Timing of Magnesium Therapy Affects Experimental Infarct Size
Background Controversy exists regarding the use of magnesium in the treatment of acute myocardial infarction (AMI) because of apparent conflicting results from clinical trials. One hypothesis to explain the various clinical observations proposes that the timing of magnesium administration significantly influences its therapeutic effect; ie, supraphysiological levels of Mg2+ must be present at the time of reperfusion for magnesium to produce clinical benefit.
Methods and Results These experiments evaluated the effect of varying the timing of magnesium administration during AMI. Female Yorkshire swine (34 to 42 kg) underwent thoracotomy and 50 minutes of left anterior descending coronary artery (LAD) occlusion, followed by 3 hours of reperfusion. In the first group, MgSO4 (250 mg of magnesium diluted in 60 cm3 saline) was infused into the LAD over 12 minutes, beginning immediately with the onset of reperfusion (n=6, Mg-early group). In the second group, MgSO4 was given after 1 hour of reperfusion (n=6, Mg-late group). Six pigs received saline instead of magnesium and served as the control group. Lethal arrhythmias were significantly reduced in the Mg-early group. Infarct size was determined by vital staining. Infarct size was 0.16±0.05 g/kg body wt (Mg-early), 0.35±0.08 g/kg (Mg-late), and 0.42±0.04 g/kg for the control group. Compared with the control group, significant (P=.029) reduction in infarct size occurred in the Mg-early group but not in the Mg-late group.
Conclusions We conclude that intracoronary MgSO4 delivered during reperfusion can significantly diminish infarct size in swine, but the timing of administration is critical.
Controversy surrounds the use of intravenous supplementation with magnesium salts during AMI because of apparent conflicting results from clinical trials. Most recently, LIMIT-2 (2316 patients) showed a significant mortality benefit,1 2 whereas ISIS-4 (54 824 patients) found no benefit.3 The large sample size in ISIS-4 has led many observers to view this trial as definitive. Other observers point to methodological differences between these trials that may explain the discordant results, thus suggesting a need to reexamine the claim that magnesium therapy is a useful adjunct in the treatment of AMI.2 4 Specifically, patients in LIMIT-2 were randomized at presentation to receive either MgSO4 or placebo, so that MgSO4 was given concurrently with thrombolytic therapy in those patients receiving thrombolytics. However, patients in ISIS-4 received thrombolytics, and the early lytic phase was completed before randomization to study drug. In ISIS-4, therefore, it is likely that reperfusion was achieved in most patients before MgSO4 was administered. The median time from onset of symptoms to randomization was 3 hours in LIMIT-2 compared with 8 hours in ISIS-4.2
The present study tested a hypothesis that could provide insight into the failure of ISIS-4 to demonstrate a mortality benefit related to magnesium administration. We hypothesized that magnesium supplementation must occur during the initial phase of reperfusion and that late administration would have no effect. Because infarct size is the primary determinant of mortality after AMI, the present study compared the effect of two treatment strategies (early and late magnesium administration) on experimental infarct size in swine. We also hypothesized that only the initial reperfusion phase is important, so a brief infusion would be enough. Additionally, to minimize potential confounding factors related to systemic effects, we limited the MgSO4 infusion to local delivery of a small dose.
A swine model was chosen for several reasons. The coronary anatomy of swine mimics the human coronary circulation closely, especially with regard to a relative absence of preexisting collateral flow.5 Also, the Yorkshire swine’s heart is large enough to produce a model of regional ischemia similar to the human condition and to allow local drug delivery.
This study was approved by the Institutional Animal Care and Use Committee of the University of Maryland. All procedures conformed to the guidelines established by the US Department of Health and Human Services and published by the NIH (NIH publication No. 85-23, revised 1985).
Twenty-six 10- to 12-week-old purebred Yorkshire female swine (34 to 42 kg) received normal swine food (Purina) and were housed at our institution for a minimum of 1 week before use. The pigs were randomized into three groups. All three groups were subjected to 50 minutes of occlusion of the LAD followed by 3 hours of reperfusion. The first group (average weight, 37.4±1.7 kg) received intracoronary normal saline at the beginning of reperfusion and served as the control group. The second group (average weight, 35.9±2.0 kg) received intracoronary MgSO4 with the onset of reperfusion that lasted 12 minutes (Mg-early group). The third group (average weight, 38.2±2.1 kg) received intracoronary MgSO4 beginning 1 hour into the reperfusion period and lasting 12 minutes (Mg-late group).
Initial sedation was achieved with 10 mg/kg IM ketamine. An ear vein was then cannulated for administration of an infusion of ketamine and thiamylal as needed to supplement anesthesia. The swine was intubated and ventilated with oxygen at a flow of 2 L/min and isofluorane in a concentration of 1.25%, which resulted in Po2 >100 mm Hg. Isofluorane was the primary anesthetic agent. Suxamethonium (3 mg IV) was given as a neuromuscular blocking agent.
A femoral vein and both femoral arteries were cannulated. Saline was infused in the femoral vein to maintain left ventricular end-diastolic pressure at baseline level. Blood samples were drawn regularly from the femoral artery for blood gas monitoring. Arterial blood pH was maintained within a physiological range of 7.40±0.05. A pressure transducer was advanced to the thoracic aorta for continuous pressure monitoring. A median sternotomy was performed. The anterior pericardium was opened, and the heart was exposed. A microtipped pressure transducer (Millar Instruments) was positioned through the left ventricular apex for measurement of left ventricular pressure and the first derivative of pressure (dP/dt). The LAD was identified, and a Silastic snare (Retracto-tape) was placed around the midportion after the first diagonal branch (about 4 cm distal to the ostium of the LAD) for the subsequent 50-minute occlusion.
A Tracker-18 infusion catheter with a radiopaque tip (Target Therapeutics) was advanced through the right femoral artery by use of a 7F guiding catheter (Medtronic) and an 8F hemostasis sheath (Daig) with fluoroscopic guidance. The infusion catheter was placed in the midportion of the LAD, distal to the surrounding snare. Limb leads were placed for ECG monitoring.
After stabilization of hemodynamics, a baseline period was recorded. Arterial blood pressure and ECG were recorded continuously. After completion of the baseline phase, a 50-minute occlusion of the LAD was performed with a plastic snare (over the catheter) followed by reperfusion for 3 hours. Simultaneous with release of the snare after the 50-minute occlusion, 60 cm3 normal saline was infused through the catheter directly into the LAD. A syringe infusion pump (Harvard Apparatus) delivered the drug solution at a rate of 5 mL/min for 12 minutes. Blood pressure, dP/dt, ECG, and left ventricular pressure were recorded on an eight-channel recorder (Gould Instruments) at a speed of 50 mm/s. Recordings were made at baseline; at 25 and 50 minutes of occlusion; during the magnesium infusion; and at 10, 15, 60, 120, and 180 minutes during reperfusion.
After the 3-hour reperfusion period, tetrazolium red dye (1.5%) was infused through the infusion catheter directly into the LAD, distal to the occlusion site. This dye stains ischemic but not necrotic tissue by undergoing reduction in the presence of dehydrogenase enzymes present in viable myocardial tissue and forming a bright red formazan precipitate.6 The red dye was restricted to the area distal to the occlusion site by reoccluding the LAD over the tracker immediately after the dye was infused. Evans blue dye (60 mL) was infused intraventricularly and perfused systemically to stain tissue not at risk. The necrotic myocardial tissue, which was also distal to the occluded infusion catheter but did not react with the red dye, remained white.
The swine was killed with 30 cm3 KCl. The heart was excised and maintained in 10% formalin for 1 week. After fixation, the heart was sliced and divided into three parts based on staining characteristics (white, necrotic tissue; red, ischemic tissue; and blue, noninjured tissue). The ventricular sections were cut by an individual unaware of the treatment protocol. The tissue was allowed to desiccate further for 1 week and then weighed. The total area at risk and the infarct size were expressed as grams per kilogram of body weight.
The LAD was occluded for 50 minutes. At the onset of reperfusion, a previously prepared magnesium sulfate (American Regent Laboratories) solution (250 mg diluted in 60 mL of 0.9% saline) was administered into the LAD for 12 minutes with a syringe infusion pump (Harvard Apparatus). Hemodynamics were monitored, and the remainder of the experiment was identical to that of the control group.
The LAD was occluded for 50 minutes. At 1 hour into the reperfusion period, magnesium (250 mg/60 cm3 normal saline) was infused at 5 mL/min over 12 minutes. Hemodynamics were recorded, and the remainder of the experiment was identical to that of the control and Mg-early groups.
All comparisons were made by use of repeated-measures ANOVA. Values are expressed as mean±SEM. A value of P<.05 was considered statistically significant. Scheffé’s test was used in comparisons involving three or more variables.
Exclusion of Swine and Arrhythmia
Of the 26 swine under investigation, 8 were excluded for ventricular fibrillation during occlusion or early reperfusion (3 of 9 in the control group, 1 of 7 in the Mg-early group, and 4 of 10 in the Mg-late group). The incidence of ventricular fibrillation in the Mg-early group was significantly lower (P=.024) than in the control and Mg-late groups (P=.011). The remaining 18 pigs, 6 from each group, were analyzed for infarct size.
Table 1⇓ summarizes these data. There were no significant differences in heart rate, left ventricular end-diastolic pressure, or left ventricular systolic pressure between control and magnesium-treated groups at baseline or during the first 120 minutes of reperfusion. At 180 minutes, the Mg-early group had a lower left ventricular end-diastolic pressure and tended to have a higher heart rate than swine in the control and Mg-late groups.
Table 2⇓ summarizes the data for each experiment. Total area of risk (ischemic plus necrotic) did not differ significantly between the Mg-early (0.52±0.10 g/kg), Mg-late (0.53±0.11 g/kg), and control (0.57±0.11 g/kg) groups. The infarct size was 0.42±0.04 g/kg for the control group and 0.16±0.05 g/kg for the Mg-early group. This represents a >50% reduction in infarct size in the Mg-early group compared with the control group (P=.029). Infarct size in the Mg-late group was 0.35±0.08 g/kg, which did not represent a statistically significant difference from that in the control group. The Figure⇓ summarizes and compares these data by groups.
Substantial improvement in mortality from AMI has been attained by the application of treatment strategies designed to achieve early reperfusion of the infarct-related artery.7 8 Despite these gains, AMI still causes significant mortality and morbidity. For years, astute clinicians and researchers have recognized that the consequences of reperfusion may not be entirely salutary.9 Cells subjected to ischemia-reperfusion may suffer deleterious effects, dubbed reperfusion injury, related solely to reperfusion. One attractive strategy to achieve further clinical gains in the treatment of AMI is to use adjunctive therapies, which may either limit reperfusion injury or prevent repeated episodes of ischemia by maintaining arterial patency. With regard to reperfusion injury, various experimental models have shown that modification of the conditions of reperfusion can alter the functional recovery.10 However, the results of clinical trials evaluating the use of adjunctive therapies to decrease reperfusion injury have been disappointing.11
Magnesium therapy showed promise as a valuable adjunctive therapy for AMI in multiple, small-scale clinical studies.12 13 The mechanism of action was unknown. It was initially hypothesized that supplemental magnesium would have an antiarrhythmic effect, reducing peri-infarct arrhythmias. Although an antiarrhythmic effect may occur, the observed decreases in peri-infarct heart failure and mortality are not explained by a decrease in arrhythmia. Other proposed mechanisms include antiplatelet effects (which may help to prevent arterial reocclusion) and vasodilatory effects (decreasing afterload and preventing spasm).14 With regard to a possible mechanism of action, magnesium is known to function as an inorganic calcium channel blocker.15 16 The site of action may be either at the level of the plasma membrane or an intracellular site. Specifically, magnesium can inhibit efflux of calcium from cardiac sarcoplasmic reticulum.17 Although the mechanism for its clinical benefit was unknown, several large clinical trials were conducted to study the effects of magnesium therapy in AMI. LIMIT-2 (2316 patients) seemingly confirmed the value of magnesium in terms of peri-infarct heart failure and both short- and long-term mortality. Then, ISIS-4 (58 050 patients) found no benefit from magnesium, apparently definitively contradicting the results of LIMIT-2.3
Concurrently, researchers in reperfusion injury demonstrated a beneficial effect of magnesium supplementation in experimental models of ischemia-reperfusion. In an isolated perfused heart, enhanced functional recovery with magnesium and various other inhibitors of cellular calcium influx was observed.18 It is generally accepted that the phenomenon of myocardial stunning is an example of reperfusion injury.10 In this laboratory, we demonstrated that myocardial stunning in a swine model is prolonged by hypomagnesemia and is ameliorated by MgSO4 supplementation.19 20
Because reperfusion injury occurs early, it is likely that an effect on reperfusion injury could account for the difference between the results of the LIMIT and ISIS studies. The reason is that magnesium was given shortly after presentation in the LIMIT-2 study design, probably before or early in reperfusion. The ISIS-4 study design, in contrast, mandated that magnesium be given after the early lytic phase, after the time when initial reperfusion probably occurred.3 Although an improvement in myocardial stunning may result in a lower incidence of heart failure during the peri-infarct period, it seems unlikely that this alone could produce the long-term mortality benefit observed in the LIMIT-2 trial. Such a benefit, we reasoned, would probably come from a reduction in infarct size.
The concept that cell necrosis can be prevented by altering reperfusion injury is more controversial than an effect on stunning. One experimental study showed decreased infarct size in rats pretreated with magnesium sulfate.21 With pretreatment, however, one cannot distinguish between an effect occurring during the ischemic or the reperfusion phase. Another recent study demonstrated a decrease in infarct size in a canine model when magnesium was administered during reperfusion.22 Infarct size was reduced when magnesium was given at 15 minutes into reperfusion. The relation of infarct size to timing of administration was concordant with our observations: a greater reduction in infarct size occurs with early administration.
Our results strongly suggest that the beneficial effect of MgSO4 on infarct size is due to an effect on reperfusion injury. The reason is that treatment was brief and limited to the reperfusion phase. Additionally, we observed no systemic hemodynamic effects from the local delivery of 250 mg MgSO4. We speculate that the physiological function of the magnesium ion as an inorganic calcium channel blocker probably is the specific mechanism of this effect. The increased Mg2+ may blunt the sudden rise in cytosolic Ca2+ that occurs early in reperfusion.23 This rise in Ca2+ may, either directly or indirectly, result in the death of cells that otherwise would remain viable.24 25 If this hypothesis is true, a late or a prolonged infusion of magnesium is unnecessary. One could further speculate that prolonged infusion of large doses of magnesium may have deleterious effects owing to systemic hypotension or negative inotropy.
In summary, supplemental MgSO4 can reduce infarct size in a swine model of AMI. However, the timing of treatment is very important, and a delay of even 1 hour results in no effect. The crucial time is early; treatment that encompasses only the first 12 minutes of reperfusion produces substantial benefit. The effectiveness of a brief infusion suggests that prolonged dosing regimens are unnecessary. Local delivery of a small dose works well, with no systemic hemodynamic effects. The timing of treatment is so important that it may explain the failure of ISIS-4 to show a clinical benefit from MgSO4 therapy. Our data are concordant with the observations of LIMIT-2, suggesting that additional benefit can be achieved beyond the benefit related to early reperfusion. Further investigation of the evident cardioprotective properties of magnesium in the setting of ischemia-reperfusion is warranted.
Selected Abbreviations and Acronyms
|AMI||=||acute myocardial infarction|
|ISIS||=||International Study of Infarct Survival|
|LAD||=||left anterior descending coronary artery|
|LIMIT||=||Liecester Intravenous Magnesium Intervention Trial|
This study was supported by a faculty seed fund of the University of Maryland School of Medicine, grant 02390510. We thank Helen Scott for excellent technical assistance. We also thank Dr Robert M. Califf for his constructive comments about the manuscript.
- Received October 31, 1994.
- Revision received April 6, 1995.
- Accepted April 25, 1995.
- Copyright © 1995 by American Heart Association
Fedor JL, McIntosh DM, Rembert JC, Greenfield JC. Coronary and transmural myocardial blood flow responses in awake domestic pigs. Am J Physiol. 1978;235:H435-H444.
Braunwald E, Kloner RA. Myocardial reperfusion: a double edged sword? J Clin Invest. 1985;76:1713-1719.
Bolli R. Mechanism of myocardial ‘stunning.’ Circulation. 1990;82:723-738.
Flaherty JT, Pitt B, Gruber JW, Heuser RR, Rothbaum DA, Burwell LR, George BS, Keriakes DJ, Deitchman D, Gustafson N, Brinker JA, Becker LC, Mancini GBJ, Topol E, Werns SW. Recombinant Human Superoxide Dismutase (h-SOD) fails to improve recovery of ventricular function in patients undergoing coronary angioplasty for acute myocardial infarction. Circulation. 1994;89:1982-1991.
Horner SM. Efficacy of intravenous magnesium in acute myocardial infarction in reducing arrhythmias and mortality. Circulation. 1992;86:774-779.
Yusuf S, Teo K, Woods K. Intravenous magnesium in acute myocardial infarction: an effective, safe, simple, and inexpensive intervention. Circulation. 1993;87:2043-2046.
Sadeh M. Action of magnesium sulfate in the treatment of preeclampsia-eclampsia. Stroke. 1989;20:1273-1275.
du Toit EF, Opie LH. Modulation of severity of reperfusion stunning in the isolated rat heart by agents altering calcium flux at onset of reperfusion. Circ Res. 1992;70:960-967.
Herzog WR, Atar D, Tong Mak I, Alyono D, MacCord C, Weglicki WW. Magnesium deficiency prolongs myocardial stunning in an open-chest swine model. Int J Cardiol. 1995;47:105-115.
Christensen CW, Rieder MA, Silverstein EC. Magnesium reduces infarct size whether administered prior to or after coronary reperfusion in a canine model of acute myocardial infarction. Circulation. 1994;90(suppl I):I-480. Abstract.
Steenbergen C, Murphy E, Levy L, London RE. Elevation in cytosolic free calcium concentration early in myocardial ischemia in perfused rat heart. Circ Res. 1987;60:700-707.