Predicting neurological outcome following cardiac arrest

doi:10.1016/j.jns.2007.04.041

Journal of the Neurological Sciences

Volume 261, Issues 1–2, 15 October 2007, Pages 108-117

https://doi.org/10.1016/j.jns.2007.04.041 Get rights and content

Abstract

Because a large number of patients will suffer cardiac arrest each year, physicians must place attention on improving care for patients in the post-resuscitative setting. Part of this effort requires setting realistic goals based on patients' potential for recovery. Recovery from cardiac arrest often depends on the extent of anoxic brain injury, and for this reason primary teams consult neurologists to offer insight into potential for awakening from post-arrest coma. In doing so, neurologists inform a decision with legal, social and ethical implications. Though inapplicable without preparation at the time of cardiac arrest, the four principles of medical ethics have a direct impact on decision making during the post-resuscitative period. A review of the literature reveals that physical examination, electrophysiology, radiology, and biochemical markers can prove useful in estimating a patient's chances for neurological recovery from cardiac arrest. These factors most reliably predict poor outcome, but do so with high specificity. However, the role of the neurology consultant must change to include guidance on strategies of neuroprotection. Aggressive efforts directed towards neuroprotection may change predictions for outcomes after cardiac arrest in the future.

Introduction

Approximately 300,000 cardiac arrests occur each year in the United States. If arrest occurs within the hospital setting, population-based research indicates that 37% of patients survive to discharge, in stark contrast to the mere 1% to 9% of patients who suffer cardiac arrest out of the hospital and live long enough to reach an emergency room [1], [2]. A distinction between sudden cardiac death and cardiac arrest is in large part due to the development of cardiopulmonary resuscitation (CPR), first described by Kouwenhoven in 1960 [3] and the introduction of the intensive care unit shortly thereafter [4]. While allowing physicians to pursue their most elemental role as life savers, medical advances during the half century since the introduction of cardiopulmonary resuscitation (CPR) and the intensive care unit (ICU) setting have come at the cost of a new class of extremely challenging patient: the patient with global anoxic brain injury due to extended periods of catastrophic cardiac failure [5].

Sadly, a half-century after the development of CPR and the ICU, achievements in survival after cardiac arrest have not been accompanied by improvements in neurological recovery. Even for patients with inpatient cardiac arrest, estimates have been as low as 10% to 30% recovering to independent living [6]. The American Heart Association now recognizes CNS injury as a crucial element in resuscitating patients from cardiac arrest. Starting in 2000, the Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care drew specific attention to this issue: “The cerebral cortex, the tissue most susceptible to hypoxia, is irreversibly damaged, resulting in death or severe neurological damage. The need to preserve cerebral vitality must be stressed in research endeavors and in practical interventions. The term cardiopulmonary–cerebral resuscitation has been used to further emphasize this need [7].”

Section snippets

Pathophysiology

Under normothermic conditions, brain oxygen is depleted within 20 seconds. CNS neurons have only sufficient stores of glucose and adenosine triphosphate for 5 minutes without perfusion [8]. Standard CPR produces 1.3 l/min of forward blood flow and 25 mm Hg of perfusion pressure [9]. This falls well below the “luxury perfusion” buffer afforded by the CNS vascular bed and the 15 ml/100 g min regional cerebral blood flow (rCBF) found to be the minimum by PET in humans [10], meaning that effects of

Decision making in the post-resuscitation setting

Almost 80% of patients who survive cardiac arrest and resuscitation remain in coma for some period of time [14], [15]. During this phase of illness, primary care teams must manage a series of complicated tasks. These include stabilizing the patient, retrieving the patient's advance directives and contacting a decision making proxy for the patient. Next, the proxy must be informed as to the patient's current condition and prognosis because this proxy will need to take a part in determining which

Ethical considerations

As the main goal of the neurology consultant is to determine whether a particular patient has a potential for recovery so low as to make medical care futile, an appreciation of the ethical dimensions of assigning prognosis bears discussion. Medical futility and withdrawal of care are issues charged with legal, cultural, and moral repercussions. More importantly a realistic prognosis proves an invaluable part of grieving and coping for those close to the patient [17]. Theory in medical ethics

Patient history

Incidence of cardiac arrest increases with age [33], and older age has been associated with poorer survival in studies of out-of-hospital arrest [34]. Conversely, of the 774 patients enrolled in the Brain Resuscitation Clinical Trials I and II, age was not a predictor for the 27% who finished the trial with good neurologic outcome, even for those patients older than 80 years.

Patient factors prior to admission have a direct impact on overall patient mortality rather than specific impact on

Multimodal approaches

Approaches to determining prognosis that combine information from multiple areas of investigation improve confidence in predictions. In particular, while many single factors have good PPV in identifying patients with poor prognosis, most have poor sensitivity, and patients with normal results die or lapse into PVS. In one of the first multimodal studies, Bassetti et al. studied patients in coma more than 6 h after ROSC and measured GCS, EEG, SSEP, ionized calcium, and NSE. Combination of GCS < 8

Expanded role of consultation

Treatment of patients after cardiac arrest has not focused on post-resuscitative encephalopathy [79], but several clinical trials have sought to minimize CNS injury due to the reperfusion syndrome. To this end, thiopental [80], glucocorticoids [81], calcium channel blockers [82], [83], barbiturates [5], magnesium and benzodiazepines [84], and vasopressors [85], [86] have all been investigated in large clinical trial but none has shown an effect in improving neurologic outcome after cardiac

Conclusion

A large number of patients suffer cardiac arrest and survive resuscitation each year. By definition, these patients have suffered anoxic brain injury, but the extent or impact of this injury is not immediately predictable. However, an understanding of a patient's neurological injury is of central importance. All decisions, medical and social, made on behalf of the patient are based on the patient's ability to recover. For this reason, much emphasis in the literature of post-resuscitative care

References (97)

R.J. Myerburg et al.
Clinical, electrophysiologic and hemodynamic profile of patients resuscitated from prehospital cardiac arrest
Am J Med
(1980)
E. Jørgensen et al.
The natural course of neurological recovery following cardiopulmonary resuscitation
Resuscitation
(1998)
J.E. Heffner et al.
End-of-life care preferences of patients enrolled in cardiovascular rehabilitation programs
Chest
(2000)
M. Mohr et al.
Ethical aspects of resuscitation
Br J Anaesth
(1997)
H. Gage et al.
Health system costs of in-hospital cardiac arrest
Resuscitation
(2002)
D.A. Shewmon et al.
Early prognosis in anoxic coma. Reliability and rationale
Neuro Clin
(1989)
W.A. Knaus et al.
The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults
Chest
(1991)
E. Edgren et al.
Assessment of neurological prognosis in comatose survivors of cardiac arrest
Lancet
(1994)
E.O. Jorgensen
Course of neurological recovery and cerebral prognostic signs during cardio-pulmonary resuscitation
Resuscitation
(1997)
E.G. Zandbergen et al.
Systematic review of early prediction of poor outcome in anoxic–ischaemic coma
Lancet
(1998)

A.H. Association

Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care

(2000)

P. Safar et al.

Cerebral blood flow promotion after prolonged cardiac arrest

Crit Care Med

(2000)

C.F.M.D.P. Babbs

CPR techniques that combine chest and abdominal compression and decompression: hemodynamic insights from a spreadsheet model

Circulation

(1999)

W.J. Powers et al.

Cerebral blood flow and cerebral metabolic rate of oxygen requirements for cerebral function and viability in humans

J Cereb Blood Flow Metab

(1985)

B.V. Maramattom et al.

Postresuscitation encephalopathy: current views, management, and prognostication

Neurologist

(2005)

O. Miyamoto et al.

Hypoxia, hyperoxia, ischemia, and brain necrosis

Neurology

(2000)

R.G. Geocadin et al.

Neurologic prognosis and withdrawal of life support after resuscitation from cardiac arrest

Neurology

(2006)

Anonymous

Medical futility in end-of-life care: report of the Council on Ethical and Judicial Affairs [see comment]

JAMA

(1999)

S.E. Bedell et al.

Survival after cardiopulmonary resuscitation in the hospital

N Engl J Med

(1983)

S.A. Mayer et al.

Withdrawal of life support in the neurological intensive care unit [see comment]

Neurology

(1999)

M. Ardagh

Futility has no utility in resuscitation medicine

J Med Ethics

(2000)

J. Addington-Hall et al.

Who should measure quality of life?

BMJ

(2001)

B. Farsides et al.

Is there such a thing as a life not worth living?

BMJ

(2001)

A.J. Carr et al.

Measuring quality of life: is quality of life determined by expectations or experience?

BMJ

(2001)

J. Cassell et al.

Surgeons, intensivists, and the covenant of care: administrative models and values affecting care at the end of life

Crit Care Med

(2003)

G.B. Young

Ethics in the intensive care unit with emphasis on medical futility in comatose survivors of cardiac arrest

J Clin Neurophysiol

(2000)

Anonymous

Medical aspects of the persistent vegetative state (1). The Multi-Society Task Force on PVS

N Engl J Med

(1994)

Anonymous

Medical aspects of the persistent vegetative state (2). The Multi-Society Task Force on PVS

N Engl J Med

(1994)

G. Rocker et al.

Clinician predictions of intensive care unit mortality

Crit Care Med

(2004)

S. Frick et al.

Medical futility: predicting outcome of intensive care unit patients by nurses and doctors—a prospective comparative study

Crit Care Med

(2003)

J. Engdahl et al.

The epidemiology of out-of-hospital ‘sudden’ cardiac arrest

Resuscitation

(2002)

T.D. Rea

Temporal trends in sudden cardiac arrest: a 25-year emergency medical services perspective

Circulation

(2003)

K. Berek et al.

Early determination of neurological outcome after prehospital cardiopulmonary resuscitation

Stroke

(1995)

Cited by (31)

Subarachnoid hemorrhage after resuscitation from out-of-hospital cardiac arrest
2014, Journal of Stroke and Cerebrovascular Diseases
Citation Excerpt :
Clinical courses of 6 patients with good outcomes reported by various groups of investigators are summarized in Table 4. First, ECGs revealed VF waveforms in most cases, and the restoration of spontaneous circulation occurred within a few minutes in almost all cases, with the cardiac arrest being either an in-hospital or in-ambulance witnessed event or an event lasting no more than a few minutes followed by ROSC before arrival at the hospital.31,32 The treatment procedure for aneurysms was clipping upon craniotomy in 4 of the 7 cases reported and embolization by coiling in 3 recently reported cases.
This study was undertaken to retrospectively investigate clinical features of subarachnoid hemorrhage (SAH) with cardiopulmonary arrest in patients achieving return of spontaneous circulation (ROSC) in order to explore the possibility of long-term survival.
Of 143 SAH patients with cardiopulmonary arrest in our hospital between April 2004 and June 2012, data on 59 (41%) patients who attained ROSC were analyzed to determine the predictive factors for neurologic recovery and outcome. Recovery of brainstem reflexes and improvement of Glasgow Coma Scale (GCS) motor score were noted (postresuscitation neurologic restorative assessment, grade I) in 5, and 2 of these patients survived.
By-grade analysis of patient background characteristics revealed a significantly shorter duration of cardiac arrest (P = .001) and a significantly smaller adrenaline dose (P = .011) for grade I patients. A logistic analysis of 1-week survival data revealed significant differences in duration of cardiac arrest (P = .022) and adrenaline dose (P = .019), with odds ratios of 0.89 and 0.25, respectively. Cox regression analysis of mortality data revealed significant differences in the duration of cardiac arrest (P = .012), adrenaline dose (P < .0001), and location of ROSC (P = .016), with hazard ratios of 1.03, 1.43, and 1.98, respectively.
Cardiac arrest caused by SAH is a disease state with a grave prognosis, but there is the possibility of a good survival outcome when the administration of a small dose of adrenaline results in the rapid recovery of brainstem reflexes.
Auditory-evoked potentials during coma: Do they improve our prediction of awakening in comatose patients?
2014, Journal of Critical Care
Citation Excerpt :
Currently, no practical tool that accurately predicts awakening is available. Some studies have suggested that patients who show higher-order cortical processing of auditory stimuli during the course of coma may be more likely to make a good recovery [7,8]. Auditory event-related potentials (ERPs) in particular, which assess the subject's ability to discriminate different acoustic stimuli [4,5,9], can identify those comatose patients whose preattentive auditory memory remains active [8].
The mismatch negativity (MMN), an auditory event-related potential, has been identified as a good indicator of recovery of consciousness during coma. We explored the predictive value of the MMN and other auditory-evoked potentials including brainstem and middle-latency potentials for predicting awakening in comatose patients after cardiac arrest or cardiogenic shock.
Auditory brainstem, middle-latency (Pa wave), and event-related potentials (N100 and MMN waves) were recorded in 17 comatose patients and 9 surgical patients matched by age and coronary artery disease. Comatose patients were followed up daily to determine recovery of consciousness and classified as awakened and nonawakened.
Among the auditory-evoked potentials, the presence or absence of MMN best discriminated between patients who awakened or those who did not. Mismatch negativity was present during coma in all patients who awakened (7/7) and in 2 of those (2/10) who did not awaken. In patients who awakened and in whom MMN was detected, 3 of those awakened between 2 and 3 days and 4 between 9 and 21 days after evoked potential examination. All awakened patients had intact N100 waves and identifiable brainstem and middle-latency waves. In nonawakened patients, N100 and Pa waves were detected in 5 cases (50%) and brainstem waves in 9 (90%).
The MMN is a good predictor of awakening in comatose patients after cardiac arrest and cardiogenic shock and can be measured days before awakening encouraging ongoing life support.
Blood pH is a useful indicator for initiation of therapeutic hypothermia in the early phase of resuscitation after comatose cardiac arrest: A retrospective study
2013, Journal of Emergency Medicine
Therapeutic hypothermia (TH) is one of the key treatments after cardiac arrest (CA). Selection of post-CA patients for TH remains problematic, as there are no clinically validated tools to determine who might benefit from the therapy.
The aim of this study was to investigate retrospectively whether laboratory findings or other patient data obtained during the early phase of hospital admission could be correlated with neurological outcome after TH in comatose survivors of CA.
Medical charts of witnessed CA patients admitted between June 2003 and July 2009 who were treated with TH were reviewed retrospectively. The subjects were grouped based on their cerebral performance category (CPC) 6 months after CA, as either good recovery (GR) for CPC 1–2 or non-good recovery (non-GR) for CPC 3–5. The following well-known determinants of outcome obtained during the early phase of hospital admission were evaluated: age, gender, body mass index, cardiac origin, presence of ventricular fibrillation (VF), time from collapse to cardiopulmonary resuscitation, time from collapse to return of spontaneous circulation, body temperature, arterial blood gases, and blood test results.
We analyzed a total of 50 (25 GR and 25 non-GR) patients. Multivariate logistic analysis showed that initial heart rhythm and pH levels were significantly higher in the GR group than in the non-GR group (ventricular tachycardia/VF rate: p = 0.055, 95% confidence interval [CI] 0.768–84.272, odds ratio [OR] 8.047; pH: 7.155 ± 0.139 vs. 6.895 ± 0.100, respectively, p < 0.001, 95% CI 1.838–25.827; OR 6.89).
These results imply that in addition to initial heart rhythm, pH level may be a good candidate for neurological outcome predictor even though previous research has found no correlation between initial pH value and neurological outcome.
Haemodynamic variables and functional outcome in hypothermic patients following out-of-hospital cardiac arrest
2013, Resuscitation
To evaluate the association between haemodynamic variables during the first 24 h after intensive care unit (ICU) admission and neurological outcome in out-of-hospital cardiac arrest (OHCA) victims undergoing therapeutic hypothermia.
In a multi-disciplinary ICU, records were reviewed for comatose OHCA patients undergoing therapeutic hypothermia. The hourly variable time integral of haemodynamic variables during the first 24 h after admission was calculated. Neurologic outcome was assessed at day 28 and graded as favourable or adverse based on the Cerebral Performance Category of 1–2 and 3–5. Bi- and multivariate regression models adjusted for confounding variables were used to evaluate the association between haemodynamic variables and functional outcome.
67/134 patients (50%) were classified as having favourable outcome. Patients with adverse outcome had a higher mean heart rate (73 [62–86] vs. 66 [60–78] bpm; p = 0.04) and received noradrenaline more frequently (n = 17 [25.4%] vs. n = 9 [6%]; p = 0.02) and at a higher dosage (128 [56–1004] vs. 13 [2–162] μg h⁻¹; p = 0.03) than patients with favourable outcome. The mean perfusion pressure (mean arterial blood pressure minus central venous blood pressure) (OR = 1.001, 95% CI = 1–1.003; p = 0.04) and cardiac index time integral (OR = 1.055, 95% CI = 1.003–1.109; p = 0.04) were independently associated with adverse outcome at day 28.
Mean perfusion pressure and cardiac index during the first 24 h after ICU admission were weakly associated with neurological outcome in an OHCA population undergoing therapeutic hypothermia. Further studies need to elucidate whether norepinephrine-induced increases in perfusion pressure and cardiac index may contribute to adverse neurologic outcome following OHCA.
Severe brain damage after punitive training technique with a choke chain collar in a German shepherd dog
2013, Journal of Veterinary Behavior: Clinical Applications and Research
Citation Excerpt :
Furthermore, choke chains and collars can cause mechanical or ischemical damage to the larynx, esophagus, thyroid, or trachea (Brammeier et al., 2006). Cerebral ischemia may also occur after general anesthesia (Jurk et al., 2001; Stiles et al., 2012) during birth (Levene et al., 1986; Dickey et al., 2011), vascular thrombosis, asphyxia, and cardiac arrest (Püttgen and Geocadin, 2007; Choi et al., 2010). It is reported in humans and animals such as cats or dogs (Panarello et al., 2004; Timm et al., 2008; Choi et al., 2010).
The features of severe ischemic brain damage after strangulation by the owner of a 1-year-old German shepherd dog are described. The dog was disciplined by the owner during training by holding the dog off the ground by his choke chain collar. At first, the dog behaved normally, but he became increasingly ataxic and started circling to the left and showed reduced consciousness. The neurological examination revealed severe disorientation, left lateral pleurothotonus, and circling. The neurological findings were consistent with a multifocal brain lesion. A magnetic resonance imaging scan was performed and showed changes in the T2- and diffusion-weighted images, consistent with severe cerebral edema resulting from ischemia. Because of the severity of the clinical features, the dog was later euthanized. To the author's knowledge, this is the first report of a severe brain ischemia after strangulation in a dog.
The changes of brain water diffusion and blood flow on diffusion-weighted and perfusion-weighted imaging in a canine model of cardiac arrest
2012, Resuscitation
Citation Excerpt :
CA continues to be a major public health challenge in USA.5 Until recently, achievements in survival after CA have not been accompanied by improvements in neurological recovery.6 Mild hypothermia has been shown to increase survival rate in the post-cardiac arrest setting3; few other specific therapeutic strategies are available to improve survival and neurologic outcome of patients requiring CPR after CA.7,8
To study the changes of brain water diffusion and cerebral haemodynamics of cortical areas using magnetic resonance imaging (MRI) in canine models of cardiac arrest (CA) and restoration of spontaneous circulation (ROSC). The secondary study objective was to evaluate whether MRI can be used to observe haemodynamic disorders in brain microcirculation.
CA was induced in six beagle dogs using electrical stimulation followed by resuscitation to spontaneous circulation 3 min later. The dogs were scanned using MRI for echo planar, diffusion-weighted imaging (DWI) and perfusion-weighted imaging (PWI) with injection of Gd-diethylene triamine pentaacetic acid (DTPA) prior to induction of CA and 3 days after ROSC. The arterial blood pressure, unilateral common carotid artery flow and intracranial microcirculation were recorded.
All dogs successfully underwent electric-induced ventricular fibrillation which lasted 3 min and were resuscitated to maintain blood pressure stability. Serial MRI scans found that cerebral blood flow (RCBF) decreased in day 1 after ROSC and returned to baseline on day 3. Apparent diffusion coefficient (ADC), however, decreased on day 1 and remained lower than baseline on day 3, with 765.8 ± 82.5 × 10⁻⁶ mm² s⁻¹ on day 1 and 770.4 ± 59.4 × 10⁻⁶ mm² s⁻¹ on day 3 comparing to 855.8 ± 43.4 × 10⁻⁶ mm² s⁻¹ on baseline.
These data provide the evidence that early MRI can be used to observe acute haemodynamic disorders in brain microcirculation in a canine model of cardiac arrest.

View all citing articles on Scopus

View full text

Predicting neurological outcome following cardiac arrest

Abstract

Introduction

Section snippets

Pathophysiology

Decision making in the post-resuscitation setting

Ethical considerations

Patient history

Multimodal approaches

Expanded role of consultation

Conclusion

Am J Med

Resuscitation

Chest

Br J Anaesth

Resuscitation

Neuro Clin

Chest

Lancet

Resuscitation

Lancet

Resuscitation

Resuscitation

Clin Neurophysiol

Resuscitation

Resuscitation

Crit Care Clin

Lancet

Chest

Resuscitation

Experiences from treatment of out-of-hospital cardiac arrest during 17 years in Goteborg

Eur Heart J

Improving survival from sudden cardiac arrest: the role of the automated external defibrillator

JAMA

Closed-chest cardiac massage

JAMA

Design of a recovery room and intensive care unit

Anesthesiology

Managing out-of-hospital cardiac arrest survivors: 1. Neurological perspective

Heart

Is this patient dead, vegetative, or severely neurologically impaired?: assessing outcome for comatose survivors of cardiac arrest

JAMA

Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care

Cerebral blood flow promotion after prolonged cardiac arrest

Crit Care Med

CPR techniques that combine chest and abdominal compression and decompression: hemodynamic insights from a spreadsheet model

Circulation

Cerebral blood flow and cerebral metabolic rate of oxygen requirements for cerebral function and viability in humans

J Cereb Blood Flow Metab

Postresuscitation encephalopathy: current views, management, and prognostication

Neurologist

Hypoxia, hyperoxia, ischemia, and brain necrosis

Neurology

Neurologic prognosis and withdrawal of life support after resuscitation from cardiac arrest

Neurology

Medical futility in end-of-life care: report of the Council on Ethical and Judicial Affairs [see comment]

JAMA

Survival after cardiopulmonary resuscitation in the hospital

N Engl J Med

Withdrawal of life support in the neurological intensive care unit [see comment]

Neurology

Futility has no utility in resuscitation medicine

J Med Ethics

Who should measure quality of life?

BMJ

Is there such a thing as a life not worth living?

BMJ

Measuring quality of life: is quality of life determined by expectations or experience?

BMJ

Surgeons, intensivists, and the covenant of care: administrative models and values affecting care at the end of life

Crit Care Med

Ethics in the intensive care unit with emphasis on medical futility in comatose survivors of cardiac arrest

J Clin Neurophysiol

Medical aspects of the persistent vegetative state (1). The Multi-Society Task Force on PVS

N Engl J Med

Medical aspects of the persistent vegetative state (2). The Multi-Society Task Force on PVS

N Engl J Med

Clinician predictions of intensive care unit mortality

Crit Care Med

Medical futility: predicting outcome of intensive care unit patients by nurses and doctors—a prospective comparative study