We searched PubMed from Jan 1, 2009, to Jan 5, 2016, with the search terms “cardiac arrest”, “hypoxic-ischemic encephalopathy”, “prognosis”, “outcome”, “prognostication”, “clinical examination”, “myoclonus”, “status myoclonus”, “brainstem reflexes”, “pupillary”, “pupillometry”, “motor reaction”, “pain”, “EEG”, “reactivity”, “background”, “epileptiform”, “status epilepticus”, “seizure”, “SSEP”, “N20”, “N70”, “MMN”, “NSE”, “S100-b”, “brain CT”, “Brain MRI”, and “diffusion”. We further searched
ReviewNeurological prognostication of outcome in patients in coma after cardiac arrest
Introduction
Cardiac arrest is a major health problem,1 with a yearly incidence of about 50–110 per 100 000 people worldwide.2 A number of factors have contributed to an overall increase in cardiac arrest survival and frequency of discharge with good neurological recovery and quality of life over the past decade. These factors include progress in advanced life support, access to emergent coronary angiography, implementation of targeted temperature management (TTM; induced mild hypothermia to 32°C or strict normothermia at 36°C for 24 h),3 optimum support of cerebral and organ perfusion, and prevention of extracerebral systemic insults (eg, hyperglycaemia and infections).4, 5, 6, 7, 8, 9 In line with this improvement, cardiac arrest has become a leading cause of coma and a frequent cause of admission to the intensive care unit. Clinicians involved in the care of adults in a coma after cardiac arrest are confronted with the increasingly optimistic expectations of next of kin and, in the early phase of intensive care management, neurologists are regularly requested to provide predictions of long-term outcome.
Brain dysfunction after cardiac arrest, predominantly resulting from global ischaemia-reperfusion injury, is the main determinant of prognosis. However, additional factors can further alter brain function—for example, sedatives used during TTM and post-cardiac arrest organ dysfunction; these aspects might substantially delay recovery of cerebral function for up to 6 days,10 leaving caregivers and families with an unacceptably long period of uncertainty. In this scenario, despite neurological examination being the first and most important step in the assessment of patients,11 a growing body of clinical evidence shows that the integration of additional methods, consisting of electrophysiological investigations, blood biomarkers of cerebral injury, and brain imaging, improves the accuracy of early (24–72 h) coma prognostication.
Neurological consultation and multimodal prognostic assessment has, nowadays, become an integral part of post-resuscitation patient management. In practice, a complete battery of tests is either not always necessary according to the clinical situation, or not available in every facility; although university medical centres might have optimal resources at their disposal, smaller or peripheral hospitals typically have to rely on fewer resources. In this Review, we critically appraise the scientific literature in this developing field and evaluate each prognostic method, emphasising false-positive rates (FPRs) and positive predictive values of poor and favourable prognosis. Experts in coma prognositication defined outcome by cerebral performance categories (CPCs; CPC 1 back to baseline, CPC 2 moderate impairment, CPC 3 severe impairment, CPC 4 vegetative or comatose, CPC 5: dead).12 Here, for the general neurology audience, we refer to good (CPC 1 or 2) or poor (CPC 3-5) outcome. Of note, however, is that CPC 3 is heterogeneous; depending on the time of assessment and the possible improvement of the patient, it may be related to a favourable outcome. We acknowledge the scarce evidence supporting the use of each prognosticator and the inevitable limitation imposed by the possibility of confirmatory bias, or the so-called self-fulfilling prophecy (ie, if a variable is, a priori, believed to be indicative of poor prognosis and leads to withdrawal of life-sustaining treatment, it will ultimately determine the outcome).13, 14
Our purpose is to provide a general neurology audience with an updated, critical review of available methods for coma prognostication after cardiac arrest in adults, summarise their respective value and potential clinical use, and suggest a stepwise multimodal paradigm, paying specific attention to appropriate timings and combination of prognosticators. Finally, an outlook on promising emerging methods will be offered.
Section snippets
Clinical examination
Neurological examination is an essential component of prognostication because it directly evaluates brain function. Assessment of brainstem reflexes, motor responses to pain, and myoclonus during the first 72 h after arrest represented the standard test before the advent of TTM15 and retains its prognostic value in patients given TTM (at target temperatures of 32–36°C).16 However, these features can be altered by TTM and residual sedation, therefore repeated assessments are often necessary.
Electroencephalography
An electroencephalographical signal is generated by cortical post-synaptic potentials. Electroencephalography is widely available, non-invasive, and inexpensive, and provides real-time investigation of electrical brain activity. It has been used routinely for decades in coma prognostication;27 however, there is increasing interest in the use of EEG in the intensive care unit, which parallels progress in post-cardiac arrest care. In the context of prognostication, EEG is very informative because
Somatosensory evoked potentials
Examination of somatosensory evoked potentials is a valuable complement to EEG, but less widely available. Early-latency evoked potentials have been consistently studied in the setting of post-cardiac arrest coma prognostication, whereas middle-latency responses have received little attention.
Biochemical markers
There is currently great interest in assessing the prognostic value of biomarkers of neuronal injury; however, because of the heterogeneity of patient cohorts, the collection and processing of biological fluids, and the detection methods of markers, the generalisability of much of the available evidence is limited.63 Although a wide range of proteins have been identified as markers of cellular injury, only neuron-specific enolase (NSE; a marker of neuronal damage) and S-100 β (a marker of
Neuroimaging
Brain imaging can help reveal structural alterations and quantify the extent of post-anoxic damage; its application for coma prognostication is quite new, and consequently it is used less frequently than other methods of prognostication. Most studies of the use of imaging in coma prognostication are retrospective and assess the value of imaging as a single method. In several studies, clinicians were not masked to imaging findings, which might have introduced some degree of confirmatory bias.
The optimum prognostic method
Clinical examination is mandatory in every patient, and electrophysiology is strongly recommended. Even if pupillary light reflexes are absent at 72 h, we believe that either EEG or sensory evoked potential recording are necessary to substantiate prognostication of poor recovery. Evidence suggests that EEG might be superior to sensory evoked potential recording for predicting poor outcome because of its higher sensitivity.36 EEG also allows for more accurate forecast of good recovery and higher
Promising new prognostic methods
The need for identification of additional prognosticators derives from two considerations: the potential usefulness of quantitative assessments and the need for robust methods not only for poor but also for good outcome. In view of their exploratory nature, studies of new variables, such as those discussed below, are less hampered by confirmatory bias because they are not yet used for prognostication in clinical practice. The most important barriers to wide applicability of these techniques are
Conclusions
Prognostication after cardiac arrest, which can be seen as an attempt to shed some light into the dark, has become an integral part of post-resuscitation care. Neurological consultation is increasingly requested, and neurologists are confronted by caregivers and families with expectations of highly accurate predictions of outcome. During the past decade, post-cardiac arrest prognostication has progressed towards a multimodal paradigm relying on clinical examination and judicious integration of
Search strategy and selection criteria
References (103)
- et al.
Global incidences of out-of-hospital cardiac arrest and survival rates: systematic review of 67 prospective studies
Resuscitation
(2010) - et al.
Clinical examination for prognostication in comatose cardiac arrest patients
Resuscitation
(2013) - et al.
Predictors of poor neurologic outcome in patients after cardiac arrest treated with hypothermia: a retrospective study
Resuscitation
(2011) - et al.
Neurologic recovery after therapeutic hypothermia in patients with post-cardiac arrest myoclonus
Resuscitation
(2012) - et al.
Predictors of poor neurological outcome in adult comatose survivors of cardiac arrest: a systematic review and meta-analysis. Part 2: patients treated with therapeutic hypothermia
Resuscitation
(2013) - et al.
Interrater variability of EEG interpretation in comatose cardiac arrest patients
Clin Neurophysiol
(2015) - et al.
Deep hypothermic circulatory arrest: I. Effects of cooling on electroencephalogram and evoked potentials
Ann Thorac Surg
(2001) - et al.
Postanoxic alpha (theta) coma: a reappraisal of its prognostic significance
Clin Neurophysiol
(2000) - et al.
Electroencephalogram for prognosis after cardiac arrest
J Crit Care
(2010) - et al.
Stimulus-induced rhythmic, periodic or ictal discharges (SIRPIDs) in comatose survivors of cardiac arrest: characteristics and prognostic value
Clin Neurophysiol
(2013)