Haemodynamic effects of adrenaline (epinephrine) depend on chest compression quality during cardiopulmonary resuscitation in pigs

Resuscitation (2006) 71, 369—378

EXPERIMENTAL PAPER

Haemodynamic effects of adrenaline (epinephrine) depend on chest compression quality during cardiopulmonary resuscitation in pigs夽 Morten Pytte a,b,∗, Jo Kramer-Johansen b,c, Joar Eilevstjønn d, Morten Eriksen b, Tævje A. Strømme b, Kristin Godang e, Lars Wik b,f, Petter Andreas Steen a, Kjetil Sunde a,b a

Department of Anaesthesiology, Ulleval University Hospital, Oslo, Norway Institute for Experimental Medical Research, Ulleval University Hospital, N-0407 Oslo, Norway c Norwegian Air Ambulance, Department of Research and Education in Acute Medicine, Drøbak, Norway d Laerdal Medical AS, Stavanger, Norway e Section of Endocrinology, National University Hospital, Oslo, Norway f National Competence Center for Emergency Medicine, Ulleval University Hospital, Oslo, Norway b

Received 6 April 2006 ; received in revised form 4 May 2006; accepted 10 May 2006 KEYWORDS Cardiopulmonary resuscitation; Epinephrine; Drugs; Ventricular fibrillation

Summary Background: Adrenaline (epinephrine) is used during cardiopulmonary resuscitation (CPR) based on animal experiments without supportive clinical data. Clinically CPR was reported recently to have much poorer quality than expected from international guidelines and what is generally done in laboratory experiments. We have studied the haemodynamic effects of adrenaline during CPR with good laboratory quality and with quality simulating clinical findings and the feasibility of monitoring these effects through VF waveform analysis. Methods and results: After 4 min of cardiac arrest, followed by 4 min of basic life support, 14 pigs were randomised to ClinicalCPR (intermittent manual chest compressions, compression-to-ventilation ratio 15:2, compression depth 30—38 mm) or LabCPR (continuous mechanical chest compressions, 12 ventilations/min, compression depth 45 mm). Adrenaline 0.02 mg/kg was administered 30 s thereafter. Plasma adrenaline concentration peaked earlier with LabCPR than with ClinicalCPR, median (range), 90 (30, 150) versus 150 (90, 270) s (p = 0.007), respectively. Coronary perfusion pressure (CPP) and cortical cerebral blood flow (CCBF) increased and femoral blood flow (FBF) decreased after adrenaline during LabCPR (mean differences

夽 A Spanish translated version of the summary of this article appears as Appendix in the online version at 10.1016/j.resuscitation.2006.05.003. ∗ Corresponding author. Fax: +47 23016799. E-mail address: [email protected] (M. Pytte).

0300-9572/$ — see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.resuscitation.2006.05.003

370

M. Pytte et al. (95% CI) CPP 17 (6, 29) mmHg (p = 0.01), FBF −5.0 (−8.8, −1.2) ml min−1 (p = 0.02) and median difference CCBF 12% of baseline (p = 0.04)). There were no significant effects during ClinicalCPR (mean differences (95% CI) CPP 4.7 (−3.2, 13) mmHg (p = 0.2), FBF −0.2 (−4.6, 4.2) ml min−1 (p = 0.9) and CCBF 3.6 (−1.8, 9.0)% of baseline (p = 0.15)). Slope VF waveform analysis reflected changes in CPP. Conclusion: Adrenaline improved haemodynamics during laboratory quality CPR in pigs, but not with quality simulating clinically reported CPR performance. © 2006 Elsevier Ireland Ltd. All rights reserved.

Introduction

Material and methods

Coronary perfusion pressure (CPP) above a certain threshold during cardiopulmonary resuscitation (CPR) predicts successful resuscitation and short term survival both in animal and clinical studies of cardiac arrest.1,2 Adrenaline (epinephrine) is used during CPR in an attempt to increase CPP via peripheral vasoconstriction3 as shown in animal experiments.4 Supportive data for the use of adrenaline in clinical settings are lacking, and some striking differences between experimental and clinical settings are evident. Experimental studies are performed under controlled conditions where chest compression depth, frequency and duration are defined, either mechanically5,6 or manually.7,8 In contrast, clinical studies lack control of CPR quality. It has only recently been possible to evaluate the quality of clinical CPR,9—11 where chest compressions were found to be too shallow10 with long hands-off periods without chest compressions and vital organ perfusion.9—11 Furthermore, in experimental studies, drugs are injected directly into the right atrium, while clinically drugs are usually delivered in a peripheral vein. This should influence the pharmacokinetics with a higher peak concentration achieved more rapidly in the laboratory.12 The discrepancy in CPR quality performed experimentally and clinically suggests that haemodynamics during clinical CPR may be inadequate and thereby negatively influence the drug effects. We investigated the haemodynamic effects of adrenaline during laboratory quality CPR (LabCPR) with continuous mechanical chest compressions, and clinical quality CPR (ClinicalCPR), performed manually, mimicking the CPR quality of recently published clinical findings.10 Furthermore, we investigated the feasibility of recognising the drug effects on CPP through analysis of the VF waveform. We hypothesised that drug delivery and drug effect depend on the quality of CPR.

Animal preparation The experiments were conducted in accordance with ‘‘Regulations on Animal Experimentation’’ under The Norwegian Animal Welfare Authority Act and approved by Norwegian Animal Research Authority. Seventeen healthy domestic pigs (27 ± 2.6 kg) of either sex were fasted over night with free access to water. They were sedated with an intramuscular injection of ketamine (30 mg kg−1 ) and atropine 1 mg. A catheter was inserted into an ear vein and anaesthesia was induced with bolus injections of propofol (1.5—2.0 mg kg−1 ) and fentanyl (0.25 mg) and maintained with continuous infusions of propofol (5—6 mg kg−1 h−1 ) and fentanyl (30—50 ␮g kg−1 h−1 ). During preparation physiological saline (30 ml kg−1 h−1 ) was administered. The pigs received a tracheotomy and were mechanically ventilated (Servo Ventilator 900 B, Siemens-Elema AB, Solna, Sweden) with ambient air at 15 breaths min−1 with positive end-expiratory pressure of 2—3 cm H2 O and tidal volume adjusted to maintain an end-tidal carbon dioxide (etCO2 ) level of 4.5—5.5 kPa as measured by the gas monitor (Datex Capnomac UltimaTM , Helsinki, Finland). Urine was drained continuously through a cystostoma, and the intra-abdominal temperature was maintained at 38.5—39.5 ◦ C throughout the experiment period using a heating pad placed under the animals. A 7F micro-tip pressure transducer catheter (Model SPC 470, Millar Instruments, Houston, TX, USA) was inserted through the right carotid artery and advanced to the proximal aorta just above the aortic valves for continuous arterial pressure monitoring. Another 7F micro-tip pressure transducer catheter (Millar Instruments) was introduced to the right atrium via the right external carotid vein. A 7.5F Swan-Ganz catheter (Baxter Healthcare Corporation, Irvine, CA, USA) was inserted into the right atrium via the right femoral vein for central venous blood sampling. Another fluid filled

Haemodynamic effects of adrenaline and different quality of chest compressions polyethylene catheter was inserted into the aorta through the right femoral artery for arterial blood sampling. An ultrasound flow meter probe (model 3SB880, Transonic Systems Inc., Ithaca, NY, USA) was applied to the left femoral artery. All invasive catheters were introduced using a cut down technique. A laser-doppler flowmetry probe (Model 407, Perimed AB, Stockholm, Sweden) was placed on the surface of the left cerebral cortex for continuous measurements of cortical blood flow as described previously by others.13—15 The technique gives an accurate real-time measurement of changes in volume flow in a small part of cerebral cortex, but is unsuitable for measurements of absolute flow levels.16 A craniotomy and duratomy of the skull was performed approximately 10 mm anterior of the coronal suture and 15 mm lateral to the sagittal suture. Care was taken to avoid placing the probe directly over visible vessels. The probe was secured to the surface of the brain with a probe holder (Model PH 07-4, Perimed AB, Stockholm, Sweden). To prevent artefacts, the probe was secured with dural sutures and the burr hole was sealed with bone wax. The readings were expressed in arbitrary perfusion units (PU). Pressures and flow signals were sampled using PC-based real time data acquisition hardware (DaqBoardsTM Model 200A, IOteck Inc., Cleveland, OH, USA) supported with DASYLab version 5.1 software (Datalog, National Instruments Company, Moenchengladback, Germany) and printed on an eight-channel thermal array recorder model TA11 (Gould Instruments Systems Inc., Ohio, USA).

Clinical quality CPR Clinical quality CPR, similar to previous clinical findings,10 was achieved with manually performed chest compressions (30—38 mm depth) with a frequency of 100 min−1 interrupted by a 9 s break every 15 compressions, during which two manual ventilations were delivered. This resulted in a 50% hands-off ratio, similar to clinical findings.10 A sternal chest pad was mounted on the lower part of the shaved sternum with double adhesive tape. The chest pad sensor provided acceleration and force signals enabling accurate measurements of compression depth, rate, and compression/decompression duty cycle, which were all recorded by a modified HeartStart 4000SP defibrillator (Philips Medical Systems, Andover, MA and Laerdal Medical AS, Stavanger, Norway). The method has previously been validated in a manikin model17 and also used in the clinical study by Wik et al.10 and Albella et al.18 Compression

371

depth and rate was continuously displayed on the screen. Two lines at 30 and 38 mm served as target for the single person who performed the chest compressions.

Laboratory quality CPR LabCPR was achieved with a previously described modified automatic hydraulic electrically driven chest compression device (Heartsaver 2000, Medreco, Bodø, Norway)6,15 maintaining consistent chest compressions of 45 mm depth, a fixed rate of 100 min−1 , and equal compression—relaxation phases (duty cycle 50%). The rate of manual ventilations was 12 min−1 .

VF analysis Standard lead II of the surface ECG was monitored and recorded via the HeartStart 4000SP defibrillator using a sample rate of 500 Hz and 16 bit resolution (1.031 ␮V/LSB). VF analysis was performed offline in MATLAB (The Math-Works Inc., Natick, MA) with the ECG down sampled to 200 Hz and pre-processed using a comb filter to remove dc and 50 Hz mains noise. Compression artefacts were filtered from the ECG using the MC-RAMP adaptive filter.19,20 Median slope (Slope) was used for VF analysis. It is defined as Median slope = median (|x(n) − x(n − 1)|)fs (mV/s) where x(n) is an ECG sample in mV in a block of L samples and fs is the sampling frequency in Hz, n = 1. . .L. A large block size L of 2000 samples (10 s) was used to smooth short time variations of slope. Median slope is reported to predict successful defibrillation defined as return of spontaneous circulation with 95% sensitivity and 50% specificity (Neurauter A. MEng, submitted data, 2005), comparable to other well used VF predictors.21,22

Experimental protocol After instrumentation, baseline measurements were obtained for all variables. The IV infusion, heating and ventilation were thereafter discontinued. VF was induced by a direct current (90 V) applied transthoracically for 3 s and confirmed by ECG changes and an abrupt fall in arterial blood pressure. After 4 min of untreated VF, mechanical CPR was started with compression depth 45 mm and 9 s pause for two ventilations with ambient air every 15th compression, simulating basic life support (BLS).

372

M. Pytte et al.

Figure 1 Graphical display of the experimental setup. After 4 min of untreated ventricular fibrillation (VF), cardiopulmonary resuscitation (CPR) was started with 15 mechanical chest compressions interrupted by two ventilations with ambient air (basic life support; BLS) for 4 min. At 8 min the pigs were randomised to receive continuous mechanical (LabCPR) or intermittent manual chest compressions (ClinicalCPR) for 6 min (advanced life support; ALS). Adrenaline was administered at 8.5 min in both groups.

Eight minutes after induction of VF the animals were randomised to receive either LabCPR or ClinicalCPR for 6 min. Adrenaline was administered in an ear vein in both groups 30 s after initiation of advanced life support (ALS). Administration of adrenaline 0.02 mg/kg was followed by a 10 ml flush and a continuous infusion of saline. Ventilations were given with a resuscitator bag (Laerdal Medical, Stavanger, Norway) filled with oxygen at a rate of 10 L/min. The experimental setup is illustrated in Figure 1. Cerebral cortical flow (CCBF), femoral artery blood flow (FBF) and blood pressures were recorded continuously throughout the experimental period. At the end of the experiment, compressions were stopped. An autopsy was performed on all animals to check for organ damage and verification of catheter placements.

Blood sampling and analysis Arterial blood gas specimens were obtained from the liquid filled femoral artery catheter and venous specimens from the right atrium at baseline and 13 min after cardiac arrest and placed on ice prior to analysis (AVL OMNITM 9, AVL List GmbH Medizintechnik, Graz, Austria). Specimens for adrenaline analysis were obtained from the right femoral artery at baseline and 7, 9—13, and 14 min post-VF (1 min before and each minute after ALS initiation) and collected in S-Monovette® containers (Sarstedt, Germany) treated with EGTAgluthatione. The samples were placed on ice before plasma was separated by centrifugation (3000 rpm, 10 min, 4 ◦ C) and frozen at −80 ◦ C until assayed. The blood samples were analysed for plasma noradrenaline (norepinephrine, NA) and adrenaline (A) by high performance liquid-

chromatography (HPLC) with a reverse phase column and glassy carbon electrochemical detector (Agilent Technologies, Colorado, USA), using a commercial kit (Chromsystems, M¨ unchen, Germany). In our hands, the intra- and interassay variations were 10.7% and 14.5% for NE and 23.2% and 10.5% for A respectively. The detection limit was 5.46 pmol/l.

Calculations All haemodynamic data were sampled with a frequency of 200 Hz and analysed by a specially made procedure in a mathematics software package (MATLAB) Coronary perfusion pressure (CPP) was calculated as the peak difference between the thoracic aortic and right atrial pressures in the relaxation phase. Femoral arterial blood flow was calculated as the mean flow signal obtained for 18 s periods. CCBF was calculated as a fraction of baseline flow. Normalised values were calculated by subtracting the mean of the sample from the value divided by the sample standard deviation. Area under the concentration curve (AUC) was calculated by the trapezoid rule. Plasma adrenaline concentrations were plotted on a semi-logarithmic scale. The slope of the straight line formed by the plotted data determined the elimination coefficient (K). Halflife (t1/2 ) of the drug was calculated from the equation t1/2 = 0.693 K−1 .

Statistical analysis SPSS version 12.0 was used for statistical analysis (SPSS Inc., Chicago, IL). Data are given as median (range) when skewed and as mean (standard deviation) when normally distributed. Mann—Whitney U test was used to determine differences between the

Haemodynamic effects of adrenaline and different quality of chest compressions Table 1

373

Mean values at baseline and 13 min after induction of cardiac arrest

Baseline

ClinicalCPR (n = 7)

LabCPR (n = 7)

p-Value

Weight (range) (kg) MAP (S.D.) (mmHg) FBF (S.D.) (ml min−1 ) Adrenaline (S.D.) (pmol) pH (S.D.) pCO2 (S.D.) (kPa) pO2 (S.D.) (kPa) BE (S.D.) (mmol l−1 )

27 (23—30) 82 (13) 95 (38) 488 (253) 7.49 (0.04) 4.6 (0.35) 11.7 (1.3) 3.2 (2.2)

28 (24—32) 83 (16) 89 (17) 543 (223) 7.47 (0.02) 4.7 (0.5) 10.8 (2.4) 2.6 (1.9)

0.6 0.9 0.95 0.70 0.27 0.68 0.38 0.55

Thirteen minutes after cardiac arrest pH (S.D.) pCO2 (S.D.) (kPa) pO2 (S.D.) (kPa) BE (S.D.) (mmol l−1 )

7.33 (0.14) 5.0 (3.6) 28.3 (25.7) −8.1 (1.4)

7.30 (0.16) 4.1 (2.0) 34.5 (23.3) −11.0 (4.0)

0.76 0.57 0.65 0.09

Mean arterial pressure (MAP), femoral artery blood flow (FBF), plasma adrenaline, arterial blood pH, carbon dioxide tension (pCO2 ), oxygen tension (pO2 ) and base excess (BE) in pigs submitted to clinically relevant CPR (ClinicalCPR) and laboratory relevant CPR (LabCPR); CPR, cardiopulmonary resuscitation.

groups. Paired t-test and Wilcoxon matched pairs signed rank test were used for determining within individual differences, as appropriate. Differences are given as mean (95% confidence interval) or median. Pearson’s correlation coefficient was used to describe the degree of association between CPP and SLOPE, based on independent normalised values. We considered values of p < 0.05 statistically significant.

Results Ventricular fibrillation was obtained in 17 pigs. Three pigs were excluded from the study due to a failure of the chest compression device in the early phase of the study (piston locked in the compression phase), causing tension pneumothorax, haemothorax and rupture of the liver with abdominal haemorrhage. After repair the machine functioned well without visceral damage found at autopsy in the remaining 14 pigs. Due to a noisy signal, FBF was omitted in two pigs. There were no significant differences in haemodynamic data or plasma adrenaline concentrations between the two groups at baseline (Table 1) or during BLS.

Plasma adrenaline concentration Endogenous levels of catecholamines prior to adrenaline administration increased to 545 (154, 1539) and 493 (221, 2226) times the pre-arrest value for adrenaline (p = 0.937), and 509 (188, 1872) and 1209 (265, 1945) times the pre-arrest values for noradrenaline (p = 0.240) during ClinicalCPR versus LabCPR, respectively. The median time from delivery of exogenous adrenaline until a measured peak in plasma concentration was longer during ClinicalCPR versus LabCPR, 150 (90, 270) s versus 90 (30, 150) s (p = 0.007), respectively, Figure 2. The median peak plasma concentration was 3560 (1551, 7381) nM and 4182 (9442, 2962) nM in ClinicalCPR versus LabCPR, respectively (p = 0.38). Similarly, median half-life of plasma adrenaline was 1.9 (1.1, 6.3) and 1.1 (0.8,

Chest compression quality Mean compression depth was 33 (1.0) versus 45 mm, mean duty cycle was 34 (2.0)% versus 50%, mean compression frequency was 100 (3.0) versus 100 min−1 and mean fraction of time without chest compressions was 0.51 (0.04) versus 0.08 during ClinicalCPR versus LabCPR, respectively.

Figure 2 Box plot of the time from administration of adrenaline to a peak plasma concentration was measured in the LabCPR and ClinicalCPR groups.

0.01 0.02 0.04 (6, 29) (−8.8, −1.2) 17 −5.0 12* 29 (15, 38) 6.1 (3.1, 12) 39 (23, 102) CPP (mmHg) (n = 7) FBF (ml min−1 ) (n = 7) CCBF (%) of baseline (n = 7) LabCPR

39 (32, 67) 1.2 (0.4, 3.3) 45 (33, 160)

0.2 0.90 0.15 (−3.2, 13) (−4.6, 4.2) (−1.8, 9.0) 4.7 −0.2 3.6 12.3 (5.5, 51) 2.5 (1.0, 5.5) 35 (19, 77)

p-Value 95% Confidence interval Mean difference

12 (2.0, 36) 1.3 (0.6, 8.2) 36 (14, 64)

The present study was designed to observe the effects of adrenaline during different CPR quality. We found that CPR quality greatly influenced the pharmacokinetics and pharmacodynamics of adrenaline injected intravenously in a peripheral vein during CPR. While adrenaline given during good quality chest compressions caused peripheral vasoconstriction with a parallel increase in CPP and cortical cerebral blood flow, no significant haemodynamic effects were seen when given during chest compressions simulating the quality reported during ALS.10 Similarly, plasma adrenaline concentrations peaked significantly later in the manual group with a significantly longer half-life. Adrenaline has been used since the first Standards for CPR were published,23 based on animal experiments showing elevated CPP,24 increased blood flow to vital organs25,26 and improved defibrillation success and survival after injection of exogenous adrenaline.1 No supporting clinical evidence has been published, however, and haemodynamic effects of the recommended repetitive

CPP (mmHg) (n = 7) FBF (ml min−1 ) (n = 5) CCBF (%) of baseline (n = 7)

Discussion

ClinicalCPR

The median values of CPP and Slope in the ClinicalCPR and LabCPR groups (n = 14) are illustrated in (Figure 4). Among the 10 pigs with an increase in CPP after adrenaline injection, mean Slope was significantly higher at the time of peak drug concentration than prior to drug administration, 8.8 (2.6) versus 7.1 (1.5), mean difference 1.7 (0.5, 3.0) (p = 0.011). The mean values of CPP and Slope in each pig are presented in a scatter plot (Figure 5). Pearson’s correlation coefficient was 0.65 (p = 0.012) including all pigs (n = 14), and 0.77 (p = 0.002) when one outlier was excluded.

Median value (range) at peak (p-adrenaline)

Slope VF analysis

Median value (range) prior to administration of adrenaline

CPP and CCBF were significantly higher and FBF significantly lower at the time of peak plasma adrenaline concentration compared to the values prior to adrenaline administration in the LabCPR group. Peak CPP was seen at median 69 (53, 83) s after drug administration. No significant drug effects were seen during ClinicalCPR (Table 2 and Figure 3).

Measured variable

Haemodynamic drug effects

Group

2.9) min (p = 0.48), respectively. There was no significant difference in area under the concentration curve.

M. Pytte et al. Table 2 Median values of coronary perfusion pressure (CPP), femoral artery blood flow (FBF), and cortical cerebral blood flow (CCBF) prior to administration of adrenaline and at the time of peak plasma concentration of adrenaline with mean (*median) differences in the ClinicalCPR and LabCPR groups

374

Haemodynamic effects of adrenaline and different quality of chest compressions

375

Figure 3 The figures illustrate haemodynamic measurements (coronary perfusion pressure, femoral blood flow and cortical cerebral blood flow) in each pig prior to administration of adrenaline (A) and at peak plasma adrenaline concentration (B) in the LabCPR and ClinicalCPR groups.

doses every three to 5 min are questionable even in animals.7,24,27,28 The optimal dose of adrenaline during CPR in cardiac arrest patients is a question of debate. The international guidelines for resuscitation recommend 1 mg adrenaline intravenously (0.014 mg kg−1 in a 70 kg man).3 Randomised clinical studies have failed to show any beneficial effect of higher doses of adrenaline.29 Some experimental studies have shown significantly increased myocardial blood flow and significantly increased number of ROSC with high dose (0.045—0.2 mg kg−1 ) versus standard dose (0.015—0.02 mg kg−1 ) adrenaline.30,31 When studying survival, however, high dose of adrenaline

Figure 4 The figure shows median normalised values (mean of sample subtracted from the value divided by the sample standard deviation) of coronary perfusion pressure (CPP) and Slope in the LabCPR (above) and ClinicalCPR (below) groups against a timeline. Error bars are omitted for clarity. The vertical line denotes the time of randomisation. Adrenaline was administered at 30 s.

Figure 5 Mean value of coronary perfusion pressure and Slope value (normalised values) presented as scatter plot (n = 14). The regression line is based on 13 pigs, excluding one outlier ().

376 (0.2 mg kg−1 versus 0.02 mg kg−1 ) was associated with increased mortality in pigs.32 In the present study we have simulated clinically relevant CPR and thus used the standard recommended dose of adrenaline (0.02 mg kg−1 ), also used in other experimental8,26,33 and clinical34 studies. In experimental studies vasoconstrictive drugs are usually delivered via a central line to the right atrium during controlled chest compressions of good quality.1,7,24,25,28 In the clinical setting adrenaline is usually delivered via a peripheral line, and the quality of CPR is documented in several studies to be much poorer.9—11 In a recent clinical study chest compressions were more shallow than guidelines recommendations and only performed half the time during ALS.10 In previous experimental studies drugs delivered via a central line reached a higher peak concentration with significantly shorter time to both first appearance and peak concentration than for peripheral injection.12 Based on differences in injection routes and quality of CPR in animal experiments 7,25,27 versus clinical manual ALS,34,35 positive effects of vasoconstrictors in animal experiments can probably not be extrapolated to the clinical situation. These factors might indeed partly explain why clinical trials of vasopressin34,35 have failed to confirm convincing experimental findings.1,33,36 Defibrillation success correlates with CPP in experimental and clinical studies,1,2 and clinically the majority of shocks do not result in spontaneous circulation.9,37 The interruptions in chest compressions in connection with unsuccessful defibrillation attempts are likely to cause further deterioration of resuscitability.5,38 To identify the optimal timing of defibrillation, VF waveform analysis may be useful as it has been reported to predict successful defibrillation and correlate with CPP and myocardial blood flow during CPR.1,21,22,39 In the present study the median VF waveform derived Slope increased after adrenaline administration as reported in animal studies for other VF shock success predictors,1 and Slope correlated with CPP both in the group simulating clinical CPR and laboratory CPR. Slope and CPP reached maximum between 1 and 2 min after peripheral adrenaline injection during good quality CPR, which fit well with the present 2 min time interval in the new guidelines for CPR.40 However, prospective studies are required to determine the feasibility of VF waveform analysis in determining the optimal timing of defibrillation. Some limitations to the study should be noted, including the small sample size that may inhibit us from detecting differences that are present (type II error). Also, animal data may not be transferable to humans. The study was not designed to evaluate the

M. Pytte et al. effectiveness of mechanical versus manual chest compressions, but rather to evaluate the haemodynamic actions of adrenaline with different chest compression quality. A mechanical device was used for LabCPR because it is the standard CPR technique in our and other laboratories to ensure consistent, good quality CPR.5,6,15 In the ClinicalCPR group a compression:ventilation ratio of 15:2 was used with 50% hands-off time. The 2005 Guidelines3 with compression:ventilation ratio 30:2, one defibrillation attempt followed by immediate chest compressions should hopefully decrease hands-off time. It is, however, important to note that there was 48% hands-off time in the recent clinical study10 where most patients were intubated and should have received continuous chest compressions without a break for ventilations also according to the previous guidelines. Thus our findings illustrate the importance of CPR quality. Furthermore, survival data are not available as the study was designed to closely observe the haemodynamic effects of adrenaline over a defined time period after peripheral injection during different quality of CPR. Finally, the majority of patients with cardiac arrest have acute or chronic heart disease, whereas our pigs were healthy. How this affects pharmacokinetics and dynamics is not known.

Conclusions In conclusion the haemodynamic effects of adrenaline depend on chest compression quality. The delivery of adrenaline was significantly delayed when simulating clinically reported CPR quality compared to good quality CPR. Adrenaline improved haemodynamics during good quality CPR in pigs, but not with quality simulating clinically reported CPR performance. Changes in CPP after administration of adrenaline during CPR may be monitored though analysis of the VF waveform.

Acknowledgements This work was supported in part by the Laerdal Foundation for Acute Medicine, the Ulleval University Hospital Scientific Advisory Council, the Norwegian Air Ambulance, Health Region East and Anders Jahre’s Foundation.

References 1. Strohmenger HU, Lindner KH, Prengel AW, Pfenninger EG, Bothner U, Lurie KG. Effects of epinephrine and vasopressin

Haemodynamic effects of adrenaline and different quality of chest compressions

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

on median fibrillation frequency and defibrillation success in a porcine model of cardiopulmonary resuscitation. Resuscitation 1996;31:65—73. Paradis NA, Martin GB, Rivers EP, et al. Coronary perfusion pressure and the return of spontaneous circulation in human cardiopulmonary resuscitation. JAMA 1990;263:1106— 13. Nolan JP, Deakin CD, Soar J, Bottiger BW, Smith G. European Resuscitation Council Guidelines for Resuscitation 2005 Section 4. Adult advanced life support. Resuscitation 2005;67(Suppl 1):S39—86. 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—35. Yu T, Weil MH, Tang W, et al. Adverse outcomes of interrupted precordial compression during automated defibrillation. Circulation 2002;106:368—72. Sunde K, Wik L, Naess P, Grund F, Nicolaysen G, Steen PA. Improved haemodynamics with increased compression—decompression rates during ACD-CPR in pigs. Resuscitation 1998;39:197—205. Wenzel V, Lindner KH, Krismer AC, Miller EA, Voelckel WG, Lingnau W. Repeated administration of vasopressin but not epinephrine maintains coronary perfusion pressure after early and late administration during prolonged cardiopulmonary resuscitation in pigs. Circulation 1999;99: 1379—84. Berg RA, Hilwig RW, Kern KB, Ewy GA. Precountershock cardiopulmonary resuscitation improves ventricular fibrillation median frequency and myocardial readiness for successful defibrillation from prolonged ventricular fibrillation: a randomized, controlled swine study. Ann Emerg Med 2002;40:563—70. Sunde K, Eftestol T, Askenberg C, Steen PA. Quality assessment of defribrillation and advanced life support using data from the medical control module of the defibrillator. Resuscitation 1999;41:237—47. Wik L, Kramer-Johansen J, Myklebust H, et al. Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA 2005;293:299—304. Valenzuela TD, Kern KB, Clark LL, et al. Interruptions of chest compressions during emergency medical systems resuscitation. Circulation 2005;112:1259—65. Talit U, Braun S, Halkin H, Shargorodsky B, Laniado S. Pharmacokinetic differences between peripheral and central drug administration during cardiopulmonary resuscitation. J Am Coll Cardiol 1985;6:1073—7. Muir JK, Boerschel M, Ellis EF. Continuous monitoring of posttraumatic cerebral blood flow using laser-Doppler flowmetry. J Neurotrauma 1992;9:355—62. Rubertsson S, Karlsten R. Increased cortical cerebral blood flow with LUCAS; a new device for mechanical chest compressions compared to standard external compressions during experimental cardiopulmonary resuscitation. Resuscitation 2005;65:357—63. Dorph E, Wik L, Stromme TA, Eriksen M, Steen PA. Oxygen delivery and return of spontaneous circulation with ventilation:compression ratio 2:30 versus chest compressions only CPR in pigs. Resuscitation 2004;60:309—18. Kirkeby OJ, Rise IR, Nordsletten L, Skjeldal S, Hall C, Risoe C. Cerebral blood flow measured with intracerebral laserDoppler flow probes and radioactive microspheres. J Appl Physiol 1995;79:1479—86. Aase SO, Myklebust H. Compression depth estimation for CPR quality assessment using DSP on accelerometer signals. IEEE Trans Biomed Eng 2002;49:263—8.

377

18. Abella BS, Alvarado JP, Myklebust H, et al. Quality of cardiopulmonary resuscitation during in-hospital cardiac arrest. JAMA 2005;293:305—10. 19. Husoy JH, Eilevstjonn J, Eftestol T, Aase SO, Myklebust H, Steen PA. Removal of cardiopulmonary resuscitation artifacts from human ECG using an efficient matching pursuitlike algorithm. IEEE Trans Bio-Med Eng 2002;49:1287— 98. 20. Eilevstjonn J, Eftestol T, Aase SO, Myklebust H, Husoy JH, Steen PA. Feasibility of shock advice analysis during CPR through removal of CPR artefacts from the human ECG. Resuscitation 2004;61:131—41. 21. Eftestol T, Sunde K, Ole Aase S, Husoy JH, Steen PA. Predicting outcome of defibrillation by spectral characterization and nonparametric classification of ventricular fibrillation in patients with out-of-hospital cardiac arrest. Circulation 2000;102:1523—9. 22. Noc M, Weil MH, Tang W, Sun S, Pernat A, Bisera J. Electrocardiographic prediction of the success of cardiac resuscitation. Crit Care Med 1999;27:708—14. 23. Standards for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). 3. Advanced life support. JAMA 1974;227:852—60. 24. Cairns CB, Niemann JT. Hemodynamic effects of repeated doses of epinephrine after prolonged cardiac arrest and CPR: preliminary observations in an animal model. Resuscitation 1998;36:181—5. 25. Lindner KH, Strohmenger HU, Prengel AW, Ensinger H, Goertz A, Weichel T. Hemodynamic and metabolic effects of epinephrine during cardiopulmonary resuscitation in a pig model. Crit Care Med 1992;20:1020—6. 26. Johansson J, Gedeborg R, Basu S, Rubertsson S. Increased cortical cerebral blood flow by continuous infusion of adrenaline (epinephrine) during experimental cardiopulmonary resuscitation. Resuscitation 2003;57:299— 307. 27. Stadlbauer KH, Wagner-Berger HG, Wenzel V, et al. Survival with full neurologic recovery after prolonged cardiopulmonary resuscitation with a combination of vasopressin and epinephrine in pigs. Anesth Analg 2003;96:1743—9. 28. Klouche K, Weil MH, Sun S, Tang W, Zhao DH. A comparison of alpha-methylnorepinephrine, vasopressin and epinephrine for cardiac resuscitation. Resuscitation 2003;57:93—100. 29. Gueugniaud PY, Mols P, Goldstein P, et al. A comparison of repeated high doses and repeated standard doses of epinephrine for cardiac arrest outside the hospital. N Engl J Med 1998;339:1595—601. 30. Brown CG, Werman HA, Davis EA, Hobson J, Hamlin RL. The effects of graded doses of epinephrine on regional myocardial blood flow during cardiopulmonary resuscitation in swine. Circulation 1987;75:491—7. 31. Lindner KH, Ahnefeld FW, Bowdler IM. Comparison of different doses of epinephrine on myocardial perfusion and resuscitation success during cardiopulmonary resuscitation in a pig model. Am J Emerg Med 1991;9:27—31. 32. Berg RA, Otto CW, Kern KB, et al. High-dose epinephrine results in greater early mortality after resuscitation from prolonged cardiac arrest in pigs: a prospective, randomized study. Crit Care Med 1994;22:282—90. 33. Babar SI, Berg RA, Hilwig RW, Kern KB, Ewy GA. Vasopressin versus epinephrine during cardiopulmonary resuscitation: a randomized swine outcome study. Resuscitation 1999;41:185—92. 34. Wenzel V, Krismer AC, Arntz HR, Sitter H, Stadlbauer KH, Lindner KH. A comparison of vasopressin and epinephrine for out-of-hospital cardiopulmonary resuscitation. N Engl J Med 2004;350:105—13.

378 35. Stiell IG, Hebert PC, Wells GA, et al. Vasopressin versus epinephrine for inhospital cardiac arrest: a randomised controlled trial. Lancet 2001;358:105—9. 36. Lindner KH, Prengel AW, Pfenninger EG, et al. Vasopressin improves vital organ blood flow during closedchest cardiopulmonary resuscitation in pigs. Circulation 1995;91:215—21. 37. Rea TD, Shah S, Kudenchuk PJ, Copass MK, Cobb LA. Automated external defibrillators: to what extent does the algorithm delay CPR? Ann Emerg Med 2005;46:132—41. 38. Eftestol T, Sunde K, Steen PA. Effects of interrupting precordial compressions on the calculated probability of defibril-

M. Pytte et al. lation success during out-of-hospital cardiac arrest. Circulation 2002;105:2270—3. 39. Strohmenger HU, Eftestol T, Sunde K, et al. The predictive value of ventricular fibrillation electrocardiogram signal frequency and amplitude variables in patients with out-of-hospital cardiac arrest. Anesth Analg 2001;93:1428— 33. 40. 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 7.2. Management of Cardiac Arrest. Circulation 2005;112:IV58—66.