American Journal of Emergency Medicine 32 (2014) 320–324
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Outcome from severe accidental hypothermia with cardiac arrest resuscitated with extracorporeal cardiopulmonary resuscitation Keigo Sawamoto, MD a,⁎, Steven B. Bird, MD b, Yoichi Katayama, MD a, Kunihiko Maekawa, MD a, Shuji Uemura, MD, PhD a, Katsutoshi Tanno, MD, PhD a, Eichi Narimatsu, MD, PhD a a b
Department of Emergency Medicine, Sapporo Medical University Department of Emergency Medicine, University of Massachusetts Medical School
a r t i c l e
i n f o
Article history: Received 22 October 2013 Received in revised form 7 December 2013 Accepted 7 December 2013
a b s t r a c t Purpose: This study aimed to identify factors of neurologic prognosis in severe accidental hypothermic patients with cardiac arrest. Basic procedures: This retrospective observational study was performed in a tertiary care university hospital in Sapporo, Japan (January 1994 to December 2012). We investigated 26 patients with accidental hypothermic cardiac arrest resuscitated with extracorporeal cardiopulmonary resuscitation (ECPR). We evaluated the neurologic outcome in patients who were resuscitated with ECPR at discharge from hospital. Main findings: In those 26 patients, their median age was 50.5 years; and 69.2% were male. The cause of hypothermia was exposure to cold air in 46.1%, submersion in 46.1%, and avalanche in 7.8%. Ten (38.5%) of these patients survived to favorable neurological outcome at discharge. Factors associated with favorable neurological outcome were a cardiac rhythm other than asystole (P = .009), nonasphyxial hypothermia (P = .006), higher pH (P = .01), and lower serum lactate (P = .01). In subgroup analyses, the patients with hypothermic cardiac arrest due to submersion or avalanche (asphyxia group) showed no factors associated with good neurological outcome, whereas the nonasphyxia group showed a significantly lower core temperature (P = .02) and a trend towards a lower serum lactate (P = .09). Principal conclusions: Patients with hypothermic cardiac arrest due to nonasphyxial hypothermia have improved neurologic outcomes when treated with ECPR compared to patients with asphyxial hypothermic cardiac arrest. Further investigation is needed to develop a prediction rule for patients with nonasphyxial hypothermic cardiac arrest to determine which patients would benefit from treatment with ECPR. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Severe accidental hypothermia (core temperature ≤ 28°C) is a condition associated with significant morbidity and mortality. Because of cerebral protection afforded by hypothermia and the possibility of return of spontaneous circulation (ROSC) despite prolonged duration of cardiac arrest, it has been recommended that resuscitation should be continued until the patient has been rewarmed to 33°C to 35°C [1–3]. Recently, there is consensus that treatment with extracorporeal cardiopulmonary resuscitation (ECPR) using extracorporeal membrane oxygenation (ECMO) is effective and safe in case of severe accidental hypothermia with cardiac arrest [4– 16]. However, efforts to identify factors associated with good neurologic outcome from accidental hypothermia have yielded conflicting results [1,4,5,17–19]. The aim of this study was to identify predictive factors associated with good neurological outcome of hypothermic cardiac arrest
⁎ Corresponding author. E-mail address: [email protected] (K. Sawamoto). 0735-6757/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajem.2013.12.023
patients resuscitated with ECPR from a single emergency department (ED) during a 19-year period.
2. Methods The present study received approval from our institutional review board. Informed consent was waived because of the life-threatening emergency situation and the absence of any therapeutic alternative. Information was delivered to the patient’s relatives after inclusion as appropriate in a life-threatening context. 2.1. Study design This study was conducted in the ED of the Sapporo Medical University Hospital, which is a major tertiary care center in Sapporo, a city with nearly 2 million people. The city is located on the northern Japanese island of Hokkaido and one of the few metropolises in the world with heavy snowfall in winter. Around 1000 critically ill patients are admitted from around Sapporo city to our ED every year. The medical records of patients who arrived with a body temperature
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
less than or equal to 35°C on arrival at the ED between January 1, 1994, and December 31, 2012, were retrospectively reviewed. Criteria for inclusion in this study were hypothermic cardiac arrest patients resuscitated with ECPR. Out-of-hospital resuscitation was delivered by emergency medical services according to the guidelines made by reference to the Japanese Resuscitation Council in study period. 2.2. Data collection and assessment of outcome Patient demographics, cause of hypothermia, core temperature at presentation, initial cardiac rhythm, estimated duration of cardiac arrest until ECPR started, arterial blood gas values and electrolytes on arrival, and outcome were collected. The mechanism of hypothermia was categorized as exposure to cold air, immersion in water, submersion in water, and avalanche. We also characterized the mechanism of hypothermia into asphyxia (submersion and avalanche) and nonasphyxia (exposure to cold air and immersion). Neurological outcome was assessed at the time of discharge from the hospital according to the Glasgow-Pittsburgh Cerebral Performance Categories scale (CPC): CPC 1, good cerebral performance; CPC 2, moderate cerebral disability; CPC 3, severe cerebral disability, conscious but dependent; CPC 4, coma; CPC 5, death. CPC 1 and 2 were classified as a good neurological outcome, whereas CPC 3, 4, and 5 were considered poor neurological outcomes. Subgroup analyses were performed in 2 groups: asphyxia and nonasphyxia. 2.3. ECMO management to resuscitate hypothermic cardiac arrest patients All patients were initially treated in the ED. The decision to initiate ECPR was dependent on the attending emergency physicians who treated the patient upon arrival. Extracorporeal membrane oxygenation was implanted in the ED by the ECPR team, which comprised emergency physicians and clinical engineers. Our ECMO device is a heparin-bounded all-in-one system with heat exchanger (CAPIOX, TERUMO Corp, Tokyo, Japan). We usually establish a percutaneous cannulation in the femoral vein and artery. Femoral cut down procedures were performed for children in this study. The pump flow started initially at 50 mL/kg/min and continued until bladder
321
core temperature reached 35°C. Heparin was infused to keep the activated clotting time between 150 and 200 seconds. Extracorporeal membrane oxygenation was gradually withdrawn with or without use of catecholamine infusion after the following patient conditions were satisfied: core body temperature ≥ 35°C, adequate oxygenation (PaO2 of at least 100 mm Hg using a ventilator with FiO2 of 0.4), and systolic blood pressure of at least 80 mm Hg. However, patients who remained unstable despite rewarming continued with ECMO or changed to ECMO via femoral and jugular veins. In this study period, targeted temperature management that cooled patients to 32°C to 35°C for 24 or 48 hours with or without ECMO in post cardiac arrest care was performed, in case that the patient had a score of less than 8 on the Glasgow Coma Scale at the time to be rewarmed to 35°C. 2.4. Statistics Continuous data are presented as means and standard deviations or medians and interquartile ranges. Categorical variables are presented as counts and percentages. Univariate analysis of comparisons for outcome was performed with χ 2 for categorical variables and the Mann-Whitney U test for continuous variables. A 2-tailed P value of less than .05 was considered to indicate statistical significance. Statistical analyses were performed using Statview 5.0 (SAS Institute Inc, Cary, NC).
3. Results A total of 198 hypothermic patients were treated during the study period. Their mean age was 54.7 years (range, 4-93 years), and 115 (58.1%) were male. The cause of hypothermia was exposure to cold air in 76.8%, immersion in cold water in 12.6%, submersion in 9.1%, and avalanche in 1.5% (Fig. 1). Of the 198 patients, 71 were hemodynamically stable on admission; and all of them were discharged home (Fig. 2). Of the remaining 127 hemodynamically unstable patients, 59 were rewarmed with ECPR and did not suffer cardiac arrest, whereas 68 patients had cardiac arrest. Of the 68 cardiac arrest patients, 6 patients did not receive attempts at resuscitation because they had excessive
Fig. 1. Distribution of accidental hypothermic patients per category by month.
322
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
Fig. 2. Flowchart of accidental hypothermic patients. ECPR, extracorporeal cardiopulmonary resuscitation; ROSC, return of spontaneous circulation.
time since the cardiac arrest (they were found frozen). Thirty-two patients with cardiac arrest were resuscitated without the use of ECPR, and 10 of them had ROSC. One of these 10 patients had a good neurological outcome. A total of 30 patients were resuscitated with ECPR, and all of them had ROSC. Four of these 30 patients had causes other than hypothermia for their cardiac arrest: 3 had an intracerebral hemorrhage, and 1 was an aortic dissection (each of these 4 patients had gone undiscovered for several hours during winter months). Thus, 26 patients were included in this study as cases of primary hypothermic cardiac arrest resuscitated with ECPR. The median age was 50.5 years (interquartile range [IQR], 28.5-58.8 years), and 18 were male (69.2%). Median core temperature was 24.4°C (IQR, 22.2°C26.5°C). The cause of hypothermia was associated with asphyxia in 53.8% of the patients (submersion, 46.1%; avalanche, 7.7%). Of the 26 patients, 10 had a good neurological outcome (CPC score upon discharge of 1 or 2) and 16 had a poor one. The characteristics of these 26 patients, categorized as good and poor neurological outcome, are presented in Table 1. Ten (38.5%) of the 26 patients were discharged from the hospital with a good neurological outcome. There were statistically significantly greater number of cases of nonasystolic rhythm on arrival (P = .009) and nonasphyxia (P = .006) in the good neurological outcome
group as compared with in the poor outcome group. Other factors shown to be associated with a good neurologic outcome in univariate analysis were higher pH (P = .01) and a lower serum lactate (P = .01). Table 2 shows the analysis of patients with hypothermic cardiac arrest caused by asphyxia. There were no significant differences in univariate analysis between good and poor neurologic outcome. Table 3 shows the analysis of patients with hypothermic cardiac arrest caused by nonasphyxia. Core temperature was significantly lower in the poor–neurological outcome group (24.6°C vs 21.3°C, P = .02). Lactate values in the poor–neurological outcome group showed a tendency to be higher than those in the good-outcome group (13.1 mmol/L vs 7.9 mmol/L, P = .09).
4. Discussions This retrospective study included 198 consecutive patients with accidental hypothermia who presented to the ED at one university hospital over a period of 19 years. Of those, we identified 26 patients who suffered from primary hypothermic cardiac arrest resuscitated with ECPR. This study is the largest series of hypothermic patients with cardiac arrest resuscitated with ECPR.
Table 1 Characteristics of primary hypothermic patients with cardiac arrest treated with ECPR
Age, y Male sex, n (%) Asystole/other rhythm Asphyxia/nonasphyxia Downtime, min Core temperature, °C pH PaCO2, mm Hg Serum potassium, mmol/L Serum lactate, mmol/L
All (n = 26)
Good neurologic outcome (n = 10)
Poor neurologic outcome (n = 16)
P value
50.5 (28.5-58.8) 18 (69.2%) 16/10 14/12 65 (51-74) 24.4 (22.2-26.5) 6.95 (6.78-7.22) 55.4 (40.8-76.2) 4.7 (3.8-6.3) 13.4 (9.4-24.3)
51.0 (40.5-58.8) 5 (50.0%) 3/7 2/8 54 (45-74) 24.6 (24.2-26.1) 7.08 (6.95-7.33) 53.7 (41.6-81.7) 3.9 (3.0-5.6) 9.0 (7.0-11.1)
49.0 (27.0-56.8) 13 (81.3%) 13/3 12/4 66 (61-79) 24.2 (20.5-26.7) 6.84 (6.64-6.97) 55.4 (43.2-69.5) 4.9 (4.7-6.7) 16.9 (13.4-24.3)
.54 .09 .009 .006 .17 .33 .01 .91 .07 .01
All values are medians with IQR. Downtime indicates estimated duration of cardiac arrest until ECPR stared.
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
323
Table 2 Characteristics of patients resuscitated with ECPR in hypothermic cardiac arrest associated with asphyxia
Age, y Male sex, n (%) Asystole/other rhythm Downtime, min Core temperature, °C pH PaCO2, mm Hg Serum potassium, mmol/L Serum lactate, mmol/L
All (n = 14)
Good neurologic outcome (n = 2)
Poor neurologic outcome (n = 12)
P value
34.0 (12.8-55.0) 12 (85.7%) 11/3 68 (64-84) 25.5 (22.5-27.0) 6.84 (6.63-6.95) 60.2 (45.7-76.2) 5.4 (4.7-7.4) 24.3 (16.6-25.6)
22.5 (14.8-30.3) 2 (100%) 1/1 80 (75-84) 25.9 (25.1-26.7) 6.93 (6.91-6.94) 98.8 (81.9-115.7) 4.8 (3.8-5.7) 25.2 (24.9-25.6)
40.0 (20.3-56.8) 10 (83.3%) 10/2 67 (61-85) 25.5 (21.5-26.9) 6.79 (6.58-6.95) 55.4 (43.2-69.5) 5.4 (4.7-7.7) 22.0 (15.6-25.1)
.47 .53 .29 .27 .65 .36 .20 .47 .27
All values are medians with IQR. Downtime indicates estimated duration of cardiac arrest until ECPR stared.
It is increasingly recognized that treatment with ECPR is effective and safe for the treatment of hypothermia [4–16]. Extracorporeal cardiopulmonary resuscitation can provide sufficient circulation and oxygenation during rewarming. Rewarming rates of 8°C/h to 12°C/h are possible with the fastest rewarming devices [4,6,20]. Extracorporeal cardiopulmonary resuscitation also provides both circulatory and respiratory support after ROSC, when there is a high incidence of severe cardiopulmonary dysfunction [4,7,20]. Several investigators have reported on prognostic factors likely to identify patients in hypothermic cardiac arrest who are likely to benefit from resuscitation with ECPR [1,4,5,8,17–20]. Farstad et al [8] analyzed 22 hypothermic cardiac arrest patients resuscitated with ECPR and suggested that extreme hyperkalemia (serum potassium N10 mmol/L) is a sign of severe cellular dysfunction and portends a dismal prognosis. Mair et al analyzed 22 hypothermic cardiac arrest patients resuscitated with ECPR and suggested that plasma potassium levels (serum potassium N 9 mmol/L), central venous pH (pH b 6.50), and activated clotting time (N400 seconds) on admission can be used to identify hypothermic arrest victims in whom death preceded the hypothermia [1]. Hauty et al [17] analyzed 10 severely hypothermic patients rescued from a snow-covered mountain and resuscitated with ECPR. They concluded that hyperkalemia (N10 mmol/L) and markedly elevated serum ammonia levels (N 250 mcmol/L) predict a dire outcome. Silfvast et al [5] analyzed 23 hypothermic cardiac arrest patients resuscitated with ECPR and concluded that, of the 23 patients, 22 could be correctly classified as survivors or nonsurvivors based on the level of serum potassium and arterial PaCO2 upon arrival to the hospital. Our study found significant differences in pH and serum lactate upon arrival to the ED between patients with good and poor neurologic outcomes. Derangements in clinical data such as pH, serum potassium, PaCO2, and lactate could reflect cardiac arrest that preceded the development of hypothermia. If asphyxiation occurs before cardiac arrest (even in patients that develop hypothermia), the prognosis is very poor, which is further supported by our data [1,21]. Subgroup analysis in both asphyxial and nonasphyxial patients was conducted to evaluate neurologic outcome in hypothermic
cardiac arrest. Asphyxial patients demonstrated worse pH, PaCO2, serum potassium, and lactate values compared with nonasphyxial patients. These results strongly suggest that hypoxia preceded the development of hypothermia in the asphyxia group. Interestingly, however, there were no significant differences in the characteristics of patients with good and poor neurologic outcomes in the asphyxial group. This suggests that the maxim “a hypothermic patient is not dead until warm and dead” may be valid. On the other hand, in the nonasphyxial group, there were a significantly lower core temperature and slightly higher lactate in the poor–neurological outcome group. We suspect that these values represent a longer period of cold exposure. These results may be useful as reference points for prognosis in such a group. There are several limitations to our study. Firstly, our study was a retrospective study from a single institution. Furthermore, although this is the largest cohort of hypothermic patients treated with ECPR, the sample size is small. There is also a risk of selection bias in our sample, as the decision to use ECPR was based solely upon the judgment of the primary emergency physician.
5. Conclusion Our data suggest that asphyxiation such as seen with submersion and avalanche in hypothermic cardiac arrest patients is predictive of a poor neurological prognosis, whereas nonasphyxial hypothermic cardiac arrest patients have a good neurologic outcome two-thirds of the time when treated with ECPR. Although a lower core body temperature was associated with a decreased likelihood of a good neurologic outcome, it is not possible to determine neurologic outcome based upon core body temperature. We therefore recommend that all cardiac arrest patients with severe hypothermia, especially nonasphyxial hypothermia, be treated with ECPR, if possible, until they are rewarmed to at least 35°C. Further investigation is needed to develop a prediction rule for good neurologic outcome.
Table 3 Characteristics of patients resuscitated with ECPR in hypothermic cardiac arrest associated with non-asphyxia
Age, y Male sex, n (%) Asystole/other rhythm Downtime, min Core temperature, °C pH PaCO2, mm Hg Serum potassium, mmol/L Serum lactate, mmol/L
All (n = 12)
Good neurologic outcome (n = 8)
Poor neurologic outcome (n = 4)
P value
52.0 (48.0-59.8) 6 (50.0%) 5/7 55 (46-68) 24.1 (22.7-24.9) 7.20 (6.95-7.33) 52.6 (36.5-65.1) 4.1 (3.1-4.6) 9.0 (6.8-11.9)
54.5 (50.3-59.8) 3 (37.5%) 2/6 48 (42-63) 24.6 (24.1-26.0) 7.22 (7.03-7.33) 49.0 (36.5-65.1) 3.9 (3.1-4.7) 7.9 (6.3-9.4)
50.5 (47.3-58.3) 3 (75.0%) 3/1 66 (61-77) 21.3 (20.0-22.5) 7.11 (6.88-7.29) 55.4 (43.9-81.1) 4.4 (3.8-4.6) 13.1 (11.2-13.4)
.80 .22 .10 .17 .02 .35 .61 .80 .09
All values are medians with IQR. Downtime indicates estimated duration of cardiac arrest until ECPR stared.
324
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
References [11] [1] Mair P, Kornberger E, Furtwaengler W, et al. Prognostic markers in patients with severe hypothermia and cardiocirculatory arrest. Resuscitation 1994;27:47–54. [2] Wollenek G, Honarwar N, Golej J, et al. Cold water submersion and cardiac arrest in treatment of severe hypothermia with cardiopulmonary bypass. Resuscitation 2002;52:255–63. [3] Bierens JJ, Knape JT, Gelissen HP. Drowning. Curr Opin Crit Care 2002;8:578–86. [4] Ruttmann E, Weissenbacher A, Ulmer H, et al. Prolonged extracorporeal membrane oxygenation-assisted support provides improved survival in hypothermic patients with cardiocirculatory arrest. J Thorac Cardiovasc Surg 2007;134: 594–600. [5] Silfvast T, Pettila V. Outcome from severe accidental hypothermia in Southern Finland—a 10-year review. Resuscitation 2003;59:285–90. [6] Walpoth BH, Walpoth-Aslan BN, Mattle HP, et al. Outcome of survivors of accidental deep hypothermia and circulatory arrest treated with extracorporeal blood warming. N Engl J Med 1997;337:1500–5. [7] Morita S, Inokuchi S, Yamagiwa T, et al. Efficacy of portable and percutaneous cardiopulmonary bypass rewarming versus that of conventional internal rewarming for patients with accidental deep hypothermia. Crit Care Med 2011;39:1064–8. [8] Farstad M, Andersen KS, Koller ME, et al. Rewarming from accidental hypothermia by extracorporeal circulation: a retrospective study. Eur J Cardiothorac Surg 2001;20:58–64. [9] Larach MG. Accidental hypothermia. Lancet 1995;345:493–8. [10] Vanden Hoek TL, Morrison LJ, Shuster M, et al. Part 12: cardiac arrest in special situations: 2010 American Heart Association guidelines for cardiopulmonary
[12] [13] [14]
[15] [16] [17] [18] [19]
[20] [21]
resuscitation and emergency cardiovascular care. Circulation 2010;122(Suppl. 3): S829–61. Soar J, Perkins GD, Abbas G, et al. European Resuscitation Council guidelines for resuscitation 2010 section 8: cardiac arrest in special circumstances: electrolyte abnormalities, poisoning, drowning, accidental hypothermia, hyperthermia, asthma, anaphylaxis, cardiac surgery, trauma, pregnancy, electrocution. Resuscitation 2010;81:1400–33. Danzl DF, Pozos RS. Accidental hypothermia. N Engl J Med 1994;331:1756–60. Althaus U, Aeberhard P, Shupbach P, et al. Management of profound accidental hypothermia with cardiorespiratory arrest. Ann Surg 1982;195:492–5. Oberhammer R, Beikircher W, Hormann C, et al. Full recovery of an avalanche victim with profound hypothermia and prolonged cardiac arrest treated by extracorporeal re-warming. Resuscitation 2008;76:474–80. Gilbert M, Busund R, Skagseth A, et al. Resuscitation from accidental hypothermia of 13.7 degrees C with circulatory arrest. Lancet 2000;355:375–6. Dobson JAR, Burgess JJ. Resuscitation of severe hypothermia by extracorporeal rewarming in a child. J Trauma 1996;40:483–5. Hauty M, Esrig BC, Hill JG, Long WB. Prognostic factors in severe accidental hypothermia: experience from the Mt Hood tragedy. J Trauma 1987;27:1107–12. Schaller M-D, Fischer A, Perret C. Hyperkalemia. A prognostic factor during acute severe hypothermia. JAMA 1990;264:1842–5. van der Ploeg GJ, Goslings JC, Walpoth BH, et al. Accidental hypothermia: rewarming treatments, complications and outcomes from one university medical centre. Resuscitation 2010;81:1550–5. Douglas JAB, Hermann B, Jeff B, et al. Accidental hypothermia. N Engl J Med 2012;367:1930–8. Schaller MD, Fischer AP, Perret CH. Hyperkalemia. A prognostic factor during acute severe hypothermia. J Am Med Assoc 1990;264:1842–5.
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Original Contribution
Outcome from severe accidental hypothermia with cardiac arrest resuscitated with extracorporeal cardiopulmonary resuscitation Keigo Sawamoto, MD a,⁎, Steven B. Bird, MD b, Yoichi Katayama, MD a, Kunihiko Maekawa, MD a, Shuji Uemura, MD, PhD a, Katsutoshi Tanno, MD, PhD a, Eichi Narimatsu, MD, PhD a a b
Department of Emergency Medicine, Sapporo Medical University Department of Emergency Medicine, University of Massachusetts Medical School
a r t i c l e
i n f o
Article history: Received 22 October 2013 Received in revised form 7 December 2013 Accepted 7 December 2013
a b s t r a c t Purpose: This study aimed to identify factors of neurologic prognosis in severe accidental hypothermic patients with cardiac arrest. Basic procedures: This retrospective observational study was performed in a tertiary care university hospital in Sapporo, Japan (January 1994 to December 2012). We investigated 26 patients with accidental hypothermic cardiac arrest resuscitated with extracorporeal cardiopulmonary resuscitation (ECPR). We evaluated the neurologic outcome in patients who were resuscitated with ECPR at discharge from hospital. Main findings: In those 26 patients, their median age was 50.5 years; and 69.2% were male. The cause of hypothermia was exposure to cold air in 46.1%, submersion in 46.1%, and avalanche in 7.8%. Ten (38.5%) of these patients survived to favorable neurological outcome at discharge. Factors associated with favorable neurological outcome were a cardiac rhythm other than asystole (P = .009), nonasphyxial hypothermia (P = .006), higher pH (P = .01), and lower serum lactate (P = .01). In subgroup analyses, the patients with hypothermic cardiac arrest due to submersion or avalanche (asphyxia group) showed no factors associated with good neurological outcome, whereas the nonasphyxia group showed a significantly lower core temperature (P = .02) and a trend towards a lower serum lactate (P = .09). Principal conclusions: Patients with hypothermic cardiac arrest due to nonasphyxial hypothermia have improved neurologic outcomes when treated with ECPR compared to patients with asphyxial hypothermic cardiac arrest. Further investigation is needed to develop a prediction rule for patients with nonasphyxial hypothermic cardiac arrest to determine which patients would benefit from treatment with ECPR. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Severe accidental hypothermia (core temperature ≤ 28°C) is a condition associated with significant morbidity and mortality. Because of cerebral protection afforded by hypothermia and the possibility of return of spontaneous circulation (ROSC) despite prolonged duration of cardiac arrest, it has been recommended that resuscitation should be continued until the patient has been rewarmed to 33°C to 35°C [1–3]. Recently, there is consensus that treatment with extracorporeal cardiopulmonary resuscitation (ECPR) using extracorporeal membrane oxygenation (ECMO) is effective and safe in case of severe accidental hypothermia with cardiac arrest [4– 16]. However, efforts to identify factors associated with good neurologic outcome from accidental hypothermia have yielded conflicting results [1,4,5,17–19]. The aim of this study was to identify predictive factors associated with good neurological outcome of hypothermic cardiac arrest
⁎ Corresponding author. E-mail address: [email protected] (K. Sawamoto). 0735-6757/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajem.2013.12.023
patients resuscitated with ECPR from a single emergency department (ED) during a 19-year period.
2. Methods The present study received approval from our institutional review board. Informed consent was waived because of the life-threatening emergency situation and the absence of any therapeutic alternative. Information was delivered to the patient’s relatives after inclusion as appropriate in a life-threatening context. 2.1. Study design This study was conducted in the ED of the Sapporo Medical University Hospital, which is a major tertiary care center in Sapporo, a city with nearly 2 million people. The city is located on the northern Japanese island of Hokkaido and one of the few metropolises in the world with heavy snowfall in winter. Around 1000 critically ill patients are admitted from around Sapporo city to our ED every year. The medical records of patients who arrived with a body temperature
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
less than or equal to 35°C on arrival at the ED between January 1, 1994, and December 31, 2012, were retrospectively reviewed. Criteria for inclusion in this study were hypothermic cardiac arrest patients resuscitated with ECPR. Out-of-hospital resuscitation was delivered by emergency medical services according to the guidelines made by reference to the Japanese Resuscitation Council in study period. 2.2. Data collection and assessment of outcome Patient demographics, cause of hypothermia, core temperature at presentation, initial cardiac rhythm, estimated duration of cardiac arrest until ECPR started, arterial blood gas values and electrolytes on arrival, and outcome were collected. The mechanism of hypothermia was categorized as exposure to cold air, immersion in water, submersion in water, and avalanche. We also characterized the mechanism of hypothermia into asphyxia (submersion and avalanche) and nonasphyxia (exposure to cold air and immersion). Neurological outcome was assessed at the time of discharge from the hospital according to the Glasgow-Pittsburgh Cerebral Performance Categories scale (CPC): CPC 1, good cerebral performance; CPC 2, moderate cerebral disability; CPC 3, severe cerebral disability, conscious but dependent; CPC 4, coma; CPC 5, death. CPC 1 and 2 were classified as a good neurological outcome, whereas CPC 3, 4, and 5 were considered poor neurological outcomes. Subgroup analyses were performed in 2 groups: asphyxia and nonasphyxia. 2.3. ECMO management to resuscitate hypothermic cardiac arrest patients All patients were initially treated in the ED. The decision to initiate ECPR was dependent on the attending emergency physicians who treated the patient upon arrival. Extracorporeal membrane oxygenation was implanted in the ED by the ECPR team, which comprised emergency physicians and clinical engineers. Our ECMO device is a heparin-bounded all-in-one system with heat exchanger (CAPIOX, TERUMO Corp, Tokyo, Japan). We usually establish a percutaneous cannulation in the femoral vein and artery. Femoral cut down procedures were performed for children in this study. The pump flow started initially at 50 mL/kg/min and continued until bladder
321
core temperature reached 35°C. Heparin was infused to keep the activated clotting time between 150 and 200 seconds. Extracorporeal membrane oxygenation was gradually withdrawn with or without use of catecholamine infusion after the following patient conditions were satisfied: core body temperature ≥ 35°C, adequate oxygenation (PaO2 of at least 100 mm Hg using a ventilator with FiO2 of 0.4), and systolic blood pressure of at least 80 mm Hg. However, patients who remained unstable despite rewarming continued with ECMO or changed to ECMO via femoral and jugular veins. In this study period, targeted temperature management that cooled patients to 32°C to 35°C for 24 or 48 hours with or without ECMO in post cardiac arrest care was performed, in case that the patient had a score of less than 8 on the Glasgow Coma Scale at the time to be rewarmed to 35°C. 2.4. Statistics Continuous data are presented as means and standard deviations or medians and interquartile ranges. Categorical variables are presented as counts and percentages. Univariate analysis of comparisons for outcome was performed with χ 2 for categorical variables and the Mann-Whitney U test for continuous variables. A 2-tailed P value of less than .05 was considered to indicate statistical significance. Statistical analyses were performed using Statview 5.0 (SAS Institute Inc, Cary, NC).
3. Results A total of 198 hypothermic patients were treated during the study period. Their mean age was 54.7 years (range, 4-93 years), and 115 (58.1%) were male. The cause of hypothermia was exposure to cold air in 76.8%, immersion in cold water in 12.6%, submersion in 9.1%, and avalanche in 1.5% (Fig. 1). Of the 198 patients, 71 were hemodynamically stable on admission; and all of them were discharged home (Fig. 2). Of the remaining 127 hemodynamically unstable patients, 59 were rewarmed with ECPR and did not suffer cardiac arrest, whereas 68 patients had cardiac arrest. Of the 68 cardiac arrest patients, 6 patients did not receive attempts at resuscitation because they had excessive
Fig. 1. Distribution of accidental hypothermic patients per category by month.
322
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
Fig. 2. Flowchart of accidental hypothermic patients. ECPR, extracorporeal cardiopulmonary resuscitation; ROSC, return of spontaneous circulation.
time since the cardiac arrest (they were found frozen). Thirty-two patients with cardiac arrest were resuscitated without the use of ECPR, and 10 of them had ROSC. One of these 10 patients had a good neurological outcome. A total of 30 patients were resuscitated with ECPR, and all of them had ROSC. Four of these 30 patients had causes other than hypothermia for their cardiac arrest: 3 had an intracerebral hemorrhage, and 1 was an aortic dissection (each of these 4 patients had gone undiscovered for several hours during winter months). Thus, 26 patients were included in this study as cases of primary hypothermic cardiac arrest resuscitated with ECPR. The median age was 50.5 years (interquartile range [IQR], 28.5-58.8 years), and 18 were male (69.2%). Median core temperature was 24.4°C (IQR, 22.2°C26.5°C). The cause of hypothermia was associated with asphyxia in 53.8% of the patients (submersion, 46.1%; avalanche, 7.7%). Of the 26 patients, 10 had a good neurological outcome (CPC score upon discharge of 1 or 2) and 16 had a poor one. The characteristics of these 26 patients, categorized as good and poor neurological outcome, are presented in Table 1. Ten (38.5%) of the 26 patients were discharged from the hospital with a good neurological outcome. There were statistically significantly greater number of cases of nonasystolic rhythm on arrival (P = .009) and nonasphyxia (P = .006) in the good neurological outcome
group as compared with in the poor outcome group. Other factors shown to be associated with a good neurologic outcome in univariate analysis were higher pH (P = .01) and a lower serum lactate (P = .01). Table 2 shows the analysis of patients with hypothermic cardiac arrest caused by asphyxia. There were no significant differences in univariate analysis between good and poor neurologic outcome. Table 3 shows the analysis of patients with hypothermic cardiac arrest caused by nonasphyxia. Core temperature was significantly lower in the poor–neurological outcome group (24.6°C vs 21.3°C, P = .02). Lactate values in the poor–neurological outcome group showed a tendency to be higher than those in the good-outcome group (13.1 mmol/L vs 7.9 mmol/L, P = .09).
4. Discussions This retrospective study included 198 consecutive patients with accidental hypothermia who presented to the ED at one university hospital over a period of 19 years. Of those, we identified 26 patients who suffered from primary hypothermic cardiac arrest resuscitated with ECPR. This study is the largest series of hypothermic patients with cardiac arrest resuscitated with ECPR.
Table 1 Characteristics of primary hypothermic patients with cardiac arrest treated with ECPR
Age, y Male sex, n (%) Asystole/other rhythm Asphyxia/nonasphyxia Downtime, min Core temperature, °C pH PaCO2, mm Hg Serum potassium, mmol/L Serum lactate, mmol/L
All (n = 26)
Good neurologic outcome (n = 10)
Poor neurologic outcome (n = 16)
P value
50.5 (28.5-58.8) 18 (69.2%) 16/10 14/12 65 (51-74) 24.4 (22.2-26.5) 6.95 (6.78-7.22) 55.4 (40.8-76.2) 4.7 (3.8-6.3) 13.4 (9.4-24.3)
51.0 (40.5-58.8) 5 (50.0%) 3/7 2/8 54 (45-74) 24.6 (24.2-26.1) 7.08 (6.95-7.33) 53.7 (41.6-81.7) 3.9 (3.0-5.6) 9.0 (7.0-11.1)
49.0 (27.0-56.8) 13 (81.3%) 13/3 12/4 66 (61-79) 24.2 (20.5-26.7) 6.84 (6.64-6.97) 55.4 (43.2-69.5) 4.9 (4.7-6.7) 16.9 (13.4-24.3)
.54 .09 .009 .006 .17 .33 .01 .91 .07 .01
All values are medians with IQR. Downtime indicates estimated duration of cardiac arrest until ECPR stared.
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
323
Table 2 Characteristics of patients resuscitated with ECPR in hypothermic cardiac arrest associated with asphyxia
Age, y Male sex, n (%) Asystole/other rhythm Downtime, min Core temperature, °C pH PaCO2, mm Hg Serum potassium, mmol/L Serum lactate, mmol/L
All (n = 14)
Good neurologic outcome (n = 2)
Poor neurologic outcome (n = 12)
P value
34.0 (12.8-55.0) 12 (85.7%) 11/3 68 (64-84) 25.5 (22.5-27.0) 6.84 (6.63-6.95) 60.2 (45.7-76.2) 5.4 (4.7-7.4) 24.3 (16.6-25.6)
22.5 (14.8-30.3) 2 (100%) 1/1 80 (75-84) 25.9 (25.1-26.7) 6.93 (6.91-6.94) 98.8 (81.9-115.7) 4.8 (3.8-5.7) 25.2 (24.9-25.6)
40.0 (20.3-56.8) 10 (83.3%) 10/2 67 (61-85) 25.5 (21.5-26.9) 6.79 (6.58-6.95) 55.4 (43.2-69.5) 5.4 (4.7-7.7) 22.0 (15.6-25.1)
.47 .53 .29 .27 .65 .36 .20 .47 .27
All values are medians with IQR. Downtime indicates estimated duration of cardiac arrest until ECPR stared.
It is increasingly recognized that treatment with ECPR is effective and safe for the treatment of hypothermia [4–16]. Extracorporeal cardiopulmonary resuscitation can provide sufficient circulation and oxygenation during rewarming. Rewarming rates of 8°C/h to 12°C/h are possible with the fastest rewarming devices [4,6,20]. Extracorporeal cardiopulmonary resuscitation also provides both circulatory and respiratory support after ROSC, when there is a high incidence of severe cardiopulmonary dysfunction [4,7,20]. Several investigators have reported on prognostic factors likely to identify patients in hypothermic cardiac arrest who are likely to benefit from resuscitation with ECPR [1,4,5,8,17–20]. Farstad et al [8] analyzed 22 hypothermic cardiac arrest patients resuscitated with ECPR and suggested that extreme hyperkalemia (serum potassium N10 mmol/L) is a sign of severe cellular dysfunction and portends a dismal prognosis. Mair et al analyzed 22 hypothermic cardiac arrest patients resuscitated with ECPR and suggested that plasma potassium levels (serum potassium N 9 mmol/L), central venous pH (pH b 6.50), and activated clotting time (N400 seconds) on admission can be used to identify hypothermic arrest victims in whom death preceded the hypothermia [1]. Hauty et al [17] analyzed 10 severely hypothermic patients rescued from a snow-covered mountain and resuscitated with ECPR. They concluded that hyperkalemia (N10 mmol/L) and markedly elevated serum ammonia levels (N 250 mcmol/L) predict a dire outcome. Silfvast et al [5] analyzed 23 hypothermic cardiac arrest patients resuscitated with ECPR and concluded that, of the 23 patients, 22 could be correctly classified as survivors or nonsurvivors based on the level of serum potassium and arterial PaCO2 upon arrival to the hospital. Our study found significant differences in pH and serum lactate upon arrival to the ED between patients with good and poor neurologic outcomes. Derangements in clinical data such as pH, serum potassium, PaCO2, and lactate could reflect cardiac arrest that preceded the development of hypothermia. If asphyxiation occurs before cardiac arrest (even in patients that develop hypothermia), the prognosis is very poor, which is further supported by our data [1,21]. Subgroup analysis in both asphyxial and nonasphyxial patients was conducted to evaluate neurologic outcome in hypothermic
cardiac arrest. Asphyxial patients demonstrated worse pH, PaCO2, serum potassium, and lactate values compared with nonasphyxial patients. These results strongly suggest that hypoxia preceded the development of hypothermia in the asphyxia group. Interestingly, however, there were no significant differences in the characteristics of patients with good and poor neurologic outcomes in the asphyxial group. This suggests that the maxim “a hypothermic patient is not dead until warm and dead” may be valid. On the other hand, in the nonasphyxial group, there were a significantly lower core temperature and slightly higher lactate in the poor–neurological outcome group. We suspect that these values represent a longer period of cold exposure. These results may be useful as reference points for prognosis in such a group. There are several limitations to our study. Firstly, our study was a retrospective study from a single institution. Furthermore, although this is the largest cohort of hypothermic patients treated with ECPR, the sample size is small. There is also a risk of selection bias in our sample, as the decision to use ECPR was based solely upon the judgment of the primary emergency physician.
5. Conclusion Our data suggest that asphyxiation such as seen with submersion and avalanche in hypothermic cardiac arrest patients is predictive of a poor neurological prognosis, whereas nonasphyxial hypothermic cardiac arrest patients have a good neurologic outcome two-thirds of the time when treated with ECPR. Although a lower core body temperature was associated with a decreased likelihood of a good neurologic outcome, it is not possible to determine neurologic outcome based upon core body temperature. We therefore recommend that all cardiac arrest patients with severe hypothermia, especially nonasphyxial hypothermia, be treated with ECPR, if possible, until they are rewarmed to at least 35°C. Further investigation is needed to develop a prediction rule for good neurologic outcome.
Table 3 Characteristics of patients resuscitated with ECPR in hypothermic cardiac arrest associated with non-asphyxia
Age, y Male sex, n (%) Asystole/other rhythm Downtime, min Core temperature, °C pH PaCO2, mm Hg Serum potassium, mmol/L Serum lactate, mmol/L
All (n = 12)
Good neurologic outcome (n = 8)
Poor neurologic outcome (n = 4)
P value
52.0 (48.0-59.8) 6 (50.0%) 5/7 55 (46-68) 24.1 (22.7-24.9) 7.20 (6.95-7.33) 52.6 (36.5-65.1) 4.1 (3.1-4.6) 9.0 (6.8-11.9)
54.5 (50.3-59.8) 3 (37.5%) 2/6 48 (42-63) 24.6 (24.1-26.0) 7.22 (7.03-7.33) 49.0 (36.5-65.1) 3.9 (3.1-4.7) 7.9 (6.3-9.4)
50.5 (47.3-58.3) 3 (75.0%) 3/1 66 (61-77) 21.3 (20.0-22.5) 7.11 (6.88-7.29) 55.4 (43.9-81.1) 4.4 (3.8-4.6) 13.1 (11.2-13.4)
.80 .22 .10 .17 .02 .35 .61 .80 .09
All values are medians with IQR. Downtime indicates estimated duration of cardiac arrest until ECPR stared.
324
K. Sawamoto et al. / American Journal of Emergency Medicine 32 (2014) 320–324
References [11] [1] Mair P, Kornberger E, Furtwaengler W, et al. Prognostic markers in patients with severe hypothermia and cardiocirculatory arrest. Resuscitation 1994;27:47–54. [2] Wollenek G, Honarwar N, Golej J, et al. Cold water submersion and cardiac arrest in treatment of severe hypothermia with cardiopulmonary bypass. Resuscitation 2002;52:255–63. [3] Bierens JJ, Knape JT, Gelissen HP. Drowning. Curr Opin Crit Care 2002;8:578–86. [4] Ruttmann E, Weissenbacher A, Ulmer H, et al. Prolonged extracorporeal membrane oxygenation-assisted support provides improved survival in hypothermic patients with cardiocirculatory arrest. J Thorac Cardiovasc Surg 2007;134: 594–600. [5] Silfvast T, Pettila V. Outcome from severe accidental hypothermia in Southern Finland—a 10-year review. Resuscitation 2003;59:285–90. [6] Walpoth BH, Walpoth-Aslan BN, Mattle HP, et al. Outcome of survivors of accidental deep hypothermia and circulatory arrest treated with extracorporeal blood warming. N Engl J Med 1997;337:1500–5. [7] Morita S, Inokuchi S, Yamagiwa T, et al. Efficacy of portable and percutaneous cardiopulmonary bypass rewarming versus that of conventional internal rewarming for patients with accidental deep hypothermia. Crit Care Med 2011;39:1064–8. [8] Farstad M, Andersen KS, Koller ME, et al. Rewarming from accidental hypothermia by extracorporeal circulation: a retrospective study. Eur J Cardiothorac Surg 2001;20:58–64. [9] Larach MG. Accidental hypothermia. Lancet 1995;345:493–8. [10] Vanden Hoek TL, Morrison LJ, Shuster M, et al. Part 12: cardiac arrest in special situations: 2010 American Heart Association guidelines for cardiopulmonary
[12] [13] [14]
[15] [16] [17] [18] [19]
[20] [21]
resuscitation and emergency cardiovascular care. Circulation 2010;122(Suppl. 3): S829–61. Soar J, Perkins GD, Abbas G, et al. European Resuscitation Council guidelines for resuscitation 2010 section 8: cardiac arrest in special circumstances: electrolyte abnormalities, poisoning, drowning, accidental hypothermia, hyperthermia, asthma, anaphylaxis, cardiac surgery, trauma, pregnancy, electrocution. Resuscitation 2010;81:1400–33. Danzl DF, Pozos RS. Accidental hypothermia. N Engl J Med 1994;331:1756–60. Althaus U, Aeberhard P, Shupbach P, et al. Management of profound accidental hypothermia with cardiorespiratory arrest. Ann Surg 1982;195:492–5. Oberhammer R, Beikircher W, Hormann C, et al. Full recovery of an avalanche victim with profound hypothermia and prolonged cardiac arrest treated by extracorporeal re-warming. Resuscitation 2008;76:474–80. Gilbert M, Busund R, Skagseth A, et al. Resuscitation from accidental hypothermia of 13.7 degrees C with circulatory arrest. Lancet 2000;355:375–6. Dobson JAR, Burgess JJ. Resuscitation of severe hypothermia by extracorporeal rewarming in a child. J Trauma 1996;40:483–5. Hauty M, Esrig BC, Hill JG, Long WB. Prognostic factors in severe accidental hypothermia: experience from the Mt Hood tragedy. J Trauma 1987;27:1107–12. Schaller M-D, Fischer A, Perret C. Hyperkalemia. A prognostic factor during acute severe hypothermia. JAMA 1990;264:1842–5. van der Ploeg GJ, Goslings JC, Walpoth BH, et al. Accidental hypothermia: rewarming treatments, complications and outcomes from one university medical centre. Resuscitation 2010;81:1550–5. Douglas JAB, Hermann B, Jeff B, et al. Accidental hypothermia. N Engl J Med 2012;367:1930–8. Schaller MD, Fischer AP, Perret CH. Hyperkalemia. A prognostic factor during acute severe hypothermia. J Am Med Assoc 1990;264:1842–5.
