Intended for healthcare professionals


Paramedics and technicians are equally successful at managing cardiac arrest outside hospital

BMJ 1995; 310 doi: (Published 29 April 1995) Cite this as: BMJ 1995;310:1091
  1. U M Guly, senior registrara,
  2. R G Mitchell, registrara,
  3. R Cook, senior house officera,
  4. D J Steedman, consultanta,
  5. C E Robertson, consultanta
  1. a Department of Accident and Emergency Medicine, Royal Infirmary of Edinburgh, Edinburgh EH3 9YW
  1. Correspondence to: Dr Robertson.
  • Accepted 17 February 1995


Objective: To examine the effect on survival of treatment by ambulance paramedics and ambulance technicians after cardiac arrest outside hospital.

Design: Prospective study over two years from 1 April 1992 to 31 March 1994.

Setting: Accident and emergency department of university teaching hospital.

Subjects: 502 consecutive adult patients with out of hospital cardiopulmonary arrest of cardiac origin.

Interventions: Treatment by ambulance technicians or paramedics both equipped with semi-automatic defibrillators.

Main outcome measures: Rate of return of spontaneous circulation, hospital admission, and survival to hospital discharge.

Results: Rates of return of spontaneous circulation, hospital admission, and survival to hospital discharge were not significantly different for patients treated by paramedics as opposed to ambulance technicians. Paramedics spent significantly longer at the scene of the arrest than technicians (P<0.0001).

Conclusions: The response of ambulance paramedics to patients with cardiopulmonary arrest outside hospital does not provide improved outcome when compared with ambulance technicians using basic techniques and equipped with semi-automatic defibrillators.

Key messages

  • Key messages

  • Ambulance paramedics undergo extensive further training in resuscitation techniques

  • No improvement in overall outcome was seen when paramedics attended the patients

  • The government plans to have at least one paramedic in every front line ambulance by the end of 1995

  • The outcome of patients treated by technicians v paramedics does not justify the government's plans


In 1984 the Department of Health authorised the extended (paramedic) training of ambulance staff by the NHS training authority (now the NHS training directorate). The directorate has recommended that from 1995 every front line ambulance in England and Wales should be staffed by one paramedic and one technician and should be equipped with a defibrillator.1 2

A similar approach was followed in Scotland but implemented differently. The Scottish Ambulance Service introduced semi-automatic defibrillators into all front line vehicles and provided limited training for the basic ambulance technicians. The provision of the 407 defibrillators, together with the eight hour training required for the ambulance technicians, was completed by July 1989.3 The results achieved by these actions have been remarkably successful, and local and national results have been reported.4 5

Paramedic training has also taken place in the Scottish Ambulance Service. This entails a 320 hour training course, and the successful candidates are then introduced to front line vehicles. Paramedics may attend patients with cardiac arrest either as a member of the crew of the first responding ambulance or as a secondary response at the request of an ambulance technician crew.

With these innovations it is appropriate to evaluate the relative contributions to patient outcome made by these two groups. We prospectively studied all out of hospital cardiac arrests presenting to this hospital over a two year period to examine how the presence of paramedic crews and the use of their extended skills affect outcome.

Patients and methods

We carried out a prospective study on all cardiac arrests occurring outside hospital that were received and treated in the accident and emergency department over the two years from 1 April 1992 to 31 March 1994. Details of prehospital intervals and interventions were obtained from the Scottish Ambulance Service's patient report form. This form includes the time of receipt of the 999 call, the time of dispatch of the ambulance crew (either technicians or paramedics), the times of arrival at and departure from the scene, and the time of arrival at hospital. These times are noted by the individual crews but are also confirmed by reference to the ambulance dispatch and control unit. The performance of interventions such as tracheal intubation and intravenous access were also obtained from this form. The presenting rhythm for the cardiac arrest episode was determined from the printout from the semi-automatic defibrillator (Laerdal Heartstart 2000 or 3000).

The decision to send an ambulance with technicians or one with paramedics is made by the central ambulance control on the basis of the available clinical information, availability of crew, and proximity. Assistance by paramedics can also be requested by ambulance technicians after evaluation of the patient at the scene.

Patients in whom return of spontaneous circulation occurred and who were admitted to hospital were treated in the coronary care or intensive care units. The rhythm on presentation to the accident and emergency department, any return of spontaneous circulation, survival to admission to hospital, and discharge from hospital were documented from the accident and emergency record sheets and the hospital records.

Patients sustaining cardiac arrest due to a non-cardiac cause (for example, drug overdose, trauma, hypovolaemia, or hypothermia) were excluded from the analysis. Within Edinburgh and the surrounding area some patients are treated by a hospital based medical team (Medic 1), which operates closely in conjunction with the Scottish Ambulance Service.6 7 These patients were also excluded from analysis.

Data were entered on a database (Microsoft Excel 4.0) with the internationally recognised Utstein template.8 Data analysis was performed by using χ2 test with Yates's correction, and differences between groups were assessed by the Mann-Whitney U test.


Over the two years a total of 565 patients were received in the accident and emergency department after out of hospital cardiac arrest of cardiac origin and treatment by ambulance technicians or paramedics. Sixty three patients were excluded from analysis: in 46 patients the arrest episode occurred after the arrival of the ambulance crew or during transit to hospital; in 17 cases the available documentation was insufficient to determine whether a paramedic or technician crew had been involved.

A total of 502 patients fulfilled the study criteria. These patients had sustained a cardiac arrest before arrival of the ambulance; their treatment was performed and documented by an ambulance paramedic or technician.

Two hundred and fifty nine patients (175 men, 84 women) received treatment from paramedics. The median age of these patients was 69 (range 29-94) years. Two hundred and forty three patients (181 men, 62 women) with a median age of 69 (30–92) years were treated by technicians.

There was no significant difference between the two groups in relation to the presenting rhythm, the proportion of patients whose cardiac arrests were witnessed, or those who received any bystander attempt at cardiopulmonary resuscitation (table I). Call response and transport intervals were identical for the two groups. The length of time spent at the scene of the arrest was significantly longer for patients treated by a paramedic (P<0.0001), and the length of time between the 999 call and arrival at hospital was also significantly longer (P<0.0001) (table II). Table III shows the interventions performed by the paramedics and their influence on these time intervals.


Details of out of hospital cardiac arrests treated by paramedics or ambulance technicians

View this table:

Median (range) time intervals (minutes) for each stage of response according to treatment by paramedics or ambulance technicians

View this table:

Interventions used and effect on median (range) time intervals (minutes) for each stage of response in out of hospital cardiac arrests attended by paremedics

View this table:

There was no significant difference in any outcome variable between the two groups. This included the proportion of patients in whom spontaneous circulation was achieved, those admitted to hospital, and those discharged alive from hospital (table IV). The proportion of patients discharged alive from hospital was greater in the group treated only by ambulance technicians (9.5% v 5.4%), but this did not reach significance (P=0.087).


Outcome in patients suffering out of hospital cardiac arrest according to treatment by paramedic or ambulance technician

View this table:

A total of 234 patients had ventricular fibrillation as their primary rhythm. For this group, 32 were subsequently discharged alive from hospital. One hundred and eleven patients with ventricular fibrillation were treated by technicians, with 18 (16%) survivors to hospital discharge. One hundred and twenty three were treated by paramedics, with 14 (11%) surviving to discharge. For episodes of ventricular fibrillation which had been witnessed by bystanders, the figures were 17/85 (20%) for ambulance technicians and 11/103 (11%) for paramedics.


The pioneering work of Pantridge and Geddes showed the efficacy of management of out of hospital cardiac arrest by doctors using a mobile intensive care unit.9 Subsequently, attention has focused on the optimal method of delivering this type of care in the community.

In North America paramedics routinely use advanced life support techniques, which include defibrillation, tracheal intubation, intravenous techniques, and drug administration.10 The training includes classroom and field experience and may last from 1000 to 3000 hours.10

In the United Kingdom, Department of Health recommendations aim to provide at least one trained paramedic in the crew of every emergency ambulance by the end of 1995.2 This policy has not, however, been based on objective evidence as to the efficacy of such interventions as compared with simpler, less expensive, and less time consuming approaches. The ability of basic life saving skills and defibrillation alone to achieve good results has been considered.11 12 13 Even in North America the need for full paramedic skills as opposed to ambulance crew using defibrillators alone has been challenged.14 15

The nature of the primary electrocardiographic rhythm for such patients is the single most important determinant of a successful outcome.10 16 Malignant ventricular tachyarrhythmias, of which ventricular fibrillation is by far the most common, are found in most patients experiencing sudden cardiac arrest.17 18 19 With the passage of time, coarse ventricular fibrillation progressively decreases in amplitude and is associated with a reduction in the probability of successful defibrillation.20 The other electrocardiographic rhythms associated with cardiac arrest—asystole and electromechanical dissociation—are accompanied by dismal outcomes irrespective of treatment given.3 5 7 10 12 18 Furthermore, defibrillation as a treatment will not influence survival in this group.21 22 For these reasons it is apparent that nearly all patients in whom a successful outcome could be anticipated are those with a primary rhythm of ventricular fibrillation and in whom defibrillating shocks can be administered with a minimal delay.


For the duration of the study paramedics were permitted to perform intravenous cannulation and tracheal intubation but did not administer specific cardioactive drugs. Although it might be argued that this distorts our conclusions, it is increasingly recognised that the administration of drugs has at best a marginal role in improving survival after cardiac arrest.

In studies on animals and humans, vasopressors such as adrenaline have been shown to have beneficial effects in relation to the haemodynamic profiles achieved during cardiopulmonary resuscitation.23 Human studies which have assessed improvement in survival to hospital discharge, however, have yielded disappointing results even with high dose regimens.24 25

The role of buffers in treating the metabolic acidosis associated with cardiac arrest has also been questioned.26 The first large double blind, randomised placebo study of the use of buffers in prehospital cardiac arrest was unable to show any benefit either in restoration of spontaneous circulation, admission to hospital, or hospital discharge rates.27

Finally, in relation to drug treatment, the ability of antiarrhythmic agents such as lignocaine or bretylium to improve outcomes remains at best not proved.28 29 30 31


The overall results achieved by ambulance service staff in our study compare favourably with other systems in the United Kingdom, Europe, and North America. The overall rate of survival to hospital discharge of 14.9% for patients with witnessed cardiac arrest and a primary rhythm of ventricular fibrillation is higher than the national figure3 4 and is similar to values from other well established programmes.10

We have not shown any improvement in outcome with the presence of a paramedic after cardiac arrest occurring out of hospital. There were small nonsignificant differences between the two groups in the proportion of patients who presented in ventricular fibrillation, whose arrests were witnessed, and who received bystander cardiopulmonary resuscitation. For all of these variables, however, the group treated by paramedics contained a higher proportion of patients who would be expected to have had a better prognosis.

The length of time at the scene of the arrest was significantly longer for patients treated by a paramedic. This is probably because of the time needed for tracheal intubation and intravenous cannulation, together with any additional time spent waiting for the paramedics to arrive when they were called as a secondary response. This additional time does not result in an improved outcome.

The challenge remains to see whether other or complementary approaches may lead to even better results. There is substantial evidence that basic life support (cardiopulmonary resuscitation) performed by bystanders before the arrival of the ambulance can positively influence the outcome.32 33 The primary reason for this improvement relates to the fact that, while cardiopulmonary resuscitation cannot itself reverse the processes occurring during cardiac arrest, it increases the proportion of patients who are still in ventricular fibrillation when the first responding ambulance team arrives.32 34 35

Hospital based teams led by physicians who are able to provide immediate support to ambulance crews is one solution which has been shown to be effective in Edinburgh6 7 and Belfast.36 More radical solutions, such as providing automatic defibrillators at public places for use by lay persons or those with minimal training, still have to obtain widespread acceptance and need to be shown to be safe and efficacious.

The results of this study do not downgrade or diminish the role of paramedics. These people are highly motivated, enthusiastic, and professional. Rather, we have shown the overwhelming need to provide rapid defibrillation for victims of sudden cardiac arrest. At present this is best achieved by ambulance technicians using basic life support skills and semi-automatic defibrillators.

We thank the technicians and paramedics of the Scottish Ambulance Service for their diligence in treating the patients under their care and completing the patient report forms.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
View Abstract