Ibuprofen-MediatedInfarct Size Reduction: Effects on Regional Myocardial Function in Canine Myocardial Infarction PHILIP C. KIRLIN, MD,* JOSEPH L. ROMSON, PhD,+ BERTRAM PITT, MD, GERALD D. ABRAMS, MD, M. ANTHONY SCHORK, PhD, and BENEDICT R. LUCCHESI, PhD, MD Normal, marginal, and central ischemic regional myocardial function were evaluated in a canine model of myocardial infarction during 90 minute left circumflex coronary artery occlusion in 25 anesthetized dogs randomly assigned to intravenous ibuprofen infusion (n = 13,5.36 mg/kg/h beginning 1 hour before occlusion) or vehicle solution as control (n = 12) and in 15 conscious, unsedated dogs 48 and 72 hours after 90 minute circumflex artery occlusion followed by reperfusion (ibuprofen, 5.36 mg/kg/h by intravenous infusion over 7 hours beginning 1 hour before occlusion, n = 7; or vehicle solution infusion as control, n = 8). Miniature subendocardial sonomicrometer crystal pairs were used to calculate left ventricular regional end-diastolic segment length, end-systolic segment length, and regional percent systolic shortening. Infarct size was estimated in 72 hour animals by a postmortem dual perfusion technique using triphenyltetrazolium histochemical dye and Evan’s blue dye for determination of infarct area, risk area, and area not at risk. Ibuprofen treatment significantly reduced infarct size expressed as percent of risk area (mean f standard deviation of 44.6 f 18 versus 64.4 f 16% for control dogs, p <0.05) but it did not improve normal, marginal, or ischemic region end-diastolic length, end-systolic length or percent systolic shortening during coronary occlusion in anesthetized dogs or after reperfusion in conscious animals at 48 and 72 hours, and it did not enhance inotropic reserve at 72 hours in conscious animals. During 90 minute circumflex occlusion in anesthetized dogs, ibuprofen was associated with increases in systemic arterial pressure and worsened ischemic regional percent systolic shortening. Thus, ibuprofen does not improve normal, marginal, or ischemic zone regional myocardial function during acute ischemia or 48 or 72 hours after myocardial reperfusion despite a significant reduction of infarct size. Reducing the extent of ischemic injury has been proposed as a rational goal in the management of acute myocardial infarction.’ Initial attempts to achieve this goal centered primarily on hemodynamic interventions which lessened myocardial oxygen consumption or improved blood flow or oxygen delivery to ischemic myocardium; more recent investigations have evaluated a variety of agents whose primary effect is not hemodynamic but rather anti-inflammatory. The hypothesis supporting the use of these agents is that a significant determinant of ultimate ischemic injury is the inflammatory process itself. Among the many agents currently under evaluation is ibuprofen, a nonsteroidal antiinflammatory agent clinically employed for rheumatic disease.* Experimental studies have demonstrated a From the Departments of Pharmacology, Internal Medicine (Cardiology myocardial protective effect of ibuprofen in a variety Division), and Pathology, University of Michigan Medical School and of animal models,3-7 presumably attributable to its the Department of Biostatistics, University of Michigan School of Public antiinflammatorv actions which affect arachidonic acid Health, Ann Arbor, Michiaan. This studv was suooorted in oart bv National Institutes of Health Research Grant HL19782-03 from zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA ihe National metabolism and-consequent physiologic events.8 Heart, Lung, and Blood Institute, Bethesda, Maryland, and by a grantHowever, few studies on myocardial infarct size rein-aid from the Upjohn Company, Kalamazoo, Michigan. Manuscript duction involving either hemodynamic or antiinflamreceived November 30, 1981; revised manuscrrpt received April 27, matory interventions have addressed the issue of the 1982, accepted May 6, 1982. functional integrity of the salvaged mvocardium or the Dr Kirlin was the recipient of a Michigan Heart Association Re_ search Fellowship Award, Lathrop Village, Michigan. effect of these agents on the function of myocardium + Dr Romson was the recipient of a Pharmaceutical Manufacturer’s removed from the area of ischemic injury. In order to Association Fellowship, Washington, D.C. assess the myocardial protective effect of ibuprofen on Address for reprints: Benedict R. Lucchesr, MD, Department of regional myocardial function, the following investigaPharmacology, M6322 Medical Science Buildina I, Universitv of Michigan Medicaikhool. Ann Arbor, Michigan 48159. tion was undertaken using implantable subendocardial l October 1982 The American Journal of CARDIOLOGY Volume 50 849 IBUPROFEN AND REGIONAL MYOCARDIAL FUNCTION-KIRLIN ET AL. sonomicrometer crystals to measure segment length changes in normal, marginal and ischemic regions9 in a canine model of left circumflex coronary artery occlusion and reperfusion. In this study, regional myocardial function was evaluated in open-chest, anesthetized dogs during coronary artery occlusion and in conscious, unsedated dogs 48 and 72 hours after circumflex occlusion and subsequent myocardial reperfusion. tricular wall to straddle the border between cyanotic and noncyanotic zones. The marginal area was defined in the present study as a region of intermediate function between normal and ischemic zones.9 Normal zone sonomicrometer crystals were implanted on the anterior left ventricular wall in the region supplied by the left anterior descending coronary artery well removed from the ischemic area as determined by the zone of cyanosis. The crystals were implanted in a plane perpendicular to the base-apex cardiac axis and secured by an epicardial suture around the connecting wires as they emerged from the myocardium. Repeated calibrations at 1 ~LS intervals were obtained to eliminate variation due to signal Methods drift. Regional Myocardial Function During Coronary In addition, a small left neck incision was made for insertion Artery Occlusion zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA of a polyethylene carotid artery catheter for blood pressure measurement and an external jugular catheter for infusion of Coronary occlusion: In order to assess the acute effects ibuprofen or control solution. The carotid artery catheter was of ibupiofen on regional function during evolving myocardial connected to a Statham P23DC pressure transducer. Mean injuiry due to coronary artery occlusion, 25 dogs were studied arterial pressure was calculated as diastolic pressure plus one in the open-chest anesthetized state. Healthy male mongrel third of the pulse pressure. dogs weighing 10.2 to 18.9 kg (mean f standard deviation 13.9 Ibuprofen infusion: Crystalline ibuprofen (sodium salt) f 1.9) were anesthetized with 30 mg/kg intravenous sodium was dissolved in 0.2 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPO M sodium carbonate and adjusted to a pH pentobarbital and mechanically ventilated on room air by a of 8 by 1.0 N HCl. Solutions were mixed fresh before each Harvard animal respirator (Harvard Apparatus, Millis, experiment and diluted to 50 ml in physiologic saline solution. Massachusetts) using an endotracheal tube. A left 5th interThe ibuprofen dose employed was 5.36 mg/kg/h (7.14 ml/h of costal space thorocotomy was performed and the pericardium ibuprofen solution) by continuous intravenous infusion beincised and sutured to form a pericardial cradle. The left cirginning 1 hour before occlusion and continuing throughout cumflex coronary artery was isolated distal to the left atria1 occlusion (90 minutes). This dose is similar to that used in a appendage proximal to major branches. A zone of cyanosis was previous investigation6 with the exception of initiation of the determined to be at least 9 cm2 during transient (10 to 15 infusion before occlusion in the present study. Vehicle solution second) coronary occlusion. Three pairs of commercially prepared in an identical manner without ibuprofen served as manufactured miniature (3 mm) piezoelectric crystals (800 a control. A Harvard infusion pump was used to provide a kHz frequency in radial mode with a resolution capability of constant infusion rate. Animals were randomly assigned to 0.1 to 0.5 mm) connected to an oscilloscope sonomicrometer their treatment groups (13 ibuprofen, 12 control). A lead II (Norland Instruments sonomicrometer crystals and NI-202 electrocardiogram was monitored throughout the experisonomicrometer) were implanted 1 to 2 cm apart in the left ment. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJ ventricular subendocardium (Fig. 1). One crystal pair (central After instrumentation, visible epicardial arterial anastoischemic zone) was implanted in the posterolateral left venmoses between the left anterior descending and circumflex tricular surface in the center of the cyanotic zone as detercoronary arteries were ligated to eliminate potentially large mined by the transient circumflex occlusion, and a 2nd pair collateral routes of blood flow.‘O A subtotal left circumflex (marginal area) was implanted in the anterolateral left vencoronary artery stenosis was produced just distal to the left atria1 appendage by tying 1-O silk suture around both the artery and an 18 gauge needle followed by immediate needle removal; subsequent total circumflex artery occlusion was performed after 10 minutes with a total duration of occlusion of 90 minutes. Three mg/kg of intravenous lidocaine and 3 mg/kg of intramuscular lidocaine were administered at the time of coronary artery occlusion as antiarrhythmic therapy.” Approximately one fourth of the animals developed ventricular fibrillation during coronary occlusion, which was readily converted to sinus rhythm within 30 to 60 seconds by low energy (10 to 20 J) direct. current cardioversion using epicardial paddles. No myocardial dysfunction has been reported at these energy levels.12 Regional myocardial segment length: Measurement of regional myocardial segment length was obtained by converting sonomicrometer crystal transit time in microseconds to distance in millimeters using a known value for the velocity of ultrasound in myocardium at the frequency employed: Distance (mm) = transit time (I.CS)X ultrasound velocity in myocardium (1.58 mm/ps).l3 Before occlusion, end-diastolic length was defined in each segment as the maximal crystal separation in millimeters before the onset of the carotid artery systolic pressure rise, and end-systolic length was defined as FIGURE 1. Diagram of subendocardial sonomicrometer crystal the minimal crystal separation in millimeters before the diplacement in normal, marginal, and central ischemic left ventricular erotic notch of the carotid artery pressure tracing. Nonsimregions. LCX = left circumflex coronary artery. See text for discusultaneous end-diastolic or end-systolic timing in marginal or sion. a50 October 1982 The American Journal of CARDIOLOGY Volume 50 IBUPROFEN AND REGIONAL MYOCARDIAL FUNCTION--KIRLIN ET AL. lengths were obtained in the conscious, unsedated state with ischemic segments in relation to normal segments was noted the animal standing upright in a nylon sling. In order to assess before occlusion and used as a reference to determine endresidual ventricular contractile function, inotropic reserve was diastole and end-systole in the presence of dyskinesia during also tested at 72 hours after coronary occlusion-reperfusion, coronary artery occlusion. Measurement of absolute end5 minutes after administration of the beta-adrenergic stimdiastolic and end-systolic length in millimeters were obtained ulant prenalterol(30 pg/kg intravenous bolus).‘* and regional percent systolic shortening defined as end-diaDetermination of infarct size and risk area: Animals stolic minus end-systolic length divided by end-diastolic were then reanesthetized with sodium pentobarbital, the chest length X100%. Subendocardial crystal placement was conwas reopened, and the hearts were electrically fibrillated and firmed by postmortem examination. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA rapidly excised for immediate postmortem in vitro determiA Grass model 7 polygraph was used to record measurenations of infarct size and risk area by a dual perfusion techments at a paper speed of 100 mm/s. Determination of carotid nique using 4% triphenyltetrazolium chloride mixed with artery pressure, heart rate, and regional segment lengths and phosphate buffer for histochemical staining of dehydrogenase percent systolic shortening were obtained before and at the enzyme and 0.5% Evan’s blue flow-dependent staining as end of 90 minutes of coronary occlusion. All measurements described in detail elsewhere. 7~s Briefly, this method involves were obtained at end-expiration and when the animal was free simultaneous selective left circumflex perfusion with triof arrhythmias. phenyltetrazolium and nonselective retrograde aortic perfuRegional Myocardial Function 48 and 72 Hours After sion with Evans blue dye at physiologic pressure (100 mg Hg), Coronary Artery Occlusion-Reperfusion with the resultant Evan’s blue staining tissue identifying the area not at risk of ischemic injury and the remaining tissue The effect of ibuprofen on regional myocardial function and representing the area at risk with red triphenyltetrazolium infarct size was evaluated in conscious, unsedated dogs after staining tissue identifying the viable, noninfarcted region, and left circumflex coronary artery occlusion with subsequent the remaining tissue (unstained) representing the area of inmyocardial reperfusion in the anesthetized state. Fifteen healthy male mongrel dogs weighing 13.6 + 1.5 kg (range 10.2 farction. After staining, the heart was sliced from apex to base in 5 mm thick transverse sections and the atria and right to 15.8) were used. Instrumentation, baseline regional function ventricle were excised. Left ventricular weight was recorded measurements, and 90 minute circumflex occlusion followed and left ventricular slices were traced on transparent plastic by reperfusion were performed using aseptic technique in sheets for hand planimeter determination of left ventricular pentobarbital anesthetized, open-chest dogs as in the acute area, area at risk, and area of infarction. Infarct size was occlusion study. Animals were randomly assigned to either ibuprofen (n = 7) or vehicle as control (n = 8). Solutions were measured as percent of area at risk and percent of total left prepared as before and administered intravenously by Harventricular area. The anatomic perfusion bed of an occluded vard infusion pump over 7 hours beginning 1 hour before occoronary has been shown to correlate well with the area at risk clusion with an ibuprofen dose of 5.36 mg/kg/h (7.14 ml/h of of infarction in experimental canine coronary artery occluibuprofen solution) or with an identical volume of vehicle as sion.lO Subendocardial crystal placement was also confirmed at postmortem evaluation. a control. Ischemic zone sonomicrometer crystals were associated with greater postoperative mortality in preliminary Light microscopic evaluation of hematoxylin and eosin studies and therefore implanted in only 4 animals in each stained left ventricular slices was performed in a blinded group. As in the acute occlusion study, a critical circumflex manner on 72 hour specimens from all animals in each group stenosis was placed 10 minutes before total occlusion using to assess the morphologic integrity of myocardial cells and the an 18 gauge needle and 1-O silk suture. A Carolina Medical inflammatory response. Electronics electromagnetic flow probe was placed around the Statistical methods: All results are expressed as mean f circumflex artery proximal to the subtotal occlusion to doc1 standard deviation. Statistical analysis was performed by ument a critical stenosis (defined as a stenosis causing at least a computerized statistical program of the University of a 50% reduction in peak reactive hyperemic flow after 10 Michigan Statistical Research Laboratory (Michigan Interseconds of coronary occlusion without altering basal flow or active Data Analysis System). Two factor analysis of variance sonomicrometer crystal segment length). This subtotal critical for repeated measures (profile analysis)i6 was employed for stenosis was left in place throughout the remainder of the the same variable measured more than twice, and paired or experiment (until the animals were killed at 72‘hours). zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA unpaired t test was performed as appropriate on data meaCoronary occlusion and reperf!sion: Total left cirsured only twice. A probability value (p) less than 0.05 was cumflex occlusion was performed using soft plastic tape for considered significant. 90 minutes and the occlusion was slowly released over 5 to 10 minutes. Intravenous lidocaine, 3 mg/kg, was administered Results zyxwvutsrqponmlkjihgfedcbaZYXW on reperfusion. Reperfusion occurred in all animals as docuRegional myocardial function during coronary mented by 3 or more of the following criteria: (1) the onset of artery occlusion: In anesthetized, open-chest dogs 90 ventricular arrhythmias within 15 to 30 minutes of occlusion release; (2) a decrease in marginal and ischemic region endminute circumflex coronary artery occlusion resulted diastolic and end-systolic segment length on release of the in a marked alteration of contractile function maniocclusion; (3) at least 5 ml/min of circumflex artery flow on fested by greatly diminished marginal and ischemic occlusion release documented by electromagnetic flow probe; segment systolic shortening and increased end-diastolic and (4) the absence of visible coronary artery thrombosis at and end-systolic segment lengths in all 3 regions, in72 hours when the animal was killed. dicative of ischemia-associated ventricular dilations After reperfusion, the chest was closed and the animals (Fig. 2, Table I). Ibuprofen administration did not dewere allowed to recouer. Intramuscular ampicillin, 6 mg/kg, crease these segment length increases or improve syswas administered twice on the day of surgery and daily tolic shortening in any region during coronary artery thereafter. No recordings were obtained until 48 hours after occlusion. In the central ischemic zone, ibuprofensurgery because of frequent ventricular arrhythmias. At 48 and 72 hours, recordings of sonomicrometer crystal segment treated animals developed more severe impairment of October 1992 The American Journal of CARDIOLOGY Volume 50 951 IBUPROFEN AND REGIONAL MYOCARDIAL FUNCTION-KIRLIN ET AL I Control,n=8-I2 QZJ Ibuprofen (37.5 mg/kg,IV), n=ll-I3 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCB T T FIGURE 2. Effect of ibuprofen on regional myocardial function after 90 minute left circumflex coronary artery occlusion. Marginal and ischemic zone percent systolic shortening was reduced in both groups compared with preocclusion (paired t test), and ibuprofen animals had significantly worse systolic shortening than controls in the ischemic zone at the end of the 90 minute occlusion (unpaired t test) ” Pre-Ocdusion SO minute Occlusion Pre-Occlusion SO minute Occlusion NORMAL ZONE MARGINAL AREA lp ( .Oi fOcctusm vs. Pre-&&S&I, **PC 123 (tbuprofm g poked t-test) Controt, unpolred t-test) systolic shortening with paradoxic systolic expansion (Table II, Fig. 4 to 6). In the conscious, unsedated state 48 and 72 hours after occlusion with myocardial reper(dyskinesia). Heart rate was not altered by ibuprofen fusion, both marginal and ischemic zone systolic during coronary occlusion; however, mean carotid artery shortening were impaired despite partial recovery, while pressure rose significantly with a rise in pressure-rate normal zone systolic function remained relatively unproduct (the product of mean arterial pressure and altered. Ibuprofen treatment did not improve systolic heart rate divided by 100) (Table I). zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA function or decrease ventricular end-diastolic or endRegional myocardial function 48 and 72 hours systolic segment lengths in normal, marginal, or ischeafter coronary artery occlusion-reperfusion: Serial mic regions at 48 hours or 72 hours before or after inochanges in regional myocardial function in a control animal before, during, and up to 72 hours after coronary tropic stimulation. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPON A significant reduction in the extent of ischemic artery occlusion-reperfusion are shown in Figure 3. injury as measured by percent of the area at risk inSimilar baseline regional myocardial function was farcted was noted in ibuprofen-treated animals (44.6 f present in ibuprofen-treated and control animals in the 18%) compared with control animals (64.4 f 169’0, p anesthetized, open-chest state before coronary occlusion TABLE I Effects of Ibuprofen During Coronary Occlusion Normal Zone Heart Rate (beats/min) Baseline 90 minute coronary occlusion Mean Arterial Pressure (mm Hg) Pressure-Rate Product Control Ibuprofen Control Ibuprofen Control 167 f25 171 *20 159 f17 174 fia 111 f14 106 f14 120 f20 137 f15’ la7 f46 178 *30 Ibuprofen 194 650 239 f44’ End-Diastolic Length (mm) Control Ibuprofen Control 12.6 f3.1 14.6 f4.0 14.7 f2.4 17.1 f3.0 9.7 f3.2 11.4 f3.9 Control Baseline 90 minute coronary occlusion 15.0 f2.9 la.1 f2.a Ibuprofen 13.9 f5 2 17.2 f6.0 End-Systolic Length (mm) % Systolic Shortening End-Diastolic Length (mm) October 1982 11.5 62.3 13.3 *3 1 Control Ibuprofen 24.3 f9.7 23.0 f9.9 21.4 f9.2 22.5 410.0 End-Systolic Length (mm) % Systolic Shortening Control Ibuprofen Control Ibuprofen Control Ibuprofen Control Ibuprofen Control Ibuprofen 12.5 43.0 17.8 f3.1 12.1 f5.3 16.8 f5.9 17.2 f6.3 1.7 f4.4 14.3 f9.7 2.1 f5.9 15.6 f4.6 la.4 15.4 17.1 f4.6 19.7 f5.2 13 6 442 la.1 f49 15.3 f4.6 20.0 f5.1 13.3 *4.2 1.3 f2.7 11.6 f5.5 -1.5 f2 37 * p <0.003; t p <0.03 (both unpaired t test). Pressure rate product = mean arterial pressure X heart rate divided by 100. Values are mean f 852 Ibuprofen % Systolic Shortening lschemic Zone Marginal Zone End-Diastolic Length (mm) End-Systolic Length (mm) The American Journal of CARDIOLOGY Volume 50 1 standard deviation IBUPROFEN AND REGIONAL MYOCARDIAL FUNCTION-KIRLIN ET AL. FKZURE 3. Serial changes in normal, marginal, and central ischemic segment length in a control animal before and at the end of 90 minute circumflex occlusion and 48 and 72 hours after reperfusion, and with inotropic stimulation (intravenous prenalterol, 30 pg/kg) at 72 hours. Acute effects of circumflex occlusion on regional myocardial function and evolutionary electrocardiographic changes of infarction are demonstrated. BP = blood pressure; EKG = lead It electrocardiogram; LCX = left circumflex coronary artery. Es50 I m I 0 :! contml, n=a eiza lbu~ofea (37.5 mg/kg, IV), n =7 m m CentId, n=7 Ibuprofen(37.5 mg/kp, IV), n ~7 il*SD tw40 8 8 T T zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 0 Rr-occkxim Pro-Ocdudon 40 bra 72 hrr 72 bra with klotmpii AQmt FIGURE 5. Effect of ibuprofen on marginal area regional function (as in Figure 4). No significant differences between groups were present 4ahrs zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA RhrS 72 hrs wifh (analysis of variance for repeated measures). lmtropic &ant FIGURE 4. Effect of ibuprofen on normal zone regional myocardial function before circumflex occlusion and in conscious, unsedated animals at 48 and 72 hours and with inotropic stimulation at 72 hours. No significant differences were present between groups (analysis of variance for repeated measures). IV = intravenous; SD = standard deviation. a pg control, n=4 Ibuprofen(37.5 mp/kg, IV), n= 4 ‘O vr 0 Pm-occ& 46 hrr 72 hn 72 hrs with lnolmpicAgent FIGURE 6. Effect of ibuprofen on central ischemic zone regional function (as in Figure 4). No significant differences between groups were present (analysis of variance for repeated measures). FIGURE 7. Effect of ibuprofen on infarct size. No differences were present between ibuprofen and control animals in left (L.) ventricular weight or area at risk; ibuprofen significantly reduced infarct size expressed as percentage of area at risk (unpaired t test). October 1982 The American Journal of CARDIOLOGY Volume 50 853 IBUPROFEN AND REGIONAL MYOCARDIAL FUNCTION-KIRLIN ET AL cO.05, unpaired t test) despite a similar area at risk and left ventricular weight (Fig. 7). The percent of total left ventricle infarcted was substantially reduced by ibuprofen (19.7 f 9.1% versus 26.0 f 5.5% for controls), but this did not reach significance (p = 0.12, unpaired t test). By visual inspection of stained ventricular slices, the predominant area of tissue salvage in ibuprofentreated animals was at the epicardial and lateral margins of the infarct. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPO Light microscopic analysis of hematoxylin and eosin stained left ventricular slices revealed no evident differences in the extent of inflammatory infiltrate or hemorrhage or in the appearance of necrotic myocardial cells. The appearance of contraction band lesions and of fibroblastic proliferation was comparable in both groups, Visual inspection of the left circumflex coronary artery after death demonstrated no detectable thrombus in ibuprofen-treated animals and visible thrombus in 1 of 8 control animals. zyxwvutsrqponmlkjihgfedcbaZYX Discussion Rationale for this model: Coronary artery occlusion with myocardial reperfusion through a critical stenosis as employed in this model may more closely simulate the evolution of human myocardial infarction than permanent coronary artery occlusion. DeWood et al.17 demonstrated a progressive decline in the prevalence of total coronary occlusion over 24 hours in patients with acute myocardial infarction, a finding consistent with coronary artery spasm superimposed on fixed obstructive disease’* or with partial resolution of a coronary artery thrombosis,17 or both. In addition, experimental animal data indicate that reperfusion may enhance survivallg and improve the likelihood of myocardial functional recovery.20 Thus, a rationale exists for the coronary artery occlusion-reperfusion model of myocardial infarction. Possible mechanisms of ibuprofen’s myocardial protective effect: The present study demonstrates a myocardial protective effect of ibuprofen during experimental myocardial infarction without improvement of normal, marginal, or ischemic regional myocardial function during coronary artery occlusion or 48 or 72 hours after reperfusion or of inotropic reserve at 72 hours. The reduction in infarct size noted with ibuprofen in this study is consistent with previous investigations”-’ employing a variety of dosage schedules and routes of administration. While the mechanisms of infarct size reduction mediated by ibuprofen have not been established, several possibilities have been evaluated. First, it is unlikely that ibuprofen exerts a hemodynamic effect through diminished myocardial oxygen consumption or increased myocardial blood flow. Jugdutt et al.fi and Romson et al? noted no significant effect of ibuprofen on blood pressure or heart rate during myocardial ischemia, and the drug has only a minimal effect on ventricular contractile force.4 In our present study, continuous ibuprofen infusion beginning 1 hour before ischemia resulted in significant elevations of systemic arterial pressure (Table I) without an effect 854 October 1982 The American Journal of CARDIOLOGY Volume 50 BUPROFEN AND REGIONAL MYOCARDIAL FUNCTION-KIRLIN ET AL tiDn,33,34 which may in part account for its anti-inon heart rate after 90 minutes of coronary occlusion. flammatory effect. This result, not found in previous investigations The use of lidocaine in our study could also reduce the employing different models of ischemia and different severity of myocardial ischemic injury.35 Our data, drug administration schedules, would increase rather however, indicate extensive infarction despite lidocaine than decrease myocardial oxygen consumption and thus administration and an additional beneficial effect of tend to worsen ischemic injury.21 In addition, directly ibuprofen if lidocaine myocardial preservation ocmeasured myocardial oxygen consumption is not altered curred. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHG by ibuprofen administration during 60 minutes of Discordance between infarct size reduction and normothermic global ischemia in the isolated, bloodfunctional recovery: The discrepancy between myoperfused feline heart.i5 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA cardial infarct size and functional recovery in this inIncreased blood flow to ischemic tissue by direct vestigation may be due to several factors. Improvement coronary artery vasodilation or enhanced collateral in regional myocardial function may occur at a time not blood flow is also an unlikely mechanism of ibuprofen’s evaluated in this study. A functionally beneficial effect beneficial effect. Tracer microsphere studies in exof ibuprofen between the onset of reperfusion and 48 perimental myocardial infarction have failed to demhours cannot be excluded; in the present study a funconstrate increased blood flow to ischemic tissue up to tional assessment was not performed during that time 6 hours6 or 24 hours15 after coronary occlusion in ibubecause of marked beat-to-beat segment length changes profen-treated experimental animals. associated with frequent reperfusion arrhythmias. AlAn antithrombotic effect of ibuprofen is a 2nd posternatively, improvement in myocardial function with sible mechanism by which ibuprofen may reduce inibuprofen could occur at a time later than 72 hours. This farct size.7 In the present study, however, coronary possibility is supported by sonomicrometer crystal data artery reperfusion occurred in all animals and only 1 of in conscious dogs studied up to 4 weeks after 2 hour 8 control animals developed visible thrombus at 72 hour postmortem examination. Also, the antithrombotic coronary artery occlusion.20 Substantial recovery of agent sulfinpyrazone does not appear to reduce infarct marginal and ischemic segment systolic shortening occurred 1 to 4 weeks after coronary artery occlusionsize.22 Thus, it is improbable that an antithrombotic effect of ibuprofen played a role in the present study. reperfusion. Further studies will be necessary to deThe potentially related phenomenon of platelet depotermine whether ibuprofen treatment alters regional sition at the site of ischemic injury also occurs in myomyocardial function more than 3 days after myocardial cardial infarction and is diminished by aspirin.23 Ibuinfarction. profen, however, does not alter indium-111-labeled Improvement in indexes of myocardial function other than subendocardial segment length changes is platelet accumulation in the ischemic myocardium.2” A 3rd possible mechanism by which ibuprofen may another possible explanation of the discordance beexert a myocardial protective effect is its antiinflamtween infarct size reduction and functional recovery in matory action, related to inhibition of the arachidonic this investigation. Roan et a1.36noted spontaneous imacid pathway.” Cyclooxygenase enzyme inhibition could provement in ischemic zone ventricular wall paradoxic have a beneficial effect in ischemic tissue through a systolic thinning between 24 hours and 1 week after reduction in the synthesis of thromboxane, a potent coronary artery occlusion. Other measures of regional vasoconstrictor. However, other nonsteroidal cyclooxfunction which were not evaluated in this study include ygenase inhibitors do not reduce the extent of experimid-myocardial and subepicardial segment length mental ischemic injury.zsJsJe An alternative hychanges, which may have been altered more than subpothesis also related to the arachidonic acid cascade endocardial functional measurements in view of prewhich may in part explain ibuprofen’s beneficial effect dominant subepicardial tissue salvage by gross inin myocardial ischemia is inhibition of the lipoxygenase spection. pathway. This branch of‘the arachidonic acid cascade A persistent inability of ibuprofen to significantly is capable of producing powerful chemotactic subimprove marginal or ischemic region myocardial stances (hydroxyeicosatetraenoic acid [HETE] derivfunction despite infarct size reduction as determined atives) and leukotrienes, which are additional mediby histochemical staining is also possible. Glucocortiators of the inflammatory response.27 Ibuprofen has coids have been implicated in “mummification” of been shown to inhibit the synthesis of 12-HETE, cataischemic myocardium (loss of nuclei with intact myolyzed by a putative peroxidase enzyme.zs Lipoxygenase fibril striations and sarcolemma).37 Such tissue may pathway inhibition may in part account for ibuprofen’s retain the ability to stain dehydrogenase with triphenability to inhibit the migration of polymorphonuclear yltetrazolium despite the loss of functional integrity. leukocytes,2g an effect noted with ibuprofen in experiThe absence of such a “mummification” effect on light mental myocardial ischemia.24 microscopic evaluation in this investigation provides A 4th mechanism of potential significance in exevidence against that possibility. plaining ibuprofen’s myocardial protective action is A recent publication by Kloner et a1.38demonstrated inflammatory cell lysosomal membrane stabilization, that a 15 minute period of coronary occlusion followed an effect suggested by data from noncardiac in vitro by reperfusion, which is not associated with the develpreparationsa0-:s2 and by indirect evidence in canine opment of necrosis, nonetheless results in prolonged cardiac ischemia.4 In addition, ibuprofen affects kinin biochemical, functional, and ultrastructural abnorand histamine release2 and superoxide radical formamalities. Wall motion, as determined by ultrasonic October 1992 The American Journal of CARDIOLOGY Volume 50 955 IBUPROFEN AND REGIONAL MYOCARDIAL FUNCTION-KIRLIN ET AL crystal measurements, was still reduced at 3 days of reperfusion, whereas creatine phosphate levels and regional myocardial blood flow recovered rapidly and had returned to normal after 90 minutes of coronary reperfusion. Brief periods (15 minutes) of regional ischemia followed by full reperfusion, therefore, can be associated with abnormalities in cardiac contractile function which persist for a relatively long period. Bush et al.sg showed that the biochemical abnormalities associated with regional ischemia after reperfusion also extend to the normally perfused myocardium. Thus, a regional ischemic event may produce an influence on a remote region of myocardium which may persist for several days after the restoration of coronary blood flow.383 In the present investigation, coronary artery occlusion was maintained for 90 minutes and was followed by reperfusion in the presence of a critical stenosis. It is conceivable that the failure of the salvaged myocardial segment to regain contractile function could be related to the relatively persistent derangement in segment length shortening which is seen even in the absence of cellular necrosis. Future attempts to assess the potential benefits of pharmacologic interventions in protecting ischemic myocardium should consider the functional recovery of the “salvaged” myocardium days to weeks after the ischemic episode. Implications: Intravenous ibuprofen infusion reduces myocardial infarct size without improving normal, marginal, or ischemic region myocardial function during coronary artery occlusion or 48 or 72 hours after myocardial reperfusion, or inotropic reserve at 72 hours. Further studies are necessary to assess ibuprofen’s effect on regional myocardial function at intervals greater than 72 hours after myocardial infarction and to further elucidate the mechanisms by which ibuprofen exerts a protective effect on ischemic myocardial tissue. Ac kno wle dg m e nt We acknowledge the assistance of James Kipnis in performing surgical procedures and Dixie Thomas for the preparation of this manuscript. Upjohn Laboratories, Kalamazoo, Michigan, kindly provided ibuprofen. Re fe re nc e s 1. Gillespie TA, Sobel BE. A rationalefor therapy of acute myocardial infarction: limitation of infarct size. Adv Intern Med 1977;22:319-353. Kantor, TG. Ibuprofen Ann Intern Med 1979;91:877-882. :: Maclean D, Fishbeln MC, Blum RI, Braunwakf E, Maroko PR. Long-term preservation of ischemic myocardium by ibuprofen after experimental coronary artery occlusion (abstr). Am J Cardiol 1978;41:394. 4. Lefer Am, Polar&y EW. Beneficial effects of ibuprofen in acute myocardial ischemia. Cardiology 1979;64:265-279 5. Rlbelro LGT, Yasuda T, Lowensteln E, Braunwafd E, Yaroko PR. Comparative effects of anatomic infarct size of verampamil, Ibuprofen, and morphine-promethazine-chlorpromazine combination (abstr). Am J Cardiol 1979;43:396. 6. Jugdutt BI, Hutchins GM, Bulkley BH, Becker LC. Salvage of ischemic myocardium by ibuprofen during infarction in the conscious dog. Am J Cardiol 1980;46:74-82. 7. Romson JL. Bush LR. Haack DW, Lucchesi BR. The beneficial effects of oral ibuprofen on coronary artery.thrombosis and myocardial ischemia in the conscious dog. J Pharmacol Exp Ther 1980;215:271-278. a. Vane JR. Prostaglandins as mediators of acute rnflammation and pain. In 856 October 1982 The American Journal of CARDIOLOGY Rainsford KD, Fort-Hutchrnson AW, eds Prostaglandins and Inflammation. Easel: Birkhauser Verlag, 1979:113-l 18. 9. lherwx P, Franklln D, Ross J Jr, Kernper WS. Regional myocardial function during acute coronary artery occlusron and Its modification by pharmacologrc agents rn the dog Circ Res 1974;35:896-908 10. Geary GG, Smith GT, McNamara JJ. Defining the anatomic perfusion bed on an occluded coronarv arterv and the reoion ., at risk to Infarctron Am J Cardiol 1981;47.1240-i247 ’ 11. Borer J’S, Harrison LA, Kent KM, Levy R, Goldstein RE, Epstein SE. Benefrcral effect of lidocaine on ventricular electrical stability and spontaneous ventricular fibrillation durrno exoerimental mvocardial infarction Am J Cardiol 1976;37:860-863. 12. Kerber RE, Martins JB, Gascho JA, Marcus ML, Grayzel J. Effect of direct-current countershocks on regional myocardial contractility and perfusion Circulation 1981,63:323-332 13. Franklin D, Kernper WS, Patrick T, McKown. D. Technique for continuous measurement of regional myocardial segment dimensions in chronic animal preparations (abstr) Fed Proc 1973;32:343 14. Kirlin PC, Pitt B, Lucchesi BR. Comparative effects of prenalterol and dobutamine In a canine model of acute.ischemrc heart failure. J Cardiovasc Pharmacol 1981.3:896-905 15. Romson JL, Bush LR,, Jolly SR, Lucchesi BR. The cardioprotectrve effects of ibuprofen rn experrmental, regional and global myocardial ischemia J Cardrovasc Pharmacol 1982;4:187-196. 16. Morrison DF. Multtvanate Statistical Methods. New York: McGraw-Hill, 1976:153-160. 205-216 17. DeWood MA, Spores J, Notske R, et al. Prevalence of total coronary occlusion during the early hours of transmural myocardial Infarction. N Engl J Med 1980,303,897-902 18. Oliva PB, Breckenridge JC. Arteriographrc evrdence of coronary arterial spasm in acute myocardral Infarction. Circulation 1977;56:366-374 19. Baughman KL, Maroko PR, Vatner SF. Effects of coronary artery reperfusion on myocardial Infarct size and survrval rn conscious dogs. Circulation 1981;63.317-323 20. Theroux P, Ross J Jr, Franklin D, KemPer WS, Sasyama S. Coronary arterial reperfusion Ill Early and late effects in regional myocardial function and dimensrons in conscious doos. Am J Cardiol 1976:38:599-606 21. Maroko PR. Braunwald E. Effecis of metabolic and pharmacoloaic interventions on myocardial infarct size following coronary occlusion. Ctculation 197653 Suool 1:1-162-l-168 22. Bolli R, Gokfstein RE,Davenport N, Epstein SE. Influence of sulfinpyrazone and naproxen on Infarct size in the dog. Am J Cardiol 1981,47:841-847. 23. Ruf W, Leinburger H, McNamara J. Influence of aspirin on platelet trapping m myocardial infarctron rn baboons (abstr). Am J Cardiol 1978:41:407. 24. Romson J. Hook B. Riaot V, Swanson D. Lucchesi B. Effect of ibuprofen on the accumulation of 111’” leukocytes and platelets In Infarcted dog mvocardium (abstrl Fed Proc 1982:41:1766. 25. B&row RO, Lipson LC, Sheehan FH, et al. Lack of effect of asprrrn on myocardial infarct size in the dog. Am J Cardiol 1981;47:258-264 26. Jugdutt BI, Hutchins GM, Bulkley BH, Pitt B, Becker LC. Effect of indomethacrn on collateral blood flow and infarct size in the conscious doa Crrculation 1979;59:734-743 27. Kuehl FA, Egan RW. Prostaglandins. arachidonrc acrd and inflammation Science 1980:210:978-984. 28. Siegel MI, McConnell RT, Porter NA, Selph JL, Truax JF, Vinegar R, Cuatrecasas P. Aspirin-like drugs inhibit arachidonic acid metabolism via Iipoxygenase and cyclooxygenase rn rat neutrophils from carrageenan pleural exudates. Biochem Brophys Res Commun 1980;92:688-695 29. Brown KA, Collins AJ. Action of nonsteroidal, antiinflammatory drugs on human and rat peripheral leukocyte migration in vitro. Ann Rheum Dis 1977;36:239-243 30. Araki H, Lefer AM. Lysosomal stabilizing effects of two non-steroidal antiinflammatory agents in the hypoxic liver. Naunyn Schmiedebergs Arch Pharmacol 1980.311 79-84 31. Nakanidri M, GotoK. Inhibitory effects of antrtnflammatory drugs on enzyme release from rabbit polymorphonuclear leukocyte lysosomes. Biochem Pharmacol 1975,24:421-424. 32. Smlfh RJ, Sabin C, Gilchresf H, Willlams S. Effect of antsnflammatory drugs on lysosomes and lysosomal enzymes from rat liver. Biochem Pharmacol 1976.25:2171-2177 33. Simchowitz L, Mehta J, Spilberg I. Chemotactic factor-Induced generation of superoxide radrcals by human neutrophils. Effect of metabolic inhibitors and antrinflammatorv druos. Arthritis Rheum 1979:22:755-783. anion production in non-stimulated 34. Ovanaaui Y. lnhibrtibn o/suoeroxide guIneapig perttoneal exudate cells by anti-inflammatory drugs. Biochem Pharmacol 1978.27:777-782. 35. Nasser FN, Waiis JT, Edwards WD, Harrison CE. Lidocaine-induced reduction in size of expenmental myocardial infarction Am J Cardiol 1980; 46 967-975 36. Roan P. Scales F. Saffer S. Buia LM. Willerson J. Functional characterlzation of left ventricular segmental responses during the initial 24 hr and 1 wk after expenmental canine myocardial infarction. J Clan Invest 1979; 641074-1088 37. Kloner RA, Fishbein MC, Lew H, Maroko P, Braunwald E. Mummificatron of the infarcted myocardrum by high dose corticosteroids. Circulation 1978;57 56-63 38. Kloner RA, DeBoer LWV, Darsee JR, lngwall JS, Hall S, Tumas J, Braunwald E. Prolonged abnormal$es of myocardium salvaged by reperfusion. Am J Physrol 1981;241:H591-599. 39. Bush LR. Shlafer M. Haack DW. Lucchesi BR. Timedeoendent chanaes rn canine cardrac mbochondrial function and ultrastructure resulting from coronary occlusron and reperfusion Basic Res Cardiol 1980:75:555571 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGF zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJ Volume 50