Retrograde Coronary Capillary Perfusion for Prevention and Reversal of Cardiogenic Shock in Experimental Myocardial Infarctionu Blase A . Carabello, M.D., Gerald M. Lemole, M.D., Kang W. Lee, M.D., a n d James F. Spann, M.D.
The coronary sinus and coronary veins offer an access route for delivery of increased oxygen to ischemic myocardium surrounding the central dead zone of heart muscle in cardiogenic shock due to myocardial infarction. This investigation was conducted to determine if transvenous retrograde coronary capillary perfusion with oxygenated blood would prevent and reverse cardiogenic shock in experimental myocardial infarction produced by acute ligation of the circumflex and anterior descending left coronary artery in dogs. Cardiac output and systemic blood pressure were maintained near control values for up to 30 minutes when total left coronary ligation was accompanied by coronary retroperfusion. Conversely, both cardiac output and systemic blood pressure fell to severe cardiogenic shock levels within 2 minutes of total left coronary artery occlusion without retrograde flow or when retrograde flow was terminated 5 to 30 minutes following simultaneous coronary ligation and institution of retrograde flow. All animals undergoing left coronary ligation without retrograde flow died within 5 minutes from ventricular fibrillation, while this occurred in only 1of 10 animals in which retrograde flow was started simultaneously with coronary occlusion. Cardiogenic shock was reversed by coronary retrograde perfusion in each of the 2 experiments in which this could be tested; oxygen was extracted by the myocardium from blood passing retrograde through the coronary capillaries. It is concluded that coronary retrograde flow will maintain cardiac pump performance and prevent the development of cardiogenic shock following experiABSTRACT
From the Sections of Cardiothoracic Surgery and Cardiology, Temple University Health Sciences Center, Philadelphia, PA. Supported in part by Grants no. lROlHL13876 and no. 5TplHL5712, National Institutes of Health. Accepted for publication Sept 3, 1975. Address reprint requests to Dr. Lemole, Section of Cardiothoracic Surgery, Temple University Health Sciences Center, Philadelphia, PA 19140.
405
mental left anterior descending and circumflex coronary artery occlusion in the working canine heart.
Fifteen percent of patients with myocardial infarction develop the clinical syndrome of cardiogenic shock including hypotension, oliguria, and cerebral dysfunction [20]. The mortality once the shock syndrome appears is 75 to 95% despite all pharmacological therapy [2]. It is now well accepted that the reduction in cardiac output that causes cardiogenic shock is due to decreased contractile performance of the left ventricular myocardium [15, 193, which in turn is related to myocardial ischemia produced by the coronary artery occlusion. Although the central area of necrosis is not retrievable, this dead area is surrounded by a zone of heart muscle with less ischemia and more potential for viability despite severe contractile dysfunction [ H I . Improvement in the oxygen supply of this potentially viable area of the heart with a resultant increase in its contractile function may reverse cardiogenic shock. Previous attempts to salvage this dysfunctioning but viable muscle have been directed principally toward reduction of oxygen demand or toward attempts to increase oxygen delivery by augmenting coronary arterial flow [171. However, since the artery through which this additional blood must flow is compromised by the atherosclerosis that caused the infarction, the arterial vessel is of limited value as a route for delivery of increased oxygen. The coronary veins offer an unobstructed route to the capillaries of the ischemic area. Accordingly, the present investigation was conducted to determine if transvenous retrograde coronary capillary perfusion with oxygenated blood would prevent and reverse cardiogenic shock in experimental myocardial infarction produced by temporary ligation of the cir-
406 The Annals of Thoracic Surgery Vol 21 No 5 May 1976
cumflex and anterior descending left coronary arteries in dogs.
temperature was maintained at 37°C by heating pads. A 16F Foley catheter was introduced into the right atrium through a pursestring incision. The left and right femoral veins were cannulated and connected to the Foley catheter and to a pump oxygenator (Travenol) to allow blood from the left femoral vein to be oxygenated and pumped either into the right femoral vein or through the Foley catheter (see Fig 1). A calibrated electromagnetic flow probe (Biotronex) was placed around the proximal ascending aorta for measurement of stroke volume and cardiac output. A cannula was introduced into the proximal aorta through the left carotid artery and attached to a pressure transducer (HewlettPackard Model 1280) for measurement of aortic pressure. A multichannel oscillograph was used to record simultaneous phasic and mean aortic pressure, high- and low-gain left ventricular pressure, the first derivative of left ventricular pressure, flow in the ascending aorta, and lead I1 of the electrocardiogram. The Foley catheter was then advanced into the coronary sinus beyond the entrance of the right coronary veins, and the catheter balloon was inflated with enough water to occlude the coronary sinus totally. Blood was drained from the coronary sinus through the catheter lumen and returned to the superior vena cava until the experimental protocol was initiated. This made possible two modes of operation. The first, with
Methods Mongrel dogs weighing approximately 18 kg were anesthetized intravenously with sodium pentobarbital (25 mglkg), intubated, and placed on a mechanical respirator. The anterior chest wall was removed following ligation of the intercostal artery and the internal mammary artery and vein, and the heart was exposed and supported in a pericardial cradle. The left anterior descending and left circumflex coronary arteries were isolated, and ligatures were placed loosely around each vessel in preparation for subsequent occlusion (Fig 1). Systemic heparinization was then accomplished with 3 mg per kilogram of sodium heparin. The electrocardiogram was obtained from four limb leads, and body Fig 1 . Experimental preparation usedfor coronary occlusion and coronary sinus retrograde perfusion. T h e coronary sinus and coronary veins are s h o w n as dashed lines. Proximal outflow f r o m the coronary sinus is prevented during retrograde f l o w b y a n occluding balloon on the perfusion catheter. (SVC = superior vend cava; RA = right atrium; RV = right ventricle; LV = left ventricle; IVC = inferior vena cava; RFV = rightfemoral vein; LFV = left femoral vein; LVEDP = left ventricular end-diastolicpressure; LVP = lcft ventricular systolicpressure; dpidt = first derivative of left ventricular pressure.)
AORTIC PRESS VALVE #3 VALVE #1 1 SVC VALVE # 2 \
'/
'
AORTIC FLOW PROBE //CIRCUMFLEX '
t
O R O N A R Y ART LIGATURES
1
h
oxvd-
..
CORONARY ARTERY
\ 'V
407 Carabello et al: Retrograde Coronary Capillary Perfusion
valves number 2 and 3 open (see Fig 1)and valve number 1closed, provided a control condition in which blood was removed from the left femoral vein, oxygenated in the extracorporeal circuit, and returned to the right femoral vein while coronary sinus drainage was returned to the superior vena cava. The second mode, with valves number 2 and 3 closed and valve number 1 open (see Fig l),provided retrograde perfusion of the coronary sinus with oxygenated blood; blood was removed from the left femoral vein, oxygenated, and pumped retrograde through the coronary sinus, veins, and capillaries. The coronary sinus retrograde flow rate was set to equal previously measured antegrade coronary sinus blood flow, which was between 50 and 100 ml per minute. Perfusion pressures in the coronary sinus setup were of systemic magnitude. Blood oxygen content was determined by the Van Slyke technique. Water content of the myocardium was determined by comparison of tissue weight before and after drying at 60°C for 72 hours in an oven. Student's paired t-test was used for statistical analysis of the data, and a p value less than 0.05 was considered significant.
Results Three groups of animals were studied: (1)cardiogenic shock-the effect of ligating the circumflex and left anterior descending coronary arteries was determined in the absence of retrograde coronary capillary perfusion; (2) cardiogenic shock prevention-both vessels were ligated, and coronary retrograde perfusion was started simultaneously and continued until it was terminated electively after 5 to 30 minutes; (3)cardiogenic shock reversal-cardiogenic shock was produced by terminating retrograde coronary capillary perfusion with both the left anterior descending and the circumflex artery ligated, and shock was reversed by reinstitution of coronary capillary retrograde perfusion. Cardiogenic Shock Experiments The 3 animals in this group demonstrated rapid deterioration of cardiac performance with a fall in aortic pressure to cardiogenic shock levels within 2 minutes of ligation of both branches of the left coronary artery, and each animal died rapidly in ventricular fibrillation. The left ven-
tricle became grossly cyanotic and hypodynamic, and the average systolic aortic pressure decreased from control values of 100 f 25 (SEM) to 38 f 8 mm Hg (p < 0.01) within 2 minutes, while mean aortic pressure fell from 71 f 15 to 24 f 6 mm Hg (p < 0.01). Cardiac output declined from the controlvalue of 0.6 k 0.2 to 0.2 f 0.1 liter per minute within 2 minutes of left coronary artery ligation (p < 0.01). Ventricular fibrillation appeared within 2 to 5 minutes of coronary ligation in each animal and prevented further studies in this group. Cardiogenic Shock Prevention Experiments In 10 animals coronary capillary retrograde perfusion was instituted simultaneous with ligation of the circumflex and left anterior descending coronary artery. Documentation of adequate retrograde flow was obtained by visualizing oxygenated blood in the distal coronary veins in each animal. In 9 of the 10 animals, cardiac function was sustained and the experiment was continued until elective termination of retrograde perfusion 5 to 30 minutes later; ventricular fibrillation did not occur in these animals. Despite retrograde perfusion 1 animal died of ventricular fibrillation within 5 minutes without data collection. To document the dependence of the left ventricle on coronary retrograde flow, coronary occlusion was continued and the coronary retrograde perfusion was electively terminated after an average of 14 minutes in 6 animals. In the other 3 animals the experiments were continued until electively terminated at 30 minutes to determine if longer support of the ventricle was possible and to obtain ventricular tissue to examine for edema. The following results were observed in the 9 successfully completed studies. Aortic pressure and cardiac output were maintained during retrograde perfusion despite circumflex and left anterior descending coronary artery ligation (Figs 2, 3). When retrograde perfusion was stopped but coronary artery occlusion continued, both aortic pressure and cardiac output fell to cardiogenic shock levels within 2 minutes. After an average of 14 minutes of left coronary ligation and retrograde perfusion, peak aortic pressure averaged 89 f 6 mm Hg compared with a control value of 100 f 8 ( p
The coronary sinus and coronary veins offer an access route for delivery of increased oxygen to ischemic myocardium surrounding the central dead zone of heart muscle in cardiogenic shock due to myocardial infarction. This investigation was conducted to determine if transvenous retrograde coronary capillary perfusion with oxygenated blood would prevent and reverse cardiogenic shock in experimental myocardial infarction produced by acute ligation of the circumflex and anterior descending left coronary artery in dogs. Cardiac output and systemic blood pressure were maintained near control values for up to 30 minutes when total left coronary ligation was accompanied by coronary retroperfusion. Conversely, both cardiac output and systemic blood pressure fell to severe cardiogenic shock levels within 2 minutes of total left coronary artery occlusion without retrograde flow or when retrograde flow was terminated 5 to 30 minutes following simultaneous coronary ligation and institution of retrograde flow. All animals undergoing left coronary ligation without retrograde flow died within 5 minutes from ventricular fibrillation, while this occurred in only 1of 10 animals in which retrograde flow was started simultaneously with coronary occlusion. Cardiogenic shock was reversed by coronary retrograde perfusion in each of the 2 experiments in which this could be tested; oxygen was extracted by the myocardium from blood passing retrograde through the coronary capillaries. It is concluded that coronary retrograde flow will maintain cardiac pump performance and prevent the development of cardiogenic shock following experiABSTRACT
From the Sections of Cardiothoracic Surgery and Cardiology, Temple University Health Sciences Center, Philadelphia, PA. Supported in part by Grants no. lROlHL13876 and no. 5TplHL5712, National Institutes of Health. Accepted for publication Sept 3, 1975. Address reprint requests to Dr. Lemole, Section of Cardiothoracic Surgery, Temple University Health Sciences Center, Philadelphia, PA 19140.
405
mental left anterior descending and circumflex coronary artery occlusion in the working canine heart.
Fifteen percent of patients with myocardial infarction develop the clinical syndrome of cardiogenic shock including hypotension, oliguria, and cerebral dysfunction [20]. The mortality once the shock syndrome appears is 75 to 95% despite all pharmacological therapy [2]. It is now well accepted that the reduction in cardiac output that causes cardiogenic shock is due to decreased contractile performance of the left ventricular myocardium [15, 193, which in turn is related to myocardial ischemia produced by the coronary artery occlusion. Although the central area of necrosis is not retrievable, this dead area is surrounded by a zone of heart muscle with less ischemia and more potential for viability despite severe contractile dysfunction [ H I . Improvement in the oxygen supply of this potentially viable area of the heart with a resultant increase in its contractile function may reverse cardiogenic shock. Previous attempts to salvage this dysfunctioning but viable muscle have been directed principally toward reduction of oxygen demand or toward attempts to increase oxygen delivery by augmenting coronary arterial flow [171. However, since the artery through which this additional blood must flow is compromised by the atherosclerosis that caused the infarction, the arterial vessel is of limited value as a route for delivery of increased oxygen. The coronary veins offer an unobstructed route to the capillaries of the ischemic area. Accordingly, the present investigation was conducted to determine if transvenous retrograde coronary capillary perfusion with oxygenated blood would prevent and reverse cardiogenic shock in experimental myocardial infarction produced by temporary ligation of the cir-
406 The Annals of Thoracic Surgery Vol 21 No 5 May 1976
cumflex and anterior descending left coronary arteries in dogs.
temperature was maintained at 37°C by heating pads. A 16F Foley catheter was introduced into the right atrium through a pursestring incision. The left and right femoral veins were cannulated and connected to the Foley catheter and to a pump oxygenator (Travenol) to allow blood from the left femoral vein to be oxygenated and pumped either into the right femoral vein or through the Foley catheter (see Fig 1). A calibrated electromagnetic flow probe (Biotronex) was placed around the proximal ascending aorta for measurement of stroke volume and cardiac output. A cannula was introduced into the proximal aorta through the left carotid artery and attached to a pressure transducer (HewlettPackard Model 1280) for measurement of aortic pressure. A multichannel oscillograph was used to record simultaneous phasic and mean aortic pressure, high- and low-gain left ventricular pressure, the first derivative of left ventricular pressure, flow in the ascending aorta, and lead I1 of the electrocardiogram. The Foley catheter was then advanced into the coronary sinus beyond the entrance of the right coronary veins, and the catheter balloon was inflated with enough water to occlude the coronary sinus totally. Blood was drained from the coronary sinus through the catheter lumen and returned to the superior vena cava until the experimental protocol was initiated. This made possible two modes of operation. The first, with
Methods Mongrel dogs weighing approximately 18 kg were anesthetized intravenously with sodium pentobarbital (25 mglkg), intubated, and placed on a mechanical respirator. The anterior chest wall was removed following ligation of the intercostal artery and the internal mammary artery and vein, and the heart was exposed and supported in a pericardial cradle. The left anterior descending and left circumflex coronary arteries were isolated, and ligatures were placed loosely around each vessel in preparation for subsequent occlusion (Fig 1). Systemic heparinization was then accomplished with 3 mg per kilogram of sodium heparin. The electrocardiogram was obtained from four limb leads, and body Fig 1 . Experimental preparation usedfor coronary occlusion and coronary sinus retrograde perfusion. T h e coronary sinus and coronary veins are s h o w n as dashed lines. Proximal outflow f r o m the coronary sinus is prevented during retrograde f l o w b y a n occluding balloon on the perfusion catheter. (SVC = superior vend cava; RA = right atrium; RV = right ventricle; LV = left ventricle; IVC = inferior vena cava; RFV = rightfemoral vein; LFV = left femoral vein; LVEDP = left ventricular end-diastolicpressure; LVP = lcft ventricular systolicpressure; dpidt = first derivative of left ventricular pressure.)
AORTIC PRESS VALVE #3 VALVE #1 1 SVC VALVE # 2 \
'/
'
AORTIC FLOW PROBE //CIRCUMFLEX '
t
O R O N A R Y ART LIGATURES
1
h
oxvd-
..
CORONARY ARTERY
\ 'V
407 Carabello et al: Retrograde Coronary Capillary Perfusion
valves number 2 and 3 open (see Fig 1)and valve number 1closed, provided a control condition in which blood was removed from the left femoral vein, oxygenated in the extracorporeal circuit, and returned to the right femoral vein while coronary sinus drainage was returned to the superior vena cava. The second mode, with valves number 2 and 3 closed and valve number 1 open (see Fig l),provided retrograde perfusion of the coronary sinus with oxygenated blood; blood was removed from the left femoral vein, oxygenated, and pumped retrograde through the coronary sinus, veins, and capillaries. The coronary sinus retrograde flow rate was set to equal previously measured antegrade coronary sinus blood flow, which was between 50 and 100 ml per minute. Perfusion pressures in the coronary sinus setup were of systemic magnitude. Blood oxygen content was determined by the Van Slyke technique. Water content of the myocardium was determined by comparison of tissue weight before and after drying at 60°C for 72 hours in an oven. Student's paired t-test was used for statistical analysis of the data, and a p value less than 0.05 was considered significant.
Results Three groups of animals were studied: (1)cardiogenic shock-the effect of ligating the circumflex and left anterior descending coronary arteries was determined in the absence of retrograde coronary capillary perfusion; (2) cardiogenic shock prevention-both vessels were ligated, and coronary retrograde perfusion was started simultaneously and continued until it was terminated electively after 5 to 30 minutes; (3)cardiogenic shock reversal-cardiogenic shock was produced by terminating retrograde coronary capillary perfusion with both the left anterior descending and the circumflex artery ligated, and shock was reversed by reinstitution of coronary capillary retrograde perfusion. Cardiogenic Shock Experiments The 3 animals in this group demonstrated rapid deterioration of cardiac performance with a fall in aortic pressure to cardiogenic shock levels within 2 minutes of ligation of both branches of the left coronary artery, and each animal died rapidly in ventricular fibrillation. The left ven-
tricle became grossly cyanotic and hypodynamic, and the average systolic aortic pressure decreased from control values of 100 f 25 (SEM) to 38 f 8 mm Hg (p < 0.01) within 2 minutes, while mean aortic pressure fell from 71 f 15 to 24 f 6 mm Hg (p < 0.01). Cardiac output declined from the controlvalue of 0.6 k 0.2 to 0.2 f 0.1 liter per minute within 2 minutes of left coronary artery ligation (p < 0.01). Ventricular fibrillation appeared within 2 to 5 minutes of coronary ligation in each animal and prevented further studies in this group. Cardiogenic Shock Prevention Experiments In 10 animals coronary capillary retrograde perfusion was instituted simultaneous with ligation of the circumflex and left anterior descending coronary artery. Documentation of adequate retrograde flow was obtained by visualizing oxygenated blood in the distal coronary veins in each animal. In 9 of the 10 animals, cardiac function was sustained and the experiment was continued until elective termination of retrograde perfusion 5 to 30 minutes later; ventricular fibrillation did not occur in these animals. Despite retrograde perfusion 1 animal died of ventricular fibrillation within 5 minutes without data collection. To document the dependence of the left ventricle on coronary retrograde flow, coronary occlusion was continued and the coronary retrograde perfusion was electively terminated after an average of 14 minutes in 6 animals. In the other 3 animals the experiments were continued until electively terminated at 30 minutes to determine if longer support of the ventricle was possible and to obtain ventricular tissue to examine for edema. The following results were observed in the 9 successfully completed studies. Aortic pressure and cardiac output were maintained during retrograde perfusion despite circumflex and left anterior descending coronary artery ligation (Figs 2, 3). When retrograde perfusion was stopped but coronary artery occlusion continued, both aortic pressure and cardiac output fell to cardiogenic shock levels within 2 minutes. After an average of 14 minutes of left coronary ligation and retrograde perfusion, peak aortic pressure averaged 89 f 6 mm Hg compared with a control value of 100 f 8 ( p
![[Myocardial infarction and cardiogenic shock].](/assets/img/hold.png)
