Article
The Outcome and Predictors of Mortality in Patients Therapeutically Cooled Postcardiac Arrest
Journal of Intensive Care Medicine 1-8 ª The Author(s) 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0885066614566792 jic.sagepub.com
John Botha, FCICM1,2, Yvette O’Brien, MBBS1, Saada Malouf, MBBS1, Elizabeth Cole, MBBS1, Erum Sahid Ansari, MBBS1, Cameron Green, Msc1, and Ravindranath Tiruvoipati, FICIM1,2
Abstract Purpose: To review the outcomes of patients postcardiac arrest admitted to a metropolitan intensive care unit (ICU) where therapeutic hypothermia is practiced. Materials and Methods: Patients admitted from 2004 to 2012 were reviewed. The management protocol included cooling to 33 C for 24 hours. The primary outcome assessed was hospital mortality. Secondary outcome measures included mortality in patients admitted to ICU after in-hospital cardiac arrest (IHCA) when compared to those with out-of-hospital cardiac arrest (OHCA) and to review initial cardiac rhythm as an indicator of mortality. Results: A total of 330 patients were included. The overall hospital mortality was 58.1%. Hospital mortality was significantly higher in patients who had OHCA when compared to IHCA (62.5% vs 51%; P ¼ .04). Patients who had asystole and pulseless electrical activity (PEA) had a higher mortality when compared to ventricular tachycardia/ventricular fibrillation (VT/VF) arrest (81.7% vs 67.8% vs 41.9%, respectively; P < .01). Conclusion: Patients admitted to ICU postcardiac arrest after therapeutic cooling have a high mortality. An initial rhythm of VT/VF confers a mortality benefit when compared to asystole and PEA. Keywords cardiac arrest, hypothermia, ventricular fibrillation, asystole, pulseless electrical activity
Introduction Cardiac arrest is a common cause of morbidity and mortality with an estimated annual incidence of 50 to 150 per 100 000 person-years for out-of-hospital cardiac arrest (OHCA) and 1 to 5/1000 admissions for in-hospital cardiac arrest (IHCA). The outcome of patients admitted with cardiac arrest may vary and numerous factors may impact on survival. Two randomized controlled trials published in 2002 compared therapeutic hypothermia (TH), (32 C-34 C for 12-24 hours) with standard treatment. In these trials, the presumed cause of the cardiac arrest was a cardiac cause with an initial shockable rhythm. Both of these trials showed an improvement in neurologic function and survival with TH.1,2 Furthermore, this treatment was found to be safe, with a nonsignificant trend toward an increased incidence of sepsis, the main concern in patients treated with TH (32 C-34 C). Therapeutic hypothermia is widely practiced and is recommended in international guidelines. Furthermore, this strategy is often used as therapy in postcardiac arrest patients with the arrest attributed to a nonshockable rhythm.3 Despite widespread support for TH, some authors have debated the benefit of this intervention.4-6 There is also uncertainty regarding the development of fever in patients after cardiac arrest, as fever
has been associated with a poorer outcome.7-9 A recent publication investigated the benefits and harms of 2 targeted temperature regimens intended to prevent fever in a broader population of patients with cardiac arrest than previously studied.10 This large European trial concluded that in unconscious survivors of OHCA of presumed cardiac cause, hypothermia at a targeted temperature of 33 did not confer a benefit as compared with a targeted temperature of 36 . These recent data compound the difficult therapeutic choices physicians have when treating patients with OHCA. Furthermore, the methods of inducing TH are evolving and it remains unclear how
1 Department of Intensive Care Medicine, Frankston Hospital, Frankston, Victoria, Australia 2 Faculty of Medicine, Nursing and Health Sciences, School of Public Health, Monash University, Melbourne, Victoria, Australia
Received June 8, 2014, and in revised form October 16, 2014. Accepted for publication November 10, 2014. Corresponding Author: Ravindranath Tiruvoipati, Department of Intensive Care Medicine, Frankston Hospital, Frankston, Victoria 3199, Australia. Email: [email protected]
Downloaded from jic.sagepub.com at University of New England on April 21, 2015
2
Journal of Intensive Care Medicine
50 45 40 35 30 25 20 15 10
0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127 133 139 145 151 157 163 169 175 181
5
Figure 1. Histogram of death by days after hospitalization.
quickly and where postcardiac arrest patients should be offered TH as a therapy.11 Our hospital services an elderly population with a significant number of patients admitted with OHCA. Since 2003, variable methods have been used to initiate TH. A review of patients admitted to our intensive care unit (ICU) who were treated with TH after cardiac arrest was conducted.
care was considered after continuous clinical assessment and neurological evaluation by the neurologists. There was no standardized approach to withdrawal of care, and somatosensory-evoked potentials (SSEPs) were not performed. Electroencephalography (EEG) and neuroimaging were performed if requested by the attending neurologist or intensivist.
Data Collection Methods A retrospective cohort study of prospectively collected data of patients who were admitted to the ICU of the Frankston Hospital post-OHCA and IHCA of presumed cardiac origin between 2004 and 2012. The arrest was considered cardiac in origin if the history and clinical presentation as assessed by the emergency physician and cardiologist in attendance made this clinical assessment. Neuroimaging or computerized tomographic pulmonary angiography was not routinely ordered unless there was a strong clinical suspicion that the arrest was as a consequence of a neurological event or pulmonary embolism. The Frankston Hospital is a major metropolitan Melbourne Hospital that provides critical care services to the city of Frankston and the Mornington Peninsula. This ICU is a 13-bed level 3 mixed medical and surgical facilities that admits about 1100 patients annually. The ICU has a broad case mix but does not admit neurosurgical or cardiothoracic cases. Ethical approval was obtained from the Peninsula Health Research and Ethics Committee prior to commencement of the study. Patients admitted to our ICU between January 2004 and September 2012 postcardiac arrest with successful resuscitation were included in the analysis. All patients referred to the ICU postresuscitation were defined as successfully resuscitated and included in the analysis irrespective of the time to return of spontaneous circulation (ROSC). Patients were excluded if their initial cardiac arrest occurred in the ICU. Withdrawal of
Data were collected from ICU and hospital databases, paperbased and online medical records, and online pathology services. Apart from ICU and hospital mortality data, the following information was collected: location of arrest, time to initiation of cardiopulmonary resuscitation (CPR), duration of CPR, time to ROSC, initial rhythm at time of arrest, demographic data (age, sex, and comorbidities), physiological data (blood pressure, Acute Physiology And Chronic Health Evaluation III [APACHE III] score, PaO2, PaCO2, fraction of inspired oxygen [FiO2], pH, heart rate and respiratory rate), and laboratory variables (sodium, HCO3, albumin, and hematocrit).
Treatment Received All patients received TH. This was achieved through externally applied icepacks, cold intravenous crystalloid solution, or both therapies from 2004 until June 2011. From July 2011, hypothermia was induced using the Arctic Sun (Medivance, Inc., Louisville, CO) cooling device. This ‘‘cold blanket’’ technology utilizes the application of gel pads infused with cold liquid to the patient’s torso and limbs. The vast majority (>85%) of patients were cooled prior to the use of a dedicated cooling device and it was considered that outcome data comparing these 2 groups would be difficult to interpret. Therapeutic hypothermia was administered for 24 hours from the time of initial cardiac arrest, with temperature measurements recorded via a nasopharyngeal
Downloaded from jic.sagepub.com at University of New England on April 21, 2015
Botha et al
3
Table 1. Comparison of Demographics and Comorbidities. Variable Age (n ¼ 312), median (IQR), years Sex ratio (M:F) Comorbidities n Chronic renal failure, n (%) Chronic obstructive pulmonary disease, n (%) Myocardial infarction, n (%) Diabetes mellitus, n (%) Hypertension, n (%) Illness severity APACHE III score
Total
Died
Discharged
P Value
212:118
71 (56-79) 120:74
67 (53-77) 92:44
.16 .28
286 25 (8.74) 54 (18.8) 56 (19.6) 49 (17.1) 137 (47.9)
171 15 (60) 36 (66.7) 31 (55.4) 29 (59.2) 82 (59.9)
115 10 (40) 18 (33.3) 25 (44.6) 20 (40.8) 55 (40.1)
.98 .25 .45 .92 .98
n Mean (SD)
191 125 (27.9)
134 97.1 (29.1)
The Outcome and Predictors of Mortality in Patients Therapeutically Cooled Postcardiac Arrest
Journal of Intensive Care Medicine 1-8 ª The Author(s) 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0885066614566792 jic.sagepub.com
John Botha, FCICM1,2, Yvette O’Brien, MBBS1, Saada Malouf, MBBS1, Elizabeth Cole, MBBS1, Erum Sahid Ansari, MBBS1, Cameron Green, Msc1, and Ravindranath Tiruvoipati, FICIM1,2
Abstract Purpose: To review the outcomes of patients postcardiac arrest admitted to a metropolitan intensive care unit (ICU) where therapeutic hypothermia is practiced. Materials and Methods: Patients admitted from 2004 to 2012 were reviewed. The management protocol included cooling to 33 C for 24 hours. The primary outcome assessed was hospital mortality. Secondary outcome measures included mortality in patients admitted to ICU after in-hospital cardiac arrest (IHCA) when compared to those with out-of-hospital cardiac arrest (OHCA) and to review initial cardiac rhythm as an indicator of mortality. Results: A total of 330 patients were included. The overall hospital mortality was 58.1%. Hospital mortality was significantly higher in patients who had OHCA when compared to IHCA (62.5% vs 51%; P ¼ .04). Patients who had asystole and pulseless electrical activity (PEA) had a higher mortality when compared to ventricular tachycardia/ventricular fibrillation (VT/VF) arrest (81.7% vs 67.8% vs 41.9%, respectively; P < .01). Conclusion: Patients admitted to ICU postcardiac arrest after therapeutic cooling have a high mortality. An initial rhythm of VT/VF confers a mortality benefit when compared to asystole and PEA. Keywords cardiac arrest, hypothermia, ventricular fibrillation, asystole, pulseless electrical activity
Introduction Cardiac arrest is a common cause of morbidity and mortality with an estimated annual incidence of 50 to 150 per 100 000 person-years for out-of-hospital cardiac arrest (OHCA) and 1 to 5/1000 admissions for in-hospital cardiac arrest (IHCA). The outcome of patients admitted with cardiac arrest may vary and numerous factors may impact on survival. Two randomized controlled trials published in 2002 compared therapeutic hypothermia (TH), (32 C-34 C for 12-24 hours) with standard treatment. In these trials, the presumed cause of the cardiac arrest was a cardiac cause with an initial shockable rhythm. Both of these trials showed an improvement in neurologic function and survival with TH.1,2 Furthermore, this treatment was found to be safe, with a nonsignificant trend toward an increased incidence of sepsis, the main concern in patients treated with TH (32 C-34 C). Therapeutic hypothermia is widely practiced and is recommended in international guidelines. Furthermore, this strategy is often used as therapy in postcardiac arrest patients with the arrest attributed to a nonshockable rhythm.3 Despite widespread support for TH, some authors have debated the benefit of this intervention.4-6 There is also uncertainty regarding the development of fever in patients after cardiac arrest, as fever
has been associated with a poorer outcome.7-9 A recent publication investigated the benefits and harms of 2 targeted temperature regimens intended to prevent fever in a broader population of patients with cardiac arrest than previously studied.10 This large European trial concluded that in unconscious survivors of OHCA of presumed cardiac cause, hypothermia at a targeted temperature of 33 did not confer a benefit as compared with a targeted temperature of 36 . These recent data compound the difficult therapeutic choices physicians have when treating patients with OHCA. Furthermore, the methods of inducing TH are evolving and it remains unclear how
1 Department of Intensive Care Medicine, Frankston Hospital, Frankston, Victoria, Australia 2 Faculty of Medicine, Nursing and Health Sciences, School of Public Health, Monash University, Melbourne, Victoria, Australia
Received June 8, 2014, and in revised form October 16, 2014. Accepted for publication November 10, 2014. Corresponding Author: Ravindranath Tiruvoipati, Department of Intensive Care Medicine, Frankston Hospital, Frankston, Victoria 3199, Australia. Email: [email protected]
Downloaded from jic.sagepub.com at University of New England on April 21, 2015
2
Journal of Intensive Care Medicine
50 45 40 35 30 25 20 15 10
0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127 133 139 145 151 157 163 169 175 181
5
Figure 1. Histogram of death by days after hospitalization.
quickly and where postcardiac arrest patients should be offered TH as a therapy.11 Our hospital services an elderly population with a significant number of patients admitted with OHCA. Since 2003, variable methods have been used to initiate TH. A review of patients admitted to our intensive care unit (ICU) who were treated with TH after cardiac arrest was conducted.
care was considered after continuous clinical assessment and neurological evaluation by the neurologists. There was no standardized approach to withdrawal of care, and somatosensory-evoked potentials (SSEPs) were not performed. Electroencephalography (EEG) and neuroimaging were performed if requested by the attending neurologist or intensivist.
Data Collection Methods A retrospective cohort study of prospectively collected data of patients who were admitted to the ICU of the Frankston Hospital post-OHCA and IHCA of presumed cardiac origin between 2004 and 2012. The arrest was considered cardiac in origin if the history and clinical presentation as assessed by the emergency physician and cardiologist in attendance made this clinical assessment. Neuroimaging or computerized tomographic pulmonary angiography was not routinely ordered unless there was a strong clinical suspicion that the arrest was as a consequence of a neurological event or pulmonary embolism. The Frankston Hospital is a major metropolitan Melbourne Hospital that provides critical care services to the city of Frankston and the Mornington Peninsula. This ICU is a 13-bed level 3 mixed medical and surgical facilities that admits about 1100 patients annually. The ICU has a broad case mix but does not admit neurosurgical or cardiothoracic cases. Ethical approval was obtained from the Peninsula Health Research and Ethics Committee prior to commencement of the study. Patients admitted to our ICU between January 2004 and September 2012 postcardiac arrest with successful resuscitation were included in the analysis. All patients referred to the ICU postresuscitation were defined as successfully resuscitated and included in the analysis irrespective of the time to return of spontaneous circulation (ROSC). Patients were excluded if their initial cardiac arrest occurred in the ICU. Withdrawal of
Data were collected from ICU and hospital databases, paperbased and online medical records, and online pathology services. Apart from ICU and hospital mortality data, the following information was collected: location of arrest, time to initiation of cardiopulmonary resuscitation (CPR), duration of CPR, time to ROSC, initial rhythm at time of arrest, demographic data (age, sex, and comorbidities), physiological data (blood pressure, Acute Physiology And Chronic Health Evaluation III [APACHE III] score, PaO2, PaCO2, fraction of inspired oxygen [FiO2], pH, heart rate and respiratory rate), and laboratory variables (sodium, HCO3, albumin, and hematocrit).
Treatment Received All patients received TH. This was achieved through externally applied icepacks, cold intravenous crystalloid solution, or both therapies from 2004 until June 2011. From July 2011, hypothermia was induced using the Arctic Sun (Medivance, Inc., Louisville, CO) cooling device. This ‘‘cold blanket’’ technology utilizes the application of gel pads infused with cold liquid to the patient’s torso and limbs. The vast majority (>85%) of patients were cooled prior to the use of a dedicated cooling device and it was considered that outcome data comparing these 2 groups would be difficult to interpret. Therapeutic hypothermia was administered for 24 hours from the time of initial cardiac arrest, with temperature measurements recorded via a nasopharyngeal
Downloaded from jic.sagepub.com at University of New England on April 21, 2015
Botha et al
3
Table 1. Comparison of Demographics and Comorbidities. Variable Age (n ¼ 312), median (IQR), years Sex ratio (M:F) Comorbidities n Chronic renal failure, n (%) Chronic obstructive pulmonary disease, n (%) Myocardial infarction, n (%) Diabetes mellitus, n (%) Hypertension, n (%) Illness severity APACHE III score
Total
Died
Discharged
P Value
212:118
71 (56-79) 120:74
67 (53-77) 92:44
.16 .28
286 25 (8.74) 54 (18.8) 56 (19.6) 49 (17.1) 137 (47.9)
171 15 (60) 36 (66.7) 31 (55.4) 29 (59.2) 82 (59.9)
115 10 (40) 18 (33.3) 25 (44.6) 20 (40.8) 55 (40.1)
.98 .25 .45 .92 .98
n Mean (SD)
191 125 (27.9)
134 97.1 (29.1)
