Consciousness as a useful concept in epilepsy classification

Summary

Impaired consciousness has important practical consequences for people living with epilepsy. Recent pathophysiologic studies show that seizures with impaired level of consciousness always affect widespread cortical networks and subcortical arousal systems. In light of these findings and their clinical significance, efforts are underway to revise the International League Against Epilepsy (ILAE) 2010 report to include impaired consciousness in the classification of seizures. Lüders and colleagues have presented one such effort, which we discuss here. We then propose an alternative classification of impaired consciousness in epilepsy based on functional neuroanatomy. Some seizures involve focal cortical regions and cause selective deficits in the content of consciousness but without impaired overall level of consciousness or awareness. These include focal aware conscious seizures (FACS) with lower order cortical deficits such as somatosensory or visual impairment as well as FACS with higher cognitive deficits including ictal aphasia or isolated epileptic amnesia. Another category applies to seizures with impaired level of consciousness leading to deficits in multiple cognitive domains. For this category, we believe the terms “dyscognitive” or “dialeptic” should be avoided because they may create confusion. Instead we propose that seizures with impaired level of consciousness be described based on underlying pathophysiology. Widespread moderately severe deficits in corticothalamic function are seen in absence seizures and in focal impaired consciousness seizures (FICS), including many temporal lobe seizures and other focal seizures with impaired consciousness. Some simple responses or automatisms may be preserved in these seizures. In contrast, generalized tonic–clonic seizures usually produce widespread severe deficits in corticothalamic function causing loss of all meaningful responses. Further work is needed to understand and prevent impaired consciousness in epilepsy, but the first step is to keep this crucial practical and physiologic aspect of seizures front-and-center in our discussions.

Consciousness is a challenging concept. Many people have strong opinions about what consciousness is or is not, perhaps because ultimately consciousness is what determines our personal individual experience throughout life. Consciousness also has tremendous clinical importance, particularly in epilepsy. Because of the great challenges in defining and understanding consciousness, some have advocated avoiding the term altogether when engaging in serious clinical or scientific discussion of epilepsy. However, loss of consciousness is a critical disability imposed by epilepsy, and one that clinicians must address in relation to safety and legal issues. Furthermore, the scientific investigation of the mechanisms of consciousness has become a legitimate and important endeavor for modern neuroscience. Fortunately, individuals such as Lüders and colleagues seek to confront the challenge.¹ Only by having the temerity to investigate and discuss consciousness can we ensure continued progress and improved understanding of this important topic.

The classification presented by Lüders and colleagues¹ has several merits but also has weaknesses. On the positive side, they have emphasized the central importance of consciousness in the description and classification of epileptic seizures. In contrast, the Blume et al.² “Glossary of Descriptive Terminology for Ictal Semiology” provides >80 descriptors of seizures, but the word “consciousness” appears not once. The 2010 International League Against Epilepsy (ILAE) report concludes that “the distinction based on impairment of consciousness or awareness, although of great pragmatic social importance (e.g., for driving), was impossible to define precisely,” and therefore the ILAE committee chose to lump all focal seizures into one undifferentiated category.³ As support for this approach, they cited Gloore's 1986 opinion piece based on his anecdotal observations of individual patients during seizures.⁴ Although Piere Gloore was certainly a respected figure in epilepsy, other well-respected clinician-scientists in epilepsy have viewed consciousness as an important area for research (e.g., Penfield and Jasper). Furthermore, there has been much scientific progress on the fundamental pathophysiology and behavioral characterization of impaired consciousness in epilepsy in recent years.^5–9 The ILAE report succeeds in finding sufficient scientific and clinical evidence to distinguish 12 types of generalized seizures, including five types of absence and three types of myoclonic seizures, yet it does not distinguish even one subtype of focal seizures and instead combines them all under one heading.³ The ILAE report does allow the use of descriptive terms to replace the traditional categories of simple partial and complex partial seizures (Table 1). A recent commentary provides more concise terms compatible with the ILAE report, namely focal impaired consciousness seizures (FICS) and focal aware conscious seizures (FACS).¹⁰

Table 1. Accepted terminology for impaired consciousness in focal or partial seizures.

ILAE 1981 Classification46	ILAE 2010 Commission Report3
Simple partial seizure	Focal seizure without impairment of consciousness or awareness. Subtypes include: With observable motor or autonomic components; Involving subjective sensory or psychic phenomenon only
Complex partial seizure	Focal seizure with impairment of consciousness or awareness. “Dyscognitive”a is a term that has been proposed for this concept
Partial seizure evolving to secondary generalized seizure	Evolving to a bilateral, convulsive seizure (involving tonic, clonic, or tonic and clonic components)b

^aThe term “dyscognitive” is confusing, as discussed in the text.

^bNote that focal seizures that evolve to bilateral atonic seizures are not included.

Lüders and colleagues go further than prior classifications by introducing descriptive categories of impaired consciousness, not just in focal seizures but also in generalized seizures and in epileptic delirium.¹ Another strength of their work is that it represents the consensus opinion of a large number of distinguished clinical epileptologists. Dr. Lüders deserves credit for promoting, discussing, and sharing his views on epilepsy classification with colleagues and collaborators over the years, and for keeping consciousness at the forefront of this discussion.

However, their proposal has two main weaknesses. The first is that insufficient data are presented to justify their proposed five types of alterations of consciousness. The second is that the proposal introduces terms that may be confusing. We will discuss each of these concerns in turn.

The Need for More Data

The proposed five types of alterations of consciousness in seizures are “auras with illusions or hallucinations, dyscognitive seizures, epileptic delirium, dialeptic seizures, and epileptic coma.”¹ Additional discussion is needed of the altered physiology and behavior in these conditions and how they relate to other disorders of consciousness. In an effort to provide this information, we will now discuss the types of impaired consciousness in epilepsy using functional neuroanatomy as a framework (Table 2).

Table 2. Proposed anatomically based classification for impaired consciousness in epilepsy.

Type of deficits	Seizure examples a	Mechanisms
Impaired content of consciousness
Selective lower-order cortical deficits	FACS b without cognitive dysfunction	Focal seizure discharge
Selective higher-order cortical deficits	FACS with selective cognitive dysfunctionc	Focal seizure discharge
Impaired level of consciousness
Widespread cortical deficits, moderate	Absence seizures and FICS d	Widespread spike-wave, or slow waves
Widespread cortical deficits, severe	Generalized tonic–clonic seizures e	Widespread polyspike discharge
Impaired content and level of consciousness
Combined selective and widespread deficits	Postictal period following focal motor seizure with secondary generalization	Focal and widespread abnormal activity

^aThe list of seizure examples provided here is not intended to be exhaustive; there are several other possible examples for each category.

^bFocal aware conscious seizures (FACS). This includes both motor and purely sensory seizures (i.e., auditory, olfactory, somatosensory, visual).

^cFocal aware conscious seizures (FACS) with focal cognitive dysfunction such as ictal aphasia, ictal neglect, ictal visual agnosia, or epileptic amnesia.

^dFocal impaired consciousness seizures (FICS). Temporal lobe seizures with impaired consciousness show widespread depressed cerebral blood flow and cortical slow waves unlike temporal lobe seizures without impaired level of consciousness.^8,32,33 Parietal lobe seizures with impaired consciousness show similar abnormal thalamocortical rhythms.⁴⁷ Other FICS (i.e., frontal and occipital) may have similar mechanisms but have not yet been thoroughly investigated.

^eMay also include secondary generalized seizures.

Plum and Posner distinguished systems contributing to the content of consciousness (including specific sensory, motor, memory, and emotional systems) from systems controlling the level of consciousness.¹¹ Lesions or dysfunction affecting widespread bilateral regions of the cerebral hemispheres, the diencephalon, or the upper brainstem arousal systems can result in impaired level of consciousness. Typically disorders such as stroke—which cause focal deficits such as somatosensory loss, motor loss, aphasia, visual changes, or memory impairment—are not considered disorders of consciousness. Although a deficit in the content of consciousness is present, these patients are considered conscious because they are aware and can respond in a voluntary (not just reflexive) manner; overall function is relatively preserved. By analogy, patients who have seizures in focal regions causing a somatosensory aura, focal motor seizure, ictal aphasia, visual distortions, or epileptic amnesia without full loss of awareness and responsiveness are considered conscious despite having selective deficits in the content of consciousness (Table 2). The mechanism for impairment in these patients is focal seizure discharge confined to a limited region of the brain.¹² This gives rise to focal deficits without impairment in the overall level of consciousness.

These categories of FACS with “impaired content of consciousness” (Table 2) are roughly analogous to the Lüders et al.¹ proposed categories of auras and dyscognitive seizures. Lüders et al. also point out that in these two categories focal deficits do not interfere with the overall level of consciousness. The only real distinction (aside from the terminology, discussed later) is that we include not just auras, which have no overt behavioral manifestations, but any focal seizures (including motor seizures) that interfere with selective aspects of sensory, motor, mnemonic, or emotional experience without disrupting the level of consciousness. In addition, because the scheme that we present is tied directly to anatomy (Table 2), some “auras” in the Lüders et al. terminology (e.g. simple visual hallucinations) fall into the category of selective lower-order cortical deficits, whereas others (e.g. formed visual hallucinations) lie in the category of higher-order cortical deficits.

Disorders of consciousness are usually considered to be conditions that affect the level of consciousness. These include coma, vegetative state, minimally conscious state, delirium, and certain types of epileptic seizures¹³ (Table 2). Again, according to classical principles of neurology described by Plum and Posner,¹¹ impaired level of consciousness occurs in disorders that interfere with subcortical arousal systems, or with widespread bilateral corticothalamic networks, or with both. The same holds true for seizures that impair overall responsiveness (Table 2).

In absence seizures, spike-wave discharges invade widespread cortical and subcortical networks,^14–16 usually causing loss of responsiveness to higher-order tasks including choice reaction time, memory encoding, and verbal or motor responses to commands,; but sometimes sparing simpler more automatic behaviors such as finger tapping.^17–20 The widespread moderately severe deficits of cortical function seen in absence seizures where some responsiveness may be preserved have been compared to the minimally conscious state.^21–23 In focal seizures with impaired level of consciousness, there are also moderately severe deficits in widespread cortical function interfering with responses to verbal and nonverbal questions and commands^24–28; but they may spare simpler responses such as reactive automatisms, visual tracking, ball grasp, or blink to visual threat.^29,30 The physiology is best understood in temporal lobe seizures where impaired consciousness is associated with widespread depressed cerebral blood flow and cortical slow waves resembling deep sleep, coma, or encephalopathy.^8,31–33 In such temporal lobe seizures, the widespread corticothalamic dysfunction is postulated to result from the spread of the focal temporal lobe seizure to subcortical structures (e.g., lateral septum, anterior hypothalamus, thalamus), which in turn depress the subcortical arousal systems (e.g., cholinergic and serotonergic).^6,34–36 In contrast, these widespread cortical-subcortical network changes are not seen in temporal lobe seizures in which consciousness is spared.^7,8,32,33

Generalized tonic–clonic seizures usually cause widespread severe cortical deficits (Table 2), leading to lack of even simple responses such as visual tracking, ball grasp, or blink to visual threat.³⁰ As noted by Lüders et al., ¹ these behavioral deficits resemble the profound unresponsiveness seen in coma except that the eyes are usually open during seizures. In terms of pathophysiology, in the ictal phase there are widespread polyspike discharges associated with abnormally increased cortical and subcortical cerebral blood flow, which evolves to polyspike-and-wave in the clonic phase.^37–40 Postictally behavioral responses remain depressed along with decreased cortical activity and widespread reduced cerebral blood flow, which, of interest, is correlated with increased cerebellar activity, suggesting a possible inhibitory role for the cerebellum in postictal depression.³⁸

Finally, in many seizures there is a combination of intense focal discharges along with more widespread abnormalities, leading to impairment in both the content and level of consciousness (Table 2). This can occur in the postictal phase of focal seizures with secondary generalization, where localized postictal deficits often occur in the region of onset (e.g., postictal hemiparesis or aphasia), together with more global impaired responsiveness.^41,42

These categories of “impaired level of consciousness” in seizures (Table 2) correspond roughly to the Lüders et al.¹ proposed categories of dialeptic seizures, epileptic delirium, and epileptic coma. The main difference is that unlike the Lüders et al. proposal, we consider these to all affect the level of consciousness to a varying degree. As we have already discussed, in focal limbic (temporal lobe) seizures with impaired consciousness, there is direct evidence of sleep-like cortical slow wave activity, depressed cerebral blood flow, and reduced subcortical cholinergic arousalproducing behavior resembling the minimally conscious state or deep sleep parasomnias such as somnambulism.^8,21,32–34 Therefore, although the eyes are usually open and some simple responses remain, we feel it is incorrect to state that these seizures do not affect arousal. In addition, since abnormal level of consciousness can occur with either impaired subcortical arousal or with widespread bilateral cortical network dysfunction, we would also consider the level of consciousness to be abnormal in absence seizures and in generalized tonic–clonic seizures, where widespread cortical dysfunction is associated with globally decreased responsiveness.

The delirium-like state that exists in nonconvulsive status epilepticus again involves widespread, moderately severe cortical dysfunction and in that sense does not differ fundamentally from other epileptic seizures that cause impaired level of consciousness (Table 2). We agree that pathophysiology of increased behavioral arousal that occurs in some cases of epileptic and nonepileptic delirium requires further investigation.

The Need for Clear Terms

There are some additional concerns with the Lüders et al.¹ proposal that relate to the specific terms used. We should note that unlike the preceding discussion these concerns are not based on data but rather on opinion, and ultimately the choice of terminology is a matter of subjective preference. Our first concern is not with the accuracy of a term but with the confusion its use may create. We fully agree with Lüders et al. that the term “dyscognitive” offered by the ILAE^2,3 is not an appropriate substitute for complex partial seizures or for “focal seizures with impaired consciousness” (FICS). As emphasized by Lüders et al. and as we have argued previously,¹⁰ selective cognitive deficits are not the same thing as impaired level of consciousness. For example, patients with ictal aphasia, memory loss, or other specific sensory, motor or cognitive dysfunction may remain fully alert and responsive, exhibiting selective deficits in content rather than overall level of consciousness (Table 2). In contrast, when focal seizures impair the level of consciousness there are widespread deficits across virtually all cognitive tasks.²⁸ “Dyscognitive” is in fact very correctly used in the proposal by Lüders et al. to mean seizures with focal cognitive deficits, but without impaired overall level of consciousness. The only problem is that this term has already been widely (and incorrectly, in our opinion) promoted by the ILAE to mean focal seizures with impaired level of consciousness. Therefore, coopting the same term to mean something else has the potential to create much confusion and we therefore advocate avoiding the term “dyscognitive” altogether.

Another potential concern is the term “dialepsis.”^1,43 Although there are similarities in the behavioral state of staring unresponsiveness seen in absence seizures and in focal seizures with impaired consciousness, there are also many important differences in pathophysiology, behavior, treatment, outcome, and patient populations. It therefore seems inadvisable to unify these seizure types under a single term, particularly one that has the potential to sound arcane to the nonspecialist (this is a matter of personal opinion).

Finally, the term “epileptic coma” for the postictal state following generalized tonic–clonic seizures may also create confusion. The term “epileptic coma” might suggest a prolonged state of coma-like unresponsiveness such as that seen in some patients in the intensive care unit with unexplained coma who turn out to be in status epilepticus.^44,45 However, Lüders et al. intended this term to refer to the relatively brief state following generalized tonic–clonic seizures that commonly lasts less than an hour. Use of the term “epileptic coma” for this transient postictal impairment seems confusing. Postictal impairment of content and/or level of consciousness would seem more appropriate terms.

Summary and Conclusions

Lüders and colleagues have made a valuable contribution by raising the general interest in altered consciousness as an important concept in epilepsy description and classification. They proposed several categories of seizures in which altered consciousness may play a role. We have endeavored to bolster this effort by proposing alternate categories of seizures based on established anatomy and physiology (Table 2) and to discuss some of the existing data to support these categories. With this approach, seizures can be classified as disrupting the content of consciousness through effects on localized cortical circuits; or the level of consciousness through effects on subcortical arousal systems and/or widespread bilateral regions of cortex. Some of our categories roughly correspond to those of Lüders and colleagues, but we did not use their proposed terms because we felt some of them may add confusion rather than clarity. Further investigation of the pathophysiology and treatment of impaired consciousness is essential for the field. Ongoing discussions of terminology and classification are also important, and fortunately are continuing through several groups including the ILAE.

Biographies

Dr. Hal Blumenfeld is Professor of Neurology, Neurobiology and Neurosurgery and Director of the Clinical Neuroscience Imaging Center (CNIC) at Yale School of Medicine.

Dr. Kimford J. Meador is Professor of Neurology & Neurological Sciences and Director of the Epilepsy Monitoring Unit at Stanford School of Medicine.