Intervention Review

You have free access to this content

Active chest compression-decompression for cardiopulmonary resuscitation

  1. Carmelo Lafuente-Lafuente1,*,
  2. María Melero-Bascones2

Editorial Group: Cochrane Heart Group

Published Online: 20 SEP 2013

Assessed as up-to-date: 14 JAN 2013

DOI: 10.1002/14651858.CD002751.pub3


How to Cite

Lafuente-Lafuente C, Melero-Bascones M. Active chest compression-decompression for cardiopulmonary resuscitation. Cochrane Database of Systematic Reviews 2013, Issue 9. Art. No.: CD002751. DOI: 10.1002/14651858.CD002751.pub3.

Author Information

  1. 1

    Groupe Hospitalier Pitié-Salpêtrière-Charles Foix, AP-HP, Université Pierre et Marie Curie (Paris 6), Service de Gériatrie à Orientation Cardiologique et Neurologique, Ivry-sur-Seine, Ile-de-France, France

  2. 2

    Complejo Hospitalario Universitario de Albacete, Servicio de Medicina Interna, Albacete, Spain

*Carmelo Lafuente-Lafuente, Service de Gériatrie à Orientation Cardiologique et Neurologique, Groupe Hospitalier Pitié-Salpêtrière-Charles Foix, AP-HP, Université Pierre et Marie Curie (Paris 6), 7 Avenue de la République, Ivry-sur-Seine, Ile-de-France, 94205, France. c.lafuente@nodo3.net. carmelo.lafuente@cfx.ap-hop.paris.fr.

Publication History

  1. Publication Status: Stable (no update expected for reasons given in 'What's new')
  2. Published Online: 20 SEP 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Goralski 1998

MethodsRandomised controlled trial


Participantsn = 150

Mean age: 61.5 years

Male: 72%

Exclusions: trauma victims

VF: 15%

Cardiac cause of heart arrest: 70%


InterventionsACDR vs STR

Physician in responder team

Mean response time: 16 minutes

Mean adrenaline dose: 12.5 mg


Outcomes
  • Immediate and at-hospital discharge mortality


NotesPrehospital
Orleans, Tours (France)
1995 to 1996


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk

Allocation concealment (selection bias)Low risk

Incomplete outcome data (attrition bias)
All outcomes
Low risk

Selective reporting (reporting bias)Low risk

Blinding of participants and personnel (performance bias)
All outcomes
High risk

Blinding of outcome assessment (detection bias)
Mortality
Low risk

Blinding of outcome assessment (detection bias)
Neurological status
Low risk

Luiz 1996

MethodsRandomised controlled trial


Participantsn = 61

Mean age: 69 years

Male: 73%

Exclusions: pregnancy, trauma victims

VF: 27%

Cardiac antecedents: 52%

Pulmonary antecedents: 12.5%


InterventionsACDR vs STR

Physician in responder team
Mean response time: 8 minutes

Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Neurological outcome
  • Complications


NotesPrehospital
Mannheim (Germany)
1993


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskParticipants "were randomised according to a schedule". No more information was available

Allocation concealment (selection bias)Unclear riskNo information about how the randomisation schedule was applied

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data

Selective reporting (reporting bias)Low riskAll outcomes analysed in the review had been reported adequately

Blinding of participants and personnel (performance bias)
All outcomes
High riskBlinding of intervention (use of ACDR device) was not possible

Blinding of outcome assessment (detection bias)
Mortality
Low riskAssessment was not blind, but we did not believe this introduced bias as death is a hard criteria, rarely misattributed

Blinding of outcome assessment (detection bias)
Neurological status
High riskNo blinding; we thought the outcome measurement was likely to have been influenced by lack of blinding

Lurie 1994

MethodsRandomised controlled trial


Participantsn = 130

Mean age: 67 years

Male: 72%

Exclusions: hypothermia, trauma victims

VF: 62%

Cardiac antecedents: 62%


InterventionsACDR vs STR

Only paramedics in responder team

Mean response time: 5.2 minutes

Mean adrenaline dose: 2.7 mg


Outcomes
  • Immediate and at-hospital discharge mortality
  • Complications


NotesPrehospital
St.Paul (USA)
1992 to 1993


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High risk

Allocation concealment (selection bias)High risk

Incomplete outcome data (attrition bias)
All outcomes
Low risk

Selective reporting (reporting bias)Low risk

Blinding of participants and personnel (performance bias)
All outcomes
High risk

Blinding of outcome assessment (detection bias)
Mortality
Low risk

Blinding of outcome assessment (detection bias)
Neurological status
Low risk

Mauer 1996

MethodsRandomised controlled trial


Participantsn = 220

Mean age: 68.7 years

Male: ?

Exclusions: pregnancy, hypothermia, drowning, overdose, trauma victims

VF: 45%

Cardiac antecedents: 61%

Pulmonary antecedents: 12.7%


InterventionsACDR vs STR

Physician in responder team

Mean response time: 9 minutes

Mean adrenaline dose: 4.3 mg


Outcomes
  • Immediate and at-hospital discharge mortality
  • Complications


NotesPrehospital
Mainz (Germany)
1992 to 1995


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskCentralised randomisation list

Allocation concealment (selection bias)Low risk24-hour centralised randomisation

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data

Selective reporting (reporting bias)Low riskAll outcomes analysed in the review had been reported adequately

Blinding of participants and personnel (performance bias)
All outcomes
High riskBlinding of intervention (use of ACDR device) was not possible

Blinding of outcome assessment (detection bias)
Mortality
Low riskAssessment was not blind, but we did not believe this introduced bias as death is a hard criteria, rarely misattributed

Blinding of outcome assessment (detection bias)
Neurological status
Low riskNeurological outcome was not studied in this trial

Nolan 1998

MethodsRandomised controlled trial


Participantsn = 674

Mean age: 67 years

Male: 68%

Exclusions: trauma victims

VF: 60%


InterventionsACDR vs STR

Only paramedics in responder team

Mean response time: 8.7 minutes

Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Neurological outcome
  • Complications


NotesPrehospital
Cardiff, Portsmouth, Bristol (UK)
1994 to 1995


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskInterventions were alternated every 2 months

Allocation concealment (selection bias)High riskAlternation

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data

Selective reporting (reporting bias)Low riskAll outcomes analysed in the review had been reported adequately

Blinding of participants and personnel (performance bias)
All outcomes
High riskBlinding of intervention (use of ACDR device) was not possible

Blinding of outcome assessment (detection bias)
Mortality
Low riskAssessment was not blind, but we did not believe this introduced bias as death is a hard criteria, rarely misattributed

Blinding of outcome assessment (detection bias)
Neurological status
Unclear riskNo information was given about how neurological outcome was assessed

Plaisance 1999

MethodsRandomised controlled trial


Participantsn = 750

Mean age: 58.5 years

Male: 68%

Exclusions: irreversible death, excessive delay

VF: 12%

Cardiac cause of heart arrest: 64%


InterventionsACDR vs STR

Physician in responder team
Mean response time: 9.2 minutes

Mean adrenaline dose: 14.3 mg


Outcomes
  • Immediate mortality
  • Complications


Intention-to-treat analysis was explicitly done


NotesPrehospital
Paris (France)
1993 to 1995


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High risk

Allocation concealment (selection bias)High risk

Incomplete outcome data (attrition bias)
All outcomes
Low risk

Selective reporting (reporting bias)Low risk

Blinding of participants and personnel (performance bias)
All outcomes
High risk

Blinding of outcome assessment (detection bias)
Mortality
Low risk

Blinding of outcome assessment (detection bias)
Neurological status
Low risk

Schwab-Fresno 1995

MethodsRandomised controlled trial


Participantsn = 253

Mean age: 65 years

Male: 61%

Exclusions: trauma victims, hypothermia

VF: 38%


InterventionsACDR vs STR

Only paramedics in responder team

Mean response time: 5.3 minutes

Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Complications


NotesPrehospital
Fresno (USA)
1992 to 1993


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskThe method of CPR was alternated on a weekly basis

Allocation concealment (selection bias)High riskAlternation

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data

Selective reporting (reporting bias)Low riskAll outcomes analysed in the review had been reported adequately

Blinding of participants and personnel (performance bias)
All outcomes
High riskBlinding of intervention (use of ACDR device) was not possible

Blinding of outcome assessment (detection bias)
Mortality
Low riskAssessment was not blind, but we did not believe this introduced bias as death is a hard criteria, rarely misattributed

Blinding of outcome assessment (detection bias)
Neurological status
Low riskNeurological outcome was not studied in this trial

Schwab-S.Francisco 1995

MethodsRandomised controlled trial


Participantsn = 607

Mean age: 65 years

Male: 67%

Exclusions: trauma victims, hypothermia

VF: 18%


InterventionsACDR vs STR
Mean response time: 5.5 minutes
Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Complications


NotesPrehospital
San Francisco (USA)
1992 to 1993


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskHalf the fire stations responding to emergency calls were randomised to ACDR CPR for the entire duration of the study

Allocation concealment (selection bias)High riskOpen allocation

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data

Selective reporting (reporting bias)Low riskAll outcomes analysed in the review had been reported adequately

Blinding of participants and personnel (performance bias)
All outcomes
High riskBlinding of intervention (use of ACDR device) was not possible

Blinding of outcome assessment (detection bias)
Mortality
Low riskAssessment was not blind, but we did not believe this introduced bias as death is a hard criteria, rarely misattributed

Blinding of outcome assessment (detection bias)
Neurological status
Unclear riskNeurological outcome was not studied in this trial

Skogvoll 1999

MethodsRandomised controlled trial


Participantsn = 306

Mean age: 71 years

Male: 73%

Exclusions: trauma victims, intoxication, airway obstruction

VF: 57%


InterventionsACDR vs STR

Physician in responder team

Mean response time: 9 minutes

Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Neurological outcome


Intention-to-treat analysis was explicitly done


NotesPrehospital
Trondheim (Norway)
1994 to 1998


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandom number table

Allocation concealment (selection bias)Low riskSequentially numbered, opaque sealed envelopes

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data

Selective reporting (reporting bias)Low riskAll outcomes analysed in the review had been reported adequately

Blinding of participants and personnel (performance bias)
All outcomes
High riskBlinding of intervention (use of ACDR device) was not possible

Blinding of outcome assessment (detection bias)
Mortality
Low riskAssessment was not blind, but we did not believe this introduced bias as death is a hard criteria, rarely misattributed

Blinding of outcome assessment (detection bias)
Neurological status
Unclear riskNo information is given about who assessed neurological outcome and how

Stiell-Inhospital 1996

MethodsRandomised controlled trial


Participantsn = 773

Mean age: 68.5 years

Male: 57%

Exclusions: trauma, recent sternotomy, operating room

VF: 32%

Cardiac antecedents: 49%

Cardiac cause of heart arrest: 77%

Respiratory cause: 16%


InterventionsACDR vs STR

Physician in responder team

Mean response time: 1.4 minutes

Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Neurological outcome
  • Complications


Intention-to-treat analysis was explicitly done


NotesIn-hospital
Otawa, London (Canada)
1993 to 1995


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk

Allocation concealment (selection bias)Low risk

Incomplete outcome data (attrition bias)
All outcomes
Low risk

Selective reporting (reporting bias)Low risk

Blinding of participants and personnel (performance bias)
All outcomes
High risk

Blinding of outcome assessment (detection bias)
Mortality
Low risk

Blinding of outcome assessment (detection bias)
Neurological status
Low risk

Stiell-Prehospital 1996

MethodsRandomised controlled trial


Participantsn = 1011

Mean age: 68 years

Male: 63%

Exclusions: trauma, excessive delay

VF: 32.5%

Cardiac antecedents: 49%

Cardiac cause of heart arrest: 86%

Respiratory cause: 8%


InterventionsACDR vs STR
Only paramedics in responder team

Mean response time: 5.2 minutes

Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Neurological outcome
  • Complications


Intention-to-treat analysis was explicitly done


NotesPrehospital
Otawa, London (Canada)
1993 to 1995


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk

Allocation concealment (selection bias)Low risk

Incomplete outcome data (attrition bias)
All outcomes
Low risk

Selective reporting (reporting bias)Low risk

Blinding of participants and personnel (performance bias)
All outcomes
High risk

Blinding of outcome assessment (detection bias)
Mortality
Low risk

Blinding of outcome assessment (detection bias)
Neurological status
Low risk

Tucker 1994

MethodsRandomised controlled trial


Participantsn = 53

Mean age: 71 years

Male: 43%

Exclusions: trauma victims, pregnancy

VF: 28%

Cardiac antecedents: 58%


InterventionsACDR vs STR

Physician in responder team

Mean response time: 2 minutes

Adrenaline dose: ?


Outcomes
  • Immediate and at-hospital discharge mortality
  • Neurological outcome


Intention-to-treat analysis was explicitly done


NotesIn-hospital
San Francisco (USA)
1992 to 1993


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High risk

Allocation concealment (selection bias)High risk

Incomplete outcome data (attrition bias)
All outcomes
Low risk

Selective reporting (reporting bias)Low riskAll outcomes analysed in the review had been reported adequately

Blinding of participants and personnel (performance bias)
All outcomes
High risk

Blinding of outcome assessment (detection bias)
Mortality
Low risk

Blinding of outcome assessment (detection bias)
Neurological status
Unclear riskNot stated

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Arai 2001Not randomised

Aufderheide 2011Therapeutic interventions other than the method of chest compression were different: ACDR was combined with an inspiratory impedance threshold device (ITD). Control group received neither ACDR nor ITD

Cohen 199360% of included participants had a known pre-existing terminal disease

Gueugniaud 1998Participants were randomised to receive high or standard doses of adrenaline. Some participants included in this study received ACDR, but they were not randomised to receive ACDR or STR

Lefrançois 1998Not randomised

Panzer 1996Retrospective, not randomised

Plaisance 2004ACDR CPR was applied to all participant groups. This trial studied the effect of a respiratory impedance threshold device compared to usual ventilation

Rivers 1993Alternation of different methods of chest compression on the same participant

Wolcke 2003Therapeutic interventions other than the method of chest compression were different: ACDR was combined with an inspiratory impedance threshold device (ITD). Control group received neither ACDR nor ITD

 
Characteristics of studies awaiting assessment [ordered by study ID]
He 2003

MethodsParticipants were allocated to ACD-CPR or standard CPR randomly (no details known)

Participants92 participants with cardiac arrest occurring out-of-hospital in Sichuan, China

InterventionsACDR CPR (cardio pump) or standard CPR

Outcomes
  • Return of spontaneous circulation and admission to hospital

NotesOnly data from abstract known. No figures given. It is said that "On arriving to the hospital, the survival rate was higher in the ACD-CPR group, but of no statistical significance." The full text of this publication is awaiting translation from Chinese

 
Comparison 1. Out-of-hospital cardiac arrests

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Immediate mortality104162Risk Ratio (M-H, Random, 95% CI)0.98 [0.94, 1.03]

 2 Mortality at hospital discharge93412Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.98, 1.01]

 3 Neurological impairment in survivors5358Risk Ratio (M-H, Fixed, 95% CI)1.74 [1.06, 2.83]

    3.1 Moderate neurological impairment
4107Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.34, 2.79]

    3.2 Severe neurological impairment
4107Risk Ratio (M-H, Fixed, 95% CI)3.11 [0.98, 9.83]

    3.3 Any neurological impairment (any severity)
5144Risk Ratio (M-H, Fixed, 95% CI)1.71 [0.90, 3.25]

 4 Complications73032Risk Ratio (M-H, Fixed, 95% CI)1.09 [0.86, 1.38]

 
Comparison 2. In-hospital cardiac arrests

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Immediate mortality2826Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.88, 1.08]

 2 Mortality at hospital discharge2826Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.95, 1.05]

 3 Neurological impairment in survivors2279Risk Ratio (M-H, Fixed, 95% CI)1.19 [0.63, 2.25]

    3.1 Moderate neurologic impairment
293Risk Ratio (M-H, Fixed, 95% CI)0.5 [0.10, 2.58]

    3.2 Severe neurologic impairment
293Risk Ratio (M-H, Fixed, 95% CI)1.95 [0.59, 6.50]

    3.3 Any neurological impairment (any severity)
293Risk Ratio (M-H, Fixed, 95% CI)1.19 [0.50, 2.86]

 4 Complications1773Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.49, 1.93]

 
Comparison 3. Subgroup analysis: physician in first responding team or only paramedics

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Immediate mortality104162Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.94, 1.01]

    1.1 Physician in first responding team
51487Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.89, 1.04]

    1.2 Only paramedics in first responding team
52675Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.95, 1.02]

 2 Mortality at hospital discharge93412Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.98, 1.01]

    2.1 Physician in first responding team
4737Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.94, 1.04]

    2.2 Only paramedics in first responding team
52675Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.98, 1.01]

 3 Neurological impairment in survivors (any severity)5144Risk Ratio (M-H, Fixed, 95% CI)1.71 [0.90, 3.25]

    3.1 Physician in first responding team
244Risk Ratio (M-H, Fixed, 95% CI)1.07 [0.41, 2.80]

    3.2 Only paramedics in first responding team
3100Risk Ratio (M-H, Fixed, 95% CI)2.19 [0.93, 5.13]

 
Comparison 4. Sensitivity analysis: best-quality studies

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Immediate mortality124988Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.95, 1.01]

    1.1 Allocation concealment adequate
52460Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.96, 1.06]

    1.2 Allocation concealment unclear
161Risk Ratio (M-H, Fixed, 95% CI)1.13 [0.77, 1.65]

    1.3 Allocation concealment inadequate
62467Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.90, 0.99]

 2 Mortality at hospital discharge114238Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.98, 1.01]

    2.1 Allocation concealment adequate
52460Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.97, 1.02]

    2.2 Allocation concealment unclear
161Risk Ratio (M-H, Fixed, 95% CI)1.04 [0.87, 1.25]

    2.3 Allocation concealment inadequate
51717Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.97, 1.02]

 3 Neurological impairment in survivors (any severity)7237Risk Ratio (M-H, Fixed, 95% CI)1.50 [0.89, 2.51]

    3.1 Allocation concealment adequate
3163Risk Ratio (M-H, Fixed, 95% CI)1.45 [0.75, 2.79]

    3.2 Allocation concealment unclear
17Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.33, 2.37]

    3.3 Allocation concealment inadequate
367Risk Ratio (M-H, Fixed, 95% CI)2.09 [0.67, 6.54]

 
Comparison 5. Sensitivity analysis: largest (n > 500) studies

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Immediate mortality53815Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.93, 1.00]

 2 Mortality at hospital discharge43065Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.98, 1.02]

 3 Neurological impairment in survivors (any severity)3159Risk Ratio (M-H, Fixed, 95% CI)1.60 [0.81, 3.16]