• 
• 

Three Part Question

In [patients in cardiac arrest] is [mechanical CPR or standard manual CPR] more effective at achieving [survival to hospital discharge]?

Clinical Scenario

A 56 year-old man suffers a witnessed out of hospital cardiac arrest. He is given immediate bystander cardiopulmonary resuscitation (CPR). A paramedic ambulance crew arrives after 8 minutes. The first recorded cardiac rhythm shows ventricular fibrillation. The ambulance crew continue CPR in accordance with current Advanced Life Support guidelines. Initial resuscitation attempts including three defibrillation attempts fail. The paramedic team is equipped with and fully trained in the use of a mechanical CPR device and this is applied and the patient transferred to the nearest emergency department. You wonder whether mechanical CPR or manual CPR is more effective at achieving a restoration of spontaneous circulation and improving the patient's chances of leaving hospital alive.

Search Strategy

Medline (between 1st January 1950 and 1st December 2007)
Embase (between 1980 and 1st December 2007)
CinAHL (between 1982 and 1st December 2007)
The Cochrane Library (Issue 4 2007)
Medline, Embase, CinAHL: [mechanical.mp OR device.mp OR automated.mp OR LUCAS.mp or AUTOPULSE.mp] AND [chest compression.mp OR CPR.mp ) AND (exp asystole.mp OR cardiac adj arrest]
Cochrane: [Mechanical OR AUTOPULSE OR LUCAS] AND [chest compression OR CPR] AND [asystole OR cardiac arrest OR arrest]

Search Outcome

The search found 429 articles in the Medline, Embase and CinAHL databases. Excluding case studies, 9 of these articles were relevant to this investigation and were analysed further.
Searching the Cochrane Library identified 15 articles, 5 of which were relevant to this investigation. All 5 of these articles had been identified by the previous search.
Eight out of the 9 studies analysed used either the AutoPulse Resuscitation System or the Lund University Cardiopulmonary Assist System (LUCAS) as a mechanical CPR device

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Axelsson et al, 2006, Sweden	328 patients suffering from cardiac arrest (159 LUCAS vs 169 to manual CPR)	Prospective, multicenter, cluster randomised trial	Return of spontaneous circulation	51% vs 51%	LUCAS device used in only 66% of cases randomised to mechanical group Numerous exclusion criteria
			Hospital admission alive	38% vs 37%
Casner et al, 2005, USA	162 patients suffering out-of-hospital cardiac arrest treated by San Francisco Fire Department (69 AutoPulse vs 93 manual)	Case match using retrospective chart review	Patient arrival to emergency department with sustained ROSC determined by measurable non-invasive blood pressure	39% vs 29% (p=0.003)	Subjective endpoint No matching for other morbidities No information about personnel training in use of AutoPulse device No measure of effectiveness of manual CPR AutoPulse only deployed with late responders
Dickinson et al, 1996, USA	20 patients suffering from out-of-hospital cardiac arrest (10 Thumper vs 10 to manual CPR)	Prospective, randomised trial	Decrease in ETCO2 from 5 minutes after starting CPR to initiation of patient transfer to hospital	0 vs 8 (p<0.04)	Questionable clinical significance (no survivors) Odd/even day randomisation 30% of patients excluded from mechanical CPR group (2 machine failure, 1 extubated)
Hallstrom et al, 2006, USA & Canada	1071 patients suffering from out-of-hospital cardiac arrest (554 AutoPulse vs 517 manual CPR)	Prospective multicentre, cluster-randomised trial	Survival to 4 hours post emergency call	28.5% vs 29.5% (p=0.74)	Study stopped following first planned interim monitoring No information on quality of manual chest compressions Longer time (2.1 minutes longer) from 911 call to first shock in VF/VT patients in AutoPulse group
			Survival to hospital discharge	5.8% vs 9.9% (p=0.06 after adjustment)
			Cerebral performance category of 1 or 2 at hospital discharge	3.1% vs 7.5% (p=0.006)
Krep et al, 2007, Germany	46 patients suffering from out of hospital cardiac arrest treated with AutoPulse	Prospective observational case series	Return of spontaneous circulation	39.1%	No comparison group Mixture of bystander CPR and no bystander CPR 3 cases AutoPulse CPR considered insufficient and replaced with manual CPR
Larsen et al, 2007, Norway	13 patients receiving concomitant CPR (for cardiac arrest or severe hypotension and bradycardia) using LUCAS and coronary intervention	Case series	Mean systolic blood pressure	81±23 mmHg	Mixture of patients with cardiac arrests and hypotension and bradycardia Only 3 patients survived procedure and none discharged from hospital
			Mean diastolic blood pressure	34±21 mm Hg
Ong et al, 2006, USA	783 patients suffering from out-of-hospital non-traumatic arrest (284 AutoPulse vs 499 manual)	Observational cohort study with intention-to-treat analysis	Return of spontaneous circulation	34.5% vs 20.2%	Only 210 patients out of the 284 patients in mechanical CPR group received mechanical CPR Faster initial response time in mechanical CPR group (mean 26 seconds faster)
			Survival to hospital admission	20.9% vs 11.1%
			Survival to hospital discharge	9.7% vs 2.9%
Steen et al, 2005, Sweden	100 patients suffering from out-of-hospital cardiac arrest treated with LUCAS	Case series	Return of spontaneous circulation	31%	No comparison group
			Survival after 30 days	7%
Timerman et al, 2004, Brazil and USA	31 patients suffering in-hospital cardiac arrest. Subjects included following 10 mins of failed standard ALS. Received alternating manual CPR (100 compressions/min) and AutoPulse CPR (60 compressions/min) for 90 secs each.	Case series	Peak aortic pressure (mmHg)	153±28 mmHg vs 115±42 mmHgv	Usable pressure signals present in only 16 patients Differing rates of compression No analysis of clinical outcome
			Peak right atrial pressure	129±32 mmHg vs 83±40 mmHg
			Coronary perfusion pressure	20±23 mmHg vs 15±11 mmHg

Comment(s)

There is strong evidence that good quality CPR improves outcome following cardiac arrest. Providing good quality CPR for long periods of time or during the transport of patients may be difficult. Mechanical CPR devices have the potential to overcome these problems. The two most commonly used mechanical CPR devices are the AutoPulse Resuscitation System and the LUCAS. Trials comparing mechanical and manual CPR vary widely in quality and often study different patient groups, mechanical CPR devices and protocols. As of yet, no trial has attempted to compensate for the Hawthorne effect on provision of manual CPR. Searching the ClinicalTrials.gov database in January 2008 identified one new ongoing studies comparing mechanical (AutoPulse) and manual CPR (NCT00597207).

Clinical Bottom Line

There is currently no evidence supporting the routine use of mechanical CPR devices for patients suffering out of hospital cardiac arrest.

References

1. Axelsson, C., Nestin, J., Svensson, L., Axelsson, Å. & Herlitz, J. Clinical consequences of the introduction of mechanical chest compression in the EMS system for treatment of out-of-hospital cardiac arrest—A pilot study. Resuscitation 2006; 71, 47-55.
2. Casner, M., Andersen, D. & Isaacs, S. The impact of a new CPR assist device on rate of return of spontaneous circulation in out-of-hospital cardiac arrest. Prehospital Emergency Care, 2005; 9, 61-67.
3. Dickinson, E., Verdile, V., Schneider, R. & Salluzzo, R. Effectiveness of mechanical versus manual chest compressions in out-of-hospital cardiac arrest resuscitation: A pilot study. The American Journal of Emergency Medicine, 1996; 16, 289-292.
4. Hallstrom, A., Rea, T. D., Sayre, M. R., Christenson, J., Anton, A. R., Mosesso Jr, V. N., Van Ottingham, L., Olsufka, M., Pennington, S. & White, L. J. Manual Chest Compression vs Use of an Automated Chest Compression Device During Resuscitation Following Out-of-Hospital Cardiac Arrest: A Randomized Trial. JAMA, 2006; 295, 2620.
5. Krep, H., Mamier, M., Breil, M., Heister, U., Fischer, M. & Hoeft, A. Out-of-hospital cardiopulmonary resuscitation with the AutoPulse™ system: A prospective observational study with a new load-distributing band chest compression device. Resuscitation, 2007; 73, 86-95.
6. Larsen, A. I., Hjornevik, A. S., Ellingsen, C. L. & Nilsen, D. W. Cardiac arrest with continuous mechanical chest compression during percutaneous coronary intervention A report on the use of the LUCAS device. Resuscitation. 2007; 75; 3: 454-459
7. Ong, M. E. H., Ornato, J. P., Edwards, D. P., Dhindsa, H. S., Best, A. M., Ines, C. S., Hickey, S., Clark, B., Williams, D. C. & Powell, R. G. Use of an Automated, Load-Distributing Band Chest Compression Device for Out-of-Hospital Cardiac Arrest Resuscitation. JAMA 2006; 295, 2629.
8. Steen, S., Sjöberg, T., Olsson, P. & Young, M. Treatment of out-of-hospital cardiac arrest with LUCAS, a new device for automatic mechanical compression and active decompression resuscitation. Resuscitation 2005; 67, 25-30.
9. Timerman, S., Cardoso, L. F., Ramires, J. A. & Halperin, H. Improved hemodynamic performance with a novel chest compression device during treatment of in-hospital cardiac arrest. Resuscitation. 2004;61, 273-280.