bs_bs_banner
THOUGHTS AND PROGRESS
171
Left Anterior Descending Coronary Artery Blood Flow and Left Ventricular Unloading During Extracorporeal Membrane Oxygenation Support in a Swine Model of Acute Cardiogenic Shock *Christoph Brehm, *Sarah Schubert, †Elizabeth Carney, *Ali Ghodsizad, *Michael Koerner, *Robert McCoach, and *Aly El-Banayosy *Heart and Vascular Institute, Penn State Milton S. Hershey Medical Center; and †Comparative Medicine, Penn State College of Medicine, Hershey, PA, USA Abstract: The impact of extracorporeal membrane oxygenation (ECMO) support on coronary blood flow and left ventricular unloading is still debated. This study aimed to further characterize the influence of ECMO on coronary artery blood flow and its ability to unload the left ventricle in a short-term model of acute cardiogenic shock. Seven anesthetized pigs were intubated and then underwent median sternotomy and cannulation for venoarterial (VA) ECMO. Flow in the left anterior descending (LAD) artery, left atrial pressure (LAP), left ventricular end-diastolic pressure (LVEDP), and mean arterial pressure (MAP) were measured before and after esmolol-induced cardiac dysfunction and after initiating VA-ECMO support. Induction of acute cardiogenic shock was associated with short-term increases in LAP from 8 ± 4 mm Hg to 18 ± 14 mm Hg (P = 0.9) and LVEDP from 5 ± 2 mm Hg to 13 ± 17 mm Hg (P = 0.9), and a decrease in MAP from 63 ± 16 mm Hg to 50 ± 24 mm Hg (P = 0.3). With VAECMO support, blood flow in the LAD increased from 28 ± 25 mL/min during acute unsupported cardiogenic shock to 67 ± 50 mL/min (P = 0.003), and LAP and LVEDP decreased to 8 + 5 mm Hg (P = 0.7) and 5 ± 3 mm Hg (P = 0.5), respectively. In this swine model of acute cardiogenic shock, VA-ECMO improved coronary blood flow and provided some degree of left ventricular unloading for the short duration of the study. Key Words: Venoarterial—Extracorporeal membrane oxygenation—Coronary artery blood flow—Cardiogenic shock—Animal model—Left ventricular unloading.
Extracorporeal membrane oxygenation (ECMO) is a modified form of cardiopulmonary bypass. The technology was first used in the 1970s to treat adult respiratory failure (1). Venoarterial ECMO (VA-ECMO) withdraws blood from the venous circulation, oxygenates the
doi:10.1111/aor.12336 Received January 2014; revised March 2014. Address correspondence and reprint requests to Dr. Christoph Brehm, Heart and Vascular Institute, Penn State Milton S. Hershey Medical Center, 500 University Drive, Hershey, PA 17033, USA. E-mail: [email protected] Artif Organs, Vol. 39, No. 2, 2015
172
THOUGHTS AND PROGRESS
blood, and returns it to the arterial circulation— either centrally through the proximal ascending aorta or peripherally through the femoral (most common), subclavian, or axillary arteries. In this configuration, systemic blood flow is a function of both the ECMO flow and native cardiac function; thus, VA-ECMO provides both respiratory and cardiac support. VA-ECMO is generally indicated in severe cases of respiratory failure and potentially reversible cardiac dysfunction that has been refractory to other therapies. The most common indications for ECMO use in the adult patient population include acute myocardial infarction, acute-on-chronic congestive heart failure, inability to wean the patient from cardiopulmonary bypass following cardiac surgery, primary graft failure after heart transplantation, and severe acute respiratory distress syndrome. The primary aims of mechanical circulatory support in acute cardiogenic shock are to restore the circulation, ensure organ perfusion, and provide left ventricular unloading to allow for myocardial recovery. Attempting to determine if ECMO does actually improve morbidity and mortality in patients with advanced, acute cardiogenic shock in an evidencebased fashion, however, is not easy, given the patients’ tenuous clinical circumstances, including hypotension, maximal inotrope usage, and signs of inadequate end-organ perfusion (e.g., elevated serum lactate). Although no randomized controlled trials examining the use of ECMO for the treatment of cardiogenic shock have been conducted, numerous case series studies have been reported in the literature; most of these indicate a mortality benefit for patients supported with ECMO. Given the high mortality traditionally associated with advanced cardiogenic shock—up to 80% mortality in some patient groups— ECMO represents a promising treatment option for some of these critically ill patients (2). In spite of the ostensible survival benefit of VA-ECMO for patients with acute cardiogenic shock, some studies have incidentally noted a reduction in left ventricular ejection indices, including stroke volume, cardiac output, aortic flow velocity, shortening fractions, and velocity of circumferential fiber length shortening in patients placed on ECMO for respiratory failure (3,4). Another study in an animal model noted an increase in left ventricular end-diastolic pressure (LVEDP) with increasing ECMO flows (5). Yet none of these investigations have directly examined the effect of ECMO on coronary artery blood flow in the setting of acute cardiogenic shock, arguably one of the few situations in which ECMO Artif Organs, Vol. 39, No. 2, 2015
support is warranted. Furthermore, the degree to which ECMO is able to unload the left ventricle is unknown and has implications for myocardial recovery. To further investigate these questions of coronary blood flow and myocardial unloading with ECMO, we developed a short-term model in swine where cardiac dysfunction was acutely induced via administration of high-dose beta-adrenergic antagonist. Given the lack of research into coronary artery flow and myocardial unloading within the context of VA-ECMO support for acute cardiogenic shock, this study primarily aimed to further characterize the influence of VA-ECMO on coronary artery blood flow and its ability to unload the left ventricle in a short-term model of acutely induced cardiac dysfunction. MATERIALS AND METHODS The Institutional Animal Care & Use Committee (IACUC) of the Penn State Milton S. Hershey Medical Center approved the research protocol. All animal care and handling was in accordance with the National Institutes of Health guidelines for use of experimental animals. Seven pigs of approximately the same age and weighing between 65.0 and 73.6 kg were deeply sedated with intramuscular tiletamine hydrochloride and zolazepam hydrochloride, butorphenol, and dexmedetomidine hydrochloride, then endotracheally intubated and mechanically ventilated. Animals were placed in the supine position on the operating room table with an underlying heating pad to maintain a body temperature between 35.8 and 38.4°C. Fentanyl, pancuronium, midazolam, and ketamine were used to maintain general anesthesia during surgery and throughout the study. The surgery was performed in four steps. Step 1: the right internal jugular vein was surgically exposed and then cannulated with a single lumen intravenous catheter, and its tip was placed in the superior vena cava as a means of measuring the central venous pressure (CVP). Step 2: the chest of the pig was opened with a median sternotomy, and the pericardium was opened and the heart exposed. The apex of the heart was mobilized, and a triple lumen catheter was inserted into the left ventricular apex to measure LVEDP. The beating heart was stabilized using a pack of surgical towels. The anterior wall was exposed and the left anterior descending (LAD) was prepared with an Overholt clamp and encircled with two 0-0 silk suture loops. A Transonic Flow probe (Transonic System, Inc., Ithaca, NY, USA) was placed on the LAD to
THOUGHTS AND PROGRESS
173
TABLE 1. Experimental protocol conditions Condition 1 Condition 2 Condition 3 Condition 4 Condition 5 Condition 6
Baseline, normal heart in sinus rhythm Induced acute cardiogenic shock without ECMO support in sinus rhythm Induced cardiogenic shock with 50% ECMO support in sinus rhythm Induced cardiogenic shock with 100% ECMO support Induced cardiogenic shock with 100% ECMO support in ventricular fibrillation Induced cardiogenic shock with 100% ECMO support in restored sinus rhythm
measure the blood flow during the study. The left atrial appendage was exposed, and a triple lumen catheter was placed to measure the left atrial pressure (LAP) and to apply esmolol throughout the study. Step 3: the left groin was opened with an incision parallel to and distal to the inguinal ligament. The femoral artery was prepared and cannulated with a Cordis Avanti 4Fr arterial cannula (Cordis Corp., Miami Lakes, FL, USA) to measure the arterial blood pressure during the study. Subsequently, unfractionated heparin was given at 300 U/kg as a bolus. Step 4: the right groin was opened as described above. The femoral vein and femoral artery were prepared, exposed, and encircled with vessel loops. The artery was then cannulated with a Medtronic Bio-Medicus 17Fr arterial cannula (Medtronic, Inc., Minneapolis, MN, USA). The vein was cannulated in the same fashion, using a Medtronic Bio-Medicus 19Fr cannula with the tip positioned in the inferior vena cava. A simplified ECMO circuit was used, composed of a magnetically levitated Levitronix CentriMag blood pump with a Levitronix CentriMag console (both from Levitronix LLC, Waltham, MA, USA) and a low resistance Maquet Quadrox D hollow fiber oxygenator (MAQUET, Inc., Bridgewater, NJ, USA). Creating acute cardiogenic shock Acute cardiogenic shock is essentially a clinical diagnosis based on severe depression of cardiac output with sustained hypotension and elevated filling pressures within the heart (6). According to the definition of the Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial, formally diagnosing cardiogenic shock requires that hypoperfusion be maintained for a certain length of time and that evidence of end-organ dysfunction exists (7). In this study, resources limited the amount of time we were able to maintain hypoperfusion and monitor for end-organ dysfunction in each pig. To achieve conditions consistent with acute cardiogenic shock, a high-dose beta-adrenergic
antagonist (esmolol bolus at 2 mg/kg) was directly infused into the left atrium. Clinically evident acute cardiogenic shock was reached in all of the pigs, with sustained hypotension and elevated filling pressures. Furthermore, contractility of the myocardium was observably decreased, as the median sternotomy had already been completed and the heart could be observed directly. In the first two pigs, the onset of cardiogenic shock was further confirmed with epicardial echo, as the pigs’ initially normal ejection fraction decreased to less than 30%. Subsequently, an esmolol drip was started at 500 mg/kg/min to maintain cardiac dysfunction throughout the trial. Experimental protocol In our experimental protocol, blood flow in the LAD and various hemodynamic parameters— including heart rate, systolic and diastolic blood pressures, LAP, CVP, and LVEDP—were measured under six different conditions: Condition 1 (baseline): before induction of acute cardiogenic shock; Condition 2: acute cardiogenic shock without support; Condition 3: acute cardiogenic shock with 50% ECMO support; Condition 4: acute cardiogenic shock with 100% ECMO support; Condition 5: electrically induced ventricular fibrillation with 100% ECMO support; and Condition 6: restored normal sinus rhythm with 100% ECMO support (Table 1). Data were collected after 3–5 min from achieving the condition to allow equilibration of measurement. Vasopressin was used to maintain mean arterial pressure (MAP) between 70 and 90 mm Hg in the conditions of cardiogenic shock without ventricular fibrillation (conditions 2, 3, 4, and 6). The esmolol drip was continued as described above. RESULTS A total of seven pigs were studied, and the weight of the pigs was similar (Table 2). The mean blood flow in the porcine LAD arteries during condition 1 was 50 ± 33 mL/min. Induction of acute cardiogenic shock was associated with a decrease in LAD flow from 50 ± 33 to 28 ± 25 mL/min (P = 0.06) and a decrease in MAP Artif Organs, Vol. 39, No. 2, 2015
174
THOUGHTS AND PROGRESS TABLE 2. Baseline characteristics of pigs
No. 1 2 3 4 5 6 7
Weight (kg)
HR (bpm)
BP (mm Hg)
MAP (mm Hg)
CVP (mm Hg)
LAP (mm Hg)
LAD blood flow (mL/min)
65.0 68.0 65.4 73.6 66.4 66.4 65.0
43 67 64 96 79 66 63
58/39 80/51 85/50 124/84 107/70 82/53 86/48
44 60 62 96 55 62 60
— 6 3 6 7 7 8
15 9 4 7 7 6 9
20 35 9 54 45 87 100
BP, blood pressure; HR, heart rate.
acute unsupported cardiac dysfunction to 67 ± 50 mL/min (P = 0.003, Fig 2). Following induction of ventricular fibrillation with concomitant ECMO support, LAD flow (69 ± 41 mL/ min) remained significantly greater than the flow in the LAD during acute, unsupported cardiac dysfunction (28 ± 25 mL/min, P = 0.03). Next, the heart was defibrillated while still receiving 100% ECMO support, and the establishment of normal sinus rhythm constituted condition 6. The flow in the LAD (91 ± 76 mL/min, P = 0.4) was greater under these conditions than under conditions of acute, unsupported cardiac dysfunction; however, the results did not reach statistical significance. DISCUSSION In patients supported with VA-ECMO for acute cardiogenic shock, the primary goals are to restore circulation, ensure organ perfusion, and unload the left ventricle to allow for myocardial and end-organ recovery. Although it is known that ECMO is restoring circulation and improving oxygen delivery to tissues, it is unknown what effects ECMO has on coronary blood flow and left ventricular unloading. Clinically, some patients in cardiogenic shock have
35
140
30
120
Blood Flow (mL/min)
Pressure (mm Hg)
from 63 ± 16 to 50 ± 24 mm Hg (P = 0.3). The heart rate of each pig slightly decreased following the induction of cardiogenic shock, explicably due to the infusion of beta-adrenergic antagonist. Furthermore, with the induction of acute cardiogenic shock, CVP increased from 6 ± 2 to 8 ± 3 mm Hg (P = 0.3), LAP increased from 8 ± 4 to 18 ± 14 mm Hg (P = 0.9), and LVEDP increased from 5 ± 2 to 13 ± 17 mm Hg (P = 0.9). Subsequently, VA-ECMO support at approximately 50% of the pigs’ estimated cardiac output was initiated. LAP decreased to 10 ± 5 mm Hg (P = 0.5), and LVEDP decreased to 10 ± 5 mm Hg (P = 0.3), while CVP remained approximately constant at 7 ± 4 mm Hg (P = 0.5). The blood flow in the LAD increased to 40 ± 41 mL/min (P = 0.02). Following completion of condition 3 measurements, ECMO pump speed was increased to 100%. With the establishment of condition 4, LAP decreased to 8 ± 5 mm Hg (P = 0.7), LVEDP decreased to 5 ± 3 mm Hg (P = 0.5, Fig. 1), and CVP decreased to 5 ± 3 mm Hg (P = 0.8). The heart rate (72 ± 22 mm Hg) was essentially unchanged from the conditions of unsupported cardiac dysfunction. Under these conditions, there was a significant increase in blood flow in the LAD from 28 ± 25 mL/min during
25 20 15 10 5
P = 0.003
P = 0.02
100 80 60 40 20
0 Baseline
ACS
ACS with 50% ECMO
ACS with 100% ECMO
0 Baseline
Left Atrial Pressure
Left Ventricular End-Diastolic Pressure
FIG. 1. LAP and left ventricular pressure (mm Hg): at baseline conditions, with induction of acute cardiogenic shock (ACS), and with 50 and 100% ECMO support. Artif Organs, Vol. 39, No. 2, 2015
ACS
ACS with 50% ECMO
ACS with 100% ECMO
FIG. 2. LAD coronary artery blood flow (mL/min): at baseline conditions, with induction of acute cardiogenic shock (ACS), and with 50 and 100% ECMO support.
THOUGHTS AND PROGRESS demonstrated a paradoxical decrease in left ventricular function with the institution of ECMO support. Such a decline in left ventricular function could be related to a variety of factors, including a change in preload, decrease of inotropes after successful ECMO initiation, or even reduced coronary artery blood flow. Another prominent concern associated with VA-ECMO support is its inability to completely unload a failing left ventricle. To examine changes in preload, a study in swine with normal cardiac function examined loadsensitive and load-insensitive parameters with VA-ECMO support and found that ECMO can change some of the load-dependent parameters of contractility. They found, however, that the intrinsic heart function was not significantly affected, as determined from load-insensitive indices of left ventricular function (8). Another possible explanation for depressed left ventricular function with VA-ECMO is a decrease in coronary artery blood flow. Because normal myocardial oxygen extraction is already greater than 75%, an increase in coronary artery blood flow is the primary mechanism by which increased myocardial oxygen demands are met. If ECMO is actually decreasing coronary artery blood flow, then ECMO support could ostensibly be counterproductive for possible myocardial recovery. Yet the effects of ECMO on coronary artery blood flow in previous studies in healthy animal models have been equivocal. A study in healthy lambs by Kinsella et al. demonstrated no significant change in coronary artery flow with ECMO support as compared with baseline. The authors did note that even with proximal placement of the arterial cannula, the coronaries are primarily perfused by blood being pumped from the left ventricle—that is, not the welloxygenated blood that is traveling through the ECMO circuit (9). Another study by Kato et al. further examined this question of coronary blood flow in a healthy dog model, and their data demonstrated that coronary artery flow decreased as ECMO flow increased (10). Unlike these previous studies which examined ECMO within the context of healthy animal models, our study aimed to characterize coronary artery flow and left ventricular unloading within the clinical context of VA-ECMO support of acute cardiogenic shock. Arguably, our results are more likely to reflect actual clinical circumstances than those results obtained in healthy animal models, as ECMO would not be instituted in healthy patients. Cardiac dysfunction was acutely induced in these swine hearts with high dose beta-adrenergic antago-
175
nist; this method of inducing cardiac dysfunction was chosen over other methods because nothing was infused directly into the coronary arteries. Introducing substances into the coronary arteries—which is a typical method of creating acute cardiogenic shock— would have interfered with the measurement of coronary blood flow. Our study demonstrated that 100% ECMO support significantly increased LAD blood flow compared with that in the unsupported heart in this shortterm model. Additionally, during acute cardiogenic shock with induced ventricular fibrillation, a significant increase in LAD blood flow occurred with 100% ECMO support. Increases in LAD blood flow were also observed with 50% ECMO support after induction of acute cardiogenic shock, but these increases were not statistically significant. These results refute an earlier study by Shen et al. which found that coronary artery blood flow was inversely related to ECMO flow (8). In our study, not only was there flow into the LAD following the initiation of ECMO, but that flow was significantly greater than that occurring in the unsupported failing heart. Because ECMO is providing blood flow to the coronary arteries and that rate of flow increases with increasing ECMO flow, this study demonstrates that ECMO is indeed enabling myocardial recovery in this swine model. By providing greater myocardial blood flow than what would be occurring in a typical failing heart, ECMO is not only sustaining the heart muscle but it is also meeting the increased oxygen demands of the myocardium as it attempts to recover from insult. Also of note in this study, flow within the LAD following acute induction of cardiac dysfunction in the swine animal model was decreased as compared with that of normal conditions prior to induction of acute cardiogenic shock, but that decrease did not reach statistical significance. The other primary finding of our study was that both 50 and 100% ECMO support was associated with a decrease in LAP and LVEDP—albeit nonstatistically significant decreases. This finding indicates that ECMO is providing some degree of left ventricular unloading—at least in the short term— under conditions of acute cardiogenic shock. Although this study was able to demonstrate a relationship between increasing ECMO flow and increasing coronary artery blood flow and ECMO’s ability to unload the left ventricle, it had limitations, including the relatively small sample size and the use of an animal model. The model of acute cardiogenic shock in this study was acutely induced; thus, many of the structural changes that occur in chronically failing Artif Organs, Vol. 39, No. 2, 2015
bs_bs_banner
176
THOUGHTS AND PROGRESS
hearts that may be supported with ECMO were not seen in these recently healthy hearts. Additionally, due to limitations in time and resources, acute cardiogenic shock was not strictly achieved according to a formal definition of cardiogenic shock. Furthermore, each pig was sequentially subjected to each experimental condition, making it possible that the early conditions could affect subsequent results. Finally, this study did not examine the oxygen content of the coronary blood. Future study in a greater number of animals with determination of coronary blood oxygen content is necessary to better characterize the influence of ECMO on coronary artery blood flow and left ventricular unloading in acute cardiogenic shock. CONCLUSION In our short-term swine model of acute cardiogenic shock, ECMO support increased blood flow in the LAD artery, and that increase was statistically significant with 100% ECMO support. Additionally, this study demonstrated that ECMO is decreasing left-sided filling pressures, indicating that ECMO has the potential to unload the left ventricle in instances of acute cardiogenic shock. REFERENCES 1. Hill JD, O’Brien TG, Murray JJ, et al. Prolonged extracorporeal oxygenation for acute post-traumatic respiratory failure (shock-lung syndrome): use of the Bramson membrane lung. N Engl J Med 1972;286:629–34. 2. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. N Engl J Med 1999;341:625–34. 3. Martin GR, Short BL. Doppler echocardiographic evaluation of cardiac performance in infants on prolonged extracorporeal membrane oxygenation. Am J Cardiol 1988;62:929–34. 4. Kimball TR, Daniels SR, Weiss RG, et al. Changes in cardiac function during extracorporeal membrane oxygenation for persistent pulmonary hypertension in the newborn infant. J Pediatr 1991;118:431–6. 5. Seo T, Ito T, Iio K, Kato J, Takagi H. Extracorporeal study on the hemodynamic effects of veno-arterial extracorporeal membrane oxygenation with an automatically driven blood pump on puppies. Artif Organs 1991;15:402–7. 6. Gheorghiade M, Filippatos GS, Felker GM. Diagnosis and management of acute heart failure syndromes. In: Bonow RO, ed. Braunwald’s Heart Disease. Philadelphia: Elsevier Saunders, 2011;517–42. 7. Hochman JS, Sleeper LA, Webb JG, et al. Should we emergently revascularize occluded coronaries for cardiogenic shock: an international randomized trial of emergency PCTA/ CABG. NEJM 1999;341:625–34. 8. Shen I, Levy FH, Vocelka CR, et al. Effect of extracorporeal membrane oxygenation on left ventricular function of swine. Ann Thorac Surg 2001;71:862–7. 9. Kinsella JP, Gerstmann DR, Rosenberg AA. The effect of extracorporeal membrane oxygenation on coronary perfusion and regional blood flow distribution. Pediatr Res 1992; 31:80–4. Artif Organs, Vol. 39, No. 2, 2015
10. Kato J, Seo T, Ando H, Takagi H, Ito T. Coronary arterial perfusion during venoarterial extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 1996;111:630–6.
THOUGHTS AND PROGRESS
171
Left Anterior Descending Coronary Artery Blood Flow and Left Ventricular Unloading During Extracorporeal Membrane Oxygenation Support in a Swine Model of Acute Cardiogenic Shock *Christoph Brehm, *Sarah Schubert, †Elizabeth Carney, *Ali Ghodsizad, *Michael Koerner, *Robert McCoach, and *Aly El-Banayosy *Heart and Vascular Institute, Penn State Milton S. Hershey Medical Center; and †Comparative Medicine, Penn State College of Medicine, Hershey, PA, USA Abstract: The impact of extracorporeal membrane oxygenation (ECMO) support on coronary blood flow and left ventricular unloading is still debated. This study aimed to further characterize the influence of ECMO on coronary artery blood flow and its ability to unload the left ventricle in a short-term model of acute cardiogenic shock. Seven anesthetized pigs were intubated and then underwent median sternotomy and cannulation for venoarterial (VA) ECMO. Flow in the left anterior descending (LAD) artery, left atrial pressure (LAP), left ventricular end-diastolic pressure (LVEDP), and mean arterial pressure (MAP) were measured before and after esmolol-induced cardiac dysfunction and after initiating VA-ECMO support. Induction of acute cardiogenic shock was associated with short-term increases in LAP from 8 ± 4 mm Hg to 18 ± 14 mm Hg (P = 0.9) and LVEDP from 5 ± 2 mm Hg to 13 ± 17 mm Hg (P = 0.9), and a decrease in MAP from 63 ± 16 mm Hg to 50 ± 24 mm Hg (P = 0.3). With VAECMO support, blood flow in the LAD increased from 28 ± 25 mL/min during acute unsupported cardiogenic shock to 67 ± 50 mL/min (P = 0.003), and LAP and LVEDP decreased to 8 + 5 mm Hg (P = 0.7) and 5 ± 3 mm Hg (P = 0.5), respectively. In this swine model of acute cardiogenic shock, VA-ECMO improved coronary blood flow and provided some degree of left ventricular unloading for the short duration of the study. Key Words: Venoarterial—Extracorporeal membrane oxygenation—Coronary artery blood flow—Cardiogenic shock—Animal model—Left ventricular unloading.
Extracorporeal membrane oxygenation (ECMO) is a modified form of cardiopulmonary bypass. The technology was first used in the 1970s to treat adult respiratory failure (1). Venoarterial ECMO (VA-ECMO) withdraws blood from the venous circulation, oxygenates the
doi:10.1111/aor.12336 Received January 2014; revised March 2014. Address correspondence and reprint requests to Dr. Christoph Brehm, Heart and Vascular Institute, Penn State Milton S. Hershey Medical Center, 500 University Drive, Hershey, PA 17033, USA. E-mail: [email protected] Artif Organs, Vol. 39, No. 2, 2015
172
THOUGHTS AND PROGRESS
blood, and returns it to the arterial circulation— either centrally through the proximal ascending aorta or peripherally through the femoral (most common), subclavian, or axillary arteries. In this configuration, systemic blood flow is a function of both the ECMO flow and native cardiac function; thus, VA-ECMO provides both respiratory and cardiac support. VA-ECMO is generally indicated in severe cases of respiratory failure and potentially reversible cardiac dysfunction that has been refractory to other therapies. The most common indications for ECMO use in the adult patient population include acute myocardial infarction, acute-on-chronic congestive heart failure, inability to wean the patient from cardiopulmonary bypass following cardiac surgery, primary graft failure after heart transplantation, and severe acute respiratory distress syndrome. The primary aims of mechanical circulatory support in acute cardiogenic shock are to restore the circulation, ensure organ perfusion, and provide left ventricular unloading to allow for myocardial recovery. Attempting to determine if ECMO does actually improve morbidity and mortality in patients with advanced, acute cardiogenic shock in an evidencebased fashion, however, is not easy, given the patients’ tenuous clinical circumstances, including hypotension, maximal inotrope usage, and signs of inadequate end-organ perfusion (e.g., elevated serum lactate). Although no randomized controlled trials examining the use of ECMO for the treatment of cardiogenic shock have been conducted, numerous case series studies have been reported in the literature; most of these indicate a mortality benefit for patients supported with ECMO. Given the high mortality traditionally associated with advanced cardiogenic shock—up to 80% mortality in some patient groups— ECMO represents a promising treatment option for some of these critically ill patients (2). In spite of the ostensible survival benefit of VA-ECMO for patients with acute cardiogenic shock, some studies have incidentally noted a reduction in left ventricular ejection indices, including stroke volume, cardiac output, aortic flow velocity, shortening fractions, and velocity of circumferential fiber length shortening in patients placed on ECMO for respiratory failure (3,4). Another study in an animal model noted an increase in left ventricular end-diastolic pressure (LVEDP) with increasing ECMO flows (5). Yet none of these investigations have directly examined the effect of ECMO on coronary artery blood flow in the setting of acute cardiogenic shock, arguably one of the few situations in which ECMO Artif Organs, Vol. 39, No. 2, 2015
support is warranted. Furthermore, the degree to which ECMO is able to unload the left ventricle is unknown and has implications for myocardial recovery. To further investigate these questions of coronary blood flow and myocardial unloading with ECMO, we developed a short-term model in swine where cardiac dysfunction was acutely induced via administration of high-dose beta-adrenergic antagonist. Given the lack of research into coronary artery flow and myocardial unloading within the context of VA-ECMO support for acute cardiogenic shock, this study primarily aimed to further characterize the influence of VA-ECMO on coronary artery blood flow and its ability to unload the left ventricle in a short-term model of acutely induced cardiac dysfunction. MATERIALS AND METHODS The Institutional Animal Care & Use Committee (IACUC) of the Penn State Milton S. Hershey Medical Center approved the research protocol. All animal care and handling was in accordance with the National Institutes of Health guidelines for use of experimental animals. Seven pigs of approximately the same age and weighing between 65.0 and 73.6 kg were deeply sedated with intramuscular tiletamine hydrochloride and zolazepam hydrochloride, butorphenol, and dexmedetomidine hydrochloride, then endotracheally intubated and mechanically ventilated. Animals were placed in the supine position on the operating room table with an underlying heating pad to maintain a body temperature between 35.8 and 38.4°C. Fentanyl, pancuronium, midazolam, and ketamine were used to maintain general anesthesia during surgery and throughout the study. The surgery was performed in four steps. Step 1: the right internal jugular vein was surgically exposed and then cannulated with a single lumen intravenous catheter, and its tip was placed in the superior vena cava as a means of measuring the central venous pressure (CVP). Step 2: the chest of the pig was opened with a median sternotomy, and the pericardium was opened and the heart exposed. The apex of the heart was mobilized, and a triple lumen catheter was inserted into the left ventricular apex to measure LVEDP. The beating heart was stabilized using a pack of surgical towels. The anterior wall was exposed and the left anterior descending (LAD) was prepared with an Overholt clamp and encircled with two 0-0 silk suture loops. A Transonic Flow probe (Transonic System, Inc., Ithaca, NY, USA) was placed on the LAD to
THOUGHTS AND PROGRESS
173
TABLE 1. Experimental protocol conditions Condition 1 Condition 2 Condition 3 Condition 4 Condition 5 Condition 6
Baseline, normal heart in sinus rhythm Induced acute cardiogenic shock without ECMO support in sinus rhythm Induced cardiogenic shock with 50% ECMO support in sinus rhythm Induced cardiogenic shock with 100% ECMO support Induced cardiogenic shock with 100% ECMO support in ventricular fibrillation Induced cardiogenic shock with 100% ECMO support in restored sinus rhythm
measure the blood flow during the study. The left atrial appendage was exposed, and a triple lumen catheter was placed to measure the left atrial pressure (LAP) and to apply esmolol throughout the study. Step 3: the left groin was opened with an incision parallel to and distal to the inguinal ligament. The femoral artery was prepared and cannulated with a Cordis Avanti 4Fr arterial cannula (Cordis Corp., Miami Lakes, FL, USA) to measure the arterial blood pressure during the study. Subsequently, unfractionated heparin was given at 300 U/kg as a bolus. Step 4: the right groin was opened as described above. The femoral vein and femoral artery were prepared, exposed, and encircled with vessel loops. The artery was then cannulated with a Medtronic Bio-Medicus 17Fr arterial cannula (Medtronic, Inc., Minneapolis, MN, USA). The vein was cannulated in the same fashion, using a Medtronic Bio-Medicus 19Fr cannula with the tip positioned in the inferior vena cava. A simplified ECMO circuit was used, composed of a magnetically levitated Levitronix CentriMag blood pump with a Levitronix CentriMag console (both from Levitronix LLC, Waltham, MA, USA) and a low resistance Maquet Quadrox D hollow fiber oxygenator (MAQUET, Inc., Bridgewater, NJ, USA). Creating acute cardiogenic shock Acute cardiogenic shock is essentially a clinical diagnosis based on severe depression of cardiac output with sustained hypotension and elevated filling pressures within the heart (6). According to the definition of the Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock (SHOCK) trial, formally diagnosing cardiogenic shock requires that hypoperfusion be maintained for a certain length of time and that evidence of end-organ dysfunction exists (7). In this study, resources limited the amount of time we were able to maintain hypoperfusion and monitor for end-organ dysfunction in each pig. To achieve conditions consistent with acute cardiogenic shock, a high-dose beta-adrenergic
antagonist (esmolol bolus at 2 mg/kg) was directly infused into the left atrium. Clinically evident acute cardiogenic shock was reached in all of the pigs, with sustained hypotension and elevated filling pressures. Furthermore, contractility of the myocardium was observably decreased, as the median sternotomy had already been completed and the heart could be observed directly. In the first two pigs, the onset of cardiogenic shock was further confirmed with epicardial echo, as the pigs’ initially normal ejection fraction decreased to less than 30%. Subsequently, an esmolol drip was started at 500 mg/kg/min to maintain cardiac dysfunction throughout the trial. Experimental protocol In our experimental protocol, blood flow in the LAD and various hemodynamic parameters— including heart rate, systolic and diastolic blood pressures, LAP, CVP, and LVEDP—were measured under six different conditions: Condition 1 (baseline): before induction of acute cardiogenic shock; Condition 2: acute cardiogenic shock without support; Condition 3: acute cardiogenic shock with 50% ECMO support; Condition 4: acute cardiogenic shock with 100% ECMO support; Condition 5: electrically induced ventricular fibrillation with 100% ECMO support; and Condition 6: restored normal sinus rhythm with 100% ECMO support (Table 1). Data were collected after 3–5 min from achieving the condition to allow equilibration of measurement. Vasopressin was used to maintain mean arterial pressure (MAP) between 70 and 90 mm Hg in the conditions of cardiogenic shock without ventricular fibrillation (conditions 2, 3, 4, and 6). The esmolol drip was continued as described above. RESULTS A total of seven pigs were studied, and the weight of the pigs was similar (Table 2). The mean blood flow in the porcine LAD arteries during condition 1 was 50 ± 33 mL/min. Induction of acute cardiogenic shock was associated with a decrease in LAD flow from 50 ± 33 to 28 ± 25 mL/min (P = 0.06) and a decrease in MAP Artif Organs, Vol. 39, No. 2, 2015
174
THOUGHTS AND PROGRESS TABLE 2. Baseline characteristics of pigs
No. 1 2 3 4 5 6 7
Weight (kg)
HR (bpm)
BP (mm Hg)
MAP (mm Hg)
CVP (mm Hg)
LAP (mm Hg)
LAD blood flow (mL/min)
65.0 68.0 65.4 73.6 66.4 66.4 65.0
43 67 64 96 79 66 63
58/39 80/51 85/50 124/84 107/70 82/53 86/48
44 60 62 96 55 62 60
— 6 3 6 7 7 8
15 9 4 7 7 6 9
20 35 9 54 45 87 100
BP, blood pressure; HR, heart rate.
acute unsupported cardiac dysfunction to 67 ± 50 mL/min (P = 0.003, Fig 2). Following induction of ventricular fibrillation with concomitant ECMO support, LAD flow (69 ± 41 mL/ min) remained significantly greater than the flow in the LAD during acute, unsupported cardiac dysfunction (28 ± 25 mL/min, P = 0.03). Next, the heart was defibrillated while still receiving 100% ECMO support, and the establishment of normal sinus rhythm constituted condition 6. The flow in the LAD (91 ± 76 mL/min, P = 0.4) was greater under these conditions than under conditions of acute, unsupported cardiac dysfunction; however, the results did not reach statistical significance. DISCUSSION In patients supported with VA-ECMO for acute cardiogenic shock, the primary goals are to restore circulation, ensure organ perfusion, and unload the left ventricle to allow for myocardial and end-organ recovery. Although it is known that ECMO is restoring circulation and improving oxygen delivery to tissues, it is unknown what effects ECMO has on coronary blood flow and left ventricular unloading. Clinically, some patients in cardiogenic shock have
35
140
30
120
Blood Flow (mL/min)
Pressure (mm Hg)
from 63 ± 16 to 50 ± 24 mm Hg (P = 0.3). The heart rate of each pig slightly decreased following the induction of cardiogenic shock, explicably due to the infusion of beta-adrenergic antagonist. Furthermore, with the induction of acute cardiogenic shock, CVP increased from 6 ± 2 to 8 ± 3 mm Hg (P = 0.3), LAP increased from 8 ± 4 to 18 ± 14 mm Hg (P = 0.9), and LVEDP increased from 5 ± 2 to 13 ± 17 mm Hg (P = 0.9). Subsequently, VA-ECMO support at approximately 50% of the pigs’ estimated cardiac output was initiated. LAP decreased to 10 ± 5 mm Hg (P = 0.5), and LVEDP decreased to 10 ± 5 mm Hg (P = 0.3), while CVP remained approximately constant at 7 ± 4 mm Hg (P = 0.5). The blood flow in the LAD increased to 40 ± 41 mL/min (P = 0.02). Following completion of condition 3 measurements, ECMO pump speed was increased to 100%. With the establishment of condition 4, LAP decreased to 8 ± 5 mm Hg (P = 0.7), LVEDP decreased to 5 ± 3 mm Hg (P = 0.5, Fig. 1), and CVP decreased to 5 ± 3 mm Hg (P = 0.8). The heart rate (72 ± 22 mm Hg) was essentially unchanged from the conditions of unsupported cardiac dysfunction. Under these conditions, there was a significant increase in blood flow in the LAD from 28 ± 25 mL/min during
25 20 15 10 5
P = 0.003
P = 0.02
100 80 60 40 20
0 Baseline
ACS
ACS with 50% ECMO
ACS with 100% ECMO
0 Baseline
Left Atrial Pressure
Left Ventricular End-Diastolic Pressure
FIG. 1. LAP and left ventricular pressure (mm Hg): at baseline conditions, with induction of acute cardiogenic shock (ACS), and with 50 and 100% ECMO support. Artif Organs, Vol. 39, No. 2, 2015
ACS
ACS with 50% ECMO
ACS with 100% ECMO
FIG. 2. LAD coronary artery blood flow (mL/min): at baseline conditions, with induction of acute cardiogenic shock (ACS), and with 50 and 100% ECMO support.
THOUGHTS AND PROGRESS demonstrated a paradoxical decrease in left ventricular function with the institution of ECMO support. Such a decline in left ventricular function could be related to a variety of factors, including a change in preload, decrease of inotropes after successful ECMO initiation, or even reduced coronary artery blood flow. Another prominent concern associated with VA-ECMO support is its inability to completely unload a failing left ventricle. To examine changes in preload, a study in swine with normal cardiac function examined loadsensitive and load-insensitive parameters with VA-ECMO support and found that ECMO can change some of the load-dependent parameters of contractility. They found, however, that the intrinsic heart function was not significantly affected, as determined from load-insensitive indices of left ventricular function (8). Another possible explanation for depressed left ventricular function with VA-ECMO is a decrease in coronary artery blood flow. Because normal myocardial oxygen extraction is already greater than 75%, an increase in coronary artery blood flow is the primary mechanism by which increased myocardial oxygen demands are met. If ECMO is actually decreasing coronary artery blood flow, then ECMO support could ostensibly be counterproductive for possible myocardial recovery. Yet the effects of ECMO on coronary artery blood flow in previous studies in healthy animal models have been equivocal. A study in healthy lambs by Kinsella et al. demonstrated no significant change in coronary artery flow with ECMO support as compared with baseline. The authors did note that even with proximal placement of the arterial cannula, the coronaries are primarily perfused by blood being pumped from the left ventricle—that is, not the welloxygenated blood that is traveling through the ECMO circuit (9). Another study by Kato et al. further examined this question of coronary blood flow in a healthy dog model, and their data demonstrated that coronary artery flow decreased as ECMO flow increased (10). Unlike these previous studies which examined ECMO within the context of healthy animal models, our study aimed to characterize coronary artery flow and left ventricular unloading within the clinical context of VA-ECMO support of acute cardiogenic shock. Arguably, our results are more likely to reflect actual clinical circumstances than those results obtained in healthy animal models, as ECMO would not be instituted in healthy patients. Cardiac dysfunction was acutely induced in these swine hearts with high dose beta-adrenergic antago-
175
nist; this method of inducing cardiac dysfunction was chosen over other methods because nothing was infused directly into the coronary arteries. Introducing substances into the coronary arteries—which is a typical method of creating acute cardiogenic shock— would have interfered with the measurement of coronary blood flow. Our study demonstrated that 100% ECMO support significantly increased LAD blood flow compared with that in the unsupported heart in this shortterm model. Additionally, during acute cardiogenic shock with induced ventricular fibrillation, a significant increase in LAD blood flow occurred with 100% ECMO support. Increases in LAD blood flow were also observed with 50% ECMO support after induction of acute cardiogenic shock, but these increases were not statistically significant. These results refute an earlier study by Shen et al. which found that coronary artery blood flow was inversely related to ECMO flow (8). In our study, not only was there flow into the LAD following the initiation of ECMO, but that flow was significantly greater than that occurring in the unsupported failing heart. Because ECMO is providing blood flow to the coronary arteries and that rate of flow increases with increasing ECMO flow, this study demonstrates that ECMO is indeed enabling myocardial recovery in this swine model. By providing greater myocardial blood flow than what would be occurring in a typical failing heart, ECMO is not only sustaining the heart muscle but it is also meeting the increased oxygen demands of the myocardium as it attempts to recover from insult. Also of note in this study, flow within the LAD following acute induction of cardiac dysfunction in the swine animal model was decreased as compared with that of normal conditions prior to induction of acute cardiogenic shock, but that decrease did not reach statistical significance. The other primary finding of our study was that both 50 and 100% ECMO support was associated with a decrease in LAP and LVEDP—albeit nonstatistically significant decreases. This finding indicates that ECMO is providing some degree of left ventricular unloading—at least in the short term— under conditions of acute cardiogenic shock. Although this study was able to demonstrate a relationship between increasing ECMO flow and increasing coronary artery blood flow and ECMO’s ability to unload the left ventricle, it had limitations, including the relatively small sample size and the use of an animal model. The model of acute cardiogenic shock in this study was acutely induced; thus, many of the structural changes that occur in chronically failing Artif Organs, Vol. 39, No. 2, 2015
bs_bs_banner
176
THOUGHTS AND PROGRESS
hearts that may be supported with ECMO were not seen in these recently healthy hearts. Additionally, due to limitations in time and resources, acute cardiogenic shock was not strictly achieved according to a formal definition of cardiogenic shock. Furthermore, each pig was sequentially subjected to each experimental condition, making it possible that the early conditions could affect subsequent results. Finally, this study did not examine the oxygen content of the coronary blood. Future study in a greater number of animals with determination of coronary blood oxygen content is necessary to better characterize the influence of ECMO on coronary artery blood flow and left ventricular unloading in acute cardiogenic shock. CONCLUSION In our short-term swine model of acute cardiogenic shock, ECMO support increased blood flow in the LAD artery, and that increase was statistically significant with 100% ECMO support. Additionally, this study demonstrated that ECMO is decreasing left-sided filling pressures, indicating that ECMO has the potential to unload the left ventricle in instances of acute cardiogenic shock. REFERENCES 1. Hill JD, O’Brien TG, Murray JJ, et al. Prolonged extracorporeal oxygenation for acute post-traumatic respiratory failure (shock-lung syndrome): use of the Bramson membrane lung. N Engl J Med 1972;286:629–34. 2. Hochman JS, Sleeper LA, Webb JG, et al. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. N Engl J Med 1999;341:625–34. 3. Martin GR, Short BL. Doppler echocardiographic evaluation of cardiac performance in infants on prolonged extracorporeal membrane oxygenation. Am J Cardiol 1988;62:929–34. 4. Kimball TR, Daniels SR, Weiss RG, et al. Changes in cardiac function during extracorporeal membrane oxygenation for persistent pulmonary hypertension in the newborn infant. J Pediatr 1991;118:431–6. 5. Seo T, Ito T, Iio K, Kato J, Takagi H. Extracorporeal study on the hemodynamic effects of veno-arterial extracorporeal membrane oxygenation with an automatically driven blood pump on puppies. Artif Organs 1991;15:402–7. 6. Gheorghiade M, Filippatos GS, Felker GM. Diagnosis and management of acute heart failure syndromes. In: Bonow RO, ed. Braunwald’s Heart Disease. Philadelphia: Elsevier Saunders, 2011;517–42. 7. Hochman JS, Sleeper LA, Webb JG, et al. Should we emergently revascularize occluded coronaries for cardiogenic shock: an international randomized trial of emergency PCTA/ CABG. NEJM 1999;341:625–34. 8. Shen I, Levy FH, Vocelka CR, et al. Effect of extracorporeal membrane oxygenation on left ventricular function of swine. Ann Thorac Surg 2001;71:862–7. 9. Kinsella JP, Gerstmann DR, Rosenberg AA. The effect of extracorporeal membrane oxygenation on coronary perfusion and regional blood flow distribution. Pediatr Res 1992; 31:80–4. Artif Organs, Vol. 39, No. 2, 2015
10. Kato J, Seo T, Ando H, Takagi H, Ito T. Coronary arterial perfusion during venoarterial extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 1996;111:630–6.
