Author + information
- Received June 25, 1999
- Revision received March 13, 2000
- Accepted April 14, 2000
- Published online September 1, 2000.
- Ken Nagao, MD∗,*,
- Nariyuki Hayashi, MD∗,
- Katsuo Kanmatsuse, MD, FACC†,
- Ken Arima, MD†,
- Jyoji Ohtsuki, MD∗,
- Kimio Kikushima, MD† and
- Ikuyoshi Watanabe, MD∗
- ↵*Reprint requests and correspondence: Dr. Ken Nagao, Surugadai Nihon University Hospital 1-8-13, Kanda Surugadai, Chiyoda-ku, Tokyo 101-8309, Japan
The purpose of this study was to evaluate the efficacy of an alternative cardiopulmonary cerebral resuscitation (CPCR) using emergency cardiopulmonary bypass (CPB), coronary reperfusion therapy and mild hypothermia.
Good recovery of patients with out-of-hospital cardiac arrest is still inadequate. An alternative therapeutic method for patients who do not respond to conventional CPCR is required.
A prospective preliminary study was performed in 50 patients with out-of-hospital cardiac arrest meeting the inclusion criteria. Patients were treated with standard CPCR and, if there was no response, by emergency CPB plus intra-aortic balloon pumping. Immediate coronary angiography for coronary reperfusion therapy was performed in patients with suspected acute coronary syndrome. Subsequently, in patients with systolic blood pressure above 90 mm Hg and Glasgow coma scale score of 3 to 5, mild hypothermia (34°C for at least two days) was induced by coil cooling. Neurologic outcome was assessed by cerebral performance categories at hospital discharge.
Thirty-six of the 50 patients were treated with emergency CPB, and 30 of 39 patients who underwent angiography suffered acute coronary artery occlusion. Return of spontaneous circulation and successful coronary reperfusion were achieved in 92% and 87%, respectively. Mild hypothermia could be induced in 23 patients, and 12 (52%) of them showed good recovery. Factors related to a good recovery were cardiac index in hypothermia and the presence of serious complications with hypothermia or CPB.
The alternative CPCR demonstrated an improvement in the incidence of good recovery. Based upon these findings, randomized studies of this hypothermia are needed.
The ultimate goal of cardiopulmonary cerebral resuscitation (CPCR) is to improve long-term outcome in patients who have suffered cardiac arrest. To reach this goal, however, there are various difficult problems such as cardiac resuscitation, cerebral resuscitation, urgent and intensive treatment of primary disease and rehabilitation and long-term treatment of primary disease. The most difficult problem is cerebral resuscitation in patients who have had cardiac arrest outside the hospital. A campaign for enlightenment in the chain of survival (1) has been performed in each country, but satisfactory effects have not been obtained yet. It is estimated that each year 8,000 people in Tokyo and 65,000 people in Japan develop cardiac arrest outside the hospital, but the rate of good recovery is extremely low (1% to 2%) (2,3). Therefore, we prepared a new protocol of CPCR with emergency cardiopulmonary bypass (CPB), coronary reperfusion therapy and mild hypothermia, and, in order to evaluate the efficacy of CPCR, a prospective preliminary study was performed in patients with out-of-hospital cardiac arrest meeting the inclusion criteria.
Tokyo emergency medical system (2)
Tokyo has an area of 1,750 km2, and its daytime and nighttime population in 1997 was estimated at 14,410,000 and 11,430,000, respectively. The public emergency medical service activity system of Tokyo consists of an ambulance unit, an aviation unit and a telephone information unit. The ambulance unit is organized with three emergency medical service personnel, at least one of whom is a licensed emergency life-saving technician (ELST) who passed the national examination after the emergency life-saving development course (≥835 h). The ELST has been authorized by the minister of health and welfare to treat for cardiac arrest through maintenance of the airway with a laryngeal mask or double balloon tube, intravenous infusion of drugs, and defibrillation with a semi-automatic defibrillator. The rules of the ambulance unit for cardiac arrest outside the hospital provide that the ELST cannot make a decision to stop cardiopulmonary resuscitation (CPR) and has to transport all patients, except for obviously dead bodies (such as a decapitation or postmortem rigidity), to the nearest emergency hospital. Consequently, each year about 96% of the patients with out-of-hospital cardiac arrest are transported under CPR by emergency medical service personnel.
Selection of patients
Patients who met all of the following new inclusion criteria of CPCR were enrolled in this study.
1. Patients age 18 to 74.
2. Patients who were witnessed having an arrest after chest pain, who cried out for help or were heard falling and were found immediately after arrest and whose first CPR was initiated within 15 min after arrest, or whose electrocardiogram recorded by the ELST demonstrated ventricular fibrillation (VF).
3. Patients who were ruled not to have acute aortic dissection and intracranial hemorrhage during resuscitation.
4. Patients with Glasgow coma scale score (4) between 3 and 5 upon arrival at an emergency room.
5. Patients whose families gave informed consent to the new CPCR therapy.
Treatment protocol in this study is shown in Figure 1. The patient with out-of-hospital cardiac arrest was assessed for the cardiac cause of arrest meeting the inclusion criteria as soon as possible, and the new CPCR, according to the treatment protocol, was initiated.
Cardiac resuscitation—advanced cardiac life support (ACLS) according to the guideline of the Emergency Cardiac Care and American Heart Association (1) was initiated in patients with cardiac arrest upon arrival at the emergency room. When return of spontaneous circulation (ROSC) could not be achieved in patients with VF after unsynchronized electric shocks, with the second administration of epinephrine, or in patients without VF after the second administration of epinephrine, emergency CPB and intra-aortic balloon pumping (IABP) were immediately performed in the emergency room. On the other hand, IABP was immediately performed in patients who achieved ROSC with standard ACLS before or after arrival at the emergency room and who did not show improved hemodynamic collapse. Then, assessment of ischemic cardiac arrest was made, and in the case of suspected acute coronary syndrome, emergency coronary angiography was subsequently performed. When acute coronary syndrome-related artery was obstructed with thrombolysis in myocardial infarction (TIMI) (5) flow grade 0 or 1, or was delayed with TIMI flow grade 2, percutaneous transluminal coronary angioplasty (PTCA) was performed.
Cerebral resuscitation—when a patient’s systolic blood pressure was increased above 90 mm Hg by ACLS and Glasgow coma scale score (4) was between 3 and 5, mild hypothermia was induced. Procedure of mild hypothermia (34°C) was performed by the direct blood cooling method with a coil (6). Pulmonary arterial blood temperature was monitored continuously as core temperature by using balloon flotation right-heart catheter. Mild hypothermia was maintained for at least two days (fundamentally for three days) and was induced by two-stage cooling (7) to reach 34°C within 6 h after initiation of cooling. Rewarming (7) was conducted slowly and gradually and took at least two days (warming by 0.5°C every 12 h then maintained at 35°C for 24 h) (Fig. 2). Patient’s condition during mild hypothermia was controlled as follows: systolic blood pressure 90 mm Hg or more, cardiac index 2.2 l/min/m2 or more, systemic oxygen delivery 520 ml/min/m2 or more, hemoglobin 12 g/dl or more, oxygen extraction ratio between 20% and 30% and activated coagulation time between 150 and 200 s by using the anticoagulant nafamostat mesilate. Anesthetics used during mild hypothermia were the gamma-aminobutyric acid-agonist midasolam and the muscle relaxant pancuronium (7). Furthermore, patients were nourished by complete parenteral nutrition during mild hypothermia. Water and electrolyte balances were controlled by continuous hemodiafiltration, and blood circulating in this hemodiafiltration cycle was cooled and used for hypothermia by coil cooling (6).
Data collection and neurologic outcome
Clinical data were obtained from a standardized form in our department according to the recommendations of the Utstein consensus conference (8). Baseline characteristics such as age, gender, location of arrest, initial cardiac rhythm, principal event-to-event intervals, Glasgow coma scale score, initial arterial blood gas and other clinically important characteristics were recorded. Acute coronary artery occlusion of patients with suspected acute coronary syndrome was assessed by at least three interventional cardiologists. The neurologic outcome at hospital discharge was assessed according to the cerebral performance categories (9) as follows: 1) good cerebral performance, 2) moderate cerebral disability, 3) severe cerebral disability, 4) coma or vegetative state, and 5) brain death or death.
Values were expressed as mean ± SD. Factors potentially relevant to good cerebral performance were assessed by univariate analysis with the chi-square test for categorical variables. A two-tailed p value of <0.05 was considered to indicate statistical significance. Predictors found to be significant on the basis of univariate analysis were then included in a forward, stepwise multiple logistic-regression model to identify the independent factors associated with a good recovery. Statistical analysis was performed with the Stata 4.0 statistical package for Macintosh (Stata, College Station, Texas).
Characteristics of the patients
From January 1996 to April 1998, 287 of the 18,247 patients with out-of-hospital cardiac arrest in Tokyo were transported to our emergency room. A total of 50 patients were treated after meeting the inclusion criteria with CPCR in this study. Characteristics of the 50 patients are shown in Table 1in the column marked “all patients.” The rate of bystander-initiated CPR was 40%, and the time interval from cardiac arrest to initiation of CPR was 7.9 ± 6.9 min. With regard to initial cardiac rhythm at the scene, 84% showed VF, and the remaining patients also showed VF subsequently. The rate of ROSC was 92% (14% of patients achieved ROSC before arrival at the hospital by emergency medical service personnel, and 78% achieved ROSC after arrival at the emergency room). The time interval from cardiac arrest to ROSC was 66.6 ± 42.5 min. The procedure of ACLS was standard ACLS in 28% and emergency CPB plus IABP in 72%. The cardiac cause of arrest was acute coronary syndrome in 78%, hypertrophic cardiomyopathy in 12% and myocarditis in 4%.
Cardiac catheterization and coronary intervention
Table 2shows angiographic data of all 39 patients with suspected acute coronary syndrome and those after PTCA for TIMI flow grades 0 to 2. The initial angiogram of acute coronary syndrome related artery revealed TIMI flow grades 0 to 2 in 77% (30/39) of the patients. Eighty-seven percent (26/30) of these arteries with TIMI flow grades 0 to 2 successfully restored anterograde flow of TIMI flow grade 3 by PTCA. In cases of non-ROSC, coronary flow was assessed by TIMI flow grade under CPB and IABP. Three of four patients who had a lesion with TIMI flow grade 3 and insignificant stenosis survived and were observed having coronary spasm by the acethylcholine load test (10) in chronic stage.
Data of mild hypothermia
Of 50 patients who underwent CPCR in this study, 46 patients continued ROSC (carotid arterial pulses were palpable and invasive arterial pressure was detectable) for more than 1 h. Mild hypothermia could be induced in 23 of these ROSC patients as systolic blood pressure increased to 90 mm Hg or more. Characteristics of the 23 patients are shown in Table 1 as patients undergoing hypothermia, and coronary angiographic data are shown in Table 2 as patients undergoing hypothermia. Table 3shows data of mild hypothermia treatment. The mean core temperature was 34.2 ± 0.8°C during the cooling stage. The mean induction time from initiation of cooling to reaching 34°C was 6.3 ± 3.4 h, the mean duration of cooling stage was 71.0 ± 49.1 h, and the mean rewarming time to 36°C was 43.9 ± 5.3 h.
Table 4shows cerebral performance categories at hospital discharge of 23 patients with mild hypothermia. The good cerebral performance rates of patients whose Glasgow coma score before hypothermia was 3, 4 or 5 were 35.7% (5/14), 100% (5/5) and 50% (2/4), respectively. The good cerebral performance rate of the entire 23 patients was 52.2% (12/23), and the survival discharge rate was 65.2% (15/23).
Cause of death
Of 36 patients who underwent emergency CPB plus IABP, four died without achieving ROSC, and 23 died without getting satisfactory improvement of hemodynamic collapse. Of 23 patients with mild hypothermia, two died of severe sepsis (11) as one of serious complications of hypothermia, five died of exacerbation of hemodynamic collapse during cooling, and two died of bleeding due to vessel injury at the insert site of CPB catheter.
Predictive factors of good recovery
Figure 3 shows the results of analysis of factors relevant to good cerebral performance in 23 patients with mild hypothermia. Cardiac index in cooling stage (odds ratio, 17.5; 95% confidence interval, 2.4 to 129.7; p < 0.01) and systemic oxygen delivery (odds ratio, 10; 95% confidence interval, 1.2 to 83.2; p = 0.03) were shown to be the factors associated with good cerebral performance. Furthermore, stepwise regression analysis including factors of cooling stage (core temperature, duration, hemoglobin, cardiac index, arterial oxygen content and oxygen extraction ratio) and the presence of serious complications with severe sepsis (11) or a large quantity of bleeding (blood transfusion of more than 2,000 ml was required) revealed that the cardiac index and serious complications were related to good cerebral performance—good cerebral performance = 0.13–0.56 (presence of serious complications with CPB or mild hypothermia) +0.18 (cardiac index in mild hypothermia at cooling stage), r = 0.683, p = 0.0091 (good cerebral performance indicates one and the other categories of cerebral performance indicate zero).
There are reports on some multicenter randomized standard ACLS clinical trials in patients with out-of-hospital cardiac arrest (12–15). The European epinephrine study (15) was conducted in 3,327 patients with out-of-hospital cardiac arrest and compared the efficacy of standard-dose epinephrine with that of high-dose epinephrine. There were no significant differences in the survival discharge rate between the two groups, and subgroup analysis revealed that the survival rate of patients with witnessed arrest with suspected cardiac cause was 3.0% (58/1,939) and that of patients with initial rhythm of VF was 7.4% (42/564). Furthermore, the survival discharge rate of patients who underwent active compression-decompression ACLS in the standard-dose epinephrine group was 3.6% (18/496). On the other hand, there was no multicenter randomized ACLS clinical trial in Japan. Katsurado et al. (3) reported that the one-month survival rate with conventional ACLS in Japan was 2.6% in all 31,206 patients with out-of-hospital cardiac arrest and 8.9% in patients with cardiac arrest who were observed and resuscitated initially by bystanders. In our previous study (16), the good recovery rate with conventional ACLS was 1.1% in all 808 patients with out-of-hospital cardiac arrest and 3.2% in VF patients upon arrival at the hospital. Therefore, it is conceivable that this may be a limit of resuscitation by conventional ACLS.
Emergency CPB and coronary reperfusion therapy
Currently, emergency CPB is performed in patients with cardiac arrest who do not respond to standard ACLS, patients with cardiogenic shock who are unresponsive to pressors and patients with pulmonary embolism (17–19). However, the efficacy of emergency CPB in cardiac arrest outside the hospital has been poor, and good recovery has rarely been obtained so far. The authors performed emergency CPB and immediate coronary reperfusion therapy in patients with acute myocardial infarction complicated with conventional ACLS-resistive VF. Return of spontaneous circulation was achieved in 88.5% of those patients, but the good recovery rate was low at 3.8% (20). Therefore, we considered that emergency CPB and immediate coronary reperfusion therapy may control a progression of myocardial ischemia and be a useful strategy for cardiac resuscitation, but further measures aimed at cerebral resuscitation are required. Spaulding et al. (21) reported that the survival discharge rate was 38.1% (32/84) in survivors with out-of-hospital cardiac arrest who received immediate angiography and PTCA, and successful PTCA was the independent factor of survival. However, Spaulding et al. (21) evaluated the efficacy in patients who had reached stable hemodynamic state upon arrival at the emergency room, and they did not describe cerebral resuscitation.
In recent years, hypothermia treatment using the surface cooling method has been performed with the aim of brain protection in patients with severe traumatic brain injury (22–25). There have been many reports on the mechanism of brain protection by hypothermia treatment, such as decrease in cerebral metabolism (26), preservation of high-energy phosphate stores (26), prevention of intracellular Ca2+ accumulation (27,28), prevention of cerebral thermo-pooling (29), preservation of blood-brain barrier (30), reduction of extracellular concentrations of excitatory neurotransmitters (25,31,32) and suppression of inflammatory response (25,33,34), but the complete mechanism remains to be elucidated. Moreover, the hypothermia treatment on procedure of cooling, degree of core temperature, duration in cooling stage and technique of induction and rewarming has not been established. By applying hypothermia treatment, we performed CPCR. Hypothermia was controlled according to the Hayashi method (7,29,35) except for monitoring of intracranial pressure, and the cooling procedure we used was the coil cooling method by direct blood cooling (6) because the coil cooling method was not laborious and a more precise control of core temperature could be exerted than with the surface cooling method. As for the overall results, the survival discharge rate and the good recovery rate were 30% (15/50) and 24% (12/50), respectively. These results were superior by two to five times those of conventional ACLS in patients with out-of-hospital VF reported so far (15,16,36). With regard to the 23 patients undergoing mild hypothermia, these rates were 65.2% (15/23) and 52.2% (12/23), respectively. Bernard et al. (37) reported that the good recovery rate at hospital discharge was 36.4% (8/22) in survivors who arrived comatose at the emergency room after out-of-hospital cardiac arrest and who were treated with moderate hypothermia (33°C for 12 h) by the surface cooling method. In comparison with Bernard’s study, the good recovery rate in our patients with mild hypothermia was better. Furthermore, 70% of our study patients with mild hypothermia did not achieve ROSC upon arrival at the emergency room. It was considered that the present patients presented with more severe conditions for resuscitation than Bernard’s patients. The analysis of factors related to good cerebral performance during mild hypothermia treatment revealed that the cardiac index in cooling stage and the presence of serious complications (severe sepsis under hypothermia and bleeding at the insert site of the CPB catheter) were the predicting factors. Infections (particularly pulmonary infection), coagulation disorders and cardiac arrhythmia have been reported as complications of hypothermia, and it has been considered that these complications may be caused by core temperature below 30°C or a cooling period longer than 24 h (38,39). In this study, we performed hypothermia treatment induced by the coil cooling method at 34°C for at least two days, which did not cause clinically significant arrhythmia or bleeding (except for bleeding due to vessel injury at the insert site of the CPB catheter), but two patients died of severe sepsis that caused pulmonary infection. Therefore, it is suggested that further infection-control measures and a specific technique to secure a vessel for CPB catheter insert with only one puncture are required.
There is a limitation in this study because this was not a randomized controlled study and the case number was small. Moreover, there is the problem that this treatment is more expensive than conventional CPCR. Linder et al. (40) reported that a significantly larger proportion of patients treated with vasopressin than of those treated with epinephrine were resuscitated successfully from out-of-hospital VF and survived for 24 h, but the neurological outcomes were similar. Therefore, cardiac resuscitation by ACLS with vasopressin and with coronary reperfusion therapy for acute coronary-artery occlusion and subsequent cerebral resuscitation by hypothermia should be evaluated hereafter. In conclusion, treatment with mild hypothermia by coil cooling, as indicated in patients with out-of-hospital cardiac cause of arrest who achieved ROSC by standard ACLS or emergency CPB plus IABP and who received coronary reperfusion therapy in acute coronary syndrome, may have improved chances of good recovery. Based on these findings, randomized studies of mild hypothermia for cerebral resuscitation in comatose survivors of out-of-hospital cardiac arrest are needed.
- advanced cardiac life support
- cardiopulmonary bypass
- cardiopulmonary cerebral resuscitation
- cardiopulmonary resuscitation
- emergency life-saving technician
- intra-aortic balloon pumping
- percutaneous transluminal coronary angioplasty
- return of spontaneous circulation
- thrombolysis in myocardial infarction
- ventricular fibrillation
- Received June 25, 1999.
- Revision received March 13, 2000.
- Accepted April 14, 2000.
- American College of Cardiology
- ↵Tokyo Fire Department. Prehospital care [In Japanese]. In: Emergency Medical Service Division of Tokyo Fire Department, editor. The Actual Situation of Emergency Medical Services in Tokyo. Tokyo: Sankei, 1998;33–43.
- Katsurada K,
- Kamijima G
- Chesebro J.H,
- Knatterud G,
- Roberts R,
- et al.
- Nagao K,
- Miki T,
- Okamoto K,
- et al.
- Hayashi N
- Yasue H,
- Horio Y,
- Nakamura N,
- et al.
- Members of the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee: American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference
- Nagao K,
- Ohba T,
- Arima K,
- et al.
- Nagao K,
- Satou K,
- Arima K,
- Watanabe I,
- Yamashita M,
- Kanmatsuse K
- Bering E.A Jr.
- Churn S.B,
- Taft W.C,
- Billingsley M.S,
- et al.
- Hayashi N,
- Hirayama T,
- Utagawa A
- Busto R,
- Globus M.Y.T,
- Dietrich W.D,
- et al.
- Kil H.Y,
- Zhang J,
- Piantadosi C.A
- Hayashi N. The adverse treatment, brain hypothermia and replacement therapy for severe brain injury. International satellite meeting of brain hypothermia (abstr). 1998;31–4.