Aqueous Oxygen Hyperbaric Reperfusion in a Porcine Model of Myocardial Infarction

Hyperbaric oxygen (HBO) administered during reperfusion has been shown experimentally to reduce tissue injury associated with ischemia/reperfusion of a wide variety of tissues,1–8 including myocardium. Potential mechanisms include inhibition of leukocyte adherence; improvement of microvascular flow; reduction of edema; and quenching of lipid peroxide radicals.9 However, implementation of such treatment is not practical for many patients in the acute coronary setting. We recently developed a new method for delivery of oxygen, based on the infusion of aqueous oxygen (AO) into blood.10,11 Hyperbaric levels of hyperoxemia are easily achievable on a regional basis with site-specific intraarterial AO infusion. Although regional hyperbaric oxygenation is also feasible with modified use of a conventional membrane oxygenator, the bulky size, large priming volumes and biocompatibility problems,12–14 including a systemic inflammatory state associated with such devices, have hindered their study and potential use in the treatment of reperfused tissues. The miniature size of silica capillaries used to deliver AO and the physiologic oxygen/blood interface (crystalloid solution mixing with blood, rather than contact with an artificial membrane) employed with AO infusion circumvent such limitations. Accordingly, the present study tested the hypothesis that intracoronary hyperbaric reperfusion with AO reduces tissue injury in a swine model of myocardial ischemia/reperfusion. METHODS Animal model/surgical preparation. Protocols were approved by the Wayne State University School of Medicine Institutional Review Board prior to initiation of studies. All animals were housed and studied in an AALAC-approved facility. Under isofluorane anesthesia in swine weighing 34 ± 5 kg (pre-anesthesia with acepromazine 1.1 mg/kg and ketamine 22 mg/kg IM), mechanical ventilation with air was provided throughout each study. The distal segment of the left anterior descending coronary artery was occluded with an angioplasty balloon. Injection of contrast medium (Ioversol) was performed to confirm the absence of flow distal to the balloon. Heparin boluses were given to maintain an activated clotting time > 300 seconds. Lidocaine boluses (1 mg/kg) were given only prior to and during balloon occlusion. When ventricular fibrillation occurred, the animals were subjected to DC cardioversion at 300 Joules, without balloon deflation, in an attempt to restore sinus rhythm. Swine were excluded from analysis if cardioversion was unsuccessful. After removal of the balloon catheter, all animals were allowed to reperfuse passively (“autoreperfusion”) at normoxemic levels for 15 minutes. In hyperoxemia treatment groups, subselective intracoronary perfusion of hyperoxemic blood was then performed at 50 mL/minute through a non-occluding 4 French (Fr) sheath, advanced through a 7 Fr guide catheter from the left femoral artery into the proximal portion of the anterior descending branch. In one hyperoxemic treatment group, AO containing 1.0 mL O2/g of lactated Ringer’s solution (3.4 MPa O2 pressure) was infused into flowing arterial blood (withdrawn with a roller pump from a 4 Fr catheter in the right femoral artery) at 1.7–2.5 mL/minute in a small (20 mL priming volume) extracorporeal circuit to produce hyperbaric levels of hyperoxemia. In a second hyperoxemic treatment group, a small, hollow prototype polypropylene fiber oxygenator (Akzo Oxyphan; 3360 fibers, 0.41 m2 blood contact surface area, 35 mL priming volume; blood flow velocity = 0.4 cm/s at 50 mL/minute) was used in the extracorporeal circuit, rather than AO, to sub-selectively perfuse the left anterior descending coronary artery with hyperbarically oxygenated blood. Oxygen pressures (PO2s) > 100 kPa (1 bar) were achievable without bubble formation by maintaining hydrostatic pressures > O2 pressure (latter = 27 kPa above atmospheric pressure) within the circuit. A 4 MHz prototype microbubble detector (Zevex), mounted on the distal end of the circuit, and transthoracic 2-D echo were employed to confirm the absence of microbubbles. At 195 minutes of reperfusion in all animals, the coronary lumen was reoccluded at the same axial location and Evans blue dye was injected immediately thereafter into the left atrium or left ventricle to define the area at risk. The animals were then euthanized with IV pentobarbital and KCl. Animal studies. In a primary study (n = 59; fourteen were excluded for refractory ventricular fibrillation), animals were randomized after 15 minutes of autoreperfusion to one of two hyperoxemic perfusion groups or to one of two normoxemic control groups. Hyperoxemic perfusion was performed for 90 minutes with either AO infusion or the prototype HFO. Autoreperfusion alone was continued at normoxemic levels throughout reperfusion in one control group. Active perfusion of normoxemic arterial blood (“normoxemic perfusion”) was performed in a manner similar to that for the AO treatment group (50 mL/minute with same extracorporeal circuit), but without AO infusion, in a second control group. In a corollary study (n = 33; eight were excluded for refractory ventricular fibrillation), animals were randomly assigned to either one of four AO groups or to a 30-minute normoxemic active perfusion control group. Each AO group consisted of hyperoxemic perfusion for either 30, 60, 90 or 180 minutes. AO treatment was initiated after 15 minutes of normoxemic autoreperfusion and performed in a manner identical to that employed in the primary study. At 195 minutes of total reperfusion in all animals, Evans blue dye was injected and the animals were euthanized as in the primary study. Contrast studies. Serial ventriculography was performed in the primary study in a standardized manner with nonionic contrast medium (Ioversol) in all animals immediately before balloon occlusion, immediately before balloon deflation, and at 15 minutes autoreperfusion, at 105 minutes reperfusion (immediately before terminating active perfusion in the normoxemic and hyperoxemic perfusion groups), and 180 minutes reperfusion. The area-length method15 was applied to digitized ventriculographic video images (3/4´´ Sony videotape) for measurement of LVEF in swine. Coronary angiograms, performed by manual injection of Ioversol through the guide catheter, were recorded immediately prior to each left ventriculogram. TIMI grade blood flow was assessed visually from the recorded images. Postmortem/histopathologic studies. After harvesting the heart, the left ventricle of all animals was sectioned in 5-mm thick transverse sections from apex to base. The slices were weighed and photographed before and after incubation in triphenyl-tetrazolium chloride (TTC) for 15 minutes at 37 ?C.16 Planimetric measurement of areas from digitized images was performed with NIH software. The percent area of infarction was determined from the ratio of the non-red area to the total LV area and multiplied by the weight of the section to provide a gram equivalent for each slice. The sum of the values of all the slices divided by the total weight of the LV provided the percent area of infarction of the LV. The area at risk was determined in a similar manner from sections prior to TTC staining, and the area of infarction was expressed as percent (area of necrosis)/(the area at risk). The extent of hemorrhage in hearts from the primary study was assessed quantitatively by computer planimetry of digitized images of all gross sections (prior to TTC stain). A hemorrhage score was defined as the percent (area of hemorrhage)/(area at risk) in a manner similar to infarct size. From animals in the primary study, 1–2 g transmural samples of the most prominent non-TTC stained sections were frozen in liquid nitrogen for subsequent analysis of myeloperoxidase levels.17 Data analysis. Two-way, repeated, unbalanced ANOVA (SPSS 9.0) was used for comparisons of mean values over time to baseline values for data shown in Figure 1. ANOVA was used for comparison of mean values between groups at equivalent time periods (Figures 2–8). A Bonferroni correction was applied to p-values, which were considered significant at values 0.05). In the primary study, the AO treatment group had a mean coronary perfusate PO2 = 111 ± 19 kPa (834 ± 104 mmHg). The HFO hyperoxemia group had a mean PO2 = 122 ± 7 kPa (912 ± 53 mmHg) in the coronary arterial perfusate. In the corollary study, the mean PO2 of the AO hyperoxemic perfusate was 105 ± 11 kPa (787 ± 82 mmHg) and did not differ significantly between the AO groups. Angiographic/ventriculographic studies. TIMI III flow was noted in all coronary angiographic recordings, with no apparent difference between contrast medium injections performed before balloon occlusion and those performed during reperfusion. By ventriculography in the primary study, acutely after balloon coronary occlusion, a significant mean decrease in LVEF was noted compared to baseline values in all groups (p 0.05). In the normoxemic control groups, no significant improvement in LVEF was observed during the 3-hour period of reperfusion compared to values noted during coronary occlusion (p > 0.05). In the primary study, hyperoxemic reperfusion was associated with a significantly greater LVEF (p 0.05) 90 minutes after termination of hyperoxemic perfusion (58 ± 6% with AO, 59 ± 6% with HFO) from the values during hyperoxemic perfusion. No significant difference in mean values of heart rate and aortic pressure between groups in either the primary study (Figures 6–8) or the corollary study was noted at equivalent time periods (p > 0.05), except for a slight but significant increase in heart rate associated with active normoxemic perfusion, compared to AO hyperbaric reperfusion in the primary study (p 0.05). DISCUSSION Many studies support the hypothesis that HBO reduces infarct size and improves left ventricular function in the setting of acute myocardial infarction18–23 and reperfusion.1,2,24–26 However, implementation of HBO is not feasible for many acutely ill patients and may be associated with undesirable side effects.27–29 Moreover, during HBO, arterial oxygen tensions cannot be controlled precisely, and, when pulmonary function is compromised, they may not be elevated to hyperbaric levels. We show here for the first time that catheter-based reperfusion of the infarct artery with hyperbarically oxygenated blood can be performed safely and is associated with acute improvement in LVEF. Perhaps more importantly, we also show that AO hyperbaric reperfusion is associated with a significant acute reduction in tissue injury, as reflected in mean AN/AR, postmortem hemorrhage score and tissue myeloperoxidase levels compared to normoxemic reperfusion, whether active (roller pump) or passive (autoreperfusion). In contrast, tissue injury was not reduced with hyperbaric reperfusion with the hollow fiber oxygenator (HFO), despite functional improvement. More studies will be required to clarify the reasons for the differences between the two methods of oxygenation. However, membrane oxygenators are known to cause a systemic inflammatory response,12–14 which may contribute to organ failure. Oxygenation by AO infusion is produced by simple mixing of the crystalloid solution with blood. The elimination of the need for a membrane requiring blood contact over a broad foreign surface may be advantageous. In addition, physiologic blood velocities are employed throughout the small AO circuit, in contrast to the potentially deleterious low velocities used in membrane oxygenators.30 Potential benefits of AO hyperbaric reperfusion. The reduction of myocardial myeloperoxidase levels in our study is consistent with the observation that HBO has been shown to inhibit leukocyte adhesion to venules in reperfused skeletal muscle, perhaps by inhibiting B2 integrins.31 Buras et al. recently demonstrated in endothelial cell culture that HBO inhibits expression of ICAM-1 induced by hypoxia and hypoglycemia, as well as inhibits leukocyte adhesion, perhaps as a result of induction of eNOS.32 HBO has also been shown in a variety of experimental and clinical studies to reduce tissue edema.33,34 The mechanism(s) responsible for this effect have not been elucidated. Ischemia-induced impairment of the permeability barrier function of the endothelium35 may be reduced by HBO. Another possibility is that addition of oxygen as a solute in plasma may produce an osmotic gradient that removes water from the extravascular compartment, as suggested by Hills.36 Focal regions of microvascular impairment and ischemia may be present in reperfused tissues,37–41 even if TIMI III grade blood flow has been restored.42,43 During reperfusion, reperfusion microvascular ischemia (i.e., low flow ischemia during reperfusion) may fail to resolve or may become exacerbated as a result of the return of physiologic perfusion pressures,44 oxygen-radical generation,45,46 and activation of leukocytes and platelets by hypoxia and low flow.47–52 Improved oxygen delivery during reperfusion by a variety of approaches has been shown to have beneficial effects on microvascular flow.53–55 Reactive oxygen species (ROS). Despite the clinical benefits of reperfusion, ROS generation upon the reintroduction of oxygen remains a potential concern.45,46 However, the relationship between oxygen partial pressure in reperfused tissues and ROS production is likely to be quite complex. Indeed, severe hypoxia accelerates lipid peroxidation under certain conditions both in vitro56 and in vivo,57 and Thom demonstrated that HBO inhibits lipid peroxidation in a rat model of hypoxic brain injury.58 In addition, the possibility exists in our model that ROS production occurs in association with reperfusion microvascular ischemia, but is reduced by attenuation of ischemia with AO hyperbaric reperfusion. Nevertheless, our observation that AN/AR is not reduced by a 3-hour period of AO hyperbaric reperfusion, in contrast to the beneficial effect of a 90-minute period of treatment, suggests that oxygen toxicity may occur with longer periods of treatment. Study limitations. Only acute effects of AO hyperbaric reperfusion were studied. The chronic effects of this approach may require a different animal model in view of the rapid growth of juvenile domestic swine. However, an FDA Phase I clinical trial of intracoronary AO hyperbaric reperfusion following stent placement in the setting of acute myocardial infarction has recently been completed.59 The results are consistent with the hypothesis that AO hyperbaric reperfusion facilitates recovery of myocardium. A Phase II randomized trial, planned for the near future, will be performed to formally test this hypothesis. Conclusion. Intracoronary hyperbaric oxygen reperfusion acutely improves left ventricular function compared to normoxemic reperfusion. Moreover, AO hyperbaric reperfusion attenuates myocardial injury acutely in this model. However, hyperbaric reperfusion with a membrane oxygenator fails to attenuate myocardial injury, perhaps as a result of inflammatory cell reactions initiated within the device. Acknowledgment. This work was supported by grants from TherOx, Inc. and the National Institute of Health (HL56436). The authors are indebted to Alice Jiang, MD, and Xiaojun Wu, BS, for their expert technical assistance.

1. Thomas MP, Brown LA, Sponseller DR, et al. Myocardial infarct size reduction by the synergistic effect of hyperbaric oxygen and recombinant tissue plasminogen activator. Am Heart J 1990;120:791–800. 2. Sterling DL, Thornton JD, Swafford A, et al. Hyperbaric oxygen limits infarct size in ischemic rabbit myocardium in vivo. Circulation 1993;88:1931–1936. 3. Haapaniemi T, Nylander G, Sirsjo A, Larsson J. Hyperbaric oxygen reduces ischemia-induced skeletal muscle injury. Plast Reconstr Surg 1996;97:602–607. 4. Chang CF, Niu KC, Hoffer BJ, et al. Hyperbaric oxygen therapy for treatment of post-ischemic stroke in adult rats. Exp Neurol 2000;166:298–306. 5. Sunami K, Takeda Y, Hashimoto M, Hirakawa M. Hyperbaric oxygen reduces infarct volume in rats by increasing oxygen supply to the ischemic periphery. Crit Care Med 2000;28:2831–2836. 6. Kolski JM, Mazolewski PJ, Stephenson LL, et al. Effect of hyperbaric oxygen on testicular ischemia-reperfusion injury. J Urol 1998;160:601–604. 7. Zamboni WA, Roth AC, Russell RC, Smoot EC. The effect of hyperbaric oxygen on reperfusion of ischemic axial skin flaps: A laser Doppler analysis. Ann Plast Surg 1992;28:339–341. 8. Bitterman H, Bitterman N, Melamed Y, Cohen L. Effects of hyperbaric oxygen in circulatory shock induced by splanchnic artery occlusion and reperfusion in rats. Can J Physiol Pharmacol 1989;67:1033–1037. 9. Buras J. Basic mechanisms of hyperbaric oxygen in the treatment of ischemia-reperfusion injury. Int Anesthesiol Clin 2000;38:91–109. 10. Spears JR, Wang B, Wu X, et al. Aqueous oxygen: A highly O2-supersaturated infusate for regional correction of hypoxemia and production of hyperoxemia. Circulation 1997;96:4385–4391. 11. Brereton GJ, Crilly RJ, Spears JR. Nucleation in small capillary tubes. Chemical Physics 1998;230:253–265. 12. Zahler S, Massoudy P, Hartl H, et al. Acute cardiac inflammatory responses to post-ischemic reperfusion during cardiopulmonary bypass. Cardiovasc Res 1999;41:722–730. 13. de Vroege R, Wagemakers M, te Velthuis H, et al. Comparison of three commercially available hollow fiber oxygenators: Gas transfer performance and biocompatibility. ASAIO J 2001;47:37–44. 14. Ashraf S, Tian Y, Cowan D, et al. Release of proinflammatory cytokines during pediatric cardiopulmonary bypass: Heparin-bonded versus non-bonded oxygenators. Ann Thorac Surg 1997;64:1790–1794. 15. Sandler H, Dodge HT. The use of single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J 1968;75:325–334. 16. Fishbein MC, Meerbaum S, Rit J, et al. Early phase acute myocardial infarct size quantification: Validation of the triphenyl tetrazolium chloride tissue enzyme staining technique. Am Heart J 1981;101:593–600. 17. Trush MA, Egner PA, Kensler TW. Myeloperoxidase as a biomarker of skin irritation and inflammation. Food Chem Toxicol 1994;32:143–147. 18. Smith G, Lawson DA. Experimental coronary arterial occlusion: Effects of the administration of oxygen under pressure. Scot Med J 1958;3:346–350. 19. Peter RH, Rau RW, Whalen RE, et al. Effects of hyperbaric oxygenation on coronary artery occlusion in pigs. Circ Res 1966;18:89–96. 20. Chardack WM, Gage AA, Federico AJ, et al. Reduction by hyperbaric oxygenation of the mortality from ventricular fibrillation following coronary artery ligation. Circ Res 1964;15:497–502. 21. Cameron AJ, Hutton I, Kenmure AC, Murdoch WR. Haemodynamic and metabolic effects of hyperbaric oxygen in myocardial infarction. Lancet 1966;2:833–837. 22. Ashfield R, Gavey CJ. Severe acute myocardial infarction treated with hyperbaric oxygen. Report on forty patients. Postgrad Med J 1969;45:648–654. 23. Thurston JGB, Greenwood TW, Bending MR, et al. A controlled investigation into the effects of hyperbaric oxygen on mortality following acute myocardial infarction. Quart J Med 1973;42:751–770. 24. Tjarnstrom J, Holmdahl L, Falk P, et al. Effects of hyperbaric oxygen on expression of fibrinolytic factors of human endothelium in a simulated ischaemia/reperfusion situation. Scand J Clin Lab Invest 2002;61:539–545. 25. Shandling AH, Ellestad MH, Hart GB, et al. Hyperbaric oxygen and thrombolysis in myocardial infarction: The “HOT MI” pilot study. Am Heart J 1997;134:544–550. 26. Stavitsky Y, Shandling AH, Ellestad MH, et al. Hyperbaric oxygen and thrombolysis in myocardial infarction: The “HOT MI” randomized multicenter study. Cardiology 1998;90:131–136. 27. Berry JM, Doursout MF, Butler BD. Effects of hyperbaric hyperoxia on cardiac on regional hemodynamics in conscious dogs. Aviat Space Environ Med 1998;69:761–765. 28. Stuhr LE, Bergo GW, Tyssebotn I. Systemic hemodynamics during hyperbaric oxygen exposure in rats. Aviat Space Environ Med 1994;65:531–538. 29. Amin HM, Hakim TS, Camporesi EM. Hematological alterations after acute exposure to hyperbaric oxygen in rats. Clin Exp Pharmacol Physiol 1995;22:21–27. 30. Gartner MJ, Wilhelm CR, Gage KL, et al. Modeling flow effects on thrombotic deposition in a membrane oxygenator. Artif Organs 2000;24:29–36. 31. Chen Q, Banick PD, Thom SR. Functional inhibition of rat polymorphonuclear leukocyte B2 integrins by hyperbaric oxygen is associated with impaired cGMP synthesis. J Pharmacol Exp Ther 1996;276:929–933. 32. Buras JA, Stahl GL, Svoboda KK, Reenstra WR. Hyperbaric oxygen downregulates ICAM-1 expression induced by hypoxia and hypoglycemia: The role of NOS. Am J Physiol Cell Physiol 2000;278:C292–C302. 33. Skyhar MJ, Hargens AR, Strauss MB, et al. Hyperbaric oxygen reduces edema and necrosis of skeletal muscle in compartment syndromes associated with hemorrhagic hypotension. J Bone Joint Surg 1986;68A:1218–1224. 34. Carl UM, Feldmeier JJ, Schmitt G, Hartmann KA. Hyperbaric oxygen for late sequelae in women receiving radiation after breast-conserving surgery. Int J Radiat Oncol Biol Phys 2001;49:1029–1031. 35. Ogawa S, Gerlach H, Esposito C, et al. Hypoxia modulates the barrier and coagulant function of cultured bovine endothelium. Increased monolayer permeability and induction of procoagulant properties. J Clin Invest 1990;85:1090–1098. 36. Hills BA. A role for oxygen-induced osmosis in hyperbaric oxygen therapy. Med Hypotheses 1999;52:259–263. 37. Jeremy RW, Links JM, Becker LC. Progressive failure of coronary flow during reperfusion of myocardial infarction: Documentation of the no reflow phenomenon with positron emission tomography. J Am Coll Cardiol 1990;16:695–704. 38. Ince C, Vink H, Wieringa PA, et al. Heterogeneous NADH fluorescence during post-anoxic reactive hyperemia in saline perfused rat heart. Adv Exp Med Biol 1990;277:477–482. 39. Weinstein H, Reinhardt CP, Leppo JA. Direct detection of regional myocardial ischemia with technetium-99m nitroimidazole in rabbits. J Nucl Med 1998;39:598–607. 40. Rochitte CD, Lima JA, Bluemke DA, et al. Magnitude and time course of microvascular obstruction and tissue injury after acute myocardial infarction. Circulation 1998;98:1006–1014. 41. Claeys MJ, Bosmans J, Veenstra L, et al. Determinants and prognostic implications of persistent ST-segment elevation after primary angioplasty for acute myocardial infarction: Importance of microvascular reperfusion injury on clinical outcome. Circulation 1999;99:1972–1977. 42. Wu KC, Zerhouni EA, Judd RM, et al. Prognostic significance of microvascular obstruction by magnetic resonance imaging in patients with acute myocardial infarction. Circulation 1998;97:765–772. 43. Gibson CM, Cannon CP, Murphy SA, et al. Relationship of TIMI myocardial perfusion grade to mortality after administration of thrombolytic drugs. Circulation 2000;101:125–130. 44. Sato H, Jordan JE, Zhan ZQ, et al. Gradual reperfusion reduces infarct size and endothelial injury but augments neutrophil accumulation. Ann Thorac Surg 1997;64:1099–1107. 45. McCord JM. Oxygen-derived radicals: A link between reperfusion injury and inflammation. Fed Proc 1987;46:2402–2406. 46. Bagchi D, Wetscher GJ, Bagchi M, et al. Interrelationship between cellular calcium homeostasis and free radical generation in myocardial reperfusion injury. Chem Biol Interact 1997;104:65–85. 47. Pinsky DJ, Yan SF, Lawson C, et al. Hypoxia and modification of the endothelium: Implications for regulation of vascular homeostatic properties. Semin Cell Biol 1995;6:283–294. 48. Al-Haboubi HA, Tomlinson DR, Ward BJ. The influence of hypoxia on transvascular leakage in the isolated rat heart: Quantitative and ultrastructural studies. J Physiol (Lond) 1995;482:157–166. 49. Mazzoni MC, Borgstrom P, Warnke KC, et al. Mechanisms and implications of capillary endothelial swelling and luminal narrowing in low-flow ischemias. Int J Microcirc Clin Exp 1995;15:265–270. 50. Ritter LS, McDonagh PF. Low-flow reperfusion after myocardial ischemia enhances leukocyte accumulation in coronary microcirculation. Am J Physiol 1997;273:H1154–H1165. 51. Kubes P. The role of shear forces in ischemia/reperfusion-induced neutrophil rolling and adhesion. J Leukoc Biol 1997;62:458–464. 52. Baudry N, Danialou G, Boczkowski J, et al. In vivo study of the effect of systemic hypoxia on leukocyte-endothelium interactions. Am J Respir Crit Care Med 1998;158:477–483. 53. Sirsjo A, Lehr HA, Nolte D, et al. Hyperbaric oxygen treatment enhances the recovery of blood flow and functional capillary density in postischemic striated muscle. Circ Shock 1993;40:9–13. 54. Zamboni WA, Roth AC, Russell RC, et al. Morphologic analysis of the microcirculation during reperfusion of ischemic skeletal muscle and the effect of hyperbaric oxygen. Plast Reconstr Surg 1993;91:1110–1123. 55. Pickelmann S, Nolte D, Leiderer R, et al. Attenuation of post-ischemic reperfusion injury in striated skin muscle by diaspirin-cross-linked Hb. Am J Physiol 1998;275:H361–H368. 56. De Groot H, Noll T. The role of physiological oxygen partial pressures in lipid peroxidation. Theoretical considerations and experimental evidence. Chem Phys Lipids 1987;44:209–226. 57. Minyailenko TD, Pozharov VP, Seredenko MM. Severe hypoxia activates lipid peroxidation in the rat brain. Chem Phys Lipids 1990;55:25–28. 58. Thom SR, Elbuken ME. Oxygen-dependent antagonism of lipid peroxidation. Free Radic Biol Med 1991;10:413–426. 59. Dixon SR, O’Neill WW, Bartorelli AL, et al. Global and regional left ventricular function after intracoronary hyperoxemic perfusion with the aqueous oxygen system in acute myocardial infarction. J Am Coll Cardiol (in press).