Rapid Ventricular Pacing Produces Myocardial Protection by Nonischemic Activation of KATP+ Channels
Background Rapid ventricular pacing reduces the incidence of ventricular arrhythmias during a subsequent sustained period of ischemia and reperfusion. We investigated whether rapid ventricular pacing also limits myocardial infarction and determined the role of KATP+ channels in the protection afforded by ventricular pacing.
Methods and Results Myocardial infarction was produced by a 60-minute coronary artery occlusion in open chest pigs. Infarct size of pigs subjected to 10 minutes of ventricular pacing at 200 beats per minute followed by 15 minutes of normal sinus rhythm before the occlusion (79±3% of the area at risk, mean±SEM) was not different from control infarct size (84±2%). Thirty-minute pacing followed by 15-minute sinus rhythm resulted in modest reductions in infarct size (71±2%, P<.05 versus control). Thirty minutes of pacing immediately preceding the occlusion without intervening sinus rhythm resulted in considerable limitation of infarct size (63±4%, P<.05), which was abolished by pretreatment with the KATP+ channel blocker glibenclamide (78±4%, P=NS). KATP+ channel activation did not appear to involve ischemia: (1) myocardial endocardial/epicardial blood flow ratio was 1.07±0.08, (2) phosphocreatine and ATP levels and arterial-coronary venous differences in pH and Pco2 were unchanged, (3) end-systolic segment length did not increase and postsystolic shortening was not observed during pacing, and (4) systolic shortening recovered immediately to baseline levels and coronary reactive hyperemia was absent after cessation of pacing. Administration of glibenclamide after 30 minutes of pacing at the onset of 15 minutes of normal sinus rhythm did not attenuate the protection (73±3%, P<.05 versus control), suggesting that KATP+ channels did not contribute to the moderate degree of protection that was still present 15 minutes after cessation of pacing.
Conclusions Rapid ventricular pacing protects the myocardium against infarction via nonischemic KATP+ channel activation. Continued activation of KATP+ channels does not appear mandatory for the protection that is still present 15 minutes after cessation of pacing.
Myocardial preconditioning can be induced by a variety of ischemic stimuli. Thus, one or more brief total1 or partial2 3 4 coronary artery occlusions can limit infarct size produced by a sustained ischemic period. Moreover, infarct size can be limited by transient ischemia in adjacent myocardium5 or even different organs.6 7 In all these studies, a temporary interruption or restriction of oxygen supply either within or outside the myocardial region of interest was required to produce protection. Recent studies suggest that nonischemic stimuli also may protect the myocardium. Thus, Ovize et al8 reported that an increase in left ventricular wall stretch produced by acute volume overload protected the myocardium against infarction during a subsequent 60-minute coronary artery occlusion. Also, two consecutive 2-minute periods of RVP in open chest dogs reduced the incidence of ventricular arrhythmias during and immediately after subsequent 25-minute coronary artery occlusion.9 In contrast, Marber et al10 failed to show a protective effect of a single 5-minute period of rapid atrial pacing against myocardial infarction in the rabbit heart. To date, no study has addressed the effect of RVP on infarct size development produced by a sustained coronary artery occlusion.
In the present study, we investigated whether RVP preceding a 60-minute TCO altered infarct size development in open chest pigs. In two groups of animals we studied the effects of either 10 or 30 minutes of RVP followed by 15 minutes of NSR on infarct size produced by 60-minute TCO, analogous to the classic preconditioning model of a brief ischemic stimulus followed by reperfusion. In view of our earlier findings that ischemia produced by a partial coronary artery occlusion can precondition the myocardium without the need of intervening reperfusion,3 we also studied a third group of animals in which the 30-minute RVP period preceded the 60-minute TCO without NSR. If RVP produces protection by causing ischemia,9 11 12 protection is likely to be distributed heterogeneously across the left ventricular wall because ischemia would occur predominantly in the inner layers. Consequently, infarct size also was determined for the outer and inner halves of the left ventricle. To explore the mechanism of myocardial protection produced by RVP, we investigated whether ischemia occurred in animals subjected to 30 minutes of RVP followed by 180 minutes of NSR without 60-minute TCO. Also, in view of evidence that ventricular pacing can activate ventricular K+ channels13 14 and that activation of KATP+ channels is cardioprotective in pigs,15 16 we also studied the role of KATP+ channels in protection produced by RVP. For this purpose, one group of animals that was subjected to 30 minutes of RVP immediately followed by 60-minute TCO were pretreated with the KATP+ channel blocker glibenclamide before the pacing period. In another group of animals that was subjected to 30 minutes of RVP followed by 15 minutes of NSR before the 60-minute TCO, glibenclamide was administered at the onset of the intervening NSR period to study the importance of continued KATP+ channel activation for the protection that was present 15 minutes after cessation of pacing.
All experiments were performed in accordance with the Guiding Principles in the Care and Use of Animals as approved by the Council of the American Physiological Society and under the regulations of the Animal Care Committee of the Erasmus University Rotterdam.
Studies were performed in a total of 59 pigs assigned to seven experimental groups (Fig 1⇓). In six groups, animals underwent a 60-minute TCO followed by 120 minutes of reperfusion. Twelve animals served as controls and underwent only a single 60-minute TCO (group 1). Twenty animals underwent a 60-minute TCO preceded by either 10 minutes (n=6, group 2) or 30 minutes (n=14, groups 3 and 4) of RVP at 200 beats per minute (bpm) followed by 15 minutes of NSR. Seven of the animals subjected to 30-minute RVP received 1 mg/kg IV glibenclamide at the onset of the 15 minutes of NSR period after the 30-minute RVP (group 4). In 20 animals, the 60-minute TCO was preceded by 30-minute RVP at 200 bpm without an intermittent period of NSR (groups 5 and 6); 8 of these animals were pretreated with glibenclamide (1 mg/kg IV) 10 minutes before the start of RVP (group 6). The dose of glibenclamide was chosen because it was previously shown to block preconditioning by a single 10-minute coronary occlusion in pigs without extending infarct size produced by 60-minute TCO in control animals.17 In these two groups of animals ventricular pacing was terminated immediately (<10 seconds) after the start of the 60-minute LADCA ligation. To evaluate whether RVP produced detectable myocardial ischemia, we determined wall function in the distribution area of the LADCA, high-energy phosphates, oxygen consumption, and regional myocardial blood flow in 7 animals throughout a 30-minute period of RVP at 200 bpm followed by 180 minutes of NSR (group 7).
Domestic Yorkshire-Landrace pigs (weight, 25 to 35 kg; HVC, Hedel, the Netherlands) were sedated with ketamine (20 mg/kg IM), anesthetized with pentobarbital (25 mg/kg IV), and instrumented for measurement of arterial and left ventricular pressure and control of arterial blood gases.18 After administration of pancuronium bromide (4 mg IV, Organon Teknika BV) and a midline thoracotomy, an electromagnetic flow probe (Skalar) was placed around the ascending aorta to measure cardiac output (Fig 2⇓). The LADCA was dissected free from the surrounding tissue to allow placement of a microvascular clamp (groups 1 through 6) and a Doppler flow probe (Crystal Biotech Inc) (groups 2 through 7). In the animals that underwent RVP, an electrode was attached to the anteriolateral left ventricular wall in the vicinity of the apex for stimulation of the myocardium by electrical monophasic stimuli with an amplitude of 2 mA and a frequency of 3.33 Hz. A small cannula was inserted into the vein accompanying the LADCA for the withdrawal of local venous blood for the determination of blood gases.
Regional Myocardial Function
In all seven groups, pairs of ultrasonic crystals (Sonotek Corp) were positioned into the midmyocardial layers of the left ventricle in the distribution areas of the LADCA and the LCXCA for the measurement of regional segment shortening by sonomicrometry17 18 (Triton Technology Inc) (Fig 2⇑). From the segment length tracings, segment length at the end of diastole (EDL, onset of positive ascending aortic flow) and the length at the end of systole (ESL, end of positive aortic flow) were determined and regional systolic segment shortening (SS) was computed as
and postsystolic segment shortening (PSS) was calculated as
Regional Myocardial Blood Flows
In the animals of group 7 (Fig 1⇑), we investigated the effects of RVP on the distribution of transmural myocardial blood flow. For this purpose, the left atrial appendage was cannulated for injection of 1 to 2 · 106 microspheres, 15±1 μm in diameter (NEN Co), labeled with either 95Nb, 103Ru, 113Sn, 46Sc, or 141Ce. Processing of myocardial tissue samples and computation of blood flow data have been described earlier.18
High-Energy Phosphate Metabolism
High-energy phosphates were measured in transmural myocardial biopsies, taken with a Tru-Cut needle (Travenol Laboratories Inc) from the area perfused by the LCXCA at baseline and immediately before the 60-minute TCO. This procedure allowed assessment of the effects of ventricular pacing on high-energy phosphate metabolism without interfering with the infarct size determination in the area perfused by the LADCA. Biopsies were immediately dipped into 0.9% NaCl at 0°C to remove adherant blood, frozen in liquid nitrogen (within 10 seconds) and stored until analysis at −80°C. Adenine nucleotides (ATP, ADP, AMP), creatine (Cr), and creatine phosphate (CrP) were measured by isocratic ion pairing high-perfomance liquid chromatography as previously described.18 From these measurements, CrP/Cr and CrP/ATP ratios were calculated to estimate changes in oxidative phosphorylation potential. Energy charge was calculated as (ATP+0.5 ADP)/(ATP+ADP+AMP).19
After completion of the instrumentation, 5000 IU of heparin was administered intravenously, and a stabilization period of at least 30 minutes was allowed before baseline data were obtained of systemic hemodynamic variables, coronary blood flow, and regional segment length changes. The animals were then subjected to one of the seven study groups (Fig 1⇑). In case of ventricular fibrillation, defibrillation was started within 10 seconds using DC countershocks (15 to 30 W). If defibrillation could not be accomplished within 2 minutes, animals were excluded from further study. Throughout the experimental protocol, body core temperature was rigorously controlled with a heating pad to maintain temperature within a narrow range (37°C to 38°C) to minimize temperature-induced infarct size variability.20 21
In the animals of groups 3 and 7, arterial and coronary venous blood samples for the determination of oxygen content and pH were withdrawn at baseline, at 10-minute, and 30-minute RVP, and at 2, 5, and 15 minutes of NSR. In the animals of group 7, measurements were also made at 60, 120, and 180 minutes of NSR. Myocardial biopsies for the measurement of high-energy phosphate levels in the LCXCA perfused area were obtained at baseline and at 30-minute RVP. The effects of RVP on the distribution of myocardial blood flow were determined in group 7 by injection of radioactive microspheres at baseline and at 30-minute RVP.
Area at Risk and Infarct Size
The methods to determine the AR and IA have been described16 17 and validated extensively.22 23 24 25 26 Briefly, after reocclusion of the LADCA, the area at risk was identified by an injection of 30 mL of a 5% (wt/wt) solution of fluorescein sodium (Sigma Chemical Co) into the left atrium. Ventricular fibrillation then was induced with a 9-V battery, and the heart was excised. Both atria, the right ventricular free wall, and the left ventricular epicardial fat were removed. The left ventricle was cooled in crushed ice, filled with alginate impression material (CAVEX Holland BV), and sliced parallel to the atrioventricular groove into five segments of approximately 1.5 cm in thickness. The cut surface(s) of each segment and the demarcated AR then were traced onto an acetate sheet under UV light. The viable myocardium then was stained dark purple by incubating the segments for 20 minutes in 0.125 g para-nitroblue tetrazolium (Sigma Chemical Co) per liter of phosphate buffer (pH 7.1) at 37°C. The nonstained pale infarcted tissue was traced onto the acetate sheet. The surface of each ring was subdivided into an endocardial (inner) half and an epicardial (outer) half by drawing a line that divided the myocardial wall into only two layers of equal thickness. Division into two layers was done because it provides information on the transmural distribution of infarct size yet preserves sufficient accuracy of infarct size determination in the two halves. Surface areas of the subendocardial and subepicardial halves and of the subendocardial and subepicardial AR and IA were determined and averaged for the apical and basal sides of each individual ring. The fractions of the ring that was infarcted and at risk then were multiplied by the weight of the ring to yield the weights of the AR and IA for that ring. Subsequently, the weights of the subendocardial and subepicardial halves and the total weight of each ring were summed to yield the LVendo, LVepi, and total LV mass. The weights of the endocardial, epicardial, and total areas at risk of each ring were summed to yield ARendo, ARepi, and total AR mass; the weights of the endocardial, epicardial, and total infarct areas of each ring were summed to yield IAendo, IAepi, and total IA mass. Endocardial, epicardial, and total AR and IA data were expressed as a percentage of LVendo, LVepi, and total LV mass, respectively.
Infarct size data have been presented by plotting the IA against AR for the endocardial and epicardial halves and for the whole left ventricular wall. Linear regression analysis was performed to determine the relation between endocardial and epicardial IA and AR in the control group. For all experimental groups, individual infarct size data points are presented. Intergroup differences between IAendo, IAepi, or total IA were analyzed by ANCOVA, with ARendo, ARepi, or total AR as the respective covariate. When a significant effect was observed, comparisons between individual groups were made with ANCOVA followed by a modified Bonferroni procedure to correct for multiple comparisons. Intragroup differences between IAendo and IAepi were analyzed with the use of ANCOVA for repeated measurements, with ARendo and ARepi as covariates. The effect of RVP on the incidence of ventricular fibrillation during 60-minute TCO was analyzed by Fisher’s exact test.
Hemodynamic and regional myocardial function data were analyzed by two-way ANOVA followed by either paired t test (intragroup) or unpaired t test (intergroup) with modified Bonferroni procedure to correct for multiple comparisons. A value of P<.05 was considered statistically significant (two-tailed). Data are presented as mean±SEM.
Ventricular Fibrillation and Mortality
In the control group, one animal was excluded because of unsuccessful defibrillation during the 60-minute TCO. All other animals that fibrillated were defibrillated successfully. Table 1⇓ shows that ventricular fibrillation occurred in 8 of the 13 control animals during the 60-minute TCO. The incidence of ventricular fibrillation in all groups that were subjected to RVP before the 60-minute TCO (groups 2 through 6) was not significantly different from the control group, indicating that in this model RVP did not protect against ventricular fibrillation during the subsequent coronary artery occlusion. Ventricular fibrillation was rare when reperfusion was reinstated, which is in agreement with previous observations that in pigs, ventricular fibrillation during reperfusion occurs predominantly after 10- and 30-minute coronary artery occlusions.27
Mean areas at risk (expressed as percentage of left ventricular mass) for the five experimental groups of animals that underwent the 60-minute TCO were not different from each other (34±2%, 33±2%, 36±2%, 33±2%, 32±2%, and 33±3% for groups 1, 2, 3, 4, 5, and 6, respectively; F=.40, P=.85).
In the 12 control animals, transmural infarct area was linearly related with the area at risk (r=.92, P<.01; Fig 3⇓). Separation of the left ventricular wall into two layers of equal thickness revealed a highly linear relation in both endocardial half (r=.92, P<.01) and epicardial half (r=.86, P<.01) of the left ventricle. Ten minutes of RVP, separated from the 60-minute TCO by a 15-minute period of NSR, failed to reduce transmural, epicardial, and endocardial infarct size compared with the control group (Fig 3⇓). When the period of RVP was extended to 30 minutes, infarct size in both the endocardial half (F=6.0, P<.05) and epicardial half (F=12.5, P<.01) was slightly but by ANCOVA significantly reduced (Fig 4⇓); the degree of protection was not significantly different between the two layers. The transmural IA/AR ratio was also significantly lower than that in the control group (71±2% versus 84±2%, P<.01).
The transmural IA in 10 of the 12 animals that underwent 30-minute RVP immediately followed by 60-minute TCO was located well below the regression line describing the relation between IA and AR in the control group (Fig 5⇓). The IA/AR of this group of animals was 63±4% (P<.01 versus control group). ANCOVA showed that 30-minute RVP immediately followed by 60-minute TCO significantly reduced the infarcted area for a given AR compared with the control group (F=21.7, P<.01). Further analysis indicated that the infarct size reduction was not significantly different between subepicardium and subendocardium (Fig 5⇓).
Mechanism of Protection by RVP
Pretreatment with glibenclamide prevented the protective effect of 30-minute RVP in both the endocardial and epicardial halves (Fig 5⇑). This is also illustrated by the IA/AR ratio, which was 78±4% (P=NS versus control group and P<.05 versus 30-minute RVP+60-minute TCO). In contrast, administration of glibenclamide at the onset of a 15-minute intervening period of sinus rhythm failed to inhibit the modest protection afforded by 30-minute RVP followed by 15 minutes of NSR (Fig 4⇑), suggesting that continued KATP+ channel activation is not mandatory for the moderate degree of protection that was still present 15 minutes after cessation of pacing.
Hemodynamic Responses to Rapid Ventricular Pacing and Total Coronary Artery Occlusion (Groups 1 through 6)
In the six groups that underwent 60-minute TCO (groups 1 through 6, n=52), there were no significant differences between baseline values of heart rate (110±2 bpm), mean aortic pressure (88±1 mm Hg), cardiac output (2.9±0.1 L/min), left ventricular dP/dtmax (1880±80 mm Hg/s), or left ventricular end-diastolic pressure (8±1 mm Hg). RVP in groups 2, 3, 4, and 5 (n=32) was associated with immediate decreases in mean arterial blood pressure (39±2%), cardiac output (44±2%), and stroke volume (70±1%), while systemic vascular resistance, left ventricular end-diastolic pressure, and LVdP/dtmax remained unchanged. This hemodynamic profile was maintained during the remainder of the ventricular pacing period (groups 3, 4, and 5, n=26) and was not different between the three groups. In the animals of groups 2 and 3 (n=13), in which pacing was terminated without an immediate occlusion of the LADCA, all variables returned to baseline within 1 minute of NSR except for heart rate, which remained slightly (≈15 bpm) elevated during the 15 minutes after cessation of ventricular pacing. In groups 1 through 5 (n=44), TCO resulted in decreases in cardiac output (16±3%) and mean aortic pressure (6±2%) and increments in heart rate (12±3%) and left ventricular end-diastolic pressure (38±6%) compared with baseline values (all P<.05); these responses were not different between the five groups. None of the hemodynamic variables recovered significantly toward baseline levels during 120 minutes of reperfusion.
Glibenclamide (group 6) produced modest increments in left ventricular end-diastolic pressure from 8±1 to 12±1 mm Hg (P<.05) and mean aortic pressure from 86±3 to 98±5 mm Hg (P<.05). The latter was due to systemic vasoconstriction as cardiac output was not altered by the KATP+ channel blocker (2.9±0.3 L/min at baseline and 3.0±0.3 L/min after administration of glibenclamide). Glibenclamide had no effect on LADCA blood flow (1.15±0.15 mL · min−1 · g−1 at baseline and 1.19±0.14 mL · min−1 · g−1 after administration of glibenclamide), LADCA vascular resistance (88±16 mm Hg · mL−1 ·min−1 · g−1 at baseline and 93±17 mm Hg · mL−1 ·min−1 · g−1 after glibenclamide) or systolic shortening in the LADCA perfused segment (16±2% at baseline and 15±2% after glibenclamide). Pretreatment with glibenclamide enhanced the pacing-induced decreases in mean aortic pressure (49±5%), cardiac output (61±3%), and coronary blood flow (36±6%) but had no effect on hemodynamic changes produced by 60-minute TCO.
Effect of Rapid Ventricular Pacing on Myocardial Performance (Groups 3 and 7)
RVP in groups 3 and 7 was associated with immediate decreases in mean arterial blood pressure and cardiac output and hence, myocardial work, whereas left ventricular dP/dtmax, systemic vascular resistance, and left ventricular end diastolic pressure were maintained (Table 2⇓). Thirty minutes of RVP decreased coronary blood flow by 19±5% (n=14), accompanied by a small increase in myocardial oxygen extraction from 72±2% to 77±2% (both P<.05, Table 3⇓). Oxygen consumption per gram of myocardium tended to decrease during RVP, but this failed to reach levels of statistical significance. Microsphere data revealed that the subendocardial to subepicardial blood flow ratio at 30-min RVP was maintained well above unity (1.07±0.08, n=6), although absolute levels were slightly lower than at baseline (1.23±0.07) (P<.05). Coronary vascular resistance (calculated as mean arterial pressure divided by coronary blood flow per gram of myocardium) was also maintained during pacing. Systolic shortening decreased markedly in both the LADCA and the LCXCA perfused segments. However, this was due to a marked decrease in end-diastolic length of both the LADCA (17±1%) and the LCXCA (15±2%) perfused segments, as end-systolic length of both LADCA (4±1%) and LCXCA (5±1%) segments decreased slightly (Table 3⇓). Furthermore, the decrease in systolic shortening during RVP was not accompanied by the appearance of postsystolic shortening. Throughout the pacing protocol, the arterial-coronary venous differences in pH (0.04±0.01 and 0.04±0.01 at baseline and 30-minute RVP, respectively) and in Pco2 (11.4±0.5 and 10.2±0.9 mm Hg at baseline and 30-minute RVP, respectively) were maintained. In further support of aerobic metabolism, we also did not observe decreases in ATP levels (36.3±1.4 μmol/g protein at baseline versus 36.5±1.4 μmol/g protein at 30-minute RVP), CrP/Cr ratio (1.24±0.12 versus 1.36±0.12), CrP/ATP ratio (1.52±0.26 versus 1.65±0.29), or energy charge (0.922±0.003 versus 0.924±0.003) at 30-minute RVP versus baseline, respectively.
Immediately after RVP was stopped, systemic hemodynamic variables recovered to baseline values except for heart rate, which remained slightly elevated after restoration to normal sinus rhythm. Mean aortic pressure increased to levels slightly higher than baseline during the first minute but had recovered to baseline levels at 15 minutes after cessation of RVP. During the first minute of postpacing, systolic shortening in both the LADCA and LCXCA perfused segments recovered to baseline values, although this was followed by a slight decrease in systolic shortening in the LCXCA area during the remainder of the protocol. Because reactive hyperemia was also absent, these findings indicate that 30-minute RVP was not associated with detectable myocardial ischemia.
The present study has yielded several important findings: (1) Infarct size after 60-minute coronary artery occlusion is limited when the occlusion is immediately preceded by a period of RVP. (2) Pretreatment was glibenclamide prevented the protective effect of RVP, suggesting the involvement of activation of KATP+ channels in producing the protection. (3) The pacing-induced activation KATP+ channels was not due to myocardial ischemia. (4) When the period of RVP was separated from the 60-minute TCO period by 15 minutes of NSR, the protective effect of RVP was markedly less, indicating that the duration of the pacing-induced protected state is much shorter than that for ischemic preconditioning. (5) However, the protection that was still present 15 minutes after cessation of pacing did not involve continued activation of KATP+ channels.
In the present study, we observed that 10 minutes of RVP followed by 15 minutes of NSR was ineffective in producing protection against myocardial infarction produced by a 60-minute coronary artery occlusion. In contrast, we previously reported that 10 minutes of coronary artery occlusion followed by 15 minutes of reperfusion produced considerable myocardial protection. Similarly, whereas a 5-minute coronary artery occlusion followed by 10 minutes of reperfusion in the rabbit heart is sufficient to precondition the myocardium,28 5 minutes of atrial pacing followed by 10 minutes of NSR failed to exert a protective effect on myocardial infarct size in rabbit hearts.10 Our finding that 30 minutes of ventricular pacing followed by 15 minutes of NSR did result in a moderate degree of protection suggests that the duration of stimulation by pacing needs to be longer than that required with a coronary artery occlusion. The duration of protection produced by RVP appears relatively short, since 15 minutes after cessation of pacing most of the protective effect had disappeared. In contrast, myocardial protection by ischemic preconditioning lasts up to at least 1 hour after a 10-minute TCO in pigs and can last up to 5 hours in individual animals.29 These findings suggest that RVP at 200 bpm represents a weaker stimulus than an equivalent period of coronary artery occlusion in pigs, requiring a longer period of stimulation and resulting in a shorter duration of the protected state.
The protection afforded by 30 minutes of RVP against irreversible myocardial damage produced by a sustained period of ischemia was prevented by pretreatment with glibenclamide, indicating that KATP+ channels are involved in the protective mechanism of RVP. Since we and others have shown that KATP+ channel blockade inhibits ischemic preconditioning in several species including rabbits,30 dogs,31 and swine,17 32 it could be hypothesized that ventricular pacing produced preconditioning via induction of myocardial ischemia. In support of this hypothesis, Szilvassy et al11 reported in rabbits that 10 minutes of RVP at 500 bpm increased left ventricular end-diastolic pressure from 4±1 to 33±3 mm Hg and produced myocardial ST-segment elevations suggestive for the occurrence of myocardial ischemia. Similarly, Vegh et al9 reported in open chest dogs that during 2 minutes of RVP in which heart rate was increased from 141±8 bpm at baseline to 300 bpm, left ventricular end-diastolic pressure increased from 9±2 to 23±3 mm Hg also accompanied by ST-segment elevations. In the present study, we did not observe an increase in left ventricular end-diastolic pressure when we paced the left ventricle at 200 bpm. The lower left ventricular filling pressure combined with the lower heart rate during pacing could explain why we did not observe signs of ischemia in metabolic, perfusion, and functional variables during RVP at 200 bpm: (1) transmural myocardial blood flow during RVP remained equally distributed across the inner and outer layers of the left ventricular wall, (2) the decrease in systolic shortening was entirely due to a decrease in end-diastolic length, (3) development of postsystolic shortening was not observed,33 and (4) no changes were observed in myocardial ATP and phosphocreatine levels, energy charge, and arterial or coronary venous pH levels.34 Furthermore, after restoration to NSR evidence for myocardial ischemia during the preceding period of RVP was also absent because (5) reactive hyperemia did not occur, (6) coronary venous oxygen tension was minimally affected after restoration to NSR, (7) systolic segment shortening recovered instantaneously to baseline levels, at which it was maintained throughout the subsequent 180-minute NSR period, and (8) there was no sustained postsystolic shortening during NSR, suggesting that post–ischemic myocardial stunning did not occur. Taken together, these findings fail to support the occurrence of significant myocardial ischemia in the present study. Although we cannot entirely exclude the occurrence of subtle subendocardial ischemia, this certainly would have been insufficient to produce ischemic preconditioning, as Ovize et al2 have shown that a 25-minute 50% flow reduction immediately preceding a 60-minute TCO (resulting in total loss of contractile function in the area perfused by the partially occluded coronary artery) failed to limit infarct size. In addition, we recently observed that 30-minute or 90-minute periods of 30% coronary blood flow reduction, associated with a 25% decrease in systolic segment shortening (due to an increase in end-systolic length), did not protect the myocardium against infarction produced by 60 minutes of TCO immediately after the 30% flow reduction.35 Therefore, if some subendocardial ischemia might have gone undetected, it is highly unlikely that this was responsible for the protective effect produced by RVP.
The exact mechanism of KATP+ channel activation by ventricular pacing cannot be determined from the present study. However, there is evidence that ventricular pacing is capable of activating transient outward K+ currents. Thus, Geller and Rosen14 observed that an increase in electrical activation rate of canine ventricular slabs from 90 to 130 pulses per minute shortened the action potential. The action potential shortening persisted for several minutes after the activation rate was lowered to 90 pulses per minute, indicative of myocardial “memory” for the activation stimulus, which could be antagonized by blockade of the transient outward K+ current. Although the specific role of KATP+ channels was not investigated in that study, our findings that 15 minutes after cessation of 30 minutes of RVP a small but statistically significant reduction in infarct size occurred could be interpreted to suggest that KATP+ channel activation by ventricular pacing might also display memory. This is supported by observations in dogs that ischemic preconditioning can induce memory for myocardial protection, which involves KATP+ channels. Thus, when glibenclamide (0.3 mg/kg) was administered intravenously in dogs immediately after a 5-minute occlusion of the LCXCA 10 minutes before a 60-minute circumflex occlusion, ischemic preconditioning was abolished,31 indicating that continued KATP+ channel activation is mandatory for the protective effect to occur even though the initial stimulus that provided the protection is no longer present. However, when we administered glibenclamide at the onset of 15-minute NSR immediately after the pacing period, the moderate degree of myocardial protection afforded by 30 minutes of RVP followed by 15 minutes of sinus rhythm was not attenuated. Thus, while the induction of myocardial protection by pacing does involve KATP+ channel activation, continued KATP+ channel activation did not appear to be mandatory for myocardial protection that was present at 15 minutes after cessation of pacing.
In the present study, infarct size produced by a 60-minute coronary artery occlusion was determined in myocardium reperfused for 2 hours using para-nitroblue tetrazolium sodium. In viable myocytes, para-nitroblue tetrazolium is reduced to form a dark purple diformazan precipitate by intracellular diaphorases that use NADH or NADPH as electron donors.23 False-positive staining is minimized by allowing myocardial reperfusion to facilitate washout of NADH and NADPH from necrotic myocardium. Schaper et al24 reported that only 30 minutes of reperfusion is sufficient for accurate detection of infarct size. Horneffer et al26 also evaluated para-nitroblue tetrazolium staining for the determination of myocardial infarct size in pigs and reported that infarct size produced by 15, 30, or 90 minutes of coronary artery occlusion was not different when either 2 or 48 hours of reperfusion was allowed. Fujiwara et al25 showed that determination of infarct size produced by a 60-minute TCO in porcine myocardium reperfused for 1 hour was identical to infarct size determined after 3 and 7 hours of reperfusion. In the latter study, histochemical analysis of infarct size correlated well with histological measurements. The available evidence clearly indicates that 2 hours of reperfusion after a 60-minute coronary artery occlusion allows accurate histochemical determination of infarct size with para-nitroblue tetrazolium staining in pigs.
In the present study, we used glibenclamide to selectively block KATP+ channels. Glibenclamide was administered intravenously in a dose of 1 mg/kg, which we have previously shown to block ischemic preconditioning in pigs without extending infarct size produced by a 60-minute coronary artery occlusion.17 Glibenclamide is known to stimulate the release of insulin and consequently decrease blood glucose levels. In view of a preliminary study in the isolated rabbit heart that reported that substitution for glucose by pyruvate in the perfusate during reperfusion abolished myocardial preconditioning,36 the possibility that glibenclamide modulated the protective effect of ventricular pacing by decreasing blood glucose levels deserves consideration. Gross and Auchampach31 reported a 16% decrease in blood glucose levels in response to administration of 0.3 mg/kg glibenclamide in dogs. Similarly, in rabbits, glibenclamide (0.15 to 3.0 mg/kg) produced a significant fall (23% to 38%) in blood glucose levels.37 In contrast, we recently observed that glibenclamide in a dose of 1 mg/kg had a minimal (<10%) effect on blood glucose in pigs.17 The reason for the minimal effect of glibenclamide on glucose in pigs is not clear but may be due to the relatively low baseline glucose levels in the overnight-fasted pigs17 or a species difference in sensitivity of the pancreatic cells for glibenclamide. Importantly, the novel cardioselective KATP+ channel blocker 5-hydroxydecanoate, which is devoid of blood glucose–lowering properties, was as effective as glibenclamide in blocking ischemic preconditioning in dogs,38 indicating that the inhibitory effect of these compounds was due to blockade of cardiac KATP+ channels and was not due to the decrease in blood glucose.31
Thirty minutes of RVP decreased myocardial infarct size produced by a 60-minute coronary artery occlusion in open chest pigs. This protective effect was prevented by pretreatment with KATP+ channel blockade, indicating that KATP+ channel activation is involved in the mechanism of protection. Since we failed to observe significant myocardial ischemia during RVP, it appears that KATP+ channel activation was produced via a nonischemic mechanism. The duration of the protected state produced by 30 minutes of RVP appears relatively brief, since much of the protection was lost when 15 minutes of NSR separated the 30-minute RVP period from the sustained occlusion. When KATP+ channel blockade was produced immediately after ventricular pacing at the onset of the intervening 15-minute period of NSR, the residual protection remained unchanged. Thus, the moderate degree of protection that was still present 15 minutes after cessation of RVP did not require continued activation of KATP+ channels.
Selected Abbreviations and Acronyms
|AR||=||area(s) at risk|
|LADCA||=||left anterior descending coronary artery|
|LCXCA||=||left circumflex coronary artery|
|NSR||=||normal sinus rhythm|
|RVP||=||rapid left ventricular pacing|
|TCO||=||total coronary artery occlusion|
This work was supported by grant 92.144 from the Netherlands Heart Foundation and by grant CIPA-CT-92-4009 of the European Economic Community. The research of Dr Duncker was made possible by a Fellowship of the Royal Netherlands Academy of Arts and Sciences. The authors gratefully acknowledge the expert technical assistance of R.H. van Bremen and S. Nieukoop.
- Received February 9, 1995.
- Revision received August 3, 1995.
- Accepted August 16, 1995.
- Copyright © 1996 by American Heart Association
Lawson CS, Downey JM. Preconditioning: state of the art myocardial protection. Cardiovasc Res. 1993;27:542-550.
Ovize M, Przyklenk K, Kloner RA. Partial coronary stenosis is sufficient and complete reperfusion is mandatory for preconditioning the canine heart. Circ Res. 1992;71:1165-1173.
Koning MMG, Simonis LAJ, De Zeeuw S, Nieukoop S, Post S, Verdouw PD. Ischaemic preconditioning by partial occlusion without intermittent reperfusion. Cardiovasc Res. 1994;28:1146-1151.
Iwamoto T, Bai XJ, Downey HF. Preconditioning with supply-demand imbalance limits infarct size in dog heart. Cardiovasc Res. 1993;27:2071-2076.
Przyklenk K, Bauer B, Ovize M, Kloner RA, Whittaker P. Regional ischemic ‘preconditioning’ protects remote virgin myocardium from subsequent sustained coronary occlusion. Circulation. 1993;87:893-899.
McClanahan TB, Nao BS, Wolke LJ, Martin BJ, Metz TE, Gallagher KP. Brief renal occlusion and reperfusion reduces myocardial size in rabbits. FASEB J. 1993;7:682. Abstract.
Gho BCG, Schoemaker RG, Van der Lee C, Sharma HS, Verdouw PD. Myocardial infarct size limitation in rat by transient renal ischemia. Circulation. 1944;90(suppl I):I-476. Abstract.
Ovize M, Kloner RA, Przyklenk K. Stretch preconditions canine myocardium. Am J Physiol. 1994;266(Heart Circ Physiol 35):H137-H146.
Vegh A, Szekeres L, Parratt JR. Transient ischaemia induced by rapid cardiac pacing results in myocardial preconditioning. Cardiovasc Res. 1991;25:1051-1053.
Marber MS, Walker DM, Eveson DJ, Walker JM, Yellon DM. A single five-minute period of rapid atrial pacing fails to limit infarct size in the in situ rabbit heart. Cardiovasc Res. 1993;27:597-601.
Szilvassy Z, Ferdinandy P, Bor P, Jakab I, Lonovics J, Koltai M. Ventricular overdrive pacing-induced anti-ischemic effect: a conscious rabbit model of preconditioning. Am J Physiol. 1994;266(Heart Circ Physiol 35):H2033-H2041.
del Balzo U, Rosen MR. T-wave changes persisting after ventricular pacing in canine heart are altered by 4-aminopyridine but not by lidocaine: implications with respect to phenomenon of cardiac ‘memory.’ Circulation. 1992;85:1464-1472.
Geller JC, Rosen MR. Persistent T-wave changes after alteration of the ventricular activation sequence: new insights into mechanisms of ‘cardiac memory.’ Circulation. 1993;88:1811-1819.
Van Winkle DM, Chien GL, Wolff RA, Soifer BE, Kuzume K, Davis RF. Cardioprotection provided by adenosine receptor activation is abolished by blockade of the KATP channel. Am J Physiol. 1994;266(Heart Circ Physiol 35):H829-H839.
Rohmann S, Weygandt H, Schelling P, Soei LK, Becker KH, Verdouw PD, Lues I, Hausler G. Effect of bimakalim (EMD 52692), an opener of ATP-sensitive potassium channels, on infarct size, coronary blood flow, regional wall function, and oxygen consumption in swine. Cardiovasc Res. 1994;28:858-863.
Sassen LMA. Bezstarorti K, Van der Giessen WJ, Lamers JMJ, Verdouw PD. L-propionylcarnitine increases post-ischemic blood flow but does not affect recovery of the energy charge. Am J Physiol. 1991;261(Heart Circ Physiol 30):H172-H180.
Atkinson DE, Walton GM. Adenosine triphosphate conservation in metabolic regulation. J Biol Chem. 1967;242;3239-3241.
Chien GL, Wolff RA, Davis RF, Van Winkle DM. ‘Normothermic range’ temperature affects myocardial infarct size. Cardiovasc Res. 1994;28:1014-1017.
Duncker DJ, Herrlinger S, Pavek T, Crampton M, Lindstrom P, Bache RJ. Effects of temperature on myocardial infarct size in swine. FASEB J. 1994;8:A830. Abstract.
Horneffer PJ, Healy B, Gott VL, Gardner TJ. The rapid evolution of a myocardial infarction in an end-artery coronary preparation. Circulation. 1987;76(suppl V):V39-V42.
Verdouw PD, Hartog JM. Provocation and suppression of ventricular arrhythmias in domestic swine. In: Stanton HC, Mersmann HJ, eds. Swine in Cardiovascular Research. Boca Raton, Fla: CRC Press Inc; 1986:121-156.
Cohen MV, Liu GS, Downey JM. Preconditioning causes improved wall motion as well as smaller infarcts after transient coronary occlusion in rabbits. Circulation. 1991;84:341-349.
Koning MMG, De Zeeuw S, Nieukoop S, De Jong JW, Verdouw PD. Is myocardial infarct size limitation by ischemic preconditioning an ‘all or nothing’ phenomenon? In: Das DK, ed. Cellular, Biochemical and Molecular Aspects of Reperfusion Injury. Ann N Y Acad Sci. 1994;723:336-336.
Toombs CF, Moore TL, Shebuski RJ. Limitation of infarct size in the rabbit by ischaemic preconditioning is reversible with glibenclamide. Cardiovasc Res. 1993;27:617-622.
Gross GJ, Auchampach JA. Blockade of ATP-sensitive potassium channels prevents myocardial preconditioning. Circ Res. 1992;70:223-233.
Schultz R, Rose J, Heusch G. Involvement of ATP-dependent potassium channels in ischemic preconditioning in swine. Am J Physiol. 1994;267(Heart Circ Physiol 36):H1341-H1352.
Morris JJ, Pellom GL, Murphy CE, Salter DR, Goldstein JP, Wechsler AS. Quantification of the contractile response to injury: assessment of the work-length relationship in the intact heart. Circulation. 1987;76:717-727.
Schamhardt HC, Verdouw PD, Saxena PR. Effects of oxyfedrine on global haemodynamics and myocardial lactate balance in acute ischaemia of the porcine heart preparation. Arzneimittelforschung/Drug Res. 1981;31:1091-1095.
Koning MMG, Gho BCG, Opstal RLJ, van Klaarwater E, Duncker DJ, Verdouw PD. Novel approaches to myocardial preconditioning in pigs. In: Parratt JR, Wainwright CL, eds. Myocardial Preconditioning. Austin, Tex: Landes RG Co. In press.
Omar B, Hanson A, Bose S, McCord J. Reperfusion with pyruvate eliminates ischemic preconditioning: an apparent role for enhanced glycolysis. FASEB J. 1991;5:A1257. Abstract.
Thornton J, Thornton CS, Sterling DL, Downey JM. Blockade of ATP-sensitive potassium channels increases infarct size but does not prevent preconditioning in rabbit hearts. Circ Res. 1993;72:44-49.
Auchampach JA, Grover GJ, Gross GJ. Blockade of ischemic preconditioning in dogs by the novel ATP dependent potassium channel antagonist sodium 5-hydroxydecanoate. Cardiovasc Res. 1994;26:1054-1062.