Resuscitation, 24 (1992) 61-71 Elsevier Scientific Publishers Ireland Ltd.
61
Use of open chest cardiopulmonary resuscitation after failure of standard closed chest CPR: illustrative cases Norman A. Paradisa, Gerard B. Martinb and Emanuel P. Riversb ‘Department of Emergency Medical Services, Bellevue Hospital Center, New York University Medical Cenier. New York and bEmergency Medicine, Henry Ford Hospital, Detroit, Michigan (USA)
(Received June lst, 1992; accepted July 3rd, 1992)
Compared to standard closed chest CPR, open chest cardiac massage improves vital organ perfusion and survival in animal models of medical cardiac arrest. Yet its use is essentially limited to the treatment of traumatic arrest. Three cases of medical cardiac arrest are presented in which open chest compression was used after failure of external chest compression. These cases illustrate the range of potential outcomes and how this therapy can be optimally applied. Approaches we have used to prevent application of futile intensive therapy in patients unlikely to be neurologically intact survivors are described, Replacement of open chest CPR by closed chest CPR as the standard of care for the in-hospital cardiac arrest was not justified by experimental data. The circumstances of refractory cardiac arrest make it unlikely that well controlled human studies will be able to demonstrate the superiority of open chest CPR in selected patients. The decision to use this therapy will likely remain within the art of medicine. Key words: cardiac arrest; resuscitation; CPR; human; open chest CPR; outcomes; cardiac massage
INTRODUCTION
On June 20th 1898 Doctor Tuftier was called to the bedside of a patient he had operated on for a gangrenous appendix 5 days earlier’. Although the patients’s post-operative course had been excellent and he was in good spirits when Dr Tuftier had spoken with him minutes before, he now was without vital signs. Tuftier immediately suspected pulmonary embolism: We felt then a certain anguish you all understand: it was hard for us to let die of mechanical accident a young man of whom all the rest of the organ systems appeared normal while we may have had a means, for which it is true we had only experimental approval, to bring him back to life. My hesitation was of short duration. I rapidly made an incision in the left third intercostal space and, taking off the pericardium with the index finger, I exercised rhythmic compressions on the ventricular region for one or two minutes: the heart undulated at first irregularly then truly contracted. The pulses returned, the patient took some deep inspirations, his eyes opened wide, his pupils contracted... Correspondence ro: Norman A. Paradis, Department of Emergency Medical Services, Bellevue Hospital Center, First Avenue and 27th Street, New York, NY 10016, USA.
0300-9572/92/%05.00 0 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
62
Although the patient succumbed to a second cardiac arrest, Dr Tuffier had applied what he believed to be the most effective therapy available, despite its experimental nature. Today, open chest cardiac CPR is not experimental. It is the treatment of choice for cardiac arrest after penetrating trauma 2. Studies in animal models demonstrate that it is also more effective than closed chest CPR in the treatment of medical cardiac arrest3*4 yet it is rarely used in such circumstances. Before popularization of external chest compression by Kouwenhoven5, thoracotomy with internal cardiac compression was more widely used6. Recent studies have demonstrated that return of spontaneous circulation after cardiac arrest is primarily a function of myocardial blood flow, the force for which is coronary perfusion pressure 7,8. In most patients, standard external CPR fails to produce a coronary perfusion pressure high enough for the return of spontaneous circulation9~10. In animal models open chest CPR consistently produces greater organ blood flow “Jo and improved outcomes 4~‘3~‘ .4A single human preliminary study reports greatly increased vital organ perfusion pressures with application of the technique3. We have periodically used internal cardiac compression after failure of external chest compression, occasionally with good outcomes. Here we report return of spontaneous circulation in three such patients. These cases do not represent a complete series, but were chosen to illustrate the circumstances under which open chest CPR may result in either good or poor clinical outcomes. CASE 1
A 39-year-old previously healthy male collapsed after intravenous drug use. Emergency personnel found the patient without vital signs. The time from collapse was unknown. Basic life support, consisting of external chest compression and bag-valvemask ventilation, was started and the patient was transported to the hospital. Transport time was approximately 5 min. On arrival at the emergency department the patient was asystolic with fixed and dilated pupils. Orotracheal intubation was performed and external chest compression and ventilation was started by a pneumatic compression device (Thumper, Michigan Instruments). A right subclavian catheter was placed. Four milligrammes of naloxone and 1 mg of epinephrine were administered without effect. An aortic catheter inserted through the femoral artery showed no arterial pulse pressure. An aortic blood gas was obtained-and was later reported as PO, = 152, Pco2 = 37‘and pH = 7.18. An arterial lactate obtained at this time was later reported to be 19 mM/l. End-tidal CO2 was 4 mm Hg. Seven minutes after arrival, a second milligram of epinephrine was administered along with 1 mg of atropine. The patient remained asystolic and these drugs were administered in the same dosage 5 min later without effect. Seventeen minutes after arrival a fourth milligram of epinephrine and sodium bicarbonate (50 mequiv.) were administered. A regular wide complex rhythm developed without associated arterial pulse”. High-dose epinephrine, 0.2 mg/kg, was administered’6*‘7. Over the next 6 min the electrical rhythm narrowed but no arterial pulse developed. Twenty-eight minutes after arrival, 1 mg of epinephrine and 10 mg calcium chloride were administered without effect. End-tidal CO2 was 11 mmHg.
63
Thirty-eight minutes after arrival, a left lateral thoracotomy was performed through the forth intercostal space. Internal cardiac massage was begun approximately 1 min later. On examination through the pericardium the heart was noted to have occasional slow ventricular contractions. Ten mg of epinephrine were administered through the central line and internal massage continued. The ventricular contractions became more frequent and within 2 min the patient had a heart rate of 140 and a blood pressure of 90/50. The end-tidal CO* increased to 28 mm Hg. Within 5 min the blood pressure began to decrease. Despite continuous infusion of epinephrine, sodium bicarbonate, and aortic cross clamping, the blood pressure continued to decline. Eventually bradycardia unresponsive to adrenergic drugs developed and progressed to asystole. A pacemaker failed to capture the myocardium and the patient was pronounced 50 min after arrival. Post mortem examination showed no anatomic abnormalities, serum analysis was positive for heroin and cocaine. CASE2
A 55-year-old female with a history of coronary artery disease called emergency medicine services because of chest pain. Shortly after initiating transport to the hospital the patient had a cardiac arrest. Basic life support was started. Transport time to the hospital was approximately 5 min. On arrival at the emergency department the patient was found to be in ventricular fibrillation. Immediate defibrillation with 200 J then 300 J was performed without success. The patient was orotracheally intubated and external chest compression and ventilation were performed by pneumatic compression device. A right subclavian catheter was placed. Three minutes after arrival, defibrillation with 360 J was attempted without success, and then again 2 min later, also without success. One milligram of epinephrine and 50 mg of lidocaine were administered and defibrillation with 360 J was attempted again without success. Fourteen minutes after arrival a left lateral thoracotomy was performed through the forth intercostal space. The aorta was cross clamped and internal cardiac massage was begun. On examination through the pericardium the heart was noted to be in fine ventricular fibrillation. This became progressively coarse with internal massage and after 2 min internal defibrillation with 20 J was performed resulting in asystole. Epinephrine (1 mg) was administered and ventricular fibrillation developed. Internal defibrillation with 20 J was again performed resulting in asystole. Four minutes after initiation of internal cardiac massage high dose epinephrine (0.2 mg/kg) was administered. Internal cardiac massage was continued and ventricular fibrillation developed approximately 5 min later. Bretylium (500 mg) was administered and internal defibrillation with 50 J was performed resulting in ventricular tachycardia with coordinated mechanical activity. Blood pressure measured from a femoral arterial catheter was 90/70 mm Hg. Total arrest time had been approximately 36 min and the time from thoracotomy to spontaneous hemodynamics had been 17 min. Maintenance of a systolic blood pressure required partial aortic cross clamping and epinephrine infusion. Calcium chloride 1 mg was administered without improvement in hemodynamics. During the next 5 min, the patient developed ventricular fibrillation four times and was suc-
64
cessfully detibrillated each time. A lidocaine infusion of 2 mg/min was begun. Aortic cross clamp was not completely discontinued until 11 min after initial return of spontaneous hemodynamics. During the next 30 min, epinephrine 70 &mm was required to maintain blood pressure above 90 mmHg systolic. Sodium bicarbonate (100 mequiv.) was administered without improvement. Norepinephrine infusion 2 &nin was started and antibiotics were administered. The patient did not become stable enough to go to the operating room for closure of the thoracotomy until more than 2 h after initial return of spontaneous circulation. Vital signs were stable on arrival at the intensive care unit, but the pupils were fixed and dilated and there was no brain stem activity. The initial hospital course was complicated by sepsis and acute renal failure. There were continued complex ventricular dysrhythmias requiring medication. Neurological improvement was slow and incomplete. Three weeks after admission the patient was ventilating spontaneously and responding appropriately to painful stimuli, but cognitive function was severely impaired. She would occasionally speak, but did not consistently recognize her family or respond appropriately to questions. She was unable to sit without assistance. A gastrostomy tube was placed for enteral feeding. Three months after initial presentation the patient was discharged to a chronic care facility. Eight months after her cardiac arrest the anoxic encephalopathy has stabilized with significant deficits in cognitive function including impairments in verbal skills, long and short term memory. CASE 3
A previously healthy 28-year-old male electrical worker came in contact with a high voltage electrical power line and collapsed immediately. CPR was instituted immediately and emergency services personnel arrived within 2 min. Basic life support was continued during the 5-min transport to the hospital. On arrival the patient was without vital signs; a small third degree burn was noted on the left anterior chest. The cardiac monitor showed line ventricular fibrillation, which converted to asystole after a defibrillation with 200 J. The patient was orally intubated and external chest compression and ventilation was performed by a pneumatic compression device. A right subclavian catheter was placed and 1 mg of epinephrine was administered. Five minutes after arrival, pulseless ventricular tachycardia developed and lidocaine 100 mg was administered. Cardioversion with 300 J was performed which resulted in a narrow complex rhythm without palpable pulses. Six minutes after arrival, 5 mg of epinephrine were administered. Ventricular fibrillation developed and defibrillation with 360 J was attempted without success. Bretylium tosylate (500 mg) was administered and defibrillation was again attempted unsuccessfully. Twelve minutes after arrival, a left lateral thoracotomy was performed through the fourth intercostal space. Internal cardiac massage was begun approximately 1 min later. On direct examination the heart was noted to be in ventricular fibrillation. Fifty milliequivalents of sodium bicarbonate, 4 mg of epinephrine and an additional 500 mg of bretylium were administered. Nineteen minutes after arrival, internal defibrillation with 30 J was performed. This resulted in a supraventricular rhythm and coordinated mechanical activity. Total arrest time had been approx-
65
imately 26 min. The blood pressure was 170/110 initially and decreased to 140/80 during the next 10 min, then stabilized. Cefazolin, 2 mg i.v., was administered. Approximately 17 min after return of spontaneous circulation, the patient began to have spontaneous respirations and 30 min later began to move his extremities. Pancuronium (5 mg) and morphine sulfate were administered. One hour after resuscitation the blood pressure had decreased to 103/76 and a dopamine infusion of 10 &kg was started. One hour and thirty minutes after resuscitation the thoracotomy incision was closed and the patient admitted to the surgical intensive care unit. He had an uneventful hospital course and was discharged to a rehabilitation facility 6 weeks after admission. At that time he was able to walk unassisted, speak and write in full sentences. Eight months after discharge he was fully functional with only mild deficits in short and long term memory. DISCUSSION
Open chest cardiac massage as a treatment for cardiac arrest is not a new technique. It was used by Hake more than 100 years ago18. In 1908 Pike commented ‘Often a faithful trial of extrathoracic massage has failed to start the heart. In nearly all of these cases direct massage has afterward proved effective”‘. During the late 1940s and early 1950s Stephenson accumulated 1200 cases of open chest cardiac compression6. Fourteen percent of these arrests occurred outside of the operating room and 17% of this subgroup survived. These results compare favorably to more recent series of in-hospital arrests treated with external chest compression*‘-*‘. With the popularization of external compression by Kouwenhoven’ and the extensive refinement of the technique by Pearson and Redding23,24, the use of open chest cardiac massage declined. Recently, its use has been limited to cardiac arrests resulting from trauma, those occurring in the operating room with the chest already open and in patients arresting shortly after cardiothoracic surgery. Controlled clinical trials comparing open and closed chest CPR were not performed before this change in the standard of care occurred. Recent animal research indicates that this may have been unfortunate. There is not agreement in regard to the mechanism of external compression CPR. It is unclear whether blood flow results from direct myocardial compression, increases in intrathoracic pressure, or a combination of both2s326.It is clear, however, that animal and human studies demonstrate inadequate coronary perfusion pressure during standard closed chest CPR”**’ and that this may underlie the poor prognosis of patients in normothermic cardiac arrest treated with standard external chest compression2s.. A study by Ditchey and associates found myocardial blood flows in the range of 1% of pre-arrest values when epinephrine was not used’. Although use of alpha adrenergic drugs may augment the coronary perfusion pressure29, in most cases it remains below the measured intramyocardial wall pressure during ventricular tibrillation3’. Inadequate myocardial perfusion is most likely responsible for the poor prognosis of patients not responding to advanced cardiac life support (ACLS). With standard closed chest CPR, myocardial hypoxia and acidosis most likely become worse with time despite therapy. Clinical experience confirms this, as the
66
rate of successful resuscitation drops dramatically after the first few minutes31-34 and the number of patients with neurologically intact long term survival is sma112*. Although restoration of spontaneous circulation is the first objective during resuscitation, maintenance of cerebral viability is of central importance. Animal studies show that external CPR provides cerebral blood flows in the range of 15-30% of pre-arrest values7. Blood flows at the low end of this range may be worse than no flow at a113’,because resultant ‘trickle flow’ may provide metabolic substrate out of proportion to oxygen, resulting in greater lactic acidosis. Animal studies not only indicate greater arterial pressure with open chest cardiac compression, but also lower intracranial pressure I2. Since cerebral perfusion pressure during CPR is equal to the arterial minus the intracranial pressure, and internal cardiac massage has a beneficial effect on both parameters, cerebral blood flow is significantly improved. During external chest compression, however, intracranial pressure increases during each compression, resulting in lower cerebral perfusion36. Studies directly comparing open and closed chest CPR have demonstrated the advantages of the open technique. Del Guercio found that, compared to closed chest compression, use of open chest CPR in humans resulted in a doubling of the cardiac index, and a twofold decrease in circulatory time37. In a preliminary report, Howard and associates found a dramatic improvement in coronary perfusion pressure during open chest CPR in humans 3. In a model that closely mimics clinical scenarios, Kern and associates randomized animals failing to defibrillate into a perfusing rhythm after I5 min of cardiac arrest with closed chest CPR into two groups 43’4. One group continued to receive external CPR, the other underwent thoracotomy and open chest cardiac massage. Animals randomized to open chest CPR had higher rates of return of spontaneous circulation and, more importantly, long term survival, Eleven of 14 animals treated with open chest cardiac compression were alive at 7 days, while only four of 14 randomised to closed chest CPR remained alive. Using this same model, they found that this benefit no longer occurred if the animals underwent thoracotomy after more than 20 min of CPR38*39. Time is the single most important variable in the treatment of cardiac arrest3i. Application of standard therapy within minutes of arrest is often effective, but the fraction of patients with return of spontaneous circulation decreases rapidly with time33, approaching 0 after 20 min of cardiac arrest ‘@.Failure to consider time factors may, in part, have been responsible for the rapid widespread adoption of closed chest CPR, since most of the early studies described animal models or in-hospital patients with unrealistically short arrest times5,23. Although thoracotomy would be expected to result in greater morbidity than closed chest compression, this conclusion may not be correct. Numerous visceral injuries have been described after closed chest compression41-43. These sequelae may not have received attention because of the poor overall prognosis of patients after cardiac arrest and the low rate of postmortem examination. Extensive use of thoracotomy for traumatic arrest has demonstrated that it can be performed quickly and without significant complications related to the procedure itself. Despite the inadequate aseptic technique employed in emergency thoracotomy, the rate of infection in successfully resuscitated patients is very 10~~. Closed chest CPR is effective in some patients in the first few minutes after loss
67
of circulation and it should remain the initial therapy of choice. Yet, with each minute of continued arrest the metabolic state of the myocardium is deteriorating, and the chance of neurological recovery decreases. Repetitive cycling through the same series of therapies which have been initially unsuccessful is not logical. Open chest CPR has the potential to resuscitate patients when closed chest CPR has failed. However, this success does not consistently result in long term survival. If standard ACLS provides a window period during which a good outcome is possible, then an alternative, such as open chest cardiac massage, may be considered as a technique which extends this window. Like standard therapy, the limitations of open chest CPR must be understood if it is to be applied appropriately. Studies in animals indicate there is no benefit in initiating internal massage after approximately 20 min of cardiac arrest38. Human studies which have failed to demonstrate improved outcomes with open chest CPR have all performed the technique beyond the time limits suggested by animal models39,45*46. Pilot studies would have demonstrated the inability of these institutions to meet the time limitation and prevented promulgation of misleading conclusions. The nature of sudden death makes prospective randomised trials of alternative therapies difficult. Since only a subset of patients are appropriate candidates, well controlled trials to evaluate open chest CPR may not be possible. Case 2 demonstrates that application of open chest CPR after prolonged arrest times portends neurological injury and dismal prognosis. If this therapy is to occasionally result in good outcomes, clinicians must not hesitate to use it immediately once standard measures have failed. This is a concept that was well understood by Dr Tuffier ‘. Case three represents what we consider to be the typical course of patients with good outcomes. Myocardial ischemia and reperfusion produce a pattern of metabolic derangements which has been termed myocardial stunning4’. This process manifests as cardiac dysfunction after resuscitation. In our experience, patients destined to do well have prompt return of myocardial function, indicating minimal stunning. These patients may even have an initial period of hypertension, probably secondary to circulating catecholamines 48. After this hypertension the need for pressor support is minimal. Refractory hypotension requiring massive pressor support indicates severe myocardial stunning and portends a poor outcome. Since the central nervous system is subjected to the same ischemic insult as the myocardium, yet is more sensitive to the deleterious effects, we have found that significant myocardial dysfunction post-resuscitation is highly correlated with severe central nervous system damage. Additionally, refractory hypotension postresuscitation most likely exacerbates the primary CNS insult. Thus, patients with return of spontaneous circulation after open chest CPR who require prolonged aortic cross clamping or large doses of vasopressors can be expected to succumb within a few hours unless there is preexisting hypothermia or intoxication. Thus, these patients’ poor prognosis is usually apparent soon after resuscitation and decisions relating to resource allocation may consider this. Case 1 demonstrates this important observation: use of open chest cardiac compression does not invariably result in wasting large amounts of resources. This is because many patients destined to expire do so before transport to the operating
room or intensive care facility. Case 2, which is unique in our experience, represents the unfortunate outcome which can occur when such patients are ‘aggressively’ managed despite clear indications that the ischemic insult was severe. It is important to understand, however, that a prolonged pre-hospital period does not mean the patient is unsalvageable. It is difficult to conclusively diagnose complete loss of vital signs in the field, especially during transport. In the more than 400 cardiac arrest patients in whom we have placed aortic catheters, a significant fraction have actually been in severe shock I5. In particular, this must be suspected in patients diagnosed as narrow complex EMD. However, as demonstrated in case one, excessive delay after arrival at the hospital consistently results in poor outcomes. The decision to perform thoracotomy must be made soon after failure of conventional therapy. It should not be used as an end stage technique after failure of prolonged standard therapy. Such an approach is unlikely to result in good outcomes, but holds the possibility of cardiac resuscitation after irreversible brain injury. Studies which assert that the transport of patients who have failed pre-hospital ACLS is futile usually have inherent flaws and inadequate sample sizes49. It seems obvious that hospital studies which apply standard ACLS to patients who have just failed identical therapy in the pre-hospital setting are going to have poor results49. As discussed above, the fear that large amounts of medical resources will be wasted on patients with post-anoxic encephalopathy is, in our experience, unfounded. In most cases, the only resources used are an ambulance trip and a surgical tray. Resuscitated patients that are going to do well can be identified within 24 hours, usually much sooner. Spontaneous ventilations usually begin within 30 min and most patients destined to do well are awake within 12 h. Continuing therapy in persistently comatose patients, as frequently happens, is more a function of poor bioethics than the primitive state of resuscitation. These patients should be allowed to complete the dying process 5o. P h ysicians should not attempt to resolve problems related to the allocation of community resources during the emergency management of sudden death. Such an approach will occasionally have tragic results that are not ultimately mitigated by our ignorance that they have occurred. We have not described our experience with open chest CPR to suggest a basic change in the current standard of care. However, cardiac arrest is a special clinical entity. Failing to use potentially effective therapies when treating a uniformly fatal condition, simply because they have not been demonstrated effective in controlled trials, is problematic. Although it is not widely recognized, most of the standard therapies used in the treatment of cardiac arrest have not been subjected to controlled study5’. Indeed, the change in standard therapy from open to closed chest CPR during the 1960s was without such support. The cases presented are not intended to substitute for controlled studies, but to highlight the inadequacy of current standards. It is hoped they can provide some guidance to clinicians confronted with a patient struck down in the midst of good health and unresponsive to the current standard of care. In summary, open chest CPR provides greater coronary blood flow than standard closed chest compression. If used expeditiously after failure of external compression it can occasionally result in neurologically intact long term survivors. Used appropriately, the technique only rarely results in neurologically injured long term sur-
69
vivors. Discontinuation of aggressive management may be considered when the stigmata of severe ischemic injury are identified. If application of this therapy is to occasionally result in good outcomes, it must be initiated early. ACKNOWLEDGEMENT
N.A. Paradis is supported by a grant from the Aaron Diamond Foundation. REFERENCES
4 5 6 I 8 9 10 11 12 13
14 15
16
17
18 19
Tuffter T, Hallion M. De la compression rythmic du coeur dans la syncope cardiaque par embolic. Bull Mem Sot Chir (Paris) 1898; 24: 937-939. American College of Surgeons, editors. Thoracic trauma. In: Advanced trauma life support course. Chicago, American College of Surgeons, 1989: 93. Howard MA, Labadie LL, Martin GB, Tomlanovich MC, Nowak RM. Improvement in coronary perfusion pressures after open-chest cardiac massage in humans: preliminary report [abstract]. Ann Emerg Med 1986; 15: 664-665. Kern KB, Sanders AB, Badylak SF, et al. Long-term survival with open-chest cardiac massage after closed-chest compression in a canine preparation. Circulation 1987; 75: 498-503. Kouwenhoven WB, Jude JR, Knickerbocker GG. Closed chest cardiac massage. J Am Med Assoc 1960; 173: 1064-1067. Stephenson HE, Corsan Reed L, Hinton JW. Some common denominators in 1200 cases of cardiac arrest. Ann Surg 1953; 137: 731-744. Lute JM, Rizk NA, Niskanen RA. Regional blood flow during cardiopulmonary resuscitation in dogs. Crit Care Med 1984; 12: 874-878. Paradis NA, Martin GB, Rivers EP, et al. Coronary perfusion pressure and return of spontaneous circulation in human cardiopulmonary resuscitation. J Am Med Assoc 1990; 263: 1106-l 113. Ditchey RV, Winkler JV, Rhodes CA. Relative lack of coronary blood flow during closed-chest resuscitation in dogs. Circ 1982; 66: 297-302. Niemann JT, Rosborough JP, Ung S, Criley JM. Coronary perfusion pressure during experimental cardiopulmonary resuscitation. Ann Emerg Med 1982; 11: 127-131. Bircher N, Safar P. Comparison of standard and “new”: closed-chest CPR and open-chest CPR in dogs. Crit Care Med 1981; 9: 384-385. Barnett WM, Alitimoff JK, Paris PM, Stewart RD, Safar P. Comparison of open-chest cardiac massage techniques in dogs. Ann Emerg Med 1986; 15: 408-411. DeBehnke DJ, Angeles MG, Leasure JE. Comparison of standard external CPR, open-chest CPR and cardiopulmonary bypass in a canine myocardial infarct model. Ann Emerg Med 1991; 20: 754-760. Sanders AB, Kern KB, Ewy GA, Atlas M, Bailey L. Improved resuscitation from cardiac arrest with open-chest massage. Ann Emerg Med 1984; 13: 672-675. Paradis NA, Martin GB, Goetting MG, Rivers EP, Feingold M, Nowak RM. Aortic pressure during human cardiac arrest. Identification of pseudo-electromechanical dissociation. Chest 1992; 101: 123-128. Paradis HA, Goetting MG, Rivers EP, Martin GB, Appleton TJ, Nowak RM. High-dose epineph rine therapy and return of spontaneous circulation during human pseudoelectromechanical dissociation. Ann Emerg Med 1990; 19: 491. Paradis NA, Martin GB, Rivers EP, Rosenberg J, Appleton TJ, Nowak RM. The effect of standardand high-dose epinephrine on coronary perfusion pressure during prolonged cardiopulmonary resuscitation. J Am Med Assoc 1991; 2650: 1139-l 144. Hake TG. Studies on ether and chloroform from Professor Schifl’s physiological laboratory. Practitioner 1874; 12: 241. Pike FH, Guthrie CC, Stewart GN. Studies in resuscitation. I. The general conditions affecting resuscitation and resuscitation of blood and the heart. J Exp Med 1908; 10: 371.
70 20 21 22 23 24 25
26
27
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
Bedell SE, Delbanco TL, Cook EF, Epstein FH. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983; 309: 569-576. Rozenbaum EA, Shenkman L. Predicting outcome of inhospital cardiopulmonary resuscitation. Crit Care Med 1988; 16: 583-586. Kyff J, Puri VK, Raheja R, Ireland T. Cardiopulmonary resuscitation in hospitalized patients: continuing problems of decision-making. Crit Care Med 1987; 15: 41-43. Redding JS, Pearson JW. Resuscitation from asphyxia. J Am Med Assoc 1962; 182: 283-286. Redding JS, Pearson JW. Evaluation of drugs for cardiac resuscitation. Anesthesia 1963; 24: 203-207. Weisfeldt ML, Chandra N, Tsitlik J. Increased intrathoracic pressure - not direct heart compression - causes the rise in intrathoracic vascular pressures during CPR in dogs and pigs. Crit Care Med 1981; 9: 377-378. Deshmukh HG, Weil MH, Gudipati CV, Trevino RP, Bisera J, Rackow EC. Mechanism of blood flow generated by precordial compression during CPR. I. Studies on closed chest precordial compression. Chest 1989; 95: 1092-1099. Martin GB, Carden DL, Nowak RM, Lewinter JR, Johnston W, Tomlanovich MC. Aortic and right atria1 pressures during standard and simulateous compression and ventilation CPR in human beings. Ann Emerg Med 1986; IS: 125-130. Becker LB, Ostrander MP, Barrett J, Kondos CT. Outcome of CPR in a large metropolitan area - where are the survivors? [see comments]. Ann Emerg Med 1991 ; 20: 355-361. Koehler RC, Michael JR, Guerci AD, et al. Beneficial effect of epinephrine infusion on cerebral and myocardial blood flows during CPR. Ann Emerg Med 1985; 14: 744-749. Downey JM, Chagrasulis RW, Hemphill V. Quantitative study of intramyocardial compression in the fibrillating heart. Am J Physiol 1979; 237: H191-H196. Pionkowski RS, Thompson BM, Gruchow HW, Aprahamian C, Darin JC. Resuscitation time in ventricular fibrillation - a prognostic indicator. Ann Emerg Med 1983; 12: 733-738. Roth R, Stewart RD, Rogers K, Cannon GM. Out of hospital cardiac arrest: factors associated with survival. Ann Emerg Med 1984; 13: 237-243. Cummins RO, Eisenberg MS, Hallstrom AP, Litwin PE. Survival of out-of-hospital cardiac arrest with early initiation of cardiopulmonary resuscitation. Am J Emerg Med 1985; 3: 114-l 19. Hargarten KM, Stueven HA, Waite EM, et al. Prehospital experience with defibrillation of coarse ventricular fibrillation: a ten-year review. Ann Emerg Med 1990; 19: 157-162. Rehncrona S, Rosen I, Siesjo BK. Excessive cellular acidosis: an important mechanism of neuronal damage in the brain. Acta Physiol Stand 1980; 110: 435-437. Rogers MC, Nugent SK, Stidham CL. Effects of closed-chest cardiac massage on intracranial pressure. Crit Care Med 1979; 7: 454-456. Del Guercio LRM, Feins NR, Cohn JD, Coomaraswamy RP, Wollman SB, State D. Comparison of blood flow during external and internal cardiac massage in man. Circ 1965; Suppl 1: I 171-1180. Kern KB, Sanders AB, Janas W, et al. Limitations of open-chest cardiac massage after prolonged, untreated cardiac arrest in dogs. Ann Emerg Med 1991; 20: 761-767. Sanders AB, Kern KB, Ewy GA. Time limitations for open-chest cardiopulmonary resuscitation from cardiac arrest. Crit Care Med 1985; 13: 897-898. Abramson NS;Safar P, Detre KM, et al. Neurologic recovery after cardiac arrest: Effect of duration of &hernia. Crit Care Med 1985; 13: 930-931. Powner DJ, Holcombe PA, Mello LA. Cardiopulmonary resuscitation-related injuries. Crit Care Med 1984; 12: 54-55. Kern KB, Carter AB, Showen RL, et al. CPR-induced trauma: comparison of three manual methods in an experimental model. Ann Emerg Med 1986; 15: 674-679. Bedell SE, F&on EJ. Unexpected findings and complications at autopsy after cardiopulmonary resuscitation (CPR). Arch Intern Med 1986; 146: 1725-1728. Ahermeier WA, Todd J. Studies on the incidence of infection following open chest cardiac massage for cardiac arrest. Ann Surg 1963; 158: 596-607. Geehr EC, Lewis FR, Auerbach PS. Failure of open-heart massage to improve survival after prehospital nontraumatic cardiac arrest [letter]. N Engl J Med 1986; 314: 1189-I 190. Sterner S, Brunette D, Ruiz E, Blake D, Croston K, Sharkey S. Open-chest verses closed-chest
71 cardiac massage in resuscitation of nontraumatic refractory cardiac arrest [abstract]. Ann Emerg Med 1991; 10: 949. 47 Kusuoka H, Porterfield JK, Weisman HF, Weisfeldt ML, Marban E. Pathophysiology and pathogenesis of stunned myocardium. Depressed Ca*+ activation of contraction as a consequence of reperfusion-induced cellular calcium overload in ferret hearts. J Clin Invest 1987; 79: 950-961. 48 Wortsman J, Frank S, Cryer PE. Adrenomedullary response to maximal stress in humans. Am J Med 1984; 77: 779-784. 49 Gray WA, Capone RJ, Most AS. Unsuccessful emergency medical resuscitation - are continued efforts in the emergency department justified? N Engl J Med 1991; 325: 1393-1398. 50 Tomlinson T, Brody H. Futility and the ethics of resuscitation. J Am Med Assoc 1990; 261: 1276-1280. 51 Paradis NA, Koscove EM. Epinephrine in cardiac arrest: a critical review. Ann Emerg Med 1990; 19: 1288-1301.
61
Use of open chest cardiopulmonary resuscitation after failure of standard closed chest CPR: illustrative cases Norman A. Paradisa, Gerard B. Martinb and Emanuel P. Riversb ‘Department of Emergency Medical Services, Bellevue Hospital Center, New York University Medical Cenier. New York and bEmergency Medicine, Henry Ford Hospital, Detroit, Michigan (USA)
(Received June lst, 1992; accepted July 3rd, 1992)
Compared to standard closed chest CPR, open chest cardiac massage improves vital organ perfusion and survival in animal models of medical cardiac arrest. Yet its use is essentially limited to the treatment of traumatic arrest. Three cases of medical cardiac arrest are presented in which open chest compression was used after failure of external chest compression. These cases illustrate the range of potential outcomes and how this therapy can be optimally applied. Approaches we have used to prevent application of futile intensive therapy in patients unlikely to be neurologically intact survivors are described, Replacement of open chest CPR by closed chest CPR as the standard of care for the in-hospital cardiac arrest was not justified by experimental data. The circumstances of refractory cardiac arrest make it unlikely that well controlled human studies will be able to demonstrate the superiority of open chest CPR in selected patients. The decision to use this therapy will likely remain within the art of medicine. Key words: cardiac arrest; resuscitation; CPR; human; open chest CPR; outcomes; cardiac massage
INTRODUCTION
On June 20th 1898 Doctor Tuftier was called to the bedside of a patient he had operated on for a gangrenous appendix 5 days earlier’. Although the patients’s post-operative course had been excellent and he was in good spirits when Dr Tuftier had spoken with him minutes before, he now was without vital signs. Tuftier immediately suspected pulmonary embolism: We felt then a certain anguish you all understand: it was hard for us to let die of mechanical accident a young man of whom all the rest of the organ systems appeared normal while we may have had a means, for which it is true we had only experimental approval, to bring him back to life. My hesitation was of short duration. I rapidly made an incision in the left third intercostal space and, taking off the pericardium with the index finger, I exercised rhythmic compressions on the ventricular region for one or two minutes: the heart undulated at first irregularly then truly contracted. The pulses returned, the patient took some deep inspirations, his eyes opened wide, his pupils contracted... Correspondence ro: Norman A. Paradis, Department of Emergency Medical Services, Bellevue Hospital Center, First Avenue and 27th Street, New York, NY 10016, USA.
0300-9572/92/%05.00 0 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
62
Although the patient succumbed to a second cardiac arrest, Dr Tuffier had applied what he believed to be the most effective therapy available, despite its experimental nature. Today, open chest cardiac CPR is not experimental. It is the treatment of choice for cardiac arrest after penetrating trauma 2. Studies in animal models demonstrate that it is also more effective than closed chest CPR in the treatment of medical cardiac arrest3*4 yet it is rarely used in such circumstances. Before popularization of external chest compression by Kouwenhoven5, thoracotomy with internal cardiac compression was more widely used6. Recent studies have demonstrated that return of spontaneous circulation after cardiac arrest is primarily a function of myocardial blood flow, the force for which is coronary perfusion pressure 7,8. In most patients, standard external CPR fails to produce a coronary perfusion pressure high enough for the return of spontaneous circulation9~10. In animal models open chest CPR consistently produces greater organ blood flow “Jo and improved outcomes 4~‘3~‘ .4A single human preliminary study reports greatly increased vital organ perfusion pressures with application of the technique3. We have periodically used internal cardiac compression after failure of external chest compression, occasionally with good outcomes. Here we report return of spontaneous circulation in three such patients. These cases do not represent a complete series, but were chosen to illustrate the circumstances under which open chest CPR may result in either good or poor clinical outcomes. CASE 1
A 39-year-old previously healthy male collapsed after intravenous drug use. Emergency personnel found the patient without vital signs. The time from collapse was unknown. Basic life support, consisting of external chest compression and bag-valvemask ventilation, was started and the patient was transported to the hospital. Transport time was approximately 5 min. On arrival at the emergency department the patient was asystolic with fixed and dilated pupils. Orotracheal intubation was performed and external chest compression and ventilation was started by a pneumatic compression device (Thumper, Michigan Instruments). A right subclavian catheter was placed. Four milligrammes of naloxone and 1 mg of epinephrine were administered without effect. An aortic catheter inserted through the femoral artery showed no arterial pulse pressure. An aortic blood gas was obtained-and was later reported as PO, = 152, Pco2 = 37‘and pH = 7.18. An arterial lactate obtained at this time was later reported to be 19 mM/l. End-tidal CO2 was 4 mm Hg. Seven minutes after arrival, a second milligram of epinephrine was administered along with 1 mg of atropine. The patient remained asystolic and these drugs were administered in the same dosage 5 min later without effect. Seventeen minutes after arrival a fourth milligram of epinephrine and sodium bicarbonate (50 mequiv.) were administered. A regular wide complex rhythm developed without associated arterial pulse”. High-dose epinephrine, 0.2 mg/kg, was administered’6*‘7. Over the next 6 min the electrical rhythm narrowed but no arterial pulse developed. Twenty-eight minutes after arrival, 1 mg of epinephrine and 10 mg calcium chloride were administered without effect. End-tidal CO2 was 11 mmHg.
63
Thirty-eight minutes after arrival, a left lateral thoracotomy was performed through the forth intercostal space. Internal cardiac massage was begun approximately 1 min later. On examination through the pericardium the heart was noted to have occasional slow ventricular contractions. Ten mg of epinephrine were administered through the central line and internal massage continued. The ventricular contractions became more frequent and within 2 min the patient had a heart rate of 140 and a blood pressure of 90/50. The end-tidal CO* increased to 28 mm Hg. Within 5 min the blood pressure began to decrease. Despite continuous infusion of epinephrine, sodium bicarbonate, and aortic cross clamping, the blood pressure continued to decline. Eventually bradycardia unresponsive to adrenergic drugs developed and progressed to asystole. A pacemaker failed to capture the myocardium and the patient was pronounced 50 min after arrival. Post mortem examination showed no anatomic abnormalities, serum analysis was positive for heroin and cocaine. CASE2
A 55-year-old female with a history of coronary artery disease called emergency medicine services because of chest pain. Shortly after initiating transport to the hospital the patient had a cardiac arrest. Basic life support was started. Transport time to the hospital was approximately 5 min. On arrival at the emergency department the patient was found to be in ventricular fibrillation. Immediate defibrillation with 200 J then 300 J was performed without success. The patient was orotracheally intubated and external chest compression and ventilation were performed by pneumatic compression device. A right subclavian catheter was placed. Three minutes after arrival, defibrillation with 360 J was attempted without success, and then again 2 min later, also without success. One milligram of epinephrine and 50 mg of lidocaine were administered and defibrillation with 360 J was attempted again without success. Fourteen minutes after arrival a left lateral thoracotomy was performed through the forth intercostal space. The aorta was cross clamped and internal cardiac massage was begun. On examination through the pericardium the heart was noted to be in fine ventricular fibrillation. This became progressively coarse with internal massage and after 2 min internal defibrillation with 20 J was performed resulting in asystole. Epinephrine (1 mg) was administered and ventricular fibrillation developed. Internal defibrillation with 20 J was again performed resulting in asystole. Four minutes after initiation of internal cardiac massage high dose epinephrine (0.2 mg/kg) was administered. Internal cardiac massage was continued and ventricular fibrillation developed approximately 5 min later. Bretylium (500 mg) was administered and internal defibrillation with 50 J was performed resulting in ventricular tachycardia with coordinated mechanical activity. Blood pressure measured from a femoral arterial catheter was 90/70 mm Hg. Total arrest time had been approximately 36 min and the time from thoracotomy to spontaneous hemodynamics had been 17 min. Maintenance of a systolic blood pressure required partial aortic cross clamping and epinephrine infusion. Calcium chloride 1 mg was administered without improvement in hemodynamics. During the next 5 min, the patient developed ventricular fibrillation four times and was suc-
64
cessfully detibrillated each time. A lidocaine infusion of 2 mg/min was begun. Aortic cross clamp was not completely discontinued until 11 min after initial return of spontaneous hemodynamics. During the next 30 min, epinephrine 70 &mm was required to maintain blood pressure above 90 mmHg systolic. Sodium bicarbonate (100 mequiv.) was administered without improvement. Norepinephrine infusion 2 &nin was started and antibiotics were administered. The patient did not become stable enough to go to the operating room for closure of the thoracotomy until more than 2 h after initial return of spontaneous circulation. Vital signs were stable on arrival at the intensive care unit, but the pupils were fixed and dilated and there was no brain stem activity. The initial hospital course was complicated by sepsis and acute renal failure. There were continued complex ventricular dysrhythmias requiring medication. Neurological improvement was slow and incomplete. Three weeks after admission the patient was ventilating spontaneously and responding appropriately to painful stimuli, but cognitive function was severely impaired. She would occasionally speak, but did not consistently recognize her family or respond appropriately to questions. She was unable to sit without assistance. A gastrostomy tube was placed for enteral feeding. Three months after initial presentation the patient was discharged to a chronic care facility. Eight months after her cardiac arrest the anoxic encephalopathy has stabilized with significant deficits in cognitive function including impairments in verbal skills, long and short term memory. CASE 3
A previously healthy 28-year-old male electrical worker came in contact with a high voltage electrical power line and collapsed immediately. CPR was instituted immediately and emergency services personnel arrived within 2 min. Basic life support was continued during the 5-min transport to the hospital. On arrival the patient was without vital signs; a small third degree burn was noted on the left anterior chest. The cardiac monitor showed line ventricular fibrillation, which converted to asystole after a defibrillation with 200 J. The patient was orally intubated and external chest compression and ventilation was performed by a pneumatic compression device. A right subclavian catheter was placed and 1 mg of epinephrine was administered. Five minutes after arrival, pulseless ventricular tachycardia developed and lidocaine 100 mg was administered. Cardioversion with 300 J was performed which resulted in a narrow complex rhythm without palpable pulses. Six minutes after arrival, 5 mg of epinephrine were administered. Ventricular fibrillation developed and defibrillation with 360 J was attempted without success. Bretylium tosylate (500 mg) was administered and defibrillation was again attempted unsuccessfully. Twelve minutes after arrival, a left lateral thoracotomy was performed through the fourth intercostal space. Internal cardiac massage was begun approximately 1 min later. On direct examination the heart was noted to be in ventricular fibrillation. Fifty milliequivalents of sodium bicarbonate, 4 mg of epinephrine and an additional 500 mg of bretylium were administered. Nineteen minutes after arrival, internal defibrillation with 30 J was performed. This resulted in a supraventricular rhythm and coordinated mechanical activity. Total arrest time had been approx-
65
imately 26 min. The blood pressure was 170/110 initially and decreased to 140/80 during the next 10 min, then stabilized. Cefazolin, 2 mg i.v., was administered. Approximately 17 min after return of spontaneous circulation, the patient began to have spontaneous respirations and 30 min later began to move his extremities. Pancuronium (5 mg) and morphine sulfate were administered. One hour after resuscitation the blood pressure had decreased to 103/76 and a dopamine infusion of 10 &kg was started. One hour and thirty minutes after resuscitation the thoracotomy incision was closed and the patient admitted to the surgical intensive care unit. He had an uneventful hospital course and was discharged to a rehabilitation facility 6 weeks after admission. At that time he was able to walk unassisted, speak and write in full sentences. Eight months after discharge he was fully functional with only mild deficits in short and long term memory. DISCUSSION
Open chest cardiac massage as a treatment for cardiac arrest is not a new technique. It was used by Hake more than 100 years ago18. In 1908 Pike commented ‘Often a faithful trial of extrathoracic massage has failed to start the heart. In nearly all of these cases direct massage has afterward proved effective”‘. During the late 1940s and early 1950s Stephenson accumulated 1200 cases of open chest cardiac compression6. Fourteen percent of these arrests occurred outside of the operating room and 17% of this subgroup survived. These results compare favorably to more recent series of in-hospital arrests treated with external chest compression*‘-*‘. With the popularization of external compression by Kouwenhoven’ and the extensive refinement of the technique by Pearson and Redding23,24, the use of open chest cardiac massage declined. Recently, its use has been limited to cardiac arrests resulting from trauma, those occurring in the operating room with the chest already open and in patients arresting shortly after cardiothoracic surgery. Controlled clinical trials comparing open and closed chest CPR were not performed before this change in the standard of care occurred. Recent animal research indicates that this may have been unfortunate. There is not agreement in regard to the mechanism of external compression CPR. It is unclear whether blood flow results from direct myocardial compression, increases in intrathoracic pressure, or a combination of both2s326.It is clear, however, that animal and human studies demonstrate inadequate coronary perfusion pressure during standard closed chest CPR”**’ and that this may underlie the poor prognosis of patients in normothermic cardiac arrest treated with standard external chest compression2s.. A study by Ditchey and associates found myocardial blood flows in the range of 1% of pre-arrest values when epinephrine was not used’. Although use of alpha adrenergic drugs may augment the coronary perfusion pressure29, in most cases it remains below the measured intramyocardial wall pressure during ventricular tibrillation3’. Inadequate myocardial perfusion is most likely responsible for the poor prognosis of patients not responding to advanced cardiac life support (ACLS). With standard closed chest CPR, myocardial hypoxia and acidosis most likely become worse with time despite therapy. Clinical experience confirms this, as the
66
rate of successful resuscitation drops dramatically after the first few minutes31-34 and the number of patients with neurologically intact long term survival is sma112*. Although restoration of spontaneous circulation is the first objective during resuscitation, maintenance of cerebral viability is of central importance. Animal studies show that external CPR provides cerebral blood flows in the range of 15-30% of pre-arrest values7. Blood flows at the low end of this range may be worse than no flow at a113’,because resultant ‘trickle flow’ may provide metabolic substrate out of proportion to oxygen, resulting in greater lactic acidosis. Animal studies not only indicate greater arterial pressure with open chest cardiac compression, but also lower intracranial pressure I2. Since cerebral perfusion pressure during CPR is equal to the arterial minus the intracranial pressure, and internal cardiac massage has a beneficial effect on both parameters, cerebral blood flow is significantly improved. During external chest compression, however, intracranial pressure increases during each compression, resulting in lower cerebral perfusion36. Studies directly comparing open and closed chest CPR have demonstrated the advantages of the open technique. Del Guercio found that, compared to closed chest compression, use of open chest CPR in humans resulted in a doubling of the cardiac index, and a twofold decrease in circulatory time37. In a preliminary report, Howard and associates found a dramatic improvement in coronary perfusion pressure during open chest CPR in humans 3. In a model that closely mimics clinical scenarios, Kern and associates randomized animals failing to defibrillate into a perfusing rhythm after I5 min of cardiac arrest with closed chest CPR into two groups 43’4. One group continued to receive external CPR, the other underwent thoracotomy and open chest cardiac massage. Animals randomized to open chest CPR had higher rates of return of spontaneous circulation and, more importantly, long term survival, Eleven of 14 animals treated with open chest cardiac compression were alive at 7 days, while only four of 14 randomised to closed chest CPR remained alive. Using this same model, they found that this benefit no longer occurred if the animals underwent thoracotomy after more than 20 min of CPR38*39. Time is the single most important variable in the treatment of cardiac arrest3i. Application of standard therapy within minutes of arrest is often effective, but the fraction of patients with return of spontaneous circulation decreases rapidly with time33, approaching 0 after 20 min of cardiac arrest ‘@.Failure to consider time factors may, in part, have been responsible for the rapid widespread adoption of closed chest CPR, since most of the early studies described animal models or in-hospital patients with unrealistically short arrest times5,23. Although thoracotomy would be expected to result in greater morbidity than closed chest compression, this conclusion may not be correct. Numerous visceral injuries have been described after closed chest compression41-43. These sequelae may not have received attention because of the poor overall prognosis of patients after cardiac arrest and the low rate of postmortem examination. Extensive use of thoracotomy for traumatic arrest has demonstrated that it can be performed quickly and without significant complications related to the procedure itself. Despite the inadequate aseptic technique employed in emergency thoracotomy, the rate of infection in successfully resuscitated patients is very 10~~. Closed chest CPR is effective in some patients in the first few minutes after loss
67
of circulation and it should remain the initial therapy of choice. Yet, with each minute of continued arrest the metabolic state of the myocardium is deteriorating, and the chance of neurological recovery decreases. Repetitive cycling through the same series of therapies which have been initially unsuccessful is not logical. Open chest CPR has the potential to resuscitate patients when closed chest CPR has failed. However, this success does not consistently result in long term survival. If standard ACLS provides a window period during which a good outcome is possible, then an alternative, such as open chest cardiac massage, may be considered as a technique which extends this window. Like standard therapy, the limitations of open chest CPR must be understood if it is to be applied appropriately. Studies in animals indicate there is no benefit in initiating internal massage after approximately 20 min of cardiac arrest38. Human studies which have failed to demonstrate improved outcomes with open chest CPR have all performed the technique beyond the time limits suggested by animal models39,45*46. Pilot studies would have demonstrated the inability of these institutions to meet the time limitation and prevented promulgation of misleading conclusions. The nature of sudden death makes prospective randomised trials of alternative therapies difficult. Since only a subset of patients are appropriate candidates, well controlled trials to evaluate open chest CPR may not be possible. Case 2 demonstrates that application of open chest CPR after prolonged arrest times portends neurological injury and dismal prognosis. If this therapy is to occasionally result in good outcomes, clinicians must not hesitate to use it immediately once standard measures have failed. This is a concept that was well understood by Dr Tuffier ‘. Case three represents what we consider to be the typical course of patients with good outcomes. Myocardial ischemia and reperfusion produce a pattern of metabolic derangements which has been termed myocardial stunning4’. This process manifests as cardiac dysfunction after resuscitation. In our experience, patients destined to do well have prompt return of myocardial function, indicating minimal stunning. These patients may even have an initial period of hypertension, probably secondary to circulating catecholamines 48. After this hypertension the need for pressor support is minimal. Refractory hypotension requiring massive pressor support indicates severe myocardial stunning and portends a poor outcome. Since the central nervous system is subjected to the same ischemic insult as the myocardium, yet is more sensitive to the deleterious effects, we have found that significant myocardial dysfunction post-resuscitation is highly correlated with severe central nervous system damage. Additionally, refractory hypotension postresuscitation most likely exacerbates the primary CNS insult. Thus, patients with return of spontaneous circulation after open chest CPR who require prolonged aortic cross clamping or large doses of vasopressors can be expected to succumb within a few hours unless there is preexisting hypothermia or intoxication. Thus, these patients’ poor prognosis is usually apparent soon after resuscitation and decisions relating to resource allocation may consider this. Case 1 demonstrates this important observation: use of open chest cardiac compression does not invariably result in wasting large amounts of resources. This is because many patients destined to expire do so before transport to the operating
room or intensive care facility. Case 2, which is unique in our experience, represents the unfortunate outcome which can occur when such patients are ‘aggressively’ managed despite clear indications that the ischemic insult was severe. It is important to understand, however, that a prolonged pre-hospital period does not mean the patient is unsalvageable. It is difficult to conclusively diagnose complete loss of vital signs in the field, especially during transport. In the more than 400 cardiac arrest patients in whom we have placed aortic catheters, a significant fraction have actually been in severe shock I5. In particular, this must be suspected in patients diagnosed as narrow complex EMD. However, as demonstrated in case one, excessive delay after arrival at the hospital consistently results in poor outcomes. The decision to perform thoracotomy must be made soon after failure of conventional therapy. It should not be used as an end stage technique after failure of prolonged standard therapy. Such an approach is unlikely to result in good outcomes, but holds the possibility of cardiac resuscitation after irreversible brain injury. Studies which assert that the transport of patients who have failed pre-hospital ACLS is futile usually have inherent flaws and inadequate sample sizes49. It seems obvious that hospital studies which apply standard ACLS to patients who have just failed identical therapy in the pre-hospital setting are going to have poor results49. As discussed above, the fear that large amounts of medical resources will be wasted on patients with post-anoxic encephalopathy is, in our experience, unfounded. In most cases, the only resources used are an ambulance trip and a surgical tray. Resuscitated patients that are going to do well can be identified within 24 hours, usually much sooner. Spontaneous ventilations usually begin within 30 min and most patients destined to do well are awake within 12 h. Continuing therapy in persistently comatose patients, as frequently happens, is more a function of poor bioethics than the primitive state of resuscitation. These patients should be allowed to complete the dying process 5o. P h ysicians should not attempt to resolve problems related to the allocation of community resources during the emergency management of sudden death. Such an approach will occasionally have tragic results that are not ultimately mitigated by our ignorance that they have occurred. We have not described our experience with open chest CPR to suggest a basic change in the current standard of care. However, cardiac arrest is a special clinical entity. Failing to use potentially effective therapies when treating a uniformly fatal condition, simply because they have not been demonstrated effective in controlled trials, is problematic. Although it is not widely recognized, most of the standard therapies used in the treatment of cardiac arrest have not been subjected to controlled study5’. Indeed, the change in standard therapy from open to closed chest CPR during the 1960s was without such support. The cases presented are not intended to substitute for controlled studies, but to highlight the inadequacy of current standards. It is hoped they can provide some guidance to clinicians confronted with a patient struck down in the midst of good health and unresponsive to the current standard of care. In summary, open chest CPR provides greater coronary blood flow than standard closed chest compression. If used expeditiously after failure of external compression it can occasionally result in neurologically intact long term survivors. Used appropriately, the technique only rarely results in neurologically injured long term sur-
69
vivors. Discontinuation of aggressive management may be considered when the stigmata of severe ischemic injury are identified. If application of this therapy is to occasionally result in good outcomes, it must be initiated early. ACKNOWLEDGEMENT
N.A. Paradis is supported by a grant from the Aaron Diamond Foundation. REFERENCES
4 5 6 I 8 9 10 11 12 13
14 15
16
17
18 19
Tuffter T, Hallion M. De la compression rythmic du coeur dans la syncope cardiaque par embolic. Bull Mem Sot Chir (Paris) 1898; 24: 937-939. American College of Surgeons, editors. Thoracic trauma. In: Advanced trauma life support course. Chicago, American College of Surgeons, 1989: 93. Howard MA, Labadie LL, Martin GB, Tomlanovich MC, Nowak RM. Improvement in coronary perfusion pressures after open-chest cardiac massage in humans: preliminary report [abstract]. Ann Emerg Med 1986; 15: 664-665. Kern KB, Sanders AB, Badylak SF, et al. Long-term survival with open-chest cardiac massage after closed-chest compression in a canine preparation. Circulation 1987; 75: 498-503. Kouwenhoven WB, Jude JR, Knickerbocker GG. Closed chest cardiac massage. J Am Med Assoc 1960; 173: 1064-1067. Stephenson HE, Corsan Reed L, Hinton JW. Some common denominators in 1200 cases of cardiac arrest. Ann Surg 1953; 137: 731-744. Lute JM, Rizk NA, Niskanen RA. Regional blood flow during cardiopulmonary resuscitation in dogs. Crit Care Med 1984; 12: 874-878. Paradis NA, Martin GB, Rivers EP, et al. Coronary perfusion pressure and return of spontaneous circulation in human cardiopulmonary resuscitation. J Am Med Assoc 1990; 263: 1106-l 113. Ditchey RV, Winkler JV, Rhodes CA. Relative lack of coronary blood flow during closed-chest resuscitation in dogs. Circ 1982; 66: 297-302. Niemann JT, Rosborough JP, Ung S, Criley JM. Coronary perfusion pressure during experimental cardiopulmonary resuscitation. Ann Emerg Med 1982; 11: 127-131. Bircher N, Safar P. Comparison of standard and “new”: closed-chest CPR and open-chest CPR in dogs. Crit Care Med 1981; 9: 384-385. Barnett WM, Alitimoff JK, Paris PM, Stewart RD, Safar P. Comparison of open-chest cardiac massage techniques in dogs. Ann Emerg Med 1986; 15: 408-411. DeBehnke DJ, Angeles MG, Leasure JE. Comparison of standard external CPR, open-chest CPR and cardiopulmonary bypass in a canine myocardial infarct model. Ann Emerg Med 1991; 20: 754-760. Sanders AB, Kern KB, Ewy GA, Atlas M, Bailey L. Improved resuscitation from cardiac arrest with open-chest massage. Ann Emerg Med 1984; 13: 672-675. Paradis NA, Martin GB, Goetting MG, Rivers EP, Feingold M, Nowak RM. Aortic pressure during human cardiac arrest. Identification of pseudo-electromechanical dissociation. Chest 1992; 101: 123-128. Paradis HA, Goetting MG, Rivers EP, Martin GB, Appleton TJ, Nowak RM. High-dose epineph rine therapy and return of spontaneous circulation during human pseudoelectromechanical dissociation. Ann Emerg Med 1990; 19: 491. Paradis NA, Martin GB, Rivers EP, Rosenberg J, Appleton TJ, Nowak RM. The effect of standardand high-dose epinephrine on coronary perfusion pressure during prolonged cardiopulmonary resuscitation. J Am Med Assoc 1991; 2650: 1139-l 144. Hake TG. Studies on ether and chloroform from Professor Schifl’s physiological laboratory. Practitioner 1874; 12: 241. Pike FH, Guthrie CC, Stewart GN. Studies in resuscitation. I. The general conditions affecting resuscitation and resuscitation of blood and the heart. J Exp Med 1908; 10: 371.
70 20 21 22 23 24 25
26
27
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
Bedell SE, Delbanco TL, Cook EF, Epstein FH. Survival after cardiopulmonary resuscitation in the hospital. N Engl J Med 1983; 309: 569-576. Rozenbaum EA, Shenkman L. Predicting outcome of inhospital cardiopulmonary resuscitation. Crit Care Med 1988; 16: 583-586. Kyff J, Puri VK, Raheja R, Ireland T. Cardiopulmonary resuscitation in hospitalized patients: continuing problems of decision-making. Crit Care Med 1987; 15: 41-43. Redding JS, Pearson JW. Resuscitation from asphyxia. J Am Med Assoc 1962; 182: 283-286. Redding JS, Pearson JW. Evaluation of drugs for cardiac resuscitation. Anesthesia 1963; 24: 203-207. Weisfeldt ML, Chandra N, Tsitlik J. Increased intrathoracic pressure - not direct heart compression - causes the rise in intrathoracic vascular pressures during CPR in dogs and pigs. Crit Care Med 1981; 9: 377-378. Deshmukh HG, Weil MH, Gudipati CV, Trevino RP, Bisera J, Rackow EC. Mechanism of blood flow generated by precordial compression during CPR. I. Studies on closed chest precordial compression. Chest 1989; 95: 1092-1099. Martin GB, Carden DL, Nowak RM, Lewinter JR, Johnston W, Tomlanovich MC. Aortic and right atria1 pressures during standard and simulateous compression and ventilation CPR in human beings. Ann Emerg Med 1986; IS: 125-130. Becker LB, Ostrander MP, Barrett J, Kondos CT. Outcome of CPR in a large metropolitan area - where are the survivors? [see comments]. Ann Emerg Med 1991 ; 20: 355-361. Koehler RC, Michael JR, Guerci AD, et al. Beneficial effect of epinephrine infusion on cerebral and myocardial blood flows during CPR. Ann Emerg Med 1985; 14: 744-749. Downey JM, Chagrasulis RW, Hemphill V. Quantitative study of intramyocardial compression in the fibrillating heart. Am J Physiol 1979; 237: H191-H196. Pionkowski RS, Thompson BM, Gruchow HW, Aprahamian C, Darin JC. Resuscitation time in ventricular fibrillation - a prognostic indicator. Ann Emerg Med 1983; 12: 733-738. Roth R, Stewart RD, Rogers K, Cannon GM. Out of hospital cardiac arrest: factors associated with survival. Ann Emerg Med 1984; 13: 237-243. Cummins RO, Eisenberg MS, Hallstrom AP, Litwin PE. Survival of out-of-hospital cardiac arrest with early initiation of cardiopulmonary resuscitation. Am J Emerg Med 1985; 3: 114-l 19. Hargarten KM, Stueven HA, Waite EM, et al. Prehospital experience with defibrillation of coarse ventricular fibrillation: a ten-year review. Ann Emerg Med 1990; 19: 157-162. Rehncrona S, Rosen I, Siesjo BK. Excessive cellular acidosis: an important mechanism of neuronal damage in the brain. Acta Physiol Stand 1980; 110: 435-437. Rogers MC, Nugent SK, Stidham CL. Effects of closed-chest cardiac massage on intracranial pressure. Crit Care Med 1979; 7: 454-456. Del Guercio LRM, Feins NR, Cohn JD, Coomaraswamy RP, Wollman SB, State D. Comparison of blood flow during external and internal cardiac massage in man. Circ 1965; Suppl 1: I 171-1180. Kern KB, Sanders AB, Janas W, et al. Limitations of open-chest cardiac massage after prolonged, untreated cardiac arrest in dogs. Ann Emerg Med 1991; 20: 761-767. Sanders AB, Kern KB, Ewy GA. Time limitations for open-chest cardiopulmonary resuscitation from cardiac arrest. Crit Care Med 1985; 13: 897-898. Abramson NS;Safar P, Detre KM, et al. Neurologic recovery after cardiac arrest: Effect of duration of &hernia. Crit Care Med 1985; 13: 930-931. Powner DJ, Holcombe PA, Mello LA. Cardiopulmonary resuscitation-related injuries. Crit Care Med 1984; 12: 54-55. Kern KB, Carter AB, Showen RL, et al. CPR-induced trauma: comparison of three manual methods in an experimental model. Ann Emerg Med 1986; 15: 674-679. Bedell SE, F&on EJ. Unexpected findings and complications at autopsy after cardiopulmonary resuscitation (CPR). Arch Intern Med 1986; 146: 1725-1728. Ahermeier WA, Todd J. Studies on the incidence of infection following open chest cardiac massage for cardiac arrest. Ann Surg 1963; 158: 596-607. Geehr EC, Lewis FR, Auerbach PS. Failure of open-heart massage to improve survival after prehospital nontraumatic cardiac arrest [letter]. N Engl J Med 1986; 314: 1189-I 190. Sterner S, Brunette D, Ruiz E, Blake D, Croston K, Sharkey S. Open-chest verses closed-chest
71 cardiac massage in resuscitation of nontraumatic refractory cardiac arrest [abstract]. Ann Emerg Med 1991; 10: 949. 47 Kusuoka H, Porterfield JK, Weisman HF, Weisfeldt ML, Marban E. Pathophysiology and pathogenesis of stunned myocardium. Depressed Ca*+ activation of contraction as a consequence of reperfusion-induced cellular calcium overload in ferret hearts. J Clin Invest 1987; 79: 950-961. 48 Wortsman J, Frank S, Cryer PE. Adrenomedullary response to maximal stress in humans. Am J Med 1984; 77: 779-784. 49 Gray WA, Capone RJ, Most AS. Unsuccessful emergency medical resuscitation - are continued efforts in the emergency department justified? N Engl J Med 1991; 325: 1393-1398. 50 Tomlinson T, Brody H. Futility and the ethics of resuscitation. J Am Med Assoc 1990; 261: 1276-1280. 51 Paradis NA, Koscove EM. Epinephrine in cardiac arrest: a critical review. Ann Emerg Med 1990; 19: 1288-1301.
