- Email: [email protected]
Bystander CPR in prehospital coarse ventricular fibrillation
ORIGINAL CONTRIBUTION cardiopulmonary resuscitation; ventricular fibrillation
Bystander CPR in Prehospital Coarse Ventricular Fibrillation Prehospital bystander cardiopulmonary resuscitation (CPR) was studied to determine if it affected the outcome of defibrillation. Four hundred twentyone consecutive witnessed cardiopulmonary arrests presenting with the initial rhythm of coarse ventricular fibrillation treated by the Milwaukee County Paramedic System from January 1980 to June 1982 were analyzed. Pediatric, trauma, and poisoning patients and those receiving intravenous or endotracheal medications before defibrillation (58) were excluded. Immediate professional bystander CPR (physician, nurs e, EMT) and citizen bystander CPR were compared to a control group receiving no bystander CPR until arrival of EMS personnel. A successful defibrillation occurred if defibrillation prior to administration of medication produced an effective cardiac rhythm with pulses. Eighty-eight of the 363 remaining patients (24%) converted with initial defibrillations. While the group receiving professional bystander CPR had a higher successful defibrillation rate than did the noCPR group (35% vs 22%, P < .04), citizen bystander CPR and no-CPR groups had similar successful defibrillation rates (24% vs 22%, no significant difference). One hundred eighty-six of the 363 patients (51%) were transported to a hospital with a rhythm and a pulse (a successful resuscitation). Ninety-seven of the 363 patients (27%) were discharged alive from the hospital (a save). Patients who were converted successfully using initial "quick-look" defibrillations were far more likely to be successfully resuscitated (79/88 [90%] vs 107/275 [39%], P > .0001) and to be ddischarged alive from the hospital (54/88 [61%] vs 43/275 [16%], P > .0001) than were those who required further advanced cardiac life support techniques. The overall bystander CPR group had the identical successful resuscitation rate (51%) and a similar save rate (26% vs 27%) as the no bystander CPR group. A patient is far more likely to have a successful resuscitation and be discharged alive from a hospital if he has a rhythm and a pulse after the initial defibrillations than when converted after further ACLS techniques. Although professional CPR was associated with a significantly improved successful defibrillation rate, there was no significant difference in successful defibrillation, resuscitation, or save rates when comparing citizen bystander CPR to no bystander CPR. Previously reported CPR save rates may be system-specific and more dependent on time of defibrillation response than previously reported. [Kowalski R, Thompson BM, Horwitz L, Stueven H, Aprahamian C, Darin JC: Bystander CPR in prehospital coarse ventricular fibrillation. Ann Emerg Med November 1984;13:1016-1020.]
Robert Kowalski, MD Bruce M Thompson, MD Louis Horwitz, MD Harlan Stueven, MD Charles Aprahamian, MD Joseph C Darin, MD Milwaukee, Wisconsin From the Section of Trauma and Emergency Medicine, Medical College of Wisconsin, The Milwaukee County Medical Complex, Milwaukee, Wisconsin. Received for publication June 13, 1983. Revision received January 23, 1984. Accepted for 15ublication May 8, 1984. Presented at the University Association for Emergency Medicine Annual Meeting in Boston, June 1983. Address for reprints: Joseph C Darin, MD, Section of Trauma and Emergency Medicine, Medical College of Wisconsin, 8700 West Wisconsin Avenue, Milwaukee, Wisconsin 53226.
INTRODUCTION External cardiopulmonary resuscitation (CPR) has long been advocated as the means to keep the heart in the state from which it can be resuscitated.l,2 Recent laboratory studies suggest limitations to its effectiveness. 8-s The traditional measure of effective CPR, "good pulses with CPR," has largely been replaced by physiologic measurements of cerebral and coronary blood flow. Ditchey et al 3 showed that external CPR on dogs produced only 1% of norreal coronary artery blood flow. MacKenzie et al 6 suggest that external CPR may be less effective in supplying blood to the heart than any other organ. 4 Rudikoff et alS have shown no significant arteriovenous pressure gradient in the chest during CPR, thus questioning the forward movement of blood flow.
13:11 November 1984
Annals of Emergency Medicine
1016/47
BYSTANDER CPR Kowalski et al
Although laboratory evidence has been equivocal, early clinical studies examining CPR i n Seattle and other paramedic systems have shown the effectiveness of CPR. z-ll These studies have looked at overall survival and have been widely interpreted to show that bystander CPR is the most significant factor in patient survival. Most of the studies, however, include such other variables as speed of advanced cardiac life support (ACLS) response, ACLS techniques and use of coronary care units which may influence the results. We have found no large series that evaluates external CPR alone in a clinical setting without other variables. We attempted to isolate the performance of CPR from other variables by evaluating outcome standards o n those patients resuscitated f r o m coarse ventricular fibrillation (VF) by CPR and countershock alone.
METHODS The Milwaukee County Paramedic System is a multi-tiered advanced life support system based on the Seattle model. Basic emergency medical technicians (EMTs) on fire rescue squads and fire engine companies respond primarily (first responders), followed by ACLS p a r a m e d i c t e a m s trained in rhythm recognition, pharmacology, intubation, intravenous and intracardiac techniques, and defibrillation. On-line medical control is performed by a single base station provided with a twoway voice radio and continuous ECG telemetry. The p a r a m e d i c base is staffed by a small group of ACLS-certiffed physicians on the faculty of the Medical College of Wisconsin. The base p h y s i c i a n s follow A m e r i c a n H e a r t A s s o c i a t i o n g u i d e l i n e s for ACLS. Paramedics, however, may administer up to four countershocks for VF prior to base physician contact or a d m i n i s t r a t i o n of medications. All paramedic run data are entered on standard forms and placed in a computer for future retrieval and analysis. In addition, all original records and rhythm strips are kept on file for verification of data. All patients with witnessed arrest presenting with an initial rhythm of coarse VF were retrieved by a computer printout and analyzed for the following data: age, sex, cardiac h i s t o ~ cardiac drugs, antiarthythmic medications, paramedic response time, citizen bystander CPR, professional bystander CPR, rhythm and pulse after 48/1017
TABLE 1. Successful prehospital defibrillation Yes 88 65 6.0
No 275 65 6.0
Sex (M:F) Cardiac history Cardiac medications history Antiarrhythmics history
72:16 (4:1) 48 (55%) 40 (45%) 13 (15%)
210:65 (3:1) 155 (56%) 112 (41%) 31 (11%)
Professional CPR Citizen CPR
20 (23%)* 21 (24%)
38 (14%)* 65 (24%)
Bystander CPR (pro and citizen) Successful resuscitations Saves *P < .05. tp < .0001. *P < .0001.
41 (47%) 79 (90%)t 54 (61%)1-
103 (38%) 107 (39%)l43 (16%)1-
No. Patients Age (y) Response time (min)
initial defibrillations, successful resuscitation, and save. Average response time for first responders and paramedics also was collected. Patients were defined as having no CPR if no bystander CPR was done prior to arrival of the first responding fire department unit. Patients were defined as professional CPR if their arrests were witnessed by, and CPR was initiated by, a physician, nurse or basic EMT. These professionals were at the scene at the t i m e of arrest. Arrests witnessed by the paramedics were not included. None of these arrests occurred in hospital settings. Patients were defined as citizen CPR if their arrests were witnessed by, and CPR was initiated by, the nonprofessional public. "Successful defibrillation" is defined as "a rhythm with a pulse" after one to four "quick-look" defibrillations given prior to m e d i c a t i o n s . "Successful resuscitation" is defined as the successful transportation of a patient to an emergency deportment with a rhythm and a pulse. A "save" is defined as discharge of the patient alive from the hospital. "Response time" is defined as the difference between the time of the telephone call following arrest to the arrival of the paramedic unit. Data were statistically analyzed by chi square or Student t test, as applicable.
RESULTS From January 1980 to June 1982 the Annals of Emergency Medicine
Milwaukee County Paramedic System attempted resuscitations on 421 patients in witnessed arrests with the initial r h y t h m of coarse ventricular fibrillation. Fifty-eight patients were excluded because of trauma, pediatric age, or administration of intravenous or endotracheal medications prior to initial successful defibrillations. The study group totaled 363 patients, of whom 88 patients had a rhythm and a pulse after initial defibrillations. Two hundred seventy-five were not successfully defibrillated initially (Table 1). The groups were statistically similar in all parameters except presence of professional CPR, rate of successful resuscitations, and saves. Average response time for first responders was 2.1 minutes, and for paramedics it was six minutes. Patients who were successfully converted using initial "quick-look" defibrillations were far more likely to be successfully resuscitated (79/88 [90%] vs 107/275 [39%], P > .0001) and to be discharged alive f r o m the hospital (54/88 [61%1 vs 43/275 [16%], P > .0001) than were those who required further ACLS techniques. The incidence of citizen bystander CPR for both groups was identical (24%). Patients who received bystander (professional and citizen) CPR were compared separately to patients not receiving bystander CPR (Table 2). No statistically significant population differences were found between these 13:11 November1984
TABLE 2. CPR groups
Professional 58 65 44:14 (3:1) 5.2 30 (52%) 22 (38%) 5 (9%) 20 (35%)* 30 (52%) 18 (31%)
No. Patients Age (y) Sex (M:F) Response time (min) Cardiac history Cardiac medications history Antiarrhythmics history Successful defibrillations Successful resuscitations Saves *P < .04.
Bystander (Pro and Citizen 144 65 109:35 (3:1) 5.6 74 (51%) 55 (38%) 18 (13%) 41 (29%) 74 (51%) 38 (26%)
Citizen 86 65 65:21 (3:1) 5.9 44 (51%) 33 (38%) 13 (15%) 21 (24%) 44 (51°/0) 20 (23%)
None 219 66 173:46 (4:1) 6.3 129 (59%) 97 (44%) 26 (12%) 47 (22%)* 112 (51%) 59 (27%)
TABLE 3. Paramedic response times (min)
No. patients Age (y) Sex (M:F) Cardiac history Cardiac medications history Antiarrhythmics history Professional CPR Citizen CPR Bystander CPR (pro and citizen) Successful defibrillations Successful resuscitations Saves *P < .04. l-p < .03.
~<4
>4
~<5
148 66 117:31 (4:1) 89 (60%) 61 (41%) 20 (14%) 31 (21%)* 30 (20%)
215 65 165:50 (3:1) 114 (53%) 91 (42%) 24 (11%) 27 (13%)* 56 (25%)
199 66 160:39 (4:1) 117 (59%) 84 (42%) 26 (13%) 40 (20%)l48 (24%)
61 (41%)
83 (38%)
33 (22%) 75 (51%) 42 (28%)
groups. Patients who were successfully defibrillated were statistically more likely to have received immediate professional CPR (23% vs 14%, P < .05). Of importance is the finding that successful resuscitation and ultimate discharge rates are similar for those patients receiving immediate bystander CPR and those not receiving any bystander CPR. W h e n separated f r o m the overall bystander group, patients receiving professional CPR had a s t a -
13:11November1984
>5
~<6
>6
164 65 122:42 (3:1) 86 (52%) 68 (42%) 18 (11%) 18 (11%)138 (23%)
232 65 183:49 (4:1) 135 (58%) 99 (43%) 28 (12%) 44 (19%) 56 (24%)
131 65 99:32 (3:1) 68 (52%) 53 (41%) 16 (12%) 14 (11%) 30 (23%)
88 (44%)
56 (34%)
100 (43%)
44 (34%)
55 (26%)
46 (23%)
42 (26%)
56 (24%)
32 (24%)
111 (51%) 55 (26%)
104 (52%) 57 (29%)
82 (50%) 40 (24%)
118 (51%) 66 (28%)
68 (52%) 31 (24%)
tistically significant improvement in rate of successful defibrillation when compared to the no bystander CPR group (35% vs 22%, P < .04). While there was a trend in the successful resuscitation and save rate, this difference was not statistically s i g n i f i c a n t . B e c a u s e of n o t e d differences, the study group was analyzed by response times. The results show, as expected, that statistically m o r e people received professional Annals of Emergency Medicine
CPR in the "less than or equal to fourm i n u t e " and "less than or equal to five-minute" groups (Table 3).
DISCUSSION Several s t u d i e s of v a r i o u s prehospital systems with both single and two-tiered responsesl~, t3 have suggested that, for patients in coarse ventricular fibrillation, the more rapidly the patient can be defibrillated, the greater the chance of conversion to a
1018/49
BYSTANDER CPR Kowalski et al
perfusing cardiac r h y t h m . Recent American Heart Association guidelines for A C L S r e c o m m e n d initial "quick-look" defibrillations with oxygenation and basic life support techniques prior to use of other ACLS drug protocols.14 The more rapidly normal perfusion to the brain can be established, the better the neurologic outcome and ultimate survival. In our study, of those patients who were successfully defibrillated with initial defibrillations, 90% were successfully resuscitated and admitted to a hospital. Two-thirds of the patients (61%) were discharged from the hospital. On the other hand those patients requiring further CPR and further ACLS techniques had significantly inferior outcomes. With the addition of potent cardiac drugs used in the advanced cardiac life support protocol, it may be unreasonable and simplistic to assume outcome related to a single variable, such as CPR. Other variables, especially "time to early defibrillation," should be analyzed when comparing prehospital systems. The role of CPR as a component of prehospital systems may be exaggerated. We did find a statistically significant difference in successful defibrillations between patients receiving professiona ! bystander CPR w h e n compared to patients who received no bystander CPR; however, patients receiving professional bystander CPR also had shorter paramedic response times than did those patients receiving no bystander CPR. This variable of time could account for the difference in defibrillation rate. Another variable that has been reported is may provide evidence that professional medical personnel may "do better CPR" than does the lay public. Other studies16,17 have shown that professionals may demonstrate less decrease in CPR skill levels over time. One could infer that experienced EMTs might remain more calm in an e m e r g e n c y and would be more able to provide a more uniform and structured approach to CPR performance than can a lay bystander. We also believe that the shortened response time found in the professional CPR group is directly related to knowledge of system access, which may have allowed shorter paramedic responses. There were, however, no statistically significant differences between the professional bystander CPR and no bystander CPR groups in relation 50/1019
to rate of successful resuscitation or hospital discharge. We can demonstate no difference between the overall bystander CPR group and the no bystander CPR group. The two groups had an identical successful resuscitation rate (51%), and a similar save rate (26% vs 27%). The predominant factor in the ultimate successful resuscitation and save rates .in any system may be the total amount of time from arrest to establ i s h m e n t of an effective r h y t h m . ACLS, and specifically early defibrillation, may be much more important in a prehospital EMS system than previously recognized. Basic cardiac life support is fraught with many problems: an erratic airway, deteriorating acid-base status, and no chance of defibrillation or use of cardiac drugs. With ACLS there exists, in addition to the ability to defibrillate coarse ventricular fibrillation, the ability to control the airway, augment oxygenation, and normalize acid-base status. Studies is in which portable defibrillators were made more available to small communities indicate that rapid defibrillation improves patient outcome. Even in a rapid response paramedic system such as ours, the most significant factor relating to successful resuscitation and hospital discharge is conversion response to initial defibrillations. While continued use of the ACLS protocol for coarse VF is beneficial, the overwhelming majority of successes occur with patients who are easily converted by initial defibrillation. We think that study data comparing prehospital systems should be much more specific in relation to time of paramedic response, time of defibrillation, and length of resuscitation in order to evaluate equitably the effectiveness of various resuscitation modalities, including CPR.
CONCLUSIONS A patient is far more likely to have a successful resuscitation and to be discharged alive from the hospital if he has a rhythm and a pulse after initial defibrillations than when converted after further ACLS techniques. Although professional CPR was associated w i t h s i g n i f i c a n t l y i m p r o v e d successful defibrillation, resuscitation or save rates when comparing citizen bystander CPR to no bystander CPR were similar. Previously reported CPR save rates may be system-specific and not applicable to all EMS systems. Annals of Emergency Medicine
Successful resuscitation and ultimate discharge outcome may be more dependent on time of defibrillation response than previously reported.
The authors thank Lauryl Pukansky and her staff for chart retrieval; Jerry Anderson, PhD (Division of Biostatisties and Clinical Epidemiology, Medical College of Wisconsin) for his assistance in statistical analysis; and Debra Vitek for her tireless preparation of this manuscript.
REFERENCES 1. Kouwenhoven WB, Jude JR, Knickerbocker GG: Closed-chest cardiac massage. l A M A 1960;173:1064-1067. 2. Jude JR, Kouwenhoven WB, Knickerbocker GG: Cardiac arrest: Report of application of external cardiac massage on 118 patients. JAMA 1961;178:1063-1070. 3. Ditchey RV, TWinkler JV, Rhodes CA: Relative lack of coronary blood flow during closed-chest resuscitation in dogs. Circulation 1982;66:297-302. 4. Niemann JT, Rosborough JP, Hausknecht M, et al: Pressure-~ynchronized cineangiography during experimental cardiopulmonary resuscitation. Circulation 1981j64:985-991. 5. Rudikoff MT, Maughan WL, Effron M, et al: Mechanisms of blood flow during cardiopulmonary resuscitation. Circulation 1980;61:345-352. 6. MacKenzie GJ, Taylor SH, McDonald AH, Donald KW: Haemodynamic effects of external cardiac compression. Lancet 1964;1:1342. 7. Thompson RG, Hallstrom AP, Cobb LA: Bystander-initiated cardiopulmonary resuscitation in the management of ventricular fibrillation. A n n I n t e r n M e d 1979;90:737-740. 8. Eisenberg MS, Bergner L, Hallstrom AP: Paramedic programs and out-of-hospital cardiac arrest: I. Factors associated with successful resuscitation. A m J Public Health 1979;69:30-38. 9. Lund I, Skulberg A: Cardiopulmonary resuscitation by lay people. Lancet 1976; 2: 702-705. 10. Tweed WA, Bristow G, Donen N: Resuscitation from cardiac arrest: Assessment of a system providing only basic life support outside of hospital. Can M e d Assoc [ 1980;122:297-300. 11. Guzy PM, Pearce ML, Greenfield S, et al: Effectiveness of citizen cardiopulmonary resuscitation during out-of-hospital emergencies in metropolitan Los Angeles. Circulation 1979;60(Part II):46. 12. Cobb LS, Baum RS, Alvarez H, et al: Resuscitation from out-of-hospital ventricular fibrillation: Four-year follow-up. Circulation 1975;51,52(suppl 3):223-228. 13:11 November 1984
13. Sherman MA: Mobile intensive care units: An evaluation of effectiveness. JAMA 1979;241:1899-1901. 14. Standards and guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). JAMA 1980; 244:483-493.
13:11 November 1984
15. Barick BL: How long do you remember lifesaving CPR skills? J Phys Ed Recreation 1977;48:62-63.
et ah Trainees' retention of cardiopulmonary resuscitation: How quickly they forget. JAMA 1979;24h901-903.
16. Watson LA: Cardiopulmonary resuscitation eompetencies of nurses. Int J Nurs Stud 1982;19:99-107.
18. Eisenberg MS, Copass MK, Hallstrom AP: Treatment of out-of-hospital cardiac arrests with rapid defibrillation by emergency medical technicians. N Engl J Med 1980;302:1379q383.
17. Weaver FJ, Ramirez AG, Dorfman SB,
Annals of Emergency Medicine
1020/51
Bystander CPR in Prehospital Coarse Ventricular Fibrillation Prehospital bystander cardiopulmonary resuscitation (CPR) was studied to determine if it affected the outcome of defibrillation. Four hundred twentyone consecutive witnessed cardiopulmonary arrests presenting with the initial rhythm of coarse ventricular fibrillation treated by the Milwaukee County Paramedic System from January 1980 to June 1982 were analyzed. Pediatric, trauma, and poisoning patients and those receiving intravenous or endotracheal medications before defibrillation (58) were excluded. Immediate professional bystander CPR (physician, nurs e, EMT) and citizen bystander CPR were compared to a control group receiving no bystander CPR until arrival of EMS personnel. A successful defibrillation occurred if defibrillation prior to administration of medication produced an effective cardiac rhythm with pulses. Eighty-eight of the 363 remaining patients (24%) converted with initial defibrillations. While the group receiving professional bystander CPR had a higher successful defibrillation rate than did the noCPR group (35% vs 22%, P < .04), citizen bystander CPR and no-CPR groups had similar successful defibrillation rates (24% vs 22%, no significant difference). One hundred eighty-six of the 363 patients (51%) were transported to a hospital with a rhythm and a pulse (a successful resuscitation). Ninety-seven of the 363 patients (27%) were discharged alive from the hospital (a save). Patients who were converted successfully using initial "quick-look" defibrillations were far more likely to be successfully resuscitated (79/88 [90%] vs 107/275 [39%], P > .0001) and to be ddischarged alive from the hospital (54/88 [61%] vs 43/275 [16%], P > .0001) than were those who required further advanced cardiac life support techniques. The overall bystander CPR group had the identical successful resuscitation rate (51%) and a similar save rate (26% vs 27%) as the no bystander CPR group. A patient is far more likely to have a successful resuscitation and be discharged alive from a hospital if he has a rhythm and a pulse after the initial defibrillations than when converted after further ACLS techniques. Although professional CPR was associated with a significantly improved successful defibrillation rate, there was no significant difference in successful defibrillation, resuscitation, or save rates when comparing citizen bystander CPR to no bystander CPR. Previously reported CPR save rates may be system-specific and more dependent on time of defibrillation response than previously reported. [Kowalski R, Thompson BM, Horwitz L, Stueven H, Aprahamian C, Darin JC: Bystander CPR in prehospital coarse ventricular fibrillation. Ann Emerg Med November 1984;13:1016-1020.]
Robert Kowalski, MD Bruce M Thompson, MD Louis Horwitz, MD Harlan Stueven, MD Charles Aprahamian, MD Joseph C Darin, MD Milwaukee, Wisconsin From the Section of Trauma and Emergency Medicine, Medical College of Wisconsin, The Milwaukee County Medical Complex, Milwaukee, Wisconsin. Received for publication June 13, 1983. Revision received January 23, 1984. Accepted for 15ublication May 8, 1984. Presented at the University Association for Emergency Medicine Annual Meeting in Boston, June 1983. Address for reprints: Joseph C Darin, MD, Section of Trauma and Emergency Medicine, Medical College of Wisconsin, 8700 West Wisconsin Avenue, Milwaukee, Wisconsin 53226.
INTRODUCTION External cardiopulmonary resuscitation (CPR) has long been advocated as the means to keep the heart in the state from which it can be resuscitated.l,2 Recent laboratory studies suggest limitations to its effectiveness. 8-s The traditional measure of effective CPR, "good pulses with CPR," has largely been replaced by physiologic measurements of cerebral and coronary blood flow. Ditchey et al 3 showed that external CPR on dogs produced only 1% of norreal coronary artery blood flow. MacKenzie et al 6 suggest that external CPR may be less effective in supplying blood to the heart than any other organ. 4 Rudikoff et alS have shown no significant arteriovenous pressure gradient in the chest during CPR, thus questioning the forward movement of blood flow.
13:11 November 1984
Annals of Emergency Medicine
1016/47
BYSTANDER CPR Kowalski et al
Although laboratory evidence has been equivocal, early clinical studies examining CPR i n Seattle and other paramedic systems have shown the effectiveness of CPR. z-ll These studies have looked at overall survival and have been widely interpreted to show that bystander CPR is the most significant factor in patient survival. Most of the studies, however, include such other variables as speed of advanced cardiac life support (ACLS) response, ACLS techniques and use of coronary care units which may influence the results. We have found no large series that evaluates external CPR alone in a clinical setting without other variables. We attempted to isolate the performance of CPR from other variables by evaluating outcome standards o n those patients resuscitated f r o m coarse ventricular fibrillation (VF) by CPR and countershock alone.
METHODS The Milwaukee County Paramedic System is a multi-tiered advanced life support system based on the Seattle model. Basic emergency medical technicians (EMTs) on fire rescue squads and fire engine companies respond primarily (first responders), followed by ACLS p a r a m e d i c t e a m s trained in rhythm recognition, pharmacology, intubation, intravenous and intracardiac techniques, and defibrillation. On-line medical control is performed by a single base station provided with a twoway voice radio and continuous ECG telemetry. The p a r a m e d i c base is staffed by a small group of ACLS-certiffed physicians on the faculty of the Medical College of Wisconsin. The base p h y s i c i a n s follow A m e r i c a n H e a r t A s s o c i a t i o n g u i d e l i n e s for ACLS. Paramedics, however, may administer up to four countershocks for VF prior to base physician contact or a d m i n i s t r a t i o n of medications. All paramedic run data are entered on standard forms and placed in a computer for future retrieval and analysis. In addition, all original records and rhythm strips are kept on file for verification of data. All patients with witnessed arrest presenting with an initial rhythm of coarse VF were retrieved by a computer printout and analyzed for the following data: age, sex, cardiac h i s t o ~ cardiac drugs, antiarthythmic medications, paramedic response time, citizen bystander CPR, professional bystander CPR, rhythm and pulse after 48/1017
TABLE 1. Successful prehospital defibrillation Yes 88 65 6.0
No 275 65 6.0
Sex (M:F) Cardiac history Cardiac medications history Antiarrhythmics history
72:16 (4:1) 48 (55%) 40 (45%) 13 (15%)
210:65 (3:1) 155 (56%) 112 (41%) 31 (11%)
Professional CPR Citizen CPR
20 (23%)* 21 (24%)
38 (14%)* 65 (24%)
Bystander CPR (pro and citizen) Successful resuscitations Saves *P < .05. tp < .0001. *P < .0001.
41 (47%) 79 (90%)t 54 (61%)1-
103 (38%) 107 (39%)l43 (16%)1-
No. Patients Age (y) Response time (min)
initial defibrillations, successful resuscitation, and save. Average response time for first responders and paramedics also was collected. Patients were defined as having no CPR if no bystander CPR was done prior to arrival of the first responding fire department unit. Patients were defined as professional CPR if their arrests were witnessed by, and CPR was initiated by, a physician, nurse or basic EMT. These professionals were at the scene at the t i m e of arrest. Arrests witnessed by the paramedics were not included. None of these arrests occurred in hospital settings. Patients were defined as citizen CPR if their arrests were witnessed by, and CPR was initiated by, the nonprofessional public. "Successful defibrillation" is defined as "a rhythm with a pulse" after one to four "quick-look" defibrillations given prior to m e d i c a t i o n s . "Successful resuscitation" is defined as the successful transportation of a patient to an emergency deportment with a rhythm and a pulse. A "save" is defined as discharge of the patient alive from the hospital. "Response time" is defined as the difference between the time of the telephone call following arrest to the arrival of the paramedic unit. Data were statistically analyzed by chi square or Student t test, as applicable.
RESULTS From January 1980 to June 1982 the Annals of Emergency Medicine
Milwaukee County Paramedic System attempted resuscitations on 421 patients in witnessed arrests with the initial r h y t h m of coarse ventricular fibrillation. Fifty-eight patients were excluded because of trauma, pediatric age, or administration of intravenous or endotracheal medications prior to initial successful defibrillations. The study group totaled 363 patients, of whom 88 patients had a rhythm and a pulse after initial defibrillations. Two hundred seventy-five were not successfully defibrillated initially (Table 1). The groups were statistically similar in all parameters except presence of professional CPR, rate of successful resuscitations, and saves. Average response time for first responders was 2.1 minutes, and for paramedics it was six minutes. Patients who were successfully converted using initial "quick-look" defibrillations were far more likely to be successfully resuscitated (79/88 [90%] vs 107/275 [39%], P > .0001) and to be discharged alive f r o m the hospital (54/88 [61%1 vs 43/275 [16%], P > .0001) than were those who required further ACLS techniques. The incidence of citizen bystander CPR for both groups was identical (24%). Patients who received bystander (professional and citizen) CPR were compared separately to patients not receiving bystander CPR (Table 2). No statistically significant population differences were found between these 13:11 November1984
TABLE 2. CPR groups
Professional 58 65 44:14 (3:1) 5.2 30 (52%) 22 (38%) 5 (9%) 20 (35%)* 30 (52%) 18 (31%)
No. Patients Age (y) Sex (M:F) Response time (min) Cardiac history Cardiac medications history Antiarrhythmics history Successful defibrillations Successful resuscitations Saves *P < .04.
Bystander (Pro and Citizen 144 65 109:35 (3:1) 5.6 74 (51%) 55 (38%) 18 (13%) 41 (29%) 74 (51%) 38 (26%)
Citizen 86 65 65:21 (3:1) 5.9 44 (51%) 33 (38%) 13 (15%) 21 (24%) 44 (51°/0) 20 (23%)
None 219 66 173:46 (4:1) 6.3 129 (59%) 97 (44%) 26 (12%) 47 (22%)* 112 (51%) 59 (27%)
TABLE 3. Paramedic response times (min)
No. patients Age (y) Sex (M:F) Cardiac history Cardiac medications history Antiarrhythmics history Professional CPR Citizen CPR Bystander CPR (pro and citizen) Successful defibrillations Successful resuscitations Saves *P < .04. l-p < .03.
~<4
>4
~<5
148 66 117:31 (4:1) 89 (60%) 61 (41%) 20 (14%) 31 (21%)* 30 (20%)
215 65 165:50 (3:1) 114 (53%) 91 (42%) 24 (11%) 27 (13%)* 56 (25%)
199 66 160:39 (4:1) 117 (59%) 84 (42%) 26 (13%) 40 (20%)l48 (24%)
61 (41%)
83 (38%)
33 (22%) 75 (51%) 42 (28%)
groups. Patients who were successfully defibrillated were statistically more likely to have received immediate professional CPR (23% vs 14%, P < .05). Of importance is the finding that successful resuscitation and ultimate discharge rates are similar for those patients receiving immediate bystander CPR and those not receiving any bystander CPR. W h e n separated f r o m the overall bystander group, patients receiving professional CPR had a s t a -
13:11November1984
>5
~<6
>6
164 65 122:42 (3:1) 86 (52%) 68 (42%) 18 (11%) 18 (11%)138 (23%)
232 65 183:49 (4:1) 135 (58%) 99 (43%) 28 (12%) 44 (19%) 56 (24%)
131 65 99:32 (3:1) 68 (52%) 53 (41%) 16 (12%) 14 (11%) 30 (23%)
88 (44%)
56 (34%)
100 (43%)
44 (34%)
55 (26%)
46 (23%)
42 (26%)
56 (24%)
32 (24%)
111 (51%) 55 (26%)
104 (52%) 57 (29%)
82 (50%) 40 (24%)
118 (51%) 66 (28%)
68 (52%) 31 (24%)
tistically significant improvement in rate of successful defibrillation when compared to the no bystander CPR group (35% vs 22%, P < .04). While there was a trend in the successful resuscitation and save rate, this difference was not statistically s i g n i f i c a n t . B e c a u s e of n o t e d differences, the study group was analyzed by response times. The results show, as expected, that statistically m o r e people received professional Annals of Emergency Medicine
CPR in the "less than or equal to fourm i n u t e " and "less than or equal to five-minute" groups (Table 3).
DISCUSSION Several s t u d i e s of v a r i o u s prehospital systems with both single and two-tiered responsesl~, t3 have suggested that, for patients in coarse ventricular fibrillation, the more rapidly the patient can be defibrillated, the greater the chance of conversion to a
1018/49
BYSTANDER CPR Kowalski et al
perfusing cardiac r h y t h m . Recent American Heart Association guidelines for A C L S r e c o m m e n d initial "quick-look" defibrillations with oxygenation and basic life support techniques prior to use of other ACLS drug protocols.14 The more rapidly normal perfusion to the brain can be established, the better the neurologic outcome and ultimate survival. In our study, of those patients who were successfully defibrillated with initial defibrillations, 90% were successfully resuscitated and admitted to a hospital. Two-thirds of the patients (61%) were discharged from the hospital. On the other hand those patients requiring further CPR and further ACLS techniques had significantly inferior outcomes. With the addition of potent cardiac drugs used in the advanced cardiac life support protocol, it may be unreasonable and simplistic to assume outcome related to a single variable, such as CPR. Other variables, especially "time to early defibrillation," should be analyzed when comparing prehospital systems. The role of CPR as a component of prehospital systems may be exaggerated. We did find a statistically significant difference in successful defibrillations between patients receiving professiona ! bystander CPR w h e n compared to patients who received no bystander CPR; however, patients receiving professional bystander CPR also had shorter paramedic response times than did those patients receiving no bystander CPR. This variable of time could account for the difference in defibrillation rate. Another variable that has been reported is may provide evidence that professional medical personnel may "do better CPR" than does the lay public. Other studies16,17 have shown that professionals may demonstrate less decrease in CPR skill levels over time. One could infer that experienced EMTs might remain more calm in an e m e r g e n c y and would be more able to provide a more uniform and structured approach to CPR performance than can a lay bystander. We also believe that the shortened response time found in the professional CPR group is directly related to knowledge of system access, which may have allowed shorter paramedic responses. There were, however, no statistically significant differences between the professional bystander CPR and no bystander CPR groups in relation 50/1019
to rate of successful resuscitation or hospital discharge. We can demonstate no difference between the overall bystander CPR group and the no bystander CPR group. The two groups had an identical successful resuscitation rate (51%), and a similar save rate (26% vs 27%). The predominant factor in the ultimate successful resuscitation and save rates .in any system may be the total amount of time from arrest to establ i s h m e n t of an effective r h y t h m . ACLS, and specifically early defibrillation, may be much more important in a prehospital EMS system than previously recognized. Basic cardiac life support is fraught with many problems: an erratic airway, deteriorating acid-base status, and no chance of defibrillation or use of cardiac drugs. With ACLS there exists, in addition to the ability to defibrillate coarse ventricular fibrillation, the ability to control the airway, augment oxygenation, and normalize acid-base status. Studies is in which portable defibrillators were made more available to small communities indicate that rapid defibrillation improves patient outcome. Even in a rapid response paramedic system such as ours, the most significant factor relating to successful resuscitation and hospital discharge is conversion response to initial defibrillations. While continued use of the ACLS protocol for coarse VF is beneficial, the overwhelming majority of successes occur with patients who are easily converted by initial defibrillation. We think that study data comparing prehospital systems should be much more specific in relation to time of paramedic response, time of defibrillation, and length of resuscitation in order to evaluate equitably the effectiveness of various resuscitation modalities, including CPR.
CONCLUSIONS A patient is far more likely to have a successful resuscitation and to be discharged alive from the hospital if he has a rhythm and a pulse after initial defibrillations than when converted after further ACLS techniques. Although professional CPR was associated w i t h s i g n i f i c a n t l y i m p r o v e d successful defibrillation, resuscitation or save rates when comparing citizen bystander CPR to no bystander CPR were similar. Previously reported CPR save rates may be system-specific and not applicable to all EMS systems. Annals of Emergency Medicine
Successful resuscitation and ultimate discharge outcome may be more dependent on time of defibrillation response than previously reported.
The authors thank Lauryl Pukansky and her staff for chart retrieval; Jerry Anderson, PhD (Division of Biostatisties and Clinical Epidemiology, Medical College of Wisconsin) for his assistance in statistical analysis; and Debra Vitek for her tireless preparation of this manuscript.
REFERENCES 1. Kouwenhoven WB, Jude JR, Knickerbocker GG: Closed-chest cardiac massage. l A M A 1960;173:1064-1067. 2. Jude JR, Kouwenhoven WB, Knickerbocker GG: Cardiac arrest: Report of application of external cardiac massage on 118 patients. JAMA 1961;178:1063-1070. 3. Ditchey RV, TWinkler JV, Rhodes CA: Relative lack of coronary blood flow during closed-chest resuscitation in dogs. Circulation 1982;66:297-302. 4. Niemann JT, Rosborough JP, Hausknecht M, et al: Pressure-~ynchronized cineangiography during experimental cardiopulmonary resuscitation. Circulation 1981j64:985-991. 5. Rudikoff MT, Maughan WL, Effron M, et al: Mechanisms of blood flow during cardiopulmonary resuscitation. Circulation 1980;61:345-352. 6. MacKenzie GJ, Taylor SH, McDonald AH, Donald KW: Haemodynamic effects of external cardiac compression. Lancet 1964;1:1342. 7. Thompson RG, Hallstrom AP, Cobb LA: Bystander-initiated cardiopulmonary resuscitation in the management of ventricular fibrillation. A n n I n t e r n M e d 1979;90:737-740. 8. Eisenberg MS, Bergner L, Hallstrom AP: Paramedic programs and out-of-hospital cardiac arrest: I. Factors associated with successful resuscitation. A m J Public Health 1979;69:30-38. 9. Lund I, Skulberg A: Cardiopulmonary resuscitation by lay people. Lancet 1976; 2: 702-705. 10. Tweed WA, Bristow G, Donen N: Resuscitation from cardiac arrest: Assessment of a system providing only basic life support outside of hospital. Can M e d Assoc [ 1980;122:297-300. 11. Guzy PM, Pearce ML, Greenfield S, et al: Effectiveness of citizen cardiopulmonary resuscitation during out-of-hospital emergencies in metropolitan Los Angeles. Circulation 1979;60(Part II):46. 12. Cobb LS, Baum RS, Alvarez H, et al: Resuscitation from out-of-hospital ventricular fibrillation: Four-year follow-up. Circulation 1975;51,52(suppl 3):223-228. 13:11 November 1984
13. Sherman MA: Mobile intensive care units: An evaluation of effectiveness. JAMA 1979;241:1899-1901. 14. Standards and guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). JAMA 1980; 244:483-493.
13:11 November 1984
15. Barick BL: How long do you remember lifesaving CPR skills? J Phys Ed Recreation 1977;48:62-63.
et ah Trainees' retention of cardiopulmonary resuscitation: How quickly they forget. JAMA 1979;24h901-903.
16. Watson LA: Cardiopulmonary resuscitation eompetencies of nurses. Int J Nurs Stud 1982;19:99-107.
18. Eisenberg MS, Copass MK, Hallstrom AP: Treatment of out-of-hospital cardiac arrests with rapid defibrillation by emergency medical technicians. N Engl J Med 1980;302:1379q383.
17. Weaver FJ, Ramirez AG, Dorfman SB,
Annals of Emergency Medicine
1020/51