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

Introduction

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.

Results

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.

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).

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.

Discussion

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.