LABORATORY INVESTIGATION open-chest cardiac massage
Limitations of Open-Chest Cardiac Massage After Prolonged, Untreated Cardiac Arrest in Dogs Study objectives: Open-chest cardiac massage is an effective method of resuscitation if instituted within 15 minutes of normothermic cardiac arrest that has failed to respond to ongoing closed-chest CPR efforts. The usefulness of invasive forms of CPR after various periods of untreated cardiac arrest is less certain. This study was performed to determine the effectiveness of open-chest resuscitation after prolonged periods of untreated cardiac arrest. Setting and design: Prospective, controlled laboratory investigation using an animal model of cardiac arrest. Open-chest cardiac massage initially was compared to standard closed-chest compression CPR. The efficacy of open-chest CPR then was evaluated after ten and 40 minutes of untreated ventricular fibrillation. Type of participants: Twenty mongrel dogs (24 + 1 kg). Measurements and main results: After 20 minutes of untreated ventricular fibrillation, open-chest resuscitation was significantly better than closed-chest efforts for the production of coronary perfusion pressure (58 +_ 14 vs 2 ± 1 m m Hg; P < .05) and initial resuscitation success (five of five vs one of five; P < .03). Open-chest cardiac massage was equally effective for initial resuscitation ff begun after ten or 20 minutes of untreated ventricular fibrillation (five of five vs five of five), but ff untreated ventricular fibrillation continued for 40 minutes prior to instituting openchest massage, no resuscitation benefit was found (none of five; P < .005). There were m a r k e d differences in 24-hour survival depending on the length of time untreated cardiac arrest continued prior to instituting open-chest resuscitation efforts. After 20 minutes of ventricular fibrillation, initial resuscitation was successful with open-chest massage, but long-term survival was poor. Conclusion: Open-chest cardiac massage did not produce long-term survival ff untreated cardiac arrest persisted for 20 or more minutes prior to invasive resuscitation efforts. [Kern KB, Sanders AB, Janas W, Nelson JR, Badylak SE Babbs CE Tacker WA, Ewy GA: Limitations of open-chest cardiac massage after prolonged, untreated cardiac arrest in dogs. Ann Emerg Med July 1991;20:761-767.]
INTRODUCTION Open-chest cardiac massage results in better circulatory support than other forms of CPR in both experimental models 1-6 and in the in-hospital setting. 7 Experimentally, this improved hemodynamic support has resulted in superior rates of successful resuscitation outcome. 8-1z Open-chest cardiac massage has been shown to be highly effective in the treatment of normothermic cardiac arrest if instituted within 15 to 20 minutes of circulatory collapse that has failed to respond to rapidly initiated but ineffective closed-chest CPR efforts.llA z Little is known about the effectiveness of open-chest cardiac massage when instituted after various periods of untreated cardiac arrest. In most instances, the duration of cardiac arrest before the institution of resuscitation efforts is unknown. Before espousing increased use of invasive CPR techniques, 13 it is important to examine the effectiveness of invasive CPR in such instances where the "downtime" is unknown and therefore possibly prolonged. Therefore, this study using a dog model was performed to examine the
86/761
Annals of Emergency Medicine
Karl B Kern, MD* Arthur B Sanders, MDt Wolfgang Janas¢ James R Nelson* Stephen F Badylak, DVM, PhD¢ Charles F Babbs, MD, PhD¢ Willis A Tacker, MD, PhD¢ Gordon A Ewy, MD* Tucson, Arizona West Lafayette, Indiana From the Section of Cardiology, Department of Internal Medicine,* and Section of Emergency Medicine, Department of Surgery,t University of Arizona College of Medicine, University Medical Center, Tucson; and the Hillenbrand Biomedical Engineering Center and Department of Veterinary Physiology and Pharmacology, Purdue University, West Lafayette, Indiana.-~ Received for publication May 30, 1990. Revision received December 31, 1990. Accepted for publication February 12, 1991. Supported by Grant HL-29398 from the National Heart, Lung and Blood Institute, a Grant-in-Aid from the Arizona Affiliate of the American Heart Association, and by a grant from the Flinn Foundation of Arizona, Phoenix. Address for reprints: Karl B Kern, MD, Section of Cardiology, University of Arizona College of Medicine, University Medical Center, 1501 N Campbell Avenue, Tucson, Arizona 85724.
20:7 July 1991
OPEN-CHEST CARDIAC MASSAGE Kern et al
Part1 ~ 0 min
TABLE1. Baseline hemodynamics
DefibrJlialion O CCCPR CCM j - - - --'-) 7 days 20 rain
9efibril/ation ) OCCMJ - - ~ Untreae~VF i 0 10 22 rain rain min
32 rain
Pad2
7 days Defibrillation > OCCM J------')Tdays 40 52 rain rain
20 min CCC*
20 rain 0CCM
40 min 0CCM
10 rain 0CCM
Aortic systolic
Pressure (mm Hg)
115 + 5
138 _+ 22
123 +_ 8
116 ± 6
Aortic diastolic
86 + 5
75 ± 0
94 ± 9
90 ± 5
5 + 1
5 ± 2
5 ± 1
3 ± 1
82 ± 6
75 +_ 1
90 ± 9
87 ± 4
Right atrial mean Coronary perfusion pressure
*CCC, closed chesl compressions; OCCM, open chest cardiac massage
FIGURE 1. Study protocol flowchart ,for both part 1 and part 2. CC CPR, c l o s e d - c h e s t CPR; OCCM, openchest cardiac massage; VF, ventricular fibrillation. effectiveness of open-chest cardiac massage in reviving victims of normothermic cardiac arrest after increasing periods of untreated ventricular fibrillation. The study was performed in two phases. In the first phase, open-chest cardiac massage was c o m p a r e d w i t h c l o s e d - c h e s t compressions after 20 minutes of untreated ventricular fibrillation. The second phase evaluated the effectiveness of open-chest cardiac massage after ten, 20, and 40 minutes of untreated ventricular fibrillation. The end points of both studies were coronary perfusion pressure, initial resuscitation success, 24-hour and sevenday survival rates, and neurologic deficit scoring for survivors.
MATERIALS AND METHODS Animal Preparation All studies were performed under the guidelines of the American Physiologic Society after approval from the Animal Care and Use Committee at Purdue University. Twenty mongrel dogs (weight, 24 -+ 1 kg) were used in this study. Initially, openchest cardiac massage was compared with standard closed-chest compression CPR (ten dogs) following 20 minutes of untreated cardiac arrest. Once it was determined that openchest cardiac massage was superior to standard external CPR, the results after 20 minutes were compared with those obtained with open-chest efforts initiated after ten and 40 minutes of ventricular fibrillation (ten) (Figure 1). Each animal was anesthetized with halothane (0.5% to 1%) and nitrous oxide in oxygen. After endotracheal intubation and attachment of the ex20:7: July 1991
pired, end-tidal carbon dioxide monitor (Hewlett-Packard, Palo Alto, California), ventilation was performed with a pressure-cycled ventilator (Michigan Instruments, Grand Rapids, Michigan) incorporated into a m e c h a n i c a l r e s u s c i t a t o r . Adjustments in ventilatory rate and tidal volumes were made to approximate a normal arterial pH (7.35 to 7.45). With the animal under full anesthesia and with the use of aseptic technique, subcutaneous skin ECG leads were placed, and skin incisions were made in the right femoral and cervical areas. Fluid-filled pressure monitoring catheters were placed in the ascending aorta and right a t r i u m through the right femoral artery and vein. The catheters were connected by pressure transducers (Cobe Lab, Denver) to a physiograph recorder (Narco-Biosystems, Houston). Calibration was accomplished with a mercury manometer. Vascular sheaths (8F) were inserted into the right carotid artery and internal jugular vein.
Experimental Protocol Collected baseline data included ECG, aortic and right atrial pressures, expired end-tidal Pco2, and arterial and mixed venous blood gases. Cardiac arrest then was produced by electrical induction of ventricular fibrillation using methods previously described. 12 Once induced, ventricular fibrillation was continued without treatment for ten, 20, or 40 minutes (Figure 1). External defibrillation was attempted after the designated ventricular fibrillation period with two defibrillation shocks of 100 and 200 J, respectively. Animals that failed to regain a perfusing rhythm were given CPR as assigned by the protocol. Animals with nondefibrillated ventricles received a bolus of epinephrine (1 mg IV) followed by an infusion of epinephrine at a rate of 4.0 ixg/kg/min and a bolus of sodium Annals of Emergency Medicine
bicarbonate (3 mEq/kg) as a part of the CPR protocol. During the CPR period, additional defibrillation attempts were made every three minutes. A maximum of four such CPR periods was allowed. In the animals receiving closed-chest compressions (performed in an anteroposterior orientation), defibrillation attempts were made with 100 and 200 J, respectively, after each three-minute period of CPR. In the groups receiving open-chest cardiac massage, defibrillation attempts were performed using internal defibrillation, first at 20 J and then at 30 J. A maximum of 12 minutes of CPR was allowed with a total of eight defibrillation attempts performed during this time. Animals that had defibrillated ventricles but were still hypotensive during the CPR period continued to receive chest compressions and pressor support for the entire 12 minutes. After the 12 minutes of CPR, resuscitation status for each animal was determined by measuring the aortic blood pressure. If the systolic blood pressure was more than 40 m m Hg, the animal was considered resuscitated; if less than 40 mm Hg, the animal was considered nonresuscitated. Resuscitated animals were monitored during an intensive care period for approximately one to two hours. During this time, any continued vasopressor support (epinephrine infusion) was weaned, and the animal was extubated. Animals receiving internal cardiac massage had their thoracotomies closed at this time using a four-layer technique. 12 After transfer to a m a i n t e n a n c e care area, surviving animals received maintenance care for a maximum of seven days. This included prophylactic antibiotic t r e a t m e n t with trimethoprim-sulfamethoxazole as well as analgesics (subcutaneous morphine sulfate) for pain control. Daily temperature and wound checks were 762/87
OPEN-CHEST CARDIAC MASSAGE Kern et al
* P < ,03 vs CC CPR 5.0
5.0
4.0
4.0
3.0
3.0
"5 ,5 2.0
* P < .03 vs CC CPR
d 2.0
z
z
1.0
1.0
0.0
0.0 20 min VF Closed-ChestCPR
10 rain VF
20 minVF
40 rain VF
20 min VF
Open-ChestCardiacMassage
2
performed. All animals surviving seven days then were e u t h a n i z e d using an IV euthanizing (T-61) solution. All animals u n d e r w e n t necropsy examination, either at the time of death or after eu[hanasia at seven days. Careful pathologic evaluation was performed by a veterinarian pathologist.
Protocol 1 Ten animals underwent 20 minutes of untreated ventricular fibrillation before a s s i g n m e n t to either closed-chest CPR or open-chest cardiac massage. The animals receiving standard closed-chest manual CPR underwent 2-in. chest compressions at a rate of 80 with a 50% duty cycle. Chest compressions were provided by a programmable mechanical resucitator (Thumper ®, Michigan Instruments, Grand Rapids, Michigan). During periods of m a n u a l cardiac massage, the ventilator functioned independently of the Thumper ®. Changes in the ventilator settings were allowed after the resuscitation of each animal and performed to normalize the postresuscitation arterial pH. The animals undergoing openchest resuscitation efforts had a lateral thoracotomy performed after the prescribed period of ventricular fibrillation. All direct cardiac massage was performed at a rate of 60 using a single-hand technique.
Protocol 2 Ten additional animals received open-chest cardiac massage. Five animals underwent ten minutes of untreated ventricular fibrillation before the institution of open-chest cardiac massage. Another five underwent 40 88/763
Closed-ChestCPR
minutes of untreated ventricular fibrillation before receiving open-chest cardiac massage. All animals were randomized in a stratified fashion.
Data Analysis All hemodynamic data are given as the mean value + SEM. Coronary perfusion pressure was calculated during resuscitation effort using the standard technique of subtracting the right atrial mid-diastolic pressure from the simultaneous aortic middiastolic pressure, lo-12 Baseline data among the four groups of study animals were compared by analysis of variance. Student's t test (two-tailed) for unpaired samples was used to compare weights and hemodynamic variables. The Bonferroni correction factor was used where m u l t i p l e t tests were used. Fisher's exact testing (one-sided) was performed for the comparison of initial resuscitation success, and 24-hour and seven-day survival rates.
RESULTS Resuscitation After 20 minutes of untreated ventricular fibrillation, open-chest cardiac massage was significantly better than closed-chest CPR for initial resuscitation success (five of five vs one of five, P < .03). No difference in initial r e s u s c i t a t i o n success was noted if open-chest cardiac massage was begun after ten or 20 minutes of ventricular fibrillation (five of five vs five of five), but if untreated ventricular fibrillation continued for 40 minutes before the institution of internal massage, no resuscitation benefit was found (none of five vs five of five, P < .005). No difference in reAnnals of Emergency Medicine
10 rain VF
20 minVF
40rain VF
0pen-ChestCardiacMassage
3
FIGURE 2. Bar graph of initial resuscitation among animals receiving either closed-chest CPR (CC CPR) or open-chest cardiac massage after various periods of untreated ventricular fibrillation (VF). FIGURE 3. Bar graph of 24-hour survival among animals that were resuscitated successfully with openchest cardiac massage after either ten or 20 minutes of untreated cardiac arrest. Survival was significantly less than initial resuscitation (one of five vs five of five) after 20 minutes of untreated ventricular fibrillation (VF), but no difference between initial resuscitation success and subsequent 24-hour survival occurred if open-chest massage was begun after only ten m i n u t e s of unt r e a t e d cardiac arrest. CC CPR, closed-chest CPR.
s u s c i t a t i o n success was seen bet w e e n c l o s e d - c h e s t CPR after 20 minutes of ventricular fibrillation and open-chest massage after 40 minutes of untreated ventricular fibrillation (one of five vs none of five, Figure 2).
Survival A m o n g the a n i m a l s r e c e i v i n g open-chest cardiac massage, marked differences occurred in 24-hour survival depending on the length of time ventricular fibrillation continued untreated before the i n s t i t u t i o n of open-chest cardiac massage. Only one of the five resuscitated animals from t h e open-chest group survived after 20 minutes of untreated ventricular fibrillation, whereas four of 20:7 July 1991
OPEN-CHEST CARDIAC MASSAGE Kern et al
F I G U R E 4, Hemodynamic tracing from one of the animals receiving open-chest cardiac massage after 20 minutes of untreated ventricular fibrillation. FIGURE 5. Bar graphs of aortic and
right atrial pressures produced during CPR with open-chest cardiac massage (OCCM) begun after ten, 20, and 40 minutes of untreated ventricular fibrillation (VF). five a n i m a l s survived 24 hours w i t h open-chest cardiac massage w h e n it was begun after only ten m i n u t e s of untreated ventricular fibrillation. The one a n i m a l that was resuscitated with closed-chest compressions after 20 m i n u t e s of ventricular fibrillation died after eight hours; none survived 24 hours or m o r e (Figure 3). Eleven of 20 a n i m a l s were resuscitated successfully. Eight of these died before s e v e n days. S e v e n of t h e s e eight animals had c o m p l e t e necropsy examinations. The three animals that survived for seven days also had complete necropsy examinations. All of the dead a n i m a l s on w h i c h necropsy was performed had microscopic and gross evidence of diffuse pulmonary edema. N o n e of the three survivors had evidence of p u l m o n a r y edema. The incidence of p u l m o n a r y edema in the a n i m a l s that died was significantly greater t h a n that in the animals that survived for seven days (seven of seven vs n o n e of three, P
Limitations of Open-Chest Cardiac Massage After Prolonged, Untreated Cardiac Arrest in Dogs Study objectives: Open-chest cardiac massage is an effective method of resuscitation if instituted within 15 minutes of normothermic cardiac arrest that has failed to respond to ongoing closed-chest CPR efforts. The usefulness of invasive forms of CPR after various periods of untreated cardiac arrest is less certain. This study was performed to determine the effectiveness of open-chest resuscitation after prolonged periods of untreated cardiac arrest. Setting and design: Prospective, controlled laboratory investigation using an animal model of cardiac arrest. Open-chest cardiac massage initially was compared to standard closed-chest compression CPR. The efficacy of open-chest CPR then was evaluated after ten and 40 minutes of untreated ventricular fibrillation. Type of participants: Twenty mongrel dogs (24 + 1 kg). Measurements and main results: After 20 minutes of untreated ventricular fibrillation, open-chest resuscitation was significantly better than closed-chest efforts for the production of coronary perfusion pressure (58 +_ 14 vs 2 ± 1 m m Hg; P < .05) and initial resuscitation success (five of five vs one of five; P < .03). Open-chest cardiac massage was equally effective for initial resuscitation ff begun after ten or 20 minutes of untreated ventricular fibrillation (five of five vs five of five), but ff untreated ventricular fibrillation continued for 40 minutes prior to instituting openchest massage, no resuscitation benefit was found (none of five; P < .005). There were m a r k e d differences in 24-hour survival depending on the length of time untreated cardiac arrest continued prior to instituting open-chest resuscitation efforts. After 20 minutes of ventricular fibrillation, initial resuscitation was successful with open-chest massage, but long-term survival was poor. Conclusion: Open-chest cardiac massage did not produce long-term survival ff untreated cardiac arrest persisted for 20 or more minutes prior to invasive resuscitation efforts. [Kern KB, Sanders AB, Janas W, Nelson JR, Badylak SE Babbs CE Tacker WA, Ewy GA: Limitations of open-chest cardiac massage after prolonged, untreated cardiac arrest in dogs. Ann Emerg Med July 1991;20:761-767.]
INTRODUCTION Open-chest cardiac massage results in better circulatory support than other forms of CPR in both experimental models 1-6 and in the in-hospital setting. 7 Experimentally, this improved hemodynamic support has resulted in superior rates of successful resuscitation outcome. 8-1z Open-chest cardiac massage has been shown to be highly effective in the treatment of normothermic cardiac arrest if instituted within 15 to 20 minutes of circulatory collapse that has failed to respond to rapidly initiated but ineffective closed-chest CPR efforts.llA z Little is known about the effectiveness of open-chest cardiac massage when instituted after various periods of untreated cardiac arrest. In most instances, the duration of cardiac arrest before the institution of resuscitation efforts is unknown. Before espousing increased use of invasive CPR techniques, 13 it is important to examine the effectiveness of invasive CPR in such instances where the "downtime" is unknown and therefore possibly prolonged. Therefore, this study using a dog model was performed to examine the
86/761
Annals of Emergency Medicine
Karl B Kern, MD* Arthur B Sanders, MDt Wolfgang Janas¢ James R Nelson* Stephen F Badylak, DVM, PhD¢ Charles F Babbs, MD, PhD¢ Willis A Tacker, MD, PhD¢ Gordon A Ewy, MD* Tucson, Arizona West Lafayette, Indiana From the Section of Cardiology, Department of Internal Medicine,* and Section of Emergency Medicine, Department of Surgery,t University of Arizona College of Medicine, University Medical Center, Tucson; and the Hillenbrand Biomedical Engineering Center and Department of Veterinary Physiology and Pharmacology, Purdue University, West Lafayette, Indiana.-~ Received for publication May 30, 1990. Revision received December 31, 1990. Accepted for publication February 12, 1991. Supported by Grant HL-29398 from the National Heart, Lung and Blood Institute, a Grant-in-Aid from the Arizona Affiliate of the American Heart Association, and by a grant from the Flinn Foundation of Arizona, Phoenix. Address for reprints: Karl B Kern, MD, Section of Cardiology, University of Arizona College of Medicine, University Medical Center, 1501 N Campbell Avenue, Tucson, Arizona 85724.
20:7 July 1991
OPEN-CHEST CARDIAC MASSAGE Kern et al
Part1 ~ 0 min
TABLE1. Baseline hemodynamics
DefibrJlialion O CCCPR CCM j - - - --'-) 7 days 20 rain
9efibril/ation ) OCCMJ - - ~ Untreae~VF i 0 10 22 rain rain min
32 rain
Pad2
7 days Defibrillation > OCCM J------')Tdays 40 52 rain rain
20 min CCC*
20 rain 0CCM
40 min 0CCM
10 rain 0CCM
Aortic systolic
Pressure (mm Hg)
115 + 5
138 _+ 22
123 +_ 8
116 ± 6
Aortic diastolic
86 + 5
75 ± 0
94 ± 9
90 ± 5
5 + 1
5 ± 2
5 ± 1
3 ± 1
82 ± 6
75 +_ 1
90 ± 9
87 ± 4
Right atrial mean Coronary perfusion pressure
*CCC, closed chesl compressions; OCCM, open chest cardiac massage
FIGURE 1. Study protocol flowchart ,for both part 1 and part 2. CC CPR, c l o s e d - c h e s t CPR; OCCM, openchest cardiac massage; VF, ventricular fibrillation. effectiveness of open-chest cardiac massage in reviving victims of normothermic cardiac arrest after increasing periods of untreated ventricular fibrillation. The study was performed in two phases. In the first phase, open-chest cardiac massage was c o m p a r e d w i t h c l o s e d - c h e s t compressions after 20 minutes of untreated ventricular fibrillation. The second phase evaluated the effectiveness of open-chest cardiac massage after ten, 20, and 40 minutes of untreated ventricular fibrillation. The end points of both studies were coronary perfusion pressure, initial resuscitation success, 24-hour and sevenday survival rates, and neurologic deficit scoring for survivors.
MATERIALS AND METHODS Animal Preparation All studies were performed under the guidelines of the American Physiologic Society after approval from the Animal Care and Use Committee at Purdue University. Twenty mongrel dogs (weight, 24 -+ 1 kg) were used in this study. Initially, openchest cardiac massage was compared with standard closed-chest compression CPR (ten dogs) following 20 minutes of untreated cardiac arrest. Once it was determined that openchest cardiac massage was superior to standard external CPR, the results after 20 minutes were compared with those obtained with open-chest efforts initiated after ten and 40 minutes of ventricular fibrillation (ten) (Figure 1). Each animal was anesthetized with halothane (0.5% to 1%) and nitrous oxide in oxygen. After endotracheal intubation and attachment of the ex20:7: July 1991
pired, end-tidal carbon dioxide monitor (Hewlett-Packard, Palo Alto, California), ventilation was performed with a pressure-cycled ventilator (Michigan Instruments, Grand Rapids, Michigan) incorporated into a m e c h a n i c a l r e s u s c i t a t o r . Adjustments in ventilatory rate and tidal volumes were made to approximate a normal arterial pH (7.35 to 7.45). With the animal under full anesthesia and with the use of aseptic technique, subcutaneous skin ECG leads were placed, and skin incisions were made in the right femoral and cervical areas. Fluid-filled pressure monitoring catheters were placed in the ascending aorta and right a t r i u m through the right femoral artery and vein. The catheters were connected by pressure transducers (Cobe Lab, Denver) to a physiograph recorder (Narco-Biosystems, Houston). Calibration was accomplished with a mercury manometer. Vascular sheaths (8F) were inserted into the right carotid artery and internal jugular vein.
Experimental Protocol Collected baseline data included ECG, aortic and right atrial pressures, expired end-tidal Pco2, and arterial and mixed venous blood gases. Cardiac arrest then was produced by electrical induction of ventricular fibrillation using methods previously described. 12 Once induced, ventricular fibrillation was continued without treatment for ten, 20, or 40 minutes (Figure 1). External defibrillation was attempted after the designated ventricular fibrillation period with two defibrillation shocks of 100 and 200 J, respectively. Animals that failed to regain a perfusing rhythm were given CPR as assigned by the protocol. Animals with nondefibrillated ventricles received a bolus of epinephrine (1 mg IV) followed by an infusion of epinephrine at a rate of 4.0 ixg/kg/min and a bolus of sodium Annals of Emergency Medicine
bicarbonate (3 mEq/kg) as a part of the CPR protocol. During the CPR period, additional defibrillation attempts were made every three minutes. A maximum of four such CPR periods was allowed. In the animals receiving closed-chest compressions (performed in an anteroposterior orientation), defibrillation attempts were made with 100 and 200 J, respectively, after each three-minute period of CPR. In the groups receiving open-chest cardiac massage, defibrillation attempts were performed using internal defibrillation, first at 20 J and then at 30 J. A maximum of 12 minutes of CPR was allowed with a total of eight defibrillation attempts performed during this time. Animals that had defibrillated ventricles but were still hypotensive during the CPR period continued to receive chest compressions and pressor support for the entire 12 minutes. After the 12 minutes of CPR, resuscitation status for each animal was determined by measuring the aortic blood pressure. If the systolic blood pressure was more than 40 m m Hg, the animal was considered resuscitated; if less than 40 mm Hg, the animal was considered nonresuscitated. Resuscitated animals were monitored during an intensive care period for approximately one to two hours. During this time, any continued vasopressor support (epinephrine infusion) was weaned, and the animal was extubated. Animals receiving internal cardiac massage had their thoracotomies closed at this time using a four-layer technique. 12 After transfer to a m a i n t e n a n c e care area, surviving animals received maintenance care for a maximum of seven days. This included prophylactic antibiotic t r e a t m e n t with trimethoprim-sulfamethoxazole as well as analgesics (subcutaneous morphine sulfate) for pain control. Daily temperature and wound checks were 762/87
OPEN-CHEST CARDIAC MASSAGE Kern et al
* P < ,03 vs CC CPR 5.0
5.0
4.0
4.0
3.0
3.0
"5 ,5 2.0
* P < .03 vs CC CPR
d 2.0
z
z
1.0
1.0
0.0
0.0 20 min VF Closed-ChestCPR
10 rain VF
20 minVF
40 rain VF
20 min VF
Open-ChestCardiacMassage
2
performed. All animals surviving seven days then were e u t h a n i z e d using an IV euthanizing (T-61) solution. All animals u n d e r w e n t necropsy examination, either at the time of death or after eu[hanasia at seven days. Careful pathologic evaluation was performed by a veterinarian pathologist.
Protocol 1 Ten animals underwent 20 minutes of untreated ventricular fibrillation before a s s i g n m e n t to either closed-chest CPR or open-chest cardiac massage. The animals receiving standard closed-chest manual CPR underwent 2-in. chest compressions at a rate of 80 with a 50% duty cycle. Chest compressions were provided by a programmable mechanical resucitator (Thumper ®, Michigan Instruments, Grand Rapids, Michigan). During periods of m a n u a l cardiac massage, the ventilator functioned independently of the Thumper ®. Changes in the ventilator settings were allowed after the resuscitation of each animal and performed to normalize the postresuscitation arterial pH. The animals undergoing openchest resuscitation efforts had a lateral thoracotomy performed after the prescribed period of ventricular fibrillation. All direct cardiac massage was performed at a rate of 60 using a single-hand technique.
Protocol 2 Ten additional animals received open-chest cardiac massage. Five animals underwent ten minutes of untreated ventricular fibrillation before the institution of open-chest cardiac massage. Another five underwent 40 88/763
Closed-ChestCPR
minutes of untreated ventricular fibrillation before receiving open-chest cardiac massage. All animals were randomized in a stratified fashion.
Data Analysis All hemodynamic data are given as the mean value + SEM. Coronary perfusion pressure was calculated during resuscitation effort using the standard technique of subtracting the right atrial mid-diastolic pressure from the simultaneous aortic middiastolic pressure, lo-12 Baseline data among the four groups of study animals were compared by analysis of variance. Student's t test (two-tailed) for unpaired samples was used to compare weights and hemodynamic variables. The Bonferroni correction factor was used where m u l t i p l e t tests were used. Fisher's exact testing (one-sided) was performed for the comparison of initial resuscitation success, and 24-hour and seven-day survival rates.
RESULTS Resuscitation After 20 minutes of untreated ventricular fibrillation, open-chest cardiac massage was significantly better than closed-chest CPR for initial resuscitation success (five of five vs one of five, P < .03). No difference in initial r e s u s c i t a t i o n success was noted if open-chest cardiac massage was begun after ten or 20 minutes of ventricular fibrillation (five of five vs five of five), but if untreated ventricular fibrillation continued for 40 minutes before the institution of internal massage, no resuscitation benefit was found (none of five vs five of five, P < .005). No difference in reAnnals of Emergency Medicine
10 rain VF
20 minVF
40rain VF
0pen-ChestCardiacMassage
3
FIGURE 2. Bar graph of initial resuscitation among animals receiving either closed-chest CPR (CC CPR) or open-chest cardiac massage after various periods of untreated ventricular fibrillation (VF). FIGURE 3. Bar graph of 24-hour survival among animals that were resuscitated successfully with openchest cardiac massage after either ten or 20 minutes of untreated cardiac arrest. Survival was significantly less than initial resuscitation (one of five vs five of five) after 20 minutes of untreated ventricular fibrillation (VF), but no difference between initial resuscitation success and subsequent 24-hour survival occurred if open-chest massage was begun after only ten m i n u t e s of unt r e a t e d cardiac arrest. CC CPR, closed-chest CPR.
s u s c i t a t i o n success was seen bet w e e n c l o s e d - c h e s t CPR after 20 minutes of ventricular fibrillation and open-chest massage after 40 minutes of untreated ventricular fibrillation (one of five vs none of five, Figure 2).
Survival A m o n g the a n i m a l s r e c e i v i n g open-chest cardiac massage, marked differences occurred in 24-hour survival depending on the length of time ventricular fibrillation continued untreated before the i n s t i t u t i o n of open-chest cardiac massage. Only one of the five resuscitated animals from t h e open-chest group survived after 20 minutes of untreated ventricular fibrillation, whereas four of 20:7 July 1991
OPEN-CHEST CARDIAC MASSAGE Kern et al
F I G U R E 4, Hemodynamic tracing from one of the animals receiving open-chest cardiac massage after 20 minutes of untreated ventricular fibrillation. FIGURE 5. Bar graphs of aortic and
right atrial pressures produced during CPR with open-chest cardiac massage (OCCM) begun after ten, 20, and 40 minutes of untreated ventricular fibrillation (VF). five a n i m a l s survived 24 hours w i t h open-chest cardiac massage w h e n it was begun after only ten m i n u t e s of untreated ventricular fibrillation. The one a n i m a l that was resuscitated with closed-chest compressions after 20 m i n u t e s of ventricular fibrillation died after eight hours; none survived 24 hours or m o r e (Figure 3). Eleven of 20 a n i m a l s were resuscitated successfully. Eight of these died before s e v e n days. S e v e n of t h e s e eight animals had c o m p l e t e necropsy examinations. The three animals that survived for seven days also had complete necropsy examinations. All of the dead a n i m a l s on w h i c h necropsy was performed had microscopic and gross evidence of diffuse pulmonary edema. N o n e of the three survivors had evidence of p u l m o n a r y edema. The incidence of p u l m o n a r y edema in the a n i m a l s that died was significantly greater t h a n that in the animals that survived for seven days (seven of seven vs n o n e of three, P
