- Email: [email protected]
Is a Pressor Necessary During Aortic Perfusion and Oxygenation Therapy of Cardiac Arrest?
LABORATORY INVESTIGATION
Is a Pressor Necessary During Aortic Perfusion and Oxygenation Therapy of Cardiac Arrest?
From the Division of Emergency Medicine, University of Colorado Health Sciences Center, Denver, CO, and The Department of Medicine, Columbia University College of Physicians and Surgeons, New York, NY.
Norman A Paradis, MD
Received for publication August 10, 1998. Revision received August 18, 1999. Accepted for publication September 5, 1999. Presented in part at the American Heart Associations 66th Scientific Sessions, New Orleans, LA, November 1993. Supported in part by a Development Grant from Reperfusion Systems Inc (Boston, MA). New York University and the principal investigator hold US Patent No. 5,334,142 covering the Selective Aortic Perfusion and Oxygen System. Address for reprints: Norman A Paradis, MD, University of Colorado Health Sciences Center–Division of Emergency Medicine, 4200 E. 9th Avenue, B215, Denver, CO 80262. Copyright © 1999 by the American College of Emergency Physicians. 0196-0644/99/$8.00 + 0 47/1/102811
Study objective: Occlusion of the descending aorta and infusion of oxygenated ultrapurified polymerized bovine hemoglobin may improve the efficacy of advanced cardiac life support (ACLS). Because selective aortic perfusion and oxygenation (SAPO) directly increases coronary perfusion pressure, exogenous epinephrine may not be required. The purpose of this study was to determine whether exogenous epinephrine is necessary during SAPO by comparing the rate of return of spontaneous circulation and aortic and coronary perfusion pressures during ACLS-SAPO in animals treated with either intra-aortic epinephrine or saline solution. Methods: A prospective, randomized, interventional beforeafter trial with a canine model of ventricular fibrillation cardiac arrest and ACLS based on external chest compression was performed. The ECG, right atrial, aortic arch, and esophageal pulse pressures were measured continuously. A descending aortic occlusion balloon catheter was placed through the femoral artery. Ventricular fibrillation was induced, and no therapy was given during the 10-minute arrest time. Basic life support was then initiated and normalized by standardization of esophageal pulse pressure and central aortic blood gases. After 3 minutes of basic life support, the aortic occlusion balloon was inflated, and 0.01 mg/kg epinephrine or saline solution was administered through the aortic catheter followed by 450 mL of ultrapurified polymerized bovine hemoglobin over 2 minutes. Defibrillation was then attempted. The outcomes and changes in intravascular pressures were compared. Results: Aortic pressures were higher during infusions in animals treated with epinephrine. During infusion, the mean aortic relaxation pressure increased by 58±5 mm Hg in animals that had received epinephrine versus 20±11 mm Hg in those that had received saline placebo. The coronary perfusion pressure during infusion increased by 52±8 mm Hg in animals that had received epinephrine versus 26±10 mm Hg in those that had
DECEMBER 1999
34:6
ANNALS OF EMERGENCY MEDICINE
6 9 7
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
received saline. Only 2 of 7 animals in the placebo group had return of spontaneous circulation versus 7 of 8 in the epinephrine group. Conclusion: The addition of epinephrine to ACLS-SAPO increases vital organ perfusion pressures and improves outcome from cardiac arrest. There appears to be a profound loss of arterial vasomotor tone after prolonged arrest. This loss of vasomotor tone may make exogenous pressors necessary for resuscitation after prolonged cardiac arrest. [Paradis NA: Is a pressor necessary during aortic perfusion and oxygenation therapy of cardiac arrest? Ann Emerg Med December 1999;34:697-702.] INTRODUCTION
Perfusion of the proximal aorta with oxygen carrying stroma-free hemoglobin may be an effective adjunct to standard advanced cardiac life support (ACLS) in canine models of cardiac arrest. This selective aortic perfusion and oxygenation (SAPO) increases CPR aortic relaxation and coronary perfusion pressures1 compared with standard external chest compression2,3 and improves the fraction of animals with return of spontaneous circulation after prolonged cardiac arrest. Vasopressors, such as epinephrine, are administered during cardiopulmonary resuscitation (CPR) to raise arterial vasomotor tone and augment perfusion pressure,4,5 particularly CPR aortic relaxation pressure, the principal determinant of coronary perfusion and return of spontaneous circulation.2,6 The presence of an aortic catheter allows administration of adrenergic agonists directly into their site of action in the arterial circulation. However, it is not known whether ACLS-SAPO, which increases perfusion pressure directly by volume expansion of the central arterial vascular compartment, requires augmented vasomotor tone to be effective. This study was performed to determine whether adjunctive vasopressor therapy is required during ACLS-SAPO. M AT E R I A L S A N D M E T H O D S
The protocol was approved by our institutional animal care and use committee and was in full compliance with US Public Health Service Policy on Humane Care and Use of Laboratory Animals. Ultrapurified polymerized bovine hemoglobin (UPBH; Biopure, Boston, MA) was chosen as an oxygen-
6 9 8
carrying material because of its potential utility in emergency settings, where cross-matching may not be possible and the cost of alternative materials may be prohibitive. UPBH was prepared from treated bovine RBCs by lysis, filtration, chromatography, and polymerization with glutaraldehyde.7 The sterile, pyrogen-free hemoglobin solution had a concentration of 13 g/dL and an oxygen halfsaturation pressure (P-50) of 21 mm Hg, a methemoglobin level of less than 10%, a mean colloid oncotic pressure of 20 mm Hg, and a molecular weight distribution ranging from 65,000 to 500,000 d. To minimize variability in epinephrine dosage, the dose administered to each animal was prepared by dilution from a common batch, which had been prepared by combining epinephrine 1:1,000 from 3 different pharmaceutical sources. Animals were weighed the day before the experiment, and drugs were administered on a milligram per kilogram basis. UPBH was preoxygenated to a PO2 of 100 to 125 mm Hg by passage through a countercurrent membrane oxygenator. Experiments were performed on colony-bred mongrel dogs, weighing 20 to 34 kg, that had negative test results for heartworm. Animals were assigned by block randomization to receive either 0.01 mg/kg epinephrine HCl or normal saline solution placebo doses. The person administering CPR was blinded as to which agent each animal had received. Animals underwent anesthetic induction with the ultrashort-acting thiopental sodium (Biotal 20 mg/kg). This was followed by a loading dose of α-chloralose (80 mg/kg). The depth of anesthesia was adjusted to a level sufficient to allow controlled ventilation. Animals were placed on a V-bracket in a dorsal recumbent position, and their limbs were secured to prevent lateral displacement during chest compression. They were intubated and ventilated with positive pressure. The fraction of inspired oxygen and minute ventilation were adjusted to achieve a normal arterial blood gas level. A standard lead II ECG was monitored throughout the experiment. Micromanometer-tipped catheters (Millar, Houston, TX) were placed in the aortic arch and right atrium through femoral artery and external jugular vein cutdowns, respectively. A custom-designed 13.5-F occluding-perfusing balloon catheter (Datascope, Oakland, NJ) was placed in the contralateral femoral artery by a combination of surgical cutdown and open guidewire techniques. A custom-designed esophageal balloon micromanometry catheter (Cook, Bloomington, IN) was placed in the esophagus at the level of the left atrium and inflated with saline solution to a pressure of 25 mm Hg. This was
ANNALS OF EMERGENCY MEDICINE
34:6 DECEMBER 1999
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
used to standardize the force of chest compression, preventing interanimal and intergroup differences. Catheter positions were confirmed by fluoroscopy and appropriateness of pressure waveforms and at postmortem examination. The pressure catheters, along with the ECG, were connected to a multichannel amplifier-recorder (Grass Instruments, Boston, MA) and recorded throughout the experiment. Amplifier output was also digitized (Dash16), displayed, and stored by using a microcomputer. Only the esophageal pulse pressure was displayed in the visual field of the person performing chest compressions. A No. 4 bipolar pacing wire was placed in the right ventricle through the jugular vein. Pancuronium, 3 mg, was administered just before induction of cardiac arrest to prevent ventilatory movement.8 Ventricular fibrillation was induced by briefly passing a 60-Hz alternating current at 25 mA through the right ventricular pacing wire. When fibrillation was confirmed by ECG and loss of arterial pressure, ventilation was discontinued. During arrest, the balloon occlusion catheter was inserted through the femoral artery introducer and positioned in the proximal descending aorta. This catheter has a large central infusion port that was connected to 450 mL of UPBH that had been preoxygenated to a PO2 of 90 to 150 mm Hg and pressurized to 300 mm Hg. The positions of all catheters were confirmed by fluoroscopy and appropriateness of waveforms and at postmortem examination. Animals remained in cardiac arrest without therapy for 10 minutes, after which standard basic life support (BLS) measures were initiated.9 This consisted of chest compressions at the rate of 80 per minute, a compression/ relaxation ratio of 0.5, and ventilation with 100% oxygen at a rate of 16 breaths/min. Manual chest compression was used because it may more accurately mimic clinical cardiac arrest than mechanical devices. The depth of compression was adjusted to achieve an aortic pulse pressure of 50 mm Hg.3 This pulse pressure was chosen because it mimics those seen in human beings during standard ACLS.3-10 Previous studies have shown that this results in an outcome that tends to mimic that seen in clinical trials in which outcomes for standard therapy are uniformly worse than the results of laboratory investigations.11 Once the predetermined pulse pressure was achieved, the person performing chest compression maintained an esophageal pressure waveform of consistent magnitude and shape. This esophageal system for standardization of chest compression has previously been
DECEMBER 1999
34:6
ANNALS OF EMERGENCY MEDICINE
described.1 Arterial blood gas analysis was performed 90 seconds after initiation of therapy, and ventilation parameters were adjusted accordingly. Thirteen minutes after initiation of arrest, the aortic balloon was inflated with 5 mL of radiopaque contrast material, and the study drug, epinephrine or saline solution, was administered as a rapid bolus into the proximal aorta through the occluding-perfusing catheter. The persons providing resuscitative therapy were blinded as to whether saline solution or epinephrine was administered. UPBH infusion was begun 13.5 minutes after arrest. Four hundred fifty milliliters of UPBH was infused as rapidly as possible, with a maximal length of infusion of approximately 1 minute. At the end of the infusion, defibrillation was attempted with 200 J and then 300 J. After countershock, the balloon was immediately deflated, and ECG rhythm and pulse were checked. If the animal remained in cardiac arrest after initial countershocks, then ACLS continued. Additional epinephrine was not given. Abnormalities in arterial blood gas parameters were treated by adjustment of minute ventilation and without use of sodium bicarbonate.12 Therapy was discontinued after 17 minutes if the arterial blood gas level was within normal limits. Animals remaining in cardiac arrest at that time were considered to have failed resuscitation. Resuscitated animals were stabilized after arrest by ventilatory and pressor support and administration of bicarbonate and crystalloid as needed on the basis of arterial blood gas values, vital signs, and central venous pressures, respectively. During the immediate postresuscitation phase, the aortic balloon was partially inflated as needed to treat hypotension refractory to pressors. An intravenous epinephrine infusion maintained the mean arterial blood pressure above 90 mm Hg. Minute ventilation was adjusted to maintain the PCO2 below 30 mm Hg. Sodium bicarbonate was administered, 1 mEq for each unit of base excess below –10, if the pH dropped below 7.1. Animals remaining alive 60 minutes after resuscitation were killed by injection of potassium chloride and additional barbiturate. All animals underwent postmortem examination to confirm catheter location, competence of the aortic valve, and absence of air emboli and to examine internal organs for gross injury. Complete occlusion by the aortic balloon was confirmed at postmortem examination. Definitions of the CPR phases and pressures have previously been described.3 Unless otherwise noted, all pressures are from the relaxation phase of CPR. Prearrest hemodynamics and arterial blood gases were measured
6 9 9
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
just before the initiation of ventricular fibrillation. BLS intravascular pressures were measured just before inflation of the obstructing balloon. ACLS-SAPO intravascular pressures were the mean of pressures measured at 5second intervals from the onset of occlusion until the end of infusion. The maximal aortic, coronary perfusion, and right atrial pressures were defined as the highest pressures measured during ACLS-SAPO and did not necessarily occur simultaneously. The change in intravascular pressures was defined as the maximal pressure during infusion minus the BLS pressure. The endpoint variables were prospectively identified as the change in aortic relaxation and coronary perfusion pressure after administration of the epinephrine or saline solution and the proportion of animals in each group with return of spontaneous circulation. Procedure failures, such as inability to place a catheter, untoward events during preparation or performance of the protocol, were prospectively listed and defined as exclusion criteria. Survival to 1 hour was not identified as part of the protocol. However, these data are reported as well. Data are reported as means±SD. All pressures are in millimeters of mercury. Normally distributed interval data were compared by using 2-sided 2-sample t tests with Bonferroni correction for multiple testing. Interval data that were not normally distributed were compared by using the Mann-Whitney rank-sum test. The Fisher exact test was used to compare proportions. The 95% confidence interval for the difference in the means is also reported for normally distributed data. Statistical significance was prospectively set at a P value of less than .05. R E S U LT S
Eighteen animals with a mean weight of 21±3 kg were studied. Three animals were excluded because of procedure failure, 2 from the epinephrine group and 1 from the Table 1.
Vital signs during spontaneous circulation before induction of cardiac arrest.
Vital Signs
Saline Solution Epinephrine
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beats/min)
152±10 131±7 132±34
165±18 131±9 137±28
95% CI, 95% confidence interval for the difference in the means.
7 0 0
95% CI
P Value
–31.7 to 5.7 –10.2 to 9.8 –46 to 34.7
.15 .97 .76
saline solution group. There were no significant differences in blood pressure, heart rate, and arterial blood gas values before initiation of ventricular fibrillation (Table 1). There were no significant differences in arterial blood gas measurements during spontaneous circulation before induction of cardiac arrest, during BLS, and just after return of spontaneous circulation (ROSC). There were no significant differences in the aortic, right atrial, and coronary perfusion pressures before initiation of SAPO (Table 2). During SAPO, the aortic and coronary perfusion pressures were all greater in animals that received epinephrine (Table 2). The difference in aortic relaxation and coronary perfusion pressures became statistically significant within 15 seconds after administration of epinephrine and remained significant until the end of the infusion (Figures 1 and 2). The change in right atrial pressure was also greater in animals that received epinephrine, and there was a trend toward greater maximal right atrial pressures during infusion in animals that received epinephrine (Table 2). The maximal aortic and coronary perfusion pressures occurred approximately halfway through infusion in animals receiving epinephrine, whereas animals treated only with saline solution had pressures that gradually increased throughout infusion. Right atrial pressures gradually increased throughout infusion in both groups. Two of seven animals in the saline solution group had ROSC compared with 7 of 8 in the epinephrine group Table 2.
Mean aortic, right atrial, and coronary perfusion pressures during BLS and representative mean aortic, maximal right atrial, and change in coronary perfusion pressures during ACLS-SAPO. Saline Solution
Pressures BLS-aortic (torr) BLS-RA (torr) BLS-CPP (torr) Aortic (torr) Maximal aortic (torr) Change aortic (torr) RA (torr) Maximal RA (torr) Change RA (torr) CPP (torr) Maximal CPP (torr) Change CPP (torr)
19±1 22±7 –3±7 31±10 39±12 20±11 25±7 30±7 8±5 7±12 23±6 26±10
Epinephrine 19±5 21±8 –3±11 67±3 76±4 58±5 30±9 39±10 18±5 36±11 49±10 52±8
95% CI
P Value
–5 to 4 –9 to 10 –12 to 11 –45 to –27 –49 to –26 –49 to –27 –15 to 5 –19 to 2 –16 to –3 –44 to –14 –36 to –15 –37 to –14
.8 .8 .95 <.001 <.001 <.001 .3 .1 .008 .001 <.001 <.001
95% CI, 95% confidence interval for the difference in the means; RA, right atrial pressure; CPP, coronary perfusion pressure. Unless otherwise indicated, data are from the relaxation phase of CPR.
ANNALS OF EMERGENCY MEDICINE
34:6 DECEMBER 1999
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
(P=.04). Two epinephrine-treated animals died of refractory hypotension before 1 hour. Postmortem examination showed all catheters to be properly located. Organ injuries, such as pulmonary atelectasis and hemorrhage and epicardial hemorrhage, were consistent with the duration of arrest and administration of external chest compression. No injuries attributable specifically to SAPO were noted. DISCUSSION
External chest compression induces a pressure differential between the arterial and central venous circulations, resulting in perfusion of the myocardium.3 However, after the first minutes of cardiac arrest, standard external chest compression requires administration of exogenous vasoactive agents to achieve perfusion pressures adequate for ROSC.13 Administration of exogenous adrenergic agonists can, however, negatively affect the oxygen supply/demand equilibrium.13,14 These data indicate that an exogenous pressor may be required even when techniques such as aortic perfusion are used. Aortic infusions have been shown to be effective in restoring spontaneous circulation when combined with aortic occlusion and external chest compression.1 Techniques such as ACLS-SAPO directly augment aortic pressure by increasing the volume of fluid within the central arterial compartment.1 If arterial compliance remains static, this increased volume results in improved perfusion of the heart and brain. During cardiac arrest and Figure 1.
Aortic relaxation phase pressures (in millimeters of mercury, mean±SD) for animals treated with epinephrine (circles) or saline solution (triangles). The vertical arrow indicates the time of drug administration.
DECEMBER 1999
34:6
ANNALS OF EMERGENCY MEDICINE
CPR, endogenously derived plasma epinephrine levels are higher than in any other pathophysiologic state,15 making it reasonable to hypothesize that exogenous epinephrine might be unnecessary during ACLS-SAPO. However, the present study indicates that after prolonged cardiac arrest, vasomotor tone has decreased so dramatically that even techniques such as ACLS-SAPO may require exogenous vasopressors. Animals not treated with epinephrine had only a moderate increase in aortic pressure, reflecting high arterial compliance and loss of infused UPBH from the central arterial compartment. The trend toward increased right atrial pressures in animals treated with epinephrine may reflect improved arterial to venous flow. Animals treated with epinephrine had a dramatic increase in aortic relaxation phase pressures during the first 30 seconds of infusion. Animals that did not receive epinephrine had a gradual rise in aortic pressure, which continued throughout the infusion (Figure 1), a pattern that seems characteristic of arterial volume infusion without vasopressor-mediated changes in aortic compliance. This study may provide indirect indication that there is a profound loss of vasomotor tone with continued cardiac arrest. After 13 minutes of cardiac arrest, arterial vasomotor tone had decayed sufficiently that even direct infusion into an occluded aorta was unable to raise perfusion pres-
Figure 2.
Coronary perfusion pressures (in millimeters of mercury, mean±SD) for animals treated with epinephrine (circles) or saline solution (triangles). The vertical arrow indicates the time of drug administration.
7 0 1
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
sure to levels adequate for ROSC. The vasopressor effect from endogenous epinephrine was therapeutically inadequate after so long an arrest. The intra-aortic route for drug administration was chosen because it has previously been shown to be more effective than the intravenous route16,17 and may prevent potential toxicity resulting from epinephrine’s passage through the pulmonary circulation.18 SAPO may prolong the venous phase of intravenously administered drug, resulting in increased metabolism before reaching the site of action. Comparison of the pressor response curves indicates that, even with fluid infusion directly into an occluded aorta, epinephrine was responsible for most of the increase in aortic pressure.17 The curve of the pressor response also provides insight into the pharmacokinetics of intra-aortic epinephrine during ACLS-SAPO therapy. The time to peak effect is only about 30 seconds, and the pressor effect has begun to subside by 1 minute. This study has a number of limitations. Increased rates of ROSC and short-term survival in animal models only indicate a therapy’s potential effect on the outcome of patients undergoing cardiac arrest. Although ROSC and short-term survival are widely accepted as surrogate variables, survival studies need to be performed before concluding that these results apply to long-term outcome. Potentially, the effect of pressors on ROSC might be offset by delayed organ toxicity. The 1-hour survival data would seem to indicate this because 2 animals were lost to refractory hypotension. However, because this portion of the study was not intended to be controlled, these data are hypothesis generating at best. In animal studies, investigators must choose a specific duration of insult, whereas human cardiac arrest is clinically variable in presentation. The results of the present study indicate the effect of epinephrine on ROSC only at this defined insult. Early in arrest, CPR alone might be adequate to achieve ROSC, whereas later, higher epinephrine dosages might be needed.
3. Paradis NA, Martin GB, Goetting MG, et al: Simultaneous aortic, jugular bulb, and right atrial pressures during cardiopulmonary resuscitation in humans. Insights into mechanisms. Circulation 1989;80:361-368. 4. Pearson JW, Redding JS: Peripheral vascular tone in cardiac resuscitation. Anesth Analg 1965;44:746-752. 5. Michael JR, Guerci AD, Koehler RC, et al: Mechanisms by which epinephrine augments cerebral and myocardial perfusion during cardiopulmonary resuscitation in dogs. Circulation 1984;69:822-835. 6. Niemann JT, Rosborough JP, Ung S, et al: Coronary perfusion pressure during experimental cardiopulmonary resuscitation. Ann Emerg Med 1982;11:127-131. 7. Sehgal LR, Gould SA, Rosen AL, et al: Polymerized pyridoxylated hemoglobin: A red cell substitute with normal oxygen capacity. Surgery 1984;95:433-438. 8. Niemann JT, Rosborough JP, Niskanen RA, et al: Mechanical “cough”: Cardiopulmonary resuscitation during cardiac arrest in dogs. Am J Cardiol 1985;55:199-204. 9. Standards and guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). JAMA 1986;255:2841-3044. 10. Rivers EP, Lozon J, Enriquez E, et al: Simultaneous radial, femoral, and aortic arterial pressures during human cardiopulmonary resuscitation. Crit Care Med 1993;21:878-883. 11. Eisenberg MS, Horwood BT, Cummins RO, et al: Cardiac arrest and resuscitation: A tale of 29 cities. Ann Emerg Med 1990;19:179-186. 12. Guidelines for cardiopulmonary resuscitation and emergency cardiac care. JAMA 1992;268:2171-2299. 13. Paradis NA, Koscove EM: Epinephrine in cardiac arrest: A critical review. Ann Emerg Med 1990;19:1288-1301. 14. Ditchey RV, Lindenfeld J: Failure of epinephrine to improve the balance between myocardial oxygen supply and demand during closed-chest resuscitation in dogs. Circulation 1988;78:382389. 15. Wortsman J, Frank S, Cryer PE: Adrenomedullary response to maximal stress in humans. Am J Med 1984;77:779-784. 16. Manning JE, Batson DN, Murphy CA, et al: Selective aortic arch perfusion with oxygenated fluorocarbons combined with aortic arch epinephrine [abstract]. Ann Emerg Med 1993;22:929930. 17. Manning JE, Murphy CA Jr, Batson DN, et al: Aortic arch versus central venous epinephrine during CPR. Ann Emerg Med 1993;22:703-708. 18. Tang WC, Weil MH, Gazmuri RJ, et al: Pulmonary ventilation/perfusion defects induced by epinephrine during cardiopulmonary resuscitation. Circulation 1991;84:2101-2107.
I thank Mr Christopher Davison, without whose assistance this research could not have been performed, and Mryon L Weisfeldt, MD, for reviewing the manuscript.
REFERENCES 1. Paradis NA, Rose MI, Gawryl M: Selective aortic perfusion and oxygenation: An effective adjunct to external chest compression-based cardiopulmonary resuscitation. J Am Coll Cardiol 1994;23:497-504. 2. Paradis NA, Martin GB, Rivers EP, et al: Coronary perfusion pressure and return of spontaneous circulation in human cardiopulmonary resuscitation. JAMA 1990;263:1106-1113.
7 0 2
ANNALS OF EMERGENCY MEDICINE
34:6 DECEMBER 1999
Is a Pressor Necessary During Aortic Perfusion and Oxygenation Therapy of Cardiac Arrest?
From the Division of Emergency Medicine, University of Colorado Health Sciences Center, Denver, CO, and The Department of Medicine, Columbia University College of Physicians and Surgeons, New York, NY.
Norman A Paradis, MD
Received for publication August 10, 1998. Revision received August 18, 1999. Accepted for publication September 5, 1999. Presented in part at the American Heart Associations 66th Scientific Sessions, New Orleans, LA, November 1993. Supported in part by a Development Grant from Reperfusion Systems Inc (Boston, MA). New York University and the principal investigator hold US Patent No. 5,334,142 covering the Selective Aortic Perfusion and Oxygen System. Address for reprints: Norman A Paradis, MD, University of Colorado Health Sciences Center–Division of Emergency Medicine, 4200 E. 9th Avenue, B215, Denver, CO 80262. Copyright © 1999 by the American College of Emergency Physicians. 0196-0644/99/$8.00 + 0 47/1/102811
Study objective: Occlusion of the descending aorta and infusion of oxygenated ultrapurified polymerized bovine hemoglobin may improve the efficacy of advanced cardiac life support (ACLS). Because selective aortic perfusion and oxygenation (SAPO) directly increases coronary perfusion pressure, exogenous epinephrine may not be required. The purpose of this study was to determine whether exogenous epinephrine is necessary during SAPO by comparing the rate of return of spontaneous circulation and aortic and coronary perfusion pressures during ACLS-SAPO in animals treated with either intra-aortic epinephrine or saline solution. Methods: A prospective, randomized, interventional beforeafter trial with a canine model of ventricular fibrillation cardiac arrest and ACLS based on external chest compression was performed. The ECG, right atrial, aortic arch, and esophageal pulse pressures were measured continuously. A descending aortic occlusion balloon catheter was placed through the femoral artery. Ventricular fibrillation was induced, and no therapy was given during the 10-minute arrest time. Basic life support was then initiated and normalized by standardization of esophageal pulse pressure and central aortic blood gases. After 3 minutes of basic life support, the aortic occlusion balloon was inflated, and 0.01 mg/kg epinephrine or saline solution was administered through the aortic catheter followed by 450 mL of ultrapurified polymerized bovine hemoglobin over 2 minutes. Defibrillation was then attempted. The outcomes and changes in intravascular pressures were compared. Results: Aortic pressures were higher during infusions in animals treated with epinephrine. During infusion, the mean aortic relaxation pressure increased by 58±5 mm Hg in animals that had received epinephrine versus 20±11 mm Hg in those that had received saline placebo. The coronary perfusion pressure during infusion increased by 52±8 mm Hg in animals that had received epinephrine versus 26±10 mm Hg in those that had
DECEMBER 1999
34:6
ANNALS OF EMERGENCY MEDICINE
6 9 7
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
received saline. Only 2 of 7 animals in the placebo group had return of spontaneous circulation versus 7 of 8 in the epinephrine group. Conclusion: The addition of epinephrine to ACLS-SAPO increases vital organ perfusion pressures and improves outcome from cardiac arrest. There appears to be a profound loss of arterial vasomotor tone after prolonged arrest. This loss of vasomotor tone may make exogenous pressors necessary for resuscitation after prolonged cardiac arrest. [Paradis NA: Is a pressor necessary during aortic perfusion and oxygenation therapy of cardiac arrest? Ann Emerg Med December 1999;34:697-702.] INTRODUCTION
Perfusion of the proximal aorta with oxygen carrying stroma-free hemoglobin may be an effective adjunct to standard advanced cardiac life support (ACLS) in canine models of cardiac arrest. This selective aortic perfusion and oxygenation (SAPO) increases CPR aortic relaxation and coronary perfusion pressures1 compared with standard external chest compression2,3 and improves the fraction of animals with return of spontaneous circulation after prolonged cardiac arrest. Vasopressors, such as epinephrine, are administered during cardiopulmonary resuscitation (CPR) to raise arterial vasomotor tone and augment perfusion pressure,4,5 particularly CPR aortic relaxation pressure, the principal determinant of coronary perfusion and return of spontaneous circulation.2,6 The presence of an aortic catheter allows administration of adrenergic agonists directly into their site of action in the arterial circulation. However, it is not known whether ACLS-SAPO, which increases perfusion pressure directly by volume expansion of the central arterial vascular compartment, requires augmented vasomotor tone to be effective. This study was performed to determine whether adjunctive vasopressor therapy is required during ACLS-SAPO. M AT E R I A L S A N D M E T H O D S
The protocol was approved by our institutional animal care and use committee and was in full compliance with US Public Health Service Policy on Humane Care and Use of Laboratory Animals. Ultrapurified polymerized bovine hemoglobin (UPBH; Biopure, Boston, MA) was chosen as an oxygen-
6 9 8
carrying material because of its potential utility in emergency settings, where cross-matching may not be possible and the cost of alternative materials may be prohibitive. UPBH was prepared from treated bovine RBCs by lysis, filtration, chromatography, and polymerization with glutaraldehyde.7 The sterile, pyrogen-free hemoglobin solution had a concentration of 13 g/dL and an oxygen halfsaturation pressure (P-50) of 21 mm Hg, a methemoglobin level of less than 10%, a mean colloid oncotic pressure of 20 mm Hg, and a molecular weight distribution ranging from 65,000 to 500,000 d. To minimize variability in epinephrine dosage, the dose administered to each animal was prepared by dilution from a common batch, which had been prepared by combining epinephrine 1:1,000 from 3 different pharmaceutical sources. Animals were weighed the day before the experiment, and drugs were administered on a milligram per kilogram basis. UPBH was preoxygenated to a PO2 of 100 to 125 mm Hg by passage through a countercurrent membrane oxygenator. Experiments were performed on colony-bred mongrel dogs, weighing 20 to 34 kg, that had negative test results for heartworm. Animals were assigned by block randomization to receive either 0.01 mg/kg epinephrine HCl or normal saline solution placebo doses. The person administering CPR was blinded as to which agent each animal had received. Animals underwent anesthetic induction with the ultrashort-acting thiopental sodium (Biotal 20 mg/kg). This was followed by a loading dose of α-chloralose (80 mg/kg). The depth of anesthesia was adjusted to a level sufficient to allow controlled ventilation. Animals were placed on a V-bracket in a dorsal recumbent position, and their limbs were secured to prevent lateral displacement during chest compression. They were intubated and ventilated with positive pressure. The fraction of inspired oxygen and minute ventilation were adjusted to achieve a normal arterial blood gas level. A standard lead II ECG was monitored throughout the experiment. Micromanometer-tipped catheters (Millar, Houston, TX) were placed in the aortic arch and right atrium through femoral artery and external jugular vein cutdowns, respectively. A custom-designed 13.5-F occluding-perfusing balloon catheter (Datascope, Oakland, NJ) was placed in the contralateral femoral artery by a combination of surgical cutdown and open guidewire techniques. A custom-designed esophageal balloon micromanometry catheter (Cook, Bloomington, IN) was placed in the esophagus at the level of the left atrium and inflated with saline solution to a pressure of 25 mm Hg. This was
ANNALS OF EMERGENCY MEDICINE
34:6 DECEMBER 1999
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
used to standardize the force of chest compression, preventing interanimal and intergroup differences. Catheter positions were confirmed by fluoroscopy and appropriateness of pressure waveforms and at postmortem examination. The pressure catheters, along with the ECG, were connected to a multichannel amplifier-recorder (Grass Instruments, Boston, MA) and recorded throughout the experiment. Amplifier output was also digitized (Dash16), displayed, and stored by using a microcomputer. Only the esophageal pulse pressure was displayed in the visual field of the person performing chest compressions. A No. 4 bipolar pacing wire was placed in the right ventricle through the jugular vein. Pancuronium, 3 mg, was administered just before induction of cardiac arrest to prevent ventilatory movement.8 Ventricular fibrillation was induced by briefly passing a 60-Hz alternating current at 25 mA through the right ventricular pacing wire. When fibrillation was confirmed by ECG and loss of arterial pressure, ventilation was discontinued. During arrest, the balloon occlusion catheter was inserted through the femoral artery introducer and positioned in the proximal descending aorta. This catheter has a large central infusion port that was connected to 450 mL of UPBH that had been preoxygenated to a PO2 of 90 to 150 mm Hg and pressurized to 300 mm Hg. The positions of all catheters were confirmed by fluoroscopy and appropriateness of waveforms and at postmortem examination. Animals remained in cardiac arrest without therapy for 10 minutes, after which standard basic life support (BLS) measures were initiated.9 This consisted of chest compressions at the rate of 80 per minute, a compression/ relaxation ratio of 0.5, and ventilation with 100% oxygen at a rate of 16 breaths/min. Manual chest compression was used because it may more accurately mimic clinical cardiac arrest than mechanical devices. The depth of compression was adjusted to achieve an aortic pulse pressure of 50 mm Hg.3 This pulse pressure was chosen because it mimics those seen in human beings during standard ACLS.3-10 Previous studies have shown that this results in an outcome that tends to mimic that seen in clinical trials in which outcomes for standard therapy are uniformly worse than the results of laboratory investigations.11 Once the predetermined pulse pressure was achieved, the person performing chest compression maintained an esophageal pressure waveform of consistent magnitude and shape. This esophageal system for standardization of chest compression has previously been
DECEMBER 1999
34:6
ANNALS OF EMERGENCY MEDICINE
described.1 Arterial blood gas analysis was performed 90 seconds after initiation of therapy, and ventilation parameters were adjusted accordingly. Thirteen minutes after initiation of arrest, the aortic balloon was inflated with 5 mL of radiopaque contrast material, and the study drug, epinephrine or saline solution, was administered as a rapid bolus into the proximal aorta through the occluding-perfusing catheter. The persons providing resuscitative therapy were blinded as to whether saline solution or epinephrine was administered. UPBH infusion was begun 13.5 minutes after arrest. Four hundred fifty milliliters of UPBH was infused as rapidly as possible, with a maximal length of infusion of approximately 1 minute. At the end of the infusion, defibrillation was attempted with 200 J and then 300 J. After countershock, the balloon was immediately deflated, and ECG rhythm and pulse were checked. If the animal remained in cardiac arrest after initial countershocks, then ACLS continued. Additional epinephrine was not given. Abnormalities in arterial blood gas parameters were treated by adjustment of minute ventilation and without use of sodium bicarbonate.12 Therapy was discontinued after 17 minutes if the arterial blood gas level was within normal limits. Animals remaining in cardiac arrest at that time were considered to have failed resuscitation. Resuscitated animals were stabilized after arrest by ventilatory and pressor support and administration of bicarbonate and crystalloid as needed on the basis of arterial blood gas values, vital signs, and central venous pressures, respectively. During the immediate postresuscitation phase, the aortic balloon was partially inflated as needed to treat hypotension refractory to pressors. An intravenous epinephrine infusion maintained the mean arterial blood pressure above 90 mm Hg. Minute ventilation was adjusted to maintain the PCO2 below 30 mm Hg. Sodium bicarbonate was administered, 1 mEq for each unit of base excess below –10, if the pH dropped below 7.1. Animals remaining alive 60 minutes after resuscitation were killed by injection of potassium chloride and additional barbiturate. All animals underwent postmortem examination to confirm catheter location, competence of the aortic valve, and absence of air emboli and to examine internal organs for gross injury. Complete occlusion by the aortic balloon was confirmed at postmortem examination. Definitions of the CPR phases and pressures have previously been described.3 Unless otherwise noted, all pressures are from the relaxation phase of CPR. Prearrest hemodynamics and arterial blood gases were measured
6 9 9
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
just before the initiation of ventricular fibrillation. BLS intravascular pressures were measured just before inflation of the obstructing balloon. ACLS-SAPO intravascular pressures were the mean of pressures measured at 5second intervals from the onset of occlusion until the end of infusion. The maximal aortic, coronary perfusion, and right atrial pressures were defined as the highest pressures measured during ACLS-SAPO and did not necessarily occur simultaneously. The change in intravascular pressures was defined as the maximal pressure during infusion minus the BLS pressure. The endpoint variables were prospectively identified as the change in aortic relaxation and coronary perfusion pressure after administration of the epinephrine or saline solution and the proportion of animals in each group with return of spontaneous circulation. Procedure failures, such as inability to place a catheter, untoward events during preparation or performance of the protocol, were prospectively listed and defined as exclusion criteria. Survival to 1 hour was not identified as part of the protocol. However, these data are reported as well. Data are reported as means±SD. All pressures are in millimeters of mercury. Normally distributed interval data were compared by using 2-sided 2-sample t tests with Bonferroni correction for multiple testing. Interval data that were not normally distributed were compared by using the Mann-Whitney rank-sum test. The Fisher exact test was used to compare proportions. The 95% confidence interval for the difference in the means is also reported for normally distributed data. Statistical significance was prospectively set at a P value of less than .05. R E S U LT S
Eighteen animals with a mean weight of 21±3 kg were studied. Three animals were excluded because of procedure failure, 2 from the epinephrine group and 1 from the Table 1.
Vital signs during spontaneous circulation before induction of cardiac arrest.
Vital Signs
Saline Solution Epinephrine
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Heart rate (beats/min)
152±10 131±7 132±34
165±18 131±9 137±28
95% CI, 95% confidence interval for the difference in the means.
7 0 0
95% CI
P Value
–31.7 to 5.7 –10.2 to 9.8 –46 to 34.7
.15 .97 .76
saline solution group. There were no significant differences in blood pressure, heart rate, and arterial blood gas values before initiation of ventricular fibrillation (Table 1). There were no significant differences in arterial blood gas measurements during spontaneous circulation before induction of cardiac arrest, during BLS, and just after return of spontaneous circulation (ROSC). There were no significant differences in the aortic, right atrial, and coronary perfusion pressures before initiation of SAPO (Table 2). During SAPO, the aortic and coronary perfusion pressures were all greater in animals that received epinephrine (Table 2). The difference in aortic relaxation and coronary perfusion pressures became statistically significant within 15 seconds after administration of epinephrine and remained significant until the end of the infusion (Figures 1 and 2). The change in right atrial pressure was also greater in animals that received epinephrine, and there was a trend toward greater maximal right atrial pressures during infusion in animals that received epinephrine (Table 2). The maximal aortic and coronary perfusion pressures occurred approximately halfway through infusion in animals receiving epinephrine, whereas animals treated only with saline solution had pressures that gradually increased throughout infusion. Right atrial pressures gradually increased throughout infusion in both groups. Two of seven animals in the saline solution group had ROSC compared with 7 of 8 in the epinephrine group Table 2.
Mean aortic, right atrial, and coronary perfusion pressures during BLS and representative mean aortic, maximal right atrial, and change in coronary perfusion pressures during ACLS-SAPO. Saline Solution
Pressures BLS-aortic (torr) BLS-RA (torr) BLS-CPP (torr) Aortic (torr) Maximal aortic (torr) Change aortic (torr) RA (torr) Maximal RA (torr) Change RA (torr) CPP (torr) Maximal CPP (torr) Change CPP (torr)
19±1 22±7 –3±7 31±10 39±12 20±11 25±7 30±7 8±5 7±12 23±6 26±10
Epinephrine 19±5 21±8 –3±11 67±3 76±4 58±5 30±9 39±10 18±5 36±11 49±10 52±8
95% CI
P Value
–5 to 4 –9 to 10 –12 to 11 –45 to –27 –49 to –26 –49 to –27 –15 to 5 –19 to 2 –16 to –3 –44 to –14 –36 to –15 –37 to –14
.8 .8 .95 <.001 <.001 <.001 .3 .1 .008 .001 <.001 <.001
95% CI, 95% confidence interval for the difference in the means; RA, right atrial pressure; CPP, coronary perfusion pressure. Unless otherwise indicated, data are from the relaxation phase of CPR.
ANNALS OF EMERGENCY MEDICINE
34:6 DECEMBER 1999
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
(P=.04). Two epinephrine-treated animals died of refractory hypotension before 1 hour. Postmortem examination showed all catheters to be properly located. Organ injuries, such as pulmonary atelectasis and hemorrhage and epicardial hemorrhage, were consistent with the duration of arrest and administration of external chest compression. No injuries attributable specifically to SAPO were noted. DISCUSSION
External chest compression induces a pressure differential between the arterial and central venous circulations, resulting in perfusion of the myocardium.3 However, after the first minutes of cardiac arrest, standard external chest compression requires administration of exogenous vasoactive agents to achieve perfusion pressures adequate for ROSC.13 Administration of exogenous adrenergic agonists can, however, negatively affect the oxygen supply/demand equilibrium.13,14 These data indicate that an exogenous pressor may be required even when techniques such as aortic perfusion are used. Aortic infusions have been shown to be effective in restoring spontaneous circulation when combined with aortic occlusion and external chest compression.1 Techniques such as ACLS-SAPO directly augment aortic pressure by increasing the volume of fluid within the central arterial compartment.1 If arterial compliance remains static, this increased volume results in improved perfusion of the heart and brain. During cardiac arrest and Figure 1.
Aortic relaxation phase pressures (in millimeters of mercury, mean±SD) for animals treated with epinephrine (circles) or saline solution (triangles). The vertical arrow indicates the time of drug administration.
DECEMBER 1999
34:6
ANNALS OF EMERGENCY MEDICINE
CPR, endogenously derived plasma epinephrine levels are higher than in any other pathophysiologic state,15 making it reasonable to hypothesize that exogenous epinephrine might be unnecessary during ACLS-SAPO. However, the present study indicates that after prolonged cardiac arrest, vasomotor tone has decreased so dramatically that even techniques such as ACLS-SAPO may require exogenous vasopressors. Animals not treated with epinephrine had only a moderate increase in aortic pressure, reflecting high arterial compliance and loss of infused UPBH from the central arterial compartment. The trend toward increased right atrial pressures in animals treated with epinephrine may reflect improved arterial to venous flow. Animals treated with epinephrine had a dramatic increase in aortic relaxation phase pressures during the first 30 seconds of infusion. Animals that did not receive epinephrine had a gradual rise in aortic pressure, which continued throughout the infusion (Figure 1), a pattern that seems characteristic of arterial volume infusion without vasopressor-mediated changes in aortic compliance. This study may provide indirect indication that there is a profound loss of vasomotor tone with continued cardiac arrest. After 13 minutes of cardiac arrest, arterial vasomotor tone had decayed sufficiently that even direct infusion into an occluded aorta was unable to raise perfusion pres-
Figure 2.
Coronary perfusion pressures (in millimeters of mercury, mean±SD) for animals treated with epinephrine (circles) or saline solution (triangles). The vertical arrow indicates the time of drug administration.
7 0 1
USE OF A PRESSOR DURING CARDIAC ARREST Paradis
sure to levels adequate for ROSC. The vasopressor effect from endogenous epinephrine was therapeutically inadequate after so long an arrest. The intra-aortic route for drug administration was chosen because it has previously been shown to be more effective than the intravenous route16,17 and may prevent potential toxicity resulting from epinephrine’s passage through the pulmonary circulation.18 SAPO may prolong the venous phase of intravenously administered drug, resulting in increased metabolism before reaching the site of action. Comparison of the pressor response curves indicates that, even with fluid infusion directly into an occluded aorta, epinephrine was responsible for most of the increase in aortic pressure.17 The curve of the pressor response also provides insight into the pharmacokinetics of intra-aortic epinephrine during ACLS-SAPO therapy. The time to peak effect is only about 30 seconds, and the pressor effect has begun to subside by 1 minute. This study has a number of limitations. Increased rates of ROSC and short-term survival in animal models only indicate a therapy’s potential effect on the outcome of patients undergoing cardiac arrest. Although ROSC and short-term survival are widely accepted as surrogate variables, survival studies need to be performed before concluding that these results apply to long-term outcome. Potentially, the effect of pressors on ROSC might be offset by delayed organ toxicity. The 1-hour survival data would seem to indicate this because 2 animals were lost to refractory hypotension. However, because this portion of the study was not intended to be controlled, these data are hypothesis generating at best. In animal studies, investigators must choose a specific duration of insult, whereas human cardiac arrest is clinically variable in presentation. The results of the present study indicate the effect of epinephrine on ROSC only at this defined insult. Early in arrest, CPR alone might be adequate to achieve ROSC, whereas later, higher epinephrine dosages might be needed.
3. Paradis NA, Martin GB, Goetting MG, et al: Simultaneous aortic, jugular bulb, and right atrial pressures during cardiopulmonary resuscitation in humans. Insights into mechanisms. Circulation 1989;80:361-368. 4. Pearson JW, Redding JS: Peripheral vascular tone in cardiac resuscitation. Anesth Analg 1965;44:746-752. 5. Michael JR, Guerci AD, Koehler RC, et al: Mechanisms by which epinephrine augments cerebral and myocardial perfusion during cardiopulmonary resuscitation in dogs. Circulation 1984;69:822-835. 6. Niemann JT, Rosborough JP, Ung S, et al: Coronary perfusion pressure during experimental cardiopulmonary resuscitation. Ann Emerg Med 1982;11:127-131. 7. Sehgal LR, Gould SA, Rosen AL, et al: Polymerized pyridoxylated hemoglobin: A red cell substitute with normal oxygen capacity. Surgery 1984;95:433-438. 8. Niemann JT, Rosborough JP, Niskanen RA, et al: Mechanical “cough”: Cardiopulmonary resuscitation during cardiac arrest in dogs. Am J Cardiol 1985;55:199-204. 9. Standards and guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). JAMA 1986;255:2841-3044. 10. Rivers EP, Lozon J, Enriquez E, et al: Simultaneous radial, femoral, and aortic arterial pressures during human cardiopulmonary resuscitation. Crit Care Med 1993;21:878-883. 11. Eisenberg MS, Horwood BT, Cummins RO, et al: Cardiac arrest and resuscitation: A tale of 29 cities. Ann Emerg Med 1990;19:179-186. 12. Guidelines for cardiopulmonary resuscitation and emergency cardiac care. JAMA 1992;268:2171-2299. 13. Paradis NA, Koscove EM: Epinephrine in cardiac arrest: A critical review. Ann Emerg Med 1990;19:1288-1301. 14. Ditchey RV, Lindenfeld J: Failure of epinephrine to improve the balance between myocardial oxygen supply and demand during closed-chest resuscitation in dogs. Circulation 1988;78:382389. 15. Wortsman J, Frank S, Cryer PE: Adrenomedullary response to maximal stress in humans. Am J Med 1984;77:779-784. 16. Manning JE, Batson DN, Murphy CA, et al: Selective aortic arch perfusion with oxygenated fluorocarbons combined with aortic arch epinephrine [abstract]. Ann Emerg Med 1993;22:929930. 17. Manning JE, Murphy CA Jr, Batson DN, et al: Aortic arch versus central venous epinephrine during CPR. Ann Emerg Med 1993;22:703-708. 18. Tang WC, Weil MH, Gazmuri RJ, et al: Pulmonary ventilation/perfusion defects induced by epinephrine during cardiopulmonary resuscitation. Circulation 1991;84:2101-2107.
I thank Mr Christopher Davison, without whose assistance this research could not have been performed, and Mryon L Weisfeldt, MD, for reviewing the manuscript.
REFERENCES 1. Paradis NA, Rose MI, Gawryl M: Selective aortic perfusion and oxygenation: An effective adjunct to external chest compression-based cardiopulmonary resuscitation. J Am Coll Cardiol 1994;23:497-504. 2. Paradis NA, Martin GB, Rivers EP, et al: Coronary perfusion pressure and return of spontaneous circulation in human cardiopulmonary resuscitation. JAMA 1990;263:1106-1113.
7 0 2
ANNALS OF EMERGENCY MEDICINE
34:6 DECEMBER 1999