RECOVER 2.0 Worksheet

QUESTION ID: ALS-05

PICO Question:
In dogs with CPA (P) does closed-chest CPR (I) compared to open chest CPR (C) improve outcome (O)?

Outcomes:
Favorable neurologic outcome,Surrogate marker(s) of perfusion,Survival to Discharge,ROSC

Prioritized Outcomes (1= most critical; final number = least important):

1. Favorable neurologic outcome
2. Survival to discharge
3. ROSC
4. Surrogate markers of perfusion

Domain chairs: Gareth Buckley, Elizabeth Rozanski, Jake Wolf

Evidence evaluators: Elizabeth Ross, Stefania Grasso,Melissa Evans

Conflicts of interest: None

Search strategy: See attached document

Evidence Review:

Study Design

Reduced Quality Factors 0 = no serious, - = serious, - - = very serious

Positive Quality Factors 0 = none, + = one, ++ = multiple

Dichotomous Outcome Summary

Non-Dichotomous Outcome Summary Brief description

Overall Quality High, moderate, low, very low, none

No of studies

Study Type

RoB

Indirectness

Imprecision

Inconsistency

Large Effect

Dose-Response

Confounder

# Intervention with Outcome

# Control with Outcome

RR (95% CI)

Outcome: Favorable neurologic outcome

0

CT

1

OB

-

–

0

0

CPA after cardiac surgery - all survived

Very low

3

ES

0

0

-

-

Improved neuro outcome with open chest CPR

Very low

Outcome: Survival to discharge
0	CT
5	OB	-	-	0	0							Equivocal whether improved survival	Very low
4	ES	0	0	-	-							Improved survival with early open chest	Moderate

Outcome: ROSC
0	CT
3	OB	–	–	0	0							No difference with open chest CPR	Very low
4	ES	0	0	-	-							Improved ROSC in dogs with open chest	Very low

Outcome: Surrogate markers of perfusion
0	CT
1	OB	–	–	0	0							No difference in ETCO2 with open chest	Very low
14	ES	0	0	0	-							Improved hemodynamics with open chest	Low

PICO Question Summary

Introduction

The 2012 RECOVER veterinary CPR guidelines advise prompt open-chest CPR (OCCPR) in specific clinical scenarios, including tension pneumothorax and pericardial effusion.1 In human medicine, emergency department thoracotomy (EDT) may be used for cardiac arrest secondary to penetrating trauma.2 However, the utility and timing of OCCPR outside these specific situations is unknown, particularly considering the cost of OCCPR and the intensity of subsequent management.

Consensus on science

Outcome 1: Favorable neurologic outcome For the most critical outcome of favorable neurologic outcome, one observational study in human adults (very low quality of evidence downgraded for serious risk of bias and indirectness) and three experimental studies in dogs were identified (very low quality of evidence downgraded for serious imprecision and inconsistency).3–6 A study by Kern et al. (1987) examined 29 mongrel dogs with induced ventricular fibrillation that received standard CPR for 15 minutes and subsequent defibrillation. Unsuccessfully defibrillated dogs were then randomized to receive 2 minutes of either OCCPR or CCCPR. The study showed no difference in neurological scores between the two groups.4 In another study in dogs with ventricular fibrillation and immediate defibrillation or 30 subsequent minutes of CCCPR or OCCPR, the authors found that OCCPR resulted in improved neurological scores when compared to CCCPR.5 In 12 dogs with cardiac arrest induced via potassium chloride that were then randomized to either OCCPR or CCCPR, all dogs with OCCPR were resuscitated and behaved normally at 72 hours.6 Only 3/7 of CCPCR dogs survived and 2 of these had incapacitating neurological deficits. Anthi et al. (1998) examined 29 human adults with cardiac arrest within 24 hours following cardiac surgery.3 In this population, CCCPR was performed for 3-5 minutes, then followed by OCCPR if needed. Thirteen people achieved ROSC with CCCPR and 14 achieved ROSC with OCCPR; all were discharged neurologically intact. However, no control population was used to compare OCCPR to CCCPR directly. Outcomes 2 and 3: Survival to discharge and ROSC For the next most critical outcomes of survival to discharge and ROSC, five observational studies in addition to the Anthi study described above were identified in people (four in adults, one in children) with traumatic cardiac arrest (very low quality of evidence downgraded for serious risk of bias and indirectness).2,3,7–10 Four experimental studies were also identified for these outcomes, all in previously healthy dogs (very low quality of evidence, downgraded for serious imprecision and inconsistency).4–6,11 The observational studies in people demonstrated little to no benefit with open-chest CPR (OCCPR) when compared to closed-chest CPR (CCCPR), while the experimental studies in dogs largely demonstrated improved survival with OCCPR. The Kern et al. study of 29 mongrel dogs with induced ventricular fibrillation showed improved ROSC frequency, 24-hour survival (12/14 v. 4/14), and 7-day survival (11/14 v. 4/14) with OCCPR.4 Similarly, the Bircher et al. study found that OCCPR resulted in improved frequency of ROSC and survival at 24 hours compared to CCCPR.5 As described above, Benson et al. found in 12 dogs with cardiac arrest induced via potassium chloride that all dogs with OCCPR were resuscitated and survived to 72 hours, while only 3/7 of CCCPR dogs achieved ROSC.6 DeBehnke et al. (1991) found in a myocardial infarct model in 26 dogs with subsequent ventricular fibrillation that there was no difference in ROSC or survival between dogs receiving OCCPR and those receiving CCCPR.11 Schulz-Drost et al. (2020) examined adults who underwent emergency department thoracotomy (EDT) for trauma, a subset of whom underwent EDT for cardiac arrest.2 For these, the survival rate was 4.8% for blunt trauma but was 20.7% for penetrating trauma. Prieto et al. (2020) analyzed patients 16 years or younger who underwent EDT within 30 minutes of arrival to a hospital.8 Of the 53 patients with no signs of life who received EDT, none survived. In a retrospective study of patients with blunt trauma undergoing CPR in the emergency department, Endo et al. (2017) found higher survival to discharge for CCCPR (3.6% v. 1.8%) and 24 hour survival (9.6% v. 5.6%) when compared to OCCPR.9 With propensity matching, significantly lower odds of survival to discharge and survival at 24 hours were found with OCCPR. However, it was difficult to determine why OCCPR was initiated in patients and made it challenging to compare the two groups. In a later study, Endo et al. (2020) found that OCCPR was associated with survival to discharge in trauma patients with signs of life upon hospital arrival when compared to CCCPR (15.2% v. 11.7%).10 This association persisted during logistic regression analysis and propensity score matching. Outcome 4: Surrogate markers of perfusion While it was not the most critical outcome examined, there have been numerous experimental studies in dogs evaluating surrogate markers of perfusion with OCCPR, many of which suggest a benefit over CCCPR (low quality of evidence, downgraded for serious imprecision). Many studies in dogs with induced ventricular fibrillation found higher arterial pressures, carotid blood flow, cardiac output, cerebral perfusion, and/or coronary perfusion pressure in OCCPR compared to CCCPR.4,5,11–16 Kern et al. (1991) demonstrated that OCCPR after 40 minutes of ventricular fibrillation in dogs resulted in better arterial pressures and coronary perfusion than CCCPR after 20 minutes of ventricular fibrillation.17 Weiser et al. (1962) found that average cardiac output was significantly higher in OCCPR (55%) when compared to CCCPR (22%).18 The difference in cardiac output between OCCPR and CCCPR was particularly pronounced in dogs greater than 10 kg. In a study by Rieder et al. (1985) of 10 dogs in which cardiac arrest was induced via potassium chloride induction while undergoing a laparotomy, OCCPR resulted in significantly higher cardiac index, MAP, and carotid blood flow when compared to CCCPR.19 A transdiaphragmatic approach in which one hand through the diaphragm compressed the heart against the sternum while the other hand compressed the sternum externally resulted in optimal hemodynamics over other techniques. Two additional studies demonstrated reduced brain injury via histopathological examination with OCCPR when compared with CCCPR.6,20

Treatment recommendation

We recommend open-chest CPR (OCCPR) in dogs and cats with abdominal organs or substantial accumulations of fluid or air in the pleural or pericardial spaces (strong recommendation, expert opinion). We recommend direct cardiac massage in dogs and cats undergoing abdominal or thoracic surgery (strong recommendation, low quality of evidence). We suggest OCCPR in dogs and cats with penetrating thoracic trauma or rib fractures at or near the chest compression point (weak recommendation, very low quality of evidence). In medium and large-breed round-chested and wide-chested dogs in which OCCPR is feasible and clients are amenable to the procedure, we recommend that CCCPR be started immediately and OCCPR be started as soon as possible (strong recommendation, low quality of evidence) We suggest not attempting OCCPR in cats and small dogs (< 15kg) that do not have pleural or pericardial disease, penetrating thoracic trauma or are not undergoing abdominal or thoracic surgery (weak recommendation, expert opinion) We recommend discussing the pros and cons of OCCPR in any dog at risk of CPA and obtaining a “CPR code” at the time of hospitalization if OCCPR is offered by the practice and is indicated. (strong recommendation, expert opinion)

Justification of treatment recommendation

Many but not all experimental studies in dogs demonstrated improved neurologic outcome, survival, ROSC, and hemodynamics with open-chest CPR (OCCPR) when compared to closed-chest CPR (CCCPR). These findings were especially profound for large dogs and dogs already undergoing laparotomy. The recommendation is complicated, however, by observational studies in people that have largely failed to demonstrate a benefit with OCCPR when compared to CCCPR. Given the positive results in the experimental studies in dogs, the committee recommends OCCPR as soon as possible in medium to large breed round-chested or wide-chested dogs in which OCCPR is feasible. Factors that could reduce feasibility of OCCPR in medium and large round-chested and wide-chested dogs include owner consent, local practice limitations that would limit the required post-ROSC care, and rescuer OCCPR procedure competence. In addition, considering the likely increased efficacy of CCCPR in keel-chested medium and large breed dogs, the committee thinks it is reasonable to default to CCCPR in these patients. Although outcomes are better with OCCPR in this subset of animals, the committee recognizes that even in practices with the skill set and facilities required for the procedure, it is likely that OCCCPR will continue to be a rarely performed procedure due to the invasiveness, client preference, and intensive after-care required. Given the likely futility of CCCPR in dogs and cats with pleural or pericardial fluid, air or abdominal organ displacement and the lack of feasibility of closed chest compressions in dogs and cats that arrest during laparotomy or thoracotomy, a stronger recommendation for OCCPR is made in these circumstances.

Knowledge gaps

The optimal timing for intervention with open-chest CPR (OCCPR) for dogs and cats with cardiopulmonary arrest is unknown. It is unknown at what weight OCCPR should be considered as a primary intervention in dogs with cardiopulmonary arrest. The diseases for which OCCPR should be considered in dogs and cats are poorly described. The appropriate time to intervene with OCCPR in dogs and cats with cardiopulmonary arrest is considered a high-priority knowledge gap in the veterinary literature.

References:

1. Fletcher DJ, Militello R, Schoeffler GL, Rogers CL. Development and evaluation of a high-fidelity canine patient simulator for veterinary clinical training. J Vet Med Educ. 2012;39(1):7-12.

2. Schulz-Drost S, Merschin D, Gümbel D, et al. Emergency department thoracotomy of severely injured patients: an analysis of the TraumaRegister DGU(®). Eur J Trauma Emerg Surg. 2020;46(3):473-485.

3. Anthi A, Tzelepis GE, Alivizatos P, Michalis A, Palatianos GM, Geroulanos S. Unexpected cardiac arrest after cardiac surgery: incidence, predisposing causes, and outcome of open chest cardiopulmonary resuscitation. Chest. 1998;113(1):15-19.

4. Kern KB, Sanders AB, Badylak SF, et al. Long-term survival with open-chest cardiac massage after ineffective closed-chest compression in a canine preparation. Circulation. 1987;75(2):498-503.

5. Bircher N, Safar P. Cerebral preservation during cardiopulmonary resuscitation. Crit Care Med. 1985;13(3):185-190.

6. Benson DM, O’Neil B, Kakish E, et al. Open-chest CPR improves survival and neurologic outcome following cardiac arrest. Resuscitation. 2005;64(2):209-217.

7. Suzuki K, Inoue S, Morita S, et al. Comparative Effectiveness of Emergency Resuscitative Thoracotomy versus Closed Chest Compressions among Patients with Critical Blunt Trauma: A Nationwide Cohort Study in Japan. PLoS One. 2016;11(1):e0145963.

8. Prieto JM, Van Gent JM, Calvo RY, et al. Nationwide analysis of resuscitative thoracotomy in pediatric trauma: Time to differentiate from adult guidelines? J Trauma Acute Care Surg. 2020;89(4):686-690.

9. Endo A, Shiraishi A, Otomo Y, Tomita M, Matsui H, Murata K. Open-chest versus closed-chest cardiopulmonary resuscitation in blunt trauma: analysis of a nationwide trauma registry. Crit Care. 2017;21(1):169.

10. Endo A, Kojima M, Hong ZJ, Otomo Y, Coimbra R. Open-chest versus closed-chest cardiopulmonary resuscitation in trauma patients with signs of life upon hospital arrival: a retrospective multicenter study. Crit Care. 2020;24(1):541.

11. DeBehnke DJ, Angelos MG, Leasure JE. Comparison of standard external CPR, open-chest CPR, and cardiopulmonary bypass in a canine myocardial infarct model. Ann Emerg Med. 1991;20(7):754-760.

12. Arai T, Dote K, Tsukahara I, Nitta K, Nagaro T. Cerebral blood flow during conventional, new and open-chest cardio-pulmonary resuscitation in dogs. Resuscitation. 1984;12(2):147-154.

13. Bircher N, Safar P, Stewart R. A comparison of standard, “MAST”-augmented, and open-chest CPR in dogs. A preliminary investigation. Crit Care Med. 1980;8(3):147-152.

14. Barsan WG, Levy RC. Experimental design for study of cardiopulmonary resuscitation in dogs. Ann Emerg Med. 1981;10(3):135-137.

15. Sanders AB, Kern KB, Ewy GA, Atlas M, Bailey L. Improved resuscitation from cardiac arrest with open-chest massage. Annals of Emergency Medicine. 1984;13(9 PART 1):672-675.

16. Fleisher G, Sagy M, Swedlow DB, Belani K. Open- versus closed-chest cardiac compressions in a canine model of pediatric cardiopulmonary resuscitation. Am J Emerg Med. 1985;3(4):305-310.

17. Kern KB, Sanders AB, Janas W, et al. Limitations of open-chest cardiac massage after prolonged, untreated cardiac arrest in dogs. Annals of Emergency Medicine. 1991;20(7):761-767.

18. Weiser FM, Adler LN, Kuhn LA. Hemodynamic effects of closed and open chest cardiac resuscitation in normal dogs and those with acute myocardial infarction. The American Journal of Cardiology. 1962;10(4):555-561.

19. Rieder CF, Crawford BG, Iliopoulos JI, Thomas JH, Pierce GE, Hermreck AS. A study of the techniques of cardiac massage with the abdomen open. Surgery. 1985;98(4):824-830.

20. Badylak SF, Kern KB, Tacker WA, Ewy GA, Janas W, Carter A. The comparative pathology of open chest vs. mechanical closed chest cardiopulmonary resuscitation in dogs. Resuscitation. 1986;13(4):249-264.

Supplemental:

Outcome: Favorable neurologic outcome

0 Clinical Trials

1 Observational studies

Anthi (1998): Unexpected cardiac arrest after cardiac surgery

● 29 patients with cardiac arrest within 24 hours after cardiac surgery

● Closed chest CPR performed initially for 3-5 min, then followed by open chest CPR if needed

● 13 resuscitated with closed chest, 14 with open chest

● All discharged neurologically intact

3 Experimental studies

Kern (1987): Long term survival with open chest cardiac massage after ineffective closed chest compression in a canine preparation

● Ventricular fibrillation induced in 29 mongrel dogs and after 3 min, standard CPR was initiated (using a machine)

● Defibrillation attempted twice after 15 min of fibrillation

● Unsuccessfully defibrillated animals randomized to receive 2 min of closed chest or 2 min of open chest

● Improved ROSC, 24 hour survival, and 7 day survival with open chest. No difference in neurological scores though small population may have limited findings

Bircher (1985): Cerebral preservation during CPR

● Ventricular fibrillation in 32 dogs for 4 minutes

● Subdivided into four groups: immediate defibrillation; 30 min of standard CPR, simultaneous ventilation compression CPR, open chest CPR

● After 30 min, drug therapy and defibrillation attempted

● Control: ROSC in all and nearly normal neuro deficit scores at 24 hours

● Standard: 6/8 restored and ⅝ had severe neuro damage and did not survive 24 hours

● SVC: ⅝ ROSC, but all brain dead and none survived 24 hours

● Open chest: 7 survived 24 hours and neuro scores not significantly different from control group

Benson (2005): Open chest CPR improves survival and neurologic outcome following cardiac arrest

● Cardiac arrest induced via KCl in 12 dogs. Received 5 min of non-intervention and then randomized to receive either closed or open chest CPR for 15 min and were then resuscitated.

● All open chest CPR dogs resuscitated and behaviorally normal at 72 hours. Only 3/7 closed chest CPR dogs survived and 2 had incapacitating deficits.

● Neuro score did include those who died in the closed chest group which may have swayed results

Outcome: Survival to discharge

0 Clinical Trials

5 Observational studies

Schulz-Drost (2020): Emergency department thoracotomy of severely injured patients: an analysis of the TraumaRegister DGU

● Focus was on all ED thoracotomy but a subset underwent EDT for cardiac arrest

● For these, survival rate of 4.8% with blunt trauma but 20.7% for penetrating trauma with EDT

Suzuki (2016): Comparative effectiveness of emergency resuscitative thoracotomy versus closed chest compressions among patients with critical blunt trauma: a nationwide cohort study in Japan

● Retrospective study from Japan in which 1377 blunt trauma patients who received CPR in the ED or OR. 484 received emergency resuscitative thoracotomy and 893 received closed chest CPR

● Lower survival with open chest (4.5% v 17.5%) and 28 day survival (1.2% v 6%). This was independently associated on multivariable analysis

Prieto (2020): Nationwide analysis of resuscitative thoracotomy in pediatric trauma: time to differentiate from adult guidelines?

● Retrospective analysis of patients 16 years or younger who underwent EDT within 30 minutes of arrival in US database

● 53 patients with no signs of life at presentation and received EDT, none of whom survived

Endo (2017): Open chest versus closed chest CPR in blunt trauma: analysis of a nationwide trauma registry

● Retrospective study for patients with blunt trauma and undergoing CPR in an ED divided into open and closed chest groups

● 6510 patients (2192 open chest, 4318 closed chest) analyzed

● Higher survival in closed chest at 24 hours (9.6% v 5.6%) and in hospital survival (3.6% v 1.8%)

● With propensity matching, significantly lower odds of survival to hospital discharge for open chest CPR and survival over 24 hours

● However, difficult to determine why open chest was initiated making it difficult to compare these two groups and application of open chest differed significantly based on hospital

Endo (2020): Open chest versus closed chest CPR in trauma patients with signs of life upon hospital arrival: a retrospective multicenter study

● Retrospective cohort study in USA with severe trauma patients who had SOL upon arrival and received CPR within first 6 hours of ED admission

● 2682 patients found (1032 open chest, 1650 closed chest). Open chest associated with survival to discharge (15.2% v 11.7%), including for logistic regression analysis and propensity score matching analysis

4 Experimental studies

Kern (1987): Long term survival with open chest cardiac massage after ineffective closed chest compression in a canine preparation

● Ventricular fibrillation induced in 29 mongrel dogs and after 3 min, standard CPR was initiated (using a machine)

● Defibrillation attempted twice after 15 min of fibrillation

● Unsuccessfully defibrillated animals randomized to receive 2 min of closed chest or 2 min of open chest

● Improved ROSC, 24 hour survival (12/14 v 4/14), and 7 day survival (11/14 v 4/14) with open chest.

Benson (2005): Open chest CPR improves survival and neurologic outcome following cardiac arrest

● Cardiac arrest induced via KCl in 12 dogs. Received 5 min of non-intervention and then randomized to receive either closed or open chest CPR for 15 min and were then resuscitated.

● All open chest CPR dogs resuscitated and behaviorally normal at 72 hours. Only 3/7 closed chest CPR dogs survived and 2 had incapacitating deficits.

DeBehnke (1991): Comparison of standard external CPR, open chest CPR, and cardiopulmonary bypass in a canine myocardial infarct model

● 26 dogs received left anterior descending coronary artery occlusion followed by four minutes of ventricular fibrillation and eight minutes of mechanical CPR. At 12 minutes, randomized to one of three groups: open chest CPR, cardiopulmonary bypass, or standard closed chest CPR

● ROSC in 9/9 bypass, 2/8 closed chest, and 6/9 open chest CPR

● Survival to four hours in 3/9 bypass and open chest and 2/8 closed chest. No difference in survival

Kern (1991): Limitations of open chest cardiac massage after prolonged, untreated cardiac arrest in dogs

● 20 mongrel dogs with 10, 20, or 40 minutes of untreated ventricular fibrillation. Open chest CPR compared to closed chest CPR for those that did not respond to initial defibrillation. Defibrillation performed every 3 minutes (closed chest received external and open chest received internal)

● Animal considered resuscitated after 12 min if systolic BP > 40 mmHg

● Open chest better for ROSC (5/5 v ⅕) after 20 min of vfib. No ROSC with open chest after 40 min

● Poor survival if open chest initiated following 20 or more minutes of untreated cardiac arrest (⅘ survived if began after 10 untreated minutes, ⅕ if after 20 minutes)

Outcome: ROSC

0 Clinical Trials

3 Observational studies

Bradley (2016): Open chest cardiac massage offers no benefit over closed chest compressions in patients with traumatic cardiac arrest

● Prospective observational study that enrolled patients with traumatic cardiac arrest. Enrolled 33 patients (16 open, 17 closed).

● No difference in ROSC between groups, but couldn’t control for confounders

DiGiacomo (2017): Thoracotomy in the ED for resuscitation of the mortally injured

● Retrospective of all patients who underwent ED RT as an adjunct to resuscitation. No control group

● 68 patients identified, 27 of whom achieved ROSC

● Patients without signs of life at the scene who arrived without signs of life did not respond to EDRT

● Only one long term survivor who had mild cognitive deficits

Anthi (1998): Unexpected cardiac arrest after cardiac surgery

● 29 patients with cardiac arrest within 24 hours after cardiac surgery

● Closed chest CPR performed initially for 3-5 min, then followed by open chest CPR if needed

● 13 resuscitated with closed chest, 14 with open chest

● All discharged neurologically intact

● No control group to determine if ROSC would have been successful with longer closed chest CPR

7 Experimental studies

Kern (1987): Long term survival with open chest cardiac massage after ineffective closed chest compression in a canine preparation

● Ventricular fibrillation induced in 29 mongrel dogs and after 3 min, standard CPR was initiated (using a machine)

● Defibrillation attempted twice after 15 min of fibrillation

● Unsuccessfully defibrillated animals randomized to receive 2 min of closed chest or 2 min of open chest

● Improved ROSC (14/14 v 5/14), 24 hour survival (12/14 v 4/14), and 7 day survival (11/14 v 4/14) with open chest.

Badylak (1986): The comparative pathology of open chest vs mechanical closed chest CPR in dogs

● VF induced in 28 healthy mongrel dogs, no treatment for 3 min, then mechanical closed chest CPR given for 12 min, followed by defibrillation twice

● Then. 2 groups: closed chest or open chest CPR with ALS

● ROSC and survival to euthanasia at 7 days: 4/14 with closed chest, 11/14 with open chest

Benson (2005): Open chest CPR improves survival and neurologic outcome following cardiac arrest

● Cardiac arrest induced via KCl in 12 dogs. Received 5 min of non-intervention and then randomized to receive either closed or open chest CPR for 15 min and were then resuscitated.

● All open chest CPR dogs resuscitated and behaviorally normal at 72 hours. Only 3/7 closed chest CPR achieved ROSC and 2 had incapacitating deficits.

Kern (1991): Limitations of open chest cardiac massage after prolonged, untreated cardiac arrest in dogs

● 20 mongrel dogs with 10, 20, or 40 minutes of untreated ventricular fibrillation. Open chest CPR compared to closed chest CPR for those that did not respond to initial defibrillation. Defibrillation performed every 3 minutes (closed chest received external and open chest received internal)

● Animal considered resuscitated after 12 min if systolic BP > 40 mmHg

● Open chest better for ROSC (5/5 v ⅕) after 20 min of vfib. No ROSC with open chest after 40 min

DeBehnke (1991): Comparison of standard external CPR, open chest CPR, and cardiopulmonary bypass in a canine myocardial infarct model

● 26 dogs received left anterior descending coronary artery occlusion follwed by four minutes of ventricular fibrillation and eight minutes of mechanical CPR. At 12 minutes, randomized to one of three groups: open chest CPR, cardiopulmonary bypass, or standard closed chest CPR

● ROSC in 9/9 bypass, 2/8 closed chest, and 6/9 open chest CPR

Sanders (1984): Improved resuscitation from cardiac arrest with open-chest massage

● 10 mongrel dogs were fibrillated. CPR was initiated and continued for 15 minutes. Patients with coronary perfusion pressures <30 mmHg were then excluded (no dogs fell into this category)

● Half of the remaining underwent open chest CPR and half underwent closed chest CPR. They were all then defibrillated at 19 minutes

● 0/5 in closed chest achieved ROSC, ⅘ with open chest achieved ROSC

Bircher (1985): Cerebral preservation during CPR

● Ventricular fibrillation in 32 dogs for 4 minutes

● Subdivided into four groups: immediate defibrillation; 30 min of standard CPR, simultaneous ventilation compression CPR, open chest CPR

● After 30 min, drug therapy and defibrillation attempted

● Control: ROSC in all and nearly normal neuro deficit scores at 24 hours

● Standard: 6/8 restored and ⅝ had severe neuro damage and did not survive 24 hours

● SVC: ⅝ ROSC, but all brain dead and none survived 24 hours

Outcome: Surrogate markers of perfusion

0 Clinical Trials

1 Observational studies

Bradley (2016): Open chest cardiac massage offers no benefit over closed chest compressions in patients with traumatic cardiac arrest

● Prospective observational study that enrolled patients with traumatic cardiac arrest. Enrolled 33 patients (16 open, 17 closed).

● No difference in ROSC between groups, but couldn’t control for confounders

● With time matched comparisons, ETCO2 was not different for initial, final, peak, mean, or median values between open and closed chest

14 Experimental studies

Arai (1984): Cerebral blood flow during conventional, new, and open chest CPR in dogs

● VF induced in 15 dogs. Chest compressions and ventilation were done manually (by one individual). Closed chest CPR was initiated for 30 seconds, then new CPR (chest compressions with high airway pressure ventilation) for 30 seconds. This was attempted 3 times and then open chest CPR was initiated

● Open chest had highest MAP, carotid blood flow, and CPP, along with sinus blood flow. No difference between closed and new CPR

Bircher (1985): Cerebral preservation during CPR

● Ventricular fibrillation in 32 dogs for 4 minutes

● Subdivided into four groups: immediate defibrillation; 30 min of standard CPR, simultaneous ventilation compression CPR, open chest CPR

● After 30 min, drug therapy and defibrillation attempted

● Highest MAP and lower CVP than other groups

Redding (1961): A comparison of open chest and closed chest cardiac massage in dogs

● VF induced in 20 healthy mongrel dogs. 30 seconds later, closed chest initiated in 10 dogs and open chest for 10 minutes for 20 minutes, followed by defibrillation

● Similar aortic and carotid flow between groups but stats not performed

Bircher (1980): A comparison of standard, MAST augmented, and open chest CPR in dogs

● VF induced in 9 dogs. Closed chest CPR initiated after 2 min of VF. After 2 hours, open chest CPR attempted

● Open chest significantly increased arterial and perfusion pressures and more than doubled common carotid arterial blood flow

Rieder (1985): A study of the techniques of cardiac massage with the abdomen open

● 10 mongrel dogs were anesthetized and had a midline laparotomy performed. KCl was administered to cause CPA

● Closed chest CPR was initiated in lateral recumbency. Six cardiac compression techniques were then trialed

● Highest CI, MAP, and Carotid blood flow with transdiaphragmatic retrocardiac massage (compression of the heart against the sternum with one hand through the diaphragm). Significantly better than closed chest

DeBehnke (1991): Comparison of standard external CPR, open chest CPR, and cardiopulmonary bypass in a canine myocardial infarct model

● 26 dogs received left anterior descending coronary artery occlusion follwed by four minutes of ventricular fibrillation and eight minutes of mechanical CPR. At 12 minutes, randomized to one of three groups: open chest CPR, cardiopulmonary bypass, or standard closed chest CPR

● Coronary perfusion pressure significantly higher with bypass and open chest compared to closed chest

● Ratio of necrotic to ischemic myocardium at 4 hours lower with bypass and open chest than closed chest

Barsan (1981): Experimental design for study of CPR in dogs

● 13 mongrel dogs with induced VF were divided into three groups: 1) 30 seconds of VF followed by closed chest compressions (n=2) 2) 2 dogs with automatic gas powered chest compressor and 3) 9 dogs with open chest (4 with left thoracotomy, 5 with median sternotomy)

● External CO approximately 17% while internal averaged 35% and also had a higher average systolic blood pressure (approximately 50 v 80 mmHg)

Weiser (1962): Hemodynamic effects of closed and open chest cardiac resuscitation in normal dogs and those with acute myocardial infarction

● 22 mongrel dogs anesthetized and divided into two groups: 12 normal and 10 induced myocardial infarction

● VF induced in those that did not develop it. Group A: 15 seconds of VF followed by closed chest for 5 min, followed by open chest compression

● Group B: 7 with only open, 1 closed then open, 2 only closed

● CO averaged 22% with closed chest and 55% in open chest. CO significantly higher once converted to open in dogs that had both procedures

● In dogs <10 kg, closed chest averaged CO of 39% while >10 kg was 13% and 75% for open chest for <10 kg and 49% for open chest for > 10 kg

Sanders (1984): Improved resuscitation from cardiac arrest with open-chest massage

● 10 mongrel dogs were fibrillated. CPR was initiated and continued for 15 minutes. Patients with coronary perfusion pressures <30 mmHg were then excluded (no dogs fell into this category)

● Half of the remaining underwent open chest CPR and half underwent closed chest CPR. They were all then defibrillated at 19 minutes

● Aortic and coronary perfusion pressures significantly higher with open chest within the first 2 minutes

Kern (1991): Limitations of open chest cardiac massage after prolonged, untreated cardiac arrest in dogs

● 20 mongrel dogs with 10, 20, or 40 minutes of untreated ventricular fibrillation. Open chest CPR compared to closed chest CPR for those that did not respond to initial defibrillation. Defibrillation performed every 3 minutes (closed chest received external and open chest received internal)

● Animal considered resuscitated after 12 min if systolic BP > 40 mmHg

● Open chest significantly increased aortic systolic and diastolic pressures and increased RAP, as well as coronary perfusion pressure

● Open chest at 40 minutes of VF provided better SAP and DAP and CPP than 20 minutes of VF followed by closed chest

Kern (1987): Long term survival with open chest cardiac massage after ineffective closed chest compression in a canine preparation

● Ventricular fibrillation induced in 29 mongrel dogs and after 3 min, standard CPR was initiated (using a machine)

● Defibrillation attempted twice after 15 min of fibrillation

● Unsuccessfully defibrillated animals randomized to receive 2 min of closed chest or 2 min of open chest

● Open chest: significantly increased SAP, DAP, CPP

Fleisher (1985): Open versus closed chest cardiac compression in a canine model of pediatric CPR

● 6-12 week old puppies who were 2-8 kg in weight were used. CPA induced with KCl and allowed for 3 minutes. CPR then initiated with open or closed chest CPR (5 in each group)

● Open chest produced greater CO and higher cerebral blood flow but no difference in SAP

● Closed chest: ⅗ experienced liver lacerations

Benson (2005): Open chest CPR improves survival and neurologic outcome following cardiac arrest

● Cardiac arrest induced via KCl in 12 dogs. Received 5 min of non-intervention and then randomized to receive either closed or open chest CPR for 15 min and were then resuscitated.

● All open chest CPR dogs resuscitated and behaviorally normal at 72 hours. Only 3/7 closed chest CPR dogs survived and 2 had incapacitating deficits.

● Histology of survivors for open chest showed little to no injury of the brain, histology of survivors of closed chest revealed moderate to severe lesions

Badylak (1986): The comparative pathology of open chest vs mechanical closed chest CPR in dogs

● VF induced in 28 healthy mongrel dogs, no treatment for 3 min, then mechanical closed chest CPR given for 12 min, followed by defibrillation twice

● Then. 2 groups: closed chest or open chest CPR with ALS

● Open chest caused more severe histopath scores to skin, chest wall muscle, SC tissues, and pleura, but lower scores for the brain