Epileptic or Nonepileptic Seizure?

HISTORY

• A right-handed woman aged 19 years is referred for intractable clinical events.
• The patient's medical history is remarkable for anxiety, depression, and attention-deficit/hyperactivity disorder; migraine headache; and Lyme disease for which she received 6 months of intravenous antibiotic therapy.
• She has no recognized epilepsy risk factors.
• In her interview, the patient and her mother describe multiple types of clinical events. Her mother recalls episodes when, during the patient's childhood, she would observe the patient sit up in bed while in a state of unresponsiveness that lasted for several minutes. In grade school, the patient recalls frequent events of loss of time, although no observers recognized these spells or loss of responsiveness. The patient also reports that for the past 2 years before presentation, she has had frequent spells of feeling "overwhelmed, as if I cannot breathe." She also reports that while reading a book on pseudoseizures, she experienced déjà vu. Several of these events proceeded to convulsive activity lasting minutes. Following the event, the patient is described as "emotional" and "agitated." She does not experience incontinence or tongue biting. The patient insists she has preserved awareness during these clinical events. Her mother agrees.
• The spells became more frequent despite treatment with carbamazepine and levetiracetam.
• Results of the neurologic examination are normal.
• MRI results using a seizure protocol are normal.
• Multiple EEGs during wakefulness were performed and were interpreted as normal. One sleep-deprived EEG was interpreted as demonstrating potentially epileptogenic activity over the right temporal head region. The Figure shows a sample of this record.

Is the patient experiencing epileptic or nonepileptic clinical events?

DIAGNOSIS: Secondarily generalized convulsive seizures without loss of consciousness

Recurrent paroxysmal clinical events may be epileptic or nonepileptic. Discerning the correct diagnosis on the basis of the clinical history, EEG, and neuroimaging is often difficult. Indeed, nonepileptic events are diagnosed in 20% to 30% of patients seen at epilepsy referral centers for medically intractable epilepsy; the vast majority of these patients have psychogenic nonepileptic spells (PNES).^1-3

Careful characterization of the clinical events and the patient's interactions in the interview may provide important clues leading to suspicion of epilepsy or PNES. However, none of these have 100% specificity or sensitivity. Physicians need to note that³:

• PNES tends not to occur in sleep, whereas epileptic seizures may awaken patients from sleep.
• Epileptic seizures are more likely to occur without the patient's knowledge. This is especially helpful when an observer reports events for which the patient is completely amnestic.
• PNES may be prolonged, lasting over 15 minutes to hours. Seizures typically last seconds to 3 minutes.
• Patients with PNES may report multiple nonstereotyped auras, involving different sensory systems (for example, smell and sight).
• Incontinence and significant injury, especially severe tongue trauma, are much more commonly associated with epileptic seizures, whereas PNES may be accompanied by bruises, "rug burns," and other minor injuries.
• Coexistent diagnoses of anxiety or depression, as in this patient, and chronic pain or chronic fatigue may be more common in PNES.

In this patient, the clinical events are indeterminate. The spells described as feeling "overwhelmed, as if I cannot breathe" may represent simple partial seizures or anxiety or panic disorder. Preserved awareness during her recent generalized convulsive events and other signs and symptoms are suspect for PNES; however, a more complete examination is due.

PERILS OF READING EEGs

Interictal epileptiform discharges (IEDs) are stereotyped deflections on the EEG record generated by a population of pathologically synchronized bursting neurons. The following are definitions of IEDs⁴:

• Epileptiform sharp waves and spikes arise abruptly out of the EEG background activity because of an abrupt change in voltage polarity. The duration of this change in voltage polarity is between 70 and 200 milliseconds for sharp waves and less than 70 milliseconds for spikes, although the clinical importance of this distinction is not clear.
• Epileptiform spikes and sharp waves are usually followed by a slow wave that disrupts the EEG background rhythm.
• Epileptiform spikes and sharp waves have a physiologic field and are not confined to a single electrode (as may be seen in electrode artifact).

The most common errors of interpretation of EEGs include benign, frequently nonspecific variant patterns that may have epileptiform morphology but are not epileptogenic in nature.⁵ Many of these benign variants are seen during drowsiness or light (stage 1) sleep and disappear in stage 2 sleep.

The wicket waves seen over the right temporal head region on this patient's EEG are a normal variant--distinct from IEDs--and do not suggest a diagnosis of epilepsy.⁶ They consist of single spike-like waveforms or runs of monophasic µ-like waveforms with a frequency of 6 to 11 Hz. They are referred to as "wickets" because the rounded EEG waves resemble croquet wickets (depending on one's perspective, a shape of the form » or ).⁷ Wicket waves are seen over the temporal head regions during drowsiness and light sleep and become apparent when the alpha rhythm and other features of the EEG in wakefulness drop out. They may be asymmetric, occurring more frequently on one side, usually the left. They mainly are seen in older patients but may be seen in patients who are in their second or third decade.

When a wicket wave is seen as a single waveform, as in this patient (see Figure) it may appear to be a temporal IED. If one carefully analyzes the waveform, however, it is apparent that the wicket wave has a similar morphology to longer runs of wicket waves that are invariably seen over the left and right temporal head region elsewhere on a prolonged sleep-deprived record.⁸ The wicket wave also is not accompanied by an after-coming slow wave or distortion of the background, as is seen in IEDs.⁷

In PNES, the routine EEG record is typically normal. Nonspecific nonepileptic background abnormalities are not uncommon. In these patients, wicket waves as well as nonspecific normal variants are frequently misinterpreted as demonstrating potentially epileptogenic activity.^3,5 Indeed, in one recent study,⁸ electroencephalographers at an epilepsy center reread EEGs of patients referred for epilepsy management and identified patients with normal variant wicket waves, as seen in the Figure. On second look, the majority of these patients were found to not have epilepsy.

VIDEO-EEG MONITORING: THE GOLD STANDARD

In patients such as ours with indeterminate clinical events, avoiding an inappropriate diagnosis of epilepsy is of critical importance. The goal of video-EEG recording is to document a patient's typical clinical event. This frequently requires that the video recording be reviewed by an observer familiar with the patient's typical event. Both the clinical semiology of the event and the EEG correlate are important.

For many patients, very high-voltage movement artifact may obscure the EEG record during a clinical event, complicating interpretation of the EEG. Alternatively, movements (for example, side-to-side head movement) may generate rhythmic artifact on the EEG record that mimics an electrographic seizure. Complicating this interpretation further is that in simple partial seizures (or auras), as well as in some seizures of frontal or neocortical onset, no change in the EEG may occur.⁹

For these reasons, the video recording of the ictal semiology is frequently at least as important to the electroencephalographer as the ictal EEG in the accurate interpretation of the patient's clinical event. Numerous clinical features have been described from video-EEG recordings that may suggest a diagnosis of PNES, although none of these clinical features is without exception in our experience. These exceptions include^3,5:

• Discontinuous activity ("stop-start").
• Irregular or asynchronous (out-of-phase) activity.
• Side-to-side ("no-no") head movement.
• Pelvic thrusting.
• Ictal stuttering.
• Ictal weeping.
• Preserved awareness during bilateral motor activity.

RESULTS OF VIDEO-EEG MONITORING IN THIS CASE

• The interictal EEG record was normal.
• The patient experienced multiple clinical events during prolonged video-EEG recording. Two of the events were characterized by focal semiology not described in the patient's interview. They consisted of forced right head turn; right gaze; and right facial contraction, proceeding to generalized, brief tonic-clonic movements. Immediately on cessation of the convulsion, she would gesture to her eyes and ears as if to say, "I can see and hear," but was unable to speak for several minutes. Subsequently, she demonstrated preserved awareness throughout the event. One of these events had an EEG correlate of a rhythmic seizure discharge, maximal over the left hemisphere.

As well as these events, the patient experienced her stereotyped clinical events of feeling as if she "could not breathe," as well as episodes of déjà vu, accompanied by a prominent sinus tachycardia. No EEG correlate was seen for these clinical events. Nonetheless, in this clinical context, these events were interpreted as compelling evidence of simple partial seizures.

Although surprising, the finding that consciousness may indeed be preserved during bilateral motor seizures has been described previously.¹⁰ In our patient, right-sided tonic posturing was followed by bilateral tonic then clonic movements. We hypothesize from this that her recorded seizures were confined to extratemporal regions representing motor function bilaterally and that seizure activity did not propagate more diffusely to those areas of the brain essential for maintaining consciousness and memory. (For a review of the neurophysiology of consciousness and seizures, see Blumenfeld H, Taylor J. Why do seizures cause loss of consciousness? Neuroscientist. 2003;9:301-310).

• Clinical events ceased when the patient was put on lamotrigine therapy.

In this patient, video-EEG recording was essential to a correct diagnosis. The case illustrates how video-EEG recording helps distinguish PNES from epileptic seizure and reminds us that every patient with a presumed diagnosis of PNES deserves video-EEG monitoring to document their clinical event.

REFERENCES1. Benbadis SR, Allen Hauser W. An estimate of the prevalence of psychogenic non-epileptic seizures. Seizure. 2000;9:280-281.
2. Benbadis SR, O'Neill E, Tatum WO, Heriaud L. Outcome of prolonged video-EEG monitoring at a typical referral epilepsy center. Epilepsia. 2004;45: 1150-1153.
3. Benbadis SR. The EEG in nonepileptic seizures. J Clin Neurophysiol. 2006;23: 340-352.
4. Worrell GA, Lagerlund TD, Buchhalter JR. Role and limitations of routine and ambulatory scalp electroencephalography in diagnosing and managing seizures. Mayo Clin Proc. 2002;77:991-998.
5. Benbadis SR, Tatum WO. Overinterpretation of EEGs and misdiagnosis of epilepsy. J Clin Neurophysiol. 2003;20:42-44.
6. Klass DW, Westmoreland BF. Nonepileptogenic epileptiform electroencephalographic activity. Ann Neurol. 1985;18:627-635.
7. Westmoreland BF. Epileptiform electroencephalographic patterns. Mayo Clin Proc. 1996;71:501-511.
8. Krauss GL, Abdallah A, Lesser R, et al. Clinical and EEG features of patients with EEG wicket rhythms misdiagnosed with epilepsy. Neurology. 2005; 64:1879-1883.
9. Devinsky O, Sato S, Kufta CV, et al. Electroencephalographic studies of simple partial seizures with subdural electrode recordings. Neurology. 1989;39: 527-533.
10. Bell WL, Walczak TS, Shin C, Radtke RA. Painful generalised clonic and tonic-clonic seizures with retained consciousness. J Neurol Neurosurg Psychiatry. 1997;63:792-795.