- Email: [email protected]
Beneficial long-term effect of intracoronary perfluorochemical on infarct size and ventricular function in a canine reperfusion model
1082
JACC Vol. 9, No.5 May 1987:1082-90
Beneficial Long-Term Effect of Intracoronary Perfluorochemical on Infarct Size and Ventricular Function in a Canine Reperfusion Model MERVYN B. FORMAN, MD, FACC, DAVID W. PUETT, BA, B. HADLEY WILSON, MD, WILLIAM K. VAUGHN, PHD, GOTTLIEB C. FRIESINGER, MD, FACC,* RENU VIRMANI, MD, FACC Nashville, Tennessee
The administration of a drug soon after reperfusion that could enhance myocardial salvage would have important clinical application. The aim of this study was to assess the long-term effect of the perfluorochemical, F1uosol DA 20%, on infarct size, infarct morphology, ventricular ectopic activity and serial regional ventricular function in a 2 week closed chest canine model. After 90 minutes of proximal left anterior descending artery occlusion, animals randomly received either oxygenated 9) or saline solution (n 9) intracorF1uosol DA (n onary at 15 ml/kg body weight over 20 to 30 minutes. Hemodynamic variables were similar in the two groups except for transient elevation of left ventricular filling pressure immediately after infusion in the treated group. Infarct size was markedly reduced in the perfluorochemical-treated animals when expressed as a percent
=
=
Although timely reperfusion has been demonstrated to reduce infarct size both in humans and in animal models (1-7), the amount of myocardium that is potentially salvageable may be limited by anatomic or metabolic events or both that occur at the time of reperfusion. The term "reperfusion injury" has been applied to this entity (8-12). Therefore the administration soon after reperfusion of an agent that could limit reperfusion injury would have important clinical applications. Perfluorochemicals are artificial blood substances with low viscosity and a high oxygen-carrying capacity. Intravenous administration of large doses of the perfluorochemFrom the Department of Medicine, Division of Cardiology, Vanderbilt University Medical Center, Nashville, Tennessee. This study was supported in part by Grant RO I HL 34079-02 from the National Institutes of Health, Bethesda, Maryland, and by Alpha Therapeutic Corporation, Pasadena, California. It was presented in part at the 35th Annual Scientific Sessions of the American College of Cardiology, March 1986. Dr. Forman is a recipient of Young Investigator Award I R 23HL-34628-01 from the National Institutes of Health. *Present address: Department of Cardiovascular Pathology, Armed Forces Institute of Pathology, Washington, DC 20306. Manuscript received July 7, 1986; revised manuscript received September 16, 1986, accepted October 28, 1986. Address for reprints: Mervyn B. Forman, MD, Division of Cardiology, Vanderbilt University Medical Center, Nashville, Tennessee 37232. © 1987 by the American College of Cardiology
of the risk region (10.8 ± 1.8% versus 28.9 ± 5.5%, P < 0.02) or as a percent of the total left ventricle (3.7 ± 1% versus 10.8 ± 8%, P < 0.006). This was associated with greater improvement in radial shortening in the jeopardized zone at 2 weeks after reperfusion (15.3 ± 2.8% versus 5.2 ± 2.1 %, p < 0.01). Histologic examination revealed adequate healing in the treated animals with an increased number of swollen mononuclear cells in the border zones. Holter electrocardiographic recordings demonstrated a low frequency of ventricular ectopic beats in both groups. This study suggests that the perfluorochemical, Fluosol DA, may be a potentially useful agent in enhancing myocardial salvage after successful reperfusion. (J Am Coli CardioI1987;9:1082-90)
ical Fluosol DA 20% (Alpha Therapeutic Corporation) resulted in a significant reduction in infarct size in a permanent occlusion canine model (13). This was associated with a decreased acute inflammatory response in the infarct border zones. In vitro studies (14,15) have demonstrated that perfluorochemicals reduce neutrophil chemotaxis, adherence and free radical production. Because neutropenia and antiinflammatory agents have been shown to reduce infarct size in reperfusion models (16,17), we initially assessed the effect of low dose intracoronary perfluorochemical in a 24 hour reperfusion canine model (18) and found a significant reduction in infarct size. However, because the long-term effect of perfluorochemical over an extended period after reperfusion has not been assessed, we extended our previous study to examine serial ventricular function over a 2 week period in a similar canine model in an effort to simulate the anticipated effects of interventional therapy in evolving myocardial infarction.
Methods Animal selection and preparation. Mongrel dogs of either sex weighing 15 to 22 kg were quarantined for 2 weeks before entering the study. During this time they were 0735-1097/87/$3.50
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL·DA
lACC Vol. 9, No.5
May 1987: 1082-90
checked to ensure that they were free of common canine diseases. Pregnant animals and those with a hematocrit of less than 35% were excluded. The animals were prepared with a thoracotomy and the heart was exposed through a pericardiotomy. The left anterior descending coronary artery was isolated just after the first diagonal branch, and a snare (surgical monofilament) enclosed in a polyethylene sleeve tubing (PE 320) was implanted and anchored to the epicardium with two sutures. The chest was then closed and the tubing was buried in a subcutaneous pocket in the left subscapular region. Dogs were given prophylactic antibiotics (penicillin, 1 million units, and streptomycin, 1 g intramuscularly) and allowed 5 to 7 days to recover from surgery. Experimental protocol (Fig. 1). On the day of the experiment, dogs were randomly assigned to one of two treatment groups and anesthetized with 30 mg/kg body weight of intravenous pentobarbital sodium (Nembutal), intubated and ventilated with a positive pressure respirator to maintain arterial pH of 7.4 ± 0.5. Anesthesia was maintained during the study with an intravenous combination of morphine, 5 mg, and valium, 5 mg. Electrocardiographic leads I and aVF and a lateral precordial lead were continuously recorded. Under aseptic conditions a 7F Cordis sheath was placed through a cutdown into the right femoral artery and was used to introduce either a 7F pigtail catheter or a modified 7F right Judkins catheter. The snare was then retrieved and the animals were allowed 45 minutes to stabilize. Heart rate and aortic pressure were monitored continuously and left ventricular end-diastolic pressure was monitored intermittently (Fig. 1). A baseline contrast ventriculogram was performed in the right anterior oblique position using 6 to 8 cc of meglumine diatrizoate (Renografin-76) injected through a power injector. The degree of rotation was carefully noted to ensure that subsequent cines filming was performed in
Figure 1. Summary of experimental protocol. B = baseline; CPK = creatine kinase; CVG = contrast ventriculography; ICSK = intracoronary streptokinase; 1 hr 0 = 1 hour after occlusion (0); PFC = perftuorochemical (Fluosol DA); PLT = platelet count; PTT = partial thromboplastin time;TLC = totalleukocyte count. I~ hr
1hr
ICSK+PFC or Saline
i.; 1t to)
Boselin{
(8l
1/2 hr
Reperfusian (Rl IhrO
I I I I I I I I I I I I I I I I I I 1
1 23
24
Hours
1 2 Weeks
1. Hemodynamic Measurements (B, I hrO,R, 1,2,3,24 hrs; 1,2 wksl 2. Blood Gases TLC,PLT, PTT,CPK (8, R, 3,24 hrs) 3. CVG(8,1 hr O, 3,24 hrs; 1,2 wksl 4. Holter Monitor (2 wks)
1083
the same projection. Blood for determination of baseline blood gases, total blood count, creatine kinase, prothrombin time and partial thromboplastin time were then drawn. We have previously observed that perfluorochemical may cause transient hypotension in the dog, and that this usually occurs as a first dose phenomenom. Therefore both groups were given a test dose of 10 cc of saline solution or Fluosol intraarterially 45 minutes before infarction. Transient hypotension occurred in approximately 70% of the perfluorochemical-treated group. Saline solution was chosen as a control agent to account for the volume, electrolyte and viscosity changes associated with perfluorochemical administration. The osmolality of perfluoroehemical is only slightly higher than 0.9 M saline solution (320 versus ± 310 mosmol/liter). Before the snare was tightened, all animals were given lidocaine, I mg/kg intravenously. After 60 minutes of occlusion, hemodynamic measurements were made and a contrast ventriculogram was performed. Perjluorochemical (Fluosol DA 20%) or saline solution was oxygenated at room temperature with 100% oxygen at 2 liters/min for 45 minutes before infusion. Just before release of the snare, a selective coronary angiogram was made to confirm total occlusion of the left anterior descending artery, and a further dose of lidocaine, I mg/kg, was then given. The snare was gradually released over 3 to 5 minutes and the animals were ventilated with 100% oxygen, which was continued for 3 hours after reperfusion. To simulate reperfusion with fibrinolytic therapy, intracoronary streptokinase, 30,000 U, was given, and a coronary angiogram was performed to confirm patency of the artery. Five to 10 minutes after reperfusion, intracoronary infusion of oxygenated perfluorochemical or saline solution was performed through a pressure bag at the rate of 15 ml/kg over 20 minutes. Blood samples were drawn after completion of infusion of the reperfusion agent and at 3 hours after reperfusion. A repeat contrast ventriculogram was then performed after which the animals were weaned from the ventilator, given prophylactic antibiotics and returned to their cages. Hemodynamic measurements and ventriculography were repeated using sterile technique at 24 hours and 1 week after reperfusion under light anesthesia (Nembutal, 15 mg/kg). Prophylactic antibiotics were again given after each procedure. All dogs underwent continuous Holter electrocardiographic monitoring for 24 hours between day 10 and day 13 after reperfusion. At 14 days the animals were reanesthetized (Nembutal, 15 mg/kg) and a ventriculogram performed. After the chest was opened, the snare was retrieved and tightened, and Monastral blue (du Pont) at 1 mllkg was injected into the descending aorta. The animals were immediately killed with an overdose of potassium chloride, and the heart was rapidly excised. Histologic analysis. The heart was fixed in 10% phosphate-buffered formaldehyde for 3 days. The left ventricle was weighed and sliced at 1 em intervals parallel to the
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
1084
atrioventricular (AV) groove (six to seven slices) and photographed to define the area at risk (ar), unstained by Monastral blue. The sections were then dehydrated and embedded in paraffin. Microscopic sections, 7 1Lm thick, were stained with hematoxylin-eosin and Mallory's trichrome stain. The area at risk (ar) was then superimposed on the corresponding histologic slide. The area at risk, area of fibrosis (af) stained blue by Mallory's stain and area of hemorrhagic infarction (ah) stained purple by Mallory's stain were then enlarged on a microscopic slide projector and measured by planimetry using a computerized program. The areas of hemorrhagic infarction and fibrosis were then summated to give the total area of infarction and were multiplied by the thickness of each section to obtain the volume of the risk and infarct region. Light microscopy was performed on all sections with the observer unaware of the treatment given to each animal. Variables assessed included degree of inflammatory infiltrate, lymphocytes and mononuclear cells, hemorrhagic infarction, the extent of dense and reticular collagen formation, fibroblast proliferation and intrarnyocardial coronary artery changes within the area of infarction. An average of 20 high power fields (400 X) per slide was evaluated. These variables were assessed semiquanitatively according to the method of Romson et al. (17) with a score of 4 + for the most dense infiltrate and 0 for rare or no changes. Analysis of contrast ventriculograms. Contrast ventriculograrns were analyzed using a computerized program
Figure 2. Hemodynamic changes during experimental protocol. No significant changes were noted in heart rate (HR) or ratepressure product(RPP). Transientelevation in left ventricular enddiastolic pressure (L VEDP) occurredin the Fluosol-treated animals immediately after reperfusion (R). B = baseline, 0 = I hour into occlusion; SBP = systolic blood pressure.
HR
~:t
0--0
FlUOSOl
-
CONTROL
T
(beals/min)
1.
100
35 30
RPP (
25
HRXSBP) 20
1000
15 10
oL
LVEDP (mmHg)
:~~
1*
""
*p<.OOI
_JW-.J< __ 1 lot r::-r a
I
I
8 0
I
R
~
I I I I I 1 3 24 1 2 HOURS WEEKS
lACC Vol. 9. No.5
May 1987:1082-90
as previously described (18). Briefly, a longitudinal axis was constructed connecting the middle of the aortic valve plane and apex of the heart for both the end-diastolic and end-systolic silhouettes. Radii were then constructed from the midpoint of this axis to the edge of the ventricular silhouette at 10° intervals (36 segments). Percent shortening of each radius was calculated according to the formula: percent shortening = (end-diastolic length - systolic length/end-diastolic length x 100). Radii that involved the mitral and aortic valves were excluded from analysis (I to 3 and 31 to 36, respectively). The ischemic zone was defined as the largest number of radii that were asynergic or dysynergic, or both, at 1 hour into occlusion. Statistical analysis. Intergroup comparisons of infarct size were performed by Students t test for unpaired data. Data from successive time points in the two groups were tested by two-way analysis of variance followed by Duncan's multiple range test. Linear regression analysis was performed to examine the relation between infarct size and global and regional radial shortening 2 weeks after reperfusion. Statistical significance was defined as probability (p) <0.05.
Results Study groups. Thirty-two dogs were entered into the study. Eight animals were excluded because of the development of ventricular fibrillation during occlusion and four (two control and two treated with perfluorochemical) because of reperfusion ventricular fibrillation. Two animals (I perfluorochemical group, I control) were found dead in their cage, presumably as a result of a ventricular arrhythmia in the first 24 hours, and were excluded from analysis. A total of 18 animals survived the 2 week study period and were included in the final analysis: nine animals in the perfluorochemical group and nine in the control group. Laboratory and hemodynamic variables (Fig. 2). No significant differences were noted in heart rate, systolic and diastolic blood pressure, mean arterial pressure or rate-pressure product between the two groups throughout the study protocol. Left ventricular end-diastolic pressure was significantly higher in the perfluorochemical-treated group immediately after reperfusion (25 ± 7 versus 13 ± 3 mm Hg; p < 0.001) but had decreased markedly I hour after reperfusion to a value similar to that of control dogs. During the intracoronary infusion of perfluorochemical, no significant effects on heart rate or blood pressure were apparent. No differences between the two treatment groups were noted in the hematocrit, white cell count, platelet count, prothrombin time and partial thromboplastin time. Both baseline partial pressure of oxygen (Po 2 ) and P0 2 at 3 hours after reperfusion were comparable in the perfluorochemicaltreated (94 ± 8 and 406 ± 24 mm Hg) and control dogs (99 ± 10 and 342 ± 29 mm Hg).
l ACC Vol. 9. No.5
FORMAN ET AL . INFARCT SIZE AND VENTRICULAR FUNCTIOII: WIT H FLUOSOL-DA
May 1987: 101l2-'IO
Table 1. Effect of Intracoronary Perfluorochemical on Infarct Size 2 Weeks After Reperfusion Control Group (saline) (n = 9) LV mass (g) LV volume (cc) Infarct volume (cc) Volume of risk region (cc) Risk region/LV (%) InfarctiLV (%) Infarct/risk region (%) Hemorrhagic infarction (% )
Fibrosis (%)
69.5 ::':: 3.8 89.7 ::':: 6.0 (6 1 to 121) 9.9 ::':: 1.7 (0.4 to (8) 27.0 ± 3. 1 (14 to 37) 31.5 ::':: 3.5 (25 to 49) 10.8 ::':: 1.8 (5.0 10 17.7) 2K .9 ± 5.5 (2.8 to 47 .6) 34.3 ::':: 6.9 (23 to 68) 65.7 ± 6 .9 (32 to 77)
Perfluorochemical Group (Fluosol DA) (n = 9) 74 .5 ::':: 96.1 ::':: (74 to 3.9 ::':: (0.4 to 38.5 ± (32 to 38.5 ::':: (24 to 3.7 ::':: (0.4 to IO.K ± ( 1.2 to 7. 1 ± (0 to 92.9 ± (66 to
7.7 5.9 127) 1.5* 15) 2.9 ' 45 ) 3.0 5 1) l.lt 11 .9) 4.6 ' 40 .6) 4 .0 t 34) 4.0 t 100)
*p < 0.02 : t p < 0 .005 . perfluorochemical versus saline. Values are mean z SEM: values in parentheses represent ranges . LV = left ventricle.
Analysis of infarct size (Table 1, Fig. 3). Although the volume of the risk region was significantly larger in the perfiuorochemical group, it was not different when ex-
Figure 3. A, Whole mount section of mid-left ventricular slice stained with Masson 's trichrome stain in a dog treated with perf1uorochemical. Note complete healing with collagen deposition (blue) well separated from normal adjoining myocardium (red). B, Section from a dog treated with saline. Note healing with collagen deposition at the margins (blue) of the infarct and hemorrhagic necrosis (purple).
1085
pressed as a percent of the left ventricle. This may be a result uf chance randomization of larger left ventricular mass and volume in the treated group. However, we cannot rule out that perftuorochemical may have played a role. Infarct size was markedly reduced in the perftuorochemical-treated group whether expressed as a percent of the risk region or as a percent of the total left ventricle. Only 10.8 ± 4.6% of the area at risk was infarcted in the perfluorochemicaltreated dogs as compared with 28.9 ± 5.5% in the control dogs (p < 0.0 2). The extent of hemorrhagic infarction was significantly larger in control animals (34.3 ±: 6.9 versus 7. 1 ± 4.0%; p < 0.005). However, the extent of fibrosis was significantly greater in the treated group (92.9 ±: 4.0 versus 65.7 ±: 6.9%; p < 0.005). Microscopic changes (Table 2, Fig. 3 and 4). Infarction was confined to the subendocardial region in both groups but tended to be more patchy in the Fluosol-treated dogs (Fig. 3). Although myocardial necrosis and hemorrhage were far more extensive in the control animals (seven of nine) than in the perfluorochemical-treated animals (three of nine), fibroblastic proliferation was similar. Mononuclear cell infiltration was more extensive in the perfluorochernical-treated animals. and the cells were swollen and had a bubbly cytoplasm (Fig. 4). Infarct healing was accelerated in animals in the treated group, probably secondary to the smaller infarct size. No differences were noted in collagen deposition. including both dense and loose reticular collagen. However, the intramural coronary vessels were thickened in fi ve of the nine animals treated with perfluorochemical, and numerous mononuclear cells were present in the walls of these vessels. Analysis of ventricular function (Fig. 5 and 6). Although control dogs demonstrated a greater regional radial shortening at baseline, global ejection fraction was similar in both groups. At I hour into coronary occlusion and at 3 hours after reperfusion, dyskinesia in the jeopardized zone
1086
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
was similar, as was the reduction in global ejection fraction. The number of radii in the jeopardized zones of both groups was also similar (13 ± 9 versus 12 ± 4; P = NS). At 24 hours after reperfusion, the control group continued to manifest dyskinesia whereas the treated animals demonstrated positive wall motion (4.8 ± 2.7 versus -2.2 ± 1.7%; p < 0.0 I). Ventricular function continued to improve in both groups at I and 2 weeks after reperfusion with the perfluorochemical-treated dogs demonstrating greater improvement as measured by both regional radial shortening and global ejection fraction. At 2 weeks, radial shortening in the jeopardized zone was 15.3 ± 2.8% in the treatment group compared with only 5.2 ± 2.1% in the control group (p < 0.01). This was associated with a significantly greater global ejection fraction in perfluorochemical-treated animals (0.48 ± 0.10 versus 0.33 ± 0.09; p < 0.05).
Correlation of infarct size and radial shortening (Fig. 7). A loose correlation was noted between infarct size expressed as a percent of the risk region or of the total left ventricle and radial shortening in the ischemic zone at 2
lACC Vol. 9, No.5 May 1987: 1082-90
Figure 4. A and B, Photomicrographs taken from the area of myocardial infarction in a dog treated with intracoronary perfluorochemical, Fluosol DA. A, Note marked mononuclear cell collection interspersed in reticular collagen. The mononuclear cells show bubbly (arrows) cytoplasm from ingested perfluorochemical particles. B, Intramyocardial coronary arteries showing increased wall thickness and narrowed lumens (arrows). C and D, Sections from a control animal showing fibroblast and mononuclear reaction in areas of healing (C). D, Intramyocardial coronary vessels showing normal wall thickness (arrows). (A to D, Original magnification x 630, reduced by 25%.)
weeks after reperfusion (r = -0.6, P < 0.02 and r = - 0.6, p < 0.01, respectively). However, a tight correlation was found between infarct size and global radial shortening at 2 weeks after reperfusion (r = -0.86, P < 0.001 and r = -0.81, P < 0.001, respectively).
Effect of perfluorochemical treatment on arrhythmias. Intracoronary infusion of perfluorochemical did not alter QRS duration or the QT interval and did not tend to
l ACCVol. 9, NO.5
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
May 1987:1082-90
1087
Table 2. Qualitati ve Morphologic Differences in Dogs Treated With Perfluorochemical (Fluosol) or Saline Solution (Control)
Group
Hemorrhagic Infarction
Fluosol Control
1+ 3+
Collagen
Mononuclear Cells
Chronic Inflammatory Cells
Fibroblasts
Intramyocardial Coronary Artery Thickening
3+ 2+
3+ 1+
1+ 1+
2+ 3+
3+ 1+
increase the incidence of reperfusion arrhythmias. At 24 hours most dogs were noted to have runs of nonsustained ventricular tachycardia. Results of continuous Holter electrocardiographic monitoring performed for 24 hours a mean of 12 days after reperfusion demonstrated no significantdifference in ventricular ectopic activity, with < I premature ventricular complex/h present in both groups. Malignant arrhythmias (frequent runs of nonsustained ventricular tachycardia) were present in two dogs, one in each group. Histologic examination of these two animals showed no differences when compared with animals withoutmalignant arrhythmias .
Discussion Rationale for use of a reperfusion model Myocardial infarction remains a majorcauseof morbidity and mortality in the United States. Early reperfusion has been shown (1- 3,19,20) to reduce infarct size, improve ventricularfunction and decrease mortality in humans . The introduction of more effective and safer thrombolytic agents and percutaneous balloon angioplasty in evolving myocardial infarction has resulted in increasing numbers of patients undergoing successful reperfusion (21,22 ). The canine model has been extensively utilized to study the pathophysiology of regional myocardial ischemia and infarction and to assess the effects of various therapeutic
Figure 5. Serial percentage radial shortening in the ischemic zone during the experimental protocol. A significant improvement in shortening was already apparent at 24 hours after reperfu sion in the treated animals (Fluosol), and they continued to show even greater improvement of regional ventricular function over the 2 week period compared with control animals. (,?
~ 30 z w IQ:UJ Oz :Co 20 tIl ...J
FLUOSOL .. p< .05 lilt p<.OI
0.60
i= 0.50 o
ex
a:: 0.40
u..
c;E
UJ U a:: UJ a,
FlUOSOl
0
N
I-z z-
6--6 CONTROL 0--0
z
O-~
«/:u
«/:UJ Q:J: UJU C)tIl «/:-
Figure 6. A significantly greater improvement in ejection fraction was noted in the perfluorochemical-treated animals at I and 2 weeks after reperfusion .
* p<.05
6--1 CONTROL
it
interventions. Reimer et at. (4) showed that irreversible injury occurs as a " wave front" phenomenon progressing from the subendocardium to the subepicardium with increasing timeof occlusion. It thus appears that a reperfusion model would be most useful in assessing the ability of a drug to limit infarct size. Drugs that have been utilized in the reperfusion model include beta-receptor blockers (timolol), calcium antagonists (verapamil), agents that limit the production of oxygen free radicals (superoxide dismutase and catalase, allopurinol) and anti-infl ammatory agents (ibuprofen) ( 17,23- 28). However, these drugs must be administered soon after occlusion or before reperfusion to significantly reduce infarct size, and this limits their clinical applicability. Present study. We have previously shown (18) that intracoronary administration of theperfluorochemical, FIuosol DA, soon after reperfusion significantly reduces infarct size in a canine reperfusion model studied at 24 hours. Although a large amount of myocardium presumably remained stunned, a small but significant improvement in regional ventricular function was noted in the treated group. We have also shown (13) that perfluorochemical administered 30 minutes after infarction in a permanent occlusion model markedly reduced leukocytic infiltration at 3 days. Because leukocytes play an important role in the chronic and the acute inflammatory response , and because ventricular function remains depressed for weeks after an ischemic
z
10
0
0 .30
i= o 0.20
...,
UJ
0
UJ
-10
Baseline
OCclusion
3hr
24 hr
lwk
2wk
0.10
o Baseline .
Occlusion 3hr
24hr
lwk
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
1088
~
illary muscles are end arteries and are known to show fibrointimal thickening with age. Similar changes have been described in vessels supplying areas of healed infarction. Because infarct healing was more complete in perfluorochemical-treated comparedwith control dogs, intimal thickening in the former may representnormalhealing. However, we cannot exclude the possibility that arterial thickening was due to perftuorochemical treatment.
A-CONTROL O-FlUOSOl
C)
Z ...JZ
r =..... 81 n = 16
«ILl an.. .
11<·001
00:: ...JO C)%: (/)
...J
20
« c 15 « a::
10 0 A
5
10
15
20
INFARCT/LEFT VENTRICLE (0/0)
~
A-CONTROL O'FlUOSOl
C)
Z ...JZ
r = -.86
n= 14 11<.001
«ILl
lD~
00:: ...Jo
C)%: (/)
...J
20
« c 15 «
a:: 10 0 B
A
A 10
20
30
x
40
JACC Vol. 9. No.5 May 1987:1082-90
50
INFARCT/RISK REGION ("I.)
Figure 7. Relation between infarct size expressed both as a percentage of left ventricle (A) and area at risk (8) and global radial shortening 2 weeks after reperfusion. A close correlation was present.
episode, we assessed the effect of intracoronary FluosolDA on infarct size, infarct morphology and serial ventricular function in a 2 week canine reperfusion model. This study demonstrates that low dose intracoronary administration of the perftuorochemical FluosolDA reduced infarct size by 63% in a reperfusion model studied at 2 weeks. The degree of salvage was comparablewith findings in our short-term study (18). Infarct healing is a function of infarct size, with smaller infarcts healing in a shorter time than larger infarcts. Therefore, areas of hemorrhagic necrosis were more often present (seven of nine) in control dogs than in the experimental group (three of nine). However, fibroblast and coIlagen deposition was histologically similar in both groups. The most striking differences were noted in the morphology and number of the mononuclear ceIls present. Perfluorochemical-treated animals had swollen mononuclear cells with bubbly cytoplasm and eccentric nuclei, most prominent in the border zones and extensively interspersed between fibroblast and reticular collagen. MononuclearceIls in control animals were not as prominent as in perfluorochemical-treated animals and were not vacuolated. Chronic inflammatory cells were infrequently seen in both groups. Intramyocardial coronary vessels within the area of infarction in perfluorochemical-treated animals (five of nine)
showed fibrointimal thickening. Vessels supplying the pap-
Infarct size reduction was associated with significantly greater improvement in ventricular function in the jeopardized zone and global ejection fraction in the treated group.
The improvement seen at 24 hours was comparable with that in our short-term study, and the perfluorochemicaltreated dogs continued to show greater improvement over the study period. By 2 weeks, radial shortening in the ischemic zone had returned to 72% of baseline shortening in the treated group compared with a 33% return in the control group. A tight correlation was present between infarct size and global radial shortening at 2 weeks. Continuous Holter monitoring performed 10 to 13 days after reperfusion showed no significant difference in ventricular ectopic activity in the two treatment groups. Two animals, one in the control group and one in the perfluorochemical group, were noted "to have malignant arrhythmias. Except for a transient elevation of left ventricular fiIling pressure immediately after reperfusion, no significant hemodynamic or electrocardiographiceffects were noted with intracoronary administration of perfluorochemical at a dose of 15 ml/kg. Reperfusion injury. Reperfusion injury refers to the anatomic and metabolic changes occurring with reperfusion that may be detrimental and thus limit the amount of potentially salvageable myocardium (8-12). Although the exact mechanisms of reperfusion injury are not understood, possibilities include abnormalities in fluid and electrolyte homeostasis, "the no reflow phenomenon," endothelialcell damage, neutrophil-mediated injury and free radical production (8-11 ,29). The "no reflow' phenomenon refers to an incomplete return of blood flow to areas of the microcirculation after reperfusion (9). Postulated mechanisms includeperivascular
swelling, interstitial hemorrhage, endothelial cell damage and plugging of the microcirculation by neutrophils (9,10,29). Endothelial cells are known to modulate the activity of the coronary vascular bed to various endogenous compounds by the release of endothelial cell-derived relaxation factor (30-33). Thirty minutes of global ischemia in the isolated perfused rabbit heart results in abnormal vascular integrity of the endothelium I hour after reperfusion (34). When arteries in an area of myocardial infarction were examined functionally, they behaved as if their endothelium had been removed (35). Neutrophil plugs may also contribute to the "no reflew" phenomenon (10). By adhering to endothelial cells, neutrophils could potentiate endothelial damage by the release of free radicals and proteolyticenzymes (36,37).
JACC Vol. 9, No.5 May 1987: 1082-90
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
The production of large amounts offree radicals during reperfusion may also contribute to reperfusion injury (JJ,38). Free radicals are highly reactive substances that are toxic to both cellular membranes and organelles. The source of free radicals has not been clearly defined but they could be generated within endothelial, myocardial or white cells (11,38). Studies using both xanthine oxidase inhibitors and specific free radical metabolizing enzymes have demonstrated a significant reduction in infarct size in reperfusion models (26,28). Mechanism of action of perfluorochemical. The exact mechanism of action of perfluorochemical in limiting infarct size in the reperfusion model is not known. Because of its small particle size and low viscosity, perfluorochemical would be expected to increase oxygen delivery to the ischemic tissue. This effect has been demonstrated in a permanent occlusion model (39). It is possible that in a reperfusion model, perfluorochemical acts by delivering oxygen to areas of the microcirculation inaccessible to red blood cells as a result of the "no reflew" phenomenon. We have demonstrated ( 18) that regional myocardial blood flow as measured by microspheres increases immediately after reperfusion in the border zone but not at I hour, thus suggesting that other mechanisms may be operative. There are several mechanisms by which neutrophils may mediate reperfusion injury: microcirculatory plugging, the release of free radicals and proteolytic enzymes and the regulation of vascular permeability by Ieukotrienes (10,11,36,37,40). Neutropenia and anti-inflammatory agents reduce neutrophil infiltration in the marginal zone and significantly reduce infarct size in a reperfusion dog model (16,17). We have previously demonstrated (13) that large doses of perfluorochemical administered intravenously significantly reduce neutrophil infiltration in the border zones of a 3 day permanent occlusion model. We have also shown (14,15) that the perfluorochemical mixture reduces neutrophil phagocytosis, chemotaxis and free radical formation in vitro. Lane and Lamkin (41) demonstrated that the detergent component of Fluosol (Pluronic F68) markedly inhibits neutrophil chemotaxis and adherence in vitro and in vivo at concentrations similar to those administered in this study. The infarct size reduction in this experiment may therefore be related to the effect of perfluorochemical on neutrophil function. By interfering with neutrophil chemotaxis, perfluorochemical may decrease cellular damage resulting from free radical or proteolytic enzyme release, or both. Also, by blocking adherence, perfluorochemical may limit microcirculatory plugging and neutrophil-endothelial interactions. It is also possible that perfluorochemical may interfere with free radical production in endothelial cells or myocytes. However, the specific mechanisms by which perfluorochemicals affect metabolic pathways producing free radicals remain to be defined. Electron microscopy has shown that perfluorochemical
1089
particles are present in endothelial cells in the permanent occlusion model (13). It is conceivable that by preserving the anatomic or functional integrity, or both, of endothelial cells, ischemic injury may be reduced especially at the microcirculatory level. In the reperfusion model the endothelial cell swelling that has been described as contributing to the "no reflew" phenomenon may actually be abolished by intracoronary perfluorochemical therapy at the time of reperfusion. Clinical implications. The discovery of fibrin-specific thrombolytic agents has resulted in reperfusion as a logical maneuver in limiting infarct size in humans. The addition of an agent at the time of reperfusion that could enhance myocardial salvage would therefore have important therapeutic implications. This study demonstrates that perfluorochemical given soon after reperfusion significantly enhances myocardial salvage compared with reperfusion alone. This salvage was associated with near normal infarct healing and improved regional ventricular function in the jeopardized zone. Intracoronary perfluorochemical may be a potentially useful agent in enhancing myocardial salvage after successful reperfusion. We thank Andrew Manlove. James Phillips and Donna Bostick for expert technical assistance and Linda Grayson for manuscript preparation.
References I. Ganz W. Geft I, Maddahi 1, et al. Nonsurgical reperfusion in evolving myocardial infarction. J Am Coli Cardiol 1983;1:1247-53. 2. Anderson JL, Marshall HW, Bray BE. et al. A randomized trial of intracoronary streptokinase in the treatment of acute myocardial infarction. N Engl J Med 1983;308:1312-8. 3. Stack RS, Phillips III HR. Grierson OS, et al. Functional improvement of jeopardized myocardium following intracoronary streptokinase infusion in acute myocardial infarction. J Clin Invest 1983;72:84-95. 4. Reimer KA. Lowe JE. Rasmussen MM, Jennings RB. The wavefront phenomenon of ischemic cell death. I. Myocardial infarct size vs. duration of coronary occlusion in dogs. Circulation 1983;56:786-94. 5. Kloner RA, Ellis SG, Lange R. Braunwald E. Studies of experimental coronary artery reperfusion: effects on infarct size. myocardial function. biochemistry, ultrastructure, and microvascular damage. Circulation 1983;68(suppl 1):1-8-15. 6. Karsch KR. Hofmann M, Rentrop KP, Blanke H. Schaper W. Thrombolysis in acute experimental myocardial infarction. J Am Coli Cardiol 1983:1:427-35. 7. Lavallee M, Cox D, Patrick TA, Vatner SF. Salvage of myocardial function by coronary artery reperfusion, I, 2 and 3 hours after occlusion in conscious dogs. Circ Res 1983;53:235-47. 8. Braunwald E, Kloner RA. Myocardial reperfusion: a double edged sword. J Clin Invest 1985;76: 1713-9. 9. Kloner RA, Ganote CE, Jennings RB. The "no-reflew" phenomenon after temporary occlusion in the dog. J Clin Invest 1974;54: 1496-508. 10. Engler RL, Schmid-Schonbein GW, Pavelec RS. Leukocyte capillary plugging in myocardial ischemia and reperfusion in the dog. Am J Pathol 1983: 111:98-111. II. McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 1985;312:159-63.
1090
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
12. Zimmerman ANE, Daems W, Hulsmann WC, et at. Morphological changes of heart muscle caused by successive perfusion with calciumfree and calcium containing solution (calcium paradox). Cardiovasc Res 1967;1:201-9. 13. Kolodgie FD, Dawson AK, Forman MB, Vinnani R. Effect of perfluorochemical Fluosol-DA on infarct morphology in dogs with permanent coronary artery occlusion. Virchows Arch [Cell Pathol) 1985;50: 119-34. 14. Virmani R, Fink LM, Gunter K, English D. Effects of perfluorochemicals on human neutrophil function. Transfusion 1984;24:343-7. 15. Virmani R, Warren D, Rees R, Fink LM, English D. Effect of perfluorochemical on phagocytic function of leukocytes. Transfusion 1983;23:512-25. 16. Romson Jl., Hook BG, Kunkel SL, et at. Reduction of the extent of ischemic myocardial injury by neutrophil depletion in the dog. Circulation 1983;67:1016-23. 17. Romson 1, Hook B, Rigot V, et at. The effect of ibuprofen on accumulation of indium-III myocardial platelets and leukocytes in experimental infarction. Circulation 1982;66: 1002-11. 18. Forman MB, Bingham S, Kopelman HA, et at. Reduction of infarct size with intracoronary perfluorochemical in a canine preparation of reperfusion. Circulation 1985;71: 1060-8. 19. Kennedy lW, Ritchie lL, Davis KB, Stadius ML, Maynard C, Fritz lK. The Western Washington randomized trial of intracoronary streptokinase in acute myocardial infarction. A 12-month followup report. N Engl 1 Med 1985;312:1073-8. 20. Koren G, Weiss AT, Hasin Y, et at. Prevention of myocardial damage in acute myocardial ischemia by early treatment with intravenous streptokinase. N Engl 1 Med 1985;312:1384-9. 21. TlMI Study Group. The thrombolysis in myocardial infarction (TIMI) trial phase I findings. N Engl 1 Med 1985;312:932-6. 22. O'Neill W, Timmis G, Bourdillon PD, et at. A prospective randomized clinical trial of intracoronary streptokinase versus coronary angioplasty for acute myocardial infarction. N Engl 1 Med 1986;314:812-8. 23. Hammerman H, Kloner RA, Briggs LL, Braunwald E. Enhancement of salvage of reperfused myocardium by early beta-adrenergic blockade (timolol). 1 Am Coli Cardiol 1984;3:1438-43. 24. Lo HM, Kloner RA, Braunwald E. Effect of intracoronary verapamil in infarct size in the ischemic, reperfusion canine heart: clinical importance of the timing of treatment. Am J Cardiol 1985;56:672-7. 25. Reimer KA, lennings RB. Verapamil in two reperfusion models of myocardial infarction. Temporary protection of severely ischemic myocardium without limitation of ultimate infarct size. Lab Invest 1984;5I :655-66. 26. lolly SR, Kane Wl, Bailie MB, Abrams GD, Lucchesi BR. Canine
lACC Vol. 9, No.5
May 1987: 1082-90
myocardial reperfusion. Its reduction by the combined administration of superoxide dismutase and catalase. Circ Res 1984;54:277-85. 27. Burton KP. Superoxide dismutase enhances recovery following myocardial ischemia. Am 1 Physiol 1985;248:H637-43. 28. Werns SW, Shea Ml, Mitsos SE, et at. Reduction of the size of infarction by allopurinol in the ischemic-reperfused heart. Circulation 1986;73:518-26. 29. lennings RB, Schaper 1, Hill ML, Steenbergen lC, Reimer KA. Effects of reperfusion late in the phase of reversible ischemic injury. Changes in cell volume, electrolytes, metabolites and ultrastructure. Circ Res 1985;56:262-78. 30. Cohen RA, Shephard JT, Vanhoutte PM. Inhibitory role of the endothelium in the response of isolated coronary arteries to platelets. Science 1983;221:272-4. 31. DeMey IG, Caeys M, Vanhoutte PM. Endothelium-dependent inhibitory effects of acetylcholine, adenosine triphosphate, thrombin and arachidonic acid in the canine femoral artery. 1 Phannacol Exp Ther 1982;222:166-73. 32. Furchgott RF. The requirement of endothelial cells in relaxation of arteries by acetylcholine and some other vasodilators. Trends Pharmacol Sci 1981;2:173-6. 33. DeMey 10, Vanhoutte PM. Contribution of the endothelium to the response to anoxia in the canine femoral artery. Arch Int Pharmacodyn Ther 198I;253:325-6. 34. Tilton RG, Larson KB, Udell lR, Sobel BE, Williamson IF. External detection of early microvascular dysfunction after no-flow ischemia followed by reperfusion in isolated rabbit hearts. Circ Res 1983;52: 210-25. 35. Ku DD. Coronary vascular reactivity after acute myocardial ischemia. Science 1982;218:576-8. 36. Sacks T, Moldow CF, Craddock PR, Bowers TK, lacob HS. Oxygen radicals mediate endothelial damage by complement-stimulated granulocytes. 1 Clin Invest 1978;61:1161-7. 37. Harlan 1M. Leukocyte-endothelial interactions. Blood 1985;65:513-25. 38. Del Maestro RF. An approach to free radicals in medicine and biology. Acta Physiol Scand 1980;492:153-68. 39. Rude RE, Glogar D, Khuri SF, et at. Effects of intravenous fluorocarbons during and without oxygen enhancement on acute myocardial ischemic injury assessed by measurement of intramyocardial gas tensions. Am Heart J 1982;103:986-95. 40. Davies P, Bailey Pl, Goldenberg MM. The role of arachidonic acid and oxygenation products in pain and inflammation. Annu Rev Immunol 1984;2:335-57. 41. Lane TA, Lamkin GE. Paralysis of phagocyte migration due to an artificial blood substitute. Blood 1984;64:400-5.
JACC Vol. 9, No.5 May 1987:1082-90
Beneficial Long-Term Effect of Intracoronary Perfluorochemical on Infarct Size and Ventricular Function in a Canine Reperfusion Model MERVYN B. FORMAN, MD, FACC, DAVID W. PUETT, BA, B. HADLEY WILSON, MD, WILLIAM K. VAUGHN, PHD, GOTTLIEB C. FRIESINGER, MD, FACC,* RENU VIRMANI, MD, FACC Nashville, Tennessee
The administration of a drug soon after reperfusion that could enhance myocardial salvage would have important clinical application. The aim of this study was to assess the long-term effect of the perfluorochemical, F1uosol DA 20%, on infarct size, infarct morphology, ventricular ectopic activity and serial regional ventricular function in a 2 week closed chest canine model. After 90 minutes of proximal left anterior descending artery occlusion, animals randomly received either oxygenated 9) or saline solution (n 9) intracorF1uosol DA (n onary at 15 ml/kg body weight over 20 to 30 minutes. Hemodynamic variables were similar in the two groups except for transient elevation of left ventricular filling pressure immediately after infusion in the treated group. Infarct size was markedly reduced in the perfluorochemical-treated animals when expressed as a percent
=
=
Although timely reperfusion has been demonstrated to reduce infarct size both in humans and in animal models (1-7), the amount of myocardium that is potentially salvageable may be limited by anatomic or metabolic events or both that occur at the time of reperfusion. The term "reperfusion injury" has been applied to this entity (8-12). Therefore the administration soon after reperfusion of an agent that could limit reperfusion injury would have important clinical applications. Perfluorochemicals are artificial blood substances with low viscosity and a high oxygen-carrying capacity. Intravenous administration of large doses of the perfluorochemFrom the Department of Medicine, Division of Cardiology, Vanderbilt University Medical Center, Nashville, Tennessee. This study was supported in part by Grant RO I HL 34079-02 from the National Institutes of Health, Bethesda, Maryland, and by Alpha Therapeutic Corporation, Pasadena, California. It was presented in part at the 35th Annual Scientific Sessions of the American College of Cardiology, March 1986. Dr. Forman is a recipient of Young Investigator Award I R 23HL-34628-01 from the National Institutes of Health. *Present address: Department of Cardiovascular Pathology, Armed Forces Institute of Pathology, Washington, DC 20306. Manuscript received July 7, 1986; revised manuscript received September 16, 1986, accepted October 28, 1986. Address for reprints: Mervyn B. Forman, MD, Division of Cardiology, Vanderbilt University Medical Center, Nashville, Tennessee 37232. © 1987 by the American College of Cardiology
of the risk region (10.8 ± 1.8% versus 28.9 ± 5.5%, P < 0.02) or as a percent of the total left ventricle (3.7 ± 1% versus 10.8 ± 8%, P < 0.006). This was associated with greater improvement in radial shortening in the jeopardized zone at 2 weeks after reperfusion (15.3 ± 2.8% versus 5.2 ± 2.1 %, p < 0.01). Histologic examination revealed adequate healing in the treated animals with an increased number of swollen mononuclear cells in the border zones. Holter electrocardiographic recordings demonstrated a low frequency of ventricular ectopic beats in both groups. This study suggests that the perfluorochemical, Fluosol DA, may be a potentially useful agent in enhancing myocardial salvage after successful reperfusion. (J Am Coli CardioI1987;9:1082-90)
ical Fluosol DA 20% (Alpha Therapeutic Corporation) resulted in a significant reduction in infarct size in a permanent occlusion canine model (13). This was associated with a decreased acute inflammatory response in the infarct border zones. In vitro studies (14,15) have demonstrated that perfluorochemicals reduce neutrophil chemotaxis, adherence and free radical production. Because neutropenia and antiinflammatory agents have been shown to reduce infarct size in reperfusion models (16,17), we initially assessed the effect of low dose intracoronary perfluorochemical in a 24 hour reperfusion canine model (18) and found a significant reduction in infarct size. However, because the long-term effect of perfluorochemical over an extended period after reperfusion has not been assessed, we extended our previous study to examine serial ventricular function over a 2 week period in a similar canine model in an effort to simulate the anticipated effects of interventional therapy in evolving myocardial infarction.
Methods Animal selection and preparation. Mongrel dogs of either sex weighing 15 to 22 kg were quarantined for 2 weeks before entering the study. During this time they were 0735-1097/87/$3.50
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL·DA
lACC Vol. 9, No.5
May 1987: 1082-90
checked to ensure that they were free of common canine diseases. Pregnant animals and those with a hematocrit of less than 35% were excluded. The animals were prepared with a thoracotomy and the heart was exposed through a pericardiotomy. The left anterior descending coronary artery was isolated just after the first diagonal branch, and a snare (surgical monofilament) enclosed in a polyethylene sleeve tubing (PE 320) was implanted and anchored to the epicardium with two sutures. The chest was then closed and the tubing was buried in a subcutaneous pocket in the left subscapular region. Dogs were given prophylactic antibiotics (penicillin, 1 million units, and streptomycin, 1 g intramuscularly) and allowed 5 to 7 days to recover from surgery. Experimental protocol (Fig. 1). On the day of the experiment, dogs were randomly assigned to one of two treatment groups and anesthetized with 30 mg/kg body weight of intravenous pentobarbital sodium (Nembutal), intubated and ventilated with a positive pressure respirator to maintain arterial pH of 7.4 ± 0.5. Anesthesia was maintained during the study with an intravenous combination of morphine, 5 mg, and valium, 5 mg. Electrocardiographic leads I and aVF and a lateral precordial lead were continuously recorded. Under aseptic conditions a 7F Cordis sheath was placed through a cutdown into the right femoral artery and was used to introduce either a 7F pigtail catheter or a modified 7F right Judkins catheter. The snare was then retrieved and the animals were allowed 45 minutes to stabilize. Heart rate and aortic pressure were monitored continuously and left ventricular end-diastolic pressure was monitored intermittently (Fig. 1). A baseline contrast ventriculogram was performed in the right anterior oblique position using 6 to 8 cc of meglumine diatrizoate (Renografin-76) injected through a power injector. The degree of rotation was carefully noted to ensure that subsequent cines filming was performed in
Figure 1. Summary of experimental protocol. B = baseline; CPK = creatine kinase; CVG = contrast ventriculography; ICSK = intracoronary streptokinase; 1 hr 0 = 1 hour after occlusion (0); PFC = perftuorochemical (Fluosol DA); PLT = platelet count; PTT = partial thromboplastin time;TLC = totalleukocyte count. I~ hr
1hr
ICSK+PFC or Saline
i.; 1t to)
Boselin{
(8l
1/2 hr
Reperfusian (Rl IhrO
I I I I I I I I I I I I I I I I I I 1
1 23
24
Hours
1 2 Weeks
1. Hemodynamic Measurements (B, I hrO,R, 1,2,3,24 hrs; 1,2 wksl 2. Blood Gases TLC,PLT, PTT,CPK (8, R, 3,24 hrs) 3. CVG(8,1 hr O, 3,24 hrs; 1,2 wksl 4. Holter Monitor (2 wks)
1083
the same projection. Blood for determination of baseline blood gases, total blood count, creatine kinase, prothrombin time and partial thromboplastin time were then drawn. We have previously observed that perfluorochemical may cause transient hypotension in the dog, and that this usually occurs as a first dose phenomenom. Therefore both groups were given a test dose of 10 cc of saline solution or Fluosol intraarterially 45 minutes before infarction. Transient hypotension occurred in approximately 70% of the perfluorochemical-treated group. Saline solution was chosen as a control agent to account for the volume, electrolyte and viscosity changes associated with perfluorochemical administration. The osmolality of perfluoroehemical is only slightly higher than 0.9 M saline solution (320 versus ± 310 mosmol/liter). Before the snare was tightened, all animals were given lidocaine, I mg/kg intravenously. After 60 minutes of occlusion, hemodynamic measurements were made and a contrast ventriculogram was performed. Perjluorochemical (Fluosol DA 20%) or saline solution was oxygenated at room temperature with 100% oxygen at 2 liters/min for 45 minutes before infusion. Just before release of the snare, a selective coronary angiogram was made to confirm total occlusion of the left anterior descending artery, and a further dose of lidocaine, I mg/kg, was then given. The snare was gradually released over 3 to 5 minutes and the animals were ventilated with 100% oxygen, which was continued for 3 hours after reperfusion. To simulate reperfusion with fibrinolytic therapy, intracoronary streptokinase, 30,000 U, was given, and a coronary angiogram was performed to confirm patency of the artery. Five to 10 minutes after reperfusion, intracoronary infusion of oxygenated perfluorochemical or saline solution was performed through a pressure bag at the rate of 15 ml/kg over 20 minutes. Blood samples were drawn after completion of infusion of the reperfusion agent and at 3 hours after reperfusion. A repeat contrast ventriculogram was then performed after which the animals were weaned from the ventilator, given prophylactic antibiotics and returned to their cages. Hemodynamic measurements and ventriculography were repeated using sterile technique at 24 hours and 1 week after reperfusion under light anesthesia (Nembutal, 15 mg/kg). Prophylactic antibiotics were again given after each procedure. All dogs underwent continuous Holter electrocardiographic monitoring for 24 hours between day 10 and day 13 after reperfusion. At 14 days the animals were reanesthetized (Nembutal, 15 mg/kg) and a ventriculogram performed. After the chest was opened, the snare was retrieved and tightened, and Monastral blue (du Pont) at 1 mllkg was injected into the descending aorta. The animals were immediately killed with an overdose of potassium chloride, and the heart was rapidly excised. Histologic analysis. The heart was fixed in 10% phosphate-buffered formaldehyde for 3 days. The left ventricle was weighed and sliced at 1 em intervals parallel to the
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
1084
atrioventricular (AV) groove (six to seven slices) and photographed to define the area at risk (ar), unstained by Monastral blue. The sections were then dehydrated and embedded in paraffin. Microscopic sections, 7 1Lm thick, were stained with hematoxylin-eosin and Mallory's trichrome stain. The area at risk (ar) was then superimposed on the corresponding histologic slide. The area at risk, area of fibrosis (af) stained blue by Mallory's stain and area of hemorrhagic infarction (ah) stained purple by Mallory's stain were then enlarged on a microscopic slide projector and measured by planimetry using a computerized program. The areas of hemorrhagic infarction and fibrosis were then summated to give the total area of infarction and were multiplied by the thickness of each section to obtain the volume of the risk and infarct region. Light microscopy was performed on all sections with the observer unaware of the treatment given to each animal. Variables assessed included degree of inflammatory infiltrate, lymphocytes and mononuclear cells, hemorrhagic infarction, the extent of dense and reticular collagen formation, fibroblast proliferation and intrarnyocardial coronary artery changes within the area of infarction. An average of 20 high power fields (400 X) per slide was evaluated. These variables were assessed semiquanitatively according to the method of Romson et al. (17) with a score of 4 + for the most dense infiltrate and 0 for rare or no changes. Analysis of contrast ventriculograms. Contrast ventriculograrns were analyzed using a computerized program
Figure 2. Hemodynamic changes during experimental protocol. No significant changes were noted in heart rate (HR) or ratepressure product(RPP). Transientelevation in left ventricular enddiastolic pressure (L VEDP) occurredin the Fluosol-treated animals immediately after reperfusion (R). B = baseline, 0 = I hour into occlusion; SBP = systolic blood pressure.
HR
~:t
0--0
FlUOSOl
-
CONTROL
T
(beals/min)
1.
100
35 30
RPP (
25
HRXSBP) 20
1000
15 10
oL
LVEDP (mmHg)
:~~
1*
""
*p<.OOI
_JW-.J< __ 1 lot r::-r a
I
I
8 0
I
R
~
I I I I I 1 3 24 1 2 HOURS WEEKS
lACC Vol. 9. No.5
May 1987:1082-90
as previously described (18). Briefly, a longitudinal axis was constructed connecting the middle of the aortic valve plane and apex of the heart for both the end-diastolic and end-systolic silhouettes. Radii were then constructed from the midpoint of this axis to the edge of the ventricular silhouette at 10° intervals (36 segments). Percent shortening of each radius was calculated according to the formula: percent shortening = (end-diastolic length - systolic length/end-diastolic length x 100). Radii that involved the mitral and aortic valves were excluded from analysis (I to 3 and 31 to 36, respectively). The ischemic zone was defined as the largest number of radii that were asynergic or dysynergic, or both, at 1 hour into occlusion. Statistical analysis. Intergroup comparisons of infarct size were performed by Students t test for unpaired data. Data from successive time points in the two groups were tested by two-way analysis of variance followed by Duncan's multiple range test. Linear regression analysis was performed to examine the relation between infarct size and global and regional radial shortening 2 weeks after reperfusion. Statistical significance was defined as probability (p) <0.05.
Results Study groups. Thirty-two dogs were entered into the study. Eight animals were excluded because of the development of ventricular fibrillation during occlusion and four (two control and two treated with perfluorochemical) because of reperfusion ventricular fibrillation. Two animals (I perfluorochemical group, I control) were found dead in their cage, presumably as a result of a ventricular arrhythmia in the first 24 hours, and were excluded from analysis. A total of 18 animals survived the 2 week study period and were included in the final analysis: nine animals in the perfluorochemical group and nine in the control group. Laboratory and hemodynamic variables (Fig. 2). No significant differences were noted in heart rate, systolic and diastolic blood pressure, mean arterial pressure or rate-pressure product between the two groups throughout the study protocol. Left ventricular end-diastolic pressure was significantly higher in the perfluorochemical-treated group immediately after reperfusion (25 ± 7 versus 13 ± 3 mm Hg; p < 0.001) but had decreased markedly I hour after reperfusion to a value similar to that of control dogs. During the intracoronary infusion of perfluorochemical, no significant effects on heart rate or blood pressure were apparent. No differences between the two treatment groups were noted in the hematocrit, white cell count, platelet count, prothrombin time and partial thromboplastin time. Both baseline partial pressure of oxygen (Po 2 ) and P0 2 at 3 hours after reperfusion were comparable in the perfluorochemicaltreated (94 ± 8 and 406 ± 24 mm Hg) and control dogs (99 ± 10 and 342 ± 29 mm Hg).
l ACC Vol. 9. No.5
FORMAN ET AL . INFARCT SIZE AND VENTRICULAR FUNCTIOII: WIT H FLUOSOL-DA
May 1987: 101l2-'IO
Table 1. Effect of Intracoronary Perfluorochemical on Infarct Size 2 Weeks After Reperfusion Control Group (saline) (n = 9) LV mass (g) LV volume (cc) Infarct volume (cc) Volume of risk region (cc) Risk region/LV (%) InfarctiLV (%) Infarct/risk region (%) Hemorrhagic infarction (% )
Fibrosis (%)
69.5 ::':: 3.8 89.7 ::':: 6.0 (6 1 to 121) 9.9 ::':: 1.7 (0.4 to (8) 27.0 ± 3. 1 (14 to 37) 31.5 ::':: 3.5 (25 to 49) 10.8 ::':: 1.8 (5.0 10 17.7) 2K .9 ± 5.5 (2.8 to 47 .6) 34.3 ::':: 6.9 (23 to 68) 65.7 ± 6 .9 (32 to 77)
Perfluorochemical Group (Fluosol DA) (n = 9) 74 .5 ::':: 96.1 ::':: (74 to 3.9 ::':: (0.4 to 38.5 ± (32 to 38.5 ::':: (24 to 3.7 ::':: (0.4 to IO.K ± ( 1.2 to 7. 1 ± (0 to 92.9 ± (66 to
7.7 5.9 127) 1.5* 15) 2.9 ' 45 ) 3.0 5 1) l.lt 11 .9) 4.6 ' 40 .6) 4 .0 t 34) 4.0 t 100)
*p < 0.02 : t p < 0 .005 . perfluorochemical versus saline. Values are mean z SEM: values in parentheses represent ranges . LV = left ventricle.
Analysis of infarct size (Table 1, Fig. 3). Although the volume of the risk region was significantly larger in the perfiuorochemical group, it was not different when ex-
Figure 3. A, Whole mount section of mid-left ventricular slice stained with Masson 's trichrome stain in a dog treated with perf1uorochemical. Note complete healing with collagen deposition (blue) well separated from normal adjoining myocardium (red). B, Section from a dog treated with saline. Note healing with collagen deposition at the margins (blue) of the infarct and hemorrhagic necrosis (purple).
1085
pressed as a percent of the left ventricle. This may be a result uf chance randomization of larger left ventricular mass and volume in the treated group. However, we cannot rule out that perftuorochemical may have played a role. Infarct size was markedly reduced in the perftuorochemical-treated group whether expressed as a percent of the risk region or as a percent of the total left ventricle. Only 10.8 ± 4.6% of the area at risk was infarcted in the perfluorochemicaltreated dogs as compared with 28.9 ± 5.5% in the control dogs (p < 0.0 2). The extent of hemorrhagic infarction was significantly larger in control animals (34.3 ±: 6.9 versus 7. 1 ± 4.0%; p < 0.005). However, the extent of fibrosis was significantly greater in the treated group (92.9 ±: 4.0 versus 65.7 ±: 6.9%; p < 0.005). Microscopic changes (Table 2, Fig. 3 and 4). Infarction was confined to the subendocardial region in both groups but tended to be more patchy in the Fluosol-treated dogs (Fig. 3). Although myocardial necrosis and hemorrhage were far more extensive in the control animals (seven of nine) than in the perfluorochemical-treated animals (three of nine), fibroblastic proliferation was similar. Mononuclear cell infiltration was more extensive in the perfluorochernical-treated animals. and the cells were swollen and had a bubbly cytoplasm (Fig. 4). Infarct healing was accelerated in animals in the treated group, probably secondary to the smaller infarct size. No differences were noted in collagen deposition. including both dense and loose reticular collagen. However, the intramural coronary vessels were thickened in fi ve of the nine animals treated with perfluorochemical, and numerous mononuclear cells were present in the walls of these vessels. Analysis of ventricular function (Fig. 5 and 6). Although control dogs demonstrated a greater regional radial shortening at baseline, global ejection fraction was similar in both groups. At I hour into coronary occlusion and at 3 hours after reperfusion, dyskinesia in the jeopardized zone
1086
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
was similar, as was the reduction in global ejection fraction. The number of radii in the jeopardized zones of both groups was also similar (13 ± 9 versus 12 ± 4; P = NS). At 24 hours after reperfusion, the control group continued to manifest dyskinesia whereas the treated animals demonstrated positive wall motion (4.8 ± 2.7 versus -2.2 ± 1.7%; p < 0.0 I). Ventricular function continued to improve in both groups at I and 2 weeks after reperfusion with the perfluorochemical-treated dogs demonstrating greater improvement as measured by both regional radial shortening and global ejection fraction. At 2 weeks, radial shortening in the jeopardized zone was 15.3 ± 2.8% in the treatment group compared with only 5.2 ± 2.1% in the control group (p < 0.01). This was associated with a significantly greater global ejection fraction in perfluorochemical-treated animals (0.48 ± 0.10 versus 0.33 ± 0.09; p < 0.05).
Correlation of infarct size and radial shortening (Fig. 7). A loose correlation was noted between infarct size expressed as a percent of the risk region or of the total left ventricle and radial shortening in the ischemic zone at 2
lACC Vol. 9, No.5 May 1987: 1082-90
Figure 4. A and B, Photomicrographs taken from the area of myocardial infarction in a dog treated with intracoronary perfluorochemical, Fluosol DA. A, Note marked mononuclear cell collection interspersed in reticular collagen. The mononuclear cells show bubbly (arrows) cytoplasm from ingested perfluorochemical particles. B, Intramyocardial coronary arteries showing increased wall thickness and narrowed lumens (arrows). C and D, Sections from a control animal showing fibroblast and mononuclear reaction in areas of healing (C). D, Intramyocardial coronary vessels showing normal wall thickness (arrows). (A to D, Original magnification x 630, reduced by 25%.)
weeks after reperfusion (r = -0.6, P < 0.02 and r = - 0.6, p < 0.01, respectively). However, a tight correlation was found between infarct size and global radial shortening at 2 weeks after reperfusion (r = -0.86, P < 0.001 and r = -0.81, P < 0.001, respectively).
Effect of perfluorochemical treatment on arrhythmias. Intracoronary infusion of perfluorochemical did not alter QRS duration or the QT interval and did not tend to
l ACCVol. 9, NO.5
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
May 1987:1082-90
1087
Table 2. Qualitati ve Morphologic Differences in Dogs Treated With Perfluorochemical (Fluosol) or Saline Solution (Control)
Group
Hemorrhagic Infarction
Fluosol Control
1+ 3+
Collagen
Mononuclear Cells
Chronic Inflammatory Cells
Fibroblasts
Intramyocardial Coronary Artery Thickening
3+ 2+
3+ 1+
1+ 1+
2+ 3+
3+ 1+
increase the incidence of reperfusion arrhythmias. At 24 hours most dogs were noted to have runs of nonsustained ventricular tachycardia. Results of continuous Holter electrocardiographic monitoring performed for 24 hours a mean of 12 days after reperfusion demonstrated no significantdifference in ventricular ectopic activity, with < I premature ventricular complex/h present in both groups. Malignant arrhythmias (frequent runs of nonsustained ventricular tachycardia) were present in two dogs, one in each group. Histologic examination of these two animals showed no differences when compared with animals withoutmalignant arrhythmias .
Discussion Rationale for use of a reperfusion model Myocardial infarction remains a majorcauseof morbidity and mortality in the United States. Early reperfusion has been shown (1- 3,19,20) to reduce infarct size, improve ventricularfunction and decrease mortality in humans . The introduction of more effective and safer thrombolytic agents and percutaneous balloon angioplasty in evolving myocardial infarction has resulted in increasing numbers of patients undergoing successful reperfusion (21,22 ). The canine model has been extensively utilized to study the pathophysiology of regional myocardial ischemia and infarction and to assess the effects of various therapeutic
Figure 5. Serial percentage radial shortening in the ischemic zone during the experimental protocol. A significant improvement in shortening was already apparent at 24 hours after reperfu sion in the treated animals (Fluosol), and they continued to show even greater improvement of regional ventricular function over the 2 week period compared with control animals. (,?
~ 30 z w IQ:UJ Oz :Co 20 tIl ...J
FLUOSOL .. p< .05 lilt p<.OI
0.60
i= 0.50 o
ex
a:: 0.40
u..
c;E
UJ U a:: UJ a,
FlUOSOl
0
N
I-z z-
6--6 CONTROL 0--0
z
O-~
«/:u
«/:UJ Q:J: UJU C)tIl «/:-
Figure 6. A significantly greater improvement in ejection fraction was noted in the perfluorochemical-treated animals at I and 2 weeks after reperfusion .
* p<.05
6--1 CONTROL
it
interventions. Reimer et at. (4) showed that irreversible injury occurs as a " wave front" phenomenon progressing from the subendocardium to the subepicardium with increasing timeof occlusion. It thus appears that a reperfusion model would be most useful in assessing the ability of a drug to limit infarct size. Drugs that have been utilized in the reperfusion model include beta-receptor blockers (timolol), calcium antagonists (verapamil), agents that limit the production of oxygen free radicals (superoxide dismutase and catalase, allopurinol) and anti-infl ammatory agents (ibuprofen) ( 17,23- 28). However, these drugs must be administered soon after occlusion or before reperfusion to significantly reduce infarct size, and this limits their clinical applicability. Present study. We have previously shown (18) that intracoronary administration of theperfluorochemical, FIuosol DA, soon after reperfusion significantly reduces infarct size in a canine reperfusion model studied at 24 hours. Although a large amount of myocardium presumably remained stunned, a small but significant improvement in regional ventricular function was noted in the treated group. We have also shown (13) that perfluorochemical administered 30 minutes after infarction in a permanent occlusion model markedly reduced leukocytic infiltration at 3 days. Because leukocytes play an important role in the chronic and the acute inflammatory response , and because ventricular function remains depressed for weeks after an ischemic
z
10
0
0 .30
i= o 0.20
...,
UJ
0
UJ
-10
Baseline
OCclusion
3hr
24 hr
lwk
2wk
0.10
o Baseline .
Occlusion 3hr
24hr
lwk
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
1088
~
illary muscles are end arteries and are known to show fibrointimal thickening with age. Similar changes have been described in vessels supplying areas of healed infarction. Because infarct healing was more complete in perfluorochemical-treated comparedwith control dogs, intimal thickening in the former may representnormalhealing. However, we cannot exclude the possibility that arterial thickening was due to perftuorochemical treatment.
A-CONTROL O-FlUOSOl
C)
Z ...JZ
r =..... 81 n = 16
«ILl an.. .
11<·001
00:: ...JO C)%: (/)
...J
20
« c 15 « a::
10 0 A
5
10
15
20
INFARCT/LEFT VENTRICLE (0/0)
~
A-CONTROL O'FlUOSOl
C)
Z ...JZ
r = -.86
n= 14 11<.001
«ILl
lD~
00:: ...Jo
C)%: (/)
...J
20
« c 15 «
a:: 10 0 B
A
A 10
20
30
x
40
JACC Vol. 9. No.5 May 1987:1082-90
50
INFARCT/RISK REGION ("I.)
Figure 7. Relation between infarct size expressed both as a percentage of left ventricle (A) and area at risk (8) and global radial shortening 2 weeks after reperfusion. A close correlation was present.
episode, we assessed the effect of intracoronary FluosolDA on infarct size, infarct morphology and serial ventricular function in a 2 week canine reperfusion model. This study demonstrates that low dose intracoronary administration of the perftuorochemical FluosolDA reduced infarct size by 63% in a reperfusion model studied at 2 weeks. The degree of salvage was comparablewith findings in our short-term study (18). Infarct healing is a function of infarct size, with smaller infarcts healing in a shorter time than larger infarcts. Therefore, areas of hemorrhagic necrosis were more often present (seven of nine) in control dogs than in the experimental group (three of nine). However, fibroblast and coIlagen deposition was histologically similar in both groups. The most striking differences were noted in the morphology and number of the mononuclear ceIls present. Perfluorochemical-treated animals had swollen mononuclear cells with bubbly cytoplasm and eccentric nuclei, most prominent in the border zones and extensively interspersed between fibroblast and reticular collagen. MononuclearceIls in control animals were not as prominent as in perfluorochemical-treated animals and were not vacuolated. Chronic inflammatory cells were infrequently seen in both groups. Intramyocardial coronary vessels within the area of infarction in perfluorochemical-treated animals (five of nine)
showed fibrointimal thickening. Vessels supplying the pap-
Infarct size reduction was associated with significantly greater improvement in ventricular function in the jeopardized zone and global ejection fraction in the treated group.
The improvement seen at 24 hours was comparable with that in our short-term study, and the perfluorochemicaltreated dogs continued to show greater improvement over the study period. By 2 weeks, radial shortening in the ischemic zone had returned to 72% of baseline shortening in the treated group compared with a 33% return in the control group. A tight correlation was present between infarct size and global radial shortening at 2 weeks. Continuous Holter monitoring performed 10 to 13 days after reperfusion showed no significant difference in ventricular ectopic activity in the two treatment groups. Two animals, one in the control group and one in the perfluorochemical group, were noted "to have malignant arrhythmias. Except for a transient elevation of left ventricular fiIling pressure immediately after reperfusion, no significant hemodynamic or electrocardiographiceffects were noted with intracoronary administration of perfluorochemical at a dose of 15 ml/kg. Reperfusion injury. Reperfusion injury refers to the anatomic and metabolic changes occurring with reperfusion that may be detrimental and thus limit the amount of potentially salvageable myocardium (8-12). Although the exact mechanisms of reperfusion injury are not understood, possibilities include abnormalities in fluid and electrolyte homeostasis, "the no reflow phenomenon," endothelialcell damage, neutrophil-mediated injury and free radical production (8-11 ,29). The "no reflow' phenomenon refers to an incomplete return of blood flow to areas of the microcirculation after reperfusion (9). Postulated mechanisms includeperivascular
swelling, interstitial hemorrhage, endothelial cell damage and plugging of the microcirculation by neutrophils (9,10,29). Endothelial cells are known to modulate the activity of the coronary vascular bed to various endogenous compounds by the release of endothelial cell-derived relaxation factor (30-33). Thirty minutes of global ischemia in the isolated perfused rabbit heart results in abnormal vascular integrity of the endothelium I hour after reperfusion (34). When arteries in an area of myocardial infarction were examined functionally, they behaved as if their endothelium had been removed (35). Neutrophil plugs may also contribute to the "no reflew" phenomenon (10). By adhering to endothelial cells, neutrophils could potentiate endothelial damage by the release of free radicals and proteolyticenzymes (36,37).
JACC Vol. 9, No.5 May 1987: 1082-90
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
The production of large amounts offree radicals during reperfusion may also contribute to reperfusion injury (JJ,38). Free radicals are highly reactive substances that are toxic to both cellular membranes and organelles. The source of free radicals has not been clearly defined but they could be generated within endothelial, myocardial or white cells (11,38). Studies using both xanthine oxidase inhibitors and specific free radical metabolizing enzymes have demonstrated a significant reduction in infarct size in reperfusion models (26,28). Mechanism of action of perfluorochemical. The exact mechanism of action of perfluorochemical in limiting infarct size in the reperfusion model is not known. Because of its small particle size and low viscosity, perfluorochemical would be expected to increase oxygen delivery to the ischemic tissue. This effect has been demonstrated in a permanent occlusion model (39). It is possible that in a reperfusion model, perfluorochemical acts by delivering oxygen to areas of the microcirculation inaccessible to red blood cells as a result of the "no reflew" phenomenon. We have demonstrated ( 18) that regional myocardial blood flow as measured by microspheres increases immediately after reperfusion in the border zone but not at I hour, thus suggesting that other mechanisms may be operative. There are several mechanisms by which neutrophils may mediate reperfusion injury: microcirculatory plugging, the release of free radicals and proteolytic enzymes and the regulation of vascular permeability by Ieukotrienes (10,11,36,37,40). Neutropenia and anti-inflammatory agents reduce neutrophil infiltration in the marginal zone and significantly reduce infarct size in a reperfusion dog model (16,17). We have previously demonstrated (13) that large doses of perfluorochemical administered intravenously significantly reduce neutrophil infiltration in the border zones of a 3 day permanent occlusion model. We have also shown (14,15) that the perfluorochemical mixture reduces neutrophil phagocytosis, chemotaxis and free radical formation in vitro. Lane and Lamkin (41) demonstrated that the detergent component of Fluosol (Pluronic F68) markedly inhibits neutrophil chemotaxis and adherence in vitro and in vivo at concentrations similar to those administered in this study. The infarct size reduction in this experiment may therefore be related to the effect of perfluorochemical on neutrophil function. By interfering with neutrophil chemotaxis, perfluorochemical may decrease cellular damage resulting from free radical or proteolytic enzyme release, or both. Also, by blocking adherence, perfluorochemical may limit microcirculatory plugging and neutrophil-endothelial interactions. It is also possible that perfluorochemical may interfere with free radical production in endothelial cells or myocytes. However, the specific mechanisms by which perfluorochemicals affect metabolic pathways producing free radicals remain to be defined. Electron microscopy has shown that perfluorochemical
1089
particles are present in endothelial cells in the permanent occlusion model (13). It is conceivable that by preserving the anatomic or functional integrity, or both, of endothelial cells, ischemic injury may be reduced especially at the microcirculatory level. In the reperfusion model the endothelial cell swelling that has been described as contributing to the "no reflew" phenomenon may actually be abolished by intracoronary perfluorochemical therapy at the time of reperfusion. Clinical implications. The discovery of fibrin-specific thrombolytic agents has resulted in reperfusion as a logical maneuver in limiting infarct size in humans. The addition of an agent at the time of reperfusion that could enhance myocardial salvage would therefore have important therapeutic implications. This study demonstrates that perfluorochemical given soon after reperfusion significantly enhances myocardial salvage compared with reperfusion alone. This salvage was associated with near normal infarct healing and improved regional ventricular function in the jeopardized zone. Intracoronary perfluorochemical may be a potentially useful agent in enhancing myocardial salvage after successful reperfusion. We thank Andrew Manlove. James Phillips and Donna Bostick for expert technical assistance and Linda Grayson for manuscript preparation.
References I. Ganz W. Geft I, Maddahi 1, et al. Nonsurgical reperfusion in evolving myocardial infarction. J Am Coli Cardiol 1983;1:1247-53. 2. Anderson JL, Marshall HW, Bray BE. et al. A randomized trial of intracoronary streptokinase in the treatment of acute myocardial infarction. N Engl J Med 1983;308:1312-8. 3. Stack RS, Phillips III HR. Grierson OS, et al. Functional improvement of jeopardized myocardium following intracoronary streptokinase infusion in acute myocardial infarction. J Clin Invest 1983;72:84-95. 4. Reimer KA. Lowe JE. Rasmussen MM, Jennings RB. The wavefront phenomenon of ischemic cell death. I. Myocardial infarct size vs. duration of coronary occlusion in dogs. Circulation 1983;56:786-94. 5. Kloner RA, Ellis SG, Lange R. Braunwald E. Studies of experimental coronary artery reperfusion: effects on infarct size. myocardial function. biochemistry, ultrastructure, and microvascular damage. Circulation 1983;68(suppl 1):1-8-15. 6. Karsch KR. Hofmann M, Rentrop KP, Blanke H. Schaper W. Thrombolysis in acute experimental myocardial infarction. J Am Coli Cardiol 1983:1:427-35. 7. Lavallee M, Cox D, Patrick TA, Vatner SF. Salvage of myocardial function by coronary artery reperfusion, I, 2 and 3 hours after occlusion in conscious dogs. Circ Res 1983;53:235-47. 8. Braunwald E, Kloner RA. Myocardial reperfusion: a double edged sword. J Clin Invest 1985;76: 1713-9. 9. Kloner RA, Ganote CE, Jennings RB. The "no-reflew" phenomenon after temporary occlusion in the dog. J Clin Invest 1974;54: 1496-508. 10. Engler RL, Schmid-Schonbein GW, Pavelec RS. Leukocyte capillary plugging in myocardial ischemia and reperfusion in the dog. Am J Pathol 1983: 111:98-111. II. McCord JM. Oxygen-derived free radicals in postischemic tissue injury. N Engl J Med 1985;312:159-63.
1090
FORMAN ET AL. INFARCT SIZE AND VENTRICULAR FUNCTION WITH FLUOSOL-DA
12. Zimmerman ANE, Daems W, Hulsmann WC, et at. Morphological changes of heart muscle caused by successive perfusion with calciumfree and calcium containing solution (calcium paradox). Cardiovasc Res 1967;1:201-9. 13. Kolodgie FD, Dawson AK, Forman MB, Vinnani R. Effect of perfluorochemical Fluosol-DA on infarct morphology in dogs with permanent coronary artery occlusion. Virchows Arch [Cell Pathol) 1985;50: 119-34. 14. Virmani R, Fink LM, Gunter K, English D. Effects of perfluorochemicals on human neutrophil function. Transfusion 1984;24:343-7. 15. Virmani R, Warren D, Rees R, Fink LM, English D. Effect of perfluorochemical on phagocytic function of leukocytes. Transfusion 1983;23:512-25. 16. Romson Jl., Hook BG, Kunkel SL, et at. Reduction of the extent of ischemic myocardial injury by neutrophil depletion in the dog. Circulation 1983;67:1016-23. 17. Romson 1, Hook B, Rigot V, et at. The effect of ibuprofen on accumulation of indium-III myocardial platelets and leukocytes in experimental infarction. Circulation 1982;66: 1002-11. 18. Forman MB, Bingham S, Kopelman HA, et at. Reduction of infarct size with intracoronary perfluorochemical in a canine preparation of reperfusion. Circulation 1985;71: 1060-8. 19. Kennedy lW, Ritchie lL, Davis KB, Stadius ML, Maynard C, Fritz lK. The Western Washington randomized trial of intracoronary streptokinase in acute myocardial infarction. A 12-month followup report. N Engl 1 Med 1985;312:1073-8. 20. Koren G, Weiss AT, Hasin Y, et at. Prevention of myocardial damage in acute myocardial ischemia by early treatment with intravenous streptokinase. N Engl 1 Med 1985;312:1384-9. 21. TlMI Study Group. The thrombolysis in myocardial infarction (TIMI) trial phase I findings. N Engl 1 Med 1985;312:932-6. 22. O'Neill W, Timmis G, Bourdillon PD, et at. A prospective randomized clinical trial of intracoronary streptokinase versus coronary angioplasty for acute myocardial infarction. N Engl 1 Med 1986;314:812-8. 23. Hammerman H, Kloner RA, Briggs LL, Braunwald E. Enhancement of salvage of reperfused myocardium by early beta-adrenergic blockade (timolol). 1 Am Coli Cardiol 1984;3:1438-43. 24. Lo HM, Kloner RA, Braunwald E. Effect of intracoronary verapamil in infarct size in the ischemic, reperfusion canine heart: clinical importance of the timing of treatment. Am J Cardiol 1985;56:672-7. 25. Reimer KA, lennings RB. Verapamil in two reperfusion models of myocardial infarction. Temporary protection of severely ischemic myocardium without limitation of ultimate infarct size. Lab Invest 1984;5I :655-66. 26. lolly SR, Kane Wl, Bailie MB, Abrams GD, Lucchesi BR. Canine
lACC Vol. 9, No.5
May 1987: 1082-90
myocardial reperfusion. Its reduction by the combined administration of superoxide dismutase and catalase. Circ Res 1984;54:277-85. 27. Burton KP. Superoxide dismutase enhances recovery following myocardial ischemia. Am 1 Physiol 1985;248:H637-43. 28. Werns SW, Shea Ml, Mitsos SE, et at. Reduction of the size of infarction by allopurinol in the ischemic-reperfused heart. Circulation 1986;73:518-26. 29. lennings RB, Schaper 1, Hill ML, Steenbergen lC, Reimer KA. Effects of reperfusion late in the phase of reversible ischemic injury. Changes in cell volume, electrolytes, metabolites and ultrastructure. Circ Res 1985;56:262-78. 30. Cohen RA, Shephard JT, Vanhoutte PM. Inhibitory role of the endothelium in the response of isolated coronary arteries to platelets. Science 1983;221:272-4. 31. DeMey IG, Caeys M, Vanhoutte PM. Endothelium-dependent inhibitory effects of acetylcholine, adenosine triphosphate, thrombin and arachidonic acid in the canine femoral artery. 1 Phannacol Exp Ther 1982;222:166-73. 32. Furchgott RF. The requirement of endothelial cells in relaxation of arteries by acetylcholine and some other vasodilators. Trends Pharmacol Sci 1981;2:173-6. 33. DeMey 10, Vanhoutte PM. Contribution of the endothelium to the response to anoxia in the canine femoral artery. Arch Int Pharmacodyn Ther 198I;253:325-6. 34. Tilton RG, Larson KB, Udell lR, Sobel BE, Williamson IF. External detection of early microvascular dysfunction after no-flow ischemia followed by reperfusion in isolated rabbit hearts. Circ Res 1983;52: 210-25. 35. Ku DD. Coronary vascular reactivity after acute myocardial ischemia. Science 1982;218:576-8. 36. Sacks T, Moldow CF, Craddock PR, Bowers TK, lacob HS. Oxygen radicals mediate endothelial damage by complement-stimulated granulocytes. 1 Clin Invest 1978;61:1161-7. 37. Harlan 1M. Leukocyte-endothelial interactions. Blood 1985;65:513-25. 38. Del Maestro RF. An approach to free radicals in medicine and biology. Acta Physiol Scand 1980;492:153-68. 39. Rude RE, Glogar D, Khuri SF, et at. Effects of intravenous fluorocarbons during and without oxygen enhancement on acute myocardial ischemic injury assessed by measurement of intramyocardial gas tensions. Am Heart J 1982;103:986-95. 40. Davies P, Bailey Pl, Goldenberg MM. The role of arachidonic acid and oxygenation products in pain and inflammation. Annu Rev Immunol 1984;2:335-57. 41. Lane TA, Lamkin GE. Paralysis of phagocyte migration due to an artificial blood substitute. Blood 1984;64:400-5.