Magnesium Sulfate Reduces Myocardial Infarct Size When Administered Before but Not After Coronary Reperfusion in a Canine Model
Background The role of magnesium in treating acute myocardial infarction (AMI) has been controversial. Several small clinical trials indicate that magnesium may have a role in treating AMI early, whereas the other results suggest that magnesium is of questionable benefit.
Methods and Results We looked at the effect of magnesium on infarct size (IS) when given during a coronary occlusion and after reperfusion. Magnesium sulfate (6-mEq bolus plus 2 mEq/h for 5 hours) was given at 15 or 45 minutes of coronary occlusion or 15 minutes of reperfusion. The left anterior descending coronary artery was occluded for 90 minutes, followed by 300 minutes of reperfusion. IS to area at risk (IS/AR) was measured by planimetry after triphenyltetrazolium chloride staining. Collateral myocardial blood flow was measured with radioactive microspheres. The IS/AR ratio in the control group was 52.3±19.6% compared with 20.5±11.7% and 21.3±6.5% at 15 and 45 minutes of occlusion, respectively (P<.05). There were no significant differences in the reduction in IS at 15 and 45 minutes of occlusion. Although there was a reduction in the IS when magnesium was administered during reperfusion (38.2±13.4%), it was not statistically significant. There was no significant difference in the AR relative to the total left ventricular weight between the four groups.
Conclusions The data suggest that magnesium infusion during a coronary occlusion has a significant benefit in reducing the IS in this model. Magnesium may have a beneficial clinical role in AMI, especially if administered before reperfusion as a bolus followed by a constant infusion.
The treatment and pharmacological management of AMI have been and continue to be difficult medical problems. Ideally, one would like to reduce the extent of MI and preserve as much myocardium as possible. Although several different pharmacological regimens (eg, β-blockers, nitrates, calcium channel blockers, angiotensin-converting enzyme inhibitors, and antithrombotic agents) have been used clinically, they have not proved to be entirely effective or universally used in AMI patients.1 2 3 Although many studies suggest that β-blockers may reduce infarct size, β-blockers have not been fully accepted by the medical community because of their myocardial depressant actions and the concern for precipitating heart failure in patients with large infarctions.1 2 3 4 Perhaps the most widely used and accepted treatment to date has been the combination therapy of thrombolytics and aspirin.5 6 However, the cost of some thrombolytics is a major concern to some institutions and health payers. Two recently published meta-analysis randomized trials (LIMIT-2 and ISIS-4) have studied the role of magnesium in the treatment of MI and AMI.7 8 Unfortunately, these studies did not draw the same conclusions and have raised a specter of doubt on the utility of magnesium in AMI. Magnesium has been suggested as a possible intervention to be used in AMI since the early 1960s mainly because it was thought to be an antiarrhythmic agent,9 although no studies have conclusively shown this to be the mechanism of action for magnesium in reducing mortality. Chang and associates10 reported that dogs with diet-induced magnesium deficiency exhibited significantly larger infarcts than normal control dogs and magnesium-supplemented control dogs. Kingma et al11 compared infarct size in dogs using a low-flow coronary infusion of crystalloid with and without magnesium.11 They failed to show any differences between the two groups; however, infarct size was reduced by the low-flow infusion compared with the total occlusion control group. To date, no definitive studies have looked at infarct size development or the reduction of an infarct by systemic administration of magnesium during infarct development. The purpose of this study was to evaluate the effect of systemic magnesium infusion on infarct size relative to LV at risk in a canine model.
The animal studies reported here conform to the guiding principles on the care and use of laboratory animals of the American Physiological Society and the American Heart Association.
Thirty-two adult mongrel dogs (weight, 20 to 30 kg) were anesthetized with sodium pentobarbital (30 mg/kg IV). The dogs were intubated, placed on a ventilator (Harvard Apparatus), and respired with room air supplemented with 100% oxygen. Arterial blood gases were maintained at a minimum Po2 of 100 mm Hg and a pH of 7.4±0.05. Catheters were placed in the femoral vein and artery for the administration of drugs as needed and for sampling of arterial blood gases and radioactive microspheres, respectively. A left thoracotomy was performed at the fifth intercostal space, and the heart was suspended in a pericardial sling. Arterial pressure and ECGs were measured and recorded on a Gould Instruments ES 1000 recorder. A catheter placed in the left atrial appendage was used for administration of radioactive microspheres to measure collateral and regional myocardial blood flows. The LAD was isolated distal to the first diagonal. All animals received heparin (500 U/kg IV) before coronary occlusion.
Drug Preparation and Administration
Control dogs received physiological saline (0.9% sodium chloride) in a volume equal to that administered to the experimental magnesium groups. The sulfate salt of magnesium was dissolved in distilled water and administered intravenously as a bolus over 5 minutes, followed by a continuous infusion for the duration of the experiment after the initial bolus (Table 1⇓). Four separate groups were studied: group 1 (n=8), control dogs; group 2 (n=8), dogs given magnesium at 15 minutes of coronary occlusion; group 3 (n=8), dogs given magnesium at 45 minutes of coronary occlusion; and group 4 (n=8), dogs given magnesium beginning at 15 minutes of reperfusion and continuing throughout the reperfusion period. All magnesium groups received a 4-mEq bolus over 5 minutes, followed by 2.0 mEq/h for the duration of the study. Total occlusion time was 90 minutes, followed by 300 minutes of reperfusion. Reperfusion was performed by totally releasing the coronary occluder into full hyperemic flow. This was done to simulate what most likely occurs in a clinical setting, where reperfusion is not controlled after angioplasty or thrombolytic reperfusion.
Regional Myocardial Blood Flow Measurements
Regional myocardial blood flow was measured in all dogs with radionuclide-labeled microspheres (15±3 μm, Du Pont–New England Nuclear). Four separate gamma labels were used: cerium-141, chromium-51, niobium-95, and ruthenium-103. Only two microspheres were used per study to reduce the amount of radioactivity that would be disposed of in the dog carcasses. Approximately 4 to 6×106 microspheres were injected into the left atrium, and reference blood samples were obtained from a catheter in the abdominal aorta. Microsphere reference sampling was obtained at a rate of 14.8 mL/min and initiated 15 seconds before microsphere injection into the left atrial catheter.12 The first microsphere was administered at 30 minutes of occlusion for measurement of coronary collateral blood flow; the second microsphere was given at 60 minutes of reperfusion. On completion of the experiment, the LAD was cannulated at the point of occlusion, and 50 mL TTC stain (37°C) was infused into the distal LAD bed, while 50 mL of methylene blue dye was simultaneously infused through the left atrial cannula and allowed to distribute into the myocardium. Once the heart appeared uniformly stained with the blue dye, it was electrically fibrillated and extirpated for evaluation. The heart was sliced bread-loaf style from the apex to the base in 1-cm-thick slices, and the region of interest as defined outside the blue staining was removed and incubated in TTC (37°C) for 20 minutes to allow full staining and demarcation of the infarct and AR. The noninfarcted myocardium and total LV were weighed, and the AR/LV ratio was calculated. The IS/AR ratio was determined by planimetry with a digitizing software program and an IBM personal computer.
ANOVA was performed on the data followed by a Student nonpaired t test. Values are reported as mean±SD unless otherwise noted. Statistical values were considered significant when P<.05.
Table 2⇑ presents the hemodynamic data. As may be observed, there were no significant differences among the four groups with regard to heart rate, arterial systolic pressure, or rate-pressure product during any of the measured periods.
Table 3⇑ gives the regional myocardial blood flow data obtained with radioactive microspheres. There were no significant differences in the collateral blood flow within the AR during coronary occlusion. There were no significant differences in the transmural blood flow at 60 minutes of reperfusion among the four groups.
ANOVA showed a significant (P<.0006 with 3 df; F=7.98; MS=455; SS=1366) difference among the four treatment groups. The Student’s t test revealed a significant difference (P<.05) in the IS/AR ratio (Table 4⇑ and Fig 1⇓) and the IS/LV ratio (Table 4⇑ and Fig 2⇓) when the magnesium treatment groups (groups 2 and 3) were compared with the control group (group 1) and the magnesium reperfusion group (group 4). The IS/AR ratio was 52.3±21.0% in the control group and 20.5±12.5% and 21.3±6.9% for the groups with infusions initiated at 15 and 45 minutes, respectively. There were no significant differences in the reduction of infarct size for the groups with treatments initiated at 15 and 45 minutes (group 2 versus group 3). Although there was a reduction in the infarct size of the animals treated after reperfusion (group 4) compared with the control group (group 1), it was not statistically significant. There was no significant difference in the AR/LV ratio among the four groups (Table 4⇑).
The intravenous administration of magnesium in the treatment group at 15 and 45 minutes of coronary occlusion and before reperfusion resulted in a significant reduction in infarct size in this canine model. The reduction in infarct size amounted to a >60% decrease at both time periods. Because the collateral coronary perfusion in all groups was similar, the data suggest that the results were not due to differences in collateral blood flow. Magnesium has been reported by Vigorito et al13 to reduce coronary vascular resistance in patients with normal coronary arteries undergoing a diagnostic catheterization. This suggests that magnesium may have increased coronary blood flow in our model; however, we did not observe any changes in collateral coronary blood flow during coronary occlusion with the magnesium infusion, and the reperfusion blood flows were not significantly different among the groups studied. In group 4, in which magnesium was infused after reperfusion, we observed a decrease in infarct size, but it was not statistically significant. In many experimental studies investigating the ability of a drug to reduce infarct size, the experimental animals are often pretreated with the drug before the coronary occlusion. The animals in this investigation were not pretreated with magnesium, and it was administered only after the coronary artery had been occluded for at least 15 minutes or after reperfusion. Such a treatment is similar to a clinical situation in which a patient presents with symptoms of AMI and chest pain. Unfortunately, most patients do not come into the clinic at the first signs or symptoms of chest pain or ischemia and would most likely not be administered magnesium in the first 15 minutes after arrival.
The fact that we did not see a difference between the groups receiving treatment at 15 and 45 minutes suggests that administering magnesium later into an AMI before reperfusion may still be of benefit. This was somewhat of a surprise because we had anticipated a greater reduction in infarct size in the group treated at 15 minutes, since the therapy was begun before cell death was expected to occur. One significant difference between the studies reported here and what was done in the ISIS-4 trials is that the ISIS-4 protocol called for magnesium to be administered any time within 24 hours of the AMI, which would indicate that the magnesium was probably given after reperfusion. Our results indicate that administering magnesium after reperfusion decreases the likelihood of infarct size reduction, and it is more effective when administered before reperfusion.
It is an accepted fact that there is a significant increase in and release of oxygen free radicals at the time of reperfusion after coronary occlusion.14 15 Many agents that block free radical formation have been shown to reduce infarct size if administered before reperfusion.16 17 In cell culture studies, magnesium has been shown to block free radical formation.18 Although free radical formation has not been investigated with magnesium in an in vivo preparation, there may be similar actions, and its administration before reperfusion would result in the suppression of free radicals on reperfusion. In group 4, in which we administered magnesium after 15 minutes of reperfusion, we observed a reduction in the IS/AR and IS/LV ratios, although the reduction was not statistically significant. When magnesium was administered before reperfusion, there was a significant reduction in infarct size. Thus, the possibility that magnesium reduces free radical formation before reperfusion is highly likely and may account for the large reduction in infarct size we observed in these studies.
The reason that magnesium was so effective in reducing infarct size is not known, but we would not attribute it to a reduction in either heart rate or systolic arterial pressure or to an effect on coronary blood flow. There was not a significant difference in the rate-pressure product; thus, we do not attribute the effects of magnesium to a reduction in afterload or wall stress. Furthermore, we would not attribute the differences to a reduction in the number of arrhythmias observed during or after coronary occlusion. Previous studies by Schechter et al19 have shown that although magnesium reduced the number of arrhythmias in patients compared with control placebo-treated patients, the difference was not statistically significant and most likely did not account for the beneficial effects of magnesium in their patient population. Before selecting the magnesium dose to use, we performed two preliminary experiments using a bolus of 8 mEq followed by 4 mEq/h constant infusion for the duration of the experiment (unpublished data). In the preliminary studies, we observed a significant reduction in heart rate and systemic arterial pressure (rate-pressure product). We therefore reduced the dose used in these studies to one half the initially tested dose to preclude the significant hemodynamic changes observed at the higher level. Gross et al20 have shown that infarct size can be reduced with bradycardic agents, and we wanted to avoid any bradycardic effects that magnesium may induce.
Magnesium is an important cofactor for many enzymatic reactions and intracellular ATPase activity and may be important in cellular recovery after an ischemic period. Haigney et al21 recently reported that magnesium levels were significantly reduced in patients with ST elevation during AMI. Previous reports have suggested that magnesium may interfere with calcium uptake and thus limit the extent of the infarction. We did not perform studies looking at the uptake of radioactive calcium and to state this as the mechanism would be purely speculative.
Many different drugs have been studied for their ability to reduce infarct size, morbidity, and mortality in the setting of AMI.7 8 LIMIT-2, based on 2316 patients, is perhaps one of the best double-blind clinical trials published to date. The results of the LIMIT-2 study showed a statistically significant lower incidence of mortality and a reduction in LV failure in the magnesium-treated groups. The incidence of MI was the same in both groups, but IS was not measured in any patients. The results of the LIMIT-2 trial suggested that the ability of magnesium to reduce mortality is comparable to that achieved with thrombolytic drugs and aspirin.8 Schechter et al19 reported in 1990 that intravenous magnesium significantly reduced in-hospital mortality compared with placebo in a controlled double-blind randomized trial. More recently, Schechter et al22 reported that magnesium administered to AMI patients (70 years of age and older) who were not candidates for thrombolytic therapy was more effective in reducing in-hospital mortality compared with a randomized double-blind placebo control population.
There is a clinical need for a good agent that would help preserve myocardium at risk in AMI. Magnesium may be such an alternative and provides a low-cost agent for treating AMI. The likelihood of an untoward event with magnesium is relatively small, especially if the loading dose is administered slowly over a 5-minute period. It should be noted that we experienced a single case of asystole in the cardiac catheterization laboratory when magnesium was injected too rapidly. The patient was immediately resuscitated and had no other complications or events. Obviously, as with any drug, caution should be exercised with intravenous administration.
Selected Abbreviations and Acronyms
|AMI||=||acute myocardial infarction|
|AR||=||area at risk|
|ISIS-4||=||International Study of Infarct Survival–4|
|LAD||=||left anterior descending coronary artery|
|LIMIT-2||=||Leicester Intravenous Magnesium Intervention Trial–2|
|LV||=||left ventricular weight|
The authors thank Dr Garrett J. Gross for his critical review and suggestions in these studies. They also thank Ms Lori Roesch for her assistance in typing the manuscript.
- Received May 18, 1995.
- Revision received July 26, 1995.
- Accepted August 15, 1995.
- Copyright © 1995 by American Heart Association
Yusuf S, Sleight P, Rossi P, Ramsdale D, Peto R, Furze L, Sterry H, Pearson M, Morwani R, Parish S, Gray R, Bennett D, Bray C. Reduction in infarct size, arrhythmias, and chest pain by early intravenous β-blockade in suspected acute myocardial infarction. Circulation. 1983;67(suppl II):II-32-II-41.
Roberts R, McCabe JB, Meister FL, Rogers WJ. Results of ISIS-4, GISSI-3, and other major clinical trials of interventional therapy. Clinical Challenges in Acute Myocardial Infarction. 1994;4:1-11.
Domenech RJ, Hoffman JIE, Noble MIM, Saunders KB, Henson JR, Subijanto S. Total regional coronary blood flow measured by radioactive microspheres in conscious and anesthetized dogs. Circ Res. 1969;25:581-596.
Gardner TJ, Stewart JR, Casale AS, Downey JM, Chambers DE. Reduction of myocardial ischemic injury with oxygen-derived free radical scavengers. Surgery. 1983;49:423-427.
Jolly SR, Kane WJ, Bailie MB, Abrams GD, Lucchesi BR. Canine myocardial reperfusion injury: its reduction by the combined administration of superoxide dismutase and catalase. Circ Res. 1984;54:277-285.
Myers ML, Bolli R, Lekich RF, Hartley CJ, Roberts R. Enhancement of recovery of myocardial function by oxygen free radical scavengers after reversible regional ischemia. Circulation. 1985;72:915-921.
Haigney MCP, Gerstenblith G, Silverman HS, Silver BB, Shapiro EP, Schulman SP. Intracellular magnesium levels in patients with ST elevation myocardial infarcts receiving intravenous magnesium therapy. Circulation. 1993;88(suppl I):I-395. Abstract.
Schechter M, Hod H, Kaplinsky E, Rabinowitz B. Magnesium as alternative therapy in acute myocardial infarction patients unsuited for thrombolysis. Circulation. 1993;88(suppl I):I-395. Abstract.