J Mol
Cell
Cardio121,
1273-1284
(1989)
Effects of Hemodynamic Variables on Myocardial K+ Balance During and After Shortlasting Ischemia Gunnar
Aksnes,
Qyvind EUingsen, and Arnfinn Iiebekk
David
L. Ruth
University of Oslo, Institute for Experimental Medical Ullev~l Hospital, Oslo, Norway (Received 23 March
Research,
1989, accepted in revisedform I August 1989)
on myocardial K+ 21, 1273-1284. Ischemia-induced myocardial potassium loss and post-ischemic potassium reuptake was quantitated in 8 open chest pigs during control conditions and during hemodynamic alterations which have been shown to increase steady state sarcolemmal potassium fluxes. Myocardial K+ balance was continuously computed before, during and after a 90 s occlusion of a branch of the circumflex artery during control (CTR), during pacing tachycardia (PACE: 34% increase in heart rate), during proximal aortic constriction (AC; 284/, increase in LVSP), and during isoprenaline infusion (ISO; 135% increase in LVdP/dt and 35% increase in heart rate). Ischemiainduced potassium loss increased significantly (40%) during IS0 only. Higher basal metabolic rate, increased sarcolemmal K+ conductance, or ischemia-induced depression of a more active Na/K-pump during IS0 are possible explanations to why increased K+ loss appeared in this situation. The maximal rate of post-ischemic potassium reuptake was not different from CTR during PACE and ISO, but it was reduced during AC, which might be due to persisting subendocardial ischemia in early reperfusion when ventricular wall stress is high. The extent of potassium restoration was not different from CTR during AC, PACE and ISO.
G. AKSNES, 0. balance during
KEY
WORDS:
ELLINGSEN, D. L. RUTLEN AND A; ILEEIEKK. Effects and after shortlasting ischemia. journal of Molecular
Myocardium;
Ischemia;
Potassium,
0022-2828/89/121273
+ 12 $03.00/O
and Cellular
In vivo; Pig; Tachycardia;
Introduction The magnitude and rate of the increase in extracellular potassium in ischemic myocardium is one important factor leading to electrophysiological derangementswhich predispose to cardiac arrhythmias during regional myocardial ischemia (Weiss and Shine 1982a; Kleber 1983; Pelleg et al., 1989). A large body of prior work suggeststhat underlying hemodynamic conditions influence potassium transport across the cell membrane. Steady state transsarcolemmal ion fluxes are increasedduring pacing tachycardia (Bassingthwaighte et al., 1976; Ilebekk et al., 1986; Webb and Poole-Wilson, 1986) and adrenergic stimulation (Gadsby 1983; Ellingsen et al., 1987a). Furthermore both increased heart rate and afterload reduces action potential duration (Lab, 1982) and effective refractory period (Reiter et al., 1988), and theseelectroPlease address all correspondance Oslo 4, Norway.
of hemodynamic
to: G. Aksnes, Institute
variables
Cardiology
Afterload;
(1989)
Contractility.
physiological alterations indicate that repolarizing sarcolemmal K+ efflux is enhanced. Hence it is likely, as indicated by the findings of Weissand Shine ( 1986), that the kinetics of K+ balance during and after ischemia are altered by several basal hemodynamic variables. Such alterations might in part explain why tachycardia (Epstein et al., 1973), pressure overload (Cove11et al., 1981; Lerman et al., 1985) and adrenergic stimulation (Nordrehaug et al., 1985) increase the risk of arrhythmias, especially during myocardial ischemia. We therefore studied the effects of heart rate, afterload and /I-adrenoceptor stimulation on myocardial potassium balance during and after coronary artery occlusionsof 90 s duration. We chose this duration of ischemia since the rapid increase in extracellular K+ which occurs in the ischemic myocardium in this period probably is imfor Experimental
Medical
Research, 0
UlIevHl
1989 Academic
Hospital,
0407
Press Limited
G. Aksnes
1274
portant for the pronounced reduction in ventricular fibrillation threshold and the high incidence of ventricular arrhythmias in the first 1-3 min of regional myocardial ischemia (Axelrod et al., 1975; Meesmann et al., 1978). Furthermore, previous experiments in our laboratory (Aksnes et al., 1989) have shown that more than five ischemic periods of this duration can be tolerated with reproducible ischemia-induced alterations in K+ balance. The objectives of the study were; (1) To determine whether tachycardia, increased afterload or fl-adrenergic stimulation affect K+ lossduring 90 s of ischemia; (2) To characterize quantitatively the K+ restorative responseduring reperfusion under thesedifferent hemodynamic conditions. Potassium balance was continuously computed before, during and after coronary artery occlusionsof 90 s duration by measuring flow through a shunt which drained the ischemic region, and by simultaneously measuring potassium concentration in the left atrium (arterial blood) and in the shunt (myocardial venous blood) with potassium sensitive valinomycin electrodes (Ellingsen et al., 1987 a,b). Methods Animal preparation
Eight domestic pigs of either sex (21-35 kg) were anesthetized with pentobarbital sodium by starting with a doseof 26-38 mg/kg intraperitoneally and continuing with an intravenous sustaining doseof 5-20 mg/kg h according to the depth of anesthesia. The pigs were artificially ventilated through a tracheostoma. We determined arterial blood gasesregularly and maintained SaOs > 95%, pH 7.3c7.50 and PC02 4.1-6.6 kPa. Body temperature was kept constant with wrappings and electrical heating pads. Urine was continuously drained through a cystostoma. We exposed the heart through a midsternal split and an incision in the fifth left intercostal space and suspendedit in a pericardial cradle. A distal part of the left anterior descendingor a branch of the circumflex coronary artery was dissected free for later intermittent occlusions with a Mayfield clip. We placed an elastic sling around the ascending aorta, and sutured pacing electrodesonto the left or right atrium.
et al.
A shunt was established from the coronary sinus to the right atrium (Andersen et al., 1983), and myocardial venous blood drained through the shunt when a preset stitch ligature placed at the entrance of the coronary sinusinto the right atrium was tightened. We administered heparin (600 IU/kg) to prevent blood clotting. Blood could be sampled from the shunt through a stopcock, and shunt flow was measured by an electromagnetic or Doppler ultrasonic probe in the shunt. We infused dextran and saline through a polyethylene catheter in a femoral vein to maintain stable hemodynamic conditions, and added KC1 to obtain a 4m~K+ concentration in all infusates. After completion of experiments, the pigs were killed by pentobarbital sodium injection. By injecting 20-30 ml 2,3,5-triphenyltetrazolium chloride (20 mg/ml) retrogradely into the shunt immediately thereafter, the drained myocardium became stained brickred. Cardiogreen injected into the distal portion of the occluded coronary artery stained the ischemic region green. After dissection, the stained areas were weighed. The shunt drained 96 (77-loo)% of the myocardium rendered ischemic, and the ischemic region constituted 2 1 ( 15-30) o/oof the drained tissue. All metabolic parameters were expressedper 100 g of drained tissue. Potassium measurements
[K+] in arterial ([K+]a) and coronary sinus ([K+]cs) blood were continuously measured by flexible valinomycin electrodes. We inserted the electrode for arterial [K+] measurements into the left atrium and secured it by a purse-string suture. The electrode for myocardial venous blood [K+] measurements was introduced into the shunt through a side hole. Ellingsen et al. (1987a,b) have described the design and calibration procedures of the electrodes and measuring system. [K+] in arterial and coronary sinus blood as well as averaged shunt flow (SF) were recorded at 3Hz sampling frequency before, during and after coronary artery occlusions. We performed subsequentcalculations on data points representing the mean over 3 s. Arterial- coronary sinus [KC] difference prior to each coronary occlusion was set to
Myocmrdial zero to obtain the net [K+]a-[K+]cs changes ([K+]a-cs) caused by ischemia and reperfusion. Hence, as seen in Figure 1, a potassium release was recorded during coronary artery occlusion and early reperfusion (between A and B in Fig. l), and thereafter a K+ reuptake ensued (between B and C) until a new steady state was reached (at C), We calculated the instantaneous rate of myocardial K+ release and reuptake (pmol/lOOg min) by the formula: Rate of K+ releaselreuptake = [K+]a-cs 1.04 (1-Hct) SF lOO/MW. The factor 1.04 corrects for underestimation of the plasma
;
cn
.-w.
4oJ
Metabolic
r; +
4.6
p
4.4
2 ‘5 ,z
4.2
g
4.0
..-. _
.j ‘I ;i .--. ” ._,.
A -4. .A_-.
-_..-.I’ y
PER100
. . .-.--_-..
_ ._~._ - ._... ‘-
.
--.
ISOPRENALINE
z
6 ?
studiej
Metabolic variables were studied in the first four occlusions.Both during CTR, AC, PACE
- _.-.._ -
E
8
fraction by the centrifugation method, Hct is the hematocrit in coronary sinus. blood, SF is shunt flow and MW is the weight of myocardial tissue drained by the shunt. We assumed that steady state was reached when [K+]a-cs changed ~0.02 mM over five 3 s data points. Time integrals of the instantaneous rate of K ’ release and reuptake represent accumulated K+ release and reuptake.
II IU II! SAMPLING
4.6
1275
Ia+ Balance
-C _c- ,-‘-
IK+l _ CK+l --
4.2 4.0
I 3.8
: + 2 a $
FIGURE I. Records from two coronary artery occlusions, one during control hemodynamics and one during isoprenaline i.v. [K+la - [K’Jcs difference prior to occlusion was set to zero to obtain the net changes in this difference induced by ischemia and reperfusion. Kf release was recorded in the period A-B and K+ reuptake during B-C. Note greater K+ loss during coronary occlusion in the isoprenahne stimulated heart. Blood sampling periods for metabolic studies are marked I-IV. Arrows mark horizontal displacement between the three tracings. The [K’]a and [K’]cs tracings have “changed places” from control to isoprenaline due to adjustments on the graph recorder, but the [I(+] values sampled by the computer was unaffected by this adjustment.
1276
G. Aksnes
and IS0 arterial and coronary sinus blood was sampled before coronary occlusion (I), during coronary occlusion (II), during the first 15 s of reperfusion (III) and concomitantly with the maximal rate of K+ reuptake (IV) (Fig. l), and myocardial oxygen consumption and lactate and phosphate balance were calculated. We determined hemoglobin (Hb) in duplicate by the cyanmethemoglobin method. 02 saturations of arterial (Sa02) and coronary sinus (ScsOz) blood were analysed on an IL 282 CO-Oximeter (Instrumentation Laboratories) and myocardial oxygen consumption was calculated as: MVOz = ((SaOz - ScsOz) 0.62 1 Hb + Pa02 - PcsOz) 1.01) SF lOO/MW Pa02 and PcsOz are partial pressures of oxygen in arterial and coronary sinus blood, respectively. We determined lactic acid concentration in arterial blood, [La]a, and in coronary sinus blood, [La]cs, by the method described by Rosenberg and Rush ( 1966). The rate of lactate loss was calculated as: Rate
of
lactate
loss = [La]cs 1OO/MW
- [La]a
SF
Arterial and coronary sinus blood gases and pH were analyzed on an AVL 945 (Automatic Blood Gas System). We determined plasma phosphate in arterial blood, [Ph]a, and in coronary sinus blood, [Ph]cs, by the method described by Bodansky (1932-33). The rate of phosphate loss was calculated as: Rate of phosphate loss = [Ph]cs - [Ph]a (1-Hct) SF lOO/MW
1.04
et al.
electromagnetic flow probe snugly fitting the ascending aorta. The electromagnetic flow probes on the aorta and in the coronary sinus shunt were connected to a two-channel flowmeter square-wave electromagnetic (model 376, Nycotron, Drammen, Norway). In five experiments shunt flow was measured by an extracorporal flow-through ultrasonic probe connected to an ultrasonic doppler flowmeter (VF-1, model PD-10, ValpeyFisher, Hopkinton, MA). We monitored ail hemodynamic measurements continuously on an eight channel galvanometric recorder (model 7758 B, Hewlett-Packard).
Exberimental
procedure
A total of five coronary artery occlusions of 90 s duration were performed at 25-35 min intervals. The first and last occlusion was done during control hemodynamics (CTR). We kept heart rate constant by atria1 pacing at 10 beats/min above spontaneous frequency during CTR. and aortic constriction (AC). Between the first and last CTR we performed thrte occlusions, one during PACE, AC and ISO. AC was alternately performed first and last of these three interventions, while PACE always preceded IS0 to allow us to adjust heart rate to the same level during IS0 and PACE. The required intravenous infusion rate for isoprenaline was 0.2-0.3 pg/kg min. The interventions AC, PACE and IS0 were maintained for 10 min before and 5-10 min after coronary artery occlusion, and the minimal recovery period between interventions was 15 min. Statistics
Hemodynamic measurements Left ventricular pressure (LVP) was recorded by a microtip pressure transducer catheter (model PC-470, Millar Instruments, Houston, TX) introduced through the apical dimple of the heart. We used the maximal positive value of the first derivative of LVP, LVdP/dt, as the contractility parameter. We measured arterial blood pressure by a Statham pressure transducer connected to a polyethylene catheter inserted into a femoral artery, and recorded aortic flow by an
Data are presented as the median values with a non-parametric 95% confidence interval (in brackets) based on Walsh numbers (Noether 1976). The Friedman test (Conover 1980) was used for comparison of: (1) parameters of potassium balance from the five coronary occlusions (CTR, AC, PACE, ISO, CTR); (2) metabolic parameters during CTR, AC, PACE and IS0 in each of the four sampling periods; (3) hemodynamic control parameters before each of the five coronary occlusions and (4) to analyse the effect of ischemia and reperfusion on LVdP/dt in each of the five situ-
Myocardid
K+ Balance
1277
ations (CTR, AC, PACE, ISO, CTR). A the five occlusions.The total ischemiainduced probability value P < 0.05 was considered K+ release was 92.7 (57.9-147.8), 94.8 statistically significant. (63.8-128.1), 101.8 (53.9155), 142.2 (88-198) and 82.3 (41.2-132.7) ~mol/lOOg during CTR, AC, PACE, IS0 and CTR, Results respectively. Potassium balance Total K+ reuptake was not statistically difOne aim in this study was to determine ferent among the five occlusions and amounwhether ischemia-induced K+ lossand/or the ted to 64-98% of the preceding K+ loss.The postischemic K+ restoration were affected by maximal rate of K+ reuptake occurred after heart rate, afterload or /I-adrenoceptor 25-35 s of repetfusion. This rate was signifistimulation. cantly reduced during AC and the last CTR A myocardial K+ losswas recorded during when compared to the first CTR (P < 0.05), coronary artery occlusion (Figs 1 and 2). On but unchanged during PACE and IS0 (Fig. reperfusion, accumulated K+ was rapidly 3). The maximal rate of K+ reuptake was 14.2 washed out from the ischemic tissue,and after (9.3-23.9), 12.4 (4.9-19.3), 14.8 (7.5-25.1), 15-20 s of reperfusion, K+ reuptake occurred. 17.7 (11.2-24) and 9.5 (5-12.1) ~mol/lOOg The K+ reuptake lasted approximately 95 s min during CTR, AC, PACE, IS0 and CTR, until a new steady state was reached. The respectively. qualitative changes in potassium balance induced by ischemia and reperfusion were the Myocardial oxygen consumption and lactate and same in all five coronary occlusions (CTR, phosphate balance AC, PACE, ISO, CTR). Figure 2 illustrates the K+ releaserecorded Since anion linked potassium extrusion is one during ischemia and early reperfusion. Beta- of the suggestedexplanations for the ischemiaadrenergic stimulation increased the Kf loss induced K+ loss(Gaspardone et al., 1986), we recorded during ischemia and the total Kf calculated myocardial balance for lactate and loss(P < 0.05), whereas the K+ lossrecorded phosphate which are two of the most importduring early reperfusion did not differ among ant anions produced during ischemia. There
:TR
FIGURE 2. CTR = control naline infusion. among the five
P < 0.05.
AC
PACE
IS0
CT?!
Illustration of K+ release during &hernia (1) and early reperlusion (2) and total K+ release (1 + 2). hemodynamics, AC = aortic constriction, PACE = pacing tachycardia and IS0 = intravenous isopreBrackets represent 95% confidence intervals. Cmresponding columns (1, 2, 1 + 2) were compared different occlusions, and asterisks denote statistically significant differences from the first CTR, *:
G. Ahncs
1278
CTR
AC
PACE
IS0
CTR
FIGURE 3. This figure shows the maximal rate of postischemic K+ reuptake. Abbreviations as in Figure 2. Asterisks denote statistically significant differences from the first CTR. *: P < 0.05.
was a myocardial lactate uptake and no significant phosphate uptake or loss before and during occlusion in both CTR, AC, PACE and ISO. We believe that the change in rate of
a -0
Sampling r--m
5or
et al.
lactate and phosphate lossfrom before occlusion to reperfusion (sampling period III and IV), is due to washout of these ions from the ischemic tissue.Hence, there appears to be no difference (Fig. 4) in lactate accumulation during ischemia between CTR, AC, PACE and ISO, whereas phosphate washout is significantly reduced during IS0 ascompared to CTR, AC and PACE (P < 0.05). Opening of ATP sensitive K+ channels is another possiblemechanism for the ischemic K+ loss (Noma, 1983). The rate of fall in intracellular ATP during ischemia is likely to vary with metabolic rate in the ischemic tissue. Hence we determined steady state myocardial oxygen consumption before coronary occlusion, and this was not statistically different among CTR, AC and PACE, but was approximately 20% higher during IS0 as compared to CTR (Table 1).
period
Sampling
period
I-IS!
_
-r--
CTR
AC PACE
IS0
CTR
AC
PACE
IS0
FIGURE 4. Changes in myocardial lactate and phosphate balance from before ischemia (I) to early reperfiusion (III and IV). These changes are induced by washout of lactate and phosphate from ischemic tissue in early reperfusion. Abbreviations as in Figure 2. Changes from I-III and I-IV during AC, PACE and IS0 were compared to those during CTR. Asterisk denotes statistically significant difference from CTR, *: P < 0.05.
Myoamlial
K+ Balance
1279
1280
G. Aksnes
et al.
extent of postischemic potassium repletion is Steady state hemodynamic parameters prior not affected by heart rate, afterload or /Iadrenergic stimulation whereas the maximal to coronary artery occlusion during CTR, AC, PACE, IS0 and CTR are listed in Table rate of K+ reuptake is reduced during in1. Peak left ventricular systolic pressure creasedafterload. (LVSP) was increased 28% compared to CTR during AC. Heart rate (HR) was maintained 34% above CTR during PACE and ISO, and maximal positive left ventricular dP/dt (LVdP/dt) during IS0 was more than double CTR value. The pressurerate product (HR x LVSP) is commonly used as an indirect measure of myocardial oxygen consumption. Steady state pressure rate product before coronary artery occlusion was 28.5 (22.1-37.8), 25.3 (14.6-36.5) and 32.6 (20.3-48.4)% above the first CTR during AC, PACE and IS0 respectively, and these increaseswere not statistically different from one another. Despite these PACE very similar rises in pressure rate product, 0 myocardial oxygen consumption increased t during IS0 only, probably because of the 2 P * significant fall in stroke volume during AC 3 and PACE. 2 Heart rate was kept constant during occlu.-300 % sion and reperfusion by atria1 pacing. Aortic 3OOr x pressure,stroke volume and LVSP fell slightly E during all five coronary occlusions, and then .-E’ regained preocclusion values within 30-120 s :: E of reperfusion. The average decline in distal blood pressure when the aortic snare was tightened was only 14 mmHg. The changes in LVDP/dt during coronary occlusion and reperfusion are shown in Figure 5. A significant decline occurred during coronary occlusion. The absolute reduction was largest during ISO. In the early reperfusion period, a transient, significant increasein LVdP/dt above preocclusion value was observed in the first and last CTR. A similar pattern was seenduring AC, PACE and IS0 as well, although the early postischemic increasein contractility did not reach statistical 1 200L significance during theseinterventions. OCCI. Hemodynamic measurements
r” 2 100
I
i-
t
4 012345 Time
( min 1
Discussion
The present study shows that the ischemiainduced myocardial potassium loss during a 90 s coronary occlusion is increased by about 40% during beta-adrenergic stimulation. The
FIGURE 5. Changes in maximal positive LVdP/dt associated with coronary artery occlusions of 90 s duration. Abbreviations as in Figure 2. Asterisks denote statistically significant changes from preocclusion values, *: P < 0.05, **: P < 0.001.
Myocardial Some methodological considerations We recorded a K+ loss during ischemia and a washout of K’ during early reperfusion (Figs 1 and 2). The removal of K+ during ischemia is possibly due to diffusion of K+ from the ischemic area into the surrounding normally perfused tissue (Coronel et al., 1988). However, with the diffusion distances involved, it is unlikely that the substantial K+ loss recorded during ischemia, which during IS0 amounts to 400/, of the total loss, occurs by diffusion alone. An additional possible explanation is that venous blood is alternately squeezed out of and sucked into the veins in the ischemic tissue during the cardiac cycle. Such retrograde venous flow was suggested by Curtis et al., (1984) to explain the entry into ischemic tissue of verapamil given during coronary occlusion in a rat model where collateral flow was low. Higher flow in the surrounding nonischemic tissue and more vigorous contractions (and hence more venous “backwash”) during IS0 might explain why the K+ loss recorded during coronary occlusion is higher in this situation (40:/, compared to 26-270/6). An important implication of our finding that 26640”/, of the total ischemia-induced K+ loss was removed from the ischemic tissue during coronary occlusion, is that extracellular recordings of K+ concentrations in similar models of regional ischemia (Gettes et al., 1986) would lead to underestimations of ischemia-induced K+ loss, especially during adrenergic stimulation. The K+ loss recorded during early reperfusion most likely represented washout of K+ accumulated in the ischemic tissue and total K+ loss is therefore the ischemia-induced K+ loss. A post-ischemic K+ reuptake after brief coronary artery occlusions is a very consistent and reproducible finding in our experimental setup. This contrasts with studies in which extracellular [K+] in the ischemic region is measured since they only inconsistently demonstrate a postischemic K+ reuptake (Wiegand et al., 1979; Hill and Gettes, 1980; Weiss and Shine, 1982a,b and 1986; Coronel et al., 1988). Damaged cells close to the extracellular electrodes and slow equilibration in the sink of extracellular fluid which surrounds those electrodes might explain the discrepacny with our observations. Since those possible shortcomings for extracellular recordings are
K+ Balance
1281
avoided by non-traumatic intravascular recordings we think that K+ restorative processes during reperfusion can be more accurately characterized in our model.
Ixhemia-induced
K’
loss
Pacing tachycardia did not enhance ischemic K+ loss, and K+ loss was enhanced during IS0 compared to PACE despite equal heart rate. These findings show that heart rate by itself, i.e. in the absence of increased myocardial oxygen consumption, is not an important determinant for ischemic K+ loss. This contrasts with the observations by Weiss and Shine (1982b and 1986) who found that the rate of extracellular K+ accumulation early during ischemia increased with heart rate. However, our observations are in accordance with the finding of Hill and Gettes (1980) that the rate of extracellular K+ accumulation in the first 2 min of ischemia is the same at a heart rate of 90 and 140 beats/min. Kleber (1983) found no increase in intracellular [Na+] during the first 15 min of ischemia, and the Na/K-pump has been shown to function in this period (Weiss and Shine 1982b). Since Na+ influx and K + elllux are coupled in an action potential, it is therefore likely that the Na/K-pump compensates for the depolarization dependent K+ loss and prevents intracellular Naf accumulation in the first minutes of ischemia (Kleber, 1984). This means that depolarization independent K+ efllux, which at a heart rate of 120 beats/min constitutes 8O”/b of the K+ efflux (Ellingsen et al., 1988), determines the rate of during early extracellular K + accumulation ischemia. Increased afterload with unchanged myocardial oxygen consumption during coronary occlusion did not affect ischemia-induced K+ loss. Hence, the increased steadv state sarcolemmal K + conductance which is thought to exist in ventricular muscle when preload or afterload is increased (Lab, 1982; Lerman et al., 1985; Reiter et al., 1988) does not seem to predispose to arryhmias by augmenting ischemia-induced K+ loss. stimulation increased total Adrenergic ischemia-induced K+ loss by about 40°, compared to both CTR, AC, PACE and CTR.
1282
G. Aksmcs et al.
This observation agreeswell with the finding by Gettes et al. (1986) that isoprenaline leads to a more rapid initial rise in extracellular [K+] during ischemia. Several possible explanations for this observation exist. Firstly, myocardial oxygen consumption (MVOZ) was increasedduring IS0 only, and MVOZ can conceivably affect ischemic K+ lossin at least two different ways: (1) If K+ lossis anion linked (Gaspardone et al., 1986) and production of lactate and phosphate were higher in the ISO-stimulated ischemic tissue, then increased K+ lossduring ischemia would be expected. However, washout of lactate and phosphate in early reperfusion was not enhanced during ISO. Hence our findings are in accordance with the recent observations by Weiss et al., (1989) that efflux of K + and anions during ischemia easily can be dissociated. (2) A higher MV02 in the ISOstimulated ischemic tissue would probably lead to a more rapid depletion of intracellular ATP, and hence to a more rapid K + efflux through ATP-sensitive K + channels. Compartmentalization of ATP and its depletion at a critical site does make this mechanism possibleeven after occlusionsas short as90 s. Secondly, during b-adrenergic stimulation increased sarcolemmal K+ conductance (gK+ ) (Gadsby 1983) is balanced by augmented Na/K-pump rate (Desilets and Baumgarten, 1986; Ellingsen et al., 1987a,b and 1989). The higher gK+ might enhance ischemia-induced K+ lossboth in the presence and absenceof preserved Na/K-pump activity. An increasein K + efflux and extracellular [K+] without a concomitant rise in intracellular sodium, which is the main regulator of Na/K-pump activity, will not stimulate even a normally functioning Na/K-pump. However, one cannot exclude that partial Na/K-pump failure contributes to the increase in ischemiainduced Kf lossin the adrenergically stimulated heart. Although in vitro experiments shows preserved Na/K-pump activity for the initial 10-l 5 min of ischemia (Weiss and Shine, 198213,Kleber 1983), it is possiblethat the approximately doubled Na/K-pump rate during isoprenaline infusion in our preparation (Ellingsen et al., 1989) cannot be maintained during ischemia.
Postischemic Kf
reupake
When ischemic tissue is reperfused accumulated ions and metabolites are washed out. However, there are additional rapid transsarcolemmal ionic shifts such as a reuptake of potassium. These rapid ionic shifts in the early post-ischemic period are probably important for the development of reperfusion arrhythmias, and hence it is not surprising that reperfusion is much more arrhythmogenie than for instance washout of localized hyperkalemia (Pelleg et al., 1989). We therefore wanted to determine how one of these ionic shifts, K+ reuptake, is affected by hemodynamic variables. In the present study we found no difference between CTR, AC, PACE, IS0 and CTR in the extent of potassium restoration during reperfusion. The maximal rate of K+ reuptake was not increased during ISO, and hence the increased Na/K-pump activity induced by /?adrenergic stimulation only seemsto balance the increase in K+ e&x. The reduction in maximal rate of K+ reuptake with AC might be due to persisting subendocardial ischemia in early reperfusion, since both systolic and diastolic wall tension is elevated during AC. Both the presentstudy and previous work in our laboratory (Aksnes et al., 1989) demonstrate a postischemicpotassiumreuptake and a concomitant enhancement of contractility after brief periods of ischemia in the open chest pig. The mechanism for K+ reuptake and hypercontractility was discussedin our previous paper. Stimulation of the Na/Kpump and favoured Na+/Ca* exchange due to increased intracellular sodium in early reperfusion was the most likely explanation. The absence of significant reperfusion hypercontractility during AC might be due to high afterload, and both during AC, PACE and IS0 persisting subendocardial ischemia in early reperfusion might prevent hypercontractility. During IS0 an alternative explanation is that the high LVdPldt makesdetection of a small increment in this variable dificult. In summary, the present study showsthat fladrenergic stimulation increases myocardial K+ lossduring shortlasting regional ischemia in open chest pigs, whereas neither increased depolarization frequency nor the enhanced
Myocardkl sarcolemmal potassium conductance which exist during tachycardia and increased afterload affect ischemic K+ loss. The increased K+ loss during /I-adrenergic stimulation might be due to a higher myocardial oxygen consumption, increased sarcolemmal K+ conductance and/or partial Na/K-pump failure. The present study also shows that the extent of postischemic K + restoration is independent of afterload, heart rate and inotropy. Acknowledgements
K+ Balance
1283
Science, Professor Carl Semb’s Medical Research Fund and The Norwegian Council for Cardiovascular Diseases.D. L. Rutlen was supported by National Institute of Health grant ROI HL34204. Skillful technical assistancewas offered by B. Austbq H. B&en, I. B. Hansen, U. Lie Henriksen, M. Ree Holte, 0. Moen, 0. Nordby, G. Steinszter Jensen and T. Verpe. We are especially grateful to’ B. Amundsen and electronic engineer S. Leraand for the manufacturing of the potassium electrode system.
This work was supported by grants from Anders Jahre’s Fund for the Promotion of References DL, ~LEB~KK A (1989) Myocardial K+ repletion and rise in contractility after brief ischemic periods in the pig. J Mel Cell Cardiol, 21: 681-690. ANDERSEN FR, SEJERSTED OM, ILEBEKK A. (1983) A model for quantitative sampling ofmyocardial venous blood in the pig. Acta Physiol Stand 119: 187-195. AXELROD PJ, VERRIER RL, LOWN B (1975) Vulnerability to ventricular fibrillation during acute coronary arterial occlusion and release. Am J Cardiol36: 776-782. BASSINCTHWAIGHTE JB, FRY CH, MCGUIGAN JAS (1975) Relationship between internal calcium and outward current in mammalian ventricular muscle; a mechanism for the control of the action potential duration? J Physiol (Land j 262: 15.-37. BODANSKY A (1932-33) Phosphate studies 1: Determination of inorganic phosphate. J Biol Chem 99: 197-206. CORONEL R, FIOLET JWT, WILMS-SCHOPMAN FJG, SCHAAPHERDER AFM, JOHNSON TA, GETTES LS, JANSE MJ ( 1988) Distribution ofextracellular potassium and its relation to electrophysiologic changes during acutc myocardial iachcmia in the isolated perfused porcine heart. Circulation 77: 1125-I 138. COVELL JW, LAB MJ, PAVELEC R (1981) Mechanical induction of paired action potentials in intact hearts uz v/u. .J Physiol (Lond) 320: 34~. CONOVER WJ (1980) Practical Nonparametric Statistics, John Wiley & Sons, New York. Cua,ns M.J, MACLEOD BA, WALKER MJA (1984) Antiarrhythmic actions of verapamil against ischaemic arrhvrhmias in the rat. Br J Pharmacol83: 373-385. pump in isolated cardiac DESILETS M, BAUMGARTEN CM (1986) Isoproterenol directly stimulates the Na’-K+ myocytes. Am J Physiol 251: H218-H225. ELLINCSEN 9, SEJERSTED OM, LERAAND S, ILEBEKK A (1987a) Catecholamine induced myocardial potassium uptakr mediated by B,-adrenoceptors and adenylate cyclase activation in the pig. Circ Res 60: 540-550. ELLINCSEN 9, VENGEN qA, ILEBEKK A (1987b) Myocardial potassium uptake during a- and /%adrrnoccptor stimulation. Am J Physiol253: H799-H810. ELLINGSEN Q, SEJERSTED OM, VENCEN QA, ILEBEKK A (1988) Potassium homeostasis ofthp intact, bratirlg htatr .A( IB Physiol &and 134 (Suppl 575): S23 (abstract). ELLINGSEN Q, SEJERSTED OM, VENCEN (PA, ILEBEKK A (1989) In &I quantification of myocardial Na-K pump rate during /?-adrenergic stimulation of intact pig hearts. Acta Physiol Stand 135: 493-503. EPSTEIN SE, GOLDSTEIN RE, REDWOOD DR, KENT KM, SMITH ER (1973) The earlv phase of acute mbocdrdial infarction: pharmacologic aspects of therapy. Ann Int Med 78: 918-936. GADSBY DC: (1983) /%Adrenoceptor agonists increase membrane K+ conductance in cardiac Purkinjr fibres. Nature 306: 691-693. GETTES LS, SYMANSKI JD, FLEET WF, JOHNSON TA, GRAEBNER C (1986) The intracellular and extracellular changes associated with ischaemia-effects of catecholamines in arrhythmogenesis. Euro Heart J 7 (Supplement Ai: 77 .84. GASPARDONE A, SHINE KI, SEABROOKE SR, POOLE-WILSON PA (1986) Potassium loss from rabbit myocardium during hypoxia: evidence for passive rfBux linked to anion extrusion. J Mel Cell Cardiol 18: 389-399. Htu JL, GETTES LS 11980) Effect of acute coronary artery occlusion on local myocardial extracellular Kf ac.ti\itv in swine. Circulation 61: 768-778. ILEBEKK A, ANDERSEN FR, SEJERSTED OM (1986) Magnitude ofmyocardial potassium changes during acute ahrrations in pacing frrbqurnry in the irr sifu pig heart. Cardiovasc Res 20: 176-181. AKSNES
G. ELLINCSEN
Q, RUTLEN
1284
G. Aksnee
et al.
AG (1984) Extracellular potassium accumulation in acute myocardial ischemia. J Mol Cell Cardiol 16: 389-394. KLEBER AG (1983) Resting membrane potential, extracellular potassium activity, and intracellular sodium activity during acute global ischemia in isolated perfused guinea pig hearts. Circ Res 52: 4422450. LAB MJ (1982) Contraction-excitation feedback in myocardium. Physiological basis and clinical relevance. Circ Res 50: KLEBER
757-766.
LERMAN BB, BURKHOFF D, YUE DT, SAGAWA K (1985) Mechanoelectrical feedback: independent role of preload and contractility in modulation of canine ventricular excitability. J Clin Invest 76: 1843-1850. MEESMANN W, WIEGAND V, MENKEN U, KOMHARD W, REHWALD U (1978) Early mortality due to ventricular fibrillation, and the vulnerability of the heart following acute experimental coronary occlusion: possible mechanisms and pharmacological prophylaxis. In: The Arterial System. Dynamics, Control, Theory and Regulation pp. 275-284. Springer Verlag: Berlin-Heidelberg-New York. NOETHER G (1976) Znlroduction to Statistics: A Nonparametric A,bproach, Edit. 2. Boston, Houghton Milllin: 128-135. NOMA A (1983) ATP-regulated K+ channels in cardiac muscle. Nature 305: 147-148. NORDREHAUC JE (1985) Malignant arrhythmia in relation to serum potassium in acute myocardial infarction. Am J Cardiol 56: 20D23D. PELLEC A, MITAMURA H, PRICE R, KAPLINSKY E, MENDUKE H, DREIFUS LS, MICHELSON EL (1989) Extracellular potassium ion dynamics and ventricular arrhythmias in the canine heart. J Am Co11 Cardiol 13: 941-950. REITER MJ, SYNHORST DP, MANN DE (1988) Electrophysiological effects of acute ventricular dilatation in the isolated rabbit heart. Circ Res 62: 554-562. ROSENBERG JC, RUSH BF (1966) An enzymatic-spectrophotometric determination of pyruvic and lactic acid in blood. Clin Chem 12: 299-307. WEBB SC, POOLE-WILSON PA (1986) Potassium exchange in the human heart during atria1 pacing and myocardial ischaemia. Br Heart J 55: 554-559. WEISS J, SHINE KI (1982a) [K’]o accumulation and electrophysiological alterations during early myocardial ischemia. Am J Physio1243: H318-H327. WEISS J, SHINE KI (1982b) Extracellular K+ accumulation during myocardial ischemia in isolated rabbit heart. Am J Physiol242: H619-H628. WEISS J, SHINE KI (1986) Effects of heart rate on extracellular [K+] accumulation during myocardial ischemia. Am J Physiol250: H982-H99 1. WEISS JN, LAMP ST, SHINE KI (1989) Cellular K” loss and anion e@lux during myocardial ischemia and metabolic inhibition. Am J Physiol256: Hl165-Hl175. WIEGAND V, Guccr M, MEESMAN W, KESSLER M, GREITSCHUS F (1979) Extracellular potassium activity changes in the canine myocardium after acute coronary occlusion and the influence of beta-blockade. Cardiovasc Res 13: 297-302.