated levels of myocardial adenosine triphosphate and creatine phosphate in ischemic hearts given fructose diphosphate. Importantly, they found the tissue lactate content in ischemic muscle
Volume 90
Imidazole-buffered cardioplegic solution 2 3 3
Number 2 August, 1985
in the treated group was twice that found in the control group, indicating improved glycolytic flux in the treated group. The experiments reported here suggest that improved buffering capacity of the extracellular and intracellular myocardial spaces by a buffer not dependent upon removal of carbon dioxide may enhance postischemic functional recovery. It is not clear, however, whether this salutory effect is the result of enhancement of glycolytic energy production and maintenance of cellular volume regulation or decreased activation of cellular lysosomes and proteolytic enzymes." Additionally, imidazole may have alternative physiological actions independent of thromboxane synthetase inhibition and proton buffering. From information available in this set of experiments, it is believed that imidazoleservesas an effective buffering agent during normothermic global myocardial ischemia and improves postischemic functional recovery. We gratefully acknowledge the cooperation of consultant statistician Kenneth Schechtman, Ph.D., Biomedical Computer Laboratory, Washington University School of Medicine. We also express our appreciation to Ms. Patricia Black for technical assistance in performing these studies. REFERENCES
2
3
4
5
6
7
8
9
Buckberg GD: A proposed "solution" to the cardioplegic controversy. J THoRAc CARDIOVASC SURG 77:803-815, 1979 Kirklin JW, Conti VR, Blackstone EH: Prevention of myocardial damage during cardiac operations. N Engl J Med 301:135-141,1979 Nayler WG: Preservation of the myocardium. Some biological considerations, Towards Safer Cardiac Surgery, DB Longmore, ed., Boston, 1981, G. K. Hall Medical Medical Publishers, pp 627-651. Kubler W, Spieckermann PG: Regulation of glycolysis in the ischemic and anoxic myocardium. J Mol Cell Cardiol 1:351-377,1970 Rovetto MJ, Lamberton WF, Neely JR: Mechanisms of glycolytic inhibition in ischemic rat hearts. Circ Res 37:742-751,1975 Bretschneider HJ, Hubner G, Knoll D, Lohr B, Nordbeck H, Spieckermann PG: Myocardial resistance and tolerance to ischemia. Physiological and biochemical basis. J Cardiovasc Surg (Torino) 16:241-260, 1975 Jones JW, Gionis TA, Nichols RL, Markov A, Webb WR: Myocardial preservation with fructose 1,6 diphosphate. Energy without oxygen. Surg Forum 31:307-309, 1980 Henderson LJ: Concerning the relationship between the strength of acids and their capacity to preserve neutrality. Am J Physiol 21:173-179,1908 Good NE, Winget GD, Winter W, Connolly TN, Izawa S,
10
11
12 13
14 15
16 17
18
19
20
21
22 23
24
25
26
Singh RMM: Hydrogen ion buffers for biological research. Biochemistry 5:467-477, 1966 Bailey LE, Ong SD: Krebs-Henseleit solution as a physiological buffer in perfused and superfused preparations. J Pharmacol Methods 1:171-175, 1978 Neely JR, Liebermaster H, Battersby EJ, Morgan HE: Effect of pressure development on oxygen consumption by isolated rat heart. Am J Physiol 212:804-814, 1967 LangendorffO: Untersuchungen am uberlebenden Saugethierherzen. Pfluegers Arch 61:291, 1895 Needleman P, Bryan B, Wyche A, Bronson SD, Eakins K, Ferrendelli JA, Minkes M: Thromboxane synthetase inhibitors as pharmacological tools. Differential biochemical and biological effects on platelet suspensions. Prostaglandins 14:897-907, 1977 Levitsky S, Feinberg H: Biochemical changes of ischemia. Ann Thorac Surg 20:21-29, 1975 Williamson JR, Schaffer SW, Ford C, Safer B: Contribution of tissue acidosis to ischemic injury in the perfused rat heart. Circulation 53:Suppl 1:3, 1976 Sobel BE: Salient biochemical features in ischemic myocardium. Circ Res 34,35:Suppl 3:173, 1974 Rovetto MJ, Whitmer JT, Neely JR: Comparison of the effects of anoxia and whole heart ischemia on carbohydrate utilization in isolated working rat hearts. Circ Res 32:699-711, 1973 Beierholm EA, Grantham RN, O'Keefe DD, Laver MB, Daggett WM: Effects of acid-base changes, hypoxia, and catecholamines on ventricular performance. Am J Physiol 228:1555-1561, 1975 Effron MB, Guarnieri T, Frederiksen JW, Greene HL, Weisfeldt ML: Effect of tris(hydroxymethyl)aminomethane on ischemic myocardium. Am J PhysioI235:HI67H174,1978 Durrett LR, Adams HR: Inotropic responsiveness of atrial myocardium bathed in Tris- or bicarbonate-buffered solutions. Am J Physiol 237:H318-H325, 1979 Liedtke AJ, Nellis SH, Neely JR, Hughes HC: Effects of treatment with pyruvate and tromethamine in experimental myocardial ischemia. Circ Res 39:378-387, 1976 Diem K, Lentner C: Buffer Solutions, Scientific Tables, Ardsley, N. Y., 1970, Geigy Pharmaceuticals, p 281 Rahn H, Reeves RB, Howell BJ: Hydrogen ion regulation, temperature, and evolution. Am Rev Respir Dis 112:165172,1975 Chapman RA, Miller DJ: Structure-activity relations for caffeine. A comparative study of the inotropic effects of the methylxanthines, imidazoles, and related compounds on the frog's heart. J Physiol (Lond) 242:615-634, 1974 Khuri SF, F1aherty JT, O'Riordan JB, Pitt B, Brawley RK, Donahoo JS, Gott VL: Changes in intramyocardial ST segment voltage and gas tensions with regional myocardial ischemia in the dog. Circ Res 37:455-463, 1975 Case RB, Felix A, Wachter M: Measurement of myocardial pC0 2 with a microelectrode. Its relation to coronary sinus pCO,. Am J Physiol 236:H29-H34, 1979
234 Vander Woude et al.
27 Needleman P, Kaley G: Cardiac and coronary prostaglandin synthesis and function. N Engl J Med 298: 1122-1128, 1978 28 De Deckere EAM, Nugteren DH, Ten Hoor F: Prostacydin is the major prostaglandin released from the isolated perfused rabbit and rat heart. Nature 268:160-163, 1977 29 F1aherty JT, Schaff HV, Goldman RA, Gott VL: Metabolic and functional effects of progressive degrees of hypothermia during global ischemia. Am J Physiol 236:H839-H845, 1979 30 Katz AM: Physiology of the Heart, New York, 1977, Raven Press, pp 35-50
The Journal of Thoracic and Cardiovascular Surgery
31 Markov AK, Oglethorpe NC, Blake TM, Lehan PH, Hellems HK: Hemodynamic, electrocardiographic, and metabolic effects of fructose diphosphate on acute myocardial ischemia. Am Heart J 100:639-646, 1980 32 Swan H: The hydroxyl-hydrogen ion concentration ratio during hypothermia. Surg Gynecol Obstet 155:897-912, 1982 33 Bersohn MM, Scheuer J: Effects of physical training on end-diastolic volume and myocardial performance of isolated rat hearts. Circ Res 40:510-516, 1977
Bound volumes available to subscribers Bound volumes of THE JOURNAL OFTHORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1985 issues from the Publisher, at a cost of $61.45 ($77.10 international) for Vol. 89 (January-June) and Vol. 90 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 30 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.
THoRAc CARDIOVASC SURG
90:225-234, 1985
Imidazole-buffered cardioplegic solution Improved myocardial preservation during global ischemia Progressive acidosis is a constant rmding in global myocardial ischemia and is associated with reduced myocardial contractility after ischemia. The hypothesis tested in these experiments was that imidazole (pKa = 6.7 at 37° C), a commonly used buffer in physiology and microbiology, would provide superior buffering capacity when used in lieu of bicarbonate (pKa = 6.1 at 37° C) in a cardioplegic solution. Twenty-eight isolated, working rabbit hearts were perfused, and preischemic and postischemic determinants of performance were measured. The 30 minute interval of normothermic global ischemia was altered by the injection at 0 and 15 minutes of 2 m1/gm wet weight of a buffered cardioplegic solution. Control bearts received a bicarbonate-buffered cardioplegic solution and experimental hearts receiveda solution buffered with imidazole. In the imidazole-buffered group, there was a superior recovery of coronary flow, developed left ventricular pressure, peak rate of rise of left ventricular pressure, peak rate of relaxation, andstroke work indices (p < 0.05). Recovery of mechanical parameters was coincident with an improved acid-base status of the coronary sinus effluent at the end of ischemia. Coronary sinus effluents in the imidazole group had significantly higher pH values and lower partial pressures of carbon dioxide than coronary sinus effluents in the bicarbonate-buffered group (p < 0.001). The data suggest that improved buffering of the extracellular and possibly intracellular space during global ischemia with a nonbicarbonate buffer is beneficial and provides improved postischemic myocardial recovery.
John C. Vander Woude, M.D., Ignacio Y. Christlieb, M.D., Gregorio A. Sicard, M.D., and Richard E. Clark, M.D., St. Louis, Mo.
In the present era of open cardiac operations most surgeons use some form of chemical cardioplegia. A variety of cardioplegic solutions are in clinical use, employing different concentrations of constituents such as sodium, potassium, and buffering agents. Many of these solutions have never been tested under experimental conditions. I The differences in composition of current solutions are not surprising given the large gaps in our understanding of the mechanisms of cell death in global ischemia.' The heart requires a continuous supply of oxygen to generate energy for contraction and for maintenance of its cellular structure and function.' During the period of From the Divisions of Cardiothoracic and General Surgery, Department of Surgery, Washington University School of Medicine, St. Louis, Mo. Received for publication Aug. 30, 1984. Accepted for publication Oct. 17, 1984. Address for reprints: John C. Vander Woude, M.D., Department of Surgery, 910 Wohl Hospital, 4960 Audubon Ave., St. Louis, Mo. 63110.
aortic cross-clamping and cardioplegic arrest in cardiac operations, metabolic perturbations occur. Myocardial oxygen reserves are depleted within 8 seconds after the initiation of ischemia, precluding continued oxidative phosphorylation in the mitochondria.' Concentrations of lactic acid rise markedly and intracellular pH falls. Glycolysis, an inefficient, anaerobic pathway of adenosine triphosphate production, is rate limited because phosphofructokinase and glyceraldehyde 3-(P) dehydrogenase activities are inhibited by the resultant intracellular acidosis." 5 Bretschneider and associates" have shown that the small energy output of anaerobiosis can maintain prolonged myocardial viability if cellular glycolytic metabolism can continue unimpeded. Investigators have shown salutory effects on ischemic myocardium by interventions within the glycolytic pathway, either by providing substrate such as fructose diphosphate or by enhancing buffering capacity with HEPES or tricine buffers.> 7 The hypothesis tested in these experiments was that imidazole (pKa = 6.7 at 37° C), a buffer not dependent on bicarbonate--carbon dioxide flux, would provide
225
The Journal of Thoracic and Cardiovascular Surgery
226 Vander Woude et al.
100
AORTIC~ o 25
LVEDP
o
EKG
100
LV
o
dP/dt
WN/0M
+2500
-2500
Fig. 1. Representative hemodynamic tracings obtained in one experiment demonstrating aortic pressure, left ventricular end-diastolic pressure (LVEDPj. electrocardiographic tracing (EKG). left ventricular pressure (LV), and ± rate of pressure development (± dPjdt). superior buffering capacity during global myocardial ischemia when used in place of bicarbonate (pKa = 6.1 at 37 0 C) in a cardioplegic solution. The rationale for the substitution was twofold. Henderson" pointed out in 1908 that a weak acid exerts maximal buffering when its dissociation constant (pKa) is equal to or near the pH of neutrality. Second, the buffering capacities of TRIS, imidazole, and HEPES have been shown to be independent of the partial pressure of carbon dioxide (Pco.), Since the pKa of the carbonic acid-bicarbonate-carbon dioxide system is 6.10, in the absence of constant replenishment of bicarbonate or the removal of carbon dioxide, bicarbonate solutions do not function as effective buffers." IO
Materials and methods Perfusion technique. These studies were performed with 28 isolated working hearts obtained from 2.5 ± OJ kg New Zealand white rabbits. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" as formulated by the National Society for Medical Research. The animals were anesthetized with intravenous pentobarbital, 30 mg/kg, 5 minutes after systemic heparinization (2,000 units per animal). Hearts were extracted rapidly through a median sternotomy, placed in iced saline, and transferred within 1 minute to an isolated heart perfusion apparatus similar to that described by Neely and colleagues.I I Retrograde perfusion was established at 38 0 C at a pressure of 48 to 49 mm Hg by the method of Langendorff." The hearts were perfused with a modified Krebs-Henseleit solution augmented with 1% bovine serum albumin and insulin, 400 mU /L; perfusate was filtered with a 0.45 ~m Millipore filter prior to use. The concentrations of the perfusate constituents (mm/ L) were as follows: NaCl 118, KCl 4J, cso, . 2H 20 2.5, MgS04 • 7H 20 1.25, KH 2P04 1.2, NaHC03 25.0, and dextrose 15.0. The perfusate was gassed with a mixture of 95% oxygen and 5% carbon dioxide through scintered glass of 15 ~m porosity to reduce foaming caused by albumin. This produced a partial pressure of oxygen (Po 2) of 450 to 525 mm Hg, a Pco, of 32 to 42 mm Hg, and a pH of 7.40 to 7.48. The entire apparatus was maintained at 38 0 C by water jackets surrounding all reservoirs and tubing. During the period of retrograde washout, a cannula was placed in the pulmonary artery and all pulmonary veins and hilar structures were ligated. A ribbed, rigid cannula, 6 mm inner diameter, was inserted into the left atrial appendage to allow left atrial inflow for the working heart perfusions. Within the inflow cannula were two, soft 0.031 inch polyethylene catheters. One terminated at the end of the inflow cannula and was used to measure left atrial pressure. The other tube extended beyond the tip of the inflow cannula and was placed through the mitral valve into the midventricle for measurement of left ventricular pressure. This latter catheter allowed high-fidelity recordings of peak rate of rise of left ventricular pressure (+LVdP/dt), peak rate of relaxation (-LVdP /dt), and left ventricular enddiastolic pressure (LVEDP) without an apex puncture or use of an isovolumic balloon (Fig. 1). The sinus node was crushed and the hearts were paced at 214 beats/min by twin electrodes sutured on the right atrium with a Grass SD-9 stimulator. The stimulus voltage was 3.0 V with a duration of 10 msec. Cardiac and solution temperatures were measured with thermistors (Yellow
Volume 90 Number 2 August. 1985
Springs Instrument Co., Yellow Springs, Ohio). Retrograde perfusion and cannulation required 20 minutes after which the heart was converted to the working mode. Perfusate from the left atrial reservoir flowed into the left atrium, and left ventricular contraction pumped the perfusate through the aortic circuit. A Starling apparatus fashioned from 15 em of V2 inch Penrose drain tubing and a glass chamber served as a compliance-resistance apparatus and was placed 20 em above the heart. The size and shape of the aortic pressure curve measured just distal to the aortic valve were altered by adjusting the air pressures in the Starling chamber. Once adjusted, mean aortic pressures were maintained at 49.0 ± 0.5 mm Hg over the range of ventricular loadingconditions (1 to 4 mm Hg). The working circuit had a recirculating volume of 1,000 rnl with aortic flow returning to the reservoir by gravity and coronary flow by means of a rotary pump. Left atrial inflow was filtered by an inline 13 tLm Pioneer filter. During a 20 minute work period, with rate and mean aortic pressure constant, ventricular function was assessed by increasing the LVEDP over the range of 1 to 4 mm Hg. A 3 to 5 minute period was required to achieve stability of the preparation at each LVEDP. The following measurements were recorded for each heart at each time point: left ventricular developed pressure; +LVdP/ dt; - LVdP/ dt; aortic and coronary flow; and reservoir and coronary sinus pH, Pco., and Paz. Phasic and mean left ventricular and aortic pressures were measured with Sanborn 267 transducers and the LVEDP was measured with a Hewlett-Packard 1280 transducer. Aortic flow was measured with a Biotronex 3.5 mm cannulatingflow probe. Coronary flow was measured volumetrically during the recording interval. The left ventricular pressure signal was differentiated on a Biotronex 620622 differentiator-integrating unit. The pH, Paz, and Pco, of the perfusate and coronary sinus effluents were determined with an IL-313 analyzer. All pressure tracings, differential signals, and bipolar electrocardiogramswere recorded on a Sanborn 8-channel strip-chart recorder at a paper speed of 100 mm/sec, Ischemic heart model. After initial work parameters had been measured, the hearts were subjected to a 30 minute period of normothermic global ischemia. Global ischemia was altered by an injection of 10 rnl (2 ml/gm wet weight per injection) of a buffered cardioplegic solution at initiation of and 15 minutes after onset of ischemia. Control hearts (n = 15) received a bicarbonate-buffered cardioplegic solution and experimental hearts (n = 13) received an imidazole-buffered cardioplegic solution. Composition of the two cardioplegic solutions is shown in Table I. The cardioplegic solutions
Imidazole-buffered cardioplegic solution
227
Table I. Composition of the cardioplegic solutions Control
Experimental
Na+ (mmol/L) K+ (rnrnol/L) Cl (rnrnol/L) HCO , (mmol/L) Imidazole (mmol/L) Dextrose (mmol/L)
143
135
Mg?" (mmol/L)
1.3 7 10
1.3
366
366
Constituent
Mannitol (rnrnol/L) O.IN HCI (rnl/L) Osmolality (calculated)
27
27
148 25
163
15
25 15 30
were designed to be as similar to each other as possible in terms of ionic content, pH, partial pressures, and osmolality. The buffers used were bicarbonate for the control group and an equimolar concentration of imidazole for the experimental group. At the onset of ischemia, the left atrial and aortic lines were clamped and 10 rnl of cardioplegic solution was infused into an aortic root cannula through a 0.45 tLm filter at 45 to 50 mm Hg pressure. All hearts stopped mechanical and electrical activity within 10 seconds. At the end of the ischemic period, hearts were reperfused at 48 to 49 mm Hg retrograde pressure while the left ventricular cavity was vented via the left ventricular pressure measurement catheter. The initial 10 rnl aliquot of reperfusate was aspirated from the pulmonary artery cannula, immediately sealed from ambient air, and analyzed for acid-base status. After the retrograde reperfusion period, hearts were converted to the working mode with fresh perfusate and postischemic indices of performance were obtained. The mean aortic pressure was maintained at 49.0 ± 0.5 mm Hg, the heart rate at 214 beats/min, and the LVEDP was varied from 1 to 4 mm Hg to obtain several data points that permitted construction of ventricular function curves for comparison of preischemic and postischemic performance. Analysis of coronary sinus effluent. Myocardial oxygen consumption was determined during the left ventricular loading tests. Recognizing the limited oxygen availability of assanguineous perfusate, we assessed ventricular performance in the limited range of 1 to 4 mm Hg LVEDP to prevent oxygen availability from being a limiting factor. Accordingly, Po.s of coronary sinus effluent were uniformly 95 to 115 mm Hg at higher LVEDPs. The sealed sample of the initial 10 rnl of reperfusate was measured for pH, Pco., and Paz and served as an indicator of the acid-base status of the extracellular space at the onset of reperfusion. As imidazole has pharmacologic activity by mechanisms other than buffering, including blockade of
228
The Journal of Thoracic and Cardiovascular Surgery
Vander Woude et al.
Table II. Hemodynamic data comparison of bicarbonate-buffered versus imidazole-buffered hearts Preischemia
HCO J (n = 60)
Parameter CFI (rnl/rnin/grn)
MVO, (ml/rnin/gm) DLVP (mm Hg) +LVdP/dt (mm Hg . sec') -LVdP/dt (mm Hg . sec-') SWI (gm-rn/gm)
63.8 0.677 88.9 +1,767 -1,534 1.10
I
Imidazole (n = 52)
Postischemia
HCO J (n = 60)
I
Pre-post difference
Imidazole (n= 52)
61.4 56.3 ± 1.0 ± 1.3 63.2 ± 1.4 ± 0.030 0.654 0.663 ± 0.032 0.610 ± 0.018 86.7 ± 1.1 82.6 79.1 ± 0.9 ± 0.9 +1,576 ± 25 + 1,589 ± 28 +1,672 ± 35 ± 26 -1,452 ± 30 -1,467 ± 18 -1,544 0.65 ± 0.03 1.10 ± 0.05 0.93 ± 0.05
HCO J (n = 60)
I
Imidazole (n = 52)
±. 1.1 7.4 ± 0.9 2.3 ± 0.8 ± 0.026 0.061 ± 0.021 0.016 ± 0.017 ± 1.0 10.1 ± 0.7 4.0 ± 0.7 ± 26 +200 ± 19 +81 ± 18 ± 25 -88 ± 31 +77 ± 30 ± 0.05 0.46 ± 0.03 0.19 ± 0.02
p Value pre-post 0.05
NS 0.01 0.05 0.05 0.001
Legend: Mean arterial pressure = 49 ± 0.5 mm Hg. Heart rate = 214 beats/min. Mean left ventricular end-diastolic pressure = 2.5 mm Hg. CFI, Coronary flow index. MVO" Myocardial oxygen consumption. DLVP, Left ventricular developed pressure. +L VdP /dt, Peak rate of rise of left ventricular pressure. -LVdP/dt, Peak rate of relaxation of left ventricular pressure. SWI, Stroke work index.
thromboxane synthetase," it was important to confirm the absence of thromboxane in the asanguineous effluents of control experiments. Reperfusion effluents in five control experiments were analyzed by radioimmunoassay for the concentration of thromboxane B2, the stable metabolite of thromboxane A 2• Reperfusate lactate determinations were performed in the experiments from each group as we wished to correlate acid-base conditions with concentrations of lactate, the end product of anaerobic, glycolytic energy production. Analysis was performed by standard spectrophotometric methods (Sigma Chemical Co., St. Louis, Mo.). Myocardial water content and hemodynamic calculations. At the conclusion of each experiment, the heart was blotted dry, debrided of all atrial and hilar tissue, and dried at 80° C to a constant dry weight (Mettler Analytical Scale, Mettler Instrument Corp., Hightstown, N. J.). Percent water content was calculated as: 100 X (wet weight - dry weightj/wet weight. Stroke work and myocardial oxygen consumption (MV02) were calculated from the following indices: Cardiac output Stroke work = - - - - = - Rate X (Mean aortic pressure - LVEDP) X 0.0136
MVO,
= Coronary
flow X Perfusate Po, - Coronary sinus Po, X 0.0031 *
Stroke work, coronary flow, and oxygen consumptions for each heart were divided by the dry weight to provide an index of comparison between groups. Data analysis. All indices of performance were paired for individual experiments regarding conditions of preload, afterload, and rate. Performance data for the *Oxygen solubility constant of perfusate at 37° C.
preischemic and postischemic states were then collated. These data were analyzed by analysis of variance, looking specifically at the preischemic minus postischemic differences. Group means for lactate concentrations and water contents were compared by the Student's t test for unpaired data. Values are expressed as means ± standard error of the mean. Differences were considered significant if the p value was less than 0.05. Results Hemodynamics. Table II provides data for preischemic and postischemic performances of hearts subjected to bicarbonate and imidazole buffering as well as the pre-post differences. The data are as stated for coronary flow indices, oxygen consumption, developed pressure (left ventricular pressure - LVEDP), +LVdP/ dt, -LVdP / dt, and stroke work indices. Analysis of preischemic performance indices showed no significant differences between the bicarbonate- and imidazole-buffered experiments. Examination of the pre-post differences showed improved recovery of the imidazole group in all parameters except for oxygen consumption (p = 0.1180). Coronary flow index after ischemia was decreased less from the preischemic level in the imidazole group than in the bicarbonate group. Developed pressure fell less after the ischemic period in the imidazole group. Contractility, as reflected by + LVdP / dt, recovered to a greater degree in the imidazole group. Diastolic relaxation, as reflected by the - LVdP / dt, was better maintained in the imidazole group and was actually greater (more negative) than its preischemic control value. Most significantly, the stroke work index of the imidazole group after the ischemic period decreased by only 16% compared to a decrease of 41% in the bicarbonate group (p < 0.001). Fig. 2 provides these results in bar graph form.
Volume 90
Imidazole-buffered cardioplegic solution
Number 2 August, 1985
70
CFI ml/min/gm
0.700
50
MV02 ml/min/gm
90
SWI gm'M/gm
mmHg
120
(+)dP/dt rnmrlq-sec -I 1800
(-)dP/dt mmHg'sec- 1
1600
65
0600 •
DLVP
229
Pre-Ischemia (all)
o Past- Ischemia (HCO;) IfJI Post -Ischemia ( Irnid)
Fig. 2. Collated hemodynamic data for the preischemic and postischemic performances of each group are represented in graphic form. eFI, Coronary flow index. MV02, Myocardial oxygen consumption. DLVP, Left ventricular developed pressure. SWl, Stroke work index. (+ jdPjdt, Rate of pressure rise. (-)dP /dt, Rate of relaxation. Imid, Imidazole.
Table m. pll, Pco;
POlo
and lactate concentration of cardioplegic infusates and coronary sinus effluents Cardioplegic solution
pH
Bicarbonate Imidazole p Value
7.78 ± 0.02 7.79 ± 0.01
NS
I
Coronary sinus ejJluent
Peal
P02
Lactate
(mmHg)
(mmHg)
(mgfdl]
19.4 ± 0.9 4.4 ± 0.3 <0.001
I
I
163 ± 3 170 ± 5
NS
Analysis of coronary sinus effluent. Effluents from five bicarbonate-buffered experiments showed no evidence of thromboxane B2 activity. All effluents contained less than 1 pg of thromboxane B2 per milliliter. Table III provides the blood gas analysis for the two groups of experimentsas well as coronary sinus lactate concentrations, The pH and P0 2 of the cardioplegic solutions were similarin the two groups. The Pco, values were different as the solutions were prepared to have similar pH (adjustment with O.1N HCl), Neither solution was gassed with carbon dioxide. A significantly higher pH of the effluent was found in the imidazole-buffered group. Carbon dioxide concentration was much higher in the
pH 6.78 ± 0.02 7.02 ± 0.02 <0.001
I
reo, (mmHg) 119 ± 5 38 ± 2 <0.001
P02
I
(mmHg) 35 ± 3 34 ± 3
NS
I
Lactate (mg/dl) 31.2 ± 1.7 38.3 ± 1.6 <0.01
effluents from the bicarbonate-buffered group than in the effluents from the imidazole group. Coronary sinus P0 2 was similar in the two groups, Lactate levels in the effluents from imidazole-buffered hearts were significantly greater than in effluents from bicarbonatebuffered hearts. Water contents were not significantly different; 80.89% ± 0.30% for the imidazole group as opposed to 81.07% ± 0.31% for the bicarbonate group. Discussion This study demonstrates the beneficial effects of using an imidazole buffer in a cardioplegic solution instead of
The Journal of Thoracic and Cardiovascular Surgery
230 Vander Woude et al.
Glucose
I t
ATP
~ADP
Glucose-6-®
t
Fructose-6-® ,r:P"""HC-::O-:CS-=-P"""HO-::-CF=R""-U-'-C-T-""'O-K-1N-A-S-'EI • A~P
t
ADP
Fructose 1,6-di® Dihydroxyocetone [email protected]
t-.. Glyceroldehyde-3~ + (j:i)...
,...NAD
r'G'---LY-C-E-R-A-L-DE-H-Y-D-E-3--....,,0~
~~+'NADH DEHYDROGENASE 1,3 dip -glycerote •
t
ADP
~ATP (2) 3- P glycerote
t t
t
Phosphoenol pyruvo te NAD NADH '--'
ADP
~ATP(2) Loctote ....Pyruvote Fig. 3. Inhibition of glycolytic, anaerobic energy production occurs at the steps catalyzed by phosphofructokinase and glyceraldehyde 3-phosphate dehydrogenase and results from increases in NADH and hydrogen ion. AT?, Adenosine triphosphate. AD?, Adenosine diphosphate. NAD, Nicotinamide-adenine dinucleotide. NADH, NAD, reduced.
a bicarbonate buffer for preservation of myocardial performance after global ischemia. Imidazole-buffered hearts displayed superior return of cardiac function coincident with more normal pH and Pco, in the coronary sinus effluent at the termination of ischemia. During cardioplegic arrest of the myocardium, energy must be supplied by anaerobic glycolysis as oxidative phosphorylation is decreased by lack of oxygen." As concentrations of acidic end products rise, intracellular pH falls, resulting in the inhibition of the glycolytic enzymes phosphofructokinase and glyceraldehyde 3-(P) dehydrogenase (Fig. 3). Both enzymes are controlled by accumulations of metabolites that regulate their activity, most notably nicotinamide-adenine dinucleotide, reduced (NADH) and hydrogen ion. As adenosine triphosphate hydrolysis produces hydrogen ions, buffering by the bicarbonate system results in the generation of one carbon dioxide molecule for each hydrogen ion buffered. The increase in intracellular hydrogen ion also alters calcium ion availability at the level of the sarcoplasmic reticulum and calcium entry into the cell. Thus,
acidosis has been shown to alter both intermediary metabolism and calcium metabolism." In the ischemic myocardium, reducing equivalents (NADH) from glycolysis accumulate in the cytoplasm and must be oxidized for the glycolytic pathway to continue." The NADH-NAD ratio, as noted by Rovetto, Whitmer, and Neely" can increase up to 16-fold in severely ischemic cardiac muscle and thus contributes to reduced glycolytic rates. The fall of the intracellular pH is a consequence of an accumulation of acidic end products of anaerobic and aerobic metabolism and strongly interferes with the buffering capacity of the bicarbonate-carbon dioxide system. This is believed to be secondary to the high levels of intracellular carbon dioxide." Bicarbonate is unable to effectively buffer and oxidize NADH as it exerts its major buffering action in the extracellular space." An additional inadequacy of bicarbonate in a "closed" system (global ischemia) is that it is dependent upon a constant Pco, for its buffering capacity. 10 Ideally, one would wish to deliver a buffer that is not dependent upon narrow ranges of Pco, for buffering
Volume 90 Number 2
Imidazole-buffered cardioplegic solution 2 3 1
August, 1985
Imidazole (Glyoxaline)
pK 7.5
pK =6.7 at 37°C = 7.0 at 25°C
7.3
7.1
6.9
H
HC-N
II
HC
'N
'CH ~
H+
Fig. 4. Imidazole buffering of protons occurs at the siteof its
electronegative nitrogen.
capacity and one that is uncharged to a significant degree, to allow passage into the intracellular cytosolic space. Tris (hydroxymethyl) aminomethane (TRIS, pKa = 7.8 at 37 0 C) has been utilized in ischemic preparations. It is 30% nonionized at pH 7.4. Effron and associates" concluded TRIS was beneficial in the ischemiccat heart by an intracellular buffering mechanism. TRIS, however, affects tissue in a manner independent of its ability to accept hydrogen ions and has been thought to interfere with the binding of calcium at the cell membrane." Liedtke and co-workers" were able to improve postischemic function and show acceleration of glycolysis in the ischemic swine heart by use of TRIS and pyruvate. The use of histidine in concentrations of 150 to 200 mmol/L has been shown by Bretschneider and colleagues" to provide extracellular buffer capacity to stimulate anaerobic energy production. The imidazole-buffered cardioplegic solution used in these experiments departs from the Bretschneider solution in that the present solution is an extracellular type, high-sodium solution with buffering imparted. by imidazole, 25 mmol/L. The buffered Bretschneider solution contains sodium, 15 mmolfL, with the majority of the osmotic space occupied by histidine, 180 mmolfL. Imidazole (pKa = 7.0 at 25 0 C) (Fig. 4) has been
......
••••.••• Phosphate
6.7
6.3
.,
...-.
.... .... ......
.
.,.... Bicarbonate
6.1
o
.............. .
10
...... . -.. ......
20
.......... ....... 30
40°C
Fig. 5. The dissociation curve of imidazole parallels that of water as temperatures trend toward hypothermia. Other COmmon buffers depart from this physiological ideal.
used as a buffering agent for many years in laboratory medicine and microbiology." Imidazole is highly soluble in water, readily available, and stable in dry form. It is not a foreign compound, but is prevalent in nature as the proton-accepting component of histidine. The imidazole moiety of histidine exists in sufficient concentration in blood (30 mmolfL in human plasma) to dominate other buffer systems." Eighty percent of the buffering capacity of blood is the result of its presence. The importance of the imidazole group to physiological systems is that it is the only buffering agent that provides alkalinity of blood relative to the neutral pH of water as temperature decreases (Fig. 5) Imidazole, like its parent histidine, may exert biological effects over and above its buffering action. In addition to acting as a proton donor/acceptor, it has been thought to influence some enzyme-substrate binding reactions and to trigger the release of calcium
2 3 2 Vander Woude et al.
from the sarcoplasmic reticulum (special conditions, pH = 9.0 at 20° C).24 Comparison of imidazole to two commonly used buffers in cardioplegic solutions, bicarbonate and TRIS, shows that there are theoretical advantages to imidazole. Bicarbonate, if infused at a pH of 7.8 at 3r C in a cardioplegic solution, would exist predominantly in the HC0 3 state by the Henderson-Hasselbach relationship. For it to act effectively as a buffer, the Pco, must remain stable. Pco, rises in the closed system of cardioplegic arrest. If TRIS (pKa 7.8 at 37° C) is infused at pH 7.8, the active nitrogen is 50% protonated and 50% nonprotonated. Thus, a significant amount is available to cross the membrane to the intracellular space for buffering during cellular ischemia." Imidazole (pKa 6.7 at 37° C), if infused at pH 7.8, exists primarily in the nonprotonated form to satisfy the HendersonHasselbach relationship. Theoretically, it would be better suited for membrane passage and intracellular buffering. All indices of performance in the postischemic state of the imidazole-buffered group were significantly better than those in the bicarbonate-buffered group except for myocardial oxygen consumption. Slightly increased postischemic +LVdP jdts, reflecting better contractility, and significantly greater - LVdP j dts, reflecting improved diastolic relaxation, were seen. Concomitantly, coronary flows were improved after ischemia. Though graphic portrayal (Fig. 2) of the raw data for +LVdPj dt would indicate no statistical difference, the analysis of variance of preischemic minus postischemic differences did indicate significance. As the effect of the buffer is dependent on both preischemic and postischemic values, the lack of a postischemic difference between groups in no way precludes the effects as described. Most importantly, recovery of stroke work indices was significantly better in the hearts given imidazole-buffered cardioplegic solution. In the absence of information about ventricular volumes, it is not known whether the enhanced performances resulted from greater fiber shortening from the same end-diastolic volume (increased contractility), from increased end-diastolic volumes due to improved compliance, or both." As myocardial and coronary venous Pcos are similar at the moment of release of coronary occlusion, as shown by the work of Case, Felix, and Wachter," we chose to examine coronary sinus effluents at the release of the cross-clamp. Volumes of effluent and times of collection were similar between the two groups, so that sampling time variation was not a cause for the differences. Analysis of coronary sinus effluents supports our
The Journal of Thoracic and Cardiovascular Surgery
conclusions about the mechanism by which imidazole provides improved buffering capacity and therefore improved myocardial protection compared to bicarbonate. The Pco, of the imidazole-buffered cardioplegic solution was initially lower than that of the bicarbonatebuffered solution owing to the affinity of imidazole for the proton of the carbonic acid intermediate. As noted in Table I, three times the volume of O.lN HCl was required for pH adjustment in the imidazole solution to bring it equal to the bicarbonate solution. The absolute value of the coronary sinus Pco, was markedly higher for the bicarbonate-buffered experiments, concordant with the lower pH. The possibility that imidazole inhibited formation of thromboxane and led to improved coronary flows and postischemic recoveries in the imidazole group was considered unlikely, because thromboxane was not detected in the bicarbonate-buffered group by a highly sensitive technique. Thromboxane synthesis has not been demonstrated in the coronary artery"; rather the majority of prostaglandin released from the isolated heart is in the form of prostacyclin, a vascular smooth muscle relaxant. 28 The finding of greater concentrations of lactate in effluents of imidazole-buffered hearts would, at first, appear contrary to the concept of an improved acid-base status of the intracellular space by imidazole. Under conditions of ischemia, anaerobic glycolysis utilizes myocardial glucose and glycogen and results in the accumulation of lactate. The production of lactate from glucose results in no excess hydrogen ions, but the hydrolysis of adenosine triphosphate does generate hydrogen ions and results in intracellular acidosis.29 With ischemia, control of the glycolytic rate shifts to later reactions in the pathway, notably that catalyzed by glylceraldehyde 3-phosphate dehydrogenase. This enzyme is regulated by product inhibition, largely NADH accumulation.v-" If a lower or more favorably NADHjNAD+ ratio can be achieved by augmented buffering capacity, glycolysis may continue on to its anaerobic product, lactate. This would be in agreement with the work of Markov and associates," who used fructose 1,6, diphosphate as a substrate-enhancing agent in a dog model of myocardial ischemia. Markov's group observed improved mechanical recovery, regression of electrocardiographic ischemic changes, and significantly elevated levels of myocardial adenosine triphosphate and creatine phosphate in ischemic hearts given fructose diphosphate. Importantly, they found the tissue lactate content in ischemic muscle
Volume 90
Imidazole-buffered cardioplegic solution 2 3 3
Number 2 August, 1985
in the treated group was twice that found in the control group, indicating improved glycolytic flux in the treated group. The experiments reported here suggest that improved buffering capacity of the extracellular and intracellular myocardial spaces by a buffer not dependent upon removal of carbon dioxide may enhance postischemic functional recovery. It is not clear, however, whether this salutory effect is the result of enhancement of glycolytic energy production and maintenance of cellular volume regulation or decreased activation of cellular lysosomes and proteolytic enzymes." Additionally, imidazole may have alternative physiological actions independent of thromboxane synthetase inhibition and proton buffering. From information available in this set of experiments, it is believed that imidazoleservesas an effective buffering agent during normothermic global myocardial ischemia and improves postischemic functional recovery. We gratefully acknowledge the cooperation of consultant statistician Kenneth Schechtman, Ph.D., Biomedical Computer Laboratory, Washington University School of Medicine. We also express our appreciation to Ms. Patricia Black for technical assistance in performing these studies. REFERENCES
2
3
4
5
6
7
8
9
Buckberg GD: A proposed "solution" to the cardioplegic controversy. J THoRAc CARDIOVASC SURG 77:803-815, 1979 Kirklin JW, Conti VR, Blackstone EH: Prevention of myocardial damage during cardiac operations. N Engl J Med 301:135-141,1979 Nayler WG: Preservation of the myocardium. Some biological considerations, Towards Safer Cardiac Surgery, DB Longmore, ed., Boston, 1981, G. K. Hall Medical Medical Publishers, pp 627-651. Kubler W, Spieckermann PG: Regulation of glycolysis in the ischemic and anoxic myocardium. J Mol Cell Cardiol 1:351-377,1970 Rovetto MJ, Lamberton WF, Neely JR: Mechanisms of glycolytic inhibition in ischemic rat hearts. Circ Res 37:742-751,1975 Bretschneider HJ, Hubner G, Knoll D, Lohr B, Nordbeck H, Spieckermann PG: Myocardial resistance and tolerance to ischemia. Physiological and biochemical basis. J Cardiovasc Surg (Torino) 16:241-260, 1975 Jones JW, Gionis TA, Nichols RL, Markov A, Webb WR: Myocardial preservation with fructose 1,6 diphosphate. Energy without oxygen. Surg Forum 31:307-309, 1980 Henderson LJ: Concerning the relationship between the strength of acids and their capacity to preserve neutrality. Am J Physiol 21:173-179,1908 Good NE, Winget GD, Winter W, Connolly TN, Izawa S,
10
11
12 13
14 15
16 17
18
19
20
21
22 23
24
25
26
Singh RMM: Hydrogen ion buffers for biological research. Biochemistry 5:467-477, 1966 Bailey LE, Ong SD: Krebs-Henseleit solution as a physiological buffer in perfused and superfused preparations. J Pharmacol Methods 1:171-175, 1978 Neely JR, Liebermaster H, Battersby EJ, Morgan HE: Effect of pressure development on oxygen consumption by isolated rat heart. Am J Physiol 212:804-814, 1967 LangendorffO: Untersuchungen am uberlebenden Saugethierherzen. Pfluegers Arch 61:291, 1895 Needleman P, Bryan B, Wyche A, Bronson SD, Eakins K, Ferrendelli JA, Minkes M: Thromboxane synthetase inhibitors as pharmacological tools. Differential biochemical and biological effects on platelet suspensions. Prostaglandins 14:897-907, 1977 Levitsky S, Feinberg H: Biochemical changes of ischemia. Ann Thorac Surg 20:21-29, 1975 Williamson JR, Schaffer SW, Ford C, Safer B: Contribution of tissue acidosis to ischemic injury in the perfused rat heart. Circulation 53:Suppl 1:3, 1976 Sobel BE: Salient biochemical features in ischemic myocardium. Circ Res 34,35:Suppl 3:173, 1974 Rovetto MJ, Whitmer JT, Neely JR: Comparison of the effects of anoxia and whole heart ischemia on carbohydrate utilization in isolated working rat hearts. Circ Res 32:699-711, 1973 Beierholm EA, Grantham RN, O'Keefe DD, Laver MB, Daggett WM: Effects of acid-base changes, hypoxia, and catecholamines on ventricular performance. Am J Physiol 228:1555-1561, 1975 Effron MB, Guarnieri T, Frederiksen JW, Greene HL, Weisfeldt ML: Effect of tris(hydroxymethyl)aminomethane on ischemic myocardium. Am J PhysioI235:HI67H174,1978 Durrett LR, Adams HR: Inotropic responsiveness of atrial myocardium bathed in Tris- or bicarbonate-buffered solutions. Am J Physiol 237:H318-H325, 1979 Liedtke AJ, Nellis SH, Neely JR, Hughes HC: Effects of treatment with pyruvate and tromethamine in experimental myocardial ischemia. Circ Res 39:378-387, 1976 Diem K, Lentner C: Buffer Solutions, Scientific Tables, Ardsley, N. Y., 1970, Geigy Pharmaceuticals, p 281 Rahn H, Reeves RB, Howell BJ: Hydrogen ion regulation, temperature, and evolution. Am Rev Respir Dis 112:165172,1975 Chapman RA, Miller DJ: Structure-activity relations for caffeine. A comparative study of the inotropic effects of the methylxanthines, imidazoles, and related compounds on the frog's heart. J Physiol (Lond) 242:615-634, 1974 Khuri SF, F1aherty JT, O'Riordan JB, Pitt B, Brawley RK, Donahoo JS, Gott VL: Changes in intramyocardial ST segment voltage and gas tensions with regional myocardial ischemia in the dog. Circ Res 37:455-463, 1975 Case RB, Felix A, Wachter M: Measurement of myocardial pC0 2 with a microelectrode. Its relation to coronary sinus pCO,. Am J Physiol 236:H29-H34, 1979
234 Vander Woude et al.
27 Needleman P, Kaley G: Cardiac and coronary prostaglandin synthesis and function. N Engl J Med 298: 1122-1128, 1978 28 De Deckere EAM, Nugteren DH, Ten Hoor F: Prostacydin is the major prostaglandin released from the isolated perfused rabbit and rat heart. Nature 268:160-163, 1977 29 F1aherty JT, Schaff HV, Goldman RA, Gott VL: Metabolic and functional effects of progressive degrees of hypothermia during global ischemia. Am J Physiol 236:H839-H845, 1979 30 Katz AM: Physiology of the Heart, New York, 1977, Raven Press, pp 35-50
The Journal of Thoracic and Cardiovascular Surgery
31 Markov AK, Oglethorpe NC, Blake TM, Lehan PH, Hellems HK: Hemodynamic, electrocardiographic, and metabolic effects of fructose diphosphate on acute myocardial ischemia. Am Heart J 100:639-646, 1980 32 Swan H: The hydroxyl-hydrogen ion concentration ratio during hypothermia. Surg Gynecol Obstet 155:897-912, 1982 33 Bersohn MM, Scheuer J: Effects of physical training on end-diastolic volume and myocardial performance of isolated rat hearts. Circ Res 40:510-516, 1977
Bound volumes available to subscribers Bound volumes of THE JOURNAL OFTHORACIC AND CARDIOVASCULAR SURGERY are available to subscribers (only) for the 1985 issues from the Publisher, at a cost of $61.45 ($77.10 international) for Vol. 89 (January-June) and Vol. 90 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 30 days after publication of the last issue of the volume. The binding is durable buckram with the JOURNAL name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact The C. V. Mosby Company, Circulation Department, 11830 Westline Industrial Drive, St. Louis, Missouri 63146, USA; phone (800) 325-4177, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOURNAL subscription.