EEG in Connection with Coma 53

Feature Neurophysiology REVIEW ARTICLE

Feature Neurophysiology Review article

EEG in connection with coma 53 – 7

BACKGROUND Coma is a dynamic condition that may have various causes. Important chan- John A. Wilson ges may take place rapidly, often with implications for treatment. The purpose of this article [email protected] Section for Clinical Neurophysiology is to provide a brief overview of EEG patterns for comas with various causes, and indicate how EEG can contribute to an assessment of the prognosis for coma patients. Helge J. Nordal Department of Neurology METHOD The article is based on many years of clinical and research experience with EEG Oslo University Hospital in connection with coma states. A self-built reference database was supplemented with searches in PubMed for relevant articles.

RESULTS EEG reveals immediate changes in coma, and can provide early information on cause and prognosis. It is the only diagnostic tool for detecting non-convulsive epileptic MAIN POINTS status. Locked-in syndrome may be overlooked without EEG. Repeated EEG recordings EEG holds a central place in the assessment increase diagnostic certainty and make it possible to monitor coma developments. and treatment of coma INTERPRETATION EEG reflects brain function continuously and therefore holds a key place Repeated EEG recordings may be useful in in the assessment and treatment of coma. connection with coma irrespective of cause Some causes and complications of coma The word «coma» stems from Greek, and standing of electroencephalography (EEG), can only be diagnosed with the aid of EEG means deep sleep. The term is first found in with an outline of the basic physiology and the works of Hippocrates (ca. 460–370 BC) some simple pathophysiological observa- and subsequently of Galen (129–199 AD) tions. We will also present an overview of (1). The term is used again by Thomas Willis EEG findings with known prognostic value (1621–1675). He localised the state of for coma states and an overview of findings lethargy (Gr. lethe = forgetfulness) and argos that have implications for treatment. (Gr. = lazy) in the cortex and that of coma beneath the cortex (i.e. subcortically) (1). Method With von Economo’s (1876– 1931) finding The overview is based on the authors’ exten- of lesions in the hypothalamus of patients sive clinical and research-related experience who died of encephalitis lethargica (2) and of EEG in comatose patients and a discre- the subsequent finding of the ascending reti- tionary selection of relevant articles on EEG cular activating system (3), it became clear in connection with coma. The articles were that coma could be caused by affection of all selected from a self-built literature database these structures (see illustration). and supplemented by searches for relevant Damage to the cortex can affect integration articles in PubMed. The search was conclu- of information, resulting in confusion, hallu- ded in October 2011. cinations or memory problems, without the degree of wakefulness necessarily being af- What does a scalp EEG measure? fected. However, a diffuse cortical affection An electrocardiogram that is recorded by may cause reduced consciousness, depending means of electrodes attached to the skin of on how much of the cortex is affected. An the head is called a scalp EEG. The excita- injury that affects the activating systems in tory pyramidal cells of the cerebral cortex the brain stem, midbrain or thalamus may be are organised in functional columns at right small, but may cause a deep coma. angles to the surface. The electrical activity Today the word coma means a patho- in these cells generates field potentials logical state of suspended consciousness and through interaction with inhibitory interneu- unresponsiveness to external or internal sti- rons. The potentials are an expression of the muli (4). The state can last for anything from total synaptic activity (6). Inhibition and hours to weeks and change to the regaining of excitation in the neurons and the neuronal consciousness with or without sequelae, to networks cause the field potentials to syn- death or to a permanent state where the pati- chronise and oscillate at different frequen- ent may be «awake», but without (or with a cies and voltages (7). The fewer the neurons minimal amount of) awareness, known as a oscillating in small networks, the higher the chronic vegetative state (4). frequency and the lower the voltage. It is An overview of clinical diagnosis, grada- this activity which is measured in an EEG. tion and causes of coma is beyond the scope Some of the activity can be registered by of this review. For this we recommend a very means of electrodes attached to the scalp, good review by Young (5). In the present but intracranial electrodes are required to article we seek to provide a principle under- register the smallest, most rapid oscillations.

Tidsskr Nor Legeforen nr. 1, 2013; 133: 53 – 7 53 REVIEW ARTICLE Feature Neurophysiology

The top right shows an anatomical illustration of the ascending reticular activating system, which consists of two parts: the ventral part, which goes through the hypothalamus, and the dorsal, which goes through the thalamus. There are five EEG printouts: A) EEG for normal sleep. The uppermost blue panel shows the frequency spectrogram (predominant frequencies along the Y-axis and time along the X-axis) through the recording, in this case throughout a 24-hour period. The last half shows the sleep period, with seven REM/NREM cycles. The EEG sequence shows 12 seconds of NREM sleep, with spindles and delta waves. Red is from the left and blue from the right side of the head. B) EEG from patient with acute demyelinising encephalomyelitis in the upper brain stem. EEG shows persistent deep sleep with preserved corticothalamic interaction (compare with printout A), but without reactivity. No alternation is seen between NREM and REM sleep (see uppermost panel with 3.5-hour spectrogram) because the REM-NREM alternation is regulated by the upper brain stem. C) EEG with status myoclonicus from patient who had been resuscitated after cardiac arrest. D) EEG from another patient who had been resuscitated after cardiac arrest. EEG shows low voltage with burst-suppression. E) EEG from patient who had been resuscitated after cardiac arrest. Periodic activity and nonconvulsive status epilepticus are seen here. The spectrogram shows a period of 18 minutes of periodic activity preceded and followed by two shorter sequences of continuous epileptic activity (marked with red). Illustration © Illumedic

54 Tidsskr Nor Legeforen nr. 1, 2013; 133 Feature Neurophysiology REVIEW ARTICLE

During sleep, the interaction of the cortex, also reduce consciousness by affecting the polyradiculitis (Guillain-Barré syndrome) thalamus, hypothalamus and brain stem activating systems in the upper brain stem and critical illness myopathy/neuropathy generates slow oscillations that may involve and midbrain (14). Repeated EEG recor- may also cause a similar condition (21). all the neurons in large networks (8–10). dings may reveal such a development. It is very important clinically that locked- This is of relevance to coma diagnosis. in syndrome is not mistaken for alpha coma. In a scalp EEG recorded in relaxed, awake Thalamus, hypothalamus, EEG is thus essential for distinguishing condition and with closed eyes, activity with mesencephalon and brain stem between these two states. a frequency of 8–13 cycles per second (alpha The corticothalamic system is involved in a activity) predominates over the posterior continuous interaction, which varies over a Ischaemic coma areas of the head. Rapid activity with a fre- 24-hour period, with the mesencephalic and With ischaemic coma, a wide range of chan- quency of 13–30 cycles per second and low hypothalamic sleep regulation systems, with ges can be seen in EEGs (22, 23). The voltage predominates over the anterior areas the frontobasal activating system and with brain’s cortical pyramid cell system is most of the head (beta activity). During light sleep, the reticular activating system (15). The cor- vulnerable to ischaemia, but there may be slower activity of 4–7 cycles per second tical activity is in constant interaction with large individual variations. Hypoxia may predominates (theta activity). During deeper the thalamus (16). This is visible in the EEG have developed rapidly, as with cardiac sleep (non-rapid eye movement sleep, as characteristic patterns during sleep. The arrest, or more slowly, as with respiratory NREM) spindle-shaped sequences of 10–16 patterns can provide an indication of which arrest, with initially preserved cerebral per- cycles per second predominate (sigma or systems are affected. fusion. Different medicines may also affect spindle activity) and slow activity of 1–4 Extensive damage to the upper brain stem the brain’s susceptability to ischaemia. Body cycles per second (delta activity) with higher may appear in the EEG almost like deep phy- temperature may be an important factor, as voltages. Both are modulated by slow back- siological sleep. However, with damage of with drowning accidents or other forms of ground oscillation of 0.2–1 cycle per second. this nature the whole corticothalamic system hypothermia. Some areas of the brain may NREM sleep alternates with REM sleep in and the frontobasal activating system in the also be more susceptible due to arterioscle- periods of about 70–80 minutes (illustration, hypothalamus will be deafferentated. The rosis with reduced perfusion. EEG printout A). sleep-physiological corticothalamic interac- In an EEG this may result in different tion may nonetheless be preserved. The EEG degrees of diffuse cortical functional im- Cortical dysfunction of such a coma does not differ from physiolo- pairment without excitatory elements. Hy- Diffuse cortical dysfunction can reduce gical sleep, but the patient cannot be woken perexcitability, or disinhibition, is also com- synaptic activity, damage networks and (illustration, EEG printout B). mon, however. This may cause epileptiform disturb the balance between excitation and Sometimes the dorsal and ventral activa- changes, varying in type and extent, such as inhibition. This usually causes the propor- ting systems may be affected to different spiked triphasic waves or periodic or persi- tion of slow waves in the EEG to be larger degrees. This is manifested in the EEG as stent epileptic activity (24). Generalised epi- than normal. When excitatory mechanisms different reactivity to sensory stimuli. This leptic activity may take the form of a so-cal- predominate, however, aberrant rhythms may result in either a general activation with led status myoclonicus pattern on the EEG may arise, such as periodic or persistent tri- faster waves or a «paradoxical» activation (the term «generalised periodic epileptiform phasic waves or clear epileptic activity. Pro- with intervals of slow waves. Such activity discharge – GPED» is often used now) nounced cortical hyperexcitability or dis- is often rhythmic, high voltage and pa- (illustration, EEG printout C). Such a condi- inhibition may be expressed in EEGs as roxysmal. A complete absence of activating tion results in persistent myoclonia which do generalised epileptic activity, and in some effect implies more extensive affection. This not respond to anti-epileptic drugs. The con- cases may be an independent cause of redu- is normally a bad sign. dition can only be brought to an end by such ced consciousness. EEG changes of this deep sedation that the EEG becomes «flat», kind are seen largely when coma has hypo- Alpha coma but this does not improve the prognosis. It xic, metabolic or toxic causes. They can also Damage to the pons often affects conscious- should be noted that EEGs may also show be seen with various progressive encephalo- ness. If the pontine tegmentum is also affec- changes of this nature when the patient does pathies that end in coma, such as Creutz- ted, the EEG may show persistent bilateral not have myoclonia. Conversely, a patient feldt-Jakob disease (11, 12). rhythmic activity. The patient is then in a so- may have myoclonia without such changes called alpha coma, theta coma or alpha-theta appearing on the EEG. These variations are Subcortical white matter coma, depending on the dominant rhythm assumed to express the fact that cortical Damage to the subcortical white matter frequency. This type of activity predominates layers are affected in different ways (25). results in functional disturbances in the cor- over the posterior areas of the head. This type tex due to deafferentation of the cortical of «rhythmic coma» can also be seen after Burst-suppression areas concerned. These will then show a ten- hypoxic damage. Alpha and theta rhythms In cases of very extensive anoxic damage dency towards pathological hypersynchro- then predominate over the anterior areas and there will be signs of combined metabolic nisation (9). In EEGs this is seen as intermit- are not reactive to sensory stimuli (17, 18). failure and cortical disinhibition (26). This is tent slow waves of relatively high ampli- called «burst-suppression» pattern. When tude, at frequencies of 3–7 per second. Locked-in syndrome this is due to anoxic damage, it is normally a Broadly speaking, the deeper the lesion, the If the damage is limited to the ventral pons terminal EEG phenomenon (illustration, slower and more high voltage the waves and and does not involve the pontine tegmen- EEG printout D). The termination of cortical the larger the area where they occur. If the tum, the patient may be fully conscious, but synaptic activity results in «flat» EEGs. damage also affects the cortex, the rhythms deafferentated. The patient is then mute and However, it should be noted that deep seda- may become epileptiform. EEG changes of almost completely paralysed. However, ver- tion and deep hypothermia – particularly in this kind may be seen after contusions, intra- tical eye movements and blinking are pre- combination – may also cause both «burst cerebral haemorrhages, infarctions, tumours served and become the only means of com- suppression» and «flat» EEGs. and encephalitis. The degree of alteration of munication (19, 20). This is a very frighte- consciousness depends on how much and ning condition for the patient. In such cases Repeated EEG is useful which region of the cortex is deafferentated the EEG shows normal reactive alpha activ- When coma lasts for hours or days, metabo- (13). Expanding intracranial processes, such ity when the patient is awake. Sleep physio- lic changes take place with corresponding as tumours, haemorrhages or abscesses, may logy is also preserved on the EEG. Extensive changes in the EEG picture. Repeated EEG

Tidsskr Nor Legeforen nr. 1, 2013; 133 55 REVIEW ARTICLE Feature Neurophysiology

examinations may be useful, partly for eva- must be excluded. The cause must be luating a prognosis, partly because epileptic known, and clinical signs of brain death BOX 1 activity that requires treatment may occur must have existed for 12 hours, but this can Synek’s gradation along the way. be reduced to six hours if it has been estab- of EEG changes associated with coma (41) lished that the cause is irreversible (35, 36). Grade 1 Predominantly post-central alpha Nonconvulsive status epilepticus Patients in a chronic vegetative state, i.e. Nonconvulsive status epilepticus exists when without awareness, but with preserved brain activity, some theta activity the EEG shows persistent epileptic activity stem structures that maintain autonomic Grade 2 Predominantly reactive theta without the comatose patient presenting functions, may sometimes lack visible cor- activity motoric symptoms. The diagnosis is used tical activity in a scalp EEG. However, the Grade 3 Predominantly extensive delta when the phenomenon is seen in a patient majority show EEG activity that is very low activity or low-amplitude, irregular with epilepsy who has entered a protracted voltage, slow, irregular and nonreactive to state of reduced consciousness or confusion. sensory stimuli. There is no variation in the and non-reactive delta activity In rare cases, the patient may be in a coma EEG over a 24-hour period even if the pa- Grade 4 Burst-suppression, generalised exclusively as a result of this. The state can tient clinically exhibits daily cycles of epileptic activity (including status only be diagnosed by means of EEG. sleeping and wakefulness. myoclonicus), non-reactive low- The diagnosis can also be used for patients amplitude activity, alpha coma who are in a coma for some reason other than Psychogenic or simulated coma and theta coma known epilepsy. It is important to distinguish In rare cases one is faced with an apparently this from nonconvulsive status epilepticus in unconscious patient with otherwise normal Grade 5 No visible EEG activity with high patients with epilepsy (27–30). clinical neurological status. Should there be sensitivity recording Nonconvulsive status epilepticus may any doubt as to whether the patient is in a occur in comatose patients during the course coma, an EEG can reveal normal wakeful- of the illness and may exacerbate the prog- ness. Thus other, resource-intensive dia- nosis. Such a state should induce attempts at gnostics can be avoided. treatment with a non-sedatory antiepileptic The general rule for all coma-related EEG drug. At best, the patient may awaken, con- Prognostic assessments pathology is that lack of reactivity to sensory currently with improvement in the EEG. When faced with a comatose patient, it is stimuli is a bad sign, unless the coma is due Unfortunately, the EEG finding is often a important to be able to decide whether ma- to deep sedation/anaesthesia (40). Several subsidiary phenomenon to the cause of the ximum and resource-intensive treatment gradation systems have been developed for coma, and is frequently a sign of serious should be started or continued (37). EEG has EEG findings to assist in making prognoses. damage with a very poor prognosis. a place in these assessments, not least be- The most widely used is Syneks gradation, Nonconvulsive status epilepticus may cause it requires limited resources, can be which is simple and has proved highly yield different EEG patterns, depending on repeated without restrictions and reflects accurate (41) (Box 1). the underlying cause. A hypoxic injury will functional changes in the brain. as a rule result in generalised EEG changes. If the prognosis is to be assessed, the Conclusion Encephalitis, cerebral contusion or cerebral cause of the coma must have been estab- EEG can provide early information about infarction will result in focal or lateralised lished and the coma must have lasted for at the cause of and prognosis in coma states. epileptic EEG changes. Periodic spiking is least 24 hours. The patient must not be seda- Repeated EEG recordings increase the dia- perceived by some as interictal epileptic ted, hypothermic below 32 °C or in circula- gnostic reliability and make it possible to activity, since it sometimes shows a transi- tory shock. Only EEGs taken in cases of follow developments in the coma, partly in tion to nonconvulsive status epilepticus coma following cardiac arrest have virtually order to assess a prognosis, partly because (illustration, EEG printout E), whereas unambiguous prognostic value (38, 39). In epileptic activity requiring treatment may others regard it as an ictal phenomenon (31, cases with other causes, particularly trau- occur along the way. The risk of developing 32) which may improve the prognosis if it is matic injury, caution should be observed nonconvulsive status epilepticus is rela- treated (33). When considering such perio- with respect to making a prognosis on the tively high in comas irrespective of cause dic activity, one should nevertheless be on basis of EEG alone. But with this reserva- (34). This stresses the importance of EEG the watch for the development of noncon- tion, the following prognostic factors may for comatose patients. vulsive status epilepticus. Sometimes the nevertheless be postulated (38, 39): condition has to be treated with deep seda- • «Burst-suppression» and status myocloni- tion with EEG monitoring until the burst- cus imply a very poor prognosis for survi- suppression pattern appears (33). val, but cases have been seen of relatively John Wilson (born 1949) The risk of developing nonconvulsive good restitution after status myoclonicus. Specialist in neurology and clinical neurophy- status epilepticus is relatively high in coma • With non-reactive rhythmic coma (alpha siology with a special interest in neurophysio- irrespective of cause (34). This stresses the or theta) restitution to a state of self-suffi- logical mechanisms and EEG in connection importance of EEG for comatose patients. ciency is extremely rare, and as a rule the with sleeping disorders, epilepsy and coma. end is death or a chronic vegetative state. Currently Senior Consultant at the Department Brain death • Persistent generalised low-voltage, slow for Complex Epilepsy, Oslo University Hospital. and chronic vegetative state activity that is nonreactive to sensory The author has completed the ICMJE form and In Norway, EEG is not used to diagnose stimuli implies a high mortality rate or reports no conflicts of interest. brain death in order to terminate treatment or survival with very poor neurological prepare for organ donation. An EEG is restitution. Helge J. Nordal nevertheless often taken in such a connec- • Periodic activity in the form of persistent Specialist in neurology and retired Senior tion, and is therefore mentioned briefly here. periods of single waves or short, often Consultant at the Neurology Department, Oslo Total cortical destruction is manifested by spiked complexes laterally or bilaterally, University Hospital, and Professor at Oslo the absence of activity in a scalp EEG. at 1–2 second intervals are signs of ser- University. When an EEG is taken, hypothermia ious damage. Any survival is then usually below 32 °C, circulatory shock and sedation with poor neurological restitution. >>>

56 Tidsskr Nor Legeforen nr. 1, 2013; 133 Feature Neurophysiology REVIEW ARTICLE

The author has completed the ICMJE form and 14. Schaul N, Gloor P, Gotman J. The EEG in deep after therapeutic hypothermia. Neurology 2009; reports the following conflicts of interest: He midline lesions. Neurology 1981; 31: 157 – 67. 72: 744 – 9. 15. Schiff N. Central thalamic contributions to arousal 31. Garzon E, Fernandes RM, Sakamoto AC. Serial has had assignments for lawyers, insurance regulation and neurological disorders of con- EEG during human status epilepticus: evidence companies and courts of law in personal injury sciousness. Ann N Y Acad Sci 2008; 1129: 105 – 18. for PLED as an ictal pattern. Neurology 2001; 57: compensation cases and has received fees 16. McCormick DA, Bal T. Sleep and arousal: thala- 1175 – 83. mocortical mechanisms. Annu Rev Neurosci 1997; 32. Akman CI, Riviello JJ jr. Generalized periodic for holding lectures in continuing professional 20: 185 – 215. epileptiform discharges in critically ill children: education courses for lawyers and doctors. 17. Young GB, Blume WT, Campbell VM et al. Alpha, a continuum of status epilepticus or an epipheno- theta and alpha-theta coma: a clinical outcome menon? J Clin Neurophysiol 2011; 28: 366 – 72. study utilizing serial recordings. Electroence- 33. Rossetti AO, Lowenstein DH. Management of phalogr Clin Neurophysiol 1994; 91: 93 – 9. refractory status epilepticus in adults: still more References 18. Kaplan PW, Genoud D, Ho TW et al. Etiology, neu- questions than answers. Lancet Neurol 2011; 10: 1. Koehler PJ, Wijdicks EF. Historical study of coma: rologic correlations, and prognosis in alpha coma. 922 – 30. looking back through medical and neurological Clin Neurophysiol 1999; 110: 205 – 13. 34. Towne AR, Waterhouse EJ, Boggs JG et al. Preva- texts. Brain 2008; 131: 877 – 89. 19. Gütling E, Isenmann S, Wichmann W. Electro- lence of nonconvulsive status epilepticus in coma- 2. von Economo C. Encephalitis lethargica. Wien Klin physiology in the locked-in-syndrome. Neurology tose patients. Neurology 2000; 54: 340 – 5. Wochenschr 1917; 30: 581 – 5. 1996; 46: 1092 – 101. 35. Buchner H, Schuchardt V. Reliability of electroen- 3. Moruzzi G, Magoun HW. Brain stem reticular 20. Bassetti C, Hess CW. Electrophysiology in locked- cephalogram in the diagnosis of brain death. Eur formation and activation of the EEG. Electroence- in syndrome. Neurology 1997; 49: 309. Neurol 1990; 30: 138 – 41. palogr Clin Neurophysiol 1949; 1: 455 – 73. 21. Vargas F, Hilbert G, Gruson D et al. Fulminant 36. Chatrian GE. Electrophysiologic evaluation of brain 4. Plum F, Posner JB. The diagnosis of stupor and Guillain-Barré syndrome mimicking cerebral death: A critical appraisal. I: Aminoff MJ, red. coma. 3. utg. Philadelphia, PA: F.A. Davis, 1980. death: case report and literature review. Intensive Electrodiagnosis in clinical neurology. New York, 5. Young GB. Coma disorders of consciousness. Ann Care Med 2000; 26: 623 – 7. NY: Churchill Livingstone, 1980: 525 – 88. N Y Acad Sci 2009; 1157: 32 – 47. 22. Drury I. The EEG in hypoxic-ischemic encephalo- 37. Young GB. Ethics in the intensive care unit with 6. Buzsaki G, Traub RD, Pedley TA. The cellular basis pathy. Am J EEG Technol 1988; 70: 1 – 8. emphasis on medical futility in comatose survivors of EEG activity. I: Ebersole JS, Pedley TA, red. Cur- 23. Brenner RP, Schaul N. Periodic EEG patterns: of cardiac arrest. J Clin Neurophysiol 2000; 17: rent practice of clinical electroencephalography. classification, clinical correlation, and pathophy- 453 – 6. 3. utg. Philadelphia, PA: Lippincott Williams and siology. J Clin Neurophysiol 1990; 7: 249 – 67. 38. Young GB, Wang JT, Connolly JF. Prognostic Wilkins, 2003: 1 – 11. 24. Young GB, Jordan KG, Doig GS. An assessment of determination in anoxic-ischemic and traumatic 7. Ebersole JS. Cortical generators and EEG voltage nonconvulsive seizures in the intensive care unit encephalopathies. J Clin Neurophysiol 2004; 21: fields. I: Ebersole JS, Pedley TA, red. Current using continuous EEG monitoring: an investigation 379 – 90. practice of clinical electroencephalography. 3. utg. of variables associated with mortality. Neurology 39. Young GB. Clinical practice. Neurologic prognosis Philadelphia, PA: Lippincott Williams and Wilkins, 1996; 47: 83 – 9. after cardiac arrest. N Engl J Med 2009; 361: 2003: 12 – 31. 25. Elger CE, Speckmann EJ, Prohaska O et al. Pat- 605 – 11. 8. Steriade M, McCormick DA, Sejnowski TJ. Thala- tern of intracortical potential distribution during 40. Logi F, Pasqualetti P, Tomaiuolo F. Predict recov- mocortical oscillations in the sleeping and focal interictal epileptiform discharges (FIED) ery of consciousness in post-acute severe brain aroused brain. Science 1993; 262: 679 – 85. and its relation to spinal field potentials in the rat. injury: the role of EEG reactivity. Brain Inj 2011; 9. McCormick DA. Cortical and subcortical gener- Electroencephalogr Clin Neurophysiol 1981; 51: 25: 972 – 9. ators of normal and abnormal rhythmicity. Int Rev 393 – 402. 41. Synek VM. Prognostically important EEG coma Neurobiol 2002; 49: 99 – 114. 26. Amzica F. Basic physiology of burst-suppression. patterns in diffuse anoxic and traumatic ence- 10. Amzica F. In vivo electrophysiological evidences Epilepsia 2009; 50 (suppl 12): 38 – 9. phalopathies in adults. J Clin Neurophysiol 1988; for cortical neuron-glia interactions during slow 27. Brenner RP. Is it status? Epilepsia 2002; 43 5: 161 – 74. (<1 Hz) and paroxysmal sleep oscillations. J Phy- (suppl 3): 103 – 13. siol Paris 2002; 96: 209 – 19. 28. Lowenstein DH, Aminoff MJ. Clinical and EEG features of status epilepticus in comatose 11. Kaplan PW. The EEG in metabolic encephalopathy Received 26 May 2011, first revision submitted and coma. J Clin Neurophysiol 2004; 21: 307 – 18. patients. Neurology 1992; 42: 100 – 4. 12. Blume WT. Drug effects on EEG. J Clin Neurophy- 29. Simon RP, Aminoff MJ. Electrographic status 8 March 2012, approved 21 June 2012. Medical siol 2006; 23: 306 – 11. epilepticus in fatal anoxic coma. Ann Neurol 1986; editor Are Brean. 13. Schaul N. Pathogenesis and significance of abnor- 20: 351 – 5. mal nonepileptiform rhythms in the EEG. J Clin 30. Rossetti AO, Oddo M, Liaudet L et al. Predictors Neurophysiol 1990; 7: 229 – 48. of awakening from postanoxic status epilepticus

Tidsskr Nor Legeforen nr. 1, 2013; 133 57