Vitamin E has been proposed to be protective against coronary heart disease (CHD) complications. Promising results have emerged from observational studies (1-3), but randomized trials of supplementation with vitamin E did not show statistically significant differences and even raised the possibility that supplements may actually have harmful effects (4,5) In the recent Cambridge Heart Antioxidant Study (CHAOS; with capsules containing 800 or 400 IU daily), however, the rate of nonfatal myocardial infarction but not of cardiac deaths was significantly reduced (6). Thus antioxidant supplementation can be benefical to certain patients but harmful to others (7). In fact, the mechanisms underlying the effects of vitamin E on CHD are poorly understood. Several hypotheses have been proposed including inhibition of oxidation of low-density lipoproteins (8), decreased platelet function (9), and protection of the ischemic myocardium against reperfusion injury (10).
In experimental cardiology, results have been conflicting, showing (to cite only recent studies) either beneficial effect (11-13) or no effect at all (14-16), depending on the experimental model, the animal species, and the end points used in these studies. There is also controversy regarding the dosage at which vitamin E should be given (17,18). In fact, in some animal models, a bell-shaped relation has been reported between vitamine E dosage and its biologic effect, indicating for instance that moderate dietary supplements preserve endothelial vasodilator function in cholesterol-fed rabbits, whereas higher doses are associated with endothelial dysfunction (19). In a recent article, we also reported, in a dog model of ischemia and reperfusion, both beneficial and deleterious effects of vitamin E, with a reduction in the incidence of lethal arrhythmias but with increased infarct size in survivors (20).
Such ambivalent effects of α-tocopherol may explain why, in humans, the effect of α-tocopherol supplementation on CHD could have been neutral, the negative effects masking the beneficial ones. Despite these unresolved questions, consumption of vitamin E in the general population remains considerable (17). It is therefore of the utmost importance to clarify the mechanisms by which α-tocopherol may be deleterious or beneficial for the ischemic heart. Knowing the potential impact of α-tocopherol on the toxicity of reactive oxygen species, it is important to examine whether the myocardial injury occurring when the myocardium is reperfused (reoxygenated) after a short period of ischemia might be influenced by a prior chronic α-tocopherol supplementation. We therefore selected the model of canine myocardial stunning, a model of reversible contractile dysfunction that occurs in the absence of cell necrosis (a confounding factor in the model of ischemia and reperfusion because it results from both ischemia and reperfusion injuries) and is known to be related, at least partly, to the abrupt generation of reactive oxygen species at reperfusion (21). Stunning theoretically should be influenced by a prior antioxidant supplementation.
In these experiments, we chose to supplement the dogs chronically and orally because, to augment the cardiomyocyte membrane content in α-tocopherol, long-term supplementation was thought to be more efficient than injecting α-tocopherol intravenously at the time of reperfusion. The 500-mg dosage was selected as the most appropriate to mimic the human use.
MATERIALS AND METHODS
All animals used in this study were maintained in accordance with the guidelines of the Committee on the Care and Use of Laboratory Animals of the American Heart Association.
Twenty anglo-norman dogs of similar age and weight (18-22 kg) and of either gender were randomized to either a vitamin E-supplemented (n = 10) or control group (n = 10). Control and supplemented dogs were fed a similar standard diet. Supplemented dogs were given 500 mg of α-tocopherol acetate (mixed with the diet) daily for 3 months.
Surgical preparation and experimental protocol
Dogs were premedicated with subcutaneous morphine (0.5 mg/kg). Anesthesia was induced with thiopental (25 mg/kg i.v.). After trachea intubation, dogs were ventilated at a rate of 15 breaths/min with a mixture of room air and oxygen 50% by using a SF4 respirator (Robert et Carrière, Paris, France). Blood gases were repeatedly checked. Tracheal pressure was monitored throughout the experiment. Anesthesia was maintained with 1-1.5% inspired halothane supplied by a Fluotec vaporizer. Temperature was maintained between 37 and 38°C by a heating element under the operating table. Catheters were inserted into the femoral artery and vein for blood-flow measurements, arterial blood pressure recording, blood samples, and administration of intravenous fluid. Electrocardiographic leads were placed subcutaneously for continuous monitoring. An aseptic thoracotomy was performed in the fifth left intercostal space, and the heart was exposed and suspended in a pericardial cradle. A 6F micromanometer-tipped catheter (Gaeltec Instrument, Edinburgh, Scotland) was inserted into the left ventricle for measurement of left ventricular pressure and its first derivative (dP/dt). For calibration, the high-fidelity transducer was first inserted in the thoracic aorta together with a fluid-filled catheter in the femoral artery. Thereafter it was positioned in the left ventricle for continuous left ventricular pressure measurement. A pair of segment length transducers consisting of 2.5-mm diameter piezoelectric crystals (Triton Technology, San Diego, CA, U.S.A.) was inserted into the myocardium through stab wounds. The crystals were inserted in the center of the soon-to-be ischemic area at risk and in the nonischemic zone, parallel to the minor axis, to a depth of ∼5-10 mm, corresponding to subendocardial myocardium and separated by 10-20 mm. The crystals were connected to a sonomicrometer (Triton Technology). The left atrial appendage was cannulated for radioactive microsphere injection. The left anterior descending artery (LAD) was isolated distal to its first diagonal branch, and a silk snare was passed around it for subsequent occlusion and reperfusion (22).
Tracheal, arterial, and left ventricular pressures, segment length, and electrocardiogram signals were digitized through an eight-channel interface (MacLab) and recorded with Macintosh LC computer (Software Chart 3.2.6). After 20-min stabilization, dogs were subjected to a 20-min LAD occlusion followed by 150 min of reperfusion. Dogs that developed lethal arrhythmia during ischemia were not defibrillated and were excluded from the study. When ventricular fibrillation occurred during reperfusion and converted with fewer than four attempts at low-energy cardioversion, dogs were included in the final analysis. We indeed demonstrated that reflow ventricular fibrillation and low-energy internal electric shocks do not significantly damage the myocardium at risk (22). Blood samples for biologic determinations were collected from the femoral vein at baseline. Hemodynamic parameters and segment shortening were measured before LAD occlusion, at 10 and 20 min of coronary occlusion, and at 30, 60, 150 min of reperfusion.
Measurement of area at risk
Immediately before killing the dogs and after having reoccluded the artery at the same site, Uniperse Blue Pigment (Ciba-Geigy, Basel, Switzerland; 0.5 g/kg) was injected into the left atrial appendage for measurement of area at risk. The hearts were stopped by lethal anesthesia and then excised. With this technique, nonischemic tissue appeared blue, whereas the previously ischemic myocardium (area at risk) remains unstained. The left ventricle was then sectioned into 8-mm-thick transverse slices from apex to base. Area at risk was measured by planimetry and expressed as percentage of left ventricle. By using the tetrazolium staining technique, we also checked the absence of subendocardial tissue necrosis (20).
Regional myocardial blood flow
Regional myocardial blood flow was measured as previously described (20). In brief, 1 million microspheres (15 μm diameter) labeled with cerium 141 and scandium 46 (New England Nuclear, Boston, MA, U.S.A.) were injected into the left atrium over a 20-s period, while a reference arterial blood flow sample was collected. Transmural segments from the central ischemic and nonischemic regions of each slice were subdivided into subendocardial and subepicardial halves. Myocardial pieces were weighed and counted for radioactivity in a gamma counter with selected energy windows. After correction of counts for background and crossover, regional myocardial blood flow was expressed in milliliters per minute per gram of tissue (ml.min−1.g−1).
Myocardial blood flow was measured in the nonischemic and in the central ischemic zones. Two measurements of regional blood flow were performed in each dog after 15 min of coronary occlusion and after 120 min of reperfusion.
Regional myocardial function
Regional myocardial function was assessed as systolic shortening of the myocardium between the ultrasonic crystals. Regional systolic shortening (RSS) was calculated as (EDL - ESL)/(EDL), where EDL is end-diastolic length and ESL the end-systolic length. The end diastole was determined by the onset of the rapid upstroke of the left ventricular dP/dt and the end systole by the peak negative dP/dt. Contractile function was normalized to baseline value, considered 100% contractility. All measurements were performed at end expiration.
Biochemical and lipid determinations
Triglyceride and total cholesterol levels in plasma were automatically determined with a Chem 1 Bayer Diagnostic Analyser (Dromont, France). Plasma α-tocopherol level was determined at baseline by high-performance liquid chromatography as described and by using an adaptation of the technique of Vuilleumier et al. (23). In brief, the extraction from plasma involved protein precipitation with ethanol and subsequent extraction into heptane. The heptane was then evaporated under nitrogen, and the concentrated extract was taken up in methanol, ready for injection. Chromatography was performed on a Spherisorb ODS2 column with a mobile phase composed of methanol and a flow rate of 1.5 ml/min. α-Tocopherol was detected at 292-nm wavelength in a single chromatographic run against external standard. Vitamin E plasma concentrations were calculated from the integrated plotted areas.
All values are means ± standard error of the mean (SEM). Hemodynamics and segment shortening in both groups were compared by using two-factor analysis of variance (ANOVA; treatment and time) with repeated measures across the second factor. Unpaired t test was used to compare baseline biologic data and infarct size, area at risk, and regional blood flows in the two groups.
Myocardial ischemia experiments
Ten supplemented dogs and 10 control dogs underwent chest surgery. There was no ventricular fibrillation during ischemia. Among the 20 dogs, however, 12 (seven supplemented and five controls) developed ventricular fibrillation at reperfusion. Eleven dogs were easily and successfully defibrillated and resuscitated. One control dog died because of intractable fibrillation and was excluded from the final analysis. When comparing the dogs that developed ventricular fibrillation with those that did not, we found no difference regarding the variables (such as the area at risk or the severity of ischemia) that could have influenced the outcome and no difference between the control and treated dogs that developed fibrillation.
Hematologic data and triglyceride and cholesterol levels were similar in the two groups (Table 1). Plasma α-tocopherol levels were higher in the supplemented than in the control group (19.1 ± 1.6 and 6.9 ± 0.6 mg/L, respectively; p < 0.001).
Table 2 shows heart rate, systolic and diastolic blood pressure, and maximal and minimal left ventricular dP/dt during the course of the experiment. At baseline and during ischemia and reperfusion, hemodynamic parameters were similar in the two groups.
Collateral blood flow and area at risk
During coronary occlusion, collateral flow in the ischemic zone expressed in milliliters per minute per grams of tissue (Table 3) or in percentage of the flow in the nonischemic zone (19 ± 4% in supplemented vs. 20 ± 6% in controls) was similar in the two groups. Myocardial blood flow was also similar in the two groups during reperfusion. The area at risk was similar in the two groups: 29.3 ± 2.6% of the left ventricle in supplemented dogs and 32.4 ± 1.9% in controls (Fig. 1). Finally, there was no evidence in any dog of subendocardial necrosis, which is not surprising in view of the relatively high collateral flow in these dogs.
Regional contractile function
Figure 2 shows the RSS in the ischemic region in both groups during ischemia and reperfusion. There was no difference between groups at baseline. The two groups also were similar during reperfusion. Postischemic regional systolic shortening in the ischemic zone at 150 min reperfusion represented 41 ± 15% of baseline in controls versus 51 ± 8% in supplemented dogs.
Summary of the results
This randomized trial in the dog demonstrated no statistically significant benefit or harm resulting from 3 months of α-tocopherol supplementation in terms of postischemic ventricular arrhythmias and myocardial dysfunction. Because of the relatively long duration of the trial and the common use of such a dose in the human setting, these findings are particularly informative. Equivocal results from recent clinical trials (4-6) and experimental studies (19,20) have indeed raised the possibility that antioxidant supplements may actually be harmful in the context of cardiovascular disease (24). Factors that could have produced a false finding of no benefit or harm include an inadequate dose of α-tocopherol or a lack of intestinal absorption of the compound. The concentration of plasma α-tocopherol was, however, 3 times higher in treated than in control dogs; such a concentration is known to be associated with α-tocopherol enrichment of myocardial tissue (25). In view of these plasma levels, poor intestinal absorption of α-tocopherol in our dogs is also not a plausible explanation for our findings.
The goal of our study was to examine in the dog the effects of a pharmacologic dose of α-tocopherol on myocardial stunning, a phenomenon independent of cell necrosis and related at least partly to the abrupt generation of reactive oxygen species in the reoxygenated myocardium (21). The potent antioxidants, superoxide dismutase and catalase, have actually been shown to attenuate myocardial stunning when injected acutely before myocardial reoxygenation (26,27), and electron spin-resonance techniques have shown increased reactive oxygen species production during reperfusion of the ischemic myocardium (21,28). Thus a causal relation seems probable. However, as shown in various experimental models, excess antioxidants could also exert prooxidant activity and result in deleterious effects (18-20,29-31).
The results of our study are quite reassuring because orally and chronically supplemented dogs did not exhibit more severe stunning than did controls. On the other hand, regarding a possible protective effect, our results confirm a previous negative report in the swine (14) and may question the theory that vitamin E can protect the postischemic myocardium against oxidative damage (32).
Limitations of the study
One possible explanation for the apparent lack of effect of α-tocopherol in this study is the plasma level of α-tocopherol in the control group. The question arises whether the control animals were not already maximally protected with a concentration of 6.9 mg/L. In fact, we do not know whether oxidative stress provoked by 20-min ischemia followed by reperfusion generates less-reactive oxygen species than oxidative stress provoked by reoxygenation after a longer period of ischemia, as for instance in our previous study in the dog with a 2-h ischemia (20). If there is a difference in the severity of the oxidative stress in the two conditions, it is likely that different concentrations of antioxidants would provide different protective effects. To address this issue, future experimental protocols should include several groups of animals with a progressive increment of dosages of α-tocopherol and perhaps a specific group maintained in an α-tocopherol-deficient diet. Such a study would need a great number of animals and is likely not feasible in dogs. In addition, our goal in this study was to compare dogs supplemented with a large dose of α-tocopherol with dogs receiving a current dietary dose rather than demonstrating that α-tocopherol supplementation is better than deficiency, in terms of myocardial protection. Further studies are nonetheless needed to explore that important point and to help to reconcile negative results of studies using α-tocopherol (14-16) with the improvement in myocardial function reported by investigators using other acutely administered antioxidants (26,27), in particular, water-soluble α-tocopherol analog (13).
Myocardial stunning and anesthesia
Halothane was the major anesthetic in our study. It is noteworthy that, given the severity of ischemia (as evaluated by the residual flow in the area at risk), myocardial dysfunction in both groups of our study was lower than expected in studies using barbiturates (10,13,14). Barbiturates are the classic anesthetics used in most experimental studies in the dog. They are, however, known to have marked cardiodepressive effects (33,34), and several studies have shown that in the conscious dog, the recovery of contractility after 15-min ischaemia followed by reperfusion is much higher than that in barbiturate-anesthetized dogs (35,36). Thus anesthesia interferes with the phenomenon of myocardial stunning, and the use of barbiturates, although classic, may represent an artefact and should not be considered a standard. On the other hand, halothane has been reported to have some cardioprotective effects (37-39). This renders the comparison of studies using different anesthetics very difficult. Thus ideally, the demonstration that one agent has a protective effect should be done in models using various anesthetics, as well as in conscious dogs.
Role of the substrates of oxidation
Regarding the effect of α-tocopherol against CHD complications, it should be remembered that the biologic effects of α-tocopherol and other antioxidants are dependent not only on their absolute levels in a given medium or tissue but also on the presence and amounts of the various substrates susceptible to be attacked by reactive oxygen species. In cell membranes, for instance, α-tocopherol essentially protects the polyunsaturated fatty acids, which are much more sensitive to oxidation than are other lipids. Recent clinical studies suggested that high levels of antioxidants were protective only among persons with high levels of serum cholesterol (40) or polyunsaturated fatty acids (41). The roles of iron and copper status also obscure the issue (42). Thus depending on the animal model, it is important to take some lipid parameters into consideration. In our study, blood cholesterol was similar in both groups, but the level of cholesterol is rather low in the dog and probably not as important as in humans. Also, the diet of animals and the subsequent fatty acid composition of cell membranes are rarely evaluated in animal cardiac experiments, although it is clear that they are important (43,44), particular when considering α-tocopherol requirements necessary to prevent enhanced lipid peroxidation (45).
Finally, when considering antioxidant drugs in general, our data do not mean that these drugs are not able to protect the postischemic myocardium. Other than α-tocopherol, antioxidants that possess additional properties such as water solubility or the ability to cross the cardiomyocyte membrane easily could be efficient. The purpose of our study was, however, to examine the effect of a currently used antioxidant in the general population. The data show that, regarding myocardial stunning, α-tocopherol was neither useful nor harmful.
Our study clearly indicates that in the halothane-anesthetized dog model of myocardial stunning, long-term and large supplementation with α-tocopherol (500 mg daily) neither augments nor decreases myocardial injury. Thus if pharmacologic doses of α-tocopherol are actually useful to improve the prognosis of cardiac patients, it is likely not by interfering with the stunning phenomenon.
1. Gey KF, Brubacher BG, Stalinn HB. Plasma levels of antioxidant vitamins in relation to haemodynamic heart disease and cancer. Am J Clin Nutr
2. Stampfer MJ, Hennekens CH, Manson JE, Colditz GA, Rosner B, Willett WC. Vitamin E consumption and the risk of coronary disease in women. N Engl J Med
3. Rimm EB, Stampfer MJ, Ascherio A, Giovannucci E, Colditz GA, Willett WC. Vitamin E consumption and the risk of coronary disease in men. N Engl J Med
4. The α-Tocopherol, Beta-Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med
5. Enström JE, Pauling L. Mortality among health-conscious elderly Californians. Proc Natl Acad Sci U S A
6. Stephens NG, Parsons A, Schofield PM, et al. Randomised controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study (CHAOS). Lancet
7. Herbert V. The antioxidant supplement myth. Am J Clin Nutr
8. Steinberg D. Antioxidants and atherosclerosis. Circulation
9. Salonen JT, Salonen R, Seppanen K, et al. Effects of antioxidant supplementation on platelet function: a randomized pair-matched, placebo-controlled, double-blind trial in men with low antioxidant status. Am J Clin Nutr
10. Kloner RA, Pzyklenk K, Whittaker P. Deleterious effects of oxygen radicals in ischemia/reperfusion. Circulation
11. Lukovic L, Petty MA, Bolkenius FN, Grisar JM, Dow J, De Jong W. Protection of infarcted, chronically reperfused hearts by an α-tocopherol analogue. Eur J Pharmacol
12. Massey KD, Burton KP. α-Tocopherol attenuates myocardial membrane-related alterations resulting from ischemia and reperfusion. Am J Physiol
13. Zughaib ME, Tang XL, Schleman M, Jeroudi MO, Bolli R. Beneficial effects of MDL 74,405, a cardioselective water soluble alpha tocopherol analogue, on the recovery of function of stunned myocardium in intact dogs. Cardiovasc Res
14. Buchwald A, Klein HH, Lindert S, et al. Effect of α-tocopherol (vitamin E) in a porcine model of stunned myocardium. J Cardiovasc Pharmacol
15. Ooiwa H, Janero DR, Stanley AWH, Downey JM. Examination of two small-molecule antiperoxidative agents in a rabbit model of postischemic myocardial infarction. J Cardiovasc Pharmacol
16. Klein HH, Pich S, Lindert-Heimberg S, Nebendahl K, Niedmann P. Failure of chronic, high-dose, oral vitamin E treatment to protect the ischemic, reperfused porcine heart. J Mol Cell Cardiol
17. Meydani M. Vitamin E. Lancet
18. Roberts HJ. Vitamin E. Lancet
19. Keany JF, Gaziano JM, Xu A, et al. Low-dose α-tocopherol improves and high-dose α-tocopherol worsens endothelial vasodilator function in cholesterol-fed rabbits. J Clin Invest
20. Sebbag L, Forrat R, Canet E, Renaud S, Delaye J, De Lorgeril M. Effects of dietary supplementation with α-tocopherol on myocardial infarct size and ventricular arrhythmias in a dog model of ischemia-reperfusion. J Am Coll Cardiol
21. Bolli R, Patel BS, Jeroudi MO, Lai EK, McCay PB. Demonstration of free radical generation in "stunned" myocardium of intact dogs with use of the spin trap alpha-phenyl N
-tert-butyl nitrone. J Clin Invest
22. De Lorgeril M, Basmadjian A, Clement R, Rousseau G, Latour JG. Influence of reflow ventricular fibrillation and electrical defibrillation on infarct size in a canine preparation of myocardial infarction. Cardiovasc Res
23. Vuilleumier JP, Keller HE, Gysel D, Hunziker F. Clinical methods for the routine assessment of the vitamin status in human population. Part 1. The fat soluble vitamins A and E and beta carotene. Int J Vitam Nutr Res
24. Hennekens CH, Buring JE. Antioxidant vitamins: benefits not yet proved. N Engl J Med
25. Machlin LJ, Gabriel E. Kinetics of tissue α-tocopherol uptake and depletion following administration of high levels of vitamin E. Ann N Y Acad Sci
26. Gross GJ, Farber NE, Hardman HF, Warltier DC. Beneficial actions of superoxide dismutase and catalase in stunned myocardium of dogs. Am J Physiol
27. Jeroudi MO, Triana FJ, Patel BS, Bolli R. Effect of superoxide dismutase and catalase, given separately, on myocardial "stunning." Am J Physiol
28. Zweier JL, Flaherty JT, Weisfeldt ML. Direct measurement of free radicals generation following reperfusion of ischemic myocardium. Proc Natl Acad Sci U S A
29. Nelson SK, Bose SK, McCord JM. The toxicity of high-dose superoxide dismutase suggests that superoxide can both initiate and terminate lipid peroxidation in the reperfused heart. Free Rad Biol Med
30. Halliwell B. Reactive oxygen species
in living systems: source, biochemistry, and role in human disease. Am J Med
31. Galinanes M, Ferrari R, Qiu Y, Cargnoni A, Ezolin A, Hearse DJ. PEG-SOD and myocardial antioxidant status during ischemia and reperfusion: dose-response studies in the isolated perfused rabbit heart. J Mol Cell Cardiol
32. Klein HH. Vitamin E cannot protect myocardium against oxidative damage. Cardiovasc Res
33. Komai H, Rusy BF. Differences in the myocardial depressant action of thiopental and halothane. Anesth Analg
34. Chamberlain JH, Seed RG, Chung DCW. Effect of thiopentone on myocardial function. Br J Anaesth
35. Triana JF, Li XY, Jamaluddin U, Thornby JI, Bolli R. Postischemic myocardial "stunning": identification of major differences between the open-chest and the conscious dog and evaluation of the oxygen radical hypothesis in the conscious dog. Circ Res
36. Heyndrickx GR, Baig H, Nellens P, Leusen I, Fishbein MC, Vatner SF. Depression of regional blood flow and wall thickening after brief coronary occlusions. Am J Physiol
37. Kanaya N, Fujita S. The effects of isoflurane on regional myocardial contractility and metabolism in "stunned" myocardium in acutely instrumented dogs. Anesth Analg
38. White JL, Myers AK, Analouei A, Kim YD. Functional recovery of stunned myocardium is greater with halothane than fentanyl anaesthesia in dogs. Br J Anaesth
39. Crystal GJ, Khoury E, Guzevicius J, Salim R. Direct effects of halothane on coronary blood flow, myocardial oxygen consumption and myocardial segmental shortening in in situ canine hearts. Anesth Analg
40. Street DA, Comstock GW, Salkeld RM, Schuep W, Klag MJ. Serum antioxidants and myocardial infarction: are low levels of carotenoids and α-tocopherol risk factors for myocardial infarction? Circulation
41. Kardinaal AFM, Aro A, Kark JD, et al. Association between beta-carotene and acute myocardial infarction depends on polyunsaturated fatty acid status. Arterioscler Thromb
42. Sullivan JL. Stored iron and ischemic heart disease: empirical support for a new paradigm. Circulation
43. Billman GE, Hallaq H, Leaf A. Prevention of ischemia-induced ventricular fibrillation by omega-3 fatty acids. Proc Natl Acad Sci U S A
44. Oskarsson HJ, Godwin J, Gunnar RM, Thomas JX. Dietary fish oil supplementation reduces myocardial infarct size in a canine model of ischemia and reperfusion. J Am Coll Cardiol
45. Cho SH, Choi YS. Lipid peroxidation and antioxidant status is affected by different vitamin E levels when feeding fish oil. Lipids