RESUSCITATION

ELSEVIER Resuscitation 33 (1997) 233 243

One-year survival after out-of-hospital cardiac arrest in Bonn city: outcome report according to the ‘Utstein style’

Matthias Fischer”**, Nicolas J. Fischer”, Ji.irgen Schiittlert’

,‘C?inic of Anaesthesiology and Intensive Care Medicine, Uniuersrty of Bonn. Sigmund-Freud Strasse 25. D-s-.? 105 BlW/ (~ermtrn,

‘Clinic of‘ Anaesthesiology. University qj’ Erlungen-Nltrenlherg Erlangen Germarn,

Received 3 February 1996; accepted 27 May 1996

Abstract

Outcome after prehospital cardiac arrest was examined in the EMS system of Bonn, a midsized urban community, and presented according to the Utstein style. The data were collected from January lst, 1989 to December 31st. 1992 by the Bonn-north ALS unit, which serves 240 000 residents. Fifty-six patients suffered from cardiac arrest of non-cardiac aetiology and were excluded; 464 patients were resuscitated after cardiac arrest of presumed cardiac aetiology (incidence of CPR attempts: 48.33 per year,‘100000 population). The collapse was unwitnessed, bystander witnessed or EMS personnel witnessed in 178, 214 or 72 patients, respectively. In these subgroups discharge rates and l-year survival accounted for 7.3% (4.5%). 22.9% (15.9%) and 16.7% (Il. 1%). respectively. Thirty-four patients were discharged without neurological deficits (cerebral performance category 1: CPC I), 22 and nine patients scored CPC 2 or CPC 3, respectively. Nine patients were comatose (CPC 4) when they were discharged and remained in this state until they died. Of the 50 l-year survivors 35 lived without neurological deficit, eight demonstrated mild (CPC 2) and five severe (CPC 3) cerebral disability at l-year after resuscitation. and. finally, two patients remained comatose for more than 1 year. The Utstein template recommends the selection of patients who were found in VF after bystander witnessed collapse. In our cohort 118 patients met these criteria. Of them 41 (35%) could be discharged from hospital and 28 (24%) lived more than 1 year. The comparison of our data with those from double-response EMS systems of other communities revealed that. in midsized urban and suburban communities the highest discharging rates could be achieved. Our study demonstrated that survival depends crucially on short response intervals and life support which will be performed by well-trained emergency technicians. paramedics and physicians. Copyright 0 1997 Elsevier Science Ireland Ltd.

Ke~vwrk: Cardiopulmonary resuscitation: Clinical investigation; Out-of-hospital cardiac arrest; [Jtstein-Style

1. Introduction

Cardiopulmonary resuscitation after preclinical car- disc arrest often results in poor neurological outcome [l]. Experimental studies are therefore required, to im- prove resuscitation techniques, to optimise pharma- cotherapy and to establish new approaches in cerebral

Ahhreviutions: ALS, advanced life support; BLS, basic life support; WC‘, cerebral performance category: CPR, cardiopulmonary resusci- tation; EMS, emergency medical service..

* Corresponding author. Tel: + + 49 228 2874114/2874138; fax: + + 49 221 2874125.

resuscitation [l]. On the other hand, clinical outcome studies are undoubtedly indispensable lo characterise determinants of outcome and to compare the effective- ness of different emergency medical services (EMS) systems. The latter, however. is hindered by variations in definitions as Eisenberg demonstrated in 1990 by reviewing resuscitation results of 29 studies [2,3]. To resolve this problem representatives from the European Resuscitation Council. the American Heart Associa- tion, the Heart and Stroke Foundation of Canada and the Australian Resuscitation Council met at Utstein Abbey (Norway) in 1990 and 5 months later at Surrey (UK). They established uniform terms and definitions

03OO-9572:97/$17.00 Copyright C 1997 Elsevier Science Ireland Ltd. All rights reserved P/l So700-9i7:(9h~01017-1

234 M. Fischer et al. I Resuscitation 33 (1997) 233-243

for out-of-hospital resuscitation and set up recommen- dations for uniform reporting of prehospital cardiac arrest data, which were published as the ‘Utstein style’ [4-71.

In regard to these guidelines a Medlinea search re- vealed that four outcome studies have been published so far. Kass et al. [8] presented data from a rural-subur- ban system. A total of 260 patients was found in ventricular fibrillation (VF), in 34% of them return of spontaneous circulation could be achieved, 8% survived up to hospital discharge and 6% were still alive after 1 year. In New York, the survival rate after out-of-hospi- tal cardiac arrest was evaluated by Lombardi et al. [9]. Of 415 patients 22 (5.3%) were discharged alive, when the collapse was witnessed and VF was found initially. In the second year of the Heartstart Scotland project VF was found in 643 patients after bystander witnessed cardiac arrest, 11% were discharged alive and 9% sur- vived for more than 1 year [lb]. Recently, Silfvast and Ekstrand published outcome data from two study peri- ods in Helsinki [1 11. In these two studies, VF was found in 120 and 130 patients after bystander witnessed car- diac arrest, respectively. Of these patients 34% and 25% could be discharged home [l 11.

The objective of our study was to estimate the effec- tiveness of the EMS system of Bonn, Germany, using outcome data after out-of-hospital cardiac arrest, which were determined and presented in accordance with the Utstein style.

2. Material and methods

2.1. Study area

The city of Bonn encompasses an area of 141 km2 with a resident population of 313 000 (48% male; 13.9% older than 65 years). Totals of 199 and 308 deaths per year/l00 000 of population occurred from ICD codes 410-414 and 410-429, respectively. Bonn is served by a double-response EMS system consisting of four basic life support (BLS) and two advanced life support (ALS) units admitting the patients to 11 hospitals. The ALS teams are manned by one physician and one emergency medical technician (EMT). All of the BLS units are staffed by two EMTs which are members of the munic- ipal fire department or ambulance organisations. One hospital-based ALS unit is located in the southern part of the city (ALS-south). The other is located in the northern part (ALS-north), in downtown Bonn. In the present study we retrospectively evaluated data from all out-of-hospital cardiac arrest patients between 1989 and 1992 in whom advanced cardiac life support was performed by the northern ALS unit. Two-thirds of the area and 75% of the population (240000) of Bonn are served by the Bonn-north ALS unit. All physicians of

this team are staff members of the department of anaesthesiology, University of Bonn, and have past, at least, 2 years of post graduate training before joining the ALS team. During their daily service in this team these physicians are not involved in other duties of the hospital.

2.2. Resuscitation algorithm

CPR was initiated or continued by the BLS team, which entered the scene first in 42% of the ALS unit’s resuscitation attempts. Defibrillation was not per- formed by the EMTs until 1991. A pilot project was then started, training the BLS teams to use external automated defibrillators. In the present study, there- fore, 17 patients are included, in which EMTs defibril- lated the patient before the ALS teams reached the scene. In all other cases the physicians started ALS by recording ECG and performing defibrillation when in- dicated. If return of spontaneous circulation (ROSC) could not be achieved external heart massage was con- tinued, the patients were intubated and received epinephrine. Initial dose of epinephrine amounted to 1 mg intravenously or, if an intravenous line was not in place, we diluted 2.5 mg of epinephrine in 5 ml 0.9% NaCl and injected it deep into the bronchial tree by using a catheter of 45 cm length [12-161. Resuscitation measures then followed the guidelines of the American Heart Association [17]. Patients were transported to hospital only when ROSC could be achieved and circu- lation could be stabilised. If spontaneous circulation could never be achieved within 30 min of CPR, physi- cians ceased their efforts and terminated CPR.

2.3. Dispatch system

Approximately 13 000 medical emergency calls per year were handled by the operators of the municipal fire department of Bonn city (fire and medical emer- gency number: 112). The BLS and ALS units were dispatched 13 000 times/year and 4500 times/year, re- spectively. The Bonn-north ALS unit treated about 3 000 patients per year. The ALS units were dispatched together with the closest available BLS unit in sus- pected life-threatening emergencies.

2.4. Documentation and follow-up

In a retrospective design all patients who were resus- citated by the Bonn-north ALS unit were identified from the collection of medical data. This database consisted of the physicians findings, diagnosis and treatment. In particular, physicians estimated the inter- val from call receipt until arrival of BLS or ALS unit (vehicle stop). On arrival at the scene, initial rhythm, pupil state and respiratory state were noted. The physi-

cians reported the occurrence of ROSC, the admission to hospital and, whether the collapse was witnessed and bystanders performed BLS. The rate of patients who were discharged alive was evaluated by reviewing the hospitals records. One-year survival data were collected by mail and phone contact with the primary physicians, the patient‘s family or the patient himself. Cerebral performance categories were determined for discharged patients and 1 -year survivors.

All data were entered into a database (dBase IV, Borland International, Scotts Valley, CA) designed ac- cording to the Utstein style. The determination of frequencies and survival rates was calculated by a com- puter program written by one of the authors (M.F.). Non-parametric and parametric variables were tested for difference using the lz-test or t-test, respectively. Two-tailed Y-values were provided and statistical sig- nificance was assumed for P < 0.05. Determinants of resuscitation success were identified by using multifac- torial regression analysis and X*-test.

3. Results

The results of the present study are summarised in Fig. I which show outcome data of all patients who suffered collapse of cardiac aetiology.

Box I Within an area of about 94 km2 240 000 residents were served by the ALS unit of the northern part of Bonn city. 48% of the population were male and 13.9% were older than 65 years. The study pe- riod covered 4 years from January 1st 1989 to December 31st, 1992. In this period about 12 000 patients were treated by the ALS team north, in 770 cases neither EMTs nor physicians started CPR due to post-mortem lividity or rigor mortis.

Box 2 In 602 patients the BLS teams confirmed cardiac arrest and started CPR. How- ever. the ALS team continued CPR in only 520 patients.

Box 3 In 82 patients the physicians terminated CPR, because the underlying disease had an unfavourable prognosis (malignancy. injuries incompatible with life) or, if the interval collapse to onset of BLS was greater than 15 min.

Box 4

Boxes 5, 6

Boxes 7~ 0

Box 10

Box 11

Box 12

Box 13

CPR according to the guidelines of the AHA was performed b>+ the ALS team in 520 patients.

The physicians confirmed another reason for cardiac arrest like hypoxia, trauma, exsanguation, intoxication and SIDS in 56 patients and, therefore. in 464 pa- tients cardiac aetiology was assumed (0% maIe, 54% > 65 years). Incidence of CPR attempts by cardiac arrest of presumed cardiac aetioiogy amounted, therefore, to 48.33 CPR attempts per year and 100 000 of population.

The collapse was unwitnessed in 178 pa- tients. bystander saw or heard the pa- tients collapse in 2 14 cases and cardiac arrest occurred after arrival of the ALS team in 72 cases. The median time inter- val from call receipt to vehicle stop, de- termined by using data of the computer-aided dispatch system. amounted to 5.5 min for the first re- sponding unit and 8.5 mm for the ALS team. 95% of the patients were reached by the first responding unit within 10 min. In 7% of all resuscitation attempts the ALS team was not dispatched to- gether with the BLS unit. which caused an additional delay. In bystander Gt- nessed cardiac arrest physicians esti- mated the time intervals from collapse to beginning of BLS and ALS. which amounted to 6.2 .k 4.8 min and 9.2 !-- 4.8 min, respectively.

?Jo electrical activ,ity of the heart was detected in 164 patients. Bystander CPR was performed in 1 ?%I of’ these asystolic patients.

VF was confirmed to be the initial rhythm in 200 out of 464 patients. By- stander CPR was performed in 23% of these patients.

Ventricular tachycardia was found in 10 patients, and 1 O’til received bystander CPR.

Other rhythms like bradycardia or elec- tromechanical dissociation were

236 hf. Fischer et al. / Resuscitation 33 (1997) 233-243

Fig. 1. Utstein template for reporting out-of-hospital cardiac arrest. Data from the EMS-system Bonn-North in the study period January Ist, 1989 to December 31st, 1992.

diagnosed in 90 out of the 464 patients. Bystander started CPR in 5 patients (4%).

Boxes 15,16 ROSC could be achieved in 248 out of the 464 patients in with collapse was assumed to be of cardiac aetiology. In 216 patients CPR was performed for at least 30 min and the physicians then ceased their efforts, because ROSC could never be achieved.

Box 17,18 Of the 464 (Box 6) 185 patients were

admitted to hospital with spontaneous circulation and measurable blood pres- sure. 163 of them survived for more than 24 h.

Box 19,20 Of 464 patients 74 were discharged from the hospital alive. 34 patients were dis- charged without neurological deficits (CPC = l), 22 and nine patients were scored to CPC 2 or CPC 3, res-pectively. Nine patients, however, were comatose when they were discharged and re- mained in a vegetative state until

18. Admitted to hospiial with 18. Admitted to hospiial with blood pressure measurable

18. Admilied to hr3spilai with

22 Alive at one year 22 Alive at one year

Fig. 2. Template for bystander witnessed cardiac arrest. Boxes I 6 are demonstrated in Fig. I data NC broken ~OM~II I>! initialI\> recorded cltrdiac rhythm.

they died. Summarising neurologi- cal outcome of discharged patients, 76% of them were independent in daily life whereas 24% suffered severe neurological deficit or remained coma- tose.

Boxes 2 1.22 One year after resuscitation 50 patients were still alive. Thirty-five of them lived without neurological deficit. Eight and five patients demonstrated mild or severe cerebral disability, respectively. Two pa- tients remained comatose for more than 1 year until they died.

In Figs. 2-4, outcome data are shown for those patients, which suffered from unwitnessed, by- stander or EMS personnel witnessed collapse. Survival rates were comparable in bystander and EMS witnessed cardiac arrest victims. Summarising all patients in these groups, ROSC could be achieved in 65% and 69(X, whereas 23% and 17% of these patients were dis- charged. Compared to the latter groups, survival rates were significantly lower, when the collapse was unwit- nessed. Spontaneous circulation could only be reached in 33% of the patients and 7% of them could be discharged.

Following the recommendations of the Utstein consensus conference the survival rates after by- stander witnessed VF were calculated. Additionally, the incidence of CPR attempts and the rate of dis- charged survivors per year,‘100 000 population were determined for all patients who suffered cardiac arrest of presumed cardiac aetiology. Our results in comparison to these of other double response systems and the Scottish study are shown in Table I. It reveals that the discharge rate after resuscitation from VF reached com- parable levels in Bonn. King County [18] and Helsinki [ll], but in EMS systems of rural and large urban communities, survival rates from cardiac arrest was significantly lower. In relation to our results, the inci- dence of CPR-attempts was significantly lower in King County [ 181 but significantly higher in Tucson 1990 [ 191, York-Adams 1993 181. New York [9] and Chicago [2O].

Survival after cardiopulmonary resuscitation is deter- mined by several factors. We. therefore, identified those determinants by multifactorial regression analysis and z-?-test (Fig. 5). First of‘ all, status of witness was found to influence outcome. Rates of ROSC. admission and long-term survival, were significantly better if the col- lapse was bystander witnessed. Secondly. bystander- CPR significantly improved short-term recovery and discharge rate, but l-year survival was not signniticantly different. Age, in contrast. did not affect initial resusci-

238 M. Fischer et al. / Resuscitation 33 (1997) 233-243

22. Alive at one year 22. Alive at one year

Fig. 3. Template for unwitnessed cardiac arrest. Boxes l-6 are demonstrated in Fig. 1, data are broken down by initially recorded cardiac rhythm.

tation success, but long-term recovery was more fre- quently observed in younger patients. Survival depends critically on delay until BLS and ALS were started. Our data revealed, that a short interval of collapse to begin- ning of ALS was associated with a higher rate of successful resuscitations. Our clinical findings support these conclusions. In fact, initial observation of VF, miotic pupils or gasping for breath correlated with better recovery and pointed to a shorter duration of ischemia, compared to those patients who were found to be asystolic, mydriatic and apnoeic. Finally, seven patients could be selected in our database, in whom all determinants corresponding to good prognosis were found. ROSC could be achieved in all of them and 57% of these patients lived for more than l-year. On the other hand, no patient was discharged from hospital if he was older than 70 years, found in asystole or EMD with mydriasis and apnoea after unwitnessed cardiac arrest without bystander CPR and ALS starting not before 6 min.

4. Discussion

The objective of the present study was to ascertain outcome rates from out-of-hospital cardiac arrest ac- cording to the recommendations of the Utstein consen- sus conference. The Utstein style template is an easy to use format to collect, report and compare outcome data after cardiopulmonary resuscitation. In this study we

collected and analysed all CPR attempts of the north- ern ALS unit of Bonn between January lst, 1989 and December 31st, 1992. We were able to gather informa- tion about all elements of the template. The numbers of bystander witnessed cardiac arrests or bystander CPR, however, might be underestimated, because these par- ticular points were not questioned by the CPR-protocol in the first year of the study. Information about the latter points, therefore, was extracted from the medical histories written by the emergency physicians or hospi- tal-based physicians.

In our study 464 patients were resuscitated by emer- gency physicians after cardiac arrest of cardiac aetiol- ogy. ROSC could be achieved in 248 (53.4%) patients, 163 (35.1%) lived for more than 24 h, 74 (15.9%) were discharged and 50 (10.8%) lived more than 1 year (Fig. 1). For comparison of different EMS systems, the Ut- stein consensus conference recommended the selection of patients who were found in VF after bystander witnessed collapse of cardiac aetiology. In our study 118 patients met these criteria, 41 of them were dis- charged alive (34.75%) and 28 (23.73%) lived for more than 1 year. Discharge rates of comparably high levels were reported from King County [18] and Helsinki [l 11, which are both midsized urban or midsized urban/sub- urban communities with double response EMS systems. In rural communities and more distinct in large cities, survival from VF is significantly less likely even if there is a double response EMS system installed. In rural communities discharge rate from witnessed VF ranged

22. Alive at one year

Fig. 3. Template for EMS-personnel witnessed cardiac arrest. Boxes I 6 are demonstrated in Vi?. I. data iIre brokcu tiown b\ initialI> recorded

cardiac rhythm.

from 8 to 20% [8,10,21], whereas in New-York [9] or Chicago [20] only 6% of the patients found in VF could be discharged.

These differences might be explainable by the inci- dence of CPR-attempts, the length of response inter- vals. qualification of personnel, sociodemographic features, and other factors. Becker et al. [22] pointed out that there is an inverse relationship between the incidence of CPR-attempts and survival after CPR. The calculated incidence accounted to 48.33 CPR attempts per year;‘100000 population for the Bonn-north ALS unit, which is comparable to the studies from Helsinki [I I]. Tucson 1992 [23] and York-Adams 1988 [21], and significantly higher than in King County [ 181 and Scot- land [IO]. In New-York [9], Chicago [20], York-Adams I993 [8] and Tucson 1990 [ 191, however, the incidences of CPR attempts were higher, which may in part ex- plain lower survival rates in these EMS systems.

From our point of view, however, it seems to be more important. that midsized urban communities pro- vided BLS within 2 6.5 min and ALS within 5-l 1.7 min after call receipt. In rural communities and large cities like New-York and Chicago, the interval from call to BLS and ALS exceeded 9 and 13 min, respec- tively. Several reports have previously demonstrated that shortening of. the intervals collapse to BLS or ALS improved outcome after resuscitation [18,24-261. These clinical observations are confirmed by experimental studies which demonstrated that reperfusion of heart

and brain was more difficult when cardiac arrest was prolonged [27,28].

The qualification of personnel and the organisation of the EMS system are further determinants which influenced outcome critically. The latter point was cx- tensively discussed by Eisenberg 121. He reviewed re- ports about 29 EMS systems and characterised five different systems. The best performance was achieved by double response systems with EMTs performing defibrillation and paramedics [2]. This partly explained poor outcome in Scotland, where ambulance crews only performed basic life support and external automatic defibrillation but not ALS. Of the cardiac arrest pa- tients. therefore, 83’::1 were transportetl to the next hospital while BLS was continued.

Within double response systems, however. differences in qualification of personnel may also exist. which influence survival rates. A previous study frorn Bonn suggested [29] that well-standardised therapeutical strategies, like early defibrillation. immediate endotra- cheal intubation and swift endobronchial epinephrine administration increased resuscitation success. In Helsinki and Bonn, like some other European countries [30], the ALS units are staffed with physicians. This might improve resuscitation success and, explain in part, better survival rates in Bonn and Helsinki [1 l] compared to Milwaukee 1311 and TUCSOIJ [19,23]. Physi- cians. possibly, better adapt resuscitation algorithms to the individual patient as paramedics might do,

Tabl

e 1

Out

com

e fro

m o

ut-o

f-hos

pita

l ca

rdia

c ar

rest

in d

oubl

e re

spon

se s

yste

ms

Stud

y Se

tting

Sy

stem

By

stan

der-

Dis

char

ged

Bon

n, 9

6 H

elsi

nki,

P I

96 [l

l] Ki

ng C

ount

y, 8

5 [1

8]

Hel

sink

i, P

II 96

[II]

Milw

auke

e, 8

6 13

11

Tucs

on, 9

0 [1

9]

Tucs

on, 9

2 [2

3]

Yor

k-A

dam

s,

88 [2

1]

Scot

land

, 93

[IO

] Y

ork-

Ada

ms,

93

[8]

New

Yor

k C

ity,

94 [9

] C

hica

go,

91 [2

0]

Mid

size

d ur

ban

Mid

size

d ur

ban

Mid

size

d ur

ban/

sub-

ur

ban

Mid

size

d ur

ban

Mid

size

d ur

ban/

sub-

ur

ban

Mid

size

d ur

ban

Mid

size

d ur

ban

Subu

rban

/ ru

ral

Subu

rban

/ ru

ral

Subu

rban

/ ru

ral

Larg

e ur

ban

Larg

e ur

ban

witn

esse

d V

F

BLS

+ AL

S 11

8 BL

S +

ALS

120

BLS

+ AL

S 95

1

BLS

+ AL

S 13

0 BL

S +

ALS

179

BLS

+ AL

S 96

BL

S +

ALS

118

BLS

+ AL

S 38

2 BL

S 64

3 BL

S +

ALS

260d

BL

S +

ALS

415

ALS

371

patie

nts

41 (

35%

) 41

(34

%)?

31

4 (3

30/o

)?

33 (2

5%)t

183

(24%

)’

21 (

22%

)*

18 (1

50/o

)**

47 (

12%

)**

71 (1

10/o

)**

21 (

8%)*

* 22

(s%

)**

15 (4

%)*

*

BLS,

bas

ic li

ve s

uppo

rt; A

LS,

adva

nced

live

sup

port;

VF,

ven

tricu

lar

fibril

latio

n;

ND

, no

t de

term

ined

. “C

all

rece

ipt

to B

LS-te

am a

rriva

l (c

all

rece

iptto

ALS

-team

arri

val).

bC

ardi

ac a

rrest

of

pres

umed

car

diac

aet

iolo

gy p

er y

ear/l

00 0

00 p

opul

atio

n.

“Dis

char

ged

patie

nts

afte

r ca

rdia

c ar

rest

of

pres

umed

car

diac

aet

iolo

gy p

er y

ear/l

00 0

00 p

opul

atio

n.

dWitn

esse

d and

unw

itnes

sed

VF

tNS

, P >

0.0

5.

*P <

0.0

5.

**p

< 0.

001.

6.5

(11.

7)

Res

pnse

inte

rval

”

60.9

8*

Inci

denc

e of

CPR

atte

mpt

sb

2.0

(5.0

)

(min

)

ND

5.5

(8.6

) 48

.33

5.4

(11.

0)

46.4

7t

4.4

(9.2

) 38

.91*

*

3.0

(5.0

) 62

.08*

4.

1 (5

.1)

50.7

3t

ND

47

.11t

9

32.8

5**

ND

73

.05*

* 9.

4 (1

3.0)

63

.52*

* 13

(ND

) N

D

7.25

t 2

ND

Dis

char

ged

%

patie

ntsC

5.21

t i

3.00

**

E 2.

98**

7.11

2.

s 3.

41**

9.80

7

z-.

4.39

*

%

9 1.

42**

2

ND

7.10

t

2 3 %

:: 2 s-

SURVIVAL BY WITNESS SURVIVAL BY RYSTANDER-CPR

100 100 NOT EMS WITNESSED

1 1 t UNWITNESSED. N=l78 1 1 1 -&-NO CPA, N=321 1

80 i *

+ WITNESSED, N=214 1

ROSC -ADMISSION ALIVb24H -DISCHARGE- ALIVE>lYR

SURVIVAL BY INITIAL CARDIAC RHYTHM

01

ROSC ADMISSION At IVE>24H DISCHARGE ALIVE (YQ

SURVIVAL BY PUPIL STATE

mn- NOT EMS WITNESSED 1 !I0 1 1 -CASYSTOLE+EMD, bk254 1 ‘“” 1 1-t MY~RIATE, ~r.209 1

80 0 VFtVT. N=ZlO I

ROSC ADMISSION ALIVE>24H DISCHARGE ALIVE>iYR

SURVIVAL BY AGE

AOSC ‘ADMISSION ‘ALIVE>P4H ‘DISCWARGE’ALIVE>‘YF”

SbRVIVAL BY RESPIRATORY STATE NOT EMS WITNESSED

1 ROSC ADMISSION ALIVE>24H DISCHARGE ALIVE>lYR ROSC ADMISSION ALIVEa24H DISCHARGE ALIVE.1 vR

100,

80.

;i s 60.

2 2 40. * 0

20 -

O-

SURVIVAL BY INTERVALL COLLAPSE ALS BYSTANDER WITNESSED

* 1 4>9 MIN. N-46 1

1 ~, ~,~ ~~_~~ ~.

ROSC ADMISSION ALIVE>24H DISCHARGE ALIVE>iYR

SURVIVAL BY THE DETERMINANTS FROM ABOVE

ROSC ADMISSION ALIVF>24H DISCHARGE At IVE>1 Y’I

Fig. 5. Determinants of outcome. frequencies in percent of resuscitated patients. *Significant difference betwren the groups ((‘hr-square test)

Finally, Lombardi et al. [93 pointed out, that sociode- mographic features can partially be responsible for poor survival rates in large cities. They described, that in New York and Chicago compared to King County less patients were found in VF/VT even in the subgroup of patients with EMS witnessed collapse. In this group the influence of response intervals was eliminated but sociodemographic features like racial composition of the cohort [32], socioeconomic status and substandard health care might have reduced rate of VF and worsen

resuscitation outcome. Interestingly we found no difference in survival after

EMS personnel compared to bystander witnessed car- diac arrest, although the delayed onset of BLS in the latter group might decrease resuscitation success. These differences could in part be explained by the lower incidence of VF in the EMS personnel witnessed group, which corresponds to a poor prognosis [3,29]. Addi- tionally, Myerburg in 1989 discriminated between at least two underlying mechanisms of cardiac arrest.

242 M. Fischer et al. / Resuscitation 33 (1997) 233-243

which correlate to different prognoses [33]. First, thromboembolic obstruction of an coronary artery might occur leading to myocardial infarction and sud- den cardiac death [33]. In this case prognosis is poor, but patients often suffered from chest pain hours before the collapse occurred, and the EMS teams, therefore, might be at scene more frequently. Second, the collapse might be caused by electrophysiological abnormalities of the heart, which may lead to sudden cardiac arrest without prodromi and better prognosis [33]. The low survival rate in our patients suffering cardiac arrest in the presence of the EMS teams, therefore, may in part be caused by the poor prognosis of patients resuscitated after myocardial infarction. Not surprising, however, was the finding, that all survivors of this group showed excellent neurological recovery, because cerebral is- chemia could be prevented by immediately initiated BLS.

Our study clearly demonstrate that outcome depends on several determinants (Fig. 5), which all, with the exception of age, are related to or influenced by the duration of cardio-circulatory arrest. This emphasises the hypothesis that under clinical conditions survival depends crucially on duration of ischemia. In detail, with prolongation of ischemia, VF convert to EMD or asystole [l&34,35], pupils more and more dilate and respiration changed from gasping for breath to apnoea. Our study clearly revealed that patients, who were found either in VF, or pupils not dilated or still gasping for breath, had a better prognosis compared to those who were found in asystole, with dilated pupils or apnoeic. Secondly, witnessed cardiac arrest victims had a better prognosis, because the interval from collapse to call will be reduced and the chain of survival will be initiated earlier. Bystander CPR, finally, improved sur- vival rate, as duration of ischemia will be reduced [l&26].

In summary, survival from out-of-hospital cardiac arrest differs widely, even if the more homogenous subgroup of patients was analysed which were found in VF. Outcome depends crucially on the duration of cardio-circulatory arrest. Survival rate increases when the EMS-system provides BLS and ALS by well-trained teams within 5 and 10 min, respectively. Discharge rates, however, greater than 30% are achievable in patients living in midsized urban communities and found in VF. Efforts, which reduce the duration of ischemia will increase the number of patients who will survive after resuscitation from cardiac arrest. Public programs, therefore, are required teaching lay persons to recognise emergencies, to initiate the chain of sur- vival immediately and to perform BLS. The communi- ties, however, are requested to optimise their EMS systems. BLS must be provided within 5 min by EMTs trained in automated defibrillation [36-381 and ALS has to be initiated within 10 min after collapse by well-trained physicians or paramedics.

References

PI

PI

131

141

PI

PI

171

PI

Safar P. Past, present and future of cardiopulmonary-cerebral resuscitation (CPCR). Minerva Anestesiol 1994; 60: 483-486. Eisenberg MS, Horwood BT, Cummins RO, Reynolds-Haertle R, Hearne TR. Cardiac arrest and resuscitation: a tale of 29 cities. Ann Emerg Med 1990; 19: 1799186. Eisenberg MS, Cummins RO, Larsen MP. Numerators, denomi- nators, and survival rates: reporting survival from out-of-hospi- tal cardiac arrest. Am J Emerg Med 1991; 9: 544546. Cummins RO, Chamberlain DA, Abramson NS, Allen M, Bas- kett PJ, Becker L, Bossaert L, Delooz HH, Dick WF, Eisenberg MS, et al. Recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest: the Utstein Style. A statement for health professionals from a task force of the American Heart Association, the European Resuscitation Coun- cil, the Heart and Stroke Foundation of Canada, and the Australian Resuscitation Council. Circulation 1991; 84: 960- 975. Cummins RO, Chamberlain DA, Abramson NS, Allen M, Bas- kett P, Becker L, Bossaert L, Delooz H, Dick W, Eisenberg M, et al. Recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest: the Utstein Style. Task Force of the American Heart Association, the European Resuscitation Council, the Heart and Stroke Foundation of Canada, and the Australian Resuscitation Council. Ann Emerg Med 1991; 20: 861-874. Recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest: the ‘Utstein style’. Prepared by a Task Force of Representatives from the European Resuscitation Council, American Heart Association, Heart and Stroke Foun- dation of Canada, Australian Resuscitation Council. Resuscita- tion 1991; 22: l-26. Chamberlain D, Cummins RO. Recommended guidelines for uniform reporting of data from out-of-hospital cardiac arrest: the ‘Utstein style’. European Resuscitation Council, American Heart Association, Heart and Stroke Foundation of Canada and Australian Resuscitation Council. Eur J Anaesthesiol 1992; 9: 245-256. Kass LE, Eitel DR, Sabulsky NK, Ogden CS, Hess DR, Peters KL. One-year survival after prehospital cardiac arrest: the Ut- stein style applied to a rural-suburban system. Am J Emerg Med 1994; 12: 17-20.

[9] Lombardi G, Gallagher J, Gennis P. Outcome of out-of-hospital cardiac arrest in New York City. The Pre-Hospital Arrest Sur- vival Evaluation (PHASE) Study. J Am Med Assoc 1994; 271: 678-683.

[lo] Sedgwick ML, Dalziel K, Watson J, Carrington DJ, Cobbe SM. Perfotmance of an established system of first responder out-of- hospital defibrillation. The results of the second year of the Heartstart Scotland Project in the ‘Utstein Style’. Resuscitation 1993; 26: 75-88.

[l l] Silfvast T, Ekstrand A. The effect of experience of on-site physicians on survival from prehospital cardiac arrest. Resuscita- tion 1996; 31: 101-105.

[12] Hornchen U, Schtittler J, Stoeckel H, Eichelkraut W, Hahn N. Endobronchial instillation of epinephrine during cardiopul- monary resuscitation. Crit Care Med 1987; 15: 1037-1039.

[13] Schtittler J, Bartsch A, Ebeling BJ, Hornchen U, Kulka P, Suhling B, Stoeckel H. Die endobronchiale Adrenalinapplikation bei prlklinischer kardiopulmonaler Reanimation. Anasth Inten- sivther Notfallmed 1987; 22: 63-68.

1141 Hornchen U, Schiittler J, Stoeckel H, Eichelkraut W. Phar- makokinetik und Dynamik von endogen freigesetztem und therapeutisch appliziertem Adrenalin unter Reanimationsbedin- gungen. Anaesthesist 1988; 37: 6155619.

[Ii]

[Iti]

/I?]

Bartsch A. Schiittlcr J. Hiirnchen U. Stoeckel H. Die Technik

der endobronchialen Adrenalinapplikation bei der kardiopul-

monalen Reanimation. Notarrt 1989: 5: 197 199.

Schiittler J. Pharmacocin&ique des mbdicaments utilisCs pour la

phase dc relance cardiaque. Ann Fr Anesth Reanim 1990; 9:

20X 211.

Standards and guidelines ibr Cardiopuhnonary Resuscitation

(CPR) and Emergency Cardiac Care (ECC). J Am Med Assoc

19X6: 25s: 2YO.5 298’).

[IX] C‘umminh RO. Eisenber, ~7 MS. Hallstrom AP. Litwin PE. Sur-

\ 1\a1 of out-of-hospital cardiac arrest with early initiation 01

c~lrdiopuln~on,~r~ resurcitatton. .\m J Emerg Med 1985; 3: I I4

1 IY]

IN]

!‘I]

1221

!23]

1141

1251

[JfJl

I I‘).

Spaite DW. Hanlon T. C‘r~\s EA. Valenzuela TD. Wright AL.

Keeley KT. Mcislin IfW. Prehospital cardiac arrest: the impact

(‘I‘ witnessed collapse and bystander CPR in a metropolitan EMS

\!stem with hhort rcspon’c tlmca. Ann Emerg Med 1990: I Y:

IX4 I?_69

Becker LB. Ostrander MP. Barrett J. Kondos CT. Outcome ol

(‘PR in il large metropolitan arca \vhere art’ the survivors’?

Ann Hnierg hlcd 1991: 20: 355 361.

Fitel DR. Walton SL, Gucrci AD. Hess DR. Sabulsky NK.

(itit-of-hosl,ital cardiac arrcat. ;I ,ix--year experience in a subur-

ban-t-ural system. /\nn Eiiiq R/led 198X: 17: 808 -812.

Becker LB. Smith DW, Rho&s KV. Incidence of cardiac arrest:

4 neglected factor in c\alu:rting \ur\ival rates. Ann Emerg Med

I’)‘): 22: X6 ‘11

Calenacla TD. Spaite DW. Meislin HW, Clark LL, Wright AL.

t v,! (;A. C’ase and survival delinitions in out-of-hospital cardiac

xrrcat. J Am Mrd Assuc IYY2: 367: 272 174.

Mullic !\. Van-Hoeyweghen R. Quet+ A. lntluence of time

Intcrb;tl\ on outcomc of CPR. The Cerebral Resuscitation Stud)

6roup Resuscitation IYXY: 15: 5199 206.

Valcntucla TD. Spaitr DW. Mcislin HW. Clark LL. Wright AL.

1 \\! (iA. Emergency vehicle intervals versus collapse-to-CPR

,111tl collapse-to-detibrillotinn intercals: monitoring emergency

medical \erviccs system performance in sudden cardiac arrest.

41111 Emcrg Med 1993; 22: 1678 16X3.

‘Martens PR, Mullie A. Callc P. ITan-Hoeyweghen R. Influence

l)n outcome after cardix arrc\t of time elapsed between call for

help and start of bystander basic CPR. The Belgian Cerebral

R~\u\c~t:tticln Stud! (iroup. Resuscitation 1993: 25: 227- 234.

P71

WI

P91

[30]

VII

(321

[33]

I341

WI

P61

Lee SK. Vaagenes P. Safer P. Sterozki SW. ‘sumIon M. f:ffect ol

cardiac arrest time on cortical cercbm! hlootl tlo\\ during subse-

quent standard external cardiopulrrlotlar? / cru\citation in i-ah-

bits. Resuscitation 1989: 17: I05 I ! 7

Fischer MS. t-Iossmrnn K-A. No-rcflo\t $tci- ~xd~ac ‘II rest

lntens Care Med I YY5: 2 I : I32 I-I I.

Schiittlcr J. Bartsch AC. Bremcr 1~. L belllis BJ. I~odisch M.

Kulka P, Pllitsch II. Effiricnr dcr pr;iklini\chcn kardiopul-

1mm~1c11 Rcaninlation. Welche Faktol-en hc\rlmmcn &,n E:rinl$l 4na~th Intensivthcr Notfallmed I YOI). 25: 3-10 347.

Bocsacl-t LL. The complexity (11‘ CornparIng dif’i;‘rent E!MS sys-

tern5 a ,une! (I/‘ FMS ysicnis in F urrqk \IIII i,tncrs ‘Maxi

1YY.J; 2’. YY 103. -. Stueven H. Troiano P, l‘homp\on B. vvl;~tca IK. Ka>tenson El I.

Tonsfeldt D. Hargartcn K, Kowalski R. .\p~-ahamixrt (I‘. D,irin

J. Bqbtandcr tir\t responder C’PR: ten !ea~-+ cxpcr.gencc in a paramedic \)stem. Ann Emerg Mcd 1Y)XO. I + *:i)T -I!).

Becker LB. Han BH. Meyer PM. Wright I A Rhode\ KV. Smith

IIW. Barrett .I Ract‘~l difference\ 111 the iixitlcnc~~ of c;~rdi;tc

arrest and \ubscqut‘rr~ sur\iLal. TII< (‘1’1~ t ‘III<~<~I Proicct. U

hyl J tlcd lYY‘*; 321): 600 606

Mycrburg KJ. Kcs?lcl- KM. Basreri ,\L. I’;:\lcl/‘ill~~, ,\. A lxo-

logicai approxtl IO \uddc‘n cardiac dcarh: \t ruL,t urc. frlncti~>n

md cauw. :Zm J ( ‘;irtliol I YXO: 63: i i III I i / h

Bro\\n fG. Dz\\onc;,~k R. Wcrman I IA. li.~mlrn RI t;ctmi;tt-

~nf lhc dut-at1011 of xntricul;ir tihrill:iiiol! :2nc I-?~crg .led

198’): 1s. II81 118’ Martin DR. Brown f-C;. I~rwoncrqb K. I ~~~quc‘nq ;~nalysi\ 01

the human and $winc electrocardiogl;~trl tiuri *:c \.t3tri~‘lil;ir tibril-

lation. Resuscitation 1901: 22: X5 ‘)I Schneider T. Maurr D. Diehl P. Dick W. H:-<hmsr I’ Juchcms

R. Kcttler D. Klcine-Zander R. Klm&x IH. ilio\x~ K. t*t II F.:rrly

delibrill;tllon by smcrgenq phy\lc’l:mh or \mcrccnL\ medrcal ~ technici;ins’! 12 contrc~llcd. prospccti\c muIll-~.,:it::i. \lud\. Rc3u5-

citation lYY4: 17. I Y-’ 206