Original Contribution

The natural biochemical changes during ventricular fibrillation with cardiopulmonary resuscitation and the onset of postdefibrillation pulseless electrical activity^B

Leslie A. Geddes PhD^a,*, Rebecca A. Roeder PhD^a, Ann E. Rundell PhD^a, Michael P. Otlewski BS^b, Andre E. Kemeny BS^b, Aaron E. Lottes BS, MBA^a

^aWeldon School of Biomedical Engineering, Purdue University, West Lafayette, IN 47907-2022, USA

^bDepartment of Basic Medical Sciences, Purdue University, West Lafayette, IN 47907-2022, USA

Received 24 August 2005; accepted 25 January 2006

Abstract

Objective: The objective of this study was to document the biochemical changes during ventricular fibrillation (VF) with cardiopulmonary resuscitation (CPR), and to identify factors associated with postdefibrillation pulseless electrical activity (PD-PEA).

Background: It has been reliably estimated that as much as 60% of out-of-hospital sudden cardiac death can be attributed to the onset of PD-PEA (Niemann JT, Cruz B, Garner D et al. Immediate countershock versus CPR before countershock in a 5-minute swine model of ventricular fibrillation arrest. Ann Emerg Med 2000;36:543-6). Previous attempts to treat Reversible causes of pulseless electrical activity have not been successful clinically (Niemann JT, Stratton SJ, Cruz B, Lewis RJ. Outcome of out-of-hospital postcountershock asystole and pulseless electrical activity versus primary asystole and pulseless electrical activity. Crit Care Med 2001;29:2366-70).

Methods: This investigation used 22 studies on 14 anesthetized pigs breathing 100% oxygen. Ventricular fibrillation was induced with a right ventricular catheter electrode, and the chest was compressed with a pneumatically driven Chest Thumper (Michigan Instruments) (80-100 lb at 60/min). The electrocardiogram and aortic pressure were recorded continuously. arterial pH, Po₂, Pco₂, Na⁺, K⁺, Ca²⁺, Cl^—, SaO₂, glucose, hematocrit, and hemoglobin level were measured at selected times. Ventricular defibrillation was achieved with transchest electrodes.

Results: Typically, during VF with CPR, mean aortic pressure was 20 to 25 mm Hg. In all cases aortic Po₂ decreased to about 20% of the initial value in 10 minutes, and aortic blood K⁺ increased by 50% in 6 minutes. By 5 to 8 minutes, the incidence of PD-PEA was 50%.

Conclusion: Ventricular fibrillation duration, arterial K⁺, and arterial Pco₂ were statistically correlated with the onset of PD-PEA in this study. In addition, trends suggest an association of mean arterial blood pressure and arterial Po₂ with the onset of PD-PEA.

D 2006

^B This study was supported in part by NIBIB grant NGAR2IEB001540 National Institutes of Health, Bethesda, MD, and a grant from the Purdue Trask Fund (Award #64333).

* Corresponding author. Tel.: +1 765 494 2997; fax: +1 765 494 1193.

E-mail address: [email protected] (L.A. Geddes).

0735-6757/$ - see front matter D 2006 doi:10.1016/j.ajem.2006.01.030

Introduction

Although cardiopulmonary resuscitation (CPR) increases survival from ventricular fibrillation (VF), it is estimated that approximately 95% of victims of sudden cardiac arrest will die before reaching a hospital [1]. In addition, the percentage of patients who are discharged from the hospital after an episode of out-of-hospital cardiac arrest is distur- bingly low [2]. The rate of survival from VF has been estimated to decrease approximately 10% for every minute that CPR is not provided [3]. Patients who suffer post- defibrillation pulseless electrical activity (PD-PEA) have a particularly poor prognosis. It has been estimated that as much as 60% of out-of-hospital sudden cardiac deaths can be attributed to the onset of PD-PEA [4]. Subjects who suffer inhospital PD-PEA have a survival rate of less than 15% [5]. Successful treatment of this condition could save a substantial number of lives, estimated as high as 25000 per year in the United States. It was the objective of this study to document the natural biochemical changes that occur during VF with CPR and to identify those factors associated with PD-PEA.

Methods

All studies were performed with the approval of the Purdue Animal Care and Use Committee. In this investiga- tion, 14 fluothane-anesthetized pigs (20-30 kg) were used to conduct 22 different studies. All animals were intubated and breathed 100% oxygen. The electrocardiogram (ECG), aortic pressure, end-tidal CO₂, and aortic blood chemistries (pH, Po₂, Pco₂, HCO^—, Na⁺, K⁺, Ca²⁺, Cl^—, SaO₂, glucose, hematocrit, and hemoglobin) were measured. Blood chem- istries for 11 animals were measured with the ISTAT (Heska Corp, Loveland, CO) analyzer and, for 3 animals, with the GEM (Mallinckrodt, Hazelwood, MO) analyzer (no Cl^— or hemoglobin readings were available from this analyzer).

3

Fig. 1 Normalized mean blood pressure values with SEMs. The value at the onset (t = 0) of VF for each animal was used as the reference value and ranged from 40 to 65 mm Hg.

Fig. 2 Normalized Po₂ and SaO₂ values with SEMs. The value of each measurement at the onset (t = 0) of VF for each animal was used as the reference value: 125 to 475 mm Hg for Po₂ and 91% to 100% for SaO₂.

Ventricular fibrillation (VF) was induced electrically with a right-ventricular, bipolar catheter electrode. Cardiopulmo- nary resuscitation was applied with the Chest Thumper (Michigan Instruments). The force of chest compression was controlled (80-100 lb); the compression rate was 60/min with a Duty cycle of 50%. Defibrillation was accomplished with transchest electrodes. If, after defibrillation, SA node arrest or atrioventricular block occurred, the ventricles were paced with the right-ventricular catheter electrode. In this way, it was possible to test for PD-PEA in all trials. We were particularly interested in the biochemistry values and VF duration associated with the onset of PD-PEA.

Fig. 3 Normalized K⁺ and iCa²⁺ values with SEMs. The value for each at the onset (t = 0) of VF for each animal was used as the reference value: 2.8 to 5.75 mmol/L for K⁺ and 1.25 to 1.53 mmol/ L for iCa²⁺.

Fig. 4 Normalized Pco₂, HCO^—, and pH values with SEMs. The value for each at the onset (t = 0) of VF for each animal was used as the reference value: 35 to 87 mm Hg for Pco₂, 27.5 to 37.5 mmol/L for HCO^—, and 7.22 to 7.60 for pH.

3

3

If K⁺ was elevated before the beginning of a study, rapidly acting insulin and dextrose 5% in water were administered intravenously to bring the K⁺ back to a normal level (about 4 mEq/L). The K⁺ level was then allowed to stabilize before beginning a study. Dextrose 5% in water was given intravenously if necessary to maintain a stable glucose level (about 85 mg/dL).

Electrocardiogram and aortic blood pressure were recorded continuously as a function of time using Labview software and a strip-chart recorder. All data were entered into a computer (Labview) for offline processing. For scenarios where 3 or more subjects were studied, normal- ized plots of the data were generated to show trends more clearly. The normalized values are presented as ratios with the zero time values as the denominator in each case. Standard error of the mean (SEM) bars appear on these plots. A 1-tailed t test was used to analyze correlations of the onset of PD-PEA with the various Biochemical parameters for significance ( P b .05).

Results

After the induction of VF, the mean aortic Blood pressure dropped rapidly to a pressure sustained with CPR of about 20 to 25 mm Hg. Fig. 1 shows the normalized (ratio) values for mean aortic pressure during VF with CPR, with the prefibrillation mean pressures ranging from 40 to 65 mm Hg. The impact of VF with CPR on the aortic Po₂ level was monitored using both the Po₂ and SaO₂ values. Fig. 2 shows the normalized Po₂ and SaO₂ values with SEM bars. Po₂ initial values ranged from 125 to 475 mm Hg, and SaO₂

decreased to about 45% of the prefibrillation (t = 0) value. The SaO₂ did not decrease appreciably until after 6 minutes of VF with CPR, signifying that SaO₂ is not a sensitive indicator of approaching hypoxia.

The normal baseline K⁺ range is 3.5 to 5.0 mEq/L in pigs. During prolonged VF with CPR, the K⁺ values increased dramatically. Fig. 3 shows normalized K⁺ and normalized Ca²⁺values with SEM bars. There was a steady increase in K⁺ of nearly 10% per minute of VF with CPR. The reference values ranged from 2.80 to 5.75 mEq/L for K⁺ and 1.25 to 1.53 mEq/L for Ca²⁺, which remained virtually constant during VF with CPR.

As indicated by the normalized pH values in Fig. 4, with reference values ranging from 7.22 to 7.60, the pH decreased slightly, but the values remained virtually constant during extended periods of VF with CPR. Related to the pH are the arterial blood-gas measurements of Pco₂ and HCO^—, and so their normalized values with SEM bars are also shown in Fig. 4, calculated from initial values of 35 to 97 mm Hg of Pco₂ and 27.5 to 37.5 mEq/L of HCO^—. It is evident that Pco₂ is relatively constant, with a slight increase at longer durations of VF with CPR. There is an eventual drop of about 20% in HCO^— over long durations of VF with CPR. The concentrations of sodium, chloride, glucose, hemo- globin, and hematocrit changed little during VF with CPR.

3

3

3

Therefore, no plots of these quantities are presented here.

1-4 2		7	22%	9
5-8	2	2	50%	4
N8	4	0	100%	4
K+ (mEq/L)
3.0-4.9	2	4	33%	6
5.0-6.9	4	2	67%	6
z7.0	2	0	100%	2
Pco2 (mm Hg)
N60	1	4	20%	5
50-60	2	2	50%	4
b50	5	2	71%	7
Average mean BP (mm Hg)
N20	3	6	33%	9
b20	4	2	67%	6
Po2 (mm Hg) N300	1	2	33%	3
100-300	3	4	43%	7
b100	3	2	60%	5
pH b7.3	3	3	50%	6
N7.3	5	5	50%	10
iCa2+ (mmol/L)
N1.30	3	3	50%	6

from 91% to 100%. By about 6 minutes, the Po₂ had

Table 1 Likelihood of PD-PEA associated with measured

factors

PEA

VF duration (min)

No PEA

% PEA

n

V1.30 5 5 50% 10

PEA indicates pulseless electrical activity.

The individual studies were categorized as exhibiting either no PD-PEA or PD-PEA present based on the postdefibrillation blood pressure records. For the directly measured blood-gas factors and VF duration, potential correlations with the onset of PD-PEA were explored. Only VF duration and arterial K⁺ and Pco₂ were found to be statistically correlated to the onset of PD-PEA in this study ( P b .05, 1-tailed t test). For episodes of PD-PEA, the mean (FSD) VF duration was 7.6 minutes (F3.2), the mean K⁺ was 5.8 mEq/L (F1.5), and the mean Pco₂ was 45.9 mm Hg (F14.5). With no PD-PEA, the mean VF duration was

3.0 minutes (F2.2), the mean K⁺ was 4.4 mEq/L (F1.3), and the mean Pco₂ was 62.4 mm Hg (F19.2). Mean aortic BP and arterial Po₂, pH, and iCa²⁺did not show a statistical correlation with the onset of PD-PEA.

Table 1 further explores potential trends by computing the percentage incidence of PD-PEA with grouped levels of the various factors. For example, PD-PEA occurred 100% of the time when the duration of VF with CPR exceeded 8 minutes, whereas it only occurred 50% of the time for durations between 5 to 8 minutes. In addition to the factors indicated previously, trends suggest a correlation between the onset of PD-PEA with mean BP and Po₂.

Discussion

From this pig study, we have identified several param- eters that may underlie the outcome of VF with CPR and the occurrence of PD-PEA; they are (1) total VF duration;

(2) mean arterial pressure; (3) arterial K⁺, Po₂, and Pco₂. Mean arterial pressure produced by the Chest Thumper, operating with 80- to 100-lb chest-compression force, 60 compressions per minute, and a 50% duty cycle, produced a typical mean aortic pressure of 20 to 25 mm Hg (Fig. 1). An increase in compression rate may increase mean pressure as advocated by the American Heart Association guidelines for CPR [6].

The correlation between VF duration and blood pressure during CPR with the occurrence of PD-PEA was expected. It is well established that the longer the duration of VF, the poorer the outcome [7,8]. Similarly, higher blood pressure during CPR with VF has long been associated with a better outcome [9,10]. As early as 1906, Crile and Dolley [11] determined that a coronary perfusion pressure of at least 30 to 40 mm Hg was necessary for successful resuscitation of patients. He used a pneumatic suit to compress the body.

A surprising finding was the decrease in arterial Po₂ (Fig. 2) during CPR in animals breathing 100% oxygen. Such an event points to impairment of oxygen transport across the alveolar membranes. Impaired O₂ alveolar transport could be caused by alveolar fluid accumulation. In 4 additional animals, we measured pulmonary artery mean pressure during VF with CPR, and it was virtually equal to aortic mean pressure, indicating the possibility of

pulmonary edema. This factor merits further investigation and suggests that during routine CPR, it may be advisable to apply an increased lung inflation pressure using 100% oxygen to increase the airway Po₂. If pulmonary hyperten- sion accompanies CPR, it could contribute to a time limitation for successful resuscitation.

The late decrease in SaO₂ (Fig. 2) is not surprising because of the nature of the oxygen-dissociation curves, which show that SaO₂ does not begin to decrease rapidly until the Po₂ falls below about 150 mm Hg, thereby making SaO₂ an insensitive indicator of approaching hypoxia.

As long ago as 1883, Ringer [12] discovered that K⁺ was needed for cardiac relaxation and Ca²⁺ was needed for contraction. Na⁺ was needed to make the solution (Ringer’s) isotonic. Because Ca²⁺ (Fig. 3) and Na⁺ did not change in this investigation, the excess K⁺ at the time of defibrillation could be a factor associated with the onset of PD-PEA.

The arterial K⁺ increased rapidly during VF with CPR, doubling in about 10 minutes (Fig. 3). It should be noted that the poor tissue perfusion afforded by the low CPR blood pressure impairs the cellular Na⁺/K⁺ ion pumps and K⁺ leaks out of all cells. Singh et al showed that with as little as 7 minutes of anoxia in the human heart, Na⁺ and Cl^— enter human myocardial cells, and K⁺ leaks out [13]. With 25 minutes of anoxia, Singh reported ultrastructural changes in the myocardium. Fish and Louie [14] reported a 15- patient, emergency department resuscitation study in which K⁺ was measured on all unsuccessfully resuscitated subjects. Measurements were made during 10 to 70 minutes of resuscitation. In all cases, K⁺ increased with prolonged resuscitation, reaching as high as 12 mEq/L (mmol/L).

The high arterial K⁺ revealed itself clearly in the postdefibrillation ECG. Winkler et al [15] showed that there is an orderly sequence of changes in the ECG as K⁺ is increased. At about 5 mEq/L, the T wave becomes peaked. At 7 mEq/L, an ST-segment shift occurs, and at about 8.5 mEq/L, atrioventricular block occurs. At 10 mEq/L intraventricular block occurs with widening of the QRS wave. Cardiac arrest occurs at 12 mEq/L. All of these ECG changes were seen in the postdefibrillation ECG in this study, and the ECG is a good and rapid indicator of the postdefibrillation level of K⁺. It is useful to recall that a high K⁺ solution is used for cardioplegia and in legal execution.

The slight rise in arterial Pco₂ after about 2 minutes of VF with CPR shown in Fig. 4 could indicate poor lung ventilation. However, these animals were all ventilated with 100% O₂, and the CO₂ absorber was changed before each study. End-tidal CO₂ has been identified as an indicator of the effectiveness of chest compressions and has been correlated with successful resuscitation from cardiac arrest [6,16]. Thus, it was interesting that in this investigation, a higher Pco₂ value was associated with less PD-PEA. The decrease in arterial bicarbonate (Fig. 4) probably reflects the slowly approaching metabolic acidosis; however, the pH changed only slightly during this study (see Fig. 4).

Summary

This pig study has shown that with VF and CPR delivered by the Chest Thumper (80-100 lb at 60/min and 50% duty cycle), with the pigs breathing 100% oxygen, a typical mean aortic pressure of 20 to 25 mm Hg was obtained. Despite breathing 100% oxygen, there was a slow but steady decrease in arterial Po₂ and a steady rise in arterial K⁺, doubling in about 10 minutes of VF with CPR. The arterial Ca²⁺ remained relatively constant, and the pH decreased slightly. PD-PEA was encountered in 50% of the trials after about 5 to 8 minutes of VF with CPR. Results from this investigation indicate that total VF duration, mean arterial pressure, arterial K⁺, Po₂, and Pco₂ may underlie the outcome of CPR with VF, in particular, the occurrence of PD-PEA. The reason for, and the consequence of, an elevated pulmonary artery pressure remains to be investigated.

References

1. American Heart Association. Sudden death from cardiac arrest- statistics. Dallas (Tex)7 American Heart Association; 2004 Accessed 10/07/05 at www.americanheart.org/downloadable/heart/ 1103835297279FS27SDCA5.pdf.
2. Wik L, Hansen TB, Fylling T, et al. Delaying defibrillation to give basic CPR to patients with out-of-hospital VF. JAMA 2003;289: 1389 - 95.
3. American Heart Association. Heart and stroke facts. Dallas (Tex)7 American Heart Association; 2003 Accessed 10/07/05 at www. americanheart.org/downloadable/heart/1056719919740HSFacts 2003text.pdf.
4. Niemann JT, Cruz B, Garner D, et al. Immediate countershock versus cardiopulmonary resuscitation before countershock in a 5-minute swine model of ventricular fibrillation arrest. Ann Emerg Med 2000;36:543 - 6.
5. Parish DC, Chandra KM, Dane FC. Success changes the problem: why ventricular fibrillation is declining, why pulseless electrical activity is emerging, and what to do about it. Resuscitation 2003; 58:31 - 5.
6. AHA. Devices to assist circulation. Circulation 2000;102(Suppl 1): I105 - 11.
7. Geddes LA, Roeder RA, Kemeny A, et al. The duration of ventricular fibrillation required to produce pulseless electrical activity. Am J Emerg Med 2005;23:138 - 41.
8. AHA. Heart and stroke facts. Dallas (Tex)7 American Heart Association; 2003.
9. Redding JS. Abdominal compression in cardiopulmonary resuscita- tion. Anesth Analg 1971;50:668 - 75.
10. Sanders AB, Kern KB, Atlas M, et al. Importance of the duration of inadequate coronary perfusion pressure on resuscitation from cardiac arrest. J Am Coll Cardiol 1985;6:113 - 8.
11. Crile G, Dolley DH. An experimental research into the resuscitation of dogs killed by anesthetics and asphyxia. J Exp Med 1906;8:713 - 25.
12. Ringer S. A further contribution regarding the influence of the different constituents of the blood on the contraction of the heart. J Physiol (London) 1883-1884;4:29 - 42.
13. Singh CM, Flear CTG, Nandra A, et al. Electrolyte changes in human myocardium after anoxic arrest. Cardiology 1971;56:128 - 35.
14. Fish A, Louie L. Laboratory tests during resuscitation. J Emerg Med 1989;7:451 - 6.
15. Winkler AW, Hoff HE, Smith PH. Cardiovascular effects of potassium, calcium, magnesium, and barium. Yale J Biol Med 1940; 13:123 - 32.
16. Kern KB, Sanders AB, Raife J, et al. A study of chest compression rates during cardiopulmonary resuscitation in humans. Arch Intern Med 1992;152:145 - 9.