Resuscitation (2008) 77, 101—110

avai lab le at www.sc iencedi rec t .com

journa l homepage: www.e lsev ier .com/ locate / resusc i ta t ion

EXPERIMENTAL PAPER

Cardiovascular response to epinephrine varieswith increasing duration of cardiac arrest�

Mark G. Angelosa,b,∗, Ryan L. Butkea, Ashish R. Panchala,Carlos A.A. Torresa, Alan Blumberga,Jim E. Schneidera, Sverre E. Aunea

a Department of Emergency Medicine, The Ohio State University, Columbus, OH, United Statesb Davis Heart and Lung Research Institute, The Ohio State University, Columbus, OH, United States

Received 19 June 2007; received in revised form 23 October 2007; accepted 26 October 2007

KEYWORDSCardiopulmonaryresuscitation (CPR);Epinephrine;Adrenaline;Return ofspontaneouscirculation;Post-resuscitationperiod

SummaryObjective: Epinephrine (adrenaline) is widely used as a primary adjuvant for improving perfusionpressure and resuscitation rates during cardiopulmonary resuscitation (CPR). Epinephrine is alsoassociated with significant myocardial dysfunction in the post-resuscitation period. We testedthe hypothesis that the cardiac effects of epinephrine vary according to the duration of cardiacarrest.Methods and materials: Cardiac arrest (CA) was induced in Sprague—Dawley rats with an IVbolus of KCl (40 �g/g). Three series of experiments were performed with CPR begun after 2,4, or 6 min of cardiac arrest. Epinephrine (0.01 mg/kg) IV or placebo was given immediately inthe 2 and 4 min CA groups. In the 6 min group, CPR was started after 6 min CA and epinephrinewas given at 15 min if no return of spontaneous circulation (ROSC) occurred. Time to ROSC wasrecorded in all groups. Cardiac function was determined with trans-thoracic echocardiography

in after ROSC.

at baseline, 5, 30 and 60 m Results: After 2 min CA, 8/8 (100%) placebo animals and 8/8 (100%) epinephrine animals attainedROSC. Cardiac index was significantly increased during the first 60 min in the epinephrine groupcompared with the placebo group (p < 0.01). After 4 min of cardiac arrest, 14/29 (48%) placeboanimals and 14/16 (88%) epinephrine animals attained ROSC (p < 0.01). Cardiac index after ROSCreturned to baseline in both groups, although tended to be lower in the epinephrine group. After6 min CA, 10/31 (32%) animals attained ROSC without epinephrine and 17/21 (81%) animals with

� A Spanish translated version of the summary of this article appears as Appendix in the final online version atdoi:10.1016/j.resuscitation.2007.10.017.

∗ Corresponding author at: Department of Emergency Medicine and Davis Heart and Lung Research Institute, The Ohio State University,146 Means Hall, 1654 Upham Drive, Columbus, OH 43210, United States. Tel.: +1 614 293 7536; fax: +1 614 293 3124.

E-mail address: angelos.1@osu.edu (M.G. Angelos).

0300-9572/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.resuscitation.2007.10.017

102 M.G. Angelos et al.

epinephrine (p < 0.01). Post-ROSC depression of cardiac index was greatest in the epinephrinegroup (p < 0.05).Conclusions: As the duration of cardiac arrest increases, a paradoxical myocardial epinephrineresponse develops, in which epinephrine becomes increasingly more important to attain ROSC,

with post-ROSC myocardial depression.All rights reserved.

I

Acfdosptpcs

pdaaitggofishrfraac

posaasiiaifcenIeta

M

M

S(gt8oob(76vr3f(shrsttltplduoca2tgrgittsconsisting of mechanical ventilation with 100% O2 at pre-arrest values and continuous arterial blood pressure, heart

but is increasingly associated© 2007 Elsevier Ireland Ltd.

ntroduction

fter cardiac arrest with initial return of spontaneous cir-ulation, significant depression of myocardial contractileunction occurs frequently.1—3 Post-resuscitation myocardialysfunction has been studied primarily in animal modelsf ventricular fibrillation and asphyxial cardiac arrest,2,4

upported by observations from clinical medicine.5 Thisost-ischemic myocardial dysfunction is a primary contribu-or to the early post-resuscitation mortality in cardiac arrestatients.6 It is speculated that preserving post-resuscitativeardiac function will improve long-term cardiac arresturvival.7

Epinephrine (adrenaline) has long been considered therincipal adrenergic agent to improve perfusion during car-iopulmonary resuscitation (CPR) as a result of its alphadrenergic agonist properties. However, the role of the betadrenergic effects of epinephrine is less well understoodn the setting of cardiac arrest and may be detrimental inhe post-resuscitative heart.8 In earlier work, epinephrineiven during CPR, was found to increase myocardial oxy-en consumption and worsen the already tenuous ratio ofxygen delivery to oxygen consumption during ventricularbrillation (VF).9,10 However, under the high flow reperfu-ion conditions of cardiopulmonary bypass during VF, weave shown that epinephrine significantly reduces time toestoration of spontaneous circulation (ROSC) and improvesunctional cardiac recovery when high flow alone fails toestore cardiac function.11 These studies suggest some vari-bility in the cardiovascular response to epinephrine inccordance with the level of circulation generated duringardiac arrest.

The effects of epinephrine are not limited to the CPReriod of cardiac arrest, but also have significant effectsn myocardial function immediately following ROSC. Pasttudies with high dose epinephrine have noted severedrenergic side effects in the post-cardiac arrest heartttributed to epinephrine.8,12 Although, not well under-tood, the effect of epinephrine on the post-arrest hearts likely to vary depending on the severity of the preced-ng ischemic injury. Indeed, in an asphyxial rat cardiacrrest model, shorting the duration of asphyxia resultsn improved recovery of contractile function and acidosisollowing initial resuscitation.4 Certainly, the duration ofardiac arrest is a principal determinant in obtaining ROSC,ven when key outcome measures of perfusion such as coro-

13

ary perfusion pressure and end tidal CO2 do not differ.n this study we tested the hypothesis that epinephrine’sffects on ROSC success and post-ROSC myocardial func-ion vary with the duration of cardiac arrest preceding itsdministration.

rgoth

aterials and methods

odel

prague—Dawley rats weighing approximately 400—450 gHarlan, Indianapolis, IN) were used in accordance with theuide for Care and Use of Laboratory Animals published byhe US National Institutes of Health (NIH publication No.5-23, revised 1996) and the approval of the University Lab-ratory Animal Resources Committee. Animals were fastedvernight and anesthetized with intraperitoneal pentobar-ital (50 mg/kg) and intubated with a 16-gauge catheterangiocath) via a tracheotomy. Animals were ventilated at5 breaths/min using a rodent ventilator (Harvard Model83, South Natick, MA) with a FiO2 of 0.21 and a tidalolume of 0.65 mL/100 mg body weight to maintain arte-ial blood gases in the normal range (pO2 > 80 mm Hg, pCO2

5—45 mm Hg, and pH 7.35—7.45). The jugular vein andemoral artery were cannulated with a 24-gauge catheterangiocath) and sutured in place. Continuous arterial pres-ure was measured in the femoral artery. Animals wereeparinized (1000 U kg−1 bolus). Arterial pressure and heartate were recorded continuously using an on-line data acqui-ition system (Digi-Med Heart Performance Analyzer). Rectalemperature was maintained between 36.5 and 37.5 ◦Chroughout the duration of the experiment with a heatingamp. Using a standardized cardiac arrest model,14 adaptedo the rat, cardiac arrest was induced by infusing a bolus ofotassium chloride (40 �g/g KCl) into the jugular vein fol-owed by a 0.2 mL bolus of saline to ensure that the full KClose was delivered to the heart. Ventilation was discontin-ed simultaneously. Cardiac arrest was confirmed by the lossf the arterial trace with a MAP of <20 mm Hg. Manual chestompressions were started after the pre-determined cardiacrrest duration at a rate of 200—220/min with ventilation at5 breaths/min. Rate and force of compression were main-ained by continuous observation of the arterial wave formenerated during the CPR period. Epinephrine, based on cur-ent recommended weight-based dose (0.01 mg/kg),15 wasiven via the jugular vein at the pre-determined time dur-ng CPR. CPR was continued until the ROSC, defined as theime of return of a spontaneous arterial blood pressureracing with a MAP of >60 mm Hg without chest compres-ions. Following ROSC, all animals received identical care

ate and temperature monitoring for 60 min. Animals wereiven a further dose of pentobarbital if there were any signsf awakening. Fluid administration consisted of intermit-ent 1 mL boluses of normal saline if the animal developedypotension (MAP < 60 mm Hg). At the conclusion of the

E

TdueasMVrtodCwT

c

Cardiovascular response to epinephrine

60 min-monitoring period, animals were sacrificed painlesslywith pentobarbital.

Study protocols

Three cardiac arrest series were performed (Figure 1) todetermine the effects of epinephrine on recovery of myocar-dial function following ROSC after different durations ofcardiac arrest. In each series of experiments, the sameKCl rat cardiac arrest model, anesthetic, single dose ofepinephrine, CPR methods and post-resuscitation supportwere used. Series were done sequentially. In series 1,epinephrine (0.01 mg/kg) or placebo (normal saline) wasgiven in a blinded fashion after 2 min of untreated cardiacarrest and CPR was provided until ROSC or the durationof cardiac arrest exceeded 15 min. In series 2, epinephrine(0.01 mg/kg) or placebo (normal saline) was given in ablinded fashion after 4 min of untreated cardiac arrest andCPR was provided until ROSC or the duration of cardiac arrestexceeded 15 min. In series 3, CPR was started after 6 min ofuntreated cardiac arrest and continued until ROSC or 15 minof cardiac arrest. If no ROSC was attained at 15 min, then

epinephrine (0.01 mg/kg), was given and CPR was continueduntil ROSC or until the duration of cardiac arrest exceeded20 min. In this series, epinephrine was given as a rescuetherapy after a longer cardiac arrest duration and outcomecompared to the shorter non-epinephrine group.

d

s

Figure 1 Experimental protocols. Three series of experiments werarrest, CPR was started and epinephrine (0.01 mg/kg) or placebo wcardiac arrest at which time CPR was started and epinephrine (0.01a 6 min KCl-induced cardiac arrest, with CPR until ROSC or 15 min aarrest, EPI = epinephrine, PL = placebo, ROSC = restoration of spontan

103

chocardiography

rans-thoracic echocardiographic examination of myocar-ial function was completed pre- and post-cardiac arrestsing standard echocardiographic methods.16 Animals werexamined in the supine position after the chest was shavednd a layer of acoustic gel was applied. 2D and M-mode mea-urements were made with a Vivid 7 Ultrasound machine (GEedical Systems, Horten, Norway) using an 11 MHz probe.iews were taken after optimization of gain, angulation, andotation. M-mode measurements were made at or just belowhe level of the papillary muscles. Ultrasound measurementsbtained included, LV posterior wall and chamber diameteruring systole (LVIDs) and diastole (LVIDd) and heart rate.alculations of cardiac output were made by machine soft-are utilizing the formula below described previously by

eichholz et al.17

ardiac output = (diastolic volume − systolic volume)

×heart rate1000

(1)

iastolic volume = 7.02.4 + LVIDd

× (LVIDd)3 (2)

ystolic volume = 7.02.4 + LVIDs

× (LVIDs)3 (3)

e performed. In series 1, following a 2 min KCl-induced cardiacas given. In series 2, animals underwent a 4 min KCl-inducedmg/kg) or placebo was given. In series 3, animals underwent

t which time epinephrine (0.01 mg/kg) was given. CA = cardiaceous circulation.

1 M.G. Angelos et al.

O

U3atibIg

D

DmUat

R

Ts(leiti

sb(

Table 1 Baseline group comparisons of CA survivors

Series 1: (2 min CA) No EPI (n = 8) EPI (n = 8)

Weight (g) 403.7 ± 6.0 399.8 ± 9.9Pentobarbital (mL) 0.60 ± 0.15 0.50 ± 0.0Fluids-pre-arrest (mL) 2.20 ± 0.14 2.60 ± 0.31Frequency of ROSC 8/8 8/8ROSC time (min) 2:40 ± 0:11 2:23 ± 0:06Fluids-post-arrest (mL) 2.30 ± 0.65 2.02 ± 0.81

Series 2: (4 min CA) No EPI (n = 14) EPI (n = 14)

Weight (g) 438.0 ± 8.7 442.8 ± 7.4Pentobarbital (mL) 0.57 ± 0.02 0.54 ± 0.02Fluids-pre-arrest (mL) 1.4 ± 0.06 1.8 ± 0.16Frequency of ROSC 14/29 14/16*ROSC time (min) 5:16 ± 0:18 5:17 ± 0:14Fluids-post-arrest (mL) 2.2 ± 0.2 2.6 ± 0.2

Series 3: (6 min CA) No EPI (n = 10) EPI (n = 17)

Weight (g) 448.8 ± 7.0 448.4 ± 7.4Pentobarbital (mL) 0.53 ± 0.10 0.56 ± 0.03Fluids-pre-arrest (mL) 2.84 ± 0.59 2.56 ± 0.29Frequency of ROSC 10/31 17/21*ROSC time (min) 11:50 ± 1:28 16:54 ± 0:08Fluids-post-arrest (mL) 7.5 ± 1.0 9.6 ± 1.3*

All animals completed the 60 min post-resuscitation protocol.ROSC rates were significantly higher in the EPI groups after both

04

utcome measures

ltrasound measurements were determined at baseline, 5,0 and 60 min following ROSC. Arterial pressure, heart ratend temperature were measured continuously throughouthe study. Arterial blood gases, drawn at baseline, follow-ng ROSC and at 60 min post-ROSC, were measured on alood gas analyzer (Critical Care Laboratory Synthesis 45,L). ROSC rates and time to ROSC were determined in eachroup.

ata analysis

ata are presented as mean ± SEM. ROSC rates were deter-ined and compared within series using Mann—Whitney-test. Parametric data were analyzed within series usingone-way analysis of variance followed by a Tukey post hoc

est with p < 0.05 set as significant.

esults

here were no baseline group differences within eacheries in weight, pentobarbital dose or fluid requirementsTable 1). Echocardiographic evaluation of left ventricu-ar function was performed in all animals. Representativechocardiographic images pre-arrest and 5 min post-ROSCllustrate the early change in LV function following resusci-ation from cardiac arrest in an animal receiving epinephrine

n the 4 min series (Figure 2).

After a short cardiac arrest duration of 2 min beforetarting CPR, ROSC with 60 min survival was 100% inoth epinephrine (8/8) and non-epinephrine (8/8) groupsFigure 3a). However, with increasing cardiac arrest dura-

4 and 6 min CA. ROSC times were similar within series except forthe 6 min CA series, in which the EPI group was longer by design.CA = cardiac arrest, EPI = epinephrine, ROSC = restoration ofspontaneous circulation (*p < 0.05 between EPI and no EPIgroups).

Table 2a 60 min survivor outcome after 2 min cardiac arrest (series 1)

Group Baseline ROSC 5 min ROSC 30 min ROSC 60 min

Heart rate Placebo 390 ± 10 391 ± 14 352 ± 30 345 ± 29Epi 378 ± 10 369 ± 42 353 ± 12 342 ± 14

SV (mL) Placebo 0.28 ± 0.03 0.18 ± 0.1 0.21 ± 0.03 0.23 ± 0.04Epi 0.25 ± 0.02 0.29 ± 0.1 0.46 ± 0.08 0.52 ± 0.04*

EF (%) Placebo 83.6 ± 3.3 95.7 ± 1.3 83.2 ± 5.3 91.8 ± 5.0Epi 88.1 ± 2.3 77.5 ± 6.3 57.7 ± 3.8* 53.0 ± 5.8*

CI (L/(min kg)) Placebo 0.270 ± 0.049 0.200 ± 0.069 0.183 ± 0.063 0.183 ± 0.050Epi 0.239 ± 0.019 0.251 ± 0.051 0.393 ± 0.034* 0.425 ± 0.032*

FS (%) Placebo 48.3 ± 3.6 74.6 ± 6.7 52.4 ± 3.5 66.4 ± 7.2Epi 53.6 ± 3.2 44.8 ± 5.5 27.3 ± 2.7* 23.5 ± 4.6*

pH Placebo 7.49 ± 0.03 7.38 ± 0.04 7.28 ± 0.05Epi 7.49 ± 0.03 7.38 ± 0.04 7.28 ± 0.05

PaCO2 Placebo 32 ± 2 50 ± 6 37 ± 6Epi 29 ± 2 32 ± 6 39 ± 2

PaO2 (FiO2) Placebo 82 ± 7 (0.21) 245 ± 79 (1.0) 402 ± 52 (1.0)Epi 91 ± 4 (0.21) 227 ± 51 (1.0) 335 ± 69 (1.0)

Hct Placebo 41 ± 2 42 ± 1 37 ± 6Epi 41 ± 2 43 ± 3 39 ± 2

Syst art press Placebo 162 ± 5 114 ± 8 102 ± 7 106 ± 8Epi 159 ± 5 161 ± 11 119 ± 5 123 ± 7

*p < 0.05 compared with placebo group (placebo group n = 8, epinephrine group n = 8). SV = stroke volume, CI = cardiac index, EF = ejectionfraction, FS = fractional shortening, Hct = hematocrit.

Cardiovascular response to epinephrine 105

Figure 2 Representative echocardiography at baseline and 60 min following ROSC in a short cardiac arrest (a) with ROSC time of2:20 and a longer cardiac arrest (b) with a ROSC time of 5:17. Cardiac output following a short cardiac arrest with epinephrine wasincreased above baseline at 60 min following ROSC (a) in contrast to a lower than baseline cardiac output after a longer cardiac

rt ax

cntcuat(ust

arrest with epinephrine (b). Trans-thoracic 2D and M-mode shopapillary muscles.

tion before CPR (4 min), ROSC rates were significantly higherwith epinephrine compared to non-epinephrine groups(Figure 3a). With still longer durations of cardiac arrestbefore beginning CPR (6 min), ROSC was successful only32% of the time without epinephrine, but 81% successfulwith epinephrine in this same group initially refractory toROSC with CPR only (Figure 3a). In all three series, ani-mals attaining ROSC survived to 60 min. ROSC times weresimilar between epinephrine and non-epinephrine groupsafter 2 and 4 min CA, but were significantly longer in theepinephrine group after 6 min of cardiac arrest by design

(see Table 1).

Relative to baseline, cardiac index at 60 min post-ROSCwas increased in the epinephrine group in the 2 min series,unchanged in the 4 min series and significantly depressed inthe 6 min series (Figure 3b). Following resuscitation from

rIbia

is views of the left ventricle were obtained at the level of the

ardiac arrest in the 2 min series, cardiac index was sig-ificantly increased in the epinephrine group relative tohe non-epinephrine group (Table 2a). This increase inardiac index was primarily due to an increase in stroke vol-me (both systolic and diastolic LV volume), but was alsoccompanied by a reduction in ejection fraction and frac-ional shortening compared with the non-epinephrine groupTable 2a). Thus, although overall CI was increased, ventric-lar efficiency (ejection fraction) was decreased, indicatingome degree of myocardial dysfunction. In the 4 min series,he cardiac index increased transiently in both groups but

eturned to baseline cardiac index by 60 min (Table 2b).n this series, stroke volume was largely unchanged inoth groups (Table 2b). In the 6 min series, the cardiacndex was significantly decreased in the non-epinephrinend epinephrine group compared with baseline; however the

106 M.G. Angelos et al.

(Co

dtean

D

Tdadawtditi

pi

rhiitSt3vino

Figure 2

epression in the epinephrine group was greater relativeo the non-epinephrine group (Table 2c). Within the 6 minpinephrine group, a significant reduction in heart rate andgreater metabolic acidosis was present compared with theon-epinephrine group.

iscussion

he present study demonstrates the variable and discor-ant effects of epinephrine on initial resuscitation ratesnd myocardial dysfunction after varying durations of car-iac arrest. Utilizing a standardized cardiac arrest modelnd current 2005 American Heart Association recommendedeight-based epinephrine dose,15 we noted equal resusci-

ation rates with or without epinephrine after very shorturations of cardiac arrest, but significant improvement innitial resuscitation rates with epinephrine, as the dura-ion of cardiac arrest increased. Concurrently, we notedncreased myocardial depression in the post-resuscitation

arhih

ntinued ).

eriod with epinephrine as the duration of cardiac arrestncreased.

In clinical cardiac arrest, the greatest percentage ofesuscitation failures consists of the inability to restart theeart, i.e. to obtain ROSC. Data from the National Reg-stry of Cardiopulmonary Resuscitation show that of 14,720n-hospital cardiac arrests, only 44% had restoration of spon-aneous circulation and 17% survived to hospital discharge.18

imilarly, in out-of-hospital cardiac arrest victims, an ini-ial return of spontaneous circulation occurs in only about0%.19 Consequently, approximately 70% of cardiac arrestictims never recover any functional contractile heart activ-ty despite CPR and other interventions. Epinephrine or theon-adrenergic agent, vasopressin, are currently the rec-mmended agents to improve initial resuscitation rates in

ll cardiac arrest rhythms, including electrically susceptiblehythms if initial defibrillation fails.15 However, epinephrineas been associated with increased myocardial depressionn the early post-resuscitation period, particularly with inigh doses.8,12

Cardiovascular response to epinephrine 107

Table 2b 60 min survivor outcome after 4 min cardiac arrest (series 2)

Group Baseline ROSC 5 min ROSC 30 min ROSC 60 min

Heart rate Placebo 358 ± 6 377 ± 20 322 ± 18 352 ± 16Epi 368 ± 6 388 ± 20 422 ± 2.3 331 ± 16

SV (mL) Placebo 0.32 ± 0.02 0.30 ± 0.04 0.34 ± 0.01 0.32 ± 0.04Epi 0.30 ± 0.02 0.14 ± 0.04 0.16 ± 0.01 0.38 ± 0.04

EF (%) Placebo 83.5 ± 1.2 86.9 ± 3.0 71.4 ± 2.6 72.8 ± 4.8Epi 79.2 ± 1.2 83.8 ± 2.9 76.0 ± 2.6 74.9 ± 4.8

CI (L/(min kg)) Placebo 0.299 ± 0.049 0.266 ± 0.069 0.280 ± 0.063 0.253 ± 0.050Epi 0.235 ± 0.027 0.141 ± 0.042 0.154 ± 0.040 0.174 ± 0.023

FS (%) Placebo 47.5 ± 1.4 59.8 ± 4.1 39.6 ± 2.5 41.5 ± 4.2Epi 42.7 ± 1.8 52.3 ± 7.6 47.4 ± 2.6 40.3 ± 2.5

pH Placebo 7.45 ± 0.03 7.07 ± 0.04 7.33 ± 0.02Epi 7.50 ± 0.01 7.04 ± 0.04 7.39 ± 0.05

PaCO2 Placebo 35 ± 3 59 ± 5 40 ± 2Epi 30 ± 1 57 ± 4 32 ± 3

PaO2 (FiO2) Placebo 148 ± 60 (0.21) 88 ± 22 (1.0) 202 ± 48(1.0)Epi 101 ± 6 (0.21) 117 ± 20 (1.0) 282 ± 33(1.0)

Hct Placebo 44 ± 2 41 ± 2 44 ± 2Epi 45 ± 1 45 ± 1 45 ± 1

Syst art press Placebo 146 ± 18 141 ± 18 132 ± 6 119 ± 30Epi 171 ± 5 114 ± 8 102 ± 7 106 ± 8

oupatocr

fL

There were no significant differences between groups (placebo grindex, EF = ejection fraction, FS = fractional shortening, Hct = hem

Using non-invasive echocardiographic measures of LVfunction, the present study demonstrates that significant

myocardial depression occurs in the early post-cardiacarrest period after relatively short durations of cardiacarrest with and without the use of epinephrine. In previ-ous studies, myocardial dysfunction following resuscitation

isom

Table 2c 60 min survivor outcome after 6 min cardiac arrest (ser

Group Baseline RO

Heart rate Placebo 353 ± 18 3Epi 358 ± 10 2

SV (mL) Placebo 0.26 ± 0.06 0.Epi 0.29 ± 0.03 0.

EF (%) Placebo 85 ± 3Epi 83 ± 2

CI (L/(min kg)) Placebo 0.213 ± 0.037 0.2Epi 0.222 ± 0.028 0.1

FS (%) Placebo 50 ± 5Epi 47 ± 2

pH Placebo 7.46 ± 0.03 6.Epi 7.39 ± 0.02 6.

PaCO2 Placebo 33 ± 3Epi 37 ± 2

PaO2 (FiO2) Placebo 120 ± 15 2Epi 87 ± 5 (0.21)

Hct Placebo 44 ± 2Epi 44 ± 1

Syst art press Placebo 147 ± 11 1Epi 137 ± 5 1

*p < 0.05 compared with placebo group (placebo group n = 10, epinEF = ejection fraction, FS = fractional shortening, Hct = hematocrit.

n = 12, epinephrine group n = 14). SV = stroke volume, CI = cardiacit.

rom cardiac arrest has been characterized by decreasedV ejection fraction, fractional shortening, dP/dt−40 and

ncreased tau (isovolumetric relaxation time).5,20 In thistudy, epinephrine was associated with an increased cardiacutput state after a very short cardiac arrest, mild reversibleyocardial depression after a slightly longer cardiac arrest

ies 3)

SC 5 min ROSC 30 min ROSC 60 min

44 ± 29 366 ± 15 351 ± 2928 ± 23* 272 ± 18* 267 ± 17*35 ± 0.11 0.22 ± 0.05 0.13 ± 0.0431 ± 0.07 0.17 ± 0.03 0.16 ± 0.0394 ± 5 88 ± 3 89 ± 472 ± 24 79 ± 5 76 ± 928 ± 0.047 0.195 ± 0.034 0.129 ± 0.03621 ± 0.025* 0.105 ± 0.030 0.085 ± 0.01977 ± 10 54 ± 5 59 ± 648 ± 5 46 ± 5 52 ± 898 ± 0.03 7.19 ± 0.0697 ± 0.03 6.87 ± 0.02*51 ± 6 40 ± 655 ± 7 32 ± 485 ± 56 (1.0) 393 ± 52 (1.0)76 ± 20 (1.0) 317 ± 36 (1.0)39 ± 2 33 ± 439 ± 2 39 ± 303 ± 17 98 ± 8 91 ± 867 ± 16* 135 ± 10* 120 ± 8*

ephrine group n = 17). SV = stroke volume, CI = cardiac index,

108

Figure 3 (a) The rate of restoration of spontaneous circu-lation (ROSC) is shown as both a fraction and percentage ineach of the three series with and without epinephrine. Therewas no difference in ROSC rates when epinephrine was givenafter 2 min of cardiac arrest. However, with lengthening cardiacarrest times, ROSC rates were significantly higher in animalsreceiving epinephrine after 4 min (*p < 0.05) and 6 min (†p < 0.01)of cardiac arrest. (b) Cardiac index at 60 min post-resuscitationperiod with and without epinephrine in survivors following car-diac arrest times of 2, 4 and 6 min before starting CPR. Thecardiac index increased significantly in the epinephrine groupcompared with the non-epinephrine group following ROSC inthe 2 min series (*p < 0.05), and then decreased relative to thenon-epinephrine group in the 4 min series (p = 0.06). In the 6 minseries, the cardiac index was depressed in both epinephrine andngb

apwfddiaot

iIic

ap>saswfmpidrprtt

aheaTiadtad2tnofiwffdwaadh

tibctttated post-arrest myocardial dysfunction. Achieving ROSC

on-epinephrine groups, but was more severe in the epinephrineroup (†p < 0.05). The cardiac index at 60 min is expressed asoth a % of baseline in each group and in L/(kg min).

nd more severe myocardial depression noted after a morerolonged cardiac arrest. As the epinephrine dose and CPRere similar in all groups, the differences in post-arrest LV

unction were primarily a function of the duration of car-iac arrest. The duration of cardiac arrest is then a keyeterminant of the cardiovascular response of epinephrine

n cardiac arrest. As, epinephrine has both important pre-nd post-ROSC effects on the myocardium in the settingf cardiac arrest, any depression in post-myocardial func-ion induced by epinephrine must be balanced against any

ittd

M.G. Angelos et al.

mprovement in ROSC rates associated with epinephrine.ndeed it is the pre-ROSC effect, for which epinephrine isndicated, although the effects of epinephrine given duringardiac arrest may persist after ROSC.

Recent studies have again highlighted the potentialdverse effects of epinephrine during cardiac arrest. In theediatric cardiac arrest population, epinephrine doses of15 �g/kg were associated with an increased incidence ofecondary VF, with worse outcome than primary VF.21 In andult out-of-hospital cardiac arrest population resuscitateduccessfully with early myocardial dysfunction, epinephrineas noted to be an independent factor for low ejection

raction.22 The effect of epinephrine on the microcirculationay be an important factor in the early post-resuscitationeriod. Utilizing direct visualization of the sublingual cap-llary bed in a swine cardiac arrest model, a significantecrease in microcirculatory blood flow was noted in animalseceiving epinephrine (25 �g/kg).23 This decrease was mostronounced in the first minutes following ROSC. If a similaresponse is found to occur in the microcirculatory flow ofhe myocardium after epinephrine, this could contribute tohe decreased myocardial function observed.

As expected with a short arrest time, there was nodvantage with epinephrine during cardiac arrest. Of note,owever, is the absence of cardio-depressant effects withpinephrine in the short arrest period as has been reportedfter longer cardiac arrests and higher doses of epinephrine.his increase in cardiac output and absence of early mortal-

ty is in contrast to a recent report of increased mortalitynd myocardial depression after a 1 min asphyxial rat car-iac arrest with 10 or 30 �g/kg epinephrine.24 Compared tohis study we used the lower epinephrine dose, 10 �g/kgnd a non-asphyxial model of cardiac arrest. However,espite the increased cardiac output state noted in themin epinephrine group, a significant increase in ven-

ricular volume and a decrease in ejection fraction wereoted simultaneously. These findings suggest some degreef myocardial dysfunction, albeit compensated, was presentollowing epinephrine. An important limitation of this studys the relatively short study time frame following ROSC,hich does not allow for assessment of the duration and

ull extent of myocardial dysfunction over time. Instead theocus of this study was on the early post-ROSC myocardialepression which occurs during the very vulnerable timehen a significant portion of cardiac arrest patients initiallychieve ROSC, but then re-arrest and die prior to hospitaldmission. In a recent study only 47% of out-of-hospital car-iac arrest patients in whom ROSC was obtained survived toospital admission.25

This study reaffirms the importance of early epinephrineo facilitate ROSC and highlights the epinephrine dichotomyn cardiac arrest. This dichotomous epinephrine response isased on the observations that the longer the duration ofardiac arrest, the more difficult to achieve ROSC and thushe greater need for epinephrine. However, as the dura-ion of cardiac arrest increases, the benefit of epinephrineo attain ROSC is offset by increased epinephrine medi-

s the first priority in cardiac arrest and only then canhe secondary problem of post-arrest myocardial dysfunc-ion be addressed. Not surprising both issues are largelyependent on the duration of cardiac arrest. Currently

1

1

1

1

1

1

11

1

1

2

2

2

2

Cardiovascular response to epinephrine

the recommended guidelines for epinephrine use duringcardiac arrest call for the same epinephrine dose for allcardiac arrest conditions. As noted in this study, the effectof this same dose concentration is likely to vary depend-ing on a number of factors including duration of cardiacarrest, but also likely vary according to various factors whichaffect the level of CPR generated blood flow. Support forthis premise is found in a recent study in a swine car-diac arrest model in which epinephrine (0.02 mg/kg) wasgiven during different chest compression protocols and sig-nificant differences in epinephrine plasma concentrations,coronary perfusion pressure, cerebral and femoral bloodflow were noted during CPR.26 Ultimately, optimization ofepinephrine (or other adrenergic agents) doses during car-diac arrest must vary in accordance with specific cardiacarrest factors, such as length of arrest. Recent studies havefocused on use of selective alpha-2 adrenergic agents topromote ROSC in cardiac arrest but minimize effects on post-arrest myocardial function.27,28 These agents may yet be analternative to epinephrine use in cardiac arrest. However,more work is needed to understand the effects of theseagents in the post-arrest setting, before they can be rec-ommended.

Conclusions

As the duration of cardiac arrest increases, a paradox-ical myocardial epinephrine response develops, in whichepinephrine becomes increasingly more important for ROSC,but is increasingly associated with post-ROSC myocardialdepression. This study demonstrates both the benefit anddetriment of epinephrine in cardiac arrest over time andre-emphasizes the need for short acting agents designedto improve ROSC without depressing post-myocardial func-tion.

Conflict of interest

None.

Acknowledgements

This research was supported by the Roessler ScholarshipFund, Ohio State University (RL Butke), SAEM InstitutionalResearch Training Award (CA Torres) and Ohio State Strate-gic Initiatives Grant and the American Heart Association OhioAffiliate (MG Angelos).

References

1. Kern KB, Hilwig RW, Berg RA, et al. Postresuscitation leftventricular systolic and diastolic dysfunction. Treatment withdobutamine. Circulation 1997;95:2610—3.

2. Tang W, Weil MH, Sun S, Gazmuri RJ, Bisera J. Progressivemyocardial dysfunction after cardiac resuscitation. Crit CareMed 1993;21:1046—50.

3. Klouche K, Weil MH, Sun S, Tang W, Zhao DH. A comparison ofalpha-methylnorepinephrine, vasopressin and epinephrine forcardiac resuscitation. Resuscitation 2003;57:93—100.

4. McCaul CL, McNamara P, Engelberts D, Slorach C, HornbergerLK, Kavanagh BP. The effect of global hypoxia on myocardial

2

2

109

function after successful cardiopulmonary resuscitation in a lab-oratory model. Resuscitation 2006;68:267—75.

5. Kern KB, Hilwig RW, Rhee KH, Berg RA. Myocardial dysfunc-tion after resuscitation from cardiac arrest: an example ofglobal myocardial stunning. J Am Coll Cardiol 1996;28:232—40.

6. Kern KB. Postresuscitation myocardial dysfunction. Cardiol Clin2002;20:89—101.

7. Sun S, Weil MH, Tang W, Kamohara T, Klouche K. Delta-opioid receptor agonist reduces severity of postresuscitationmyocardial dysfunction. Am J Physiol Heart Circ Physiol2004;287:H969—74.

8. Tang W, Weil MH, Sun S, Noc M, Yang L, Gazmuri RJ. Epinephrineincreases the severity of postresuscitation myocardial dysfunc-tion. Circulation 1995;92:3089—93.

9. Ditchey RV, Lindenfeld J. Failure of epinephrine to improve thebalance between myocardial oxygen supply and demand dur-ing closed-chest resuscitation in dogs. Circulation 1988;78:382—9.

0. Lindner KH, Ahnefeld FW, Schuermann W, Bowdler IM.Epinephrine and norepinephrine in cardiopulmonary resuscita-tion. Effects on myocardial oxygen delivery and consumption.Chest 1990;97:1458—62.

1. Angelos MG, DeBehnke DJ. Epinephrine and high-flow reperfu-sion after cardiac arrest in a canine model. Ann Emerg Med1995;1926:208—15.

2. Berg RA, Otto CW, Kern KB, et al. High-dose epinephrine resultsin greater early mortality after resuscitation from prolongedcardiac arrest in pigs: a prospective, randomized study. CritCare Med 1994;22:282—90.

3. Duggal C, Weil MH, Tang W, Gazmuri RJ, Sun S. Effect of arresttime on the hemodynamic efficacy of precordial compression.Crit Care Med 1995;23:1233—6.

4. Abella BS, Zhao D, Alvarado J, Hamann K, Vanden Hoek TL,Becker LB. Intra-arrest cooling improves outcomes in a murinecardiac arrest model. Circulation 2004;109:2786—91.

5. American Heart Association (2005). Guidelines for cardiopul-monary resuscitation and emergency cardiovascular care.Circulation 2005;112(IV1):203.

6. Feigenbaum H. Echocardiography. Philadelphia, PA; 1993.7. Teichholz LE, Kreulen T, Herman MV, Gorlin R. Problems in

echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence of absence ofasynergy. Am J Cardiol 1976;37:7—11.

8. Peberdy MA, Kaye W, Ornato JP, et al. Cardiopulmonary resusci-tation of adults in the hospital: a report of 14720 cardiac arrestsfrom the National Registry of Cardiopulmonary Resuscitation.Resuscitation 2003;58:297—308.

9. Eisenberg MS, Mengert TJ. Cardiac resuscitation. N Engl J Med2001;344:1304—13.

0. Gazmuri RJ, Weil MH, Bisera J, Tang W, Fukui M, McKee D.Myocardial dysfunction after successful resuscitation from car-diac arrest. Crit Care Med 1996;24:992—1000.

1. Tibballs J, Kinney S. A prospective study of outcome ofin-patient paediatric cardiopulmonary arrest. Resuscitation2006;71:310—8.

2. Chang WT, Ma MH, Chien KL, et al. Postresuscitation myocardialdysfunction: correlated factors and prognostic implications.Intensive Care Med 2006. November 15 [epub ahead ofprint].

3. Fries M, Weil MH, Chang YT, Castillo C, Tang W. Microcircu-lation during cardiac arrest and resuscitation. Crit Care Med2006;34:S454—7.

4. McCaul CL, McNamara PJ, Engelberts D, et al. Epinephrineincreases mortality after brief asphyxial cardiac arrest in anin vivo rat model. Anesth Analg 2006;102:542—8.

5. Ong ME, Ornato JP, Edwards DP, et al. Use of an automated, load-distributing band chest compression device for out-of-hospital

1

2

2

10

cardiac arrest resuscitation. JAMA 2006;295(22):2629—37. June

14.

6. Pytte M, Kramer-Johansen J, Eilevstjonn J, et al. Haemody-namic effects of adrenaline (epinephrine) depend on chestcompression quality during cardiopulmonary resuscitation inpigs. Resuscitation 2006;71:369—78.

2

M.G. Angelos et al.

7. Klouch K, Weil MH, Tang W, et al. A selective alpha2-

adrenergic agonist for cardiac resuscitation. J Lab Clin Med2002;140:27—34.

8. Klouche K, Weil MH, Sun S, et al. A comparison of alpha-methylnorepinephrine, vasopressin and epinephrine for cardiacresuscitation. Resuscitation 2003;57:93—100.