Improved Patient Survival After Cardiac Arrest Using a Cardiopulmonary Support System Robert T. Reichman, MD, Colin I. Joyo, MD, Walter P. Dembitsky, MD, Lee D. Griffith, MD, Robert M. Adamson, MD, Pat 0. Daily, MD, Paul A. Overlie, MD, Sidney C. Smith, Jr, MD, and Brian E. Jaski, MD Sharp Memorial Hospital, University of California at San Diego Medical Center, San Diego, California
A portable cardiopulmonary bypass system that can be rapidly deployed in a nonsurgical setting using nursing staff was used in 38 patients with cardiovascular collapse refractory to ACLS protocol. Percutaneous or cutdown cannulation sites were: femoral vein-femoral artery (n = 18), right internal jugular vein-femoral artery (n = 2), right atrium-ascending aorta (n = 121, or a combination approach (n = 4).Two patients could not be cannulated. Patient diagnoses were pulmonary emboli (n = 3), failed coronary angioplasty (n = 7), myocardial infarction with cardiogenic shock (n = 5), trauma (n = 7), aortic stenosis (n = 2), postcardiotomy deterioration (n = lo), deterioration after cardiac transplantation (n = 21, cardiomyopathy with shock (n = l), and ruptured ascending aortic dissection (n = 1). Ninety-five percent of patients (36 of 38) were successfully resuscitated to a stable rhythm. Eight diagnostic procedures (coronary angiography, n = 4; pulmonary angiography, n = 3; and aortography, n = 1) were performed while patients were on cardiopulmonary support. Early deaths resulted from massive hemorrhage (n = 81, inability to cannulate (n = 2), and irreversible myocardial injury (n = 10). Sixty-six percent
(24 of 36) of patients successfully cannulated underwent conversion to standard cardiopulmonary bypass with attendant operative procedure or placement of ventricular assist device or total artificial heart. Fifty percent (18 of 36) of patients cannulated were successfully weaned from cardiopulmonary support, and 17% (6136) are longterm survivors. Postweaning deaths resulted from central nervous system failure (n = 5), multisystem failure (n = 3 ) , anemia (n = 1,Jehovah's Witness), peripheral pulmonary emboli (n = 11, support withdrawn (n = 11, and sudden death due to arrhythmia (n = 1). Cardiopulmonary support has salvaged 6 patients in our series of 38 who were not able to be resuscitated by conventional techniques. The cardiopulmonary support system supports patients who have experienced sudden death and allows diagnostic and therapeutic interventions to be applied. Survival is coupled to early implementation and reversibility of the conditions that led to patient death. Cardiopulmonary support is ineffective in late cardiogenic shock resulting from massive myocardial infarction or in trauma patients with uncontrolled hemorrhage. (Ann Thoruc Surg 2990;49:202-5)
C
Material and Methods Cardiopulmonary Support System
ardiac arrest and sudden cardiac death account for approximately 300,000 deaths annually in the United States, nearly 50% of all cardiovascular deaths [l]. Techniques of closed chest massage for cardiopulmonary resuscitation (CPR) were popularized by Dr William Kouwenhoven and associates of Johns Hopkins in 1960 [2]. Since then, different methods of circulatory assistance [3, 41 have been applied to CPR techniques in an effort to improve survival. Results of CPR are a function of the mechanism of arrest, location of arrest, and underlying pathology. In 1983, Bedell and colleagues [5]reported that only 14% of patients who survived in-hospital cardiac arrest were successfully discharged. Increased understanding of extracorporeal circulation and perfusion techniques lends itself to application in the setting of cardiac arrest that is believed to be reversible with or without operative management.
Presented at the Twenty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Sep 26-28, 1988. Address reprint requests to Dr Reichman, 8010 Frost St, Suite 501, San Diego, CA 92123.
0 1990 by The Society of Thoracic Surgeons
The cardiopulmonary support (CPS) system used in all cases is manufactured by C.R. Bard. The CPS mobile cart houses a Bio-Medicus centrifugal pump controller, a 28-V battery source along with charger, a Normotherm heater, and an E cylinder (oxygen). The disposable software includes thin-walled 20F Teflon venous and arterial cannulas with sideholes, a Bio-Medicus centrifugal pump head, a Scimed heat exchanger, and a hollow-fiber oxygenator, all prepackaged with connecting tubing. Stored along with the hardware and software package are sterile instrument trays for vascular cutdown procedures and a chest tray with a USCI 1969 venous cannula and Sarns 6.5-mm high-flow aortic cannulas. With this equipment, cannulation sites may be individualized, with venous drainage through the right or left femoral veins, right internal jugular vein, right atrium, or combinations of these. Ideally, the tip of the venous cannula should be in the right atrium. Similarly, arterial return may be accomplished through the right or left femoral artery or the ascending aorta. Peripheral cannulation may be performed by surgical exposure of the surface of the access 0003-4975/90/$3.50
102
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
Ann Thorac Surg 1990;49:101-5
Table 1. Patient Characteristics"
Table 2. Patients Transferred to Operating Room
Sex
Range
Mean
F M
12-83
51 56
g78
Patients in whom cardiopulmonary support was attempted between June 1986 and August 1988. a
vessels or by percutaneous puncture using modified Seldinger technique [6, 71. The cart and instrument packs are stored in the intensive care unit. If the device is required elsewhere in the hospital, it is transported by a trained intensive care unit nurse who is responsible for priming and conduct of CPS. Placement of cannulas is performed by cardiovascular surgeons, trauma surgeons, or cardiologists. A perfusionist takes out-of-hospital call responsibility and is readily available. Flows of up to 6 L/min have been obtained with satisfactorily placed cannulas. Diagnostic studies such as pulmonary angiography, coronary angiography, or aortography can be performed without difficulty while the patient is on CPS. Definitive operative procedures such as pulmonary embolectomy, coronary artery bypass grafting, major arterial repair, or placement of assist devices (left ventricular, right ventricular, biventricular, or total artificial heart) are easily performed. If conversion to standard cardiopulmonary bypass (CPB) is chosen, a Hemocron filter is used to lessen the problems of fluid overload and excess hemodilution.
Patients From June 1986, to August 1988, CPS was used in 38 patients with cardiac arrest at Sharp Memorial Hospital, University of California at San Diego Medical Center, or Palomar Memorial Hospital. Each of these 38 patients had sudden catastrophic loss of their cardiac rhythm and hemodynamics and could not be resuscitated by ACLS protocol [4]. Of the 38 patients in the series, 13 (34%)were female and 25 (66%)were male. Ages ranged from 8 to 83 years (mean, 57 years) (Table 1). Among the 38 patients, 10 (26%) had postcardiotomy deterioration, 7 (18%) had failed coronary angioplasty, 7 (18%)had major trauma, 5 (14%)had myocardial infarction with cardiogenic shock, 3 (8%)had massive pulmonary embolus, 2 (5%)had aortic stenosis, 2 (5%)had deterioration after cardiac transplantation, 1 (3%)had cardiomyopathy with shock, and 1 (3%) had a ruptured acute dissection of the ascending aorta. Cannulation approach was individualized. Twelve patients (32%) in whom a recent sternotomy or lateral thoracotomy was accessible were cannulated through the right atrium and ascending aorta directly. Femoral vein and femoral artery cannulation were used in 18 patients (47%).The right internal jugular vein and femoral artery were used in 2 patients (5%). A combination of approaches was used in 6 patients (16%). Eight diagnostic procedures were performed while pa-
Early Survivors
Late Survivors
3 3 1
6 5 2 2 1
3 3 0 0 0
1 24
1 17
0 6
Procedure
n
Control of hemorrhage Coronary artery bypass grafting Pulmonary embolectomy Repair of aorta
9
7
Placement of left ventricular assist device Placement of Jarvik-7 Total
tients were on CPS. Four patients had coronary angiograms, 3 had pulmonary angiograms, and 1 had aortography.
Results Of 38 patients in whom CPS was attempted, 36 (95%) were successfully resuscitated to a stable rhythm. The remaining 2 patients represent our early experience in which cannulation through the femoral vein was unsuccessful and no further attempts were made at venous access. Eighteen of the 36 patients in whom CPS was successfully placed were successfully weaned. Of those who could not be weaned, 8 had massive uncontrollable hemorrhage and 10 had massive myocardial infarction. Twenty-four patients were taken to the operating room, 9 for control of hemorrhage, 7 for coronary artery bypass grafting, 3 for pulmonary embolectomy, 3 for repair of the aorta, 1 for placement of a left ventricular assist device, and 1 for explanation of a transplant graft and implantation of a Jarvik-7 artificial heart. Sixteen of these 24 patients were transferred to standard CPB (Tables 2, 3). Twelve patients who were weaned from CPS died later in their hospital course: 5 died of central nervous system failure, 3 of multisystem failure, 1 of anemia, and 1 of peripheral pulmonary emboli not surgically accessible; in 1 patient support was withdrawn, and there was one sudden death. One of the central nervous system failures resulted from anoxic insult in which oxygen was not delivered to the perfusion circuit. The patient with severe
Table 3 . Results in 36 Surviving Patients Who Received Cardiopulmonary Support (n = 36) Result
n (%)
Stable hemodynamics
34 (94) 24 (67) 16 (44) 18 (SOY
Taken to operating room Converted to full CPB Early survivors (weaned from CPS) Late survivors
6 (17)
One Jarvik-7, one left ventricular assist device, one biventricular assist device.
a
CPB = cardiopulmonary bypass;
CPS
=
cardiopulmonary support.
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
Ann Thorac Surg 1990;49:101-5
anemia refused blood transfusion on the basis of religious beliefs. The patient from whom support was withdrawn had a ruptured coronary artery after angioplasty. This patient and her family declined operative intervention. She died 60 minutes later. Six patients survived long term and were discharged from the hospital. Three of these patients underwent coronary artery bypass grafting after failed angioplasty. One of these patients had previously undergone coronary artery bypass grafting. Sternotomy in this patient while on CPS was greatly facilitated in that the heart was decompressed and all adhesions were stretched. The other 3 patients who survived long term were resuscitated from massive hemorrhage after cardiotomy.
Comment Modifications of CPR techniques have evolved based on anecdotal data and have come under scrutiny by the scientific community [8]. Closed cardiac massage can provide only 10% of normal cerebral blood flow and only 5% of myocardial blood flow [8]. Use of open chest cardiac massage may well provide better blood flow to these critical areas. Decades have passed since the inception of CPR, but in-hospital resuscitation still has a poor prognosis, with only 14% of resuscitated patients living long enough to leave the hospital [5]. An external power source to provide CPS has long been desired. The concept of extracorporeal CPS support was set forth by LeGallois in 1813 [9], although it did not become a clinical reality until 1953 as performed by Gibbon [lo]. The initial impetus to develop CPB was to provide adequate systemic circulation in patients with a massive pulmonary embolus. The simple pump-oxygenator rapidly grew into massive electronic consoles requiring four roller-pump heads [ll, 121. This, coupled with unacceptable embolic rates and disruption of formed blood elements, slowed clinical progress [131. Configurations of venovenous and venoarterial bypass were evaluated. Goldman and colleagues [14] showed the superiority of venoarterial pumping in ventricular fibrillation. Because access to the venous reservoir provided improved preload to the system, higher flows were obtained. Use of membrane oxygenators lessened the problem of disruption of the formed blood elements with prolonged perfusion [ 151. Techniques of percutaneous femoral cannulation [16, 171 along with improved understanding of extracorporeal perfusion methods ushered in the concept of a portable CPS system to be used in the hospital. We used CPS in patients in whom standard CPR was ineffective and in whom a fatal arrhythmia was believed to be reversible. We also used it in cardiac arrest patients in whom the diagnosis was unclear. In such patients, CPS provided time to perform diagnostic studies. When aortography is performed, the CPS must be slowed and the heart allowed to eject. Similarly, pulmonary angiography requires that the contrast medium be injected distal to the pulmonary valve with the right heart ejecting. Because of the additional data obtained from these studies, patients
103
with no reversible pathology were spared an unnecessary thoracic procedure. Mattox and Beall [9] reported 39 patients in a moribund state who were placed on a portable CPB: 15 were long-term survivors; 13 of them had had massive pulmonary embolus. Phillips and colleagues reported 21 patients with refractory cardiac arrest. Seventeen patients could be resuscitated; 4 trauma patients could not [Phillips SJ, personal communication]. The clinical effectiveness of CPS is difficult to evaluate. Use of CPS improves tissue perfusion, as evidenced by correction of acidosis and ability to cardiovert the heart to a regular rhythm. We estimate that mortality would be 100%in our group of patients if they had not been treated. Sixteen percent of our treated patients survived. These results closely paralleled the early results of a 1973 study [181 using intraaortic balloon counterpulsation to treat cardiogenic shock: of 87 patients, there were 35 early survivors but only 15 survivors left the hospital. It is natural for physicians to be hesitant to apply any new mechanical support devices with unproven track records in only the most desperate situations in hopelessly ill patients. We trust that these results will not delay use of this technology in select patients.
Conclusion Our early experience in these patients was punctuated with difficulty in cannulation as well as a high proportion of central nervous system dysfunction. Design changes have remedied the former problem. Late institution of CPS coupled with ineffective CPR have prompted us to initiate CPS earlier and may produce less central nervous system dysfunction. Use of CPS in trauma patients has been disappointing in our series as well as those of other investigators [17, Phillips SJ, personal communication]. Heparinization tends to worsen the grave situation. With the advent of heparinless oxygenators around the corner, careful evaluation of CPS in these patients is warranted. We had limited success using CPS in patients with acute myocardial infarction before onset of cardiogenic shock. Resuscitation of the postcardiotomy patient who exsanguinates is performed easily with CPS. Cardiopulmonary support may be used in hypothermic patients or those with cardiodepressant drug toxicity. Although mortality may not decrease with more widespread use of CPS in cardiac arrest, we recommend a careful clinical trial involving earlier implementation in select critically ill patients.
References 1. Lown 8. Cardiovascular collapse and sudden cardiac death. In: Braunwald E, ed. Heart disease. A textbook of cardiovascular medicine; 2nd ed. Philadelphia: WB Saunders, 1984: 774-806. 2. Kouwenhoven WB, JudeJR, Knickerbocker GG. Closed chest cardiac massage. JAMA 1960;173:1064-7. 3. Austen WG, Buckley MJ, Mundth ED, Sanders CA. Assisted circulation: intraaortic balloon pumping. Transplant Proc 1971;3:1474.
104
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
4. American Heart Association. Standards and guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). JAMA 1980;244:462-8. 5. Bedell SE, Delbanco TL, Cook EF, Epstein FH. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983;309:569-608. 6. Seldinger 51. Catheter replacement of the needle in percutaneous arteriography. Acta Radiologica 1953;39:36%76. 7. Phillips SJ, Ballentine B, Slonine D. Percutaneous initiation of cardiopulmonary bypass. Ann Thorac Surg 1983;36:223-5. 8. Wexler HR, Gelb AW. Controversies in cardiopulmonary resuscitation. Crit Care Clin 1986;2:2335-48. 9. Mattox KL, Beall AC Jr. Resuscitation of the moribund patient using portable cardiopulmonary bypass. Ann Thorac Surg 1976;22:436-42. 10. Gibbon JH Jr. Artificial maintenance of circulation during experimental occlusion of pulmonary artery. Arch Surg 1937; 34:1105. 11. Miller BJ, Gibbon JH Jr, Fineberg C. An improved mechanical
Ann Thorac Surg 1990;49:101-5
12. 13. 14. 15. 16. 17. 18.
heart and lung apparatus. Med Clin North Am 1953; 3 7 1603-24. Jones RF, Donald DE, Swan HJC, Harshbarger HC, Kirklin JW, Wood EH. Apparatus of the Gibbon type for mechanical bypass of the heart and lungs. Mayo Clin Proc 1955;3010515. Miller BJ. The development of heart lung machines. Surg Gynecol Obstet 1982;154:403-14. Goldman A, Boszormenyi E, Utsu F, Enescu V, Swan HJC, Corday E. Veno-arterial pulsatile partial bypass for circulatory assist. Dis Chest 1966;50:63?40. Lande AJ. Assisted circulation with membrane oxygenators. Transplant Proc 1971;3:1479-89. Bregman D, Casarella WJ. Percutaneous intraaortic balloon pumping: initial clinical experience. Ann Thorac Surg 1979; 29: 153-5. Phillips SJ. Percutaneous cardiopulmonary bypass and innovations in clinical counterpulsation. Crit Care Clin 1986; 2:297-318. Scheidt S, Wilmer G, Mueller H, et al. Intraaortic balloon counterpulsation in cardiogenic shock. N Engl J Med 1973; 288:979-84.
DISCUSSION DR WALTER E. PAE, JR (Hershey, PA): I congratulate Dr Reichman and the group at Sharp Memorial for their fine presentation and thank them for the privilege of reviewing their manuscript before this presentation. At The Pennsylvania State University Hospital in conjunction with our colleagues at the Harrisburg Hospital, using this same CPS system during the last 3 years, we initiated bypass outside the operating room in 15 patients after failure of standard advanced life support measures. The patients’ diagnoses were quite similar to those of the patients of Dr Reichman and colleagues. All 15 patients were stabilized hemodynamically, with mean bypass flows of approximately 4L/min; they had average bypass times of 110 minutes. Similar to the experience of Reichman and associates, 7 of our 15 patients, or about 50%, were ultimately weaned. Only 2 of the 7 patients had cardiac and neurological recovery, however. Nonetheless, there were no survivors when the system was used in patients receiving cardiopulmonary resuscitation longer than 30 or 40 minutes before bypass or in patients who had an unwitnessed cardiac arrest. In our experience, cannulation and therefore rapidity of deployment has been most optimal in the cardiac catheterization laboratory, using a modified Seldinger technique such as Reichman and associates described, in which the femoral vessels were surgically visualized-before puncture and fluoroscopy was available. Cannulation is also easily achieved in the intensive care unit, where the chest of a postoperative cardiac patient can be rapidly reopened and direct cannulation can be achieved. Our early experience using a true Seldinger technique, although successful in achieving cannulation, was more cumbersome and time-consuming in patients without palpable pulses and resulted in some complex vascular injuries. This technology does represent a powerful adjunctive therapy in properly selected patients to allow subsequent diagnostic and therapeutic maneuvers, but unless the uncovered pathology is correctable and the intervention itself does not lead to significant morbidity, meaningful recovery will probably not occur. I would like to direct several questions to Dr Reichman. First, in terms of patient selection, d o you believe that this technology has any application in the treatment of an unwitnessed cardiac
arrest or have you placed a time limit on pre-bypass CPR as a patient selection criterion? Second, in deference to our cardiology colleagues and despite improvements in cannula design, do you believe that direct visualization of the peripheral vessels is important to limit complex vascular injuries and cannula malposition, and to increase the chances of a rapid successful cannulation? Last, do you foresee an expanded role or have unreported experience in use of this particular technology as a n adjunct to complex angioplasty or valvoplasty procedures in carefully, mutually selected high-risk patients? DR WILLIAM B. IAMS (Camp Hill, PA): I am concerned that this technology might be expanded to noncardiac surgeons and be run by nonperfusionists. I know that there is a commercial impetus to do this, but my particular concern is that the patient will be killed by the machine rather than saved. As shown by this series and our series, the salvage rate is small even when CPS is performed by cardiac surgeons and perfusionists. What if the CPS system is placed by less than experienced people and is run by less than experienced perfusionists? It is CPB, and I believe that this issue of credentials will arise in your own hospitals. How many survivors were cannulated from the right atrium to ascending aorta? I believe that 12 of your patients were treated in this way. Could conventional CPB not be instituted in these cases? This technology has expanded with the great interest in providing femorofemoral bypass with excellent cannulas. These cannulas, in my experience, have been very helpful in redo coronary artery bypass patients: I can place a small catheter in the femoral artery and a small catheter in the femoral vein and, if I need to, I can cannulate percutaneously with these cannulas in a matter of minutes if a redo is decompensating and I cannot get the heart free of adhesions. Finally, I believe that this technology is best used in the deteriorating patient but that it should not be instituted in the obviously dead patient. DR HOOSHANG BOLOOKI (Miami, FL): I congratulate Dr Reichman and his associates for a great deal of work. This
Ann Thorac Surg 1990;49:101-5
procedure requires a lot of manpower, but, as you note, provides little reward. So long as CPS equipment is used in the cardiac catheterization laboratory for complications of angioplasty and possibly prophylactically for some of the difficult angioplasties or for left main disease, I believe it will be a helpful system. But when the indication for its use is expanded, I think we will have trouble. At the University of Miami between 1968 and 1971, we used a very simple protocol. We thought that a patient who was doing well a few minutes before cardiac arrest would respond to resuscitation if we could reestablish cardiac output with a pump oxygenator. We had a portable heart-lung machine in the emergency room and used it in select young patients with outof-hospital cardiac arrest if cardiac massage was not effective within ten to 15 minutes. We placed the patient on the heart-lung machine-femorofemoral bypass with oxygenation-and simultaneously opened the left chest to perform open cardiac massage. This we believed mandatory because we discovered that as soon as we stopped the external cardiac massage, the arrested heart dilated, and pulmonary edema, hemorrhage, and hypoxia followed. The protocol called for taking the patient to the operating room, discontinuing cardiac support gradually, and closing the chest. We were also seeking an acute coronary obstruction that we could fix or bypass. We performed such procedures as on-table coronary angiography. Some of the results on this work were previously reported [Circulation 1971;44:103442]. We initially used the portable heart-lung machine in 7 patients. All had had out-of-hospital cardiac arrests. We had 1 survivor; this provided the impetus for us to use the device more often. Unfortunately, in the next year we used it in 7 other patients with in-hospital cardiac arrest. We expanded the indication to patients with suspected pulmonary emboli and patients with myocardial infarction and cardiac arrest, but we had no other survivors despite a great deal of effort. Our final conclusion was that we must use some type of case stratification. Which patient would get the best result? With the small number of survivors so far, we cannot draw any conclusions. We can, however, identify patients who do not respond to this support system. We discovered that patients in the coronary care unit with cardiac arrest after cardiogenic shock were the worst candidates for this method of support. Our 1 patient who eventually did well had no coronary artery disease and had ventricular fibrillation due to cardiomyopathy that had not responded to ten minutes of cardiac massage. Dr Reichman, can you indicate which patient group benefited most from this support system? You had 6 survivors, 3 with postoperative hemorrhage leading to cardiac arrest and 3 with cardiac arrest after coronary bypass. These would be surgical patients in the intensive care unit. Are you thinking of limiting your effort to this group of patients instead of using this device in all patients with cardiac arrest in the hospital setting?
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
105
DR HENDRICK B. BARNER (St Louis, MO): The abstract indicates that nursing personnel were used to initiate CPS. Dr Reichman, could you speak to the issue of using nursing personnel versus perfusionists for managing the CPS system? D R REICHMAN Dr Pae, out initial impression, although we do not have hard data to substantiate this, is that prolonged CPR and unwitnessed arrest are poor predictors of outcome. In our patients, all cardiac arrests were in-hospital arrests. No patient in our series arrived at the emergency room in an arrested state. Furthermore, if restoration of a rhythm was made possible by CPR, use of CPS to stabilize hemodynamics met with better results. We have placed cannulas through percutaneous puncture of the right internal jugular vein, femoral vein, and femoral artery in select cases. I must caution you about this approach. One of our patients had percutaneous femoral cannulation. This patient’s groin was so thick that some of the sideholes in the arterial cannulas were extravascular. This created a lethal situation. Removal of the arterial cannulas is a procedure requiring direct suture repair. We have removed jugular venous cannulas without cutdown. As for use of the procedure in the catheterization laboratory, we have used CPS in 6 patients for elective angioplasty or valvoplasty who were considered nonsurgical candidates. These data are not included in our current report. Three of the patients had valvoplasty, 2 had coronary angioplasty, and 1 patient had combined valvoplasty and angioplasty. We performed cutdown of the femoral vessels on the right side of CPS insertion and cutdown of the left femoral artery for placement of the angioplasty or valvoplasty catheter. In an effort to decrease myocardial oxygen consumption, we cooled each patient’s temperature to 34°C. At the conclusion of the procedure, each vessel was repaired primarily. Each of these catheterization laboratory patients involved a team effort. The cardiac surgeon provided vascular access, conduct of CPB was maintained by the cardiac surgeon and perfusionist, and only the formal balloon procedure was performed by the cardiologist. Dr Bolooki, we have our unit stationed in the intensive care unit on a battery charger. Sterile vascular cutdown and reentry sternotomy trays are stored with the unit. If it is summoned to a location outside the intensive care unit, such as the medical ward or the emergency room, an intensive care unit nurse responds with the unit. Set-up and priming of the unit is her responsibility. Cannulation is performed by the surgeon; we have had trauma surgeons, cardiac surgeons, and cardiologists place the unit. Further management is performed by an on-call perfusionist who is summoned from home. The machine can be placed initially by a trained noncardiac surgeon, and initial conduct of the machine can be performed by an intensive care unit nurse with a perfusionist on call. It is inefficient to have perfusionists in house in each of these three hospitals on a 24-hour basis. Dr lams, of our 6 long-term survivors, 3 were cannulated through the chest, 2 through the femoral artery and vein, and 1 through internal jugular vein and femoral artery.
A portable cardiopulmonary bypass system that can be rapidly deployed in a nonsurgical setting using nursing staff was used in 38 patients with cardiovascular collapse refractory to ACLS protocol. Percutaneous or cutdown cannulation sites were: femoral vein-femoral artery (n = 18), right internal jugular vein-femoral artery (n = 2), right atrium-ascending aorta (n = 121, or a combination approach (n = 4).Two patients could not be cannulated. Patient diagnoses were pulmonary emboli (n = 3), failed coronary angioplasty (n = 7), myocardial infarction with cardiogenic shock (n = 5), trauma (n = 7), aortic stenosis (n = 2), postcardiotomy deterioration (n = lo), deterioration after cardiac transplantation (n = 21, cardiomyopathy with shock (n = l), and ruptured ascending aortic dissection (n = 1). Ninety-five percent of patients (36 of 38) were successfully resuscitated to a stable rhythm. Eight diagnostic procedures (coronary angiography, n = 4; pulmonary angiography, n = 3; and aortography, n = 1) were performed while patients were on cardiopulmonary support. Early deaths resulted from massive hemorrhage (n = 81, inability to cannulate (n = 2), and irreversible myocardial injury (n = 10). Sixty-six percent
(24 of 36) of patients successfully cannulated underwent conversion to standard cardiopulmonary bypass with attendant operative procedure or placement of ventricular assist device or total artificial heart. Fifty percent (18 of 36) of patients cannulated were successfully weaned from cardiopulmonary support, and 17% (6136) are longterm survivors. Postweaning deaths resulted from central nervous system failure (n = 5), multisystem failure (n = 3 ) , anemia (n = 1,Jehovah's Witness), peripheral pulmonary emboli (n = 11, support withdrawn (n = 11, and sudden death due to arrhythmia (n = 1). Cardiopulmonary support has salvaged 6 patients in our series of 38 who were not able to be resuscitated by conventional techniques. The cardiopulmonary support system supports patients who have experienced sudden death and allows diagnostic and therapeutic interventions to be applied. Survival is coupled to early implementation and reversibility of the conditions that led to patient death. Cardiopulmonary support is ineffective in late cardiogenic shock resulting from massive myocardial infarction or in trauma patients with uncontrolled hemorrhage. (Ann Thoruc Surg 2990;49:202-5)
C
Material and Methods Cardiopulmonary Support System
ardiac arrest and sudden cardiac death account for approximately 300,000 deaths annually in the United States, nearly 50% of all cardiovascular deaths [l]. Techniques of closed chest massage for cardiopulmonary resuscitation (CPR) were popularized by Dr William Kouwenhoven and associates of Johns Hopkins in 1960 [2]. Since then, different methods of circulatory assistance [3, 41 have been applied to CPR techniques in an effort to improve survival. Results of CPR are a function of the mechanism of arrest, location of arrest, and underlying pathology. In 1983, Bedell and colleagues [5]reported that only 14% of patients who survived in-hospital cardiac arrest were successfully discharged. Increased understanding of extracorporeal circulation and perfusion techniques lends itself to application in the setting of cardiac arrest that is believed to be reversible with or without operative management.
Presented at the Twenty-fourth Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Sep 26-28, 1988. Address reprint requests to Dr Reichman, 8010 Frost St, Suite 501, San Diego, CA 92123.
0 1990 by The Society of Thoracic Surgeons
The cardiopulmonary support (CPS) system used in all cases is manufactured by C.R. Bard. The CPS mobile cart houses a Bio-Medicus centrifugal pump controller, a 28-V battery source along with charger, a Normotherm heater, and an E cylinder (oxygen). The disposable software includes thin-walled 20F Teflon venous and arterial cannulas with sideholes, a Bio-Medicus centrifugal pump head, a Scimed heat exchanger, and a hollow-fiber oxygenator, all prepackaged with connecting tubing. Stored along with the hardware and software package are sterile instrument trays for vascular cutdown procedures and a chest tray with a USCI 1969 venous cannula and Sarns 6.5-mm high-flow aortic cannulas. With this equipment, cannulation sites may be individualized, with venous drainage through the right or left femoral veins, right internal jugular vein, right atrium, or combinations of these. Ideally, the tip of the venous cannula should be in the right atrium. Similarly, arterial return may be accomplished through the right or left femoral artery or the ascending aorta. Peripheral cannulation may be performed by surgical exposure of the surface of the access 0003-4975/90/$3.50
102
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
Ann Thorac Surg 1990;49:101-5
Table 1. Patient Characteristics"
Table 2. Patients Transferred to Operating Room
Sex
Range
Mean
F M
12-83
51 56
g78
Patients in whom cardiopulmonary support was attempted between June 1986 and August 1988. a
vessels or by percutaneous puncture using modified Seldinger technique [6, 71. The cart and instrument packs are stored in the intensive care unit. If the device is required elsewhere in the hospital, it is transported by a trained intensive care unit nurse who is responsible for priming and conduct of CPS. Placement of cannulas is performed by cardiovascular surgeons, trauma surgeons, or cardiologists. A perfusionist takes out-of-hospital call responsibility and is readily available. Flows of up to 6 L/min have been obtained with satisfactorily placed cannulas. Diagnostic studies such as pulmonary angiography, coronary angiography, or aortography can be performed without difficulty while the patient is on CPS. Definitive operative procedures such as pulmonary embolectomy, coronary artery bypass grafting, major arterial repair, or placement of assist devices (left ventricular, right ventricular, biventricular, or total artificial heart) are easily performed. If conversion to standard cardiopulmonary bypass (CPB) is chosen, a Hemocron filter is used to lessen the problems of fluid overload and excess hemodilution.
Patients From June 1986, to August 1988, CPS was used in 38 patients with cardiac arrest at Sharp Memorial Hospital, University of California at San Diego Medical Center, or Palomar Memorial Hospital. Each of these 38 patients had sudden catastrophic loss of their cardiac rhythm and hemodynamics and could not be resuscitated by ACLS protocol [4]. Of the 38 patients in the series, 13 (34%)were female and 25 (66%)were male. Ages ranged from 8 to 83 years (mean, 57 years) (Table 1). Among the 38 patients, 10 (26%) had postcardiotomy deterioration, 7 (18%) had failed coronary angioplasty, 7 (18%)had major trauma, 5 (14%)had myocardial infarction with cardiogenic shock, 3 (8%)had massive pulmonary embolus, 2 (5%)had aortic stenosis, 2 (5%)had deterioration after cardiac transplantation, 1 (3%)had cardiomyopathy with shock, and 1 (3%) had a ruptured acute dissection of the ascending aorta. Cannulation approach was individualized. Twelve patients (32%) in whom a recent sternotomy or lateral thoracotomy was accessible were cannulated through the right atrium and ascending aorta directly. Femoral vein and femoral artery cannulation were used in 18 patients (47%).The right internal jugular vein and femoral artery were used in 2 patients (5%). A combination of approaches was used in 6 patients (16%). Eight diagnostic procedures were performed while pa-
Early Survivors
Late Survivors
3 3 1
6 5 2 2 1
3 3 0 0 0
1 24
1 17
0 6
Procedure
n
Control of hemorrhage Coronary artery bypass grafting Pulmonary embolectomy Repair of aorta
9
7
Placement of left ventricular assist device Placement of Jarvik-7 Total
tients were on CPS. Four patients had coronary angiograms, 3 had pulmonary angiograms, and 1 had aortography.
Results Of 38 patients in whom CPS was attempted, 36 (95%) were successfully resuscitated to a stable rhythm. The remaining 2 patients represent our early experience in which cannulation through the femoral vein was unsuccessful and no further attempts were made at venous access. Eighteen of the 36 patients in whom CPS was successfully placed were successfully weaned. Of those who could not be weaned, 8 had massive uncontrollable hemorrhage and 10 had massive myocardial infarction. Twenty-four patients were taken to the operating room, 9 for control of hemorrhage, 7 for coronary artery bypass grafting, 3 for pulmonary embolectomy, 3 for repair of the aorta, 1 for placement of a left ventricular assist device, and 1 for explanation of a transplant graft and implantation of a Jarvik-7 artificial heart. Sixteen of these 24 patients were transferred to standard CPB (Tables 2, 3). Twelve patients who were weaned from CPS died later in their hospital course: 5 died of central nervous system failure, 3 of multisystem failure, 1 of anemia, and 1 of peripheral pulmonary emboli not surgically accessible; in 1 patient support was withdrawn, and there was one sudden death. One of the central nervous system failures resulted from anoxic insult in which oxygen was not delivered to the perfusion circuit. The patient with severe
Table 3 . Results in 36 Surviving Patients Who Received Cardiopulmonary Support (n = 36) Result
n (%)
Stable hemodynamics
34 (94) 24 (67) 16 (44) 18 (SOY
Taken to operating room Converted to full CPB Early survivors (weaned from CPS) Late survivors
6 (17)
One Jarvik-7, one left ventricular assist device, one biventricular assist device.
a
CPB = cardiopulmonary bypass;
CPS
=
cardiopulmonary support.
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
Ann Thorac Surg 1990;49:101-5
anemia refused blood transfusion on the basis of religious beliefs. The patient from whom support was withdrawn had a ruptured coronary artery after angioplasty. This patient and her family declined operative intervention. She died 60 minutes later. Six patients survived long term and were discharged from the hospital. Three of these patients underwent coronary artery bypass grafting after failed angioplasty. One of these patients had previously undergone coronary artery bypass grafting. Sternotomy in this patient while on CPS was greatly facilitated in that the heart was decompressed and all adhesions were stretched. The other 3 patients who survived long term were resuscitated from massive hemorrhage after cardiotomy.
Comment Modifications of CPR techniques have evolved based on anecdotal data and have come under scrutiny by the scientific community [8]. Closed cardiac massage can provide only 10% of normal cerebral blood flow and only 5% of myocardial blood flow [8]. Use of open chest cardiac massage may well provide better blood flow to these critical areas. Decades have passed since the inception of CPR, but in-hospital resuscitation still has a poor prognosis, with only 14% of resuscitated patients living long enough to leave the hospital [5]. An external power source to provide CPS has long been desired. The concept of extracorporeal CPS support was set forth by LeGallois in 1813 [9], although it did not become a clinical reality until 1953 as performed by Gibbon [lo]. The initial impetus to develop CPB was to provide adequate systemic circulation in patients with a massive pulmonary embolus. The simple pump-oxygenator rapidly grew into massive electronic consoles requiring four roller-pump heads [ll, 121. This, coupled with unacceptable embolic rates and disruption of formed blood elements, slowed clinical progress [131. Configurations of venovenous and venoarterial bypass were evaluated. Goldman and colleagues [14] showed the superiority of venoarterial pumping in ventricular fibrillation. Because access to the venous reservoir provided improved preload to the system, higher flows were obtained. Use of membrane oxygenators lessened the problem of disruption of the formed blood elements with prolonged perfusion [ 151. Techniques of percutaneous femoral cannulation [16, 171 along with improved understanding of extracorporeal perfusion methods ushered in the concept of a portable CPS system to be used in the hospital. We used CPS in patients in whom standard CPR was ineffective and in whom a fatal arrhythmia was believed to be reversible. We also used it in cardiac arrest patients in whom the diagnosis was unclear. In such patients, CPS provided time to perform diagnostic studies. When aortography is performed, the CPS must be slowed and the heart allowed to eject. Similarly, pulmonary angiography requires that the contrast medium be injected distal to the pulmonary valve with the right heart ejecting. Because of the additional data obtained from these studies, patients
103
with no reversible pathology were spared an unnecessary thoracic procedure. Mattox and Beall [9] reported 39 patients in a moribund state who were placed on a portable CPB: 15 were long-term survivors; 13 of them had had massive pulmonary embolus. Phillips and colleagues reported 21 patients with refractory cardiac arrest. Seventeen patients could be resuscitated; 4 trauma patients could not [Phillips SJ, personal communication]. The clinical effectiveness of CPS is difficult to evaluate. Use of CPS improves tissue perfusion, as evidenced by correction of acidosis and ability to cardiovert the heart to a regular rhythm. We estimate that mortality would be 100%in our group of patients if they had not been treated. Sixteen percent of our treated patients survived. These results closely paralleled the early results of a 1973 study [181 using intraaortic balloon counterpulsation to treat cardiogenic shock: of 87 patients, there were 35 early survivors but only 15 survivors left the hospital. It is natural for physicians to be hesitant to apply any new mechanical support devices with unproven track records in only the most desperate situations in hopelessly ill patients. We trust that these results will not delay use of this technology in select patients.
Conclusion Our early experience in these patients was punctuated with difficulty in cannulation as well as a high proportion of central nervous system dysfunction. Design changes have remedied the former problem. Late institution of CPS coupled with ineffective CPR have prompted us to initiate CPS earlier and may produce less central nervous system dysfunction. Use of CPS in trauma patients has been disappointing in our series as well as those of other investigators [17, Phillips SJ, personal communication]. Heparinization tends to worsen the grave situation. With the advent of heparinless oxygenators around the corner, careful evaluation of CPS in these patients is warranted. We had limited success using CPS in patients with acute myocardial infarction before onset of cardiogenic shock. Resuscitation of the postcardiotomy patient who exsanguinates is performed easily with CPS. Cardiopulmonary support may be used in hypothermic patients or those with cardiodepressant drug toxicity. Although mortality may not decrease with more widespread use of CPS in cardiac arrest, we recommend a careful clinical trial involving earlier implementation in select critically ill patients.
References 1. Lown 8. Cardiovascular collapse and sudden cardiac death. In: Braunwald E, ed. Heart disease. A textbook of cardiovascular medicine; 2nd ed. Philadelphia: WB Saunders, 1984: 774-806. 2. Kouwenhoven WB, JudeJR, Knickerbocker GG. Closed chest cardiac massage. JAMA 1960;173:1064-7. 3. Austen WG, Buckley MJ, Mundth ED, Sanders CA. Assisted circulation: intraaortic balloon pumping. Transplant Proc 1971;3:1474.
104
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
4. American Heart Association. Standards and guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). JAMA 1980;244:462-8. 5. Bedell SE, Delbanco TL, Cook EF, Epstein FH. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983;309:569-608. 6. Seldinger 51. Catheter replacement of the needle in percutaneous arteriography. Acta Radiologica 1953;39:36%76. 7. Phillips SJ, Ballentine B, Slonine D. Percutaneous initiation of cardiopulmonary bypass. Ann Thorac Surg 1983;36:223-5. 8. Wexler HR, Gelb AW. Controversies in cardiopulmonary resuscitation. Crit Care Clin 1986;2:2335-48. 9. Mattox KL, Beall AC Jr. Resuscitation of the moribund patient using portable cardiopulmonary bypass. Ann Thorac Surg 1976;22:436-42. 10. Gibbon JH Jr. Artificial maintenance of circulation during experimental occlusion of pulmonary artery. Arch Surg 1937; 34:1105. 11. Miller BJ, Gibbon JH Jr, Fineberg C. An improved mechanical
Ann Thorac Surg 1990;49:101-5
12. 13. 14. 15. 16. 17. 18.
heart and lung apparatus. Med Clin North Am 1953; 3 7 1603-24. Jones RF, Donald DE, Swan HJC, Harshbarger HC, Kirklin JW, Wood EH. Apparatus of the Gibbon type for mechanical bypass of the heart and lungs. Mayo Clin Proc 1955;3010515. Miller BJ. The development of heart lung machines. Surg Gynecol Obstet 1982;154:403-14. Goldman A, Boszormenyi E, Utsu F, Enescu V, Swan HJC, Corday E. Veno-arterial pulsatile partial bypass for circulatory assist. Dis Chest 1966;50:63?40. Lande AJ. Assisted circulation with membrane oxygenators. Transplant Proc 1971;3:1479-89. Bregman D, Casarella WJ. Percutaneous intraaortic balloon pumping: initial clinical experience. Ann Thorac Surg 1979; 29: 153-5. Phillips SJ. Percutaneous cardiopulmonary bypass and innovations in clinical counterpulsation. Crit Care Clin 1986; 2:297-318. Scheidt S, Wilmer G, Mueller H, et al. Intraaortic balloon counterpulsation in cardiogenic shock. N Engl J Med 1973; 288:979-84.
DISCUSSION DR WALTER E. PAE, JR (Hershey, PA): I congratulate Dr Reichman and the group at Sharp Memorial for their fine presentation and thank them for the privilege of reviewing their manuscript before this presentation. At The Pennsylvania State University Hospital in conjunction with our colleagues at the Harrisburg Hospital, using this same CPS system during the last 3 years, we initiated bypass outside the operating room in 15 patients after failure of standard advanced life support measures. The patients’ diagnoses were quite similar to those of the patients of Dr Reichman and colleagues. All 15 patients were stabilized hemodynamically, with mean bypass flows of approximately 4L/min; they had average bypass times of 110 minutes. Similar to the experience of Reichman and associates, 7 of our 15 patients, or about 50%, were ultimately weaned. Only 2 of the 7 patients had cardiac and neurological recovery, however. Nonetheless, there were no survivors when the system was used in patients receiving cardiopulmonary resuscitation longer than 30 or 40 minutes before bypass or in patients who had an unwitnessed cardiac arrest. In our experience, cannulation and therefore rapidity of deployment has been most optimal in the cardiac catheterization laboratory, using a modified Seldinger technique such as Reichman and associates described, in which the femoral vessels were surgically visualized-before puncture and fluoroscopy was available. Cannulation is also easily achieved in the intensive care unit, where the chest of a postoperative cardiac patient can be rapidly reopened and direct cannulation can be achieved. Our early experience using a true Seldinger technique, although successful in achieving cannulation, was more cumbersome and time-consuming in patients without palpable pulses and resulted in some complex vascular injuries. This technology does represent a powerful adjunctive therapy in properly selected patients to allow subsequent diagnostic and therapeutic maneuvers, but unless the uncovered pathology is correctable and the intervention itself does not lead to significant morbidity, meaningful recovery will probably not occur. I would like to direct several questions to Dr Reichman. First, in terms of patient selection, d o you believe that this technology has any application in the treatment of an unwitnessed cardiac
arrest or have you placed a time limit on pre-bypass CPR as a patient selection criterion? Second, in deference to our cardiology colleagues and despite improvements in cannula design, do you believe that direct visualization of the peripheral vessels is important to limit complex vascular injuries and cannula malposition, and to increase the chances of a rapid successful cannulation? Last, do you foresee an expanded role or have unreported experience in use of this particular technology as a n adjunct to complex angioplasty or valvoplasty procedures in carefully, mutually selected high-risk patients? DR WILLIAM B. IAMS (Camp Hill, PA): I am concerned that this technology might be expanded to noncardiac surgeons and be run by nonperfusionists. I know that there is a commercial impetus to do this, but my particular concern is that the patient will be killed by the machine rather than saved. As shown by this series and our series, the salvage rate is small even when CPS is performed by cardiac surgeons and perfusionists. What if the CPS system is placed by less than experienced people and is run by less than experienced perfusionists? It is CPB, and I believe that this issue of credentials will arise in your own hospitals. How many survivors were cannulated from the right atrium to ascending aorta? I believe that 12 of your patients were treated in this way. Could conventional CPB not be instituted in these cases? This technology has expanded with the great interest in providing femorofemoral bypass with excellent cannulas. These cannulas, in my experience, have been very helpful in redo coronary artery bypass patients: I can place a small catheter in the femoral artery and a small catheter in the femoral vein and, if I need to, I can cannulate percutaneously with these cannulas in a matter of minutes if a redo is decompensating and I cannot get the heart free of adhesions. Finally, I believe that this technology is best used in the deteriorating patient but that it should not be instituted in the obviously dead patient. DR HOOSHANG BOLOOKI (Miami, FL): I congratulate Dr Reichman and his associates for a great deal of work. This
Ann Thorac Surg 1990;49:101-5
procedure requires a lot of manpower, but, as you note, provides little reward. So long as CPS equipment is used in the cardiac catheterization laboratory for complications of angioplasty and possibly prophylactically for some of the difficult angioplasties or for left main disease, I believe it will be a helpful system. But when the indication for its use is expanded, I think we will have trouble. At the University of Miami between 1968 and 1971, we used a very simple protocol. We thought that a patient who was doing well a few minutes before cardiac arrest would respond to resuscitation if we could reestablish cardiac output with a pump oxygenator. We had a portable heart-lung machine in the emergency room and used it in select young patients with outof-hospital cardiac arrest if cardiac massage was not effective within ten to 15 minutes. We placed the patient on the heart-lung machine-femorofemoral bypass with oxygenation-and simultaneously opened the left chest to perform open cardiac massage. This we believed mandatory because we discovered that as soon as we stopped the external cardiac massage, the arrested heart dilated, and pulmonary edema, hemorrhage, and hypoxia followed. The protocol called for taking the patient to the operating room, discontinuing cardiac support gradually, and closing the chest. We were also seeking an acute coronary obstruction that we could fix or bypass. We performed such procedures as on-table coronary angiography. Some of the results on this work were previously reported [Circulation 1971;44:103442]. We initially used the portable heart-lung machine in 7 patients. All had had out-of-hospital cardiac arrests. We had 1 survivor; this provided the impetus for us to use the device more often. Unfortunately, in the next year we used it in 7 other patients with in-hospital cardiac arrest. We expanded the indication to patients with suspected pulmonary emboli and patients with myocardial infarction and cardiac arrest, but we had no other survivors despite a great deal of effort. Our final conclusion was that we must use some type of case stratification. Which patient would get the best result? With the small number of survivors so far, we cannot draw any conclusions. We can, however, identify patients who do not respond to this support system. We discovered that patients in the coronary care unit with cardiac arrest after cardiogenic shock were the worst candidates for this method of support. Our 1 patient who eventually did well had no coronary artery disease and had ventricular fibrillation due to cardiomyopathy that had not responded to ten minutes of cardiac massage. Dr Reichman, can you indicate which patient group benefited most from this support system? You had 6 survivors, 3 with postoperative hemorrhage leading to cardiac arrest and 3 with cardiac arrest after coronary bypass. These would be surgical patients in the intensive care unit. Are you thinking of limiting your effort to this group of patients instead of using this device in all patients with cardiac arrest in the hospital setting?
REICHMAN ET AL IMPROVED PATIENT SURVIVAL WITH CPS
105
DR HENDRICK B. BARNER (St Louis, MO): The abstract indicates that nursing personnel were used to initiate CPS. Dr Reichman, could you speak to the issue of using nursing personnel versus perfusionists for managing the CPS system? D R REICHMAN Dr Pae, out initial impression, although we do not have hard data to substantiate this, is that prolonged CPR and unwitnessed arrest are poor predictors of outcome. In our patients, all cardiac arrests were in-hospital arrests. No patient in our series arrived at the emergency room in an arrested state. Furthermore, if restoration of a rhythm was made possible by CPR, use of CPS to stabilize hemodynamics met with better results. We have placed cannulas through percutaneous puncture of the right internal jugular vein, femoral vein, and femoral artery in select cases. I must caution you about this approach. One of our patients had percutaneous femoral cannulation. This patient’s groin was so thick that some of the sideholes in the arterial cannulas were extravascular. This created a lethal situation. Removal of the arterial cannulas is a procedure requiring direct suture repair. We have removed jugular venous cannulas without cutdown. As for use of the procedure in the catheterization laboratory, we have used CPS in 6 patients for elective angioplasty or valvoplasty who were considered nonsurgical candidates. These data are not included in our current report. Three of the patients had valvoplasty, 2 had coronary angioplasty, and 1 patient had combined valvoplasty and angioplasty. We performed cutdown of the femoral vessels on the right side of CPS insertion and cutdown of the left femoral artery for placement of the angioplasty or valvoplasty catheter. In an effort to decrease myocardial oxygen consumption, we cooled each patient’s temperature to 34°C. At the conclusion of the procedure, each vessel was repaired primarily. Each of these catheterization laboratory patients involved a team effort. The cardiac surgeon provided vascular access, conduct of CPB was maintained by the cardiac surgeon and perfusionist, and only the formal balloon procedure was performed by the cardiologist. Dr Bolooki, we have our unit stationed in the intensive care unit on a battery charger. Sterile vascular cutdown and reentry sternotomy trays are stored with the unit. If it is summoned to a location outside the intensive care unit, such as the medical ward or the emergency room, an intensive care unit nurse responds with the unit. Set-up and priming of the unit is her responsibility. Cannulation is performed by the surgeon; we have had trauma surgeons, cardiac surgeons, and cardiologists place the unit. Further management is performed by an on-call perfusionist who is summoned from home. The machine can be placed initially by a trained noncardiac surgeon, and initial conduct of the machine can be performed by an intensive care unit nurse with a perfusionist on call. It is inefficient to have perfusionists in house in each of these three hospitals on a 24-hour basis. Dr lams, of our 6 long-term survivors, 3 were cannulated through the chest, 2 through the femoral artery and vein, and 1 through internal jugular vein and femoral artery.
