Resuscitation, 19 (1990) l-16 Elsevier Scientific Publishers Ireland Ltd.

Collective Review

Adrenergic agonists during cardiopulmonary resuscitation Charles G. Brown and Howard A. Werman Divtsion of Emergency Medicine. The Ohio State University, 450 West Tenth Avenue, Columbus, OH 43210 (U.S.A.) A number of studies have suggested that following a prolonged cardiopulmonary arrest, large doses of alpha-adrenergic agonists that possess post-synaptic alpha-2 agonist properties, i.e. epinephrine and norepinephrine, may be required to enhance myocardial and cerebral hemodynamics. While initial human studies using large doses of epinephrine have shown improved hemodynamics over standard therapy, hospital discharge rates and neurological outcome have been discouraging. This probably reflects the fact that the administration of epinephrine was employed late in the resuscitation effort. Future studies using larger doses of epinephrine as the initial pharmacologic intervention during cardiopulmonary resuscitation (CPR) will help to determine whether there is any therapeutic benefit. In addition, a number of questions still remain unanswered in delineating the specific alpha and beta adrenergic agonist components which will maximally enhance hemodynamics and resuscitation rates during CPR. This will help determine whether norepinephrine or a yet unsynthesized adrenergic agonist may be more beneficial for use during cardiac arrest. Cardiac arrest - Adrenergic agonists - Catecholamines INTRODUCTION

Several studies have shown that myocardial and cerebral blood flow during closed-chest cardiopulmonary resuscitation (CPR) is insufficient to meet the metabolic demands of the heart and brain. Alpha-adrenergic agonists such as epinephrine, norepinephrine, phenylephrine and methoxamine, have ail been used during CPR to improve myocardial and cerebral blood flow. Only recently have systematic investigations been conducted in order to determine the most effective adrenergic agonist and its dose during CPR. This paper reviews those studies that have looked at the dose-response effects of adrenergic agonists during CPR. In addition, to fully understand which adrenergic agonists may be most effective, this paper also discusses the experimental work which has defined the myocardial and cerebral blood flow requirements, as well as the rationale for use of alpha-adrenergic agonists during CPR. DEFINING

MYOCARDIAL

CARDIOPULMONARY

AND

CEREBRAL

BLOOD FLOW

REQUIREMENTS

DURING

RESUSCITATION

The total duration of time that the brain and heart can withstand a global ischemic normothermic cardiopulmonary arrest before irreversible tissue injury 0300-9572/90/$03.50 0 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

H Aorta+ CoronaryArtery

E

A

CoronarySinus +

I Oxygen

Content

=

(HGB x 1.39 x YO saturation) MVO, = MBF (C,O, - C,, 0,)

+ (0.003

x

POP)

MDO, = MBF x C,O, Fig. 1. Model for measuring myocardial oxygen consumption (MVO,) and oxygen delivery (MDO,). CaO,, arterial oxygen confent; Cer02, coronary sinus oxygen content; HGB, hemoglobin; MBF, myocardial blood flow; PO,, partial pressure of oxygen.

occurs has not been defined. Although early reports suggested a 4-lO-min limit for successful myocardial and cerebral resuscitation [l-3], more recent studies in animals [4,5] and clinical reports in humans [6], have demonstrated that the heart may tolerate 15-20 min, and the brain up to 22 min, of normothermic global ischemia before the onset of irreversible injury. While investigators continue to define these time limits more precisely, it is evident that both the brain and heart can survive more prolonged ischemic insults than previously thought. Other investigative work has attempted to determine the minimal myocardial and cerebral blood flow requirements needed to maintain tissue viability and to restore spontaneous circulation during CPR. One approach that has been employed to define the myocardial blood flow requirements during CPR is to measure the myocardial oxygen requirements during cardiac arrest (Fig. 1). This can be calculated in animal models by determining the difference in the oxygen content of the blood entering (CaO,; arterial oxygen content) and leaving (C,,O*; coronary sinus oxygen content) the heart over time. This difference in oxygen content gives an estimate of myocardial oxygen consumption (MVO,). By knowing MVO,, one has in fact defined the amount of oxygen that needs to be delivered to the heart to meet its metabolic demands. This is termed myocardial oxygen delivery (MDO,) requirements. Since MDO, is the product of myocardial blood flow (MBF) and CaO, and estimates of CaO, can be made (Fig. l), one can calculate the level of MBF required during CPR to meet the metabolic demands of the heart in each of the cardiac dysrhythmias (ventricular fibrillation, asystole and electrical-mechanical dissociation) during cardiac arrest. These estimates of MVO, and the corresponding calculation of MBF requirements during these cardiac dysrhythmias are listed in table one [7-l 11. One problem with prior measurements of MVO, during ventricular fibrillation is that they were determined immediately at the onset of cardiac arrest. The MVO, is not a static parameter during cardiac arrest but changes over time. In patients who have a cardiac arrest outside the hospital, the average duration of time between the onset of the cardiac arrest and the institution of advanced cardiac life support is approxi-

3 Table I. Myocardial oxygen consumption (MVO,) and myocardial blood flow (MBF) requirements during various cardiac dysrhythmias. Normal sinus rhythm (NSR) shown for comparison. EMD, electrical mechanical dissociation; VF, ventricular fibrillation. MVO, in ml O,/min/lOO g; MBF in ml/min/ loog. Cardiac rhythm

MVO,

MBF

NSR Asystole EMD VF (early) VF (late)

15.0 2.0 2.04 5.0 10.0

75.0 10.0 10.2 25.0 50.0

ten minutes [ 121. Thus prior animal studies which determined MVO, at the onset of cardiac arrest may not reflect the oxygen debt that has occurred in cardiac arrests of longer duration. More recently, studies reporting MVO, following lo-15 min of a normothermic global ischemic arrest and 3 min of CPR during ventricular fibrillation, have calculated the MVO, to be approximately 10 ml O,/min/lOO g, or twice that previously reported when measured immediately after the onset of cardiac arrest [ 13,141. Using this value, and estimates of arterial oxygen content, the MBF requirements during CPR following a prolonged fibrillatory cardiac arrest approach 50 ml/min/lOO g (Table I). Similar estimates of cerebral oxygen consumption (CVO,) following a prolonged cardiac arrest have not been made. Thus the level of cerebral blood flow (CBF) required during CPR is less clear. Although determination of CVO, has been made with shorter arrests [15], our best estimates of the CBF requirements come from electroencephalographic (EEG) studies in humans undergoing carotid endarterectomy. This work suggests that EEG abnormalities occur when CBF falls below 15 ml/min/ 100 g [ 161. In addition, further animal studies have noted that neuronal cell membranes depolarize with CBF below 10 ml/min/lOO g [17]. Therefore the minimal CBF requirements during CPR needed to maintain neuronal viability appear to be approximately lo-15 ml/min/lOO g.

mately

CEREBRAL AND MYOCARDIAL BLOOD FLOW DURING CPR

Measurements of cerebral and myocardial blood flow during CPR have been made in several animal models. The variability in the levels of CBF and MBF that has been reported is due in part to differences in the duration of time between the onset of cardiac arrest and the initiation of CPR. This duration of time is referred to as downtime. One study has shown that as downtime increases, the level of CBF decreases with closed-chest CPR [18]. With downtimes of 9 min in this model, CBF approached zero [la]. No such study has been conducted to measure the change in MBF with varying downtimes. Extrapolating from several animal studies using downtimes of 0 [19] and 10 min [20], one can also hypothesize that the level of MBF generated with closed-chest CPR decreases as downtime increases; with levels of

4

Aortic Diastolic Pressure (mmHg) (100

Flow (ml/min/lOO grams) 2001

Cortex

Diastolic Pressure

Fig. 2. Aortic diastolic pressures, myocardial and cerebral blood flows during CPR. NSR values shown for comparison.

about 30% of baseline at 0 downtimes, and approaching < 10 ml/min/lOO g, or about 6% of baseline with a lo-min downtime. Regional cerebral blood flow measurements following a lo-min downtime have shown that the cerebral corticies receive < 3% of normal sinus rhythm (NSR) flows or about 1 ml/min/lOO g [21] (Fig. 2). Regional blood flows to the cerebellum and brainstem are slightly better, but on the average remain < 10 ml/min/lOO g [21]. Measurements of MDO, in the same model are < 1 ml O,/min/ 100 g [22]. Therefore, for the cardiac arrest occurring outside the hospital, where downtimes average 10 min, the level of cerebral and myocardial blood flow, and myocardial oxygen delivery generated with closed-chest CPR is exceedingly low, and probably insufficient to maintain cerebral and myocardial viability, or repay the oxygen debt that has occurred during cardiac arrest. RATIONALE FOR ALPHA-ADRENERGIC

AGONIST DRUGS DURING CPR

Blood flow to the myocardium is the result of the pressure differential across the vascular bed divided by the vascular resistance [23]. MBF occurs during diastole, thus the pressure differential across the myocardium is the difference between coronary artery and coronary sinus pressures during diastole (relaxation phase during closed-chest CPR). Since these arterial and venous pressures are not easily measured during CPR, the vascular beds from which they originate and drain into respectively; are measured instead. Thus, differences in the aortic diastolic pressure (ADP) and right atria1 diastolic pressure (RADP) are used to measure the pressure gradient across the myocardium [24]. The term myocardial perfusion pressure (MPP) reflects this difference: MPP = ADP - RADP

(1)

5

When the resistance (R) across the vascular bed of the myocardium consideration, the equation for MBF becomes:

MBF =

is taken into

ADP - RADP (2) R

During closed-chest CPR when myocardial hypoxia is severe, myocardial resistance vessels are probably maximally vasodilated, and thus, it is logical to assume that the resistance across the myocardial vascular bed’is low. Right atria1 diastolic pressure is also low during closed-chest [20], thus, the main driving force for MBF during CPR is ADP. Studies in both animals [20] and humans [25] have demonstrated that ADP is low (< 20 mmHg) during closed-chest CPR following a prolonged cardiopulmonary arrest when no vasopressors are used: Therefore interventions that increase ADP should be effective in improving MBF during CPR. Since pressure is dependent upon volume and capacitance:

Pressure =

Volume (3) Capacitance

investigative efforts have focused on volume loading or decreasing arterial capacitance during CPR, in order to increase ADP. Studies which have looked at volume loading have shown no definitive benefit [26-281. As a result, a major effort has focused on the use of alpha-adrenergic agonist drugs during closed-chest CPR. These drugs vasoconstrict the arterial vasculature, and decrease arterial capacitance, thus increasing ADP [29,30]. Improvements in ADP would increase MBF and MDO, and improve efforts at maintaining adequate perfusion. This would help repay the oxygen debt and exhance efforts at preserving myocardial tissue viability. The rationale for using alpha-adrenergic agonists to improve CBF has also been demonstrated. Vasoconstriction of the peripheral circulation, in particular, decreasing flow to the skin and skeletal muscle, allows blood to be shunted away from the extra-cerebral vasculature and non-vital organs to the intracerebal vasculature during CPR [31]. There is also evidence to suggest that the intrathoracic portion of the carotid artery collapses during CPR, which can be reversed with alphaadrenergic agonists [32,33]. DOSE-RESPONSE

EFFECTS OF ALPHA-ADRENERGIC

AGONISTS

DURING

CPR

Although much work has supported the use of alpha-adrenergic agonists like epinephrine, norepinephrine, methoxamine and phenylephrine during CPR, [34-441 and in particular the use of alpha-adrenergic agonists over adrenergic drugs with pure beta-agonist properties [45-481, only a few studies have tried to determine which alpha-adrenergic agonist is most beneficial in this setting [49--531. The doseresponse effects of alpha-adrenergic agonists on myocardial hemodynamics and cerebral blood flow have recently been systematically investigated [21,22,54,62].

6

30. 25. 20. 15. 10. 50

0.015

0.020

0.045

0.075

0.150

0.200

Dose (mg/kg) Fig. 3. Comparison of the change in aortic diastolic pressure following varying doses of epinephrine during CPR [22,55,62].

The current recommended alpha-adrenergic agonist for use during CPR is epinephrine, at a dose of 0.5-1.0 mg in adults [63]. This recommendation appears to be derived from one study conducted in 1906 [64]. Although the 1.O-mg dose used in this study has been extrapolated to adults, and thus translates into a dose of appproximately 0.01-0.02 mg/kg, no dose-response effects were actually investigated in this original study [64]. Recent animal work has begun to systematically

60

E

E

4030-

:

a

2010O-

.02

CPR

0.2 Eplnephrlne

.06 .12 .16 0.1 1.0 Noreplnephrlne Phenylephrine

0.1

1.0 10.0

Methoxamtne

Dose (mg/Kg) Comparison of aortic diastolic pressures during CPR, and following varying doses of epineFig. 4. phrine, norepinephrine. phenylephrine and methoxamine [22,55-571.

examine the dose-response effects of various alpha-adrenergic agonists, specifically epinephrine, norepinephrine, methoxamine and phenylephrine, on ADP, MBF, MDO,, MVO,, myocardial extraction ratio (MVOJMDO,), CBF and resuscitation rates during CPR. (Figs. 2-9). Aortic diastolic pressure during CPR following a prolonged cardiopulmonary arrest averages 20 mmHg. It is evident from Fig. 3 that doses of epinephrine from 0.015 to 0.150 mg/kg, on the average, do not raise ADP to clinically significant levels. Aortic diastolic pressures above 30 mmHg have been correlated with successful resuscitation in various animal models [64,65]. Epinephrine 0.20 mg/kg raises ADP by approximately 40 mmHg [55-571. Similar benefits have been found with norepinephrine in doses of 0.12-0.16 mg/kg [57] (Fig. 4). Despite large doses of methoxamine (0.10-10.0 mg/kg) and phenylephrine (0.10-l .O mg/kg), improvements in ADP were not consistently demonstrated in a model of prolonged cardiac arrest [55,56] (Fig. 4). Also, considering the minimal myocardial oxygen delivery (Fig. 1) and myocardial and cerebral blood flow criteria that was discussed earlier, only epinephrine (0.20 mg/kg) and norepinephrine (0.12-0.16 mg/kg) consistently improved MBF, MDO, and CBF to levels that would maintain tissue viability [21,54-611 (Figs. 5-8). Epinephrine (0.20 mg/kg) and norepinephrine (0.12-0.16 mg/kg) improved MDO, over MVO, as characterized by a decrease in the oxygen extraction ratio following drug administration, and by an improvement in resuscitation rates [55-571 (Figs. 6 and 9). The differences in hemodynamic response between the alpha-adrenergic agonists epinephrine, norepinephrine, methoxamine and phenylephrine may be related, in part, to their alpha-l and alpha-2 agonist properties. There is a growing body of evidence that show that both alpha-l and alpha-2 adrenoreceptors are located post-

100 90 80

en

0”

?O-

=

60-

2

50-

;

40-

2

3020 10 01 I CPR

I

.02 0.2 Eplnephrlne

ALlI

0.1 1.0 .08 Noreplnephrine Phenylephrlne

0.1 1.0 10.0

Methoxamine

Dose (mg/Kg) Comparison of myocardial blood flows during CPR, and following varying doses of epineFig. 5. phrine, norepinephrine, phenylephrine and methoxamine [22,54--571.

20-

8420-l CPR

.02 0.2 Eplnephrlne

I .08 .12 .18 0.1 1.0 Noreplnephrlne Phenylephrine

0.1 1.0 10.0 Methoxamlne

Dose (mg/Kg) Fig. 6. Comparison of myocardial oxygen consumption/delivery during CPR, and following varying doses of epinephrine, norepinephrine, phenylephrine and methoxamine [22,55--571.

junctionally in vascular smooth muscle [66]. The response to both alpha-l and alpha-2 stimulation post-junctionally is vasoconstriction [67]. The exact localization of these receptors within the vascular smooth muscle is of importance (Fig. 10). Investigators have demonstrated that intrajunctional alpha-adrenoreceptors are

24-1 --

I

222018181412lo884 2 .32 0.2 CPR

Eplnephrlne

.08 .12 .18 0.1 1.0 Noreplnephrlne Phenylephrlne

I 0.1 1.0 10.0 Yethoxamlne

Dose (mg/Kg) Fig. 1. Comparison of cerebral cortical blood flows during CPR. and following varying doses of epinephrine, norepinephrine. phenylephrine and methoxamine [21.58-611.

601

50ul

40-

8 =: Q,

30-

‘j

E E : E

2010OI

I

CPR

.02 Eplnephrlne

0.1 1.0 Noreplnephrlne Phenylephrlne

0.1 1.0 10.0 Methoxamlne

Dose (mg/Kg) Fig. 8.

Comparison of brainstem blood flows during CPR, and following varying doses of epinephrine, norepinephrine, phenylephrine, and methoxamine [21,58-61).

predominantly of the alpha-l subtype, and are located at the Ievel of the intima near the vessel lumen. The extrajunctional adrenoreceptors are located near the vessel lumen in the intima, and are predominately of the alpha-2 subtype. [68,69]. The intrajunctional alpha-l adrenoreceptor is activated primarily by neuronally released

100 90 80 70

I)

i E

60

g & Q

50 40 30

1

10 20 0 LCPR

I .02 0.2 Eplpephrlne

.08 .18 0.1 1.0 Noreplnephrlne Phenylephrlne

0.1 1.0 10.0 Methoxamlne

Dose (mg/Kg) Fig. 9.

Comparison of resuscitation rates following varying doses of epinephrine, norepinephrine, phenylephrine and methoxamine [22,55--571.

10

Adwntitio -

CROSS-SECTION OF RSISTANCE VESSEL

Medio

-

lntima

-

Lumen A ALPHA-2 Zh.

. ALPHA-1 F+mt-_I

H ALPHA-2 Pre-junctional InhibitsNA release

Cotechoklmines Fig. 10. Cross-section of resistance vessel with adrenergic nerve synapsing. Post-junctional and alpha-2 receptors depicted. NA, noradrenaline, norepinephrine.

alpha-l

norepinephrine, while the extrajunctional alpha-2 adrenoreceptor responds to circulating catecholamines [66]. During ischemic states there is evidence suggesting a relative decrease in the number of alpha- 1 binding sites [70]. Despite large increases in circulating endogenous catecholamines during CPR [71-741, ADP, MBF and CBF are low. Thus it is hypothesized that desensitization or tolerance develops at the level of the alpha-adrenoreceptor in response to the elevated levels of circulating endogenous catecholamines. In order to effect a hemodynamic response, that is, vasoconstriction of the peripheral circulation, large doses of alpha-adrenergic agonists may be needed to activate the desensitized receptor. In addition, since there appears to be a decrease in the number of post-junctional alpha-l binding sites, alpha-adrenergic drugs with post-synaptic alpha-2 agonist properties may be more effective in the ischemic state. Thus alpha-adrenergic agonists that possess alpha-2 agonist properties like epinephrine and norepinephrine, as opposed to pure alpha-l agonists methoxamine and phenylephrine [66], would be capable of mediating this hemodynamic response during CPR. While several human studies have compared epinephrine to phenylephrine [75] and methoxamine [76,77], the doses of alpha-adrenergic agonists used were too low to produce a significant hemodynamic response based on the above animal studies. Additional work has failed to show any benefit from standard doses of epinephrine in prehospital resuscitation [78]. More recently, dose-response improvements in arterial diastolic pressure and myocardial perfusion presssure have been demonstrated with larger doses of epinephrine in humans [79,80]. Although employed very late into the resuscitation effort, with incremental doses of epinephrine at 1.0, 3.0 and 5.0 mg, arterial diastolic pressure was maximally improved during CPR with 5.0 mg of epinephrine [79]. Improvements in myocardial perfusion pressure in humans have also been reported late in the resuscitation effort with 0.2 mg/kg of epine-

11

phrine, while a ‘standard’ dose of epinephrine 0.02 mg/kg, had little effect on this parameter [80]. While these larger doses of epinephrine did improve myocardial hemodynamics, the most important outcomes, hospital discharge rates and improved neurological outcome, were not realized. It must be noted that in these studies, this larger dose of epinephrine was not employed until very late in the resuscitation effort [79,80]. Although the exact limits of cerebral and myocardial viability have not been defined following a global ischemic arrest, it is clear that if larger doses of alpha-2 adrenergic agonists are to be of benefit, they must be employed earlier in the resuscitation effort. Despite recent case reports of successful resuscitation and normal neurological outcome using high-dose epinephrine, [81] larger doses of alpha-2 adrenergic agonists can not be recommended until more definitive human comparative trials are completed. Currently, a prospective, prehospital multi-center trial, comparing standard (0.02 mg/kg) and high-dose (0.20 mg/kg) epinephrine early in the resuscitation is underway. This study will evaluate not only resuscitation and hospital discharge rates, but neurological outcome as well. DEFINING

THE SPECIFIC

ADRENERGIC

AGONIST

REQUIREMENTS

DURING

CPR

The above animal and human studies have given some support to the concept that larger doses of alpha-adrenergic agonists with peripheral post-synaptic alpha-2 agonist properties are necessary to improve myocardial and cerebral hemodynamics during CPR. Several questions remain unanswered in defining the specific adrenergic requirements during CPR: (1) Are the beta-l and/or beta-2 agonist properties needed in addition to the alpha-2 agonist properties? (2) Are the alpha-l and alpha-2 agonist properties synergistic? The first of these questions can be answered in part. Several studies have looked at the comparative effects of equipotent doses of epinephrine and norepinephrine [57,61]. While both drugs have alpha-l,2 and beta-l agonist properties [29], norepinephrine is relatively devoid of beta-2 agonist effects on peripheral smooth muscle vasculature [82]. When comparing the hemodynamic responses of these two drugs, a trend toward improved MBF, CBF and resuscitation rates is noted with norepinephrine compared to epinephrine. Since beta-2 agonist properties may diminish or counteract the vasoconstrictive alpha-agonist effect on the peripheral smooth muscle vasculature [29], it appears that the beta-2 agonist effects may not be required to mediate the desired hemodynamic response during CPR. What remains unanswered is whether the beta-l agonist component is needed. Although earlier studies suggested that the beta-l components were detrimental in that they increased MVO, and adversely affected high-energy phosphate metabolism, there were several methodologic problems with these studies [83,84]. Most notable was the maintenance of a constant coronary perfusion pressure during drug administration in a cardiopulmonary bypass model of cardiac arrest [83]. As a result, epinephrine was prevented from increasing myocardial perfusion pressure, and thus, its full therapeutic benefit could not be realized. Similarly, a study which failed to show a benefit from large doses of epinephrine on high-energy phosphates during CPR did not adequately account for prolonged biopsy times and the extreme

12

Phenylethylamine

Imidazoline R