- Email: [email protected]
Pre-CPR conditions and the final outcome of CPR
Resweitatkm, 17, Suppl. (1989) Sll - 521 Elsevier Scientific Publishers Ireland Ltd.
Sll
PRE-CPR CONDITIONS AND THE FINAL OUTCOME OF-CPR
A. MULLIE’, P. LEWP, R. VAN HOEYWEGHEN’ and THE CEREBRAL RESUSCITATION STUDY GROUP “Department of Critic01 Care Medicine, Algemeen Ziekenhuis Sint Jan, Ruddershove 10, 8000 Brugge, bInfonnatkm Science Department, Janeeen Reeearch Fowzdutima, Janeeen Pharmaceutics NV, B-2340, Beerse and ‘Department of Intensive Care and Emergency Medicine, University Antwerp - U.I.A., Universiteitsplein 1, 2610Antwerp lBelgium)
SUMMARY
Outcome of cardiac arrest (CA) is very much influenced by pre-CPR conditions. To assess the importance of these pre-CPR factors, an analysis of the Belgian CPCR registry was made according to some pre-CPR conditions. In this registry, several variables related to pre-arrest, arrest, CPR and post CPR period have been recorded in 4548 patients. The pre-CPR conditions studied were: age, witnessed event or not, pre-arrest health state, underlying disease, site of cardiac arrest, type of respiratory arrest and type of cardiac arrest. Age did not influence outcome significantly. The importance of witnessing is very significant. Severe pre-arrest disability reduces chances on long-term survival (LTSl to half and overall health status longterm survivors is clearly less. Intoxication and metabolic origin of CA have good prognosis (LTS, 21%). Trauma/exsanguination, drowning, SIDS and sepsis have bad prognosis (LTS, 1 -3%). Cardiac (LTS , 12%) and respiratory (LTS, 14*,411 origin have similar outcome, although significant difference exists in occurence of cerebral failure, suggesting that post-ischemic encephalopathy is more severe in respiratory than in cardiac origin. The most frequent site of CA, the home of the patient. has poor outcome results (LTS, 5%). Gasping is significantly related to successful outcome. In the out-of-hospital setting the type of CA was 25% VF (LTS, 77%1,65% asystole (LTS, 4%) and 10% EMD (LTS, 3%). Outcome of the subgroup out-of-hospital, witnessed, VF is comparable to other reports. This sub-group seems to us the most appropriate for clinical trials.
Key words: Cardiopulmonary resuscitation - Pre-CPR conditions - Outcome
0300-9672/89/$03.50 0 1989 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
s12 INTRODUCTION
Reports on the survival of cardiac arrest (CA) vary widely from less than 5% [l] to 600/b[2] It is generally accepted that CPR results are more influenced by the pre-CPR conditions than by the CPR treatment itself. The CA-CPR syndrome is a complex interplay of what has been called ‘fate factors’ (age, underlying disease...) and ‘program factors’ (response times of basic and advanced life support..) [3]. The results of CPR outcome studies are therefore difficult to compare. Several authors have noticed the lack of clarity and precision in language used to report CPR study results and patient selection criteria; they made a plea for a uniform reporting system of outcome (shortterm as well as long-term survival), in order to improve comparability and accurate measurement of the impact of CPR treatment and EMS organisation [3-61. The objective of this report is to outline the influence of the important pre-CPR factors. This should allow a reasonable comparison between outcome studies in different centers, as well as an outline of an optimal patient category for multi-institutional therapeutic CPCR trials.
METHOD
The registration method has been described before [7-g]. In summary, several variables related to the pre-arrest, arrest, CPR and post-CPR period have been registered prospectively on a uniform simple data collection form, on all CA cases from 7 major hospitals attended by their prehospital EMS-MICU services in Belgium. Post- and pre-arrest health status was classified with the simple Jennett-Bond scale [lo]. The outcome has been stratified in 4 classes: Class 1, immediate CPR failure, spontaneous circulation cannot be restarted; Class 2a, CPCR failure due to cerebral failure patients whose circulation can be restored but die within 14 days from cerebral failure or remain in a vegetative state at 14 days post-CPR; Class 2b, CPCR failure due to cardiac, septic or other failure patients whose circulation can be restarted but die within 14 days from non-cerebral failure; Class 3, CPCR successes, patients awake 14 days postCPR either normal (good recovery), moderately disabled (disabled but independent) or severely disabled (dependent on others). The studied pre-CPR variables have been restricted to the following: age, place of CA, pre-arrest health state, underlying disease preceding CA, whether the CA was witnessed or not and, finally, the type of CA and respiratory arrest (RA) on arrival of the MICU team. Other variables, e.g. the hour of the day, have been registered and studied but were without major influence on outcome, contrary to some previous reports [ll. Another very important variable, bystander CPR will be reported extensively elsewhere in this journal. Diseases, underlying the CA were grouped as follows: cardiac, trauma and
s13
exsanguination, respiratory, cerebral, intoxication, metabolic, drowning, septic, sudden infant death syndrome @IDS). Among other variables the pupillary signs and quality of CPR performance was scored. The pupillary signs pre, during and post CPR were grouped in 2 types: bad quality when pupils were dilated, pinpoint or unequal without light reaction and good quality when normal or unequal pupils had slow or normal light reaction. The pupillary signs were evaluated by the MICU team, on arrival at the scene. The quality of CPR by the different performers was categorised as efficient (correct technique with good effect) or inefficient (bad technique or correct technique with weak effect1 according to the chest movements by insufflation and the quality of the palpated pulse under external cardiac compressions (ECC) on arrival of the MICU team. Arrests were considered to be witnessed according to the time of CA to call, a CA-CALL time (1 min, was used as the criterium that someone was present during the event. This interpretative term of witnessed or not was tested during 1986- 1987 with more explicit interrogation of bystanders and found to be correct in over 90% of the cases. CA-CPR is the time interval between CA and the start of basic life support either by a bystander, EMT, nurse or medical MICU team. Qualitative data have been analysed for significance by means of the x2 test; quantitative data have been analysed by means of the Student’st-test. Significance is assumed at the 0.05 level of probability. Average values are reported as mean + S.D. RESULTS
From 1983 to 1987, 4548 patients with CA have been treated by the study centers. Overall results are summarised in Table I. According to the outcome criteria used, CPCR is successful in 18% of the in-hospital CA’s and in 7% of outhospital CA’s
TABLE 1 CPCR REGISTRY - BELGIUM (1983- 1987) - OVERALL RESULTS Comparison in outcome between in-hospital and out-of-hospital cardiac arrest patients. The numbers in between brackets are percentages.
In out In x Out
Class 1 (CPR failure)
Class 2a (Cerebr. failure)
Class 2b (Other failure)
Class 3 Total (CPCR success)
959 (65) 2445 (79) 3404f75)
8163) 267 (9) 348 (7)
155 (11) 153 (5) 308 (7)
270 (18) 218 (7) 488 (11)
1465 3083 4548
P< 0,OOOl
s14
Influence of age Mean age was not systematically related to outcome. Classes 1,2a, 2b and 3 were, respectively, 61 f 19,58 -c 21,67 f 15,63 f 14 years old. Considering only CA of cardiac origin, mean age for classes 1,2a, 2b and 3 was 67 + 13.65 f 14,70 f 11,65 f 13years. Pre-arres t health status In 1433 in-hospital CAs, pre-arrest health status was recorded as normal moderately disabled or severely disabled in, respectively, l5%, 51% and 34% of the patients. In 3066 out-of-hospital CAs, the pre-arrest status was normal in 57%. moderately disabled in 36% and severely disabled in 7%. Both an influence on overall outcome and an influence on the degree of disability in survivors, can be illustrated. The infuence of pre-arrest health status on outcome in the 1433 in-hospital patients (with a significant occurrence of severe disability pre-arrest) is summarised in Table IIa; the influence of the pre-arrest health status on the degree of disability in the 485 surviving patients (in- and out-hospital) is summarised in Table IIb.
TABLE II PRE-ARREST HEALTH STATUS - OUTCOME /al Iwhoupital In = 14.991.Comparison in outcome between I433 in-hospital CA patients in normal health status, moderate disability and severe disability pre-arrest. The numbers in between bracketa are the percentages of patients in that particular health status pre-arrest. The numbers in the table are the percentages, ending in each outcome category, of the total number (166%) in that particular health status. Class 1
Class 2a (Cerebr. failure)
Class 2b (Other failure)
Class 3 (CPRC success)
62 62 73
6 5 6
8 12 10
24 22 11 P< 0.001
(CPR failure) Normal (151 Mod. disabled (511 Sev. disabled (341
(bl Class 3 outcome patients only h = 4&V. Comparison in percentage of severe disability postarrest in 485 surviving CA patients, in either normal health status, moderate disability and severe disability pre-arrest. The numbers in between brackets are the percentages of the patients in that particular health status pre-arrest. The numbers in the table are the percentages of the patients, in that particular health status pre-arrest, ending up with severe disability post-arrest. Severely disabled post-arrest (%I Normal (361 Mod. disabled (50) Sev. disabled (14)
16 26 92 P< 0.01
s15
Underlying disease The disease preceding the CA was cardiac in 76%. trauma and/or exsanguination in 9%, respiratory in 60/b, cerebral in 2% intoxication in 2% metabolic in 2W, drowning in l%, sepsis in 1% and SIDS in 1%. The influence of the underlying disease on outcome is tabulated in Table IIIa, and is highly significant. Although CPCR success is as frequent in respiratory as in cardiac origin patients, cerebral failure is more than twice as frequent in the respiratory origin group. Focussing on differences between cardiac and respiratory origin patients, the following relatively minor differences could be found: ventricular fibrillation was present in 33% in CA of cardiac origin, 8%~in respiratory origin; CPR was performed by bystanders in 38O/bin CA of cardiac origin, 25% respiratory origin; the effect of ECC, judged by MICU as good quality, was 85% in cardiac origin, 84O/brespiratory origin; good pupillary signs during CPR was present in 25O/bin cardiac origin, in 18% in respiratory origin; good pupillary signs 10 min post CPR were present in 62% in cardiac origin and in 56% in respiratory origin. Table IIIb summarises the significant proportional difference in outcome between patients of cardiac and respiratory origin, even with similar neurological status 10 min post initially successful CPR (as judged by the reaction of the pupils). In the 24-h period post CPR, blood gas abnormalities (7.3 < pH < 7.6; Pao, < 50 mmHg; 25 mmHg < PCO, < 50 mmHg1 were more frequently present in respiratory cases (42%) than in cardiac cases (30%) and hyperthermia (t/O > 38.5OCl occurs in 18O/b of respiratory cases vs. 9% in cardiac origin. Place of CA The MICU teams of the study groups attended both in- and out-of-hospital CA situations. Although there are some differences in the in-hospital emergency call system between different centers, overall occurrences of CA in-hospital vs. outside was about 1 to 2. In-hospital, CA occurs in 54%~on the ward, 23Ok,in the emergency room (ER), 200/o in the ICUlCCU and 3?b during transport. Out-of-hospital CA occurred in 16% on the roadside, 660/b at home, 100~ at public places, 2% at work and 6% during transport. Table IV summarises the influence of the place of CA on outcome, the influence is highly significant. Witnessed vs non-witnessed In-hospital CA is witnessed in 81% of the cases. Out-of-hospital CA is witnessed in 53OEo of the cases. Table IV summarises the influence on outcome both in all out-of-hospital patients and in out-of-hospital ventricular fibrillation patients only, the influence is highly significant. Initial ECG rhythms In the in-hospital setting VF, asystole and electromechanical dissociation (EMDl are present, respectively, in 3OW, 49% and 21°h. Outside the hospital VF, asystole and EMD are present in 25%, 65% and 10%. When only cases of
Cardaic (761 Trauma Exanguination(91 Respiratory (56) Cerebral (21 Intoxication (21 Metabolic (21 Drowing (1) Sepsis (11 SIDS (11 83 94
31 7 6
18 26 9
65 63 59 50
7 3
74 95
Class 2a (Cerebr. failure)
11 10 0
4 1 12
7 1
Class 2b (other failure)
8 0 0
13 10 20
12 1
Class 3 (CPCR success)
86 90 80
a4 85 89
8.5 a.3 8.0 17.0 3.3 42.3 P< 0.0001
85 58
ECC-MICU good quality 6.0 6.7
CA-CPR time
1. 2.
Cardiac (621 Respiratory (661
16 36
Class 2a (Cerebral fail)
26 10
Class 2b (other failure)
58 54 P< 0.001
Class 3 (CPCR success)
(b) Difference card& vs. respiratory - outcome - pup&by sigus good 10 min post CPR In = 575). Comparison in outcome between patients with similar neurological status 10 min post-arrest in 528 CA patients, cardiac origin vs. 47 CA patients, respiratory origin. The number in between brackets is the percentage of the patients, in that particular CA origin group, with good pupillary signs 10 min post-arrest. The numbers in the table are the percentages, ending in each outcome class, of the total number (100%) of initially successful CPR patients, with good pupillary signs 10 min post CPR, in that particular CA origin group.
6. 7. 8.
3. 4. 5.
1. 2.
Class 1 (CPR failure)
laJAll patients /n = 45021.Comparison in outcome between the different underlying diseases, preceeding the CA. The number in between brackets is the percentage, of the total number of patients, with that particular underlying disease. The numbers in the table are the percentages, ending in each outcome category, of the total number (100%) in that particular underlying disease. The added 5th and 6th column are the duration of CA (in min) and the percentage of patients resuscitated with good palpable pulses during ECC, in each underlying disease.
UNDERLYING DISEASE - OUTCOME
TABLE III
s17 TABLE IV OTHER PRE-CPR VARIABLES - OUTCOME Comparison in outcome between witnessed and non-witnessed CA patients. The number between brackets is the percentage of patients in that particular category; for the site of the arrest, relating to a total 1OOI in hospital and 100% out-of-hospital. The numbers in the table are the percentages, ending in each outcome class, of the total number of patients @O%) in that particular category. Class 1 (CPR failure)
Class 2a (Cereb. failure)
Class 2b (Other failure)
Class 3 (CPCR success)
10 8 13 13
16 18 25 11
Site of cardiac arrest ln = 4556) Inside During transport (3) 74 67 Ward (54) 57 E.R. (23) 72 ICUlCCU (20) Outeide 82 Road side (10) 81 Home (66) 70 Public place (10) 85 Work (2) During transport (6) 70
6 9 8 6 11
5 5 6 3 7 P< 0.0001
7 5 16 6 13
Witnessing outside tn = 2898) 73 Witnessed (65) 86 Non-witnessed (45)
10 8
7 3 P< 0.001
10 3
Witnessing outside, VFhL = 7641 56 Witnessed (67) 71 Non-witnessed (33)
12 9
11 8 P< 0.002
20 12
0 6 5 4
Type of cardiac arrest, outside hr = SO67) 12 60 V.F. (25) 7 86 Asystole (66) 8 87 E.M.D. (10)
11 3 2 P 0.0001 Type of cardiac arrest, outside. duration CA less than 5 min In = 26121 14 7 49 V.F. (28) 7 8 .75 Asystole (61) 5 5 82 E.M.D. (11) P< 0.0001 Type of respiratory arrest, outside In = 90671 4 9 82 Complete RA (83) 9 6 66 Gasping 07) P< 0.0001 Type of re@mtory arrest, outside, duration CA less than 5 min ln = 2613) 6 10 75 Complete RA (81) 11 4 60 Gasping (19) P< 0.0001
17 4 3
30 10 8
5 20
9 23
S18
CA-CPR less than 5 min are considered the proportion is 28%, 61% and 1lOrb. Table IV summarises the influence of the type of CA on outcome both in all outof-hospital patients and CA cases of less than 5 min duration only, the influence is highly significant. Initial respiratory pattern Inside the hospital the presence of gasping is about as frequent (46%) as complete respiratory arrest (54%). However, out-of-hospital gasping is infrequent (170/b).Considering only cases with CA-CPR less than 5 min, the proportion gasping vs. complete respiratory arrest is 19% vs. 81%. Table IV summarises the influence of the type of RA on outcome, both for all out-of-hospital CA cases, for CA of less than 5 min duration; the influence is highly significant. DISCUSSION
The overall results, as given in Table I, confirm the high mortality of the CACPR syndrome, certainly when considering the long-term outcome of all CA cases, as has been reported by others [l- 6,11- 131. As has been reported before, advanced age is not a predictor of bad outcome in CA [14 - 181.Although this is certainly influenced by the disease underlying the CA in the different age groups, the non-influence of age remains when only cases of the most frequent type of underlying disease, cardiac origin, are considered. Outcome is significantly influenced by pre-arrest health status. Severe disability pre-arrest diminishes chances on long-term survival (Class 31 to half. The overall health status of successful outcome patients is also clearly less in patients with severe disability pre-arrest. From an additional analysis on a subgroup of our own patients and from other reports [19-2241the mortality of patients, surviving CA with severe disability, is about 50% in the first 6 months after the arrest. Cardiac origin is the most frequent cause of CA as stated by others [15,21,25,26]. There is an impressive difference in outcome between the cardiac origin patients and the other groups. Some groups (intoxication and metabolic origin1 have a better outcome as reported by others [19]. Other groups (trauma, exsanguination, drowning, SIDS, sepsis) have very bad outcome. Some of the differences in outcome between the underlying disease groups are related to obvious explanations i.e. drowning has a remarkably long CA-CPR time compared to the cardiac origin group (17.0 vs. 6.1 mini, trauma and exsanguination have a high percentage of inefficient CPR as judged clinically from the ineffective arterial pulse during ECC. CA of respiratory origin has about a similar successful outcome (Class 31as CA of cardiac origin, contrary to some other reports [28]. However between the cardiac and respiratory group there is a clear difference in the percentage of cerebral failure (Class 21patients. Animal experiments have outlined the differences in pathophysiology and treatment of post-ischemic encephalopathy after asphyxial and ventricular fibrillation CA [9]. Parameters, such as CA-CPR time, quality of ECC during CPR by MICU, pupillary changes during and 10 min post
s19
CPR, were not significantly different from the cardiac and respiratory origin cases and even when considering the subgroup of patients with similar neurological signs 10 min post CPR, the outcome differences in cerebral failure remain. Therefore, whatever the mechanism, post-ischemic encephalopathy in respiratory origin cases is indeed clearly different from cardiac origin cases. The reported percentages of the sites where CA occurs, are somewhat influenced by the national and local emergency medical system, and certainly the in-hospital data should be interpreted accordingly. However, the number clearly illustrates that the preferential place of the sudden death syndrome is the home of the patient, theoretically amenable to CPR [27] but out of reach for easy and immediate intervention of public health services as the Belgian EMS lOO-system. CPR remains therefore a difficult organisational problem. Differences in outcome, according to the site, have been reported before. Probably, the difference between in- and out-of-hospital CA [28,30-321 is related to the clearly shorter CA-CPR time in the hospital. The ER, with 24ch of success, proves its usefulness for immediate advanced life support of critically ill patients. The problem is, however, far from being optimally treated at home (5% success). Contrary to the home, public places seem to have about similar CPR results as in-hospital events. Several reports claim the major importance of whether the arrest was witnessed or not [3 - 6,30,33]. In our overall out-of-hospital group and in the more selective VF group, we can confirm this reported significant difference. The occurrence of VF vs. other rythms in our patients is lower than in other reports [34,35], clearly related to the relatively long CA-CPR time in our patients and the inclusion of CA of all origin. Numerous reports have demonstrated the better prognosis of VF vs. asystole [1,5,30,36-391. However prognosis of asystole is not necessarily fatal [S], as reported by other European investigators [40] and once the circulation is restarted, the long-term survival is equivalent in both groups according to some reports [41], although this is not confirmed by our data. The outcome of the group of patients with VF, out-of-hospital, witnessed arrests, is comparable to other reports and has about a 20% long-term survival. This is the largest group of patients with more or less uniform pathophysiology and seems to us the most indicated group of patients for multiinstitutional clinical trials. The type of respiratory arrest in the CA-CPR syndrome, has not been reported so frequently, although this parameter is also significantly related to outcome. Similar to the distribution of the type of CA, the type of RA is somewhat related to the duration of CA to CPR, more than as independant markers of the pre-CPR setting. As such gasping is infrequently encountered in the prehospital setting. Again, similar to the asystole, complete RA has a bad prognosis, although clearly not uniformly fatal. CONCLUSION Except for age, all studied pre-CPR parameters influence outcome significantly. The bad outcome groups, however, still have success rates of about So,&and, therefore in our opinion, contrary to others [36], an initial CPR
520
effort is always worthwhile. Pre-arrest health status is of importance to a lesser degree. Taking all the above variables into account, our results are comparable with published results of other centers. Although therapeutic clinical trials claim to ‘iron-out’ the influence of these parameters by randomisation, it still appears to us that the study groups should be stratified according to the above pre-CPR conditions. In order to plan studies on the therapeutic effect of drugs on the outcome of CPR, it may be mandatory to prevent dilution of possible beneficial effects with cases of an almost hopeless outcome anyway. In our setting we propose the following selected study group: out-of-hospital, ventricular fibrillation, witnessed cardiac arrest patients. REFERENCES 1
M. Eliastam, T. Duralde, F. Martinez and D. Schwartz, Cardiac arrest in the emergency medical service system guidelines for resuscitation, J. Am. Coll. Emerg. Phys., 6 (197’7)525529. JR. Jude, N.B. Kouwenhove and G.G. Knickenbocher, Report of application of external cardiac massage on 118 patients, J. Am. Med. Assoc., 178 (196111063- 1070. R.O. Cummins and M.S. Eisenberg, Prehospital cardiopulmonary resuscitation. Is it effective? J. Am. Med. Assoc., 253 (198512408-2412. C.A. Polnitsky, R.J. Capone, D.E. Gagnon and K.K. McGinnis. Prehospital coronary care. Proposal for a uniform reporting system, J. Am. Med. Assoc., 237 (19771134- 137. M.S. Eisenberg, L. Bergner and T. Hearne, Out-of-hospital cardiac arrest: A review of major studies and a proposed uniform reporting system, Am. J. Pharmacol., 70 (1980)236-240. B.M. Thompson, H.A. Steuven, J.R. Mateer, C.C. Aprahamian, J.F. Tucker and J.C. Darin, Comparison of clinical CPR studies in Milwaukee and elsewhere in the United States, Ann. Emerg. Med, 14 (19851750-754. 7 A. Mullie, Studygroup ‘Cerebral Resuscitation’, Belgian society of intensive care (SIZ). Clinical study of cerebral and cardiopulmonary resuscitation in Belgian mobile intensive care units, Rianimazione,2 (19851279- 310. 8 The studygroup ‘Cerebral Resuscitation’of the Belgian Society of Intensive Care Medecine (SIZl, A multicenter registration project of cardiopulmonary resuscitation in Belgium, Acta Anesthesiol. Belg.. 35 (Suppl.l(19841129- 135. 9 H. Delooz and A. Mullie, Cardiopulmonary and cerebral resuscitation registry by some Belgian Mobile Intensive Care Units (MICU),Les Comas, (1986171-74. 10 B. Jennett and M. Bond, Assessment of outcome after severe brain damage a practical scale, Lancet, March 1, (19751480- 484. 11 B. Messert and C.E. Quaglieni, Cardiopulmonary Resuscitation Lance& ii (1976)410 - 412. 12 M. Wernberg and A. Thomassen, Prognosis after cardiac arrest occuring outside intensive care and coronary units, Acta Anesth. Stand, 23 (1979169-77. 13 R.J. Meyerburg, K.M. Kessler, L. Zaman, C.A. Conde and A. Castellanos, Survivors of prehospital cardiac arrest, J. Am. Med. Assoc., 247 (198211485- 1490. 14 G. Rochswald, P. Sharn and E. Krir, Follow-up of 514 consecutive patients with cardiopulmonary arrest outside the hospital. J. Am. Coll. Emerg. Phys., 8 (1979)216-220. 15 K. Suljaga-Veehtel, E. Goldburg, P. Strychon. M. Berger and M.L. Scovron, Cardiopulmonary resuscitation in a hospitalised population: prospective study of factors associated with outcome, Resuscitation, 12 (1984177- 95. 16 A.G. Gulati, G.L. Bhan and M.A. Horan, Cardiopulmonary resuscitation of old people, Lancet, July 30 (19831267- 269. 17 J.E. Clinton, J. McGill, G. Irwin, G. Paterson, G.P. Libgon and E. Ruise. Cardiac arrest under age 40: etiology and prognosis, Ann. Emerg. Med., 13 (198411011- 1015.
521 18 19 20
21
22 23 24 25 26
27 28 29
30 31 32 33 34 35 36 37
38
39
40
41
N. Eisenberg, L. Bergner and A. Hallstrom. Epidemiology of cardiac arrest and resuscitation in children, Ann. Emerg. Med., 12 (19831672- 674. R.C. Peatfield, D. Taylor, R.W. Sillett and M.W. McNicoll, Survival after cardiac arrest in hospital, Lancet, June 11(197711223- 1225. M.P. Earnest, J.C. Brickenridge, P.R. Yarnell and P.B. Oliva, Quality of survival after out-ofhospital cardiac arrest predictive value of early neurologic evaluation, Neurology, 29 (1979)56 -60. N. Wernberg and A. Thomassen, Long- and short-term mortality rates in patients who primarily survive cardiac arrest compared with a normal population, Acta Anesth. Stand., 23 (19791211-216. N. Szczygiel, R. Wright, E. Wagner and M.S. Holcomb, Prognostic indicators of ultimate longterm survival following advanced life support, Ann. Emerg. Med., 10 (1981)566-570. M.S. Eisenberg, A. Hallstrom and L. Bergner, Longterm survival after out-of-hospital cardiac arrest. N. Engl. J. Med., 306 (198211340- 1343. M.S. Eisenberg. L. Bergner and A. Hallstrom, Survivors of out-of-hospital cardiac arrest: morbidity and longterm survival, Am. J. Emerg. Med., 2 (19841189- 192. M.S. Eisenberg and L. Bergner and A. Hallstrom. Epidemiology of cardiac arrest and resuscitation in a suburban community, J. Am. Coll. Emerg. Phys., 8 (197912- 6. H.S. Rosman, S. Goldstein, J.R. Landis, R. Leighton, G. Ritter, C.M. Vasu. R.A. Wolfe, A. Acheson and S. Vandenburg Medendrop, Clinical characteristics and survival of out-of-hospital cardiac arrest victims without coronary heart disease, Eur. Heart J., 9 (1988117- 23. W.B. Kouwenhoven, J.R. Jude and G.G. Knickerbocher, Closed chest cardiac massage, J. Am. Natl. Acad., 173 (196011063- 1067. S.E. Bedell, D.E. Delbanco, E.F. Cook and F.H. Epstein, Survival after cardiopulmonary resuscitation in the hospital, N. Engl. J. Med., 209 (19831569-608. P. Vaagenes, R. Cantadore, P. Safar et al., Amelioration of brain damage by lidollazine after prolonged ventricular fibrillation in cardiac arrests in dogs, Crit. Care Med., 12 (1984) 846855. R. Roth, R.D. Stewart, K. Rogers and G.M. Cannon, Out-of-hospital cardiac arrest factors associated with survival, A Ann. Emerg. Med., 13 (19841237-243. R.R. Liberthson, E.L. Nagel, J.C. Hirschman and S.R. Nussenfeld, Prehospital ventricular fibrillation prognosis and follow-up course, N. Engl. J. Med., 291(19741317-321. M. Wernburg and A. Thomassen, Prognosis after cardiac arrest outside intensive care and coronary unit, Acta Anesth. Stand., 23 (1979169- 77. W.D. Weaver, L.A. Cobb, A.P. Hallstrom, C. Fahrenbuch, M.K. Copass and R. Ray, Factors influencing survival after out-of-hospitalcardiac arrest, J. Am. Coll. Cardiol., 7 (19861752-757. F.C. Kempf and M.E. Josephson, Cardiac arrest recorded on ambulatory electrocardiogram, Am. J. Cardiol., 53 (198411577-82. W.T. Longstreth, T.S. Inui, L.A. Cobb and N.K. Copass, Neurologic recovery after out of hospital cardiac arrest, Ann. Int. Med., 98 (1983)588- 592. G.R. Sowden, D.W. Robbins and P.J.F. Basket& Factors associated with survival and eventual cerebral status following cardiac arrest, Anesthesia, 39 (1984129- 43. R.J. Myerburgh, C.A. Conde, R.L. Sung, A. Mayora-Contes, G.M. Miller, D.S. Sheps, R.A. Appel and A. Castellanos, Clinical, electrophysical and hemodynamic profile of patients, Am. J. Med., 68 (19801569-662. M.S. Eisenberg, L. Bergner and A. Hallstrom, Paramedic programs and out of hospital cardiac arrest: I. Factors associated with successful resuscitation, Am. J. Publ. Health, 69 (197913034. B.D. Snyder, M. Ramirez-Lassepas and D.N. Lippert, Neurologic status and prognosis after cardiopulmonary arrest. I. A retrospective study, Neurology, 27 (19731807-811. E. Edgren, U. Hedstrand, N.S. Abramson et al., Relationship of presenting ECG rhythm to subsequent mortality and neuroiogic recovery in comatose cardiac arrest survivors, Crit. Care Med., 14 (19861349. A. Thomassen and M. Wernberg, Prevalence and prognostic significance of coma after cardiac arrest outside intensive care and coronary units, Acta. Anesth. &and., 23 (19791143- 148.
Sll
PRE-CPR CONDITIONS AND THE FINAL OUTCOME OF-CPR
A. MULLIE’, P. LEWP, R. VAN HOEYWEGHEN’ and THE CEREBRAL RESUSCITATION STUDY GROUP “Department of Critic01 Care Medicine, Algemeen Ziekenhuis Sint Jan, Ruddershove 10, 8000 Brugge, bInfonnatkm Science Department, Janeeen Reeearch Fowzdutima, Janeeen Pharmaceutics NV, B-2340, Beerse and ‘Department of Intensive Care and Emergency Medicine, University Antwerp - U.I.A., Universiteitsplein 1, 2610Antwerp lBelgium)
SUMMARY
Outcome of cardiac arrest (CA) is very much influenced by pre-CPR conditions. To assess the importance of these pre-CPR factors, an analysis of the Belgian CPCR registry was made according to some pre-CPR conditions. In this registry, several variables related to pre-arrest, arrest, CPR and post CPR period have been recorded in 4548 patients. The pre-CPR conditions studied were: age, witnessed event or not, pre-arrest health state, underlying disease, site of cardiac arrest, type of respiratory arrest and type of cardiac arrest. Age did not influence outcome significantly. The importance of witnessing is very significant. Severe pre-arrest disability reduces chances on long-term survival (LTSl to half and overall health status longterm survivors is clearly less. Intoxication and metabolic origin of CA have good prognosis (LTS, 21%). Trauma/exsanguination, drowning, SIDS and sepsis have bad prognosis (LTS, 1 -3%). Cardiac (LTS , 12%) and respiratory (LTS, 14*,411 origin have similar outcome, although significant difference exists in occurence of cerebral failure, suggesting that post-ischemic encephalopathy is more severe in respiratory than in cardiac origin. The most frequent site of CA, the home of the patient. has poor outcome results (LTS, 5%). Gasping is significantly related to successful outcome. In the out-of-hospital setting the type of CA was 25% VF (LTS, 77%1,65% asystole (LTS, 4%) and 10% EMD (LTS, 3%). Outcome of the subgroup out-of-hospital, witnessed, VF is comparable to other reports. This sub-group seems to us the most appropriate for clinical trials.
Key words: Cardiopulmonary resuscitation - Pre-CPR conditions - Outcome
0300-9672/89/$03.50 0 1989 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
s12 INTRODUCTION
Reports on the survival of cardiac arrest (CA) vary widely from less than 5% [l] to 600/b[2] It is generally accepted that CPR results are more influenced by the pre-CPR conditions than by the CPR treatment itself. The CA-CPR syndrome is a complex interplay of what has been called ‘fate factors’ (age, underlying disease...) and ‘program factors’ (response times of basic and advanced life support..) [3]. The results of CPR outcome studies are therefore difficult to compare. Several authors have noticed the lack of clarity and precision in language used to report CPR study results and patient selection criteria; they made a plea for a uniform reporting system of outcome (shortterm as well as long-term survival), in order to improve comparability and accurate measurement of the impact of CPR treatment and EMS organisation [3-61. The objective of this report is to outline the influence of the important pre-CPR factors. This should allow a reasonable comparison between outcome studies in different centers, as well as an outline of an optimal patient category for multi-institutional therapeutic CPCR trials.
METHOD
The registration method has been described before [7-g]. In summary, several variables related to the pre-arrest, arrest, CPR and post-CPR period have been registered prospectively on a uniform simple data collection form, on all CA cases from 7 major hospitals attended by their prehospital EMS-MICU services in Belgium. Post- and pre-arrest health status was classified with the simple Jennett-Bond scale [lo]. The outcome has been stratified in 4 classes: Class 1, immediate CPR failure, spontaneous circulation cannot be restarted; Class 2a, CPCR failure due to cerebral failure patients whose circulation can be restored but die within 14 days from cerebral failure or remain in a vegetative state at 14 days post-CPR; Class 2b, CPCR failure due to cardiac, septic or other failure patients whose circulation can be restarted but die within 14 days from non-cerebral failure; Class 3, CPCR successes, patients awake 14 days postCPR either normal (good recovery), moderately disabled (disabled but independent) or severely disabled (dependent on others). The studied pre-CPR variables have been restricted to the following: age, place of CA, pre-arrest health state, underlying disease preceding CA, whether the CA was witnessed or not and, finally, the type of CA and respiratory arrest (RA) on arrival of the MICU team. Other variables, e.g. the hour of the day, have been registered and studied but were without major influence on outcome, contrary to some previous reports [ll. Another very important variable, bystander CPR will be reported extensively elsewhere in this journal. Diseases, underlying the CA were grouped as follows: cardiac, trauma and
s13
exsanguination, respiratory, cerebral, intoxication, metabolic, drowning, septic, sudden infant death syndrome @IDS). Among other variables the pupillary signs and quality of CPR performance was scored. The pupillary signs pre, during and post CPR were grouped in 2 types: bad quality when pupils were dilated, pinpoint or unequal without light reaction and good quality when normal or unequal pupils had slow or normal light reaction. The pupillary signs were evaluated by the MICU team, on arrival at the scene. The quality of CPR by the different performers was categorised as efficient (correct technique with good effect) or inefficient (bad technique or correct technique with weak effect1 according to the chest movements by insufflation and the quality of the palpated pulse under external cardiac compressions (ECC) on arrival of the MICU team. Arrests were considered to be witnessed according to the time of CA to call, a CA-CALL time (1 min, was used as the criterium that someone was present during the event. This interpretative term of witnessed or not was tested during 1986- 1987 with more explicit interrogation of bystanders and found to be correct in over 90% of the cases. CA-CPR is the time interval between CA and the start of basic life support either by a bystander, EMT, nurse or medical MICU team. Qualitative data have been analysed for significance by means of the x2 test; quantitative data have been analysed by means of the Student’st-test. Significance is assumed at the 0.05 level of probability. Average values are reported as mean + S.D. RESULTS
From 1983 to 1987, 4548 patients with CA have been treated by the study centers. Overall results are summarised in Table I. According to the outcome criteria used, CPCR is successful in 18% of the in-hospital CA’s and in 7% of outhospital CA’s
TABLE 1 CPCR REGISTRY - BELGIUM (1983- 1987) - OVERALL RESULTS Comparison in outcome between in-hospital and out-of-hospital cardiac arrest patients. The numbers in between brackets are percentages.
In out In x Out
Class 1 (CPR failure)
Class 2a (Cerebr. failure)
Class 2b (Other failure)
Class 3 Total (CPCR success)
959 (65) 2445 (79) 3404f75)
8163) 267 (9) 348 (7)
155 (11) 153 (5) 308 (7)
270 (18) 218 (7) 488 (11)
1465 3083 4548
P< 0,OOOl
s14
Influence of age Mean age was not systematically related to outcome. Classes 1,2a, 2b and 3 were, respectively, 61 f 19,58 -c 21,67 f 15,63 f 14 years old. Considering only CA of cardiac origin, mean age for classes 1,2a, 2b and 3 was 67 + 13.65 f 14,70 f 11,65 f 13years. Pre-arres t health status In 1433 in-hospital CAs, pre-arrest health status was recorded as normal moderately disabled or severely disabled in, respectively, l5%, 51% and 34% of the patients. In 3066 out-of-hospital CAs, the pre-arrest status was normal in 57%. moderately disabled in 36% and severely disabled in 7%. Both an influence on overall outcome and an influence on the degree of disability in survivors, can be illustrated. The infuence of pre-arrest health status on outcome in the 1433 in-hospital patients (with a significant occurrence of severe disability pre-arrest) is summarised in Table IIa; the influence of the pre-arrest health status on the degree of disability in the 485 surviving patients (in- and out-hospital) is summarised in Table IIb.
TABLE II PRE-ARREST HEALTH STATUS - OUTCOME /al Iwhoupital In = 14.991.Comparison in outcome between I433 in-hospital CA patients in normal health status, moderate disability and severe disability pre-arrest. The numbers in between bracketa are the percentages of patients in that particular health status pre-arrest. The numbers in the table are the percentages, ending in each outcome category, of the total number (166%) in that particular health status. Class 1
Class 2a (Cerebr. failure)
Class 2b (Other failure)
Class 3 (CPRC success)
62 62 73
6 5 6
8 12 10
24 22 11 P< 0.001
(CPR failure) Normal (151 Mod. disabled (511 Sev. disabled (341
(bl Class 3 outcome patients only h = 4&V. Comparison in percentage of severe disability postarrest in 485 surviving CA patients, in either normal health status, moderate disability and severe disability pre-arrest. The numbers in between brackets are the percentages of the patients in that particular health status pre-arrest. The numbers in the table are the percentages of the patients, in that particular health status pre-arrest, ending up with severe disability post-arrest. Severely disabled post-arrest (%I Normal (361 Mod. disabled (50) Sev. disabled (14)
16 26 92 P< 0.01
s15
Underlying disease The disease preceding the CA was cardiac in 76%. trauma and/or exsanguination in 9%, respiratory in 60/b, cerebral in 2% intoxication in 2% metabolic in 2W, drowning in l%, sepsis in 1% and SIDS in 1%. The influence of the underlying disease on outcome is tabulated in Table IIIa, and is highly significant. Although CPCR success is as frequent in respiratory as in cardiac origin patients, cerebral failure is more than twice as frequent in the respiratory origin group. Focussing on differences between cardiac and respiratory origin patients, the following relatively minor differences could be found: ventricular fibrillation was present in 33% in CA of cardiac origin, 8%~in respiratory origin; CPR was performed by bystanders in 38O/bin CA of cardiac origin, 25% respiratory origin; the effect of ECC, judged by MICU as good quality, was 85% in cardiac origin, 84O/brespiratory origin; good pupillary signs during CPR was present in 25O/bin cardiac origin, in 18% in respiratory origin; good pupillary signs 10 min post CPR were present in 62% in cardiac origin and in 56% in respiratory origin. Table IIIb summarises the significant proportional difference in outcome between patients of cardiac and respiratory origin, even with similar neurological status 10 min post initially successful CPR (as judged by the reaction of the pupils). In the 24-h period post CPR, blood gas abnormalities (7.3 < pH < 7.6; Pao, < 50 mmHg; 25 mmHg < PCO, < 50 mmHg1 were more frequently present in respiratory cases (42%) than in cardiac cases (30%) and hyperthermia (t/O > 38.5OCl occurs in 18O/b of respiratory cases vs. 9% in cardiac origin. Place of CA The MICU teams of the study groups attended both in- and out-of-hospital CA situations. Although there are some differences in the in-hospital emergency call system between different centers, overall occurrences of CA in-hospital vs. outside was about 1 to 2. In-hospital, CA occurs in 54%~on the ward, 23Ok,in the emergency room (ER), 200/o in the ICUlCCU and 3?b during transport. Out-of-hospital CA occurred in 16% on the roadside, 660/b at home, 100~ at public places, 2% at work and 6% during transport. Table IV summarises the influence of the place of CA on outcome, the influence is highly significant. Witnessed vs non-witnessed In-hospital CA is witnessed in 81% of the cases. Out-of-hospital CA is witnessed in 53OEo of the cases. Table IV summarises the influence on outcome both in all out-of-hospital patients and in out-of-hospital ventricular fibrillation patients only, the influence is highly significant. Initial ECG rhythms In the in-hospital setting VF, asystole and electromechanical dissociation (EMDl are present, respectively, in 3OW, 49% and 21°h. Outside the hospital VF, asystole and EMD are present in 25%, 65% and 10%. When only cases of
Cardaic (761 Trauma Exanguination(91 Respiratory (56) Cerebral (21 Intoxication (21 Metabolic (21 Drowing (1) Sepsis (11 SIDS (11 83 94
31 7 6
18 26 9
65 63 59 50
7 3
74 95
Class 2a (Cerebr. failure)
11 10 0
4 1 12
7 1
Class 2b (other failure)
8 0 0
13 10 20
12 1
Class 3 (CPCR success)
86 90 80
a4 85 89
8.5 a.3 8.0 17.0 3.3 42.3 P< 0.0001
85 58
ECC-MICU good quality 6.0 6.7
CA-CPR time
1. 2.
Cardiac (621 Respiratory (661
16 36
Class 2a (Cerebral fail)
26 10
Class 2b (other failure)
58 54 P< 0.001
Class 3 (CPCR success)
(b) Difference card& vs. respiratory - outcome - pup&by sigus good 10 min post CPR In = 575). Comparison in outcome between patients with similar neurological status 10 min post-arrest in 528 CA patients, cardiac origin vs. 47 CA patients, respiratory origin. The number in between brackets is the percentage of the patients, in that particular CA origin group, with good pupillary signs 10 min post-arrest. The numbers in the table are the percentages, ending in each outcome class, of the total number (100%) of initially successful CPR patients, with good pupillary signs 10 min post CPR, in that particular CA origin group.
6. 7. 8.
3. 4. 5.
1. 2.
Class 1 (CPR failure)
laJAll patients /n = 45021.Comparison in outcome between the different underlying diseases, preceeding the CA. The number in between brackets is the percentage, of the total number of patients, with that particular underlying disease. The numbers in the table are the percentages, ending in each outcome category, of the total number (100%) in that particular underlying disease. The added 5th and 6th column are the duration of CA (in min) and the percentage of patients resuscitated with good palpable pulses during ECC, in each underlying disease.
UNDERLYING DISEASE - OUTCOME
TABLE III
s17 TABLE IV OTHER PRE-CPR VARIABLES - OUTCOME Comparison in outcome between witnessed and non-witnessed CA patients. The number between brackets is the percentage of patients in that particular category; for the site of the arrest, relating to a total 1OOI in hospital and 100% out-of-hospital. The numbers in the table are the percentages, ending in each outcome class, of the total number of patients @O%) in that particular category. Class 1 (CPR failure)
Class 2a (Cereb. failure)
Class 2b (Other failure)
Class 3 (CPCR success)
10 8 13 13
16 18 25 11
Site of cardiac arrest ln = 4556) Inside During transport (3) 74 67 Ward (54) 57 E.R. (23) 72 ICUlCCU (20) Outeide 82 Road side (10) 81 Home (66) 70 Public place (10) 85 Work (2) During transport (6) 70
6 9 8 6 11
5 5 6 3 7 P< 0.0001
7 5 16 6 13
Witnessing outside tn = 2898) 73 Witnessed (65) 86 Non-witnessed (45)
10 8
7 3 P< 0.001
10 3
Witnessing outside, VFhL = 7641 56 Witnessed (67) 71 Non-witnessed (33)
12 9
11 8 P< 0.002
20 12
0 6 5 4
Type of cardiac arrest, outside hr = SO67) 12 60 V.F. (25) 7 86 Asystole (66) 8 87 E.M.D. (10)
11 3 2 P 0.0001 Type of cardiac arrest, outside. duration CA less than 5 min In = 26121 14 7 49 V.F. (28) 7 8 .75 Asystole (61) 5 5 82 E.M.D. (11) P< 0.0001 Type of respiratory arrest, outside In = 90671 4 9 82 Complete RA (83) 9 6 66 Gasping 07) P< 0.0001 Type of re@mtory arrest, outside, duration CA less than 5 min ln = 2613) 6 10 75 Complete RA (81) 11 4 60 Gasping (19) P< 0.0001
17 4 3
30 10 8
5 20
9 23
S18
CA-CPR less than 5 min are considered the proportion is 28%, 61% and 1lOrb. Table IV summarises the influence of the type of CA on outcome both in all outof-hospital patients and CA cases of less than 5 min duration only, the influence is highly significant. Initial respiratory pattern Inside the hospital the presence of gasping is about as frequent (46%) as complete respiratory arrest (54%). However, out-of-hospital gasping is infrequent (170/b).Considering only cases with CA-CPR less than 5 min, the proportion gasping vs. complete respiratory arrest is 19% vs. 81%. Table IV summarises the influence of the type of RA on outcome, both for all out-of-hospital CA cases, for CA of less than 5 min duration; the influence is highly significant. DISCUSSION
The overall results, as given in Table I, confirm the high mortality of the CACPR syndrome, certainly when considering the long-term outcome of all CA cases, as has been reported by others [l- 6,11- 131. As has been reported before, advanced age is not a predictor of bad outcome in CA [14 - 181.Although this is certainly influenced by the disease underlying the CA in the different age groups, the non-influence of age remains when only cases of the most frequent type of underlying disease, cardiac origin, are considered. Outcome is significantly influenced by pre-arrest health status. Severe disability pre-arrest diminishes chances on long-term survival (Class 31 to half. The overall health status of successful outcome patients is also clearly less in patients with severe disability pre-arrest. From an additional analysis on a subgroup of our own patients and from other reports [19-2241the mortality of patients, surviving CA with severe disability, is about 50% in the first 6 months after the arrest. Cardiac origin is the most frequent cause of CA as stated by others [15,21,25,26]. There is an impressive difference in outcome between the cardiac origin patients and the other groups. Some groups (intoxication and metabolic origin1 have a better outcome as reported by others [19]. Other groups (trauma, exsanguination, drowning, SIDS, sepsis) have very bad outcome. Some of the differences in outcome between the underlying disease groups are related to obvious explanations i.e. drowning has a remarkably long CA-CPR time compared to the cardiac origin group (17.0 vs. 6.1 mini, trauma and exsanguination have a high percentage of inefficient CPR as judged clinically from the ineffective arterial pulse during ECC. CA of respiratory origin has about a similar successful outcome (Class 31as CA of cardiac origin, contrary to some other reports [28]. However between the cardiac and respiratory group there is a clear difference in the percentage of cerebral failure (Class 21patients. Animal experiments have outlined the differences in pathophysiology and treatment of post-ischemic encephalopathy after asphyxial and ventricular fibrillation CA [9]. Parameters, such as CA-CPR time, quality of ECC during CPR by MICU, pupillary changes during and 10 min post
s19
CPR, were not significantly different from the cardiac and respiratory origin cases and even when considering the subgroup of patients with similar neurological signs 10 min post CPR, the outcome differences in cerebral failure remain. Therefore, whatever the mechanism, post-ischemic encephalopathy in respiratory origin cases is indeed clearly different from cardiac origin cases. The reported percentages of the sites where CA occurs, are somewhat influenced by the national and local emergency medical system, and certainly the in-hospital data should be interpreted accordingly. However, the number clearly illustrates that the preferential place of the sudden death syndrome is the home of the patient, theoretically amenable to CPR [27] but out of reach for easy and immediate intervention of public health services as the Belgian EMS lOO-system. CPR remains therefore a difficult organisational problem. Differences in outcome, according to the site, have been reported before. Probably, the difference between in- and out-of-hospital CA [28,30-321 is related to the clearly shorter CA-CPR time in the hospital. The ER, with 24ch of success, proves its usefulness for immediate advanced life support of critically ill patients. The problem is, however, far from being optimally treated at home (5% success). Contrary to the home, public places seem to have about similar CPR results as in-hospital events. Several reports claim the major importance of whether the arrest was witnessed or not [3 - 6,30,33]. In our overall out-of-hospital group and in the more selective VF group, we can confirm this reported significant difference. The occurrence of VF vs. other rythms in our patients is lower than in other reports [34,35], clearly related to the relatively long CA-CPR time in our patients and the inclusion of CA of all origin. Numerous reports have demonstrated the better prognosis of VF vs. asystole [1,5,30,36-391. However prognosis of asystole is not necessarily fatal [S], as reported by other European investigators [40] and once the circulation is restarted, the long-term survival is equivalent in both groups according to some reports [41], although this is not confirmed by our data. The outcome of the group of patients with VF, out-of-hospital, witnessed arrests, is comparable to other reports and has about a 20% long-term survival. This is the largest group of patients with more or less uniform pathophysiology and seems to us the most indicated group of patients for multiinstitutional clinical trials. The type of respiratory arrest in the CA-CPR syndrome, has not been reported so frequently, although this parameter is also significantly related to outcome. Similar to the distribution of the type of CA, the type of RA is somewhat related to the duration of CA to CPR, more than as independant markers of the pre-CPR setting. As such gasping is infrequently encountered in the prehospital setting. Again, similar to the asystole, complete RA has a bad prognosis, although clearly not uniformly fatal. CONCLUSION Except for age, all studied pre-CPR parameters influence outcome significantly. The bad outcome groups, however, still have success rates of about So,&and, therefore in our opinion, contrary to others [36], an initial CPR
520
effort is always worthwhile. Pre-arrest health status is of importance to a lesser degree. Taking all the above variables into account, our results are comparable with published results of other centers. Although therapeutic clinical trials claim to ‘iron-out’ the influence of these parameters by randomisation, it still appears to us that the study groups should be stratified according to the above pre-CPR conditions. In order to plan studies on the therapeutic effect of drugs on the outcome of CPR, it may be mandatory to prevent dilution of possible beneficial effects with cases of an almost hopeless outcome anyway. In our setting we propose the following selected study group: out-of-hospital, ventricular fibrillation, witnessed cardiac arrest patients. REFERENCES 1
M. Eliastam, T. Duralde, F. Martinez and D. Schwartz, Cardiac arrest in the emergency medical service system guidelines for resuscitation, J. Am. Coll. Emerg. Phys., 6 (197’7)525529. JR. Jude, N.B. Kouwenhove and G.G. Knickenbocher, Report of application of external cardiac massage on 118 patients, J. Am. Med. Assoc., 178 (196111063- 1070. R.O. Cummins and M.S. Eisenberg, Prehospital cardiopulmonary resuscitation. Is it effective? J. Am. Med. Assoc., 253 (198512408-2412. C.A. Polnitsky, R.J. Capone, D.E. Gagnon and K.K. McGinnis. Prehospital coronary care. Proposal for a uniform reporting system, J. Am. Med. Assoc., 237 (19771134- 137. M.S. Eisenberg, L. Bergner and T. Hearne, Out-of-hospital cardiac arrest: A review of major studies and a proposed uniform reporting system, Am. J. Pharmacol., 70 (1980)236-240. B.M. Thompson, H.A. Steuven, J.R. Mateer, C.C. Aprahamian, J.F. Tucker and J.C. Darin, Comparison of clinical CPR studies in Milwaukee and elsewhere in the United States, Ann. Emerg. Med, 14 (19851750-754. 7 A. Mullie, Studygroup ‘Cerebral Resuscitation’, Belgian society of intensive care (SIZ). Clinical study of cerebral and cardiopulmonary resuscitation in Belgian mobile intensive care units, Rianimazione,2 (19851279- 310. 8 The studygroup ‘Cerebral Resuscitation’of the Belgian Society of Intensive Care Medecine (SIZl, A multicenter registration project of cardiopulmonary resuscitation in Belgium, Acta Anesthesiol. Belg.. 35 (Suppl.l(19841129- 135. 9 H. Delooz and A. Mullie, Cardiopulmonary and cerebral resuscitation registry by some Belgian Mobile Intensive Care Units (MICU),Les Comas, (1986171-74. 10 B. Jennett and M. Bond, Assessment of outcome after severe brain damage a practical scale, Lancet, March 1, (19751480- 484. 11 B. Messert and C.E. Quaglieni, Cardiopulmonary Resuscitation Lance& ii (1976)410 - 412. 12 M. Wernberg and A. Thomassen, Prognosis after cardiac arrest occuring outside intensive care and coronary units, Acta Anesth. Stand, 23 (1979169-77. 13 R.J. Meyerburg, K.M. Kessler, L. Zaman, C.A. Conde and A. Castellanos, Survivors of prehospital cardiac arrest, J. Am. Med. Assoc., 247 (198211485- 1490. 14 G. Rochswald, P. Sharn and E. Krir, Follow-up of 514 consecutive patients with cardiopulmonary arrest outside the hospital. J. Am. Coll. Emerg. Phys., 8 (1979)216-220. 15 K. Suljaga-Veehtel, E. Goldburg, P. Strychon. M. Berger and M.L. Scovron, Cardiopulmonary resuscitation in a hospitalised population: prospective study of factors associated with outcome, Resuscitation, 12 (1984177- 95. 16 A.G. Gulati, G.L. Bhan and M.A. Horan, Cardiopulmonary resuscitation of old people, Lancet, July 30 (19831267- 269. 17 J.E. Clinton, J. McGill, G. Irwin, G. Paterson, G.P. Libgon and E. Ruise. Cardiac arrest under age 40: etiology and prognosis, Ann. Emerg. Med., 13 (198411011- 1015.
521 18 19 20
21
22 23 24 25 26
27 28 29
30 31 32 33 34 35 36 37
38
39
40
41
N. Eisenberg, L. Bergner and A. Hallstrom. Epidemiology of cardiac arrest and resuscitation in children, Ann. Emerg. Med., 12 (19831672- 674. R.C. Peatfield, D. Taylor, R.W. Sillett and M.W. McNicoll, Survival after cardiac arrest in hospital, Lancet, June 11(197711223- 1225. M.P. Earnest, J.C. Brickenridge, P.R. Yarnell and P.B. Oliva, Quality of survival after out-ofhospital cardiac arrest predictive value of early neurologic evaluation, Neurology, 29 (1979)56 -60. N. Wernberg and A. Thomassen, Long- and short-term mortality rates in patients who primarily survive cardiac arrest compared with a normal population, Acta Anesth. Stand., 23 (19791211-216. N. Szczygiel, R. Wright, E. Wagner and M.S. Holcomb, Prognostic indicators of ultimate longterm survival following advanced life support, Ann. Emerg. Med., 10 (1981)566-570. M.S. Eisenberg, A. Hallstrom and L. Bergner, Longterm survival after out-of-hospital cardiac arrest. N. Engl. J. Med., 306 (198211340- 1343. M.S. Eisenberg. L. Bergner and A. Hallstrom, Survivors of out-of-hospital cardiac arrest: morbidity and longterm survival, Am. J. Emerg. Med., 2 (19841189- 192. M.S. Eisenberg and L. Bergner and A. Hallstrom. Epidemiology of cardiac arrest and resuscitation in a suburban community, J. Am. Coll. Emerg. Phys., 8 (197912- 6. H.S. Rosman, S. Goldstein, J.R. Landis, R. Leighton, G. Ritter, C.M. Vasu. R.A. Wolfe, A. Acheson and S. Vandenburg Medendrop, Clinical characteristics and survival of out-of-hospital cardiac arrest victims without coronary heart disease, Eur. Heart J., 9 (1988117- 23. W.B. Kouwenhoven, J.R. Jude and G.G. Knickerbocher, Closed chest cardiac massage, J. Am. Natl. Acad., 173 (196011063- 1067. S.E. Bedell, D.E. Delbanco, E.F. Cook and F.H. Epstein, Survival after cardiopulmonary resuscitation in the hospital, N. Engl. J. Med., 209 (19831569-608. P. Vaagenes, R. Cantadore, P. Safar et al., Amelioration of brain damage by lidollazine after prolonged ventricular fibrillation in cardiac arrests in dogs, Crit. Care Med., 12 (1984) 846855. R. Roth, R.D. Stewart, K. Rogers and G.M. Cannon, Out-of-hospital cardiac arrest factors associated with survival, A Ann. Emerg. Med., 13 (19841237-243. R.R. Liberthson, E.L. Nagel, J.C. Hirschman and S.R. Nussenfeld, Prehospital ventricular fibrillation prognosis and follow-up course, N. Engl. J. Med., 291(19741317-321. M. Wernburg and A. Thomassen, Prognosis after cardiac arrest outside intensive care and coronary unit, Acta Anesth. Stand., 23 (1979169- 77. W.D. Weaver, L.A. Cobb, A.P. Hallstrom, C. Fahrenbuch, M.K. Copass and R. Ray, Factors influencing survival after out-of-hospitalcardiac arrest, J. Am. Coll. Cardiol., 7 (19861752-757. F.C. Kempf and M.E. Josephson, Cardiac arrest recorded on ambulatory electrocardiogram, Am. J. Cardiol., 53 (198411577-82. W.T. Longstreth, T.S. Inui, L.A. Cobb and N.K. Copass, Neurologic recovery after out of hospital cardiac arrest, Ann. Int. Med., 98 (1983)588- 592. G.R. Sowden, D.W. Robbins and P.J.F. Basket& Factors associated with survival and eventual cerebral status following cardiac arrest, Anesthesia, 39 (1984129- 43. R.J. Myerburgh, C.A. Conde, R.L. Sung, A. Mayora-Contes, G.M. Miller, D.S. Sheps, R.A. Appel and A. Castellanos, Clinical, electrophysical and hemodynamic profile of patients, Am. J. Med., 68 (19801569-662. M.S. Eisenberg, L. Bergner and A. Hallstrom, Paramedic programs and out of hospital cardiac arrest: I. Factors associated with successful resuscitation, Am. J. Publ. Health, 69 (197913034. B.D. Snyder, M. Ramirez-Lassepas and D.N. Lippert, Neurologic status and prognosis after cardiopulmonary arrest. I. A retrospective study, Neurology, 27 (19731807-811. E. Edgren, U. Hedstrand, N.S. Abramson et al., Relationship of presenting ECG rhythm to subsequent mortality and neuroiogic recovery in comatose cardiac arrest survivors, Crit. Care Med., 14 (19861349. A. Thomassen and M. Wernberg, Prevalence and prognostic significance of coma after cardiac arrest outside intensive care and coronary units, Acta. Anesth. &and., 23 (19791143- 148.