- Email: [email protected]
Clinical experience with portable cardiopulmonary bypass in cardiac arrest patients
Clinical Experience With Portable Cardiopulmonary Bypass in Cardiac Arrest Patients Renee Hartz, MD, Joseph LoCicero 111, MD, John H. Sanders, Jr, MD, James W. Frederiksen, MD, Axel W. Joob, MD, and Lawrence L. Michaelis, MD Division of Cardiothoracic Surgery, Department of Surgery, Northwestern University Medical School, Chicago, Illinois
To evaluate the use of portable cardiopulmonary bypass as a resuscitative tool and its impact on long-term survival of patients in cardiac arrest, we reviewed the results of 32 consecutive patients resuscitated by cardiopulmonary bypass for cardiac arrest or severe hemodynamic compromise at Northwestern Memorial Hospital over a 2-year period. Overall survival was 12.5%. Only 1(3.4%) of the 29 patients who had cardiac arrest survived and left the hospital. All 3 patients who had severe hemodynamic compromise but not cardiac arrest were long-term
survivors. Our study suggests that portable cardiopulmonary support systems used as a resuscitative tool do not prolong the survival of most cardiac arrest patients but may be useful for patients with shock due to mechanical causes and for those with profound hemodynamic compromise due to ischemia or myocardial infarction. Portable heart-lung machines can provide patients with excellent hemodynamic support; however, neurological or cardiac recovery is unlikely once cardiac arrest occurs. (Ann Thorac Surg 1990;50:43741)
ith the advent of large-bore, thin-walled cannulas and portable heart-lung support machines, cardiopulmonary bypass (CPB) can now be instituted quickly and efficiently in virtually any hospital setting. Portable CPB is a simple and effective system for controlling a patient’s circulatory and respiratory functions on an emergency basis until physicians have time to evaluate the cause of the patient’s hemodynamic collapse and to initiate further treatment. The commercially available cardiopulmonary support (CPS) systems are portable units composed of a centrifugal pump, an oxygenator, heater, pump tubing, and venous and arterial cannulas. Because this new technology is portable and easy to apply, cardiothoracic surgeons have renewed their interest in the use of CPB as an emergency resuscitative tool for patients in cardiac arrest. Although this therapy can provide excellent cardiopulmonary resuscitation and support for a variety of patients, previous studies have not clearly shown it to be of benefit to cardiac arrest patients [l-51. To evaluate the impact of portable CPB on the longterm survival of patients with cardiac arrest, we studied the results in a series of 32 consecutive patients who had cardiac arrest or severe hemodynamic compromise and were resuscitated by emergency CPB at Northwestern University Medical Center from March 1987 through February 1989. Two thirds of the patients subsequently underwent aggressive diagnostic and therapeutic interventions directed at correcting the cause of the cardiac arrest. Our findings form the basis of this report.
Material and Methods
Accepted for publication April 11, 1990. Address reprint requests to Dr Hartz, Division of Cardiothoracic Surgery, Department of Surgery, Northwestern University Medical School, 303 E Chicago Ave, Ward Bldg, Suite 9-105, Chicago, IL 60611-3008.
0 1990 by The
Society of Thoracic Surgeons
We began using the Bard CPS system (Bard Cardiosurgery Division, C.R. Bard, Billerica, MA) at Northwestern Memorial Hospital in March 1987. Over a 2-year period from March 1987 through February 1989,32 patients were placed on CPB within a mean time of 9.5 minutes after telephone request for this procedure. Either a nurse trained in CPB or, on occasion, a registered perfusionist assembled, deaired, and primed the system with crystalloid solution. Through a right groin cutdown, an experienced cardiovascular surgeon injected sodium heparin into the femoral veins of these patients and inserted thin-walled, 20F Teflon cannulas into the femoral arteries and veins without gaining proximal or distal control of either vessel (Fig 1). With use of a modified Seldinger technique (61, guidewires and cannulas were inserted as soon as the common femoral artery and vein were identified. When cannulation was complete, the surgeon connected the appropriate bypass pump tubing. Once the bypass equipment was connected to the patient, pump flow was maximized by adjusting the venous cannula (which is longer than the arterial cannula and extends to the right atrium) and by infusing volume. We then decreased ventilation to one breath every 15 to 20 seconds, discontinued cardiopulmonary resuscitation, and defibrillated patients when necessary. Radial or femoral arterial lines were used for monitoring blood pressure in all patients. Swan-Ganz catheters were used for hemodynamic monitoring only in those patients who underwent additional procedures (Table 1). When central venous access was obtained in these patients, we decreased the bypass flow momentarily to prevent air from being sucked into the system. Left ventricular venting was not used. Thirty-six requests for CPB for cardiac arrest patients 0003-4975/90/$3.50
438
HARTZ ET AL PORTABLE CARDIOPULMONARY BYPASS
Ann Thorac Surg 1990;50:43741
a A
B Fig 1. (A) An experienced cardiovascular surgeon performs a right groin cut-down followed by systemic heparinization, 18-gauge femoral venipuncture, and 18-gauge femoral arterial puncture. Guidewires are inserted (0.038, 250 cm). ( B ) A 20F, 65-cm, thin-walled Teflon cannula is advanced and adjusted with endholes in right atrium.
were made in the 2-year period. Three of the requests were denied by the attending cardiothoracic surgeon. Among the remaining 33 patients, severe bilateral iliofemoral occlusive disease precluded cannulation in 1 patient and difficulties encountered in the use of the cannulas early in the series led to three other failures in instituting the bypass system. The manufacturer provided improved cannulas and guidewires in June 1987, however, and we had no subsequent cannulation difficulties. Ultimately, we were successful in connecting the CPS bypass system to 29 cardiac arrest patients. Nineteen of these 29 patients (59.0%)had cardiac arrest due to cardiogenic shock caused by ongoing myocardial infarction. In the remaining 10, cardiac arrest was due to a variety of other causes: hypothermia in 2 patients, pulmonary embolus in 2, and amniotic fluid embolus,
aortic dissection, hyperkalemia, sepsis, ventricular tachycardia, and unexplained cardiovascular collapse in 1 each (Table 2). The cardiac arrest episodes occurred in an intensive care unit in 9 patients, outside the hospital in 6, in the emergency room in 6, and in an operating room in 5; one episode each occurred in the catheterization laboratory, the radiology suite, and the hospital ward room. In contrast, 5 additional patients were in profound shock and were placed on bypass using the 20F cannula with conventional bypass equipment. Two of these patients had cardiac arrest, but the other 3 did not. Of the 2 patients who were undergoing radiopulmonary resuscitation, 1 had a reperfusion injury that occurred 15 minutes after a coronary bypass procedure was completed, and the other had a massive myocardial infarction that occurred during a thoracotomy and lobectomy for cancer.
HARTZ ET AL PORTABLE CARDIOPULMONARY BYPASS
Ann Thorac Surg 1990;50:43741
These 5 patients were treated with a CPB system consisting of thin-walled cannulas and a standard roller-pump bypass circuit with gravity drainage. This alternative method was used because 3 of the patients experienced hemodynamic collapse in or near the cardiac surgical suite and the other 2 had right ventricular lacerations that precluded the use of centrifugal pumping.
Results Cardiopulmonary bypass was used as an emergency resuscitative procedure in 29 patients with cardiac arrest and in another 3 patients in profound shock. For cardiac arrest patients, CPB was instituted in 10 of the 29 patients within 15 minutes of cardiac arrest, in a range of -15 to 30 minutes in another 9 patients, and in 30 minutes or more after cardiac arrest in the remaining 10 patients. Once the telephone request for CPB was made, a mean time of 9.5 minutes was required for the support team to place these patients on bypass (range, 5 to 35 minutes). The mean blood flow achieved on the CPS system was 4.2 L (range, 3.0 to 5.2 L). Patients were maintained on CPS for a mean duration of 2 hours 10 minutes (range, 12 minutes to 6.5 hours). Portable CPB provided good hemodynamic support in all but 1 patient (who had undergone an aortic dissection and was placed inappropriately on the bypass system). In all patients, either ventricular fibrillation spontaneously converted to a more stable rhythm shortly after initiation of CPB (sinus, junctional, or idioventricular) or such a rhythm was easily generated with one or two electrical countershocks. Nineteen of the 29 cardiac patients (65%) underwent subsequent diagnostic or therapeutic procedures, all of which we performed while the patients were on CPB. Seven patients underwent cardiac catheterization, 2 of whom had percutaneous transluminal coronary angio-
Table 2 . Diagnostic and Therapeutic Interventions During Cardiopulmonary Bypass" Intervention None Cardiac catheterization CABG IAB Pulmonary artery exploration Centrifugal VAD Valve replacement PTCA Rewarming Pneumatic VAD Dia1y sis Endocardia] resection a
No. of Patients 10 7 6 5 4 2 2 2
2 1 1 1
Most patients had more than one intervention.
IAB = intraaortic balloon; CABG = coronary artery bypass grafting; PTCA = percutaneous translurninal coronary angioplasty; VAD = ventricular assist device.
439
Table 2 . Cause of Cardiac Arrest Cause Cardiogenic shock Hypothermia Pulmonary embolus Amniotic fluid embolus Aortic dissection Hyperkalemia Unexplained cardiovascular collapse Sepsis Ventricular tachycardia
No. of Patients 19 2 2 1 1 1 1 1 1
plasty while on bypass. Cardiac surgical procedures included coronary artery bypass grafting in 6 patients, valve replacement in 2, endocardia1 resection in 1, and pulmonary artery exploration in 4. Intraaortic balloon counterpulsation was instituted in 5 of the patients. One patient had a centrifugal left ventricular assist device placed after coronary artery bypass grafting and 1 had both a centrifugal right ventricular assist device and a pneumatic left ventricular assist device. The 2 patients who were hypothermic were rewarmed on CPB, and 1 patient with hyperkalemia had dialysis through the arterial and venous bypass lines (Table 1). For patients who required a cardiac operation, full hemodynamic monitoring was instituted and bypass support was converted to right atrial and aortic cannulation with gravity drainage once the patient's chest was opened. The remaining 10 patients were determined to be clinically brain dead, and no further measures were taken. Before initiation of the subsequent procedures, patients had additional invasive monitoring, including Swan-Ganz catheter and arterial line placement. Weaning from portable CPB necessitated accurate measurement of pulmonary capillary wedge pressure, central venous pressure, and arterial pressure. Weaning was accomplished by decreasing pump flow gradually with concomitant hemodynamic observation and measurement of thermodilution cardiac outputs. Patients with adequate hemodynamics at flows of 1 to 1.5 L were then removed from portable CPB. Twenty-two of the 29 cardiac arrest patients could not be removed from CPB despite the aggressive measures shown in Table 1. Only 1 of the 7 patients who were successfully weaned from bypass (a young woman with unexplained cardiovascular collapse during a gynecological procedure) survived cardiac arrest and was able to leave the hospital. Thus, the long-term survival rate among this group of 27 cardiac arrest patients was a mere 3.7%. Among the 6 short-term (less than a week) survivors of cardiac arrest, 1patient remained neurologically intact but died of a cardiac arrhythmia 72 hours after being removed from CPB, 2 showed signs of neurological improvement but died of low cardiac output, 1 was believed to be brain dead and subsequently died of ventilatory failure, and 1
440
HARTZETAL PORTABLE CARDIOPULMONARY BYPASS
died of hemorrhagic shock from retroperitoneal hemorrhage due to a cannula complication that occurred shortly after successful pulmonary embolectomy. The last patient, a young woman with a global reperfusion injury 15 minutes after coronary artery bypass grafting, was placed back on bypass to facilitate placement of a pneumatic left ventricular assist device and centrifugal right ventricular assist device. Even though this patient was neurologically intact and showed signs of substantial recovery of cardiac function, she died 5 days later of low cardiac output and multiorgan system failure shortly after the ventricular assist devices were removed. Her family had denied a request to place her back on circulatory support. All 5 patients treated with CPS cannulas and conventional bypass equipment had good hemodynamic support, with a mean pump flow of 3.5 L. Two of these patients were placed on bypass because they had sustained right ventricular lacerations while undergoing pericardiectomy. As soon as we placed the venous cannulas in these patients, massive volume infusion was possible and we were able to complete the pericardiectomy with the patient on CPB. Both patients survived and left the hospital in good condition. The other 3 patients had hemodynamic collapse in or immediately adjacent to the operating room where conventional bypass equipment was more readily available than the CPS system. In the entire series of 32 patients who were resuscitated by portable CPB over a 2-year period, only 1 of the 29 patients who had cardiac arrest survived (3.4%),whereas all 3 patients who underwent emergency CPB and did not suffer cardiac arrest survived (100%).
Comment Portable CPB is a powerful resuscitative tool that can be used on an emergency basis almost anywhere in a hospital by an experienced surgical team. Currently available systems are safe and easy to assemble. The femoral arteriotomy created by the cannula can easily be closed primarily. Major vascular complications are slight; in our experience, only 1 patient had a major vascular complication (retroperitoneal hemorrhage from a cannula perforation) and it is not certain whether the cannula or the insertion technique actually caused the complication. Because the mortality rates of cardiogenic shock and cardiac arrest are so high [7-111, we were very aggressive in our early application of this new technology. We considered all patients with witnessed cardiac arrest who had received cardiopulmonary resuscitation immediately to be candidates for continued resuscitation with CPB. Decisions concerning further diagnosis and treatment could then be made in a controlled setting. We were able to provide this service in our hospital on a 24-hour basis. When the team was consulted, preparations for CPB were started even while the history was being obtained. Although 23 of the 29 patients had cardiac arrest in the hospital and even though CPB was instituted on the average less than 10 minutes after the request for it was made, only 1 patient with cardiac arrest survived more than 1 week. On the other hand, as expected, CPB
Ann Thorac Surg
1990;50:43741
proved an extremely efficacious method of resuscitating patients who had profound hemodynamic compromise but did not actually suffer a cardiac arrest. All 3 such patients in this series, including 1 with a massive myocardial infarction, survived, left the hospital, and are well more than 6 months after discharge. An obvious question that arises from this experience and one that poses an ethical dilemma is whether these patients had too much or too little resuscitation. Cardiopulmonary bypass cannot be continued indefinitely, and current FDA regulations allow use of the CPS equipment for a maximum of 6 hours. Should we, as physicians,. continue to support the cardiovascular systems of patients with possible brain death when this would involve major surgical procedures and expensive ventricular assist devices? Will the brains of these patients recover even when they have had absolutely no return of brainstem reflexes after 2 to 4 hours of excellent hemodynamic support? Because we do not know the answers to these questions, and because patient consent for CPB was implied in the first place, we could not recommend further diagnostic or therapeutic measures to the families of 10 of our patients who appeared to have both cardiac and brain death. These patients were slowly separated from the bypass circuit, had no evidence of cardiac ejection, and died. It is entirely possible that both cardiac and neurological activity might resume in these patients if they were supported for a longer period of time. For example. venting the left ventricle may facilitate cardiac recovery; although we did not note progressive ventricular distention in our patients, 2 of them who had left main coronary occlusions and who underwent cardiac catheterization had no evidence of cardiac mechanical activity despite the presence of spontaneous cardiac rhythms. This angiographic correlate of electromechanical dissociation was improved in 1 patient by left main coronary angioplasty, but the patient never had cardiac ejection sufficient to sustain life. Twelve hours later, the patient had diffuse hemorrhage from prolonged CPB and no neurological function and died. This patient is a typical example of the problems involved with resuscitating the cardiovascular system but not the patient. Ideally, a multicenter trial of resuscitation with CPB will be undertaken, and patients such as this one will be randomized to various treatment arms after informed consent is obtained from the family. In the meantime, we currently use CPB as a resuscitative tool in patients with a mechanical cause for shock (such as ventricular injury during repeat sternotomy) and for patients who are profoundly ischemic but who have not yet had a cardiac arrest. Patients who have an out-ofhospital arrest are no longer routinely considered, nor are those who have cardiac arrest in the hospital but who had cardiopulmonary resuscitation for more than 30 minutes. The thin-walled cannulas themselves can be extremely useful, and good bypass flows can be achieved even with gravity drainage. For this reason, we now place angiocatheters in both the femoral artery and vein of all our patients undergoing redo cardiac operations. The CPS cannulas are in the room, the pump is primed, and the perfusionist is present. Surgical exposure of the femoral
Ann Thorac Surg 1990;50:437-41
vessels is therefore unnecessary for rapid institution of femoral bypass. As with most new technologies, the exact role of portable CPB systems will take some time to clarify. We are not discouraged by our early results, however, because 4 of the 32 patients who would have died without this intervention did survive (including one young woman with a cardiac arrest). In the future, we will try to apply this technology earlier in the course of treatment. We also hope to be part of a multicenter trial organized to define not only patient selection but also the degree and length of cardiac resuscitation that should be used in cardiac arrest patients.
References 1. Egan TM, Duffin J, Glynn MFX, et al. Ten-year experience with extracorporeal membrane oxygenation for severe respiratory failure. Chest 1988;94:681-7. 2. Pennington DG, Merjavy JP, Codd JE, Swartz MT, Miller LL, Williams GA. Extracorporeal membrane oxygenation for patients with cardiogenic shock. Circulation 1984;7O(Suppl l): 13C7. 3 . Phillips SJ, Ballentine B, Slonine D, et al. Percutaneous initiation of cardiopulmonary bypass. Ann Thorac Surg 1983; 36:22%5.
HARTZETAL PORTABLE CARDIOPULMONARY BYPASS
441
4. Phillips SJ, Zeff RH, Kongtahworn C, et al. Percutaneous cardiopulmonary bypass: application and indication for use. Ann Thorac Surg 1989;47121-3. 5. Vogel RA, Tommaso CL, Gundry SR. Initial experience with coronary angioplasty and aortic valvuloplasty using elective semipercutaneous cardiopulmonary support. Am J Cardiol 1988;62:811-3. 6. Seldinger SI. Catheter replacement of the needle in percutaneous angiography: a new technique. Acta Radio1 1953;39: 368. 7. Kanter KR, Ruzevich SA, Pennington DG, McBride LR, Swartz MT, Willman VL. Follow-up of survivors of mechanical circulatory support. J Thorac Cardiovasc Surg 1988;96:7280. 8. Pennock JL, Pierce WS, Wisman CB, Bull AP, Waldhausen JA. Survival and complications following ventricular assist pumping for cardiogenic shock. Ann Surg 1983;198:469-78. 9. Bedell SE, Delbanco TL, Cook EF, Epstein FH. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983;309:569-76. 10. Kellerman AL, Staves DR, Hackman BB. In-hospital resuscitation following unsuccessful prehospital advanced cardiac life support: “heroic efforts” or an exercise in futility? Ann Emerg Med 1988;17:589-94. 11. McGrath RB. In-house cardiopulmonary resuscitation-after a quarter of a century. Ann Emerg Med 1987;16:1365-8.
To evaluate the use of portable cardiopulmonary bypass as a resuscitative tool and its impact on long-term survival of patients in cardiac arrest, we reviewed the results of 32 consecutive patients resuscitated by cardiopulmonary bypass for cardiac arrest or severe hemodynamic compromise at Northwestern Memorial Hospital over a 2-year period. Overall survival was 12.5%. Only 1(3.4%) of the 29 patients who had cardiac arrest survived and left the hospital. All 3 patients who had severe hemodynamic compromise but not cardiac arrest were long-term
survivors. Our study suggests that portable cardiopulmonary support systems used as a resuscitative tool do not prolong the survival of most cardiac arrest patients but may be useful for patients with shock due to mechanical causes and for those with profound hemodynamic compromise due to ischemia or myocardial infarction. Portable heart-lung machines can provide patients with excellent hemodynamic support; however, neurological or cardiac recovery is unlikely once cardiac arrest occurs. (Ann Thorac Surg 1990;50:43741)
ith the advent of large-bore, thin-walled cannulas and portable heart-lung support machines, cardiopulmonary bypass (CPB) can now be instituted quickly and efficiently in virtually any hospital setting. Portable CPB is a simple and effective system for controlling a patient’s circulatory and respiratory functions on an emergency basis until physicians have time to evaluate the cause of the patient’s hemodynamic collapse and to initiate further treatment. The commercially available cardiopulmonary support (CPS) systems are portable units composed of a centrifugal pump, an oxygenator, heater, pump tubing, and venous and arterial cannulas. Because this new technology is portable and easy to apply, cardiothoracic surgeons have renewed their interest in the use of CPB as an emergency resuscitative tool for patients in cardiac arrest. Although this therapy can provide excellent cardiopulmonary resuscitation and support for a variety of patients, previous studies have not clearly shown it to be of benefit to cardiac arrest patients [l-51. To evaluate the impact of portable CPB on the longterm survival of patients with cardiac arrest, we studied the results in a series of 32 consecutive patients who had cardiac arrest or severe hemodynamic compromise and were resuscitated by emergency CPB at Northwestern University Medical Center from March 1987 through February 1989. Two thirds of the patients subsequently underwent aggressive diagnostic and therapeutic interventions directed at correcting the cause of the cardiac arrest. Our findings form the basis of this report.
Material and Methods
Accepted for publication April 11, 1990. Address reprint requests to Dr Hartz, Division of Cardiothoracic Surgery, Department of Surgery, Northwestern University Medical School, 303 E Chicago Ave, Ward Bldg, Suite 9-105, Chicago, IL 60611-3008.
0 1990 by The
Society of Thoracic Surgeons
We began using the Bard CPS system (Bard Cardiosurgery Division, C.R. Bard, Billerica, MA) at Northwestern Memorial Hospital in March 1987. Over a 2-year period from March 1987 through February 1989,32 patients were placed on CPB within a mean time of 9.5 minutes after telephone request for this procedure. Either a nurse trained in CPB or, on occasion, a registered perfusionist assembled, deaired, and primed the system with crystalloid solution. Through a right groin cutdown, an experienced cardiovascular surgeon injected sodium heparin into the femoral veins of these patients and inserted thin-walled, 20F Teflon cannulas into the femoral arteries and veins without gaining proximal or distal control of either vessel (Fig 1). With use of a modified Seldinger technique (61, guidewires and cannulas were inserted as soon as the common femoral artery and vein were identified. When cannulation was complete, the surgeon connected the appropriate bypass pump tubing. Once the bypass equipment was connected to the patient, pump flow was maximized by adjusting the venous cannula (which is longer than the arterial cannula and extends to the right atrium) and by infusing volume. We then decreased ventilation to one breath every 15 to 20 seconds, discontinued cardiopulmonary resuscitation, and defibrillated patients when necessary. Radial or femoral arterial lines were used for monitoring blood pressure in all patients. Swan-Ganz catheters were used for hemodynamic monitoring only in those patients who underwent additional procedures (Table 1). When central venous access was obtained in these patients, we decreased the bypass flow momentarily to prevent air from being sucked into the system. Left ventricular venting was not used. Thirty-six requests for CPB for cardiac arrest patients 0003-4975/90/$3.50
438
HARTZ ET AL PORTABLE CARDIOPULMONARY BYPASS
Ann Thorac Surg 1990;50:43741
a A
B Fig 1. (A) An experienced cardiovascular surgeon performs a right groin cut-down followed by systemic heparinization, 18-gauge femoral venipuncture, and 18-gauge femoral arterial puncture. Guidewires are inserted (0.038, 250 cm). ( B ) A 20F, 65-cm, thin-walled Teflon cannula is advanced and adjusted with endholes in right atrium.
were made in the 2-year period. Three of the requests were denied by the attending cardiothoracic surgeon. Among the remaining 33 patients, severe bilateral iliofemoral occlusive disease precluded cannulation in 1 patient and difficulties encountered in the use of the cannulas early in the series led to three other failures in instituting the bypass system. The manufacturer provided improved cannulas and guidewires in June 1987, however, and we had no subsequent cannulation difficulties. Ultimately, we were successful in connecting the CPS bypass system to 29 cardiac arrest patients. Nineteen of these 29 patients (59.0%)had cardiac arrest due to cardiogenic shock caused by ongoing myocardial infarction. In the remaining 10, cardiac arrest was due to a variety of other causes: hypothermia in 2 patients, pulmonary embolus in 2, and amniotic fluid embolus,
aortic dissection, hyperkalemia, sepsis, ventricular tachycardia, and unexplained cardiovascular collapse in 1 each (Table 2). The cardiac arrest episodes occurred in an intensive care unit in 9 patients, outside the hospital in 6, in the emergency room in 6, and in an operating room in 5; one episode each occurred in the catheterization laboratory, the radiology suite, and the hospital ward room. In contrast, 5 additional patients were in profound shock and were placed on bypass using the 20F cannula with conventional bypass equipment. Two of these patients had cardiac arrest, but the other 3 did not. Of the 2 patients who were undergoing radiopulmonary resuscitation, 1 had a reperfusion injury that occurred 15 minutes after a coronary bypass procedure was completed, and the other had a massive myocardial infarction that occurred during a thoracotomy and lobectomy for cancer.
HARTZ ET AL PORTABLE CARDIOPULMONARY BYPASS
Ann Thorac Surg 1990;50:43741
These 5 patients were treated with a CPB system consisting of thin-walled cannulas and a standard roller-pump bypass circuit with gravity drainage. This alternative method was used because 3 of the patients experienced hemodynamic collapse in or near the cardiac surgical suite and the other 2 had right ventricular lacerations that precluded the use of centrifugal pumping.
Results Cardiopulmonary bypass was used as an emergency resuscitative procedure in 29 patients with cardiac arrest and in another 3 patients in profound shock. For cardiac arrest patients, CPB was instituted in 10 of the 29 patients within 15 minutes of cardiac arrest, in a range of -15 to 30 minutes in another 9 patients, and in 30 minutes or more after cardiac arrest in the remaining 10 patients. Once the telephone request for CPB was made, a mean time of 9.5 minutes was required for the support team to place these patients on bypass (range, 5 to 35 minutes). The mean blood flow achieved on the CPS system was 4.2 L (range, 3.0 to 5.2 L). Patients were maintained on CPS for a mean duration of 2 hours 10 minutes (range, 12 minutes to 6.5 hours). Portable CPB provided good hemodynamic support in all but 1 patient (who had undergone an aortic dissection and was placed inappropriately on the bypass system). In all patients, either ventricular fibrillation spontaneously converted to a more stable rhythm shortly after initiation of CPB (sinus, junctional, or idioventricular) or such a rhythm was easily generated with one or two electrical countershocks. Nineteen of the 29 cardiac patients (65%) underwent subsequent diagnostic or therapeutic procedures, all of which we performed while the patients were on CPB. Seven patients underwent cardiac catheterization, 2 of whom had percutaneous transluminal coronary angio-
Table 2 . Diagnostic and Therapeutic Interventions During Cardiopulmonary Bypass" Intervention None Cardiac catheterization CABG IAB Pulmonary artery exploration Centrifugal VAD Valve replacement PTCA Rewarming Pneumatic VAD Dia1y sis Endocardia] resection a
No. of Patients 10 7 6 5 4 2 2 2
2 1 1 1
Most patients had more than one intervention.
IAB = intraaortic balloon; CABG = coronary artery bypass grafting; PTCA = percutaneous translurninal coronary angioplasty; VAD = ventricular assist device.
439
Table 2 . Cause of Cardiac Arrest Cause Cardiogenic shock Hypothermia Pulmonary embolus Amniotic fluid embolus Aortic dissection Hyperkalemia Unexplained cardiovascular collapse Sepsis Ventricular tachycardia
No. of Patients 19 2 2 1 1 1 1 1 1
plasty while on bypass. Cardiac surgical procedures included coronary artery bypass grafting in 6 patients, valve replacement in 2, endocardia1 resection in 1, and pulmonary artery exploration in 4. Intraaortic balloon counterpulsation was instituted in 5 of the patients. One patient had a centrifugal left ventricular assist device placed after coronary artery bypass grafting and 1 had both a centrifugal right ventricular assist device and a pneumatic left ventricular assist device. The 2 patients who were hypothermic were rewarmed on CPB, and 1 patient with hyperkalemia had dialysis through the arterial and venous bypass lines (Table 1). For patients who required a cardiac operation, full hemodynamic monitoring was instituted and bypass support was converted to right atrial and aortic cannulation with gravity drainage once the patient's chest was opened. The remaining 10 patients were determined to be clinically brain dead, and no further measures were taken. Before initiation of the subsequent procedures, patients had additional invasive monitoring, including Swan-Ganz catheter and arterial line placement. Weaning from portable CPB necessitated accurate measurement of pulmonary capillary wedge pressure, central venous pressure, and arterial pressure. Weaning was accomplished by decreasing pump flow gradually with concomitant hemodynamic observation and measurement of thermodilution cardiac outputs. Patients with adequate hemodynamics at flows of 1 to 1.5 L were then removed from portable CPB. Twenty-two of the 29 cardiac arrest patients could not be removed from CPB despite the aggressive measures shown in Table 1. Only 1 of the 7 patients who were successfully weaned from bypass (a young woman with unexplained cardiovascular collapse during a gynecological procedure) survived cardiac arrest and was able to leave the hospital. Thus, the long-term survival rate among this group of 27 cardiac arrest patients was a mere 3.7%. Among the 6 short-term (less than a week) survivors of cardiac arrest, 1patient remained neurologically intact but died of a cardiac arrhythmia 72 hours after being removed from CPB, 2 showed signs of neurological improvement but died of low cardiac output, 1 was believed to be brain dead and subsequently died of ventilatory failure, and 1
440
HARTZETAL PORTABLE CARDIOPULMONARY BYPASS
died of hemorrhagic shock from retroperitoneal hemorrhage due to a cannula complication that occurred shortly after successful pulmonary embolectomy. The last patient, a young woman with a global reperfusion injury 15 minutes after coronary artery bypass grafting, was placed back on bypass to facilitate placement of a pneumatic left ventricular assist device and centrifugal right ventricular assist device. Even though this patient was neurologically intact and showed signs of substantial recovery of cardiac function, she died 5 days later of low cardiac output and multiorgan system failure shortly after the ventricular assist devices were removed. Her family had denied a request to place her back on circulatory support. All 5 patients treated with CPS cannulas and conventional bypass equipment had good hemodynamic support, with a mean pump flow of 3.5 L. Two of these patients were placed on bypass because they had sustained right ventricular lacerations while undergoing pericardiectomy. As soon as we placed the venous cannulas in these patients, massive volume infusion was possible and we were able to complete the pericardiectomy with the patient on CPB. Both patients survived and left the hospital in good condition. The other 3 patients had hemodynamic collapse in or immediately adjacent to the operating room where conventional bypass equipment was more readily available than the CPS system. In the entire series of 32 patients who were resuscitated by portable CPB over a 2-year period, only 1 of the 29 patients who had cardiac arrest survived (3.4%),whereas all 3 patients who underwent emergency CPB and did not suffer cardiac arrest survived (100%).
Comment Portable CPB is a powerful resuscitative tool that can be used on an emergency basis almost anywhere in a hospital by an experienced surgical team. Currently available systems are safe and easy to assemble. The femoral arteriotomy created by the cannula can easily be closed primarily. Major vascular complications are slight; in our experience, only 1 patient had a major vascular complication (retroperitoneal hemorrhage from a cannula perforation) and it is not certain whether the cannula or the insertion technique actually caused the complication. Because the mortality rates of cardiogenic shock and cardiac arrest are so high [7-111, we were very aggressive in our early application of this new technology. We considered all patients with witnessed cardiac arrest who had received cardiopulmonary resuscitation immediately to be candidates for continued resuscitation with CPB. Decisions concerning further diagnosis and treatment could then be made in a controlled setting. We were able to provide this service in our hospital on a 24-hour basis. When the team was consulted, preparations for CPB were started even while the history was being obtained. Although 23 of the 29 patients had cardiac arrest in the hospital and even though CPB was instituted on the average less than 10 minutes after the request for it was made, only 1 patient with cardiac arrest survived more than 1 week. On the other hand, as expected, CPB
Ann Thorac Surg
1990;50:43741
proved an extremely efficacious method of resuscitating patients who had profound hemodynamic compromise but did not actually suffer a cardiac arrest. All 3 such patients in this series, including 1 with a massive myocardial infarction, survived, left the hospital, and are well more than 6 months after discharge. An obvious question that arises from this experience and one that poses an ethical dilemma is whether these patients had too much or too little resuscitation. Cardiopulmonary bypass cannot be continued indefinitely, and current FDA regulations allow use of the CPS equipment for a maximum of 6 hours. Should we, as physicians,. continue to support the cardiovascular systems of patients with possible brain death when this would involve major surgical procedures and expensive ventricular assist devices? Will the brains of these patients recover even when they have had absolutely no return of brainstem reflexes after 2 to 4 hours of excellent hemodynamic support? Because we do not know the answers to these questions, and because patient consent for CPB was implied in the first place, we could not recommend further diagnostic or therapeutic measures to the families of 10 of our patients who appeared to have both cardiac and brain death. These patients were slowly separated from the bypass circuit, had no evidence of cardiac ejection, and died. It is entirely possible that both cardiac and neurological activity might resume in these patients if they were supported for a longer period of time. For example. venting the left ventricle may facilitate cardiac recovery; although we did not note progressive ventricular distention in our patients, 2 of them who had left main coronary occlusions and who underwent cardiac catheterization had no evidence of cardiac mechanical activity despite the presence of spontaneous cardiac rhythms. This angiographic correlate of electromechanical dissociation was improved in 1 patient by left main coronary angioplasty, but the patient never had cardiac ejection sufficient to sustain life. Twelve hours later, the patient had diffuse hemorrhage from prolonged CPB and no neurological function and died. This patient is a typical example of the problems involved with resuscitating the cardiovascular system but not the patient. Ideally, a multicenter trial of resuscitation with CPB will be undertaken, and patients such as this one will be randomized to various treatment arms after informed consent is obtained from the family. In the meantime, we currently use CPB as a resuscitative tool in patients with a mechanical cause for shock (such as ventricular injury during repeat sternotomy) and for patients who are profoundly ischemic but who have not yet had a cardiac arrest. Patients who have an out-ofhospital arrest are no longer routinely considered, nor are those who have cardiac arrest in the hospital but who had cardiopulmonary resuscitation for more than 30 minutes. The thin-walled cannulas themselves can be extremely useful, and good bypass flows can be achieved even with gravity drainage. For this reason, we now place angiocatheters in both the femoral artery and vein of all our patients undergoing redo cardiac operations. The CPS cannulas are in the room, the pump is primed, and the perfusionist is present. Surgical exposure of the femoral
Ann Thorac Surg 1990;50:437-41
vessels is therefore unnecessary for rapid institution of femoral bypass. As with most new technologies, the exact role of portable CPB systems will take some time to clarify. We are not discouraged by our early results, however, because 4 of the 32 patients who would have died without this intervention did survive (including one young woman with a cardiac arrest). In the future, we will try to apply this technology earlier in the course of treatment. We also hope to be part of a multicenter trial organized to define not only patient selection but also the degree and length of cardiac resuscitation that should be used in cardiac arrest patients.
References 1. Egan TM, Duffin J, Glynn MFX, et al. Ten-year experience with extracorporeal membrane oxygenation for severe respiratory failure. Chest 1988;94:681-7. 2. Pennington DG, Merjavy JP, Codd JE, Swartz MT, Miller LL, Williams GA. Extracorporeal membrane oxygenation for patients with cardiogenic shock. Circulation 1984;7O(Suppl l): 13C7. 3 . Phillips SJ, Ballentine B, Slonine D, et al. Percutaneous initiation of cardiopulmonary bypass. Ann Thorac Surg 1983; 36:22%5.
HARTZETAL PORTABLE CARDIOPULMONARY BYPASS
441
4. Phillips SJ, Zeff RH, Kongtahworn C, et al. Percutaneous cardiopulmonary bypass: application and indication for use. Ann Thorac Surg 1989;47121-3. 5. Vogel RA, Tommaso CL, Gundry SR. Initial experience with coronary angioplasty and aortic valvuloplasty using elective semipercutaneous cardiopulmonary support. Am J Cardiol 1988;62:811-3. 6. Seldinger SI. Catheter replacement of the needle in percutaneous angiography: a new technique. Acta Radio1 1953;39: 368. 7. Kanter KR, Ruzevich SA, Pennington DG, McBride LR, Swartz MT, Willman VL. Follow-up of survivors of mechanical circulatory support. J Thorac Cardiovasc Surg 1988;96:7280. 8. Pennock JL, Pierce WS, Wisman CB, Bull AP, Waldhausen JA. Survival and complications following ventricular assist pumping for cardiogenic shock. Ann Surg 1983;198:469-78. 9. Bedell SE, Delbanco TL, Cook EF, Epstein FH. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983;309:569-76. 10. Kellerman AL, Staves DR, Hackman BB. In-hospital resuscitation following unsuccessful prehospital advanced cardiac life support: “heroic efforts” or an exercise in futility? Ann Emerg Med 1988;17:589-94. 11. McGrath RB. In-house cardiopulmonary resuscitation-after a quarter of a century. Ann Emerg Med 1987;16:1365-8.