Role of Delayed lntraaortic Balloon Pumping in Treatment of Experimental Myocardial Infarction
ARTHUR J. ROBERTS, MD’ DANIEL R. ALONSO, MD JOHN R. COMBES, BA JEROME G. JACOBSTEIN, MD MARTIN R. POST, MD, FACC PATRICK T. CAHILL, PhD SHEAN-LAN T. HO, PhD RONALD M. ABEL, MD VALAVANUR A. SUBRAMANIAN, MD, FACC WILLIAM A. GAY, Jr., MD, FACC New York. New York
lntrsaortk balloon pumplng Improves coronary blood flow characteristics while simultaneously reducing myocardlal oxygen demands by reducing aortlc systolic pressure. Clinical applkatkn of lntraaortk balkon pumping has largely been In the “high risk” patient (cardkgenlc shock, postlnfarctlon anglna, left main coronary artery disease and unstable angina) for support during dlagnostlc studles or cardiac surgery, or both. In addltlon, there Is some evidence that balloon pumping lmmedlately after coronary occlusion reduces the size of experimentally Induced myocardlal Infarcts. In this study, myocardlal Infarcts were produced by ligation of the left anterlor descending coronary artery In 12 dogs, 6 of which were treated wlth balloon counterpulsatlon beglnnlng 3 hours after coronary occlusion. All dogs were kllled 8 hours after coronary ligation. lntraaortlc balloon pumplng resulted In the expected hemodynamlc changes (decreased aortlc systolic pressure, left ventricular end-dlastollc pressure and heart rate and Increased aortk peak diastolic pressure). In addition, there was a slgnlflcant reduction In Infarct size In the group wlth balloon pumplng as determlned with eplcardlal S-T segment mapplng, myocardlal Imaging wlth technetium-ggmglucoheptonate and hlstochemlcal stalnlng with nltroblue tetrazollum. These results suggest that even when Instituted as long as 3 hours after coronary occlusion, lntraaortlc balloon pumping results in significant reduction in Infarct size and, It might be speculated, the mortality and morbidity associated with acute myocardlal Infarction may also be decreased.
balloon pumping has been useful in the treatment of cardiogenic shock resulting from myocardial infarction.14 The effects of balloon counterpulsation on cardiac performance and oxygen consumption have been documented5*6and are related to a reduction of left ventricular afterload. However, the reported effects of balloon pumping on total coronary blood flow and the redistribution of myocardial blood flow during acute ischemia are variable.T-9 In recent years, a variety of therapeutic measures, including intraaortic balloon pumping, have been shown to preserve ischemic myocardium and reduce infarct size when applied at the time of experimental coronary occlusion.l”-I2 This study was designed to determine whether it is possible to salvage ischemic myocardium with balloon counterpulsation instituted as late as 3 hours after coronary occlusion, and whether this beneficial effect can be documented and quantified by means of scintigraphic and pathologic methods developed by our group.13J4The 3 hour interval was chosen because it represents a realistic equivalent to the delay that might occur between the onset of symptoms and institution of balloon pumping in patients with acute myocardial infarction. Intraaortic
Fromthe departmentsof Surgery(Cardlothoracic SurgeryDivision),Radiology(Nuclear Medicine), Internal Medicine(Cardiology),and Pathologyat The New York Hospital-CornellMedical Center, New York, New York. ManuscriptreceivedAugust 31, 1977; revisedmanuscriptreceivedDecember 5, 1977, accepted December 7. 1977. Pmsentad&essandad&essforreprlnts:Arthur J. Roberts, MD. Department of Surgery, Northwestern Memorial Hospital,SuperiorStreet and FairbanksCourt,Chlcago,llllnols60611.
1202
June 1978
The Amwlcan Journal of CARMOLOGY
Volume 41
DELAYED BALLOON PUMPING IN MYOCARDIAL INFARCTION-ROBERTS
ET AL.
Methods Twelve mongrel dogs were anesthetized with intravenous sodium pentobarbitol (30 mg/kg) and ventilated through a cuffed endotracheal tube with a Harvard respirator using room air. Six dogs were included in the group treated with intraaortic balloon pumping, and six served as control animals. The control dogs had a sustained ligation of the left anterior descending coronary artery and were killed 8 hours later. In the treated dogs, balloon pumping was initiated 3 hours after coronary ligation and maintained for 5 additional hours; the animals were then killed. The left carotid artery was cannulated with a rigid polyethylene catheter to monitor pressure in the aortic arch proximal to the intraaortic balloon in those animals receiving balloon pumping. A no. 6F Lehman catheter was advanced retrograde from the femoral artery into the ascending aorta in all animals and, when needed, advanced across the aortic valve into the left ventricle under direct fluoroscopic control to obtain pressure recordings. Intraaortic balloon pump: A catheter-mounted Avcothane intraaortic balloon (Avco-Everett Research Laboratory, Everett, Massachusetts) 20 cm long and 1.1 cm in diameter was introduced through the left femoral artery in balloontreated animals and positioned under fluoroscopic guidance so that its tip was located immediately distal to the aortic arch near the origin of the left subclavian artery. The balloon was inflated with 8 to 10 cc of helium triggered by an appropriate delay after the R wave of the electrocardiogram using the model IABP-7 Avco balloon pump console. Accurate synchronization of balloon inflation and deflation was accomplished by observing the arterial pressure wave form proximal to the balloon to be certain that the balloon-induced pulse wave followed the naturally occurring dicrotic notch. Anticoagulation was not used during the experiment in either the control or the balloon-treated dogs. Central aortic pressure and left ventricular end-diastolic pressure were recorded before and at regular intervals after coronary occlusion with a Statham P23Db transducer connected to an eight channel Electronics for Medicine oscillographic recorder. The standard electrocardiogram was continuously monitored in all experiments from subdermal electrodes. Epicardial S-T segment mapping: Median sternotomy was performed and the heart suspended in a pericardial cradle. The left anterior descending coronary artery was permanently occluded with a ligature at the junction of its proximal and middle thirds. A 36-square cloth grid was moistened and molded to the shape of the epicardial surface of the heart, and a hand-held hollow metal insulated electrode was gently applied to the center of each square to record epicardial electrograms 15 minutes after coronary occlusion (Fig. 1). Earlier investigations15Je have validated a direct relation between S-T segment elevation after 15 minutes of coronary occlusion and the degree of myocardial necrosis determined 24 hours after occlusion. In our laboratory, a high correlation between the magnitude of S-T segment elevation 15 minutes after coronary occlusion and infarct weight at autopsy was also found (r = 0.93).17 On this basis, epicardial S-T elevation was chosen as an index for predicting subsequent ischemic myocardial damage. Unipolar epicardial electrograms w&e obtained at 0.1 sensitivity and a permanent record made on a Hewlett-Packard 7712 recorder. The number of sites with S-T segment elevation greater than or equal to 2 mm as well as the sum of all the S-T segment elevations on the epicardial grid were noted.
Point occlus
FIGURE 1. Diagram showing the method of epicardial S-T segment mapping. The fixed point of coronary ligation and its relation to the epicardial grid constaM from dog to dog are shown. Each epicardial site measures 0.5 cm*.
On the basis of comparable S-T segment elevation recorded 15 minutes after coronary occlusion, the animals were randomized into a treatment group (6 dogs) and a control group (6 dogs). Epicardial S-T segment maps were repeated in the balloon-treated group to assess electrophysiologic improvement secondary to balloon pump&g. Values recorded before starting counterpulsation 3 hours after coronary occlusion were compared with levels observed 5,15 and 30 minutes after the initiation of balloon pumping. The control group underwent repeat epicardial S-T segment mapping in a manner similar to balloon-treated animals. Left ventriculography: This procedure was performed before and 1 hour after coronary ligation in all animals by hand injection of 10 cc of Renografin-76”. The ventriculograms were recorded on 16 mm film with a General Electric tine camera and, when projected, the ventricular outline was traced during consecutive beats at end-systole and enddiastole. Each heart tracing was divided into anterior and posterior segments by a !ine drawn from the mid point of the aortic valve to the left ventricular apex. Each segmental area was then measured with planimetry, and regional ejection fractions were determined. Thus, an index of regional ventricular function was determined for each of the two myocardial segments (anterior and posterior) by comparing the percent change in area from end-diastole and end-systole before and 1 hour after coronary occlusion. In addition, further ventriculograms were obtained in balloon-treated dogs for detailed analysis using a new skewed coordinate system’s designed in our laboratories (Cahill PT, et al.& bepublished). This new method was used in the study of serial ventriculograms that were taken in the balloontreated group before coronary occlusion, 1 and 3 hours after coronary ligation but before the initiation of balloon pumping, and finally after 3 hours of continuous counterpulsation. Although additional ventriculograms were not performed in the control group, the possible depressant effects of additional contrast material in balloon-treated animals might tend to increase rather thandecrease infarct size. With use of an acoustical graph pen, the borders of the left ventricle in the balloon-treated group were digitized and stored on disc for future computer analysis. For each of the ventriculographic images from end-diastole to end-systole (normally 12), a set of radii spanning the entire left ventricle (360”) were calculated and referenced to this new skewed coordinate system,
June 1979
The Amerlcsn Journal of CARMOLOGY
Volume 41
1203
DELAYED BALLOON PUMPING IN MYOCARDIAL INFARCTION-ROBERTS
ET AL.
FIGURE 2. Method for determining the weight of a myocardial infarct using nkroblue tetrazolium myocardial staining and planlmetry. (Reproduced from Roberts et al.*’ by permision of the American Heart Association, Inc.)
and the normalized fractional changes in length and velocity were determined for each image from end-systole and enddiastole. Local ejection fractions were also calculated at 36’ intervals around the left ventricle, and were normalized by taking the difference in volume between end-diastole and end-systole and dividing by the volume at end-diitole. These serial ventriculograms were also analyzed using conventional methods for determining ejection fraction,lg but only the new method using a skewed coordinate system could correlate abnormalities in left ventricular wall motion with location of infarct as determined from pathologic studies. Myocardial imaging and delineation of myocardial necrosis: One hour after coronary occlusion, 20 millicuries of technetium-99m glucoheptonate was injected intravenously, and images of the myocardial uptake of the radiopharmaceutical agent were obtained 7 hours after injection using a Picker Dyna Camera 2C gamma camera with a low energy high-sensitivity collimator. After the animals were killed their hearts were removed, imaged and sectioned horizontally at 7 to 10 mm intervals from apex to base, reimaged as heart slices and incubated in buffered nitroblue tetrazolium to delineate areas of myocardial necrosis.20 Color transparencies were taken of the heart slices and projected onto tracing paper to life size, and
1204
June 1978
The American Journal of CARDIOLOGY
boundaries were traced. Planimetry of the area occupied by the infarct in ventricular slices of known weight permitted calculation of the total weight of the infarct (Fig. 2). Scintigraphic infarct size was determined with planimetry of the cross-sectional area of radiopharmaceutical uptake in the right anterior oblique projection. Myocardial imaging with technetium-99m glucoheptonate allowed early and accurate delineation of myocardial necrosis (Fig. 3). The accuracy of these techniques in evaluating experimental infarct size has been previously documented.13J4J7*21 Statistical analysis: All data were expressed as mean f standard error of the mean. Hemodynamic data and S-T segment mapping were evaluated with the analysis of variance of repeated measures. The changes in the ventriculograms and the quantitation of infarct size estimated with technetium99m glucoheptonate and nitroblue tetrazolium were analyzed with the Wilcoxan test.
Results Hemodynamic changes: namic differences between balloon treated groups were blood pressure, heart rate or
Voluma 41
No significant hemodycontrol and intraaortic seen in mean values for left ventricular end-dia-
DELAYED
stolic pressure before or up to 3 hours after coronary occlusion. However, there were significant differences when values in the balloon-treated group 3 hours after coronary occlusion but before counterpulsation were compared with values 1 hour after balloon pumping: After pumping, peak diastolic pressure increased from 99 f 5 to 117 f 6 mm Hg (P
ARTHUR J. ROBERTS, MD’ DANIEL R. ALONSO, MD JOHN R. COMBES, BA JEROME G. JACOBSTEIN, MD MARTIN R. POST, MD, FACC PATRICK T. CAHILL, PhD SHEAN-LAN T. HO, PhD RONALD M. ABEL, MD VALAVANUR A. SUBRAMANIAN, MD, FACC WILLIAM A. GAY, Jr., MD, FACC New York. New York
lntrsaortk balloon pumplng Improves coronary blood flow characteristics while simultaneously reducing myocardlal oxygen demands by reducing aortlc systolic pressure. Clinical applkatkn of lntraaortk balkon pumping has largely been In the “high risk” patient (cardkgenlc shock, postlnfarctlon anglna, left main coronary artery disease and unstable angina) for support during dlagnostlc studles or cardiac surgery, or both. In addltlon, there Is some evidence that balloon pumping lmmedlately after coronary occlusion reduces the size of experimentally Induced myocardlal Infarcts. In this study, myocardlal Infarcts were produced by ligation of the left anterlor descending coronary artery In 12 dogs, 6 of which were treated wlth balloon counterpulsatlon beglnnlng 3 hours after coronary occlusion. All dogs were kllled 8 hours after coronary ligation. lntraaortlc balloon pumplng resulted In the expected hemodynamlc changes (decreased aortlc systolic pressure, left ventricular end-dlastollc pressure and heart rate and Increased aortk peak diastolic pressure). In addition, there was a slgnlflcant reduction In Infarct size In the group wlth balloon pumplng as determlned with eplcardlal S-T segment mapplng, myocardlal Imaging wlth technetium-ggmglucoheptonate and hlstochemlcal stalnlng with nltroblue tetrazollum. These results suggest that even when Instituted as long as 3 hours after coronary occlusion, lntraaortlc balloon pumping results in significant reduction in Infarct size and, It might be speculated, the mortality and morbidity associated with acute myocardlal Infarction may also be decreased.
balloon pumping has been useful in the treatment of cardiogenic shock resulting from myocardial infarction.14 The effects of balloon counterpulsation on cardiac performance and oxygen consumption have been documented5*6and are related to a reduction of left ventricular afterload. However, the reported effects of balloon pumping on total coronary blood flow and the redistribution of myocardial blood flow during acute ischemia are variable.T-9 In recent years, a variety of therapeutic measures, including intraaortic balloon pumping, have been shown to preserve ischemic myocardium and reduce infarct size when applied at the time of experimental coronary occlusion.l”-I2 This study was designed to determine whether it is possible to salvage ischemic myocardium with balloon counterpulsation instituted as late as 3 hours after coronary occlusion, and whether this beneficial effect can be documented and quantified by means of scintigraphic and pathologic methods developed by our group.13J4The 3 hour interval was chosen because it represents a realistic equivalent to the delay that might occur between the onset of symptoms and institution of balloon pumping in patients with acute myocardial infarction. Intraaortic
Fromthe departmentsof Surgery(Cardlothoracic SurgeryDivision),Radiology(Nuclear Medicine), Internal Medicine(Cardiology),and Pathologyat The New York Hospital-CornellMedical Center, New York, New York. ManuscriptreceivedAugust 31, 1977; revisedmanuscriptreceivedDecember 5, 1977, accepted December 7. 1977. Pmsentad&essandad&essforreprlnts:Arthur J. Roberts, MD. Department of Surgery, Northwestern Memorial Hospital,SuperiorStreet and FairbanksCourt,Chlcago,llllnols60611.
1202
June 1978
The Amwlcan Journal of CARMOLOGY
Volume 41
DELAYED BALLOON PUMPING IN MYOCARDIAL INFARCTION-ROBERTS
ET AL.
Methods Twelve mongrel dogs were anesthetized with intravenous sodium pentobarbitol (30 mg/kg) and ventilated through a cuffed endotracheal tube with a Harvard respirator using room air. Six dogs were included in the group treated with intraaortic balloon pumping, and six served as control animals. The control dogs had a sustained ligation of the left anterior descending coronary artery and were killed 8 hours later. In the treated dogs, balloon pumping was initiated 3 hours after coronary ligation and maintained for 5 additional hours; the animals were then killed. The left carotid artery was cannulated with a rigid polyethylene catheter to monitor pressure in the aortic arch proximal to the intraaortic balloon in those animals receiving balloon pumping. A no. 6F Lehman catheter was advanced retrograde from the femoral artery into the ascending aorta in all animals and, when needed, advanced across the aortic valve into the left ventricle under direct fluoroscopic control to obtain pressure recordings. Intraaortic balloon pump: A catheter-mounted Avcothane intraaortic balloon (Avco-Everett Research Laboratory, Everett, Massachusetts) 20 cm long and 1.1 cm in diameter was introduced through the left femoral artery in balloontreated animals and positioned under fluoroscopic guidance so that its tip was located immediately distal to the aortic arch near the origin of the left subclavian artery. The balloon was inflated with 8 to 10 cc of helium triggered by an appropriate delay after the R wave of the electrocardiogram using the model IABP-7 Avco balloon pump console. Accurate synchronization of balloon inflation and deflation was accomplished by observing the arterial pressure wave form proximal to the balloon to be certain that the balloon-induced pulse wave followed the naturally occurring dicrotic notch. Anticoagulation was not used during the experiment in either the control or the balloon-treated dogs. Central aortic pressure and left ventricular end-diastolic pressure were recorded before and at regular intervals after coronary occlusion with a Statham P23Db transducer connected to an eight channel Electronics for Medicine oscillographic recorder. The standard electrocardiogram was continuously monitored in all experiments from subdermal electrodes. Epicardial S-T segment mapping: Median sternotomy was performed and the heart suspended in a pericardial cradle. The left anterior descending coronary artery was permanently occluded with a ligature at the junction of its proximal and middle thirds. A 36-square cloth grid was moistened and molded to the shape of the epicardial surface of the heart, and a hand-held hollow metal insulated electrode was gently applied to the center of each square to record epicardial electrograms 15 minutes after coronary occlusion (Fig. 1). Earlier investigations15Je have validated a direct relation between S-T segment elevation after 15 minutes of coronary occlusion and the degree of myocardial necrosis determined 24 hours after occlusion. In our laboratory, a high correlation between the magnitude of S-T segment elevation 15 minutes after coronary occlusion and infarct weight at autopsy was also found (r = 0.93).17 On this basis, epicardial S-T elevation was chosen as an index for predicting subsequent ischemic myocardial damage. Unipolar epicardial electrograms w&e obtained at 0.1 sensitivity and a permanent record made on a Hewlett-Packard 7712 recorder. The number of sites with S-T segment elevation greater than or equal to 2 mm as well as the sum of all the S-T segment elevations on the epicardial grid were noted.
Point occlus
FIGURE 1. Diagram showing the method of epicardial S-T segment mapping. The fixed point of coronary ligation and its relation to the epicardial grid constaM from dog to dog are shown. Each epicardial site measures 0.5 cm*.
On the basis of comparable S-T segment elevation recorded 15 minutes after coronary occlusion, the animals were randomized into a treatment group (6 dogs) and a control group (6 dogs). Epicardial S-T segment maps were repeated in the balloon-treated group to assess electrophysiologic improvement secondary to balloon pump&g. Values recorded before starting counterpulsation 3 hours after coronary occlusion were compared with levels observed 5,15 and 30 minutes after the initiation of balloon pumping. The control group underwent repeat epicardial S-T segment mapping in a manner similar to balloon-treated animals. Left ventriculography: This procedure was performed before and 1 hour after coronary ligation in all animals by hand injection of 10 cc of Renografin-76”. The ventriculograms were recorded on 16 mm film with a General Electric tine camera and, when projected, the ventricular outline was traced during consecutive beats at end-systole and enddiastole. Each heart tracing was divided into anterior and posterior segments by a !ine drawn from the mid point of the aortic valve to the left ventricular apex. Each segmental area was then measured with planimetry, and regional ejection fractions were determined. Thus, an index of regional ventricular function was determined for each of the two myocardial segments (anterior and posterior) by comparing the percent change in area from end-diastole and end-systole before and 1 hour after coronary occlusion. In addition, further ventriculograms were obtained in balloon-treated dogs for detailed analysis using a new skewed coordinate system’s designed in our laboratories (Cahill PT, et al.& bepublished). This new method was used in the study of serial ventriculograms that were taken in the balloontreated group before coronary occlusion, 1 and 3 hours after coronary ligation but before the initiation of balloon pumping, and finally after 3 hours of continuous counterpulsation. Although additional ventriculograms were not performed in the control group, the possible depressant effects of additional contrast material in balloon-treated animals might tend to increase rather thandecrease infarct size. With use of an acoustical graph pen, the borders of the left ventricle in the balloon-treated group were digitized and stored on disc for future computer analysis. For each of the ventriculographic images from end-diastole to end-systole (normally 12), a set of radii spanning the entire left ventricle (360”) were calculated and referenced to this new skewed coordinate system,
June 1979
The Amerlcsn Journal of CARMOLOGY
Volume 41
1203
DELAYED BALLOON PUMPING IN MYOCARDIAL INFARCTION-ROBERTS
ET AL.
FIGURE 2. Method for determining the weight of a myocardial infarct using nkroblue tetrazolium myocardial staining and planlmetry. (Reproduced from Roberts et al.*’ by permision of the American Heart Association, Inc.)
and the normalized fractional changes in length and velocity were determined for each image from end-systole and enddiastole. Local ejection fractions were also calculated at 36’ intervals around the left ventricle, and were normalized by taking the difference in volume between end-diastole and end-systole and dividing by the volume at end-diitole. These serial ventriculograms were also analyzed using conventional methods for determining ejection fraction,lg but only the new method using a skewed coordinate system could correlate abnormalities in left ventricular wall motion with location of infarct as determined from pathologic studies. Myocardial imaging and delineation of myocardial necrosis: One hour after coronary occlusion, 20 millicuries of technetium-99m glucoheptonate was injected intravenously, and images of the myocardial uptake of the radiopharmaceutical agent were obtained 7 hours after injection using a Picker Dyna Camera 2C gamma camera with a low energy high-sensitivity collimator. After the animals were killed their hearts were removed, imaged and sectioned horizontally at 7 to 10 mm intervals from apex to base, reimaged as heart slices and incubated in buffered nitroblue tetrazolium to delineate areas of myocardial necrosis.20 Color transparencies were taken of the heart slices and projected onto tracing paper to life size, and
1204
June 1978
The American Journal of CARDIOLOGY
boundaries were traced. Planimetry of the area occupied by the infarct in ventricular slices of known weight permitted calculation of the total weight of the infarct (Fig. 2). Scintigraphic infarct size was determined with planimetry of the cross-sectional area of radiopharmaceutical uptake in the right anterior oblique projection. Myocardial imaging with technetium-99m glucoheptonate allowed early and accurate delineation of myocardial necrosis (Fig. 3). The accuracy of these techniques in evaluating experimental infarct size has been previously documented.13J4J7*21 Statistical analysis: All data were expressed as mean f standard error of the mean. Hemodynamic data and S-T segment mapping were evaluated with the analysis of variance of repeated measures. The changes in the ventriculograms and the quantitation of infarct size estimated with technetium99m glucoheptonate and nitroblue tetrazolium were analyzed with the Wilcoxan test.
Results Hemodynamic changes: namic differences between balloon treated groups were blood pressure, heart rate or
Voluma 41
No significant hemodycontrol and intraaortic seen in mean values for left ventricular end-dia-
DELAYED
stolic pressure before or up to 3 hours after coronary occlusion. However, there were significant differences when values in the balloon-treated group 3 hours after coronary occlusion but before counterpulsation were compared with values 1 hour after balloon pumping: After pumping, peak diastolic pressure increased from 99 f 5 to 117 f 6 mm Hg (P
