A 48-Year-Old Man With Temporal Lobe Epilepsy and Psychiatric Illness

DR DELBANCO: Mr C is a 48-year-old man with a long history of cognitive and psychological problems, marked by mood swings, memory loss, and seizures involving the frontal and temporal lobes of the brain. A primary physician, a neurologist with special interest in seizures (Dr J), and a psychiatrist (Dr N) manage Mr C's care. He has private insurance, works part-time, and lives near Boston with his wife.

Mr C's problems likely date to birth. He was the product of an uncomplicated pregnancy but during labor there may have been some anoxia. He recalls a "foggy feeling" in his head from the first time he can remember. As a child, he fell out of a tree and hit his head without notable sequelae. In elementary school, he did well but had difficulty with his memory. He continued a successful academic experience in high school, during which time he had intermittent abdominal pains, and then was graduated from a top university with extraordinarily high academic achievement. In college, he describes having had difficulty understanding lectures and remembering details. He read avidly and worked endless hours but felt his ability to learn was hampered by a poor memory. He had difficulty handling several different tasks at once. In graduate school, depression because of his struggles with his cognitive limitations led to a suicide attempt, which caused a contusion to the right side of the head. Thereafter, Mr C was hospitalized; lithium followed by carbamazepine and phenytoin gave him more, but not complete, control of his symptoms.

Neurologic evaluation at that time showed no persistent physical findings or lateralization. He was ambidextrous, and his right hand was noted to be a bit smaller than his left. Reflexes were diminished but bilaterally symmetrical. There were no motor or sensory findings. Multiple electroencephalogram (EEG), magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET) evaluations showed inconsistent findings. There were no clear ictal patterns on EEG.

Over the next years, he was treated with multiple medications. Multiple MRI, CT, and PET scans, along with serial EEGs, continued to show inconsistent findings, with frequent abnormalities identified in the left frontal and temporal lobes. Mr C fared increasingly better, working at various jobs and getting married, until serious cognitive and affective disruption again supervened. Multiple high-dose antidepressants, augmentation strategies, neuroleptics, and atypical neuroleptics were tried without effect. However, over a period of several years, his mood and functionality slowly improved. In recent years, Mr C has worked part-time in jobs far less demanding than his superior intellectual attainment might otherwise anticipate.

In 2000, Mr C contemplated surgery for possible focal epileptiform lesions in the brain, but depth electrode investigation revealed abnormalities that were multifocal and involved the anterior cingulate and supplementary motor areas. The abnormalities were more active on the left than on the right. He also had a left temporal lobe slow wave focus. Electrical stimulation through the left supplementary motor area electrode at subseizure threshold levels elicited his characteristic paroxysmal, seizurelike cognitive and affective symptoms.

In addition to seizures, Mr C notes very poor memory, which has led to great difficulty in recognizing people and has put him in awkward social circumstances on many occasions. A talented musician, Mr C plays an instrument with considerable competence but has great difficulty remembering pieces even after extensive practice. He feels worse during the winter months and notes that lack of sleep exacerbates the symptoms that are likely partial seizures. Sexual activity at times exacerbates his symptoms. He notes also long-standing tinnitus in the right ear, with a sense of white noise coming from his left ear. During seizure activity, he feels he is spinning backward in space and that his limbs become very long.

For decades, Mr C developed an intermittent rash on his right arm, and recent evaluation demonstrated herpes simplex virus 2 infection. Appearance of the eruption appears to coincide with exacerbations of his seizures, and, in the last year, treatment with valacyclovir diminished his seizure symptoms. In recent months, however, this treatment appeared less effective and was discontinued, to be restarted pending observation.

His family history is notable for affective disorder in his paternal and maternal forebears, including a parent in whom formal neuropsychological testing revealed left temporal dysfunction, and a sibling with affective disorder. He has had a cholecystectomy, tonsillectomy/adenoidectomy, repair of an anterior cruciate ligament, and hernia repair, but otherwise he has a generally negative review of systems. He has consumed very little alcohol, finding that it exacerbates his seizures. He smoked cigarettes in the past. Formal neuropsychological testing has revealed markedly superior intelligence, with some features pointing toward abnormalities in the left temporal lobe.

Examination reveals a highly intelligent, engaging man who speaks in long sentences. There are no focal neurologic abnormalities, though his neurologist notes right arm posturing on stressed gait. His psychiatrist observes asymmetry of his facial muscles and poor articulation of speech during periods of worsening seizures. His vital signs and physical examination are unremarkable. Currently, Mr C takes cytidine diphosphocholine (CDP-choline) (to replenish choline stores in the brain), multivitamins, and carbamazepine. He continues to work hard with his neurologist and psychiatrist to come up with a therapeutic regimen that will diminish his symptoms and enable him to function more effectively in the future. In addition, he continues to pursue what he and his psychiatrist consider highly effective and valuable psychotherapy.

There are many everyday examples of ways in which my memory is a problem. I can learn a lot of things, but I can't hold on to them—and over time, the use of information deteriorates. It's as if the more I use the information, it gets spotty and holes in it, especially if I'm trying to learn something else. There's a rate at which information for me deteriorates that outpaces what it does with other people. People don't realize, I think, exactly how much information about a whole lot of things is what allows your brain power to function over—that makes the information useful. And without that ready access to a lot of information, it's actually very hard to carry on a career life.

I am constantly struggling with this sense of wanting to do things that I would like to do with my life and being hampered and then having that level of distress and disruption. My understanding personally of it is that there is some kind of as yet not understood underlying disorder here that is expressed in me through things that have a very seizurelike component, and things that have a highly cognitively disruptive component, and things that have an emotionally disruptive component, and things that have on the edge of psychoticlike components.

The events that I've had that sound like seizure events are a sense I have of spinning backwards in space. It's as if there's an axis through the middle of my body and my whole body spins backwards. And I know that this isn't happening because I can open my eyes and look around and see that it's not happening. If I close my eyes, it seems much more real. Another sense that I have that, likewise, I know isn't actually happening but seems very real, particularly if I close my eyes, is that it feels as if parts of my body, particularly my arms, get very large and very heavy. Another symptom is that I get the sense that there's an argument going on in my head, although there are not actually any voices or content. What I've come to understand is that some kind of emotional part of my brain is being affected by the seizure activity.

One unique thing about my illness is that I always retain a sense of what is real and what isn't real. When I was contemplating suicide and I was looking out of this window of this 10-story building, I was having so much difficulty retaining my memory that I really decided that if I couldn't get my brain to function in the way that it needed to, I really didn't want to live anymore. So then, I was looking out this window, and in all of the windows across the way I could see the face of this Edvard Munch painting, The Scream—which very readily expressed how I was feeling in my own life—and they were all there sort of shimmering. I guess that was a kind of hallucination, although I basically knew that there was something awry in my brain and that that was what this was about. And I never really lost that sense.

The obvious crossroads that one is at constantly with a disease like this is how to live life, and what to expect of oneself, and what to go forward to try to accomplish and not try to accomplish. I have given an enormous amount of thought to what is happening in my brain, and the way that I am affected by this illness and the way that I can use my own sense to disentangle some of the snafus that happen from the way my brain works. It's a lot of work always, but in some ways I have done that. I guess I think that it's always the case that any successful psychotherapy is as much due to the sense of the patient as the sense of the doctor—a doctor cannot wield intelligent psychotherapy on someone who isn't there to dance with him.

Mr C has presented with a number of behaviors that really tax our ability to understand what is psychiatric and what is neurologic in origin. He is an extremely bright, capable man who, for most of his adult life, has had significant impairment to his ability to function—or at least to maintain a steady level of high function. We have learned, I think with him, to note that what we think of as classic psychiatric problems have been very closely tied to the neurophysiology.

When Mr C was admitted to the hospital for stereotactic EEG, we blinded him as to whether he would know when and where we were stimulating him. And we found that very, very minimal currents of stimulation triggered the subjective feelings that we, for many years, have been saying were probably convulsive, and that was this sort of whooshing feeling that he would get in his head. With very low currents, stimulation reproduced that sensation in the left supplementary motor cortex.

The question I would ask Dr Devinsky has to do primarily with interplay among the disciplines of psychiatry, neurology, and epileptology. When are we, as physicians, going to start trying to figure out what factors lead to the severe psychiatric complications to this disorder so that we can start treating those at a much earlier stage and start treating them in a more prospective, rather than reactive, fashion?

Mr C has not had dramatic relief from any of the agents we've tried. Psychopharmacology is wonderful because it's like surgery—when it works, it really works. He's not had that. He has had periods when it looks like things are better, but then they often get worse again. We have not seen sustained response to any of the antiepileptic agents or any of the mood agents. And we've tried everything and then some.

I would like to see if we could localize the source of his illness and arrive at what I would call a classic neurologic diagnosis. Is it something we could suppress enough with medication so his cognitive function would improve and, therefore, his daily life function would improve, his emotional status would calm down, and he would have an easier time of it?

Should we try other imaging studies now or in the future? Given this patient's current primary complaint of cognition and mood—not so much spell symptoms—what, if any, other medicines or medicine combinations would Dr Devinsky offer to help Mr C?

What is temporal lobe epilepsy (TLE)? What is the epidemiology of TLE in the United States today? What are the clinical manifestations of TLE? What is the differential diagnosis, and how is the diagnosis established? What are the treatment options? What are their risks and benefits? What are the financial, emotional, and social costs of TLE? In terms of diagnosis and management, what does the future hold? Finally, what do you recommend for Mr C?

DR DEVINSKY: Mr C's history exemplifies the typical paroxysmal features of temporal lobe epilepsy syndrome and highlights the expanse of interictal cognitive and behavioral disorders. Since early childhood, he has had undiagnosed problems with learning and memory, abdominal discomfort, and sadness. Sleep deprivation and stress during graduate school exacerbated his depression, leading to a suicide attempt accompanied by visual hallucinations. Psychotic depression, partial seizures, or both may have contributed. The fall contused his right temporal lobe and likely caused contrecoup left frontotemporal damage. Subsequently, Mr C developed partial seizures of temporal lobe origin: vertigo, distortion of limb size, déjà vu, and derealization that can progress to staring, followed by lethargy. This constellation of symptoms suggests a right lateral temporal focus. Problems with autobiographical and geographic memory and failure to "emotionally relate to information" are also consistent with right temporal dysfunction. Stimulating the left supplementary motor area (contrecoup to the injury) reproduced another seizure symptom, a "whooshing" feeling in his head.

Mr C insightfully observes that increased seizure activity can exacerbate memory and emotional problems. Antiepileptic drugs, initially used to treat affective disorder, reduced the frequency of his seizures and improved neurobehavioral function. However, conventional and investigational drugs failed to control his seizures fully. He and his physicians considered an underused procedure, surgical therapy for refractory TLE, but a clear focus could not be identified. Mr C's concatenation of family and personal history of major depression, disabling cognitive symptoms, and epilepsy yield a dynamic disorder requiring collaboration among the patient, psychiatrist, and neurologist. Mr C's rescue from the depths of illness marks the journey of a determined human spirit, partnered with compassionate medical resources.

Temporal lobe epilepsy is a syndrome characterized by simple partial and complex partial seizures. Symptoms result from activation and inhibition of temporal lobe or neighboring areas involved in seizure spread. For many patients, seizures secondarily generalize into convulsions. Especially in sleep, this generalization may occur without prior symptoms, making it difficult to distinguish TLE from generalized or other partial epilepsies.Conscious awareness is preserved during simple partial seizures (auras) that usually cause autonomic, sensory, emotional, or cognitive symptoms ( Article ).¹ Complex partial seizures impair consciousness and memory and typically are associated with automatic movements of the lips, face, pharynx, or hands. Seizures are usually stereotypic for individual patients, both in their manifestations and duration. A typical sequence is an aura followed by arrest of motor behavior, blank stare, and automatisms. Motor activity in the hand helps lateralize seizure onset: semipurposeful automatisms are usually ipsilateral, while dystonic postures or clonic jerks are usually contralateral to the focus.² The duration is 20 to 180 seconds; most last 60 to 120 seconds. These events are often followed by transient confusion, tiredness, and amnesia. Mr C's symptoms are typical of TLE. In addition to staring, confusion, and subsequent amnesia, he also muttered short incomprehensible phrases and wandered aimlessly.

Premonitory (Hours or Days Before Seizure)

Affective (depression, irritability, aggression)
Cognitive (confusion, impaired memory)
Behavioral (withdrawal)
Somatic (headache; change in body temperature, appetite, or thirst)

Simple Partial Symptoms

Sensory (illusions or hallucinations in olfactory, auditory, visual, or gustatory modes)
Autonomic/visceral (rising abdominal feeling, chest sensa tion, palpitation, piloerection)
Affective (depression, fear, anxiety, elation, laughing, cry ing, religious or sexual feeling)
Cognitive/experiential (confusion, altered familiarity [déjà vu, jamais vu], dreamy state, depersonalization, forced thoughts, distortion of time or body image)

Complex Partial Symptoms

Impaired consciousness (blank stare, arrest of ongoing behavior)
Automatisms (lip smacking, swallowing, fumbling or re petitive movements of hands, repetitive phrases, con tinuation of ongoing behavior [eg, walking, dealing cards])
Dystonic posturing of extremity, usually the hand

Postictal

Headache
Affective (depression, mania, aggression*)
Cognitive (confusion, amnesia, anomia, aphasia)
Psychosis

Interictal

Affective (depression, anxiety, irritability, hypomania, in creased or decreased emotionality, aggression)
Cognitive (amnesia, anomia, psychomotor slowing, im paired executive and social functions)
Sexual (reduced libido, impotence, anorgasmia)
Personality (circumstantiality, humorlessness, hyper graphia, hypermoralism, obsessionalism, paranoia, reli giosity, viscosity)
Psychosis

*Usually provoked by restraint.

The border between simple and complex partial seizures is fluid. Mr C's description of "watching the world and myself," and visual perceptions becoming "gray and 2-dimensional, as if on television" exemplifies the challenges in defining the level of consciousness. Some patients are unaware of the lapses and adamantly report preserved consciousness. Since physicians rarely observe seizures, family members or friends can be instructed to test responsiveness and memory during seizures. This distinction is very important, since physicians must decide if a patient is safe to drive or operate dangerous equipment. To maintain driving privileges, many patients, usually men, deny seizure activity.^{3 - 4} About 31% of patients with refractory epilepsy drive, violating state laws and endangering themselves and others.⁴ Alternatively, unnecessary driving restrictions impose dire social and economic consequences.

In TLE patients, behavioral alterations can occur before, after, and between seizures. Premonitory symptoms occur hours or days before a seizure. When premonitory symptoms consistently precede a seizure, they warn patients to seek safety and make short-term medication changes to prevent the impending seizure. During the interictal period, patients with TLE commonly experience continuous symptoms. These include changes in cognitive function, personality, affect, and drive-related behaviors such as libido.⁵ Mr C illustrates these changes, as well as the difficulties identifying their cause: structural brain changes, epilepsy, medication effects, and primary psychiatric and environmental factors.

The epidemiology of TLE is poorly defined. The prevalence of epilepsy in the United States is 0.7%, affecting 2.1 million people. The incidence of all forms of epilepsy is about 40 cases per 100 000 patient-years. Cumulative incidence from birth to age 20 years is about 1%, increasing to nearly 3% at age 75 years.⁶ The incidence of partial epilepsy is fairly consistent from birth to age 65 years (20 cases per 100 000 patient-years) but then rises briskly. Overall, the percentage of new patients with partial epilepsy among all types of epilepsy increases with age.⁷

Partial epilepsy comprises 50% of incident and 60% of prevalent epilepsy cases.⁸ Focal epileptiform discharges are most frequently recorded over the temporal lobe.⁹ Temporal lobe epilepsy is the most common epilepsy syndrome.¹⁰ Unfortunately, syndrome classification and localization of epilepsy to a specific lobe are problematic in epidemiologic studies.¹¹ Conservatively, among newly diagnosed partial epilepsies, 27% had TLE and another 6% had temporal and frontal foci.¹² Thus, of the 1.2 million Americans with partial epilepsy, TLE probably occurs in at least 400 000.

Patients with epilepsy face an enormous burden. As the disorder removes control over one's life, it may lead to a deep stigma and heavy financial burden and can negatively affect mental and physical functioning. For patients whose seizures are well controlled with a single antiepileptic drug (AED), the costs of epilepsy are modest. For those with difficult-to-control seizures, direct medical costs of epilepsy are very high.¹³ Annually, epilepsy incurs $12.5 billion in costs in the United States, 85% of which are indirect.¹³

Individuals with refractory epilepsy are often underemployed or unemployed. Restrictions on driving limit social and work opportunities. Patients with refractory epilepsy require intensive use of medical resources, including physician visits, changes in AEDs, use of expensive AEDs, frequent measurement of plasma drug levels or other special laboratory tests, emergency department care, travel to specialized epilepsy centers, extensive inpatient testing and, possibly, surgical therapy.

The combined effects of seizures, AEDs, and psychosocial handicaps sap emotional vigor, causing fatigue, depression, anxiety, personality disorders, paranoia and psychosis, fears concerning seizures or adverse effects of medications, and sexual dysfunction and diminished libido. Cognitive blemishes affect mental processing speed, attention, memory and learning, language, arithmetic ability, judgment, and reasoning. Social costs include stigma, reduced self-esteem and independence, social isolation, embarrassment of public seizures and the fear that they instill, and perceived and real discrimination. Impairments in social cognition, a complex function that profoundly influences one's life, can be devastating.

Seizures can arise from medial or lateral regions (Table 1 and Figure 1). Medial TLE is more than twice as frequent as lateral TLE.¹⁴ Patients with medial TLE often have had febrile seizures; meningitis, encephalitis, and head trauma are less common risk factors.^{15 - 16} Mesial temporal sclerosis, the most common lesion, is usually refractory to medical therapy, but surgery can be effective.¹⁵ A latency of 5 to 10 years is common between the central nervous system insult and onset of seizures.^{16 - 17} The principal seizures are complex partial; common auras are abdominal sensations or fear. Patients with lateral TLE rarely have a history of febrile seizures or a long interval between cerebral insult and epilepsy.¹⁸ Common etiologies include malformations of cortical development, head trauma, low-grade tumors, and infection. Surgery is often effective when medications fail, especially when MRI reveals a lesion.¹⁸ In lateral TLE, complex partial seizures predominate, with auras such as déjà vu, depersonalization, abdominal sensation, tinnitus, and vertigo. Throughout his clinical course, Mr C has reported all of these symptoms, and he likely has lateral TLE.

The first simple partial seizure (aura) symptom often reveals the area where the seizure arises; the progression of sensory, autonomic, and experiential symptoms and objective signs reflect the ictal spread. When seizures arise in "silent" areas, symptoms result from spread to other regions. Auras may be evoked by activation of areas that are strongly networked to the seizure focus.¹⁹ Seizures can cause negative symptoms. For example, hippocampal seizures impair memory by disrupting the creation of "directories" and "address books" that bind and find fragments of experience.

Patients frequently identify factors that provoke seizures. In decreasing order of frequency, the most common are missed medication, stress, sleep deprivation, sleep, fever or illness, fatigue, and alcohol withdrawal.^{20 - 21} Mr C notes worsening of seizures and cognition with sleep deprivation and stress. For women older than 12 years, seizures are more frequent during premenstrual/early menstrual and ovulatory periods as opposed to follicular or luteal periods.^{21 - 22} The effects of menopause on seizure frequency are unpredictable.²³

Geschwind and colleagues^{24 - 25} defined an interictal syndrome of deepened emotions, circumstantiality, altered religious and sexual concerns, and hypergraphia. They emphasized behavioral change rather than disorder, since some features were not maladaptive or negative. Geschwind²⁶ recognized that considering behavior more broadly and asking questions correctly are critical:

A young boy with temporal lobe epilepsy was presented to me by the resident who said, "We don't think he is particularly religious." The patient came in and I said, "Are you religious?" He said, "No." I said, "Why not?" He said, "Because the rabbi in my congregation is not theologically sound."

Mr C reports positive features: an "attractive personality"; "people enjoy my company." Deepening of emotions, often with intense interpersonal contact and sustained intense affect and will, can occur and may contribute to brilliance, as in Alexander the Great, Julius Caesar, Napoleon, Dostoyevsky, van Gogh, and others.²⁷ The interictal behavioral changes in TLE do not comprise a neatly defined syndrome but overlap with behaviors in patients with other neurologic and psychiatric disorders. For example, diverse behavioral changes often occur in patients with absence, juvenile myoclonic, and frontal lobe epilepsies.²⁸

In TLE, the behavioral spectrum is diversified but skewed negatively, leading to intellectual and emotional disorders. Mr C exemplifies these disorders, as well as the complex interplay of biological and psychological factors. Interictal amnesia can result from structural lesions,²⁹ neuronal dysfunction or loss, interictal epileptiform discharges,³⁰ recurrent seizures (see below), and AEDs.³¹ Family history, mesial temporal sclerosis, AEDs, and psychosocial-reactive mechanisms contribute to depression.^{32 - 33} Left or bitemporal seizure foci, family history, female sex, dysplasia or low-grade tumors, and seizure clusters are risk factors for psychosis.^{34 - 37}

Interictal depression occurs in one third of TLE patients.^{38 - 39} Among those with refractory epilepsy, depression potently reduces quality of life; half are depressed and nearly 20% have suicidal ideation.⁴⁰ Suicide rates in TLE patients are more than 5 times higher than in healthy populations.^{38 ,41 - 43} Mr C's depression imposed an enormous burden on him, and nearly proved fatal. Most patients can be successfully treated with antidepressants without exacerbating epilepsy.⁴⁴ Failure to recognize and treat depression in epilepsy patients remains an enormous problem.^{40 ,45}

Anxiety, phobias, and panic attacks also are common interictal problems in TLE patients.⁴⁶ Serotonergic agents (eg, buspirone, selective serotonin reuptake inhibitors) are often helpful. Benzodiazepines may cause tolerance and can cause abuse or mood and memory disorders. Furthermore, dose reductions can provoke seizures.

Impaired short-term memory is also common. Patients report difficulty recalling recently learned information, especially names and details. As Mr C describes, remembering autobiographical memories of vacations and important events also can be difficult. Left-sided temporal foci primarily impair verbal memory; right-sided foci affect recently acquired visual, spatial, and geographic information. Although standard tests with 30-minute delays reveal memory impairments, longer delays reveal even greater deficits.⁴⁷

Interictal psychosis occurs in approximately 7% of all epilepsy patients and exceeds 10% in those with refractory TLE.^{35 ,48} In contrast with schizophrenia, interictal psychosis is associated primarily with visual rather than auditory hallucinations, better preserved affect and personality, less thought disorder and fewer negative symptoms, and more favorable outcome.^{35 ,37} Psychosis usually develops more than a decade after the onset of epilepsy.

Patients often ask, "Are seizures bad for my brain?" Most physicians respond no. The evidence, however, suggests that recurrent tonic-clonic and prolonged complex partial seizures can cause structural and functional damage.^{49 - 51} The long-term consequences of recurrent complex partial seizures remains uncertain, but in susceptible individuals, cognitive and behavioral disorders may develop.^{52 - 53} Gowers' intuition that seizures beget seizures is likely true in some cases.⁵⁴ Patients who have 20 seizures before therapy or who have an inadequate response to initial treatment with AEDs often develop refractory epilepsy.⁵⁵

Unfortunately, epilepsy is often a progressive disorder.⁵⁶ In animals, brief seizures and excitotoxic lesions induce progressive hippocampal neuronal loss and memory deficits.^{57 - 58} Long-standing TLE is associated with progressive hippocampal atrophy^{49 - 50 ,59} and slowly deteriorating cognition.^{60 - 61} Early structural injury can cause unilateral hippocampal atrophy, and subsequent generalized seizures can cause progressive neuronal loss.^{49 - 50 ,62} However, most patients with partial epilepsy show no MRI changes over brief intervals.⁶³ The long latency before development of intractable epilepsy suggests that if patients are at risk and the causative mechanisms are identified, preventive interventions may be found.¹⁷

Temporal lobe epilepsy is diagnosed primarily by history of paroxysmal behavioral events consistent with temporal lobe seizures. The stereotypical brief paroxysmal behavioral changes of partial seizures must be distinguished from similar symptoms that accompany medical and psychiatric disorders. As in the case of Mr C's paroxysmal abdominal discomfort in childhood, the diagnosis of partial epilepsy remains uncertain. Similar diagnostic challenges are illustrated in the following examples:

At 25 years of age, George Gershwin likely developed partial seizures manifest as abdominal discomfort. Extensive evaluations were unrevealing, and he was diagnosed with neurosis. Episodes persisted. At 38 years of age, he developed depression and episodic olfactory hallucinations, followed by unresponsiveness. Within 6 months, he died of a right temporal lobe glioma.⁶⁴

A 54-year-old woman with recurrent chest pain was admitted more than 40 times to exclude a myocardial infarction. Cardiac evaluation with angiography was repeatedly normal. Cardiac and gastrointestinal medications were ineffective. When specifically asked, she reported olfactory hallucinations. Sleep-deprived EEG revealed bitemporal epileptiform discharges. Antiepileptic drugs dramatically diminished chest pain episodes.⁶⁵

The electroencephalogram is critical in evaluation. Only a seizure recorded during an EEG confirms the diagnosis. Electroencephalographic evidence of epileptiform activity (spikes or sharp waves) over one or both temporal regions strongly supports diagnosis, but is neither necessary nor sufficient for diagnosis. Epileptiform activity appears in approximately 50% of initial waking recordings of adults with epilepsy and in about 85% of patients undergoing 2 recordings.⁶⁶ Sleep deprivation, hyperventilation, and photic stimulation increase the sensitivity, but repeatedly normal EEGs do not exclude epilepsy, and epileptiform activity occurs in 0.4% of adults and 1.5% to 3.5% of children without a history of seizures or other neurologic disorders.^{67 - 68} Normal patterns and variants misread as epileptiform contribute to the overdiagnosis of epilepsy.

Video-EEG monitoring simultaneously captures behavior and EEG and has revolutionized diagnosis. It both confirms the diagnosis and excludes comorbid disorders such as nonepileptic psychogenic seizures. However, only 25% of temporal lobe simple partial seizures have an EEG correlate. In contrast, more than 90% of temporal lobe complex partial seizures show paroxysmal changes on scalp EEG.⁶⁹

Magnetic resonance imaging is essential for evaluating patients with TLE.⁷⁰ It can identify common, surgically treatable lesions that are not visualized on CT scans, including mesial temporal sclerosis, low-grade gliomas, cortical dysplasia, and cavernous angiomas (Figure 1). When a patient with epilepsy reports that results of a prior CT scan or MRI of the brain were normal, the results should be confirmed. If seizure frequency or severity has increased since the last imaging study without identifiable cause, such as noncompliance, repeat MRI should be considered.

For all physicians, paroxysmal events should be diagnosed cautiously ( Article ). Errors and uncertainties are common, even among experts. Referral to an epilepsy center should be considered when seizures are not fully controlled, since the diagnosis of epilepsy or seizure type may be incorrect. Patients are frequently confused or unconscious during part of the event, and witnesses are often frightened, leading to inaccurate reports. Witnesses often overestimate episode duration and err in describing clinical details. Most jerks, stares, and unusual behaviors are not seizures. Nonepileptic paroxysms are common and can imitate any epileptic seizure.

Diagnostic

Misdiagnosis of epilepsy
Medical disorders
Syncope
Psychiatric disorders
Conversion Hyperventilation syndrome
Panic disorder
Neurologic disorders
Migraine
Transient ischemic attack Misdiagnosis of epilepsy syndrome (eg, generalized syn drome misdiagnosed as a partial syndrome)
Failure to identify progressive disorder (eg, tumor)

Therapeutic

Problems with compliance (may be related to disabling adverse effects such as sedating drugs, polytherapy, and peak dose effects)
Incorrect AED selection
Failure to titrate slowly to maximally tolerated AED dose
Failure to coordinate peak dose or level of AED and circadian susceptibility
Use of prescribed or over-the-counter medications that interact with AED or lower seizure threshold
Failure to consider surgical therapy

Lifestyle Issues
Sleep deprivation
Alcohol use

*AED indicates antiepileptic drug.

Psychogenic episodes and syncope are most often mistaken for epileptic seizures. Nonepileptic psychogenic seizures (NES) affect nearly 20% of patients diagnosed as having refractory epilepsy.⁷¹ Sexual and physical abuse and mild head injury are common antecedents. Diagnosis of NES often depends on video-EEG documentation because the history can be misleading and no single feature confirms or refutes the diagnosis. Further confusing the diagnosis, about 20% of patients with NES have a current or past history of epilepsy.⁷²

Syncope is likely if episodic loss or impairment of consciousness is precipitated by acute pain, anxiety, or standing upright or is accompanied by facial pallor and diaphoresis. It is a common error to diagnose as a seizure an episode of lost consciousness with brief tonic or clonic activity, especially if motor activity is more prominent on one side. When syncope reduces blood flow to the brain for a longer period, more intense and prolonged tonic-clonic activity can occur.⁷³

For the majority of TLE patients, medications are the mainstay of therapy (Table 2). Approximately 50% of patients respond to maximally tolerated doses of a single drug.^{55 ,74} If the first agent does not control seizures fully, another trial of monotherapy or combination therapy can be pursued. If adverse effects prevent attainment of high therapeutic levels, another trial of monotherapy is indicated.^{55 ,75} Mr C, like many patients, is very sensitive to adverse effects of medications. When seizures persist despite high plasma levels, a trial of 2 AEDs is appropriate. When monotherapy fails, 2 AEDs improve seizure control in nearly 40% of patients and fully control seizures in 3% to 10%.⁷⁶ Newer AEDs (eg, lamotrigine, levetiracetam, oxcarbazepine, topiramate, zonisamide) offer some patients improved seizure control or fewer adverse effects, although full seizure control is achieved in fewer than 10% when these agents are used to treat refractory partial epilepsy.^{76 - 77}

Managing the care of patients with refractory epilepsy is difficult. The morbidity of seizures and AEDs is often underestimated. For many physicians, patients are considered "stable" when they return for a 6-month follow-up and report 3 partial seizures a month and an occasional tonic-clonic seizure with missed medication. However, such patients cannot drive and have problems with employment, social life, and emotional and cognitive function. After a single drug fails, therapeutic challenges multiply. When a second drug is added, the costs and benefits must be assessed after some interval. For example, if a patient takes 14 pills per day instead of 6 and becomes more tired, dizzy, and constipated, and complex partial seizure frequency decreases from 3 to 2 per month, the cost exceeds the benefit. The decline in seizure frequency will not change that patient's life, but the increase in adverse effects will. Mr C has had extensive AED and psychotropic medication trials, but his physicians ultimately achieved an excellent outcome with a small medication burden. Overtreatment remains a common problem.⁷⁸ Benzodiazepines help stop prolonged seizures or seizure clusters but play a limited role in chronic therapy because adverse effects usually outweigh benefits.

Epilepsy surgery is an underused procedure that is often considered years or decades after seizures are medically refractory.When seizures or adverse effects of AEDs significantly impair quality of life in patients with partial epilepsy, surgery should be considered. Those with temporal lobe foci or localized structural lesions are the best candidates.⁷⁹ With surgery, approximately 70% of carefully selected TLE patients become seizure free, and many others enjoy significantly reduced seizure burden and improved quality of life.^{80 - 82} A randomized controlled study of 80 patients with TLE found that 58% were seizure free following surgery compared with only 8% of those treated with AEDs.⁸³ Adverse effects of surgery are primarily transient and disappear with time. Serious complications, including infections, hemorrhage, neurologic deficits, or psychiatric symptoms occur in 1% to 5% of patients; very rarely, death occurs.⁸⁴ Mr C's evaluation for surgery was timely but, unfortunately, revealed that his disorder is multifocal.

The vagus nerve stimulator (VNS) is an adjunctive therapy for patients with refractory partial epilepsy. This pacemakerlike device is implanted in the chest and stimulates the left vagus nerve. Stimulation parameters range from 30 seconds on and 5 minutes off to 7 seconds on and 14 seconds off. Transient hoarseness is the most common adverse effect. Other possible adverse effects include infection, voice alteration, throat or neck pain, cough, and dyspnea. These effects are reported as mild to moderate and can often be diminished by adjusting stimulation parameters.⁸⁵ In a controlled trial of 114 patients, approximately 20% of patients treated with a VNS enjoyed a greater than 50% reduction in seizure frequency compared with controls.⁸⁶ The VNS rarely provides full seizure control. In treatment-resistant depression, open studies found positive results,⁸⁷ but no placebo-controlled study has confirmed these findings. As a potential treatment for Mr C, if controlled studies show that the VNS has efficacy for depression, as demonstrated in epilepsy, this device would be an attractive option.

Mr C has a deep need to understand his disorder. However, dissecting the contributions of different factors to behavior is often impossible. His case exemplifies this conundrum: a strong family history of affective disorder, neuropsychiatric symptoms since childhood, environmental stressors, severe head injury, and epilepsy. Part of Mr C's—and his physicians'—journey is to understand their relative contributions. This could provide therapeutic insights and satisfy an intellectual thirst partly fueled by the disorder itself. Mr C's disorder is primarily biological. Society and medicine draw artificial lines between brain disorders that are labeled neurologic and psychiatric, causing troubling consequences. For example, most epilepsy and Parkinson disease patients take medication faithfully, while many depressed patients decline treatment. The misunderstanding of mechanism and therapy is that a physical disorder warrants therapy while an emotional disorder does not. Mr C has a complex limbic system disorder.

Mr C's care is exemplary. Therapeutic goals have been largely attained—controlling depression and seizures, minimizing his psychotropic drug and AED burden, and maximizing his function and quality of life. His care highlights the value of sensitive ears and creative minds on the parts of physicians. The future for Mr C will hopefully identify more effective and safer therapies for his seizure, cognitive, and emotional disorders.

DR J (MR C'S NEUROLOGIST): The memory disorder that Mr C displays is commonly seen in this population—that is, "enhanced forgetting." Can you comment on the mechanisms behind that phenomenon?

DR DEVINSKY: The concept of enhanced forgetting, which is mainly a retrieval issue, is one of the major functions of the temporal lobe. Information from the neocortex funnels into the hippocampus via the entorhinal cortex and the parahippocampal gyrus. The temporal neocortex is critical. In addition, for autobiographical memories, the right temporal lobe, which was the most contused area of Mr C's brain, may be a very critical structure. Some remarkable case reports appear in the literature describing individuals with predominantly right temporal neocortical damage whose verbal memory is quite intact, as is their retention of factual information. However, they have tremendous difficulty, as Mr C eloquently describes, retaining autobiographical information.⁸⁸ Enhanced forgetting is common in people with temporal lobe epilepsy. Neuropsychological tests during a 30-minute interval reveal problems; tests during an 8-week interval uncover much greater problems.⁴⁷

DR N (MR C'S PSYCHIATRIST): An element in the history, for those who are interested, is that the depression that Mr C experiences is also mirrored by times when he goes too fast. One question I've had with him over the years is whether this is some sort of psychiatric disorder, and if so, what? Other than a mood disorder due to epilepsy, what would your suggestion be?

DR DEVINSKY: The question is, who's the chicken and who's the egg? If I had to speculate, I would say that the primary psychiatric disorder was major depression, possibly with psychotic features, that led to the initial symptoms, the suicide attempt, and the major head injury. After the major head injury, the neurologic features have predominated. In addition to a strong genetic predisposition and affective disorder, Mr C acquired structural brain disease, probably on both sides, with epilepsy and medication now superimposed. At this point, teasing it out becomes less of the issue—although I think we all intellectually want to understand it. It would be nice to be able to say that his memory problem is "X" percent head injury, "Y" percent epilepsy, and "Z" percent medication. Obviously, those fluctuate, but they all are interwoven. For Mr C, I believe that the neurologic illness at this point predominates, although I think the psychiatric one is present too.

Of note, we observed all of our patients in the epilepsy monitoring unit who developed postictal psychosis.^{36 ,89} They had a cluster of a few complex, partial, or tonic-clonic seizures and, 2 to 72 hours later, became psychotic. Apart from known risk factors, such as seizure clusters, bitemporal epileptiform discharges, or structural abnormalities (eg, head injury or encephalitis), the biggest factor in predicting postictal psychosis was a family history of major depressive disorder. So these biological risk factors may tremendously influence the neurologic ones.

A PHYSICIAN: Mr C plays a musical instrument. George Gershwin was obviously a brilliant musician. Could you comment on TLE and musical ability?

DR DEVINSKY: We think of the right temporal lobe as a very important structure in musical processing,⁹⁰ and, despite its injury, Mr C has remained musically talented. However, he describes problems in which he needs to create a cognitive superstructure to learn musical pieces. Many people are destined for brilliance, and some of them get epilepsy as well. It may hold them back. In Gershwin's case it didn't. Despite epilepsy and a right temporal lobe glioma, he continued writing masterpieces until the last 6 months of his life.⁶⁴ Did epilepsy fuel his brilliance? Possibly, but the evidence is stronger in the case of Dostoyevsky. In letters to his brother, he wrote eloquently that after seizures he became more motivated and inspired and thought that he wrote better than he did at any other time.⁹¹ But it's difficult to know for certain.