Results from the first 12 months of a fire first-responder program in Australia

Resuscitation 49 (2001) 143– 150 www.elsevier.com/locate/resuscitation

Results from the first 12 months of a fire first-responder program in Australia K.L. Smith *, A. Peeters, J.J. McNeil Monash Medical School, Monash Uni6ersity, Alfred Hospital, Commercial Rd., Prahran, Vic. 3181, Australia Received 20 July 2000; received in revised form 3 October 2000; accepted 10 October 2000

Abstract Study objecti6e: We aimed to reduce response times and time to defibrillation for out-of-hospital cardiac arrest patients through fire first-responders equipped with automatic external defibrillators (AEDs). The fire first-responders were added as an extra tier to the existing two-tired ambulance response. Methods: This prospective controlled trial set in Melbourne, Australia, consisted of a control area (277 km2, population density 2343/km2-ambulance only dispatch) and a pilot area (171 km2, population density 2290/km2-ambulance and fire first-responder dispatch). The main outcome measures were time to emergency medical service (EMS) arrival at scene for all cardiac arrest patients and time to defibrillation for cardiac arrest patients presenting in ventricular fibrillation (VF). The study participants were patients who suffered a cardiac arrest of presumed cardiac aetiology for which a priority 0 emergency response was activated. A total of 268 patients were located in the control area and 161 in the pilot (intervention) area. Results: The mean response time to arrival at scene was reduced by 1.60 (95% CI 1.21, 1.99) min, P B 0.001. A large reduction in prolonged responses ( ] 10 min) to cardiac arrests was also observed in the pilot area (2%) compared with the control area (18%), = 23.19, PB 0.001. Mean time to defibrillation was reduced by 1.43 (95% CI 0.11, 2.98) min, P =0.068. Conclusion: The results from this study suggest that fire officers can be successfully trained in the use of AEDs and can integrate well into a medical response role. The combined response of ambulance and fire personnel significantly reduced the response interval and reduced time to defibrillation. This suggests that in appropriate situations other agencies could be considered for involvement in co-ordinated first-responder programs. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Ambulance response; Automatic external defibrillators; Cardiac arrest

Resumo Objecti6o: Os autores tiveram por intenc¸a˜o a reduc¸a˜o do tempo de resposta e do tempo para desfibrilac¸a˜o de doentes em paragem cardı´aca fora do hospital atrave´s da implementac¸a˜o de um programa de resposta ra´pida por bombeiros equipados com desfibriladores automa´ticos externos (DAEs). O programa de resposta ra´pida por bombeiros entrou de modo complementar ao modelo existente, de um sistema de ambulaˆncias organizadas ate ali em dupla resposta. Me´todo: Este estudo prospectivo controlado montado em Melbourne na Austra´lia consistiu numa a´rea de controle (277 km2, densidade populacional de 2343/km2-ambulaˆncia isolada) e numa a´rea piloto (171 km2, densidade populacional de 2290/km2- ambulaˆncia mais resposta ra´pida). Os principais resultados a medir foram o tempo para chegada ao local para todos os doentes em paragem cardı´aca e de tempo para desfibrilac¸a˜o para os doentes em paragem cardı´aca que se apresentassem em fibrilac¸a˜o ventricular. Os participantes foram doentes que sofreram paragem de etiologia presumida como cardı´aca para qual foi activada a prioridade 0 no sistema de resposta de emergeˆncia. Um total de 268 doentes localizaram-se na a´rea de controle e 161 na a´rea piloto (de intervenc¸a˜o). Resultados: O tempo me´dio de resposta para chegada ao local foi reduzido em 1.60 min (95% IC 1.21, 1.99) P B 0.001. Uma grande reduc¸a˜o nas respostas prolongadas ( \10 min) a`s paragens tambe´m foi observado na a´rea piloto (2%) comparativamente a` a´rea de controle (18%) x= 23.19 P B0.001. O tempo me´dio para desfibrilac¸a˜o foi reduzido em 1,43 min (95% IC 0.11, 2.98) P= 0.068. Conclusa˜o: Os resultados deste estudo sugerem que os bombeiros podem ser treinados com sucesso no uso de DAE e



On behalf of the EMR steering committee. * Corresponding author. Tel: +61-3-99030951; fax: + 61-3-99030556. E-mail address: [email protected] (K.L. Smith).

0300-9572/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 0 - 9 5 7 2 ( 0 0 ) 0 0 3 5 5 - 5

144

K.L. Smith et al. / Resuscitation 49 (2001) 143–150

podem integrar-se bem dentro de um papel da resposta me´ dica. A resposta combinada de ambulaˆ ncia mais bombeiros reduziu de modo significativo o tempo de resposta e o tempo para desfibrilac¸ a˜ o. Isto sugere que em situac¸ o˜ es apropriadas outras instituic¸ o˜ es podem ser consideradas para envolvimento num programa coordenado de resposta ra´ pida. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Pala6ras cha6e: Resposta por ambulaˆ ncia; Desfibriladores automa´ ticos externos; Paragem cardı´aca

1. Introduction Sudden onset of ventricular fibrillation (VF) is responsible for about 70% of cardiovascular deaths [1–3]. During the first 5–10 min after the onset the disorganized electrical activity is potentially reversible with electrical defibrillation [4]. The likelihood of survival after cardiac arrest is, therefore, determined by the rapid availability of trained individuals equipped with defibrillators [5]. Recent improvements in technology have allowed the development of new lightweight defibrillators with software capable of analyzing a patients cardiac rhythm and determining whether electrical cardioversion is appropriate[6]. Unlike earlier defibrillators, this equipment can be successfully applied by individuals with relatively little formal training. These features potentially allow for their use by a wide range of individuals likely to find themselves on the scene of a medical emergency [7]. In North America and Europe, various police and fire department personnel have been equipped with automatic external defibrillators (AEDs) and trained to be effective ‘first-responders’ to cardiac arrests[8– 14]. Several studies have suggested that this strategy is effective in improving survival from cardiac arrest [8,10,12,15]. However, a variety of methodological weaknesses including retrospective data collection, the use of historical controls and a lack of adherence to protocol have made the results difficult to interpret. The present study examined the effect of adding fire first responders (trained and equipped with AEDs) to an existing two tiered emergency medical service (EMS) system on the outcome of cardiac arrest in Melbourne. The study was a planned prospective investigation, which compared outcomes in defined intervention and control areas within the city. The primary outcome was time to defibrillation for patients presenting in VF.

2. Materials and methods

2.1. Study setting Melbourne has a population of approximately 3 million, which is covered for medical emergencies by the Metropolitan Ambulance Service (MAS). The ambulance service operates a 2-tiered system of basic life

support (BLS) ambulances and advanced life support (ALS) Mobile Intensive Care Ambulances (MICA). At the time of this study there were 108 ambulances and 31 MICA. All ambulance units are equipped with semi-automatic defibrillators. The Metropolitan Fire and Emergency Services Board (MFESB) primary role is responding to fire calls and has not previously responded to medical emergencies. Both the ambulance and fire are dispatched from a central communications center run by Intergraph Bureau of Emergency Services Telecommunications Victoria (IBV) which has a central emergency phone number of 000 for the entire State. When an individual dials 000, the call is directed to the IBV Central Communications Center. The highest medical emergency are priority 0, which are those where the subject is suspected to be unconscious and/or non-breathing. The IBV Central Communications Center receives around 4200 priority 0 calls a year. On identifying a priority 0 event, the dispatcher activates a full-tiered ambulance response. The closest available ambulance and MICA units are dispatched in prompt succession.

2.2. Inter6ention Fire officers were trained in BLS and the use of automatic defibrillators (Laerdal HeartStart FR) in an 8-day training course. A pilot area was defined which included approximately 20% of Melbourne’s population (648 570) and covered 171 km2. Seven fire stations were located within this area. From 14 July 1998 to 22 June 1999, a dual dispatch of fire-first responders and normal ambulance response occurred to priority 0 events in the pilot area. A control area, which was peripheral to the pilot area, was also defined (population 648 570 over 277 km2). The control area was covered by a normal ambulance response only.

2.3. Data Collection The Central Communications Center created a daily report of all priority 0 events. This report included data on crew type, event location and response times. Patient care records for each event were completed by both fire and ambulance personnel for events in the pilot area and ambulance personnel only for events in the control area. Data was abstracted from patient care

K.L. Smith et al. / Resuscitation 49 (2001) 143–150

records on patient details, event details and response times. Ambulance personnel followed up hospital discharge details of cardiac arrest patients through the relevant hospitals. An interruption to simultaneous dispatch of fire and ambulance, resulting from computer changes at IBV Central Communications resulted in data collection for the study running just under a 12-month period.

2.4. Quality control As a quality control measure, medical students travelled with random ambulance crews and noted their compliance to protocol in recording of time and calling in of time to IBV (mean difference to Eastern Standard Time was B1 min for all the cases observed).

2.5. Data processing Data for the study was recorded into an Access database. All statistical analyses were performed using the statistical package STATA [16]. The pilot and control groups were compared using univariate statistical tests. Continuous data was analyzed with the Student’s t-test and discrete variables were analyzed using the  2-test and Proportion test. A P-value of B 0.05 was considered statistically significant.

3. Results Results from data collected on cardiac arrest patients for 2 months prior to the start of the study, demonstrate the comparability of the control and pilot areas. Both areas were similar in population density, proportion of the population aged over 64, incidence of cardiac arrest, survival from cardiac arrest and mean

145

ambulance response time (P \ 0.05 for all comparisons). Data on all patients presenting in VF to ambulance personnel in 1997 was also compared and no statistically significant difference in incidence of VF or response time of ambulance in each area was observed.

3.1. E6ents During the study period there were 907 priority 0 events in the control area and 454 in the pilot area. The main cause of priority 0 events in both areas was cardiac arrests of presumed cardiac cause. No difference was observed between the pilot and control areas with respect to the age or sex of cardiac arrest patients, or the proportion who were witnessed or received bystander CPR (P\ 0.05). Both areas had a low percentage of patients presenting in VF (15% control and 13% pilot), which may be due to the low numbers of arrests that were witnessed (see Table 1). The main site of arrests was in the patients home (79%).

3.2. Response times The time taken for ambulance vehicles to arrive at the scene of cardiac arrest patients did not alter significantly during the pilot study. The mean response time remained similar to that in the historical areas and the control area (P\ 0.05). However, the combined EMS response time (either ambulance or fire) was 1.60 min shorter in the pilot area compared with the ambulance only response in the control area (5.85 vs. 7.45 min, PB 0.001). For witnessed arrests during the study period the mean time to arrival at scene was 1.18 min shorter for survivors than for non-survivors (6.28 min vs. 7.46, P=0.072). A large reduction in the proportion of prolonged responses (]10 min) to cardiac arrests was also observed in the pilot area (2%) compared with

Table 1 Cardiac arrests of presumed cardiac cause during the study period Control area All arrests

Number of arrestsa Mean age (year) (S.D.) Initial rhythm VF/VT (%) Resuscitation initited byEMS (%) Bystander CPR Survived to hospital discharge (%)b Survived to hospital discharge initial rhythm VF (%)b

268 (41/100 000 pop) 69 ( 9 16) 39 (15) 92 (34) 67 (25) 11 (4) 7/39 (18–20)

Pilot area Witnessed arrests 78 69 31 61 46 11 7

( 915) (40) (78) (59) (14) (23)

All arrests

161 (41/100 000 pop) 71 ( 914) 21 (13) 67 (42) 33 (20) 6 (4) 6/21 (29)

Witnessed arrests

35 69 ( 9 13) 14 (40) 30 (86) 14 (40) 5 (14) 5 (36)

a Three arrests were excluded from the control area due to resuscitation prior to the arrival of the EMS. One patient in the control area and one in the pilot area were excluded as their arrests were witnessed by the EMS. b One unwitnessed VF patient in the control area, who was transporeted to hospital, could not be traced in the hospital system and therefore survival data is missing for this patient. The range provides a percentage survival ‘if the patient died’ to ‘if the patient survived’.

146

K.L. Smith et al. / Resuscitation 49 (2001) 143–150

Table 2 Mean response and defibrillation intervals for the control and pilot areas during the study period Control area min (S.D.) Ambulance arrival at scene for cardiac 7.45 ( 9 2.41) arrests a Combined EMS arrival at scene for cardiac 7.45 ( 9 2.41) arrests a,b Ambulance time to defibrillation c 10.05 ( 93.21) Combined EMS time to defibrillation b,d 10.05 ( 9 3.21)

Pilot area min (S.D.)

Difference min

Power to detect observed difference

P

7.61 ( 92.94)

0.16

0.560

0.08

5.85 ( 91.66)

−1.60

B0.001

1.00

8.38 ( 92.80) 8.62 ( 92.56)

−1.67 −1.43

0.0643 0.0675

0.48 0.46

a

Calculated from 268 events in the control area and 161 in the pilot area. First EMS vehicle on scene (ambulance or fire). c Calculated from 37 events involving defibrillation in the control area and 16 in the pilot area (initial rhythm VF). d Calculated from 37 events involving defibrillation in the control area and 21 in the pilot area (initial rhythm VF). b

the control area (18%),  = 23.19, P B 0.001 (see Table 2, Fig. 1). The mean time taken to institute defibrillation in the pilot area was reduced by 1.43 min compared with the ambulance time to defibrillation in the control area (8.62 vs. 10.05 min). This reduction in defibrillation interval approached statistical significance (P = 0.068). A reduction was also observed in ambulance only time to defibrillation in the pilot area compared with the control area (8.38 vs. 10.05, P =0.064) (see Table 2). During the study period only 60 patients presented in VF (39 in the control and 21 in the pilot areas).

3.3. Patient sur6i6al During the study period survival from all rhythms was 4% for both areas. Survival from witnessed VF was 23 (7/31) and 36% (5/14) in the control and pilot areas respectively (P = 0.363). One out of two hundred and fifteen patients presenting in asystole in the control area also survived to hospital discharge (see Table 1).

4. Discussion Defibrillation is rarely successful if the duration of VF is greater than 10 min. A previous study in Melbourne showed that the mean ambulance response time to cardiac arrest patients was 9.49 3.6 min [17]. This is the response interval from when an event is recorded at the central communications center to the arrival at scene of an ambulance. With the addition of the delay of the public in dialing 000 and the delay from arrival at scene to defibrillation, early defibrillation (B10 min from collapse) would appear to be impossible for over half of cases. Two previous studies have examined the addition of first-responders as an extra tier to their EMS [10,12]. One utilized a retrospective design [10] and the other utilized historical controls [12], which place limitations on the internal validity of the results. This is the first time that the addition of first responders equipped with AEDs added as an extra tier, has been examined in a controlled prospective study (see Table 3). Our pilot study aimed to decrease EMS response times to cardiac arrest patients and time to defibrillation through the implementation of a fire-first respon-

3.4. Cost The cost of implementing a fire first-responder program across all of metropolitan Melbourne covered by the fire brigade (approximately 2 million people) was determined by monitoring all capital and recurrent costs involved in establishing the program. A capital once off cost of $A1.627 million included the equipping of 53 vehicles and 46 fire stations with defibrillators and oxygen equipment (the batteries in the AEDs have a shelf life of 5 years) and the cost of training 1400 fire fighters at $A658.50 per person. In addition to the capital cost, we estimated that appliance maintenance and consumables will result in a recurrent cost of $A70.97 per attendance and refresher training will result in a cost of $A57 600 per year.

Fig. 1. Response time to arrival at scene for cardiac arrests.

K.L. Smith et al. / Resuscitation 49 (2001) 143–150

der system. We concentrated on response time to enable analysis of the effect of our intervention after a 12month period. We also focused on a combined system response rather than comparing the responses of the two EMS organizations, as an ambulance and MICA were always dispatched when a fire first-responder was dispatched. The combined system response formed the basis of decisions regarding the future the fire brigade acting as first responders. The results of the fire first-responder pilot show a decrease in response time to cardiac arrests of 1.60 min on average (PB 0.001) and time to defibrillation decreased by 1.43 min (P =0.0675). Prolonged responses ( ] 10 min) to cardiac arrests decreased from 18 to 2%, P B0.001. The effect of a reduction in time to arrival at scene of 1.60 min and a reduction in time to defibrillation of 1.43 min can be modeled to assess the likely influence on patient survival over a much larger duration. A logistic regression model proposed by Vanezuela suggests that our reduction in defibrillation interval of 1.43 min could potentially lead to an increase in survival of 64% [4]. In keeping with this prediction it was observed that survival from witnessed VF in the pilot area was 57% greater than that in the control area. However, this improvement did not reach statistical significance. Two recent studies examining the effect of the addition of first responders to time to defibrillation, have reported defibrillation intervals similar to our combined defibrillation interval of 8.629 56) min. Mossesso et al. equipped police officers with AEDs and examined the effect on defibrillation interval in a prospective cohort with historical controls [12]. Their defibrillation interval decreased from 11.80 (94.70) min to 8.70 (93.70) min (PB 0.001). A similar observation was reported by Shuster et al., who utilized a retrospective cohort to analyze the impact of equipping fire first-responders with AEDs, on defibrillation interval. The addition of AEDs decreased their defibrillation interval from 11.96 ( 9 4.79) min to 8.50 (94.35) min (P B 0.001) [10]. The reduction in defibrillation interval observed in our study appears to be related to a reduction in ambulance defibrillation interval in the pilot area. During the study period ambulance time to defibrillation decreased from 10.05 (93.21) min in the control area to 8.38 (9 2.80) min in the pilot area. This is probably due to the fact that patient care by first-responders was handed over to ambulance personnel as soon as they arrived on scene. The fire first responders appear to have contributed to a decrease in the ambulance defibrillation interval, by placing the patient in the appropriate position, initiating CPR and placing the defibrillator pads on the patient’s chest, prior to the arrival of the ambulance. It is for reasons such as this that we have concentrated on the combined EMS response.

147

A total of 21 patients presented in VF in the pilot area. Fire first-responders arrived first for ten patients, both services arrived simultaneously for five patients and ambulance arrived first for six. While first-responders arrived first for ten patients presenting in VF, they only defibrillated five (50%) of these patients. This suggests that there was some uncertainty and hesitation within the first-responders to initiate defibrillation. Also, anecdotal evidence has suggested that the first responders were often taking over 2 min from arrival at patient to place the defibrillator paddles on the patient’s chests. This suggests that the defibrillation interval could be further reduced through improving the action of fire officers at the scene. Both of the officer’s apparent lack of confidence and delays in attaching the AEDs are being addressed in current ongoing training programs. Controlling accurate reporting of response times is very difficult to achieve and is a limitation of this study. Ambulance and fire personnel were not blinded during this study. We abstracted response times from patient care records and Intergraph Central Communications records and without the use of mobile data terminals, time to arrival at scene is open to some subjectivity. It is possible that some of the reduction in response times observed in the pilot area were the result of increased enthusiasm of ambulance personnel. However, the average ambulance only response time in the pilot area did not differ significantly from the average response time in the control area. Also, as a quality control measure, medical students traveled with random ambulance crews and noted their compliance to protocol in recording of time and calling in of time to Intergraph (mean difference to Eastern Standard Time was B 1 min for all cases observed). We are still not achieving defibrillation within the narrow time frame (B 5 min) observed in some EMS systems within North America. A recent meta-analysis, which included 39 EMS systems, obtained an overall mean defibrillation response time interval of 6.1 min, with out-lying mean defibrillation response time intervals observed in 1-tier BLS-D systems [18]. This would suggest that Melbourne is still lagging behind best practice for 2-tiered systems. In the meta-analysis analysis by Nichols et al. survival was constant if defibrillation was achieved within 6 min of receipt of call, decreased as the interval increased from 6 to 11 min and leveled off after 11 min (PB 0.01). Successful resuscitation is dependent on the system providing an adequate response at each link in the ‘chain of survival’ from cardiac arrest. We observed a low percentage of patients presenting in VF which may be attributed to extended delays from collapse to activation of the EMS. Lack of bystander CPR for witnessed arrests was also noted. The Victorian Department of Human Services is currently addressing

148

Table 3 Previous studies reporting results from use of AEDs in the pre-hospital settinga Location

Design

Results

Jaggarao et al., 1982 [19]

Brighton, UK

An observational study evaluating the use of AEDs by highly trained, experienced ambulance staff

Of 11 consecutive patients treated with the AED, 5 presented in VF, of which 4/5 (80%) were discharged from hospital

Stults et al., 1986 [20]

Iowa, USA

Controlled trial comparing the effectiveness of EMT use of AEDs compared with EMT use of standard defibrillators for patients in cardiac arrestd

AEDs were able to deliver shocks more quickly than with the manual defibrillators (1.56 vs. 2.77 min; PB0.001). Of patients presenting in VF 6/35 (17%) in the AED group were discharged and 7/53 (13%) in the manual group

Cummins et al., 1987 [21]

King County, Washington, USA

A randomized control trial comparing the effectiveness of EMT use of AEDs compared with EMT use of standard defibrillators for patients in cardiac arrestc

28% (13/46) of VF patients were discharged from the AED group compared with 23% (15/64) from the standard group (P\0.05). Mean time from power on to first shock in the AED group was 1 min versus 2.0 min in standard group, P= 0.05

Gray et al., 1987 [22]

Stockport, UK

Observational study of the effect of equipping ambulances with AEDs on survival from cardiac arrest. No control groupd

Ambulance personnel equipped with AEDs attended 113 patients in cardiac arrest of which 165 were in VF and none survived. Mean ambulance response time was 4.5 min

Weaver et al., 1988 Seattle, USA [23]

Prospective controlled trial equipping fire first-responders. 22 fire companies equipped with AEDs, 18 fire companies provided BLS only. Examined effect on survival from VFc,f

19% survival (44/228) prior to the addition of AEDs compared with 30% survival (74/276) after the addition, PB0.05. Mean response time for EMTs equipped with AEDs was 3.6 9 1.5 min compared with a mean paramedic response of 5.1 92.3 min, P= 0.001

Vukov et al., 1988 [24]

Prospective controlled trial, 14 rural communities were randomized to participate as treatment (AED) or control (BLS) for 1 year then crossed over for 1 year. Examined survival from VFd

4% (1/27) survival with AEDs compared with 17% (6/36) survival without AEDs, P\0.05. None of the BLS only group was defibrillated in less than 10 min compared with 46% of the AED group

Walters et al., 1990 London, UK [25]

Controlled prospective trial comparing outcome of patients treated by AED plus BLS or BLS by ambulance staffd

Survival at 6 months from all rhythms was 2% (6/186) in the control group compared with 3% (4/212) in AED group, P\0.05

Sedgwick et al., 1993 [26]

Scotland, UK

Observational study of the effect of equipping EMTs with AEDs on survival from cardiac arrest. No control groupd

Survival from witnessed VF was 11% (71/643). Median response time was 7 min

Kellerman et al., 1993 [9]

Memphis, Tennessee, USA

Prospective controlled trial equipping fire first-responders with AEDs and examining affect on survival. AEDs carried for 75-day rotationsc

12.9% (12/93) survival from VF before the addition of AEDs compared with 17.7% (22/124) after the addition of AEDs, P\0.05. In intervention group the mean EMT response time was 3.5 90.7 min compared with a mean paramedic response of 6.0 92.8 min

Rural, SE Minnesota, USA

K.L. Smith et al. / Resuscitation 49 (2001) 143–150

Author

149

Table 3 (continued) Rural, SE Minnesota, Location

then crossed over for 1 year. Examined survival Design

with 23% (15/64) from the standard group (P\0.05). Mean time from power on to first Results

Shuster et al., 1993 [10]

Hamilton-Wentworth, Ontario, Canada

Retrospective cohort analyzing the impact of equipping fire first-responders with AEDs on survival from VFb

12% (3/25) survival prior to equipping first-responders with AEDs compared with 19% (7/37) after equipping first-responders with AEDs, P\0.05. Introduction of fire-first responder decreased the mean defibrillation interval from 11.96 min to 8.50 min, PB0.001

Mols et al., 1994 [27]

Brussels, Belgium

Observational study of the effect of equipping EMTs with AEDs on survival from cardiac arrest. No control groupc

Survival form VF was 13% (14/105). Mean response time for EMT-D to patients in VF/VT was 5.28 90.21 min

Calle et al., 1997 [11]

Gent, Aalter and Brugge, Belgium.

Observational study of defibrillation by fire officers equipped with AEDs over 3–5 years. No control groupsc

Fire EMTs attended 219 patents in VF/VT of which 46 (21%) survived

Mosesso et al., 1998 [12]

Pennsylvania, USA

Prospective cohort study examining the affect of equipping police officers with AEDs on survival from VF. Historical data used as controlb,e

6.3% (5/80) survival before the addition of AEDS compared with 14.2% (18/127) after the addition of AEDs, P\0.05. Mean time to defibrillation decreased from 11.8 9 4.7 min to 8.7 9 3.7 min, PB0.0001

Sweeney et al., 1998 [13]

Charlotte, North Carolina, USA

Prospective controlled trial examining the effect of equipping fire first-responders with AEDs on survival from VF. AEDs carried for 1–2 months rotationsc,f

7.6% (8/105) survival before the addition of AEDs compared with 6.5% (5/77) after the addition of AEDs, P\0.05. In the AED group mean response time for fire fighters was 4.3 9 1.8 min compared with 11.1 9 3.6 min for paramedics

White et al., 1998 [14]

Rochester, USA

Retrospective observational outcome study of consecutive adult cardiac arrest patients. Examination of the effect police first-responders equipped with AEDs on survival from VFc

37% (31/84) of patients were first shocked by police of which 18 (58%) survived. 23 (43%) patients first shocked by paramedics survived. Mean call to shock time was 5.6 min when police defibrillated compared with 6.3 min when paramedics defibrillated (P = 0.038)

a

EMT, emergency medical technician, BLS, basic life support, AED, automatic external defibrillator, VF, ventricular fibrillation. EMS systems were two tiered (BLS plus ALS units). First responders with AED were added as an extra tier. c EMS systems were two tiered (BLS plus ALS), AEDs added to a proportion of first-tier. d Single tier EMS (BLS only). e Selected police departments that arrived at least 2 min. prior to EMS in \50% calls. f Selected fire companies with the largest response time differences between fire-first responders and paramedics. b

K.L. Smith et al. / Resuscitation 49 (2001) 143–150

Cummins et al., Author

150

K.L. Smith et al. / Resuscitation 49 (2001) 143–150

this with the ‘Key to Survival’, a program which provides free CPR lessons to Victorians. The impact of this program on levels of bystander CPR participation is yet to be ascertained. First-responders have been equipped with defibrillators in many countries as a means of decreasing time to defibrillation. It is possible that a wider dissemination of AEDs will aid in decreasing response times. The question is how defibrillators should be disseminated and who should be trained and equipped to deploy them [7]. The results from this study suggest that fire officers, who have not previously responded to medical emergencies, can be successfully trained in the use of AEDs and can integrate well into a medical response role. Other agencies such as police and security guards may also warrant consideration for involvement in co-ordinated first-responder programs.

[9]

[10] [11]

[12]

[13]

[14]

[15]

5. Conclusion The implementation of fire first-responders equipped with defibrillators has been an important addition to an EMS system where early defibrillation was not regularly provided. The combined response of ambulance and fire personnel significantly reduced the response interval and reduced time to defibrillation.

[16] [17] [18]

[19]

[20]

References [1] Reichenbach DD, Moss NS, Meyer E. Pathology of the heart in sudden cardiac death. Am J Cardiol 1977;39(6):865. [2] Eisenberg MS, Horwood BT, Cummins RO, et al. Cardiac arrest and resuscitation: a tale of 29 cities. Ann Emerg Med 1999;19(2):179 – 86. [3] Meyer ADM, Cameron PA, Smith KL, et al. Out-of-hospital cardiac arrest. Med J Aust 2000;172(2):73 –6. [4] Valenzuela TD, Roe DJ, Cretin S, et al. Estimating effectiveness of cardiac arrest interventions. A logistic regression survival model. Circulation 1997;96(10):3308 –13. [5] Cummins RO, Ornato JP, Thies WH, et al. Improving survival from sudden cardiac arrest: the ‘chain of survival’ concept. A statement for health professionals from the Advanced Cardiac Life Support Subcommittee and the Emergency Cardiac Care Committee, American Heart Association. Circulation 1991;83(5):1832 – 47. [6] Robertson C, Nichol NM. Recent advances in defibrillation therapy. Curr Opin Crit Care 1997;3:214 – 7. [7] Nichol G, Hallstrom AP, Kerber R, et al. American Heart Association report on the second public access defibrillation conference, April 17 –19, 1997. Circulation 1998;97(13):1309 – 14. [8] Weaver WD, Copass MK, Hill DL, et al. Cardiac arrest treated

.

[21]

[22]

[23]

[24] [25]

[26]

[27]

with a new automatic external defibrillator by out-of-hospital first responders. Am J Cardiol 1986;57(13):1017 – 21. Kellerman AL, Hackman BB, Somes G, et al. Impact of first-responder defibrillation in an urban emergency medical services system [see comments]. J Am Med Assoc 1993;270(14):1708 –13. Shuster M, Keller JL. Effect of fire department first-responder automated defibrillation. Ann Emerg Med 1993;22(4):721 –7. Calle PA, Verbeke A, Vanhaute O, et al. The effect of semi-automatic external defibrillation by emergency medical technicians on survival after out-of-hospital cardiac arrest: an observational study in urban and rural areas in Belgium. Acta Clin Belg 1997;52(2):72 – 83. Mosesso VN Jr, Davis EA, Auble TE, et al. Use of automated external defibrillators by police officers for treatment of out-ofhospital cardiac arrest. Ann Emerg Med 1998;32(2):200 –7. Sweeney TA, Runge JW, Gibbs MA, et al. EMT defibrillation does not increase survival from sudden cardiac death in a two-tiered urban-suburban EMS system. Ann Emerg Med 1998;31(2):234 – 40. White RD, Hankins DG, Bugliosi TF. Seven years’ experience with early defibrillation by police and paramedics in an emergency medical services system. Resuscitation 1998;39(3):145 – 51. Smith KL, Cameron PA, Peeters A, et al. Automatic external defibrillators: changing the way we manage ventricular fibrillation. Med J Aust 2000;172(8):384 – 8. StataCorp. Stata statistical software. In: Release, sixth ed. College Station: Stata Corporation, 1999. Bernard S. Outcome from prehospital cardiac arrest in Melbourne, Australia. Emerg Med 1998;10(1):25 – 9. Nichol G, Stiell IG, Laupacis A, et al. A cumulative meta-analysis of the effectiveness of defibrillator-capable emergency medical services for victims of out-of-hospital cardiac arrest. Ann Emerg Med 1999;34(4):517 – 25. Jaggarao NSV, Heber H, Grainger R. Use of an automated external defibrillator-pacemaker by ambulance staff. Lancet 1982;2(8289):73 – 5. Stults KR, Brown DD, Kerber RE. Efficacy of an automated external defibrillator in the management of out-of-hospital cardiac arrest: validation of the diagnostic algorithm and initial clinical experience in a rural environment. Circulation 1986;73(4):701 –9. Cummins RO, Eisenberg MS, Litwin PE, et al. Automatic external defibrillators used by emergency medical technicians. A controlled clinical trial. J Am Med Assoc 1987;257(12):1605 – 10. Gray AJ, Redmond AD, Martin MA. Use of the automatic external defibrillator-pacemaker by ambulance personnel: the Stockport experience. Br Med J 1987;294(6580):1133 – 5. Weaver WD, Hill D, Fahrenbruch CE, et al. Use of the automatic external defibrillator in the management of out-of-hospital cardiac arrest. New Engl J Med 1988;319(11):661 – 6. Vukov LF, White RD, Bachman JW, et al. New perspectives on rural EMT defibrillation. Ann Emerg Med 1988;17(4):318 – 21. Walters G, D’Auria D, Glucksman EE. Controlled trial of automated external defibrillators in the London Ambulance Service. J Royal Soc Med 1990;83(9):563 – 5. Sedgwick ML, Dalziel K, Watson J, et al. Performance of an established system of first-responder out-of-hospital defibrillation. The results of the second year of the Heratstart Scotland Project in the ‘Utstein Style’. Resuscitation 1993;26(1):75 – 88. Mols P, Beaucarne E, Bruyninx J, et al. Early defibrillation by EMTs: the Brussels experience. Resuscitation 1994;27(2):129 –36.