Presence of a Critical Coronary Artery Stenosis Does Not Abolish the Protective Effect of Ischemic Preconditioning
Background Episodic, severe coronary artery flow restriction preconditions the myocardium much like brief occlusions. The necessity for full reperfusion after a preconditioning intervention to elicit the preconditioning response is unclear. This study investigated in closed-chest swine the effect of a persistent critical coronary stenosis with moderate flow reduction on ischemic preconditioning.
Methods and Results Farm pigs (n=23) assigned to one of four groups—(1) control, (2) stenosis, (3) preconditioned (PC), or (4) preconditioned plus stenosis (PC/S)—underwent percutaneous instrumentation with a percutaneous transluminal coronary angioplasty catheter advanced to the mid–left anterior descending coronary artery. An artificial coronary stenosis (82% diameter reduction) was mounted on the catheter just proximal to the balloon in the two stenosis groups. Preconditioning stimulus consisted of two 10-minute balloon occlusions followed by 15 minutes of reperfusion. All groups subsequently underwent 45 minutes of occlusion followed by 120 minutes of reperfusion. Baseline regional myocardial blood flow in the area at risk (AAR), measured with colored microspheres, was lowest in the stenosis groups, with flow expressed as a percentage of normal zone flow. Infarct size (percent of AAR), determined by staining slices of the heart with triphenyltetrazolium, was significantly reduced in PC compared with control pigs (15.1±5.9% versus 66.8±6.4%, respectively; P<.001). Infarct size in PC/S pigs was also significantly reduced (29.7±7.1%, P=.004 versus control) but was not different in degree from PC pigs (P=.6). The stenosis by itself conferred no preconditioning benefit (percent of AAR=69.0±5.4%).
Conclusions A moderate flow-limiting stenosis did not prevent preconditioning but may have attenuated the effect. This may be analogous to the clinical scenario in which intermittent coronary occlusion and reperfusion superimposed on a critical stenosis precede a prolonged occlusion treated with thrombolysis.
One or more brief episodes of coronary occlusion have been shown to limit myocardial infarct size after a prolonged occlusion. Murry et al1 called this phenomenon “preconditioning.” Initially demonstrated in dogs, preconditioning has been seen in other animal species, including pigs.2 Ischemic preconditioning may also be an important adaptive response in humans in the setting of percutaneous transluminal coronary angioplasty with repeated inflations,3 4 myocardial protection during coronary artery bypass graft surgery,5 and preinfarction angina.6 7
Recent studies suggest moderately severe coronary blood flow reduction at rest (<50% baseline) may be sufficient to elicit the preconditioning response.8 9 In the initial report of this finding, full reperfusion after the preconditioning flow restriction was necessary for preconditioning to occur in dogs8 but was not required in two recent swine studies.9 10 A common clinical condition relevant to this phenomenon is the presence of a persistent coronary stenosis. Two aspects of a persistent, flow-limiting stenosis have not been studied experimentally: the effect of the stenosis as a preconditioning agent and the effect of the continuous, flow-limiting stenosis on a demonstrated preconditioning intervention. This condition is analogous to the clinical situation in which brief episodes of coronary occlusion in the setting of a preexisting coronary lesion precede a myocardial infarction. In such a setting, reperfusion after each preconditioning insult may be restricted by a fixed stenosis. Furthermore, once the infarct-evoking occlusion reperfuses, either spontaneously or aided by an intervention, a stenosis may persist that continues to limit reperfusion. Accordingly, this clinical analogue was studied in closed-chest swine with an artificial intraluminal coronary stenosis that reduced the coronary artery diameter by 82%. Flow restriction continued in the reperfusion phases once a stenosis was placed, a condition similar to the clinical scenario and unlike previous studies.
Farm-bred domestic swine (weight, 34 to 71 kg) were premedicated with xylazine (1 mg/kg IM), ketamine hydrochloride (6 mg/kg IM), and thiopental sodium (0.5 to 1.0 g IV) after an overnight fast, intubated, and anesthetized with nitrous oxide (60:40 mixture with O2) and isoflurane (1% to 1.5%). The pigs were mechanically ventilated with a respirator on room air supplemented with 2 to 3 L/min O2. Frequent arterial blood gases were measured to guide adjustment of ventilator settings. After induction of anesthesia, each pig was anticoagulated with heparin (10 000-U bolus IV) and aspirin 325 mg dissolved in saline and passed through a 0.22-μm filter for intravenous injection. Full anticoagulation was maintained by administration of heparin 3000 U before balloon catheter placement and 5000 U every hour thereafter.
Pigs were instrumented for the study as follows. The femoral arteries were cannulated with 8F sheaths and a femoral vein with a multilumen catheter. The vein was used to administer fluids and medications during the study. The right carotid artery and right jugular vein were isolated by cutdown. An 8F pigtail catheter was advanced by fluoroscopic guidance from the right femoral artery to the left ventricle and then retrograde across the mitral valve into the left atrium. This catheter was used to administer colored microspheres for measurement of regional myocardial blood flow and to inject Evan's blue dye for myocardial staining of the AAR. Next, a 7F end-hole catheter was inserted into the left femoral artery and advanced to the arch of the aorta. This line was used to monitor arterial pressure and obtain blood for pH, Po2, Pco2, and reference sample for determination of regional myocardial blood flow. An 8F right Judkins catheter was inserted into the coronary sinus through the right jugular vein and positioned in the great cardiac vein. A 3F Teflon catheter was inserted over a guide wire into the Judkins catheter and advanced into the AIV. The Judkins catheter and guide wire were then removed.
An 8F Amplatz catheter was inserted into the right carotid artery and advanced under fluoroscopic guidance to the ostium of the LAD. Catheter position was verified with contrast injection. A guide wire was passed through the catheter into the distal LAD. The 8F Amplatz catheter was removed over the wire, and a 3.6F angioplasty catheter (20-mm-long balloon, 2.5- to 3.5-mm inflated diameter) was passed over the wire into the mid-LAD. A similar technique was used for placement of the artificial coronary artery stenosis (see below) that was mounted immediately proximal to the balloon on the distal end of an angioplasty catheter.
After instrumentation was complete, inhaled anesthetics were discontinued, and the pig was allowed to awaken sufficiently to breathe spontaneously and exhibit modest tremulousness. A constant intravenous infusion of a 2.5% solution of thiopental sodium was begun at 5 to 10 mL/h to maintain sedation and eliminate pain. Once the pig had stabilized, the experimental protocol was initiated.
Fig 1⇓ depicts the experimental protocol. The pigs were assigned to one of four groups: control, stenosis, PC, or PC/S. All pigs were allowed to equilibrate for 60 minutes after the initial instrumentation. During this period, stability was documented by measuring arterial and AIV blood gases and recording hemodynamic variables (ie, heart rate, aortic blood pressure, left atrial pressure, and coronary artery pressure). The treatment period occurred during the subsequent 50 to 60 minutes.
The control group received no intervention during the treatment period, which was followed by 45 minutes of total coronary artery occlusion with the angioplasty balloon inflated followed by balloon deflation and 2 hours of full reperfusion.
This group had a radiolucent plastic stenosis in the shape of a truncated cone that measured 5 mm long and had an outer diameter tapering from 3.25 to 3.0 mm that was mounted on an angioplasty catheter as described earlier and inserted into the mid-LAD. Two lumens were present within the cone. The central lumen accommodated the angioplasty catheter. Another lumen, placed off-center, resulted in an 82% diameter reduction where the stenosis came to rest (cone outer diameter=vessel internal diameter).11 12 The stenosis remained in position for the entire experiment. There was no additional intervention during the treatment period, which was followed by 45 minutes of total coronary occlusion with the angioplasty balloon inflated. Next, the balloon was deflated, followed by reperfusion with the stenosis in place.
During the treatment period, two 10-minute occlusions, each followed by 15 minutes of full reperfusion, were performed before the final 45-minute occlusion and 2 hours of full reperfusion.
This group was similar to the PC group except that the artificial stenosis was inserted as an initial intervention and remained in position for the entire experiment. It should be noted that only the control and PC groups had potential for full reperfusion.
At the conclusion of the study, the LAD was reoccluded by balloon inflation and 50 mL of a 5% Evan's blue solution was injected into the left atrium through the 8F pigtail catheter to delineate the AAR. The ischemic myocardium (AAR) remained unstained and the nonischemic myocardium appeared blue. The pigs were euthanatized under deep sedation by a left atrial injection of 20 mL concentrated potassium chloride solution. A midline thoracotomy was performed, and the heart was excised for further analysis. Before excision, the location of the distal end of the AIV catheter was noted with respect to the LAD catheter position (to ensure maximal sampling selectivity), and a tie was placed around the LAD at the point of balloon location. All catheters were then removed, and the tie around the LAD was secured to prevent leaching of the blue dye into the AAR.
Heart rate and aortic, left atrial, and distal coronary artery pressures were measured at baseline (60 minutes after instrumentation), preocclusion (just before the long occlusion), occlusion (45 minutes into the long occlusion), reperfusion (5 minutes into the final reperfusion), and the conclusion of the study (“final”). Intracardiac and intravascular pressures were recorded from fluid-filled catheters by standard pressure transducers, and the analog signal was digitized by and displayed on a Macintosh II Ci computer using the Acknowledge 3.0 Waveform Data Analysis program by BioPAC Systems, Inc. Thirty seconds of data for each hemodynamic variable was stored for each sample period and averaged (forward plus reverse) for numerical analysis.
Paired samples of arterial and anterior interventricular venous blood were obtained for determination of both oxygen content (Lex-O2-CON Instrument, Lexington Instruments) and lactate concentration (Calbiochem Rapid Lactate Reagents, Calbiochem-Behring) by use of previously described methods.12 These measurements were made at baseline, preocclusion, occlusion, reperfusion, and final. Regional oxygen consumption (milliliter per minute per 100 g) was calculated as the product of transmural regional myocardial blood flow (see below) and the arterial and venous oxygen difference. Regional lactate consumption was calculated as the product of transmural regional myocardial blood flow and the arterial and anterior interventricular venous lactate difference.
Regional Myocardial Blood Flow
Blood flow measurements were made by use of multiple colored microspheres as described by Hale et al.13 In all pigs, blood flow was measured at baseline, occlusion, reperfusion, and final. Transmural flow in the AAR is used in the analysis to allow inclusion of septal myocardial flow. Both absolute AAR flow (milliliters per minute per gram) and the ratio of flow in AAR to remote, normal zone flow (AAR/NL) are shown.
AAR and Infarct Size
All hearts were cut from the apex to the tip of the balloon catheter into four or five transverse slices, each 1.0 to 1.5 cm thick. The heart slices were weighed after right ventricular myocardium was removed, and the AAR–normal myocardial border was marked with an incision. Each slice was incubated for 15 minutes in a 1% solution of triphenyltetrazolium chloride at 37° centigrade. This method distinguishes necrotic from viable myocardium.14 Next, the slices were bathed in normal saline for a minimum of 2 days. Subsequently, they were imaged with a JE 3662-RGB video camera (Javelin Electronics) linked to a Perceptics Pixel-HR-24 video capture device with a 24-bit color display, digitizer board and window, and capture software. Data were collected on a Macintosh II Fx computer. Image processing was performed on a Power Macintosh 6100/60 with Adobe Photoshop 2.5 and NIH Image 1.52 computer software programs. The AAR and AN were expressed as a percentage of the left ventricular area as determined by planimetry. Also, AN was expressed as a percentage of the AAR.
Comparison of hemodynamics, myocardial blood flow, and metabolic variables among groups was performed by two-way ANOVA. Single time measurements, eg, infarct size (AN/AAR) and AAR/left ventricle, were compared by use of one-way ANOVA. A value of P<.05 was used to assess the statistical significance of differences between groups. If significant F ratios were obtained, post hoc pairwise comparisons were then carried out to identify the group(s) to which this difference could be attributed. The Bonferroni-Dunn method was used to adjust for multiple comparisons. A value of P<.0083 was the threshold for significance. Data for significantly different groups were then compared at each time interval by use of the t test. All measurements are expressed as group mean±SEM.
Exclusions and Mortality
Fifty-three pigs were initially assigned to the control (n=10), stenosis (n=9), PC (n=18), or PC/S (n=16) group. Five control pigs (50%) were excluded: 2 for refractory ventricular fibrillation (20%), 2 for technical problems (angioplasty balloon rupture and inability to insert AIV and coronary catheters), and 1 for coronary artery spasm. This left 5 pigs in the control group. Four stenosis pigs (44%) were excluded: 2 for refractory ventricular fibrillation (22%), and 2 for technical problems (indeterminate risk area secondary to poor myocardial staining and inability to insert the AIV catheter). Therefore, 5 pigs were left in the stenosis group. Eleven PC pigs (44%) were excluded: 5 for refractory ventricular fibrillation (28%); 2 for technical problems (inability to place the LAD catheter); 1 for improper placement of the LAD catheter in a large diagonal branch, unrecognized until autopsy, leading to a lateral wall infarction; and 3 owing to transmural blood flow in the AAR >0.25 mL·g−1·min−1 during the long occlusion. Seven pigs remained in the PC group. Ten PC/S pigs (63%) were excluded: 5 for refractory ventricular fibrillation (31%), 1 for excessive supraventricular tachycardia, 3 for clotting of the artificial stenosis evidenced by very low distal coronary artery pressures, and 1 for erratic flow with unstable distal pressure during the final reperfusion consistent with spasm distal to or intermittent thrombotic occlusion of the artificial stenosis. Consequently, 6 pigs formed the PC/S group.
Table 1⇓ summarizes the hemodynamic data. There was no significant difference between groups for heart rate, MAP or mean left atrial pressure. Stenosis group MCAPd was lower than those in the control and PC groups at all time periods except for the occlusion period. MCAPd was lower in the PC/S than in the PC group at baseline and reperfusion but not at preocclusion (P=.053) or final (P=.10). There were no significant differences in any group during occlusion. The mean pressure gradient (MAP–MCAPd), excluding occlusion, ranged from 25 to 42 mm Hg in the stenosis and PC/S groups. These values differed significantly (P<.005 for all except P<.05 for the final time period) at all time periods when groups with and without the stenosis were compared.
Data on MV˙o2 and lactate consumption are summarized in Tables 2⇓ and 3,⇓ respectively. There were no statistical differences in MV˙o2 between the groups (P=.085) although there was a trend toward a higher MV˙o2 in the PC group during the final measurement that exceeded that in all groups.
Lactate consumption was not statistically different between groups at each time period, including baseline. All groups showed net lactate consumption at baseline and net lactate production during early reperfusion. However, net lactate consumption returned to baseline in the PC and PC/S groups at the final time period in contrast to the control and stenosis groups.
Regional Myocardial Blood Flow
Table 4⇓ shows regional myocardial blood flow in the AAR at four time periods both as absolute flow and flow normalized to the normal zone (AAR/NL). Baseline flow was lowest in the groups with a stenosis, but only the difference between PC and stenosis was significant. Flow in PC/S pigs was exceeded by PC pigs at occlusion, during early reperfusion, and at final reperfusion. Reperfusion hyperemia was noted in only PC pigs (twice baseline). Flow in PC/S pigs was blunted (76% of baseline) during the same time period. Flow differences persisted until the end of the study and were higher in PC pigs compared with all groups at early reperfusion and final. Flow in PC/S pigs did not differ from that in either control or stenosis pigs at any time period.
AAR and Infarct Size
The AAR (as a percentage of the left ventricle) did not differ significantly between the groups: 38±4.5% in control, 32.2±2.4% in stenosis, 30.3±3.0% in PC/S, and 25.9±3.0% in PC (Fig 2⇓). Infarct size (AN/AAR) did not differ between control (66.8±5.4%) and stenosis (69.0±5.4%) pigs. Infarct size (Fig 2⇓) was significantly reduced in the PC and PC/S groups compared with the control and stenosis groups (PC versus control, P<.001; PC/S versus control, P=.004; PC versus stenosis, P<.001; PC/S versus stenosis, P=.002). No other pairwise differences were found.
This study indicates that a critical coronary artery stenosis may limit but does not abolish the benefits of ischemic preconditioning. Furthermore, it suggests that a critical stenosis may not be sufficient in itself to confer the preconditioning effect. In the usual ischemic preconditioning experiment, no flow restriction was imposed after either the brief preconditioning occlusion(s) or the longer infarct-inducing occlusion. This scenario is rarely encountered in the clinical situation of unstable coronary artery disease in which a significant stenosis is often the underlying and persistent culprit lesion even when reperfusion of a significant degree is restored. The impact of flow limitation during reperfusion was studied by Ovize et al.8 When severe flow restriction (<50% baseline flow, 15-minute duration) but not occlusion was used as the preconditioning stimulus, the preconditioning effect was eliminated if full reperfusion did not follow the brief period of flow restriction. In that study, full reperfusion also followed the infarct-inducing occlusion. In contrast, Koning et al,9 using a more prolonged period of severe flow limitation (<30% baseline flow, 30-minute duration) without an intervening period of flow restoration, found significant preconditioning benefit. Schulz et al10 also demonstrated preconditioning without an intervening reperfusion period. Using a swine model with a prolonged (90 minute) 70% reduction in coronary blood flow, they found that 10 minutes of “no-flow” ischemia immediately preceding the severe ischemic period resulted in ≈50% reduction in infarct size compared with the group without antecedent no-flow ischemia. The role of an intervening reperfusion period in the setting of a preconditioning intervention thus far has been a matter of debate. The weight of current evidence, however, suggests that full reperfusion may be less critical than previously believed in terms of a preconditioning benefit.
Although there was evidence for preconditioning in the PC/S group, there was no blood flow advantage during reperfusion compared with the control group. This raises some doubt that hyperemic reperfusion blood flow is necessary for preconditioning. Complete reperfusion may be a double-edged sword. Abrupt, full reperfusion may hasten myocardial salvage by rapid washout of accumulated toxic metabolites. However, oxygen free radicals formed during ischemia and reperfusion have the potential to cause microvascular and myocardial injury through membrane damage by lipid peroxidation.15 Partial or graded reperfusion theoretically may be protective in this regard. Although reperfusion was incomplete between ischemic episodes produced by brief spontaneous thrombotic episodes in the cyclic flow variations protocol by Ovize et al,16 preconditioning was clearly demonstrated. Full reperfusion, however, was established after the sustained occlusion, which is different from the present study.
The effect of various degrees of flow restriction both as a preconditioning stimulus and as a condition superimposed on a known preconditioning intervention needs to be fully defined. Ovize et al8 and Koning et al9 demonstrated that >50% flow reduction could serve as a preconditioning stimulus although the groups differed on the importance of reperfusion in the immediate preocclusion period. Previous studies had used only total coronary occlusion to study preconditioning. In the present study, an 82% coronary artery stenosis present before, during, and after a 45-minute occlusion conferred no preconditioning benefit. The flow limitation imposed by such a stenosis (limiting baseline AAR transmural flow to 55% to 65% of normal zone flow and 65% to 97% of control AAR flow) is significantly less than that imposed in either of the two previous studies. Furthermore, the stenosis limited reperfusion at every stage of the protocol after an occlusion was relieved. Given the relatively mild degree of flow restriction, the absence of preconditioning resulting from the stenosis alone is not surprising. Although net chemical lactate balance indicated extraction in all groups before the occlusion, this measurement may underestimate actual lactate release.17 Underlying ischemia with increased lactate release may be seen with levels of net chemical lactate consumption as low as those reported in this study at baseline.17 A similar protocol with the addition of stress ischemia, eg, pacing or dobutamine, may provide further insight into the possible significance of such a stenosis as a clinical preconditioning variable.12 18 One important intermediate condition was defined in this study: A critical stenosis (82%) coupled with brief occlusions did not prevent preconditioning, although the benefit may have been attenuated.
Differences in MV˙o2 were observed among the groups. Final MV˙o2 in the PC/S group was not greater than that in either the control or stenosis group despite greater myocardial salvage. This is in contrast to the observed increase in final MV˙o2 in the PC group. Apparently, myocardial salvage alone did not result in a higher final MV˙o2. Whether the greater MV˙o2 seen in the PC group reflects myocardial salvage and/or the phenomenon of stunning with paradoxically increased oxygen consumption19 cannot be determined because regional wall motion was not analyzed in this study. Although not specifically assessed in this study, MV˙o2 measured just before the infarct-producing ischemia may be a more important factor than measurements made during the ischemic or reperfusion period in terms of determining infarct size. In a swine infarct study using gradual versus abrupt severe blood flow reduction, Ito20 found that gradual ischemia (without intermittent reperfusion) resulted in a marked reduction of infarct size, similar in magnitude to classic preconditioning models. In contrast to the abrupt ischemia group, less mismatch between MV˙o2 and oxygen delivery was seen in the gradual ischemia group at the onset of severe ischemia. Thus, metabolic downregulation of the myocardium, through either graded ischemia as demonstrated by Ito20 or traditional preconditioning with brief episodes of ischemia and reperfusion, may play a crucial role in governing final infarct size.
The return of lactate metabolism toward normal compared with baseline values in the two preconditioned groups at the end of reperfusion is further evidence suggesting a protective effect. The PC/S group closely resembled the PC group in this regard, further suggesting that the stenosis did not significantly limit preconditioning.
Pigs and dogs differ considerably in their responses to coronary occlusion and flow restriction.2 8 9 10 Differences in collateralization are well recognized but may not be critical because several species with limited collaterals exhibit preconditioning. In the preconditioning literature, the AN as a percentage of the AAR (AN/AAR) varies widely among species. Values <10% have been reported in dogs, and values as high as 60% have been seen in rats. In pigs, Schott et al2 reported 48.0±12.7%, and Strasser et al21 reported 71.3±4.4%. In the present study, AN/AAR was 66.8±5.4% in the control group and 69.0±5.4% in the stenosis group. Infarct size was reduced to 15.1±5.9% in the PC group and to 29.7±7.1% in the PC/S group. Similar preconditioning benefits have been seen in rats and rabbits with comparable AN/AAR control values. Given the seeming universality of preconditioning in the laboratory, there is little reason to doubt its relevance in humans. Kloner and Yellon22 reviewed this subject recently and concluded, “There is now convincing evidence that human myocardial tissue can be preconditioned.”
This closed-chest swine model closely resembles human clinical physiology in the setting of obstructive coronary artery disease with similar coronary anatomy, paucity of significant collaterals, and a fixed intralumenal stenosis. The phenomenon of ischemic preconditioning observed with a fixed stenosis in place has relevance for the common clinical scenario seen with thrombolytic therapy in acute myocardial infarction. Prodromal angina may be indicative of a preconditioning stimulus and would be analogous to the brief conditioning occlusions used in preconditioning protocols. In this context, a fixed 82% intraluminal stenosis takes on special significance. A recent clinical report6 found that patients with new-onset, prodromal angina had smaller infarcts as assessed by cardiac enzymes and ventriculography compared with patients without prodromal angina. Of interest, patients in this study did not have angiographically visible collaterals to the AAR, and the percent stenosis of the infarct-related artery was 75% to 79%. Thrombolytic therapy with tissue plasminogen activator was the method used to induce reperfusion. Similarly, the TIMI 4 study reported “beneficial in-hospital effects of previous angina” in a select subset of patients eligible for thrombolysis, ie, less heart failure, shock, and smaller infarcts as assessed by serum creatine kinase.7 These beneficial effects were also noted to be independent of angiographically visible collaterals.
Interestingly, mean residual stenosis after thrombolysis in most clinical studies has ranged from 68% to 82%,23 similar to the 82% diameter stenosis used in this study. Thus, human studies and this report suggest that the presence of a fixed stenosis during both the preconditioning phase (angina) and reperfusion (thrombolysis) does not abolish the preconditioning response. Although significant infarct size reduction was demonstrated in the PC/S group, it may have been attenuated in magnitude compared with the PC group. Thus, insufficient by itself to precondition the myocardium, a partial coronary stenosis still allows the protective effect of preconditioning to be elicited with brief episodes of severe ischemia.
Lactate consumption/production by the heart is a useful measure of relative oxidative/nonoxidative metabolism, but chemical measures alone of lactate in the artery and coronary sinus provide only limited information in this regard. Both the oxidative and nonoxidative pathways of lactate metabolism coexist to various degrees in normal and ischemic myocardium. Isotopic lactate release from nonischemic myocardium and isotopic lactate extraction in the face of net chemical production have been demonstrated.17 Chemical measure of lactate balance may therefore fail to detect a degree of ischemia underlying a reduced net chemical extraction. In the case of the two stenosis groups, particularly the PC/S group, the slightly lower level of lactate extraction at baseline may indicate ischemia. From the report by Guth et al,17 calculated chemical subendocardial lactate consumption suggests that at the levels of consumption observed in this study, increased lactate release, a marker of ischemia, could have been present. The degree of ischemia, however, may have been too mild to have a measurable preconditioning effect. Stress ischemia may be necessary to more fully define the preconditioning potential of the stenosis in clinically relevant situations.
Variability of transmural flow in the AAR with respect to the normal zone was seen in all groups. Reduced flow in the AAR could be attributed to the presence of an in-dwelling balloon catheter in the LAD throughout the study. Although low-profile balloons were used, frequent recycling of balloons may have altered its characteristics, thus mildly impeding flow compared with the normal zone. Nonetheless, only a minimal coronary gradient was observed in the groups without the stenosis. Close correlation with blood flow measurements by use of radioactive microspheres has been reported with colored microspheres,13 but only two different-colored spheres were used in that validation study. In the present study, four different-colored spheres were used in each pig. Although the method should be reliable for more than two different colors based on the two-sphere study, this has not been validated. Two recent studies,24 25 however, in which colored spheres were used and four to six injections were made in each animal report data that suggest reasonable consistency with a counting method similar to that used in the present study. Hale et al13 noted that the correlation with radioactive microsphere results was very good until high flows were produced, at which point the colored spheres yielded flows in excess of those seen with radioactive spheres. The present study did not analyze any high flow conditions, with the possible exception of the reperfusion state in the PC group.
Differences were noted in the extent of AAR among the groups although these differences were not statistically significant. These variations in AAR size are not unexpected in a closed-chest model in which consistent placement of the occlusive balloon in the same portion of the LAD was not possible. Despite this technical problem, infarct size was independent of AAR size in the PC group.
Presence of the stenosis during the final reperfusion did not allow a comparison of the stenosis and the PC/S groups to the commonly used occlusion/reperfusion (PC) protocol. Although the flow limitation during final reperfusion adds to our understanding of that condition, the extensive literature evaluating preconditioning interventions has used a model in which full reperfusion follows the prolonged occlusion.
Stenosis patency throughout the study was ensured by monitoring pressure distal to the stenosis. Both the pressure and the waveform were closely monitored for evidence of stenosis closure. As with external occlusion techniques, absolute certainty of continuous patency could not be ensured. Persistent evidence of closure or near-occlusion was a basis for exclusion from the study, but transient and even mild reductions in lumen patency could have gone undetected.
The high attrition rate (57%) resulted in a study group of pigs that had survived a demanding protocol in which 26% died of ventricular fibrillation. The impact of such “natural selection” on the study outcome cannot be determined. It should be noted that the percentage of deaths attributed to ventricular fibrillation was 29% in the preconditioned groups and 24% in the nonpreconditioned animals.
Ischemic preconditioning was observed in closed-chest swine. The presence of an intraluminal coronary stenosis with 82% diameter reduction during the preconditioning, occlusion, and final reperfusion phases moderately attenuated but did not eliminate the protective effect of preconditioning. This may mimic the clinical situation in which preinfarction angina serves as a preconditioning agent, conferring myocardial protection when reperfusion is induced with thrombolysis despite the presence of a residual stenosis.
Selected Abbreviations and Acronyms
|AAR||=||area at risk|
|AIV||=||anterior interventricular vein|
|AN||=||area of necrosis|
|LAD||=||left anterior coronary artery|
|MAP||=||mean aortic pressure|
|MCAPd||=||mean distal coronary artery pressure|
|MV˙o2||=||myocardial oxygen consumption|
|PC/S||=||preconditioned plus stenosis|
We would like to thank the Cardiac Research Laboratory staff (Lorraine Schofield, Patricia Mastrofrancesco, Tammy Donahay, and Leonard Chaves) for their hard work and dedication without which this study would not have been possible. We also thank Steven Reinert and Ranjini Natarajan for statistical advice and assistance.
- Received October 14, 1996.
- Accepted October 27, 1996.
- Copyright © 1997 by American Heart Association
Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986;75:1124-1136.
Schott RJ, Rohman S, Braun ER, Schaper W. Ischemic preconditioning reduces infarct size in swine myocardium. Circ Res. 1990;66:1133-1142.
Deutsch E, Berger M, Kussmaul WG, Hirshfield JW, Herrman HC, Laskey WL. Adaptation to ischemia during percutaneous transluminal coronary angioplasty: clinical, hemodynamic, and metabolic features. Circulation. 1990;82:2044-2051.
Tomai F, Crea F, Gaspardone A, Versaci F, de Paulis R, de Peppo AP, Chiariello L, Gioffre PA. Ischemic preconditioning during coronary angioplasty is prevented by glibenclamide, a selective ATP-sensitive K+ channel blocker. Circulation. 1994;90:700-705.
Ottani F, Galvani M, Ferrini D, Sorbello F, Limonetti P, Pantoli D, Rusticali F. Prodromal angina limits myocardial infarct size: a role for ischemic preconditioning. Circulation. 1995;91:291-297.
Kloner RA, Shook T, Przyklenk K, Davis VG, Junio V, Matthews RV, Burstein S, Gibson M, Poole K, Cannon CP, McCabe CH, Braunwald E, for the TIMI 4 Investigators. Previous angina alters in-hospital outcome in TIMI 4. Circulation. 1995;91:37-47.
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 MM, Simonis LA, de Zeeuw S, Nieukoop S, Post S, Verdouw PD. Ischemic preconditioning by partial occlusion without intermittent reperfusion. Cardiovasc Res. 1994;28:1146-1151.
Schulz R, Post H, Sakks S, Wallbridge DR, Heusch G. Intraischemic preconditioning: increase in tolerance to sustained low-flow ischemia by a brief episode of no-flow ischemia without intermittent reperfusion. Circ Res. 1995;76:942-950.
Fedele FA, Gewirtz H, Capone RJ, Sharaf B, Most AS. Metabolic response to prolonged reduction of myocardial blood flow distal to a severe coronary artery stenosis. Circulation. 1988;78:729-735.
Hale SL, Alker KJ, Kloner RA. Evaluation of nonradioactive, colored microspheres for measurement of regional myocardial blood flow in dogs. Circulation. 1988;78:428-434.
Ovize M, Kloner RA, Hale SL, Przyklenk K. Coronary cyclic flow variations ‘precondition’ ischemic myocardium. Circulation. 1992;85:779-789.
Guth BD, Wisneski JA, Neese RA, White FC, Heusch G, Mazer CD, Gertz EW. Myocardial lactate release during ischemia in swine: relation to regional blood flow. Circulation. 1990;81:1948-1958.
Ito BR. Gradual onset of myocardial ischemia results in reduced myocardial infarction. Circulation. 1995;91:2058-2070.
Strasser R, Avras M, Vogt A, Elsasser A, Schwarz ER, Schlepper M, Schaper W. Preconditioning of porcine myocardium: how much ischemia is required for induction? Circulation. 1994;90(suppl I):I-109. Abstract.
Lefkowitz CA, Pace DP, Gallagher KP, Buda A. The effects of a critical stenosis on myocardial blood flow, ventricular function, and infarct size after coronary reperfusion. Circulation. 1988;77:915-926.
Ishikawa K, Kamata N, Nakai S, Akiyama H, Koka H, Ogawa I, Katori R. Preservation of high regional blood flow at epicardial rim after coronary occlusion. Am J Physiol 1994;267(Heart Circ Physiol 36):H528-H534.
Kobayashi N, Kobayashi K, Kouno K, Horinaka S, Yagi S. Effects of intra-arterial injection of colored microspheres on systemic hemodynamics and regional blood flow in rats. Am J Physiol. 1994;266(Heart Circ Physiol 35):H1910-H1917.