Skip navigation

Defibrillator use in out-of-hospital cardiac arrest


Cardiac arrest occurring outside hospital will be treated using emergency medical services that might have a range of resources available, including advanced life support systems with cardiopulmonary resuscitation, defibrillation, intubation, artificial ventilation and intravenous administration of medication. A new meta-analysis [1] has explored the literature to see whether any particular technique or combination is better.


The review searched MEDLINE up to mid 1977 for articles in English, and examined bibliographies of papers. It included only published studies that looked at a several different systems (Table 1), and which had data on survival to hospital discharge. This information was extracted, plus response time interval, the proportion of patients to whom bystanders had applied cardiopulmonary resuscitation and the type of system used to treat patients.

Table 1: Definitions of different systems of life support after cardiac arrest examined in the study

System of support Definition
Basic life support (BLS) Administration of oxygen and cardio-pulmonary resuscitation
BLS with defibrillation (BLS-D) Additional use of automatic or manual defibrillators
Advanced life support (ALS) Providers trained to perform endotracheal intubation and administer intravenous medications
BLS + ALS Where basic life support is followed by advanced life support
BLS-D + ALS Where basic life support with defibrillator is followed by advanced life support


Thirty-seven articles had information on over 33,000 people suffering cardiac arrest out of hospital. None of the studies was randomised. The analysis investigated the effect of independent variables - the proportion of bystander cardiopulmonary resuscitation, defibrillator response interval and type of system used - on the survival to hospital discharge.

Table 2 shows both the crude survival figures and the odds ratios after making allowance for other variables. Greater survival to hospital discharge was associated with the type of system used, and also with increases in bystander cardiopulmonary resuscitation and reduced defibrillator response time interval.

Table 2: Outcomes with different systems of life support after cardiac arrest examined in the study, using both crude survival percentage and odds ratios after allowance for differing rates of bystander cardiopulmonary resuscitation and other confounders

System of support Survivors / Total Percent (95% CI) Odds ratio (95% CI)
BLS-D 815/12433 6.6 (6.1 to 7.0) 1
ALS 560/10072 5.6 (5.1 to 6.0) 1.71 (1.09 to 2.70)
BLS + ALS 842/7502 11.2 (10.5 to 11.9) 1.47 (0.89 to 2.42)
BLS-D + ALS 221/2359 9.4 (8.2 to 10.5) 2.31 (1.47 to 3.62)

For bystander cardiopulmonary resuscitation, every 5% increase was associated in an absolute increase in survival of between 0.3% and 1%. A 1 minute decrease in the defibrillator response time was associated with an absolute increase in survival of 0.7% to 2.1%.


The review perhaps emphasises what we might have guessed. Getting in early with cardiopulmonary resuscitation, defibrillation and pre-hospital advanced life support all contribute to improved chances of survival for someone suffering from cardiac arrest outside hospital. There may be no randomised trials, but this is the best information we have. It is a carefully and cleverly done review that would help anyone responsible for designing or delivering emergency services (though readers in Wales may be distressed to find that it claims south Glamorgan for England!). There is nothing here about quality of survival. What proportion of the survivors had hypoxic brain damage?


  1. G Nichol et al. A cumulative meta-analysis of the effectiveness of defibrillator-capable emergency medical services for victims of out-of-hospital cardiac arrest. Annals of Emergency Medicine 1999 34: 517-525.
previous or next story in this issue