Keywords
The erroneous diagnosis of epilepsy is common. At a typical epilepsy center, 20% to 40% of patients previously diagnosed with epilepsy, and whose seizures are not responding to drugs, are found to be misdiagnosed.
1
, 2
, 3
Most patients who are misdiagnosed as having epilepsy are eventually shown to have psychogenic nonepileptic seizures (PNES), or (more rarely) syncope4
or parasomnias.5
Occasionally, other paroxysmal conditions can be misdiagnosed as epilepsy, but they are uncommonly seen in a seizure monitoring unit (SMU). Once the diagnosis of seizures is made, it is easily perpetuated without being questioned, which explains the usual diagnostic delay6
, 7
and its cost.8
, 9
, 10
PNES are time-limited, paroxysmal changes in movements, sensations, behaviors, or consciousness, that can resemble epileptic seizures, but they are not associated with epileptiform activity. Of the 1% of the US population diagnosed with seizures, presumed to be epilepsy, 5% to 20% have PNES.
11
On average, 7 years elapse between a patient’s onset of PNES and the correct diagnosis.6
The misdiagnosis of PNES is costly to patients, the health care system, and to society. Repeated workups and treatments for what is mistakenly believed to be epilepsy are estimated to incur $100 to $900 million per year in medical services.9
Patients with PNES are prescribed antiepileptic drugs (AEDs) that do not treat, and may exacerbate PNES,12
have multiple tests performed, and may not receive the necessary mental health care that could benefit them. Delayed diagnosis could lead to adverse effects from unneeded AEDs, iatrogenic complications from invasive procedures in continuous PNES (nonepileptic psychogenic status),13
medical costs from unnecessary hospitalization treatment and workup, delayed referral to appropriate psychiatric treatment, and employment difficulties and disability.14
The first step in PNES treatment is proper diagnosis. Video electroencephalography (EEG) remains the gold standard for PNES diagnosis. Certain seizure types, such as those seen in frontal lobe epilepsy (FLE), may mimic PNES semiology, and, conversely, ictal characteristics of PNES may resemble epileptic seizures (ES). New diagnostic techniques may help distinguish stereotypic semiology seen in FLE that are not seen in PNES. Bedside observations may also be of benefit in augmenting the video EEG interpretation to establish the PNES diagnosis. The use of other diagnostic measures to augment video EEG diagnosis is examined in this article. The safety of discontinuing AEDs in lone NES is discussed.
Nonepileptic seizures (NES) can be physiologic or psychogenic in origin and can be difficult to distinguish from ES, with both seizure types showing alterations in behavior, consciousness, sensation, and perception.
15
Recent research has yielded clinically useful differentiating features at bedside and on video EEG. Appropriate diagnosis then informs potential treatments.Diagnosis: distinguishing NES from ES
The diagnosis and treatment of patients with PNES has long confounded neurologists, psychiatrists, and emergency physicians. As an adjunct to anamnesis and video EEG, ictal semiology, neurophysiologic tests, patient characteristics, and neuropsychological testing contribute to making the diagnosis of PNES.
Differentiating NES from ES is the first step in appropriate treatment.
16
PNES can appear similar to ES, and, to distinguish the 2 types of seizures, the gold standard is video EEG.17
Because there is no tissue confirmation against which to measure the accuracy of video EEG, the next best measure is interrater reliability (IRR). Video EEG has been shown to have substantial IRR for epilepsy and moderate IRR for PNES, as discussed later.18
The IRR would almost certainly be higher when incorporating supplemental information including history, physical, and more ictal segments. Other techniques can be used as adjuncts to make the diagnosis of NES,19
but admission to an SMU is the key. Video EEG not only provides a definitive diagnosis in almost 90% of patients but also rectifies an incorrect diagnosis of epilepsy, and results in treatment change in 79% of patients.20
Monitoring a patient with seizures in the SMU may also help to identify the 10% of patients with PNES who also have epilepsy.21
Higher numbers of mixed ES/PNES were given in the literature in the past, but these seem to have been overestimates based on less-stringent criteria for differentiating epilepsy from PNES.Some physical observations of the ictal semiology used in differentiating PNES from epilepsy are noted later (Table 1).
22
Table 1Behaviors to distinguish between psychogenic nonepileptic and epileptic seizures
Reproduced from Benbadis SR, LaFrance WC Jr. Clinical features and the role of video-EEG monitoring. In: Schachter SC, LaFrance WC Jr, editors. Gates and Rowan's Nonepileptic Seizures. 3rd edition. Cambridge: Cambridge University Press; 2010. p. 38–50. Chapter 4; with permission. Data from Refs.
22
, 23
, 24
, 25
, 26
, 27
, 28
, 29
, 30
, 31
, 32
, 36
, 37
, 38
, 39
, 40
| Observation | PNES | ES |
|---|---|---|
| Situational onset | Common | Rare |
| Gradual onset | Common | Rare |
| Precipitated by stimuli (noise, light) | Occasional | Rare |
| Purposeful movements | Occasional | Very rare |
| Opisthotonus (arc de cercle) | Occasional | Very rare |
| Tongue biting (tip) | Occasional | Rare |
| Tongue biting (side) | Very Rare | Common |
| Prolonged ictal atonia | Occasional | Very rare |
| Vocalization during tonic-clonic phase | Occasional | Very rare |
| Reactivity during unconsciousness | Occasional | Very rare |
| Rapid postictal reorientation | Common | Unusual |
| Undulating motor activity | Common | Very rare |
| Asynchronous limb movements | Common | Rare |
| Rhythmical pelvic movements | Occasional | Rare |
| Side-to-side head shaking | Common | Rare |
| Ictal crying | Occasional | Very rare |
| Ictal stuttering | Occasional | Rare |
| Postictal whispering | Occasional | Not present |
| Closed mouth in tonic phase | Occasional | Very rare |
| Closed eyelids during seizure onset | Very common | Rare |
| Convulsion >2 min | Common | Very rare |
| Resisted lid opening | Common | Very rare |
| Pupillary light reflex | Usually retained | Commonly absent |
| Cyanosis | Rare | Common |
| Ictal grasping | Rare | Occurs in FLE and TLE |
| Postictal nose rubbing | Not present | Can occur in TLE |
| Stertorous breathing postictally | Not present | Common |
| Self-injury | May be present (especially excoriations) | May be present (especially lacerations) |
| Incontinence | May be present | May be present |
Abbreviations: FLE, frontal lobe epilepsy; TLE, temporal lobe epilepsy.
Eye and Facial Findings
Using data from video EEG monitoring, researchers found that 50 of 52 patients with PNES (96%) closed their eyes during the seizure, compared with 152 of 156 of patients with ES (97%), who had their eyes open at the beginning of their seizure.
23
This information may help clinicians differentiate between PNES and ES, particularly when the 2 types of seizures occur in the same patient. Also, other observers, such as family members, could report to physicians whether the patient’s eyes were open or closed during the ictal event. However, this observation has been challenged by other investigators, who prospectively assessed whether observer or self-report eye closure could predict NES, before video EEG monitoring.24
In the monitoring unit, 112 met study criteria and had either PNES (n = 43, 38.4%) or epilepsy (n = 84, 75%). The investigators recorded eye closure as a percentage of episode duration, rather than the previously studied dichotomous, absent or present. Self-report of eye closure more accurately predicted actual videorecorded eye closure than observer report. The study confirmed that video-recorded eye closure was 92% specific for PNES identification, but not as sensitive (only 64%) as previously reported.Patients with PNES may also exhibit geotropic eye movements, in which the eyes deviate downward to the side that the head is turned.
25
Eyelids are typically closed for a longer duration (20 seconds) compared with temporal lobe epilepsy (TLE) or FLE (∼2 seconds).26
Weeping is also a characteristic with PNES.27
, 28
Ictal stuttering and postictal whispering voice are seen in PNES.
29
, 30
Postictal nose rubbing and cough have been observed in TLE but not in PNES.31
Similarly, noisy or stertorous breathing can be seen after ictus in epileptic convulsions but was not observed following PNES convulsions.32
Although it helps to differentiate convulsive epilepsy from convulsive PNES, this finding does not apply to partial seizures.Pelvic thrusting is reportedly as common in FLE as in PNES.
33
, 34
, 35
Other ictal features associated with PNES are out-of-phase or side-to-side oscillatory movements or chaotic and disorganized thrashing.15
In contrast, seizures of FLE typically arise from sleep, are brief, and often involve vocalization and quick tonic posturing.36
, 37
Occasionally, whole body trembling may be observed with PNES. These behaviors may wax and wane (including stop and go), and vary in direction and rhythm, which is atypical for ES (ES tend to follow a stereotyped evolution).Physical injury during an ictus was once believed to occur only in patients with epilepsy, but research shows that more than 50% of patients with PNES are injured during seizures.
38
The character of the injury is helpful in differentiating ES from PNES. Excoriations on long bone surfaces, such as the arm, leg, or cheek, are seen in PNES,39
as opposed to lacerations from epilepsy. Tongue biting, self-injury, and incontinence are commonly associated with ES but are also reported by up to two-thirds of patients with PNES, rendering these signs less specific than was once believed,40
especially when they are reported rather than documented. Objectively documented injuries or incontinence are of higher diagnostic value (specificity).Observing what the patients bring into the SMU also has some value. One group found that, of those admitted for monitoring, patients with PNES brought a toy stuffed animal with them.
41
In their study, 381 patients with PNES were compared with 453 patients with epilepsy. Of 23 patients (2.5%) who had toy animals during admission, 20 were diagnosed with PNES, and 3 were diagnosed with epilepsy (P<.001). The 3 patients with epilepsy had a history of a psychiatric disorder. Sensitivity was 5.2% and specificity was 99.3%, with a positive predictive power of 87%, and a negative predictive power of 55%. The investigators proposed that such behaviors may represent nonverbal expressions of attachment desires, dependency needs, or other psychological traits.Diagnostic measures
EEG
NES diagnosis is most accurately established by coregistering EEG neurophysiologic testing with video. Video EEG, in which the patient’s seizure is observed visually with simultaneous EEG, allows data about behavior to be coupled with EEG rhythms. With the history and examination, the absence of expected epileptiform patterns before, during, and after the ictus points to a NES diagnosis. Occasionally, EEG-negative epilepsy on scalp EEG occurs, in which a simple partial seizure, a frontal lobe epileptic seizure, or a deep temporal lobe epileptic seizure does not generate an ictal epileptiform pattern.
42
The EEG can also be uninterpretable because of movement or electromyographic artifact. Without video EEG, neurologists’ ability to differentiate ES from NES by history alone has a specificity of 50%.43
One study described a method for diagnosing frontal lobe epileptic seizures by comparing the video EEGs in a synchronized, side-by-side view.
44
Split-screen synchronized display was found to be a simple and valid technique for studying and presenting particular semiological aspects of ES. Using this methodology to diagnose NES may also be of value. Research suggests that magnetoencephalography (MEG) may be useful for identification and localization of FLE,45
and MEG studies are attempting to develop optimal procedures for localizing interictal epileptiform discharges of patients with localization-related FLE. For the scalp EEG–negative cases in which the differential diagnosis of FLE versus PNES is present, future research could examine the potential to screen for spikes in FLE that are not seen in PNES.Neurohumoral Measures
The use of serum prolactin (PRL) drawn within 30 minutes of the ictus onset is helpful for differentiating generalized tonic-clonic (GTC) epileptic seizures and complex partial epileptic seizures (CPS) from PNES, as summarized in a report from the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology.
46
Trimble47
first showed that GTC epileptic seizures, but not PNES, raised serum PRL. Pooling the available data of the 10 studies meeting inclusion criteria, the subcommittee investigators found a sensitivity of 60% for GTC and 46% for CPS, and a specificity of 96% for both. They found a positive predictive value of 93% to 99%. Cragar and colleagues19
similarly found lack of PRL increase has an average 89% sensitivity to PNES. Clinically, this translates into a strong confirmation of a diagnosis of ES when an increased PRL is found in patients with GTC- or CPS-like events suspected of being PNES. The investigators concluded that an increase in serum PRL level is probably a useful adjunct to differentiate GTC or CPS from PNES. However, PRL is of limited usefulness for differentiating FLE of any semiology from PNES.Neuroimaging
Structural neuroimaging abnormalities neither confirm nor exclude ES or PNES. PNES can occur in the presence of structural lesions,
48
and about 10% of patients with PNES alone have structural abnormalities on magnetic resonance imaging scans.49
Functional imaging is not useful for the diagnosis of PNES versus FLE. A negative ictal single-photon emission computed tomography (SPECT) scan does not imply a diagnosis of PNES nor does an abnormal scan mean epilepsy is present. A small series of ictal and interictal SPECT scans of patients with PNES revealed a few scans with lateralized perfusion abnormalities, but the findings did not change when the ictal and interictal images were compared.50
In contrast, patients with epilepsy have dynamic changes when ictal and interictal changes are compared on functional neuroimaging.Characteristics of patients with PNES and epilepsy
Neuropsychological Measures
Many studies exist describing the cognitive, emotional, personality, and psychomotor differences between the ES and PNES groups. Cragar and colleagues
19
reviewed the literature on adjunctive tests for diagnosing PNES and reported sensitivity and specificity of the different measures. A summary of their findings noted that patients with ES and PNES perform roughly the same on neuropsychological (NP) measures but worse than healthy controls. PNES tended to perform better than patients with ES on certain NP tests, as described later.A summary of studies examining intelligence, psychomotor function, motivational measures, and personality features in PNES
51
suggests that, for cognitive measures, patients with ES and PNES show no significant differences on tests of intelligence, learning, and memory but score lower than healthy control subjects.52
On psychomotor measures, patients with PNES show reduced motor speed and grip strength, compared with healthy controls.53
Motivational measures reveal that patients with PNES score lower than patients with ES on some motivational measures, perhaps reflecting a lack of psychological resources necessary to persist with a challenging NP battery. Some studies show comparable failure rates in PNES and ES groups on symptom validity batteries. Frank malingering is believed to occur rarely in PNES,54
, 55
, 56
but malingering is probably underdiagnosed in general, because it is an accusation rather than a diagnosis. The Minnesota Multiphasic Personality Inventory (MMPI) has been used for more than 20 years in assessing patients with PNES. Personality testing performed with instruments such as MMPI-2 studies show increases in hypochondria, hysteria, and depression scores in PNES.19
, 57
Intelligence Measures and Cognitive Testing
Comparing patients with PNES with those with ES, Binder and colleagues
52
found no significant differences on tests of intelligence or learning and memory, including the Wechsler Adult Intelligence Scale - Revised, Wisconsin Card Sort Test, or Rey Auditory Verbal Learning Test. Control subjects were significantly superior to the PNES and the ES group. Bortz and colleagues58
studied the California Verbal Learning Test results in patients with PNES and ES and the investigators suggested that “failure to explicitly recognize words following repeated exposure” may be reflective of a negative response bias and psychological denial in patients with PNES.Psychomotor Measures
Kalogjera-Sackellares and Sackellares
53
evaluated patients with PNES compared with matched normal controls and found reduced motor speed and grip strength in the patients with PNES. Some have interpreted this as a manifestation of motivation, which is discussed later. Dodrill and Holmes59
reported that patients with PNES performed better than those with ES on measures from the Halstead-Reitan Battery, with differences between Tactual Performance Test, Seashore Tonal Memory, and Trail-making Part B. Although finger tapping and grooved pegboard differed between controls compared with PNES and ES groups, Binder and colleagues52
did not find differences between the ES and PNES groups on these measures.Motivational Measures
Motivational tests include the Portland Digit Recognition Test (PDRT), the Test of Memory Malingering (TOMM), and others, and are used to detect inadequate performance on NP testing. The presence of unconscious psychological stress is hypothesized as an explanation for variable effort in patients with PNES.
60
Binder and colleagues52
, 54
found that patients with PNES performed poorly compared with patients with ES on the PDRT. The investigators noted that frank malingering occurs rarely in PNES, and that the poorer performance in PNES may reflect a lack of psychological resources necessary to persist with a challenging NP battery. More recent symptom validity tests of patients with PNES and patients with ES show discrepant findings. Drane and colleagues55
concluded that many patients with PNES do not put forth maximal effort on neuropsychological tests, based on the Word Memory Test failure rate in their sample (51.2% in NES, vs 8.1% in ES). More extensive batteries, including the Digit Memory Test, Letter Memory Test, TOMM, and PDRT, failed to show differences between the ES and PNES groups, with 22% of patients with epilepsy and 24% of patients with PNES performing suboptimally on 1 or more effort measures.56
Personality Testing: MMPI/MMPI-2 and Clinical Psychological Profiles
The MMPI has been used for more than 20 years in assessing patients with PNES. Most MMPI studies in PNES report the conversion V profile, with increases in scales 1 (Hs [hypochondriasis]) and 3 (Hy [hysteria]), and depressions in Scale 2 (D [depression]).
19
Cragar and colleagues19
also reported an average sensitivity of 70% and specificity of 73% to PNES diagnosis using MMPI-2 decision rules. Using the MMPI-2 and clinical variables yielded 80% sensitivity and specificity for PNES diagnosis between 57 patients with ES and 51 with PNES.57
Along with others, the study shows the MMPI-2 to be a useful adjunct to video EEG diagnosis.The dramatic personality of patients with PNES was illustrated in a blinded pilot study of artwork drawn by patients with ES and NES.
61
The investigators calculated that an 80% positive predictive value for PNES existed if subjects used 10 or more colors to draw their seizures. Galimberti and colleagues62
administered the cognitive behavioral assessment (CBA) psychometric battery to patients with lone NES, mixed ES/PNES, and ES controls. The CBA, which assesses personality characteristics and emotional adjustment, comprises scales rating introversion-extroversion, neuroticism, psychoticism, state-trait anxiety, psychophysiologic distress, and depressive and other anxiety symptoms, and found that the mean scores on the psychophysiologic distress scale for the PNES and the ES/PNES groups were higher than the mean scores of the ES control group.Family and Patient Traits
Studies comparing family functioning in patients with ES and PNES reveal that individuals with PNES consider their families to be more dysfunctional, particularly in regard to communication, and that family members of patients with PNES reported difficulties defining roles.
63
Individuals with PNES score higher on symptom checklists (a measure of somatic complaints) compared with other patients with seizures.64
Pain disorders are also common in patients with PNES. Among patients in epilepsy clinics, a diagnosis of fibromyalgia or chronic pain has an 85% positive predictive value for PNES.65
PNES could be described as a disorder of communication, in which internal distress is conveyed somatically rather than verbally.Patients with PNES have several psychiatric diagnosis comorbidities, including depression, anxiety, posttraumatic stress disorder, or a personality disorder. Bowman and colleagues
66
found that up to half of patients with PNES had 1 of these diagnoses on clinical interview using the DSM-III-R. We found similar results using DSM-IV criteria in patients with PNES.67
The simultaneous presence of depression, a history of abuse, and a personality disorder may portend a worse prognosis for patients with PNES.68
Comparisons between outcomes and prognosis in patients with ES and in patients with PNES could be of benefit to discern the effect of comorbidities on seizures.To summarize, compared with healthy controls, patients with ES and with PNES perform worse on several NP measures, but there are few differences between ES and PNES groups on tests that would reliably differentiate ES from PNES. The impairments are believed to be caused by at least 3 factors: (1) patients with both ES and PNES were on AEDs, which may affect cognition; (2) structural lesions in the patients with ES and in some of the patients with PNES and ES, and (3) emotional factors contributing to cognitive impairment in the PNES group.
60
The psychological makeup of patients with PNES seems to be that they have personalities with anxiety, cognitive, and somatic distress. Along with these comorbidities, they have difficulty expressing and communicating distress verbally; they express themselves somatically. Family dysfunction is also present in PNES.Overall, neuropsychological measures are not useful to make a diagnosis of PNES (vs epilepsy), because they are neither sensitive nor specific. However, psychological testing may be a useful adjunct once the diagnosis of PNES has been made, to characterize the psychopathology and mechanisms underlying the symptoms and clarify the psychiatric diagnosis (eg, somatoform disorder, dissociation).
Limitations and pitfalls of video EEG
In most cases, the diagnosis of PNES with video EEG is clear and can be made with a high degree of confidence. There are limitations to video EEG, and it is important to be familiar with them to avoid serious diagnostic errors.
Ictal EEG has limitations because it may be negative in simple partial ES
69
, 70
and in some complex partial ES, especially those of frontal lobe onset.67
Ictal EEG may also be uninterpretable or difficult to read if movements generate excessive artifact. Knowing what type of clinical seizures may be unaccompanied by ictal EEG changes is critical. The most common types of ES that are unaccompanied by ictal EEG changes are those without impairment of awareness (ie, simple partial ES). This type includes all simple partial ES with subjective phenomena (ie, auras), which can involve the 5 senses as well as psychic or experiential sensations.The other types of simple partial ES that are commonly unaccompanied by ictal EEG changes are brief tonic phenomena such as those typical of frontal lobe ES. These phenomena are typically brief (5–30 seconds) and tonic, and may be hypermotor, but not usually as dramatically flailing or thrashing as PNES. In hypermotor seizures in which semiology is suspected of being psychogenic, given that both FLE and PNES are scalp EEG–negative, it can be impossible to prove based on EEG that such episodes are psychogenic. Brief episodes of deja vu, or fear, or tonic stiffening with no EEG changes, can be epileptic. Conversely, in favor of PNES is when the events never progress to clear ES, and if there is suggestibility (triggering them with placebo maneuvers). This situation is similar to psychogenic movement disorders (PMD), in which the diagnosis rests solely on phenomenology (ie, there is no equivalent of the EEG in PMD), and response to placebo or suggestion is considered a diagnostic criterion for a definite psychogenic mechanism.
68
A solid rule is that psychogenic events do not occur out of physiologic sleep, so that events that arise out of EEG-verified sleep are related to neurologic disorders (ES or parasomnias). ES with altered awareness and no EEG changes are rare, and if the clinical events are strongly suggestive of seizures, it is best to err on the side of treating them as epileptic, at least initially. More recently, video split-screen techniques have been shown to be helpful in diagnosing ES.44
Lack of ictal EEG changes only indicates that the episodes are nonepileptic, and nonepileptic does not always mean psychogenic. Other diagnoses must be considered before making a diagnosis of PNES. The most common diagnoses to consider are physiologic nonepileptic events, such as syncope, for episodes that occur during waking, and parasomnias for episodes that occur in sleep. When syncope (convulsive or not) is recorded on video EEG, the EEG proceeds through a stereotyped pattern of changes (delta slowing and suppression caused by lack of cerebral blood flow).
71
, 72
Occasionally, in the absence of ictal EEG changes, the differentiation between seizure and parasomnia can be difficult. Hypnic jerks or sleep starts are benign myoclonic jerks that everyone has experienced on occasion. Although they resemble the jerks of myoclonic seizures, their occurrence only on falling asleep defines them as benign nonepileptic phenomena. Parasomnias are easily identified on video EEG by their occurrence in waking to stage 1 transition and having no EEG correlate associated with the jerks.73
Restless legs syndrome and periodic limb movements of sleep also may interfere with sleep; however, no ictal EEG changes are seen with these disorders.A common myth is that a recorded episode with a negative EEG is all that is required to make a diagnosis of PNES, and this is grossly inaccurate. A negative EEG can only be interpreted in the context of the semiology of the event in question. Thus, both the video and EEG must be available. (The diagnosis could be more accurate with video alone than with EEG alone, when differences in ictal semiology are classified by the informed observer.)
Unlike the definitive diagnosis of brain tumors, the closest test to a biopsy for distinguishing epilepsy from PNES would be intracranial monitoring. The risk and morbidity associated with craniotomy and grid or depth electrode placement outweighs the use in patients with a suspicion of PNES. In the absence of the definitive confirmation of the diagnosis, there is no way to prove that the PNES diagnosis is correct even when there is a high degree of certainty. Ramsay and colleagues
74
described the limitations of scalp EEG and reported the use of depth electrodes on scalp-negative EEGs. Subsequent EEG monitoring revealed that patients with epilepsy had an epileptic focus in either the mesial (n = 8) or inferior frontal (n = 2) areas.As noted earlier, we conducted a study of the interrater reliability of the diagnosis by video EEG,
18
sampling a group of epileptologists, and found that, for the diagnosis of epilepsy, there was substantial agreement (κ = 0.69, 95% confidence interval [CI] 0.51–0.86). For the diagnosis of PNES, there was moderate agreement (κ = 0.57, 95% CI 0.39–0.76). For physiologic nonepileptic events, the agreement was low (κ = 0.09, 95% CI 0.02–0.27). The overall κ statistic across all 3 diagnostic categories was moderate at 0.56 (95% CI 0.41–0.73).18
The investigators noted that the diagnosis in this study was, intentionally but artificially, based solely on video EEG recordings, which does not reflect clinical reality, whereas the actual diagnosis of PNES is made by a combination of patient history (neurologic and psychiatric), examination, and video EEG monitoring. The study underscored that there is a certain component of subjective, artful judgment. When used properly, video EEG allows the diagnosis of paroxysmal seizure–like events, and in particular the diagnosis of PNES, with a high degree of confidence.False Positives in EEG
Having a report of a prior abnormal EEG is a common problem. Many patients with PNES seen at epilepsy centers have had previous EEGs interpreted as epileptiform. A common error is that the episodic symptoms are not really suggestive of seizures (ie, nonspecific symptoms such as light-headedness, dizziness, and numbness), and the diagnosis of seizures is entirely based on the (over-read) EEG. In this situation, it is essential to obtain and review the tracing previously read as epileptiform, because no amount of normal subsequent EEGs will cancel the previous “abnormal” one. When reviewed, most will turn out to show normal variants that were overinterpreted as epileptiform.
75
, 76
, 77
However, obtaining prior EEGs can be difficult. First, records are not always available or accessible, and second, not all digital EEG systems are compatible. In this regard, software that allows one to read any digital EEG format is valuable. The most common errors in EEG interpretation, and the main source of over-reading, are benign temporal sharp transients, or wicket spikes.77
In general, the threshold for considering a sharp transient epileptiform must be high. Criteria include asymmetric contour (upslope steeper than downslope), different frequency and amplitude than ongoing background, diphasic or triphasic morphology, after-going slow wave, and disruption background. In addition, contrary to a common misconception, phase reversal is not one of the criteria. Phase reversals only reflect the maximum location of a discharge.75
, 76
, 77
In children, an additional issue is the frequent coexisting benign focal epileptiform discharges (BFEDC), which are frequently seen in asymptomatic children.The usefulness of AEDs for refractory seizures
Treatment of PNES is beyond the scope of this article
78
; however, given that patients with FLE can be refractory to AEDs, and that AEDs do not treat PNES79
and can exacerbate PNES,12
it does address the recent research on AEDs. One pharmacologic question that has been addressed recently has been the effect of withdrawing AEDs from patients with lone PNES. Given that most patients with PNES are prescribed AEDs, Oto and colleagues80
studied whether withdrawal of AEDs can be performed safely in patients with PNES in a prospective evaluation of safety and outcome. Seventy-eight patients with PNES who satisfied a standardized set of criteria for excluding the diagnosis of coexisting or underlying epilepsy had their AEDs withdrawn (64 as outpatients, 14 as inpatients). PNES frequency declined in the group as a whole during the period of the study (follow-up 6–12 months) in all individuals except for 8 patients in whom there was a transient increase. Fourteen patients reported new physical symptoms after withdrawal; however, no serious adverse events were reported. The investigators concluded that, with appropriate diagnostic investigation and surveillance during follow-up, withdrawal of AED can be achieved safely in patients with PNES.Summary
Differentiating PNES from FLE can be challenging, and the diagnosis of PNES can be difficult to distinguish by history alone. Physical signs, patient characteristics, and neuropsychological testing are helpful adjuncts to the video EEG to confirm the diagnosis. The first phase of treatment begins with the neurologist, as the findings of the video EEG monitoring are shared in a positive nonpejorative manner. Neurologists, psychiatrists, and psychologists must then continue to work together to diagnose comorbidities and effectively treat this difficult population. Interdisciplinary research and discussion, along with collaborative sponsorship between neurologic and psychiatric institutes, will help move the field forward to address these difficult-to-treat populations with seizures.
References
- Outcome of prolonged video-EEG monitoring at a typical referral epilepsy center.Epilepsia. 2004; 45: 1150-1153
- The misdiagnosis of epilepsy and the management of refractory epilepsy in a specialist clinic.QJM. 1999; 92: 15-23
- The misdiagnosis of epilepsy: findings of a population study.Seizure. 1998; 7: 403-406
- Misdiagnosis of epilepsy: many seizure-like attacks have a cardiovascular cause.J Am Coll Cardiol. 2000; 36: 181-184
- Movement disorders in sleep: guidelines for differentiating epileptic from non-epileptic motor phenomena arising from sleep.Sleep Med Rev. 2007; 11: 255-267
- Diagnostic delay in psychogenic nonepileptic seizures.Neurology. 2002; 58: 493-495
- Non-epileptic seizures: patients’ understanding and reaction to the diagnosis and impact on outcome.Seizure. 2003; 12: 287-294
- Epilepsy: a costly misdiagnosis.Clin Electroencephalogr. 1997; 28: 225-228
- Improved health care resource utilization following video-EEG-confirmed diagnosis of nonepileptic psychogenic seizures.Seizure. 1998; 7: 385-390
- Nonepileptic seizures treatment workshop summary.Epilepsy Behav. 2006; 8: 451-461
- The classification and treatment of nonepileptic events.in: Devinsky O. Theodore W.H. Epilepsy and behavior. Wiley-Liss, New York1991: 251-263 (Chapter 18)
- Classical hysterical seizures facilitated by anticonvulsant toxicity.Psychiatr Clin (Basel). 1970; 3: 71-84
- Nonepileptic psychogenic status: markedly prolonged psychogenic nonepileptic seizures.Epilepsy Behav. 2010; 19: 65-68
- Avoiding the costs of unrecognized psychological nonepileptic seizures.Neurology. 2006; 66: 1620-1621
- Ictal characteristics of pseudoseizures.Arch Neurol. 1985; 42: 1183-1187
- Treatment of nonepileptic seizures.Epilepsy Behav. 2002; 3: S19-S23
- Video-EEG telemetry can be a crucial tool for neurologists experienced in epilepsy when diagnosing seizure disorders.Seizure. 2004; 13: 32-34
- Interrater reliability of EEG-video monitoring.Neurology. 2009; 73: 843-846
- A review of diagnostic techniques in the differential diagnosis of epileptic and nonepileptic seizures.Neuropsychol Rev. 2002; 12: 31-64
- Diagnostic utility of an epilepsy monitoring unit.Am J Med Qual. 2007; 22: 117-122
- How many patients with psychogenic nonepileptic seizures also have epilepsy?.Neurology. 2001; 57: 915-917
- Clinical features and the role of video-EEG monitoring.in: Schachter S.C. LaFrance Jr., W.C. Gates and Rowan’s Nonepileptic Seizures. 3rd edition. Cambridge University Press, Cambridge2010: 38-50 (Chapter 4)
- Ictal eye closure is a reliable indicator for psychogenic nonepileptic seizures.Neurology. 2006; 66: 1730-1731
- Do observer and self-reports of ictal eye closure predict psychogenic nonepileptic seizures?.Epilepsia. 2008; 49: 898-904
- A diagnostic sign in states of apparent unconsciousness.Lancet. 1978; 2: 920-921
- Eyelids position - during epileptic versus psychogenic seizures [abstract: OPL145].J Neurol Sci. 2005; 238: S82-S83
- Closed eyes during a seizure indicate psychogenic etiology: a study with suggestive seizure provocation.J Epilepsy. 1996; 9: 165-169
- Weeping as a common element of pseudoseizures.Arch Neurol. 1993; 50: 1059-1060
- Ictal stuttering: a sign suggestive of psychogenic nonepileptic seizures.Neurology. 2004; 63: 516-519
- Postictal behaviors associated with psychogenic nonepileptic seizures.Epilepsy Behav. 2006; 9: 307-311
- Postictal coughing and noserubbing coexist in temporal lobe epilepsy.Neurology. 2001; 56: 133-134
- Stertorous breathing is a reliably identified sign that helps in the differentiation of epileptic from psychogenic non-epileptic convulsions: an audit.Epilepsy Res. 2007; 77: 62-64
- Cluster analysis of clinical seizure semiology of psychogenic nonepileptic seizures.Epilepsia. 2000; 41: 610-614
- The value of pelvic thrusting in the diagnosis of seizures and pseudoseizures.Neurology. 2000; 54: 227-229
- Frontal lobe partial seizures and psychogenic seizures: comparison of clinical and ictal characteristics.Neurology. 1992; 42: 1274-1277
- Supplementary motor seizures mimicking pseudoseizures: some clinical differences.Neurology. 1990; 40: 1404-1407
- Frontal lobe seizures.Psychiatr Clin North Am. 2005; 28: 635-651
- Outcome in psychogenic nonepileptic seizures: 1 to 10-year follow-up in 164 patients.Ann Neurol. 2003; 53: 305-311
- Non-epileptic seizures.in: Halligan P.W. Bass C.M. Marshall J.C. Contemporary approaches to the study of hysteria: clinical and theoretical perspectives. Oxford University Press, Oxford2001: 143-154 (Chapter 10)
- Non-epileptic seizures: delayed diagnosis in patients presenting with electroencephalographic (EEG) or clinical signs of epileptic seizures.Seizure. 2002; 11: 193-197
- Teddy bears: an observational finding in patients with non-epileptic events.Neurology. 2003; 61: 714-715
- Clinical and electroencephalographic features of simple partial seizures.Neurology. 1988; 38: 1347-1352
- Seizure identification by clinical description in temporal lobe epilepsy: how accurate are we?.Neurology. 2003; 61: 1686-1689
- Split-screen synchronized display. A useful video-EEG technique for studying paroxysmal phenomena.Epileptic Disord. 2004; 6: 27-30
- Magnetoencephalography is more successful for screening and localizing frontal lobe epilepsy than electroencephalography.Epilepsia. 2007; 48: 2139-2149
- Use of serum prolactin in diagnosing epileptic seizures: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology.Neurology. 2005; 65: 668-675
- Serum prolactin in epilepsy and hysteria.Br Med J. 1978; 2: 1682
- Correspondence: MRI evidence of mesial temporal sclerosis in patients with psychogenic nonepileptic seizures.Neurology. 2001; 56: 821-823
- Evidence of brain abnormality in patients with psychogenic nonepileptic seizures.Epilepsy Behav. 2002; 3: 249-254
- Postictal SPECT in epileptic versus nonepileptic seizures.J Epilepsy. 1998; 11: 67-73
- Psychogenic nonepileptic seizures.Curr Opin Neurol. 2008; 21: 195-201
- Neuropsychologic impairment in patients with nonepileptic seizures.Arch Clin Neuropsychol. 1998; 13: 513-522
- Impaired motor function in patients with psychogenic pseudoseizures.Epilepsia. 2001; 42: 1600-1606
- Psychological correlates of psychogenic seizures.J Clin Exp Neuropsychol. 1994; 16: 524-530
- Cognitive impairment is not equal in patients with epileptic and psychogenic nonepileptic seizures.Epilepsia. 2006; 47: 1879-1886
- Performance of patients with epilepsy or psychogenic non-epileptic seizures on four measures of effort.Clin Neuropsychol. 2006; 20: 552-566
- Using the Minnesota Multiphasic Inventory 2, EEGs, and clinical data to predict nonepileptic events.Epilepsy Behav. 2007; 11: 343-346
- Differential response characteristics in nonepileptic and epileptic seizure patients on a test of verbal learning and memory.Neurology. 1995; 45: 2029-2034
- Part summary: psychological and neuropsychological evaluation of the patient with non-epileptic seizures.in: Gates J.R. Rowan A.J. Non-epileptic seizures. 2nd edition. Butterworth-Heinemann, Boston2000: 169-181 (Chapter 13)
- Cognitive and psychological functioning in patients with non-epileptic seizures.in: Gates J.R. Rowan A.J. Non-epileptic Seizures. 2nd edition. Butterworth-Heinemann, Boston2000: 123-137 (Chapter 9)
- A blinded pilot study of artwork in a comprehensive epilepsy center population.Epilepsy Behav. 2005; 6: 196-202
- Patients with psychogenic nonepileptic seizures, alone or epilepsy-associated, share a psychological profile distinct from that of epilepsy patients.J Neurol. 2003; 250: 338-346
- Family functioning in subjects with pseudoseizures and epilepsy.J Nerv Ment Dis. 2001; 189: 38-43
- Psychological characteristics of patients with newly developed psychogenic seizures.J Neurol Neurosurg Psychiatry. 2004; 75: 1175-1177
- A spell in the epilepsy clinic and a history of “chronic pain” or “fibromyalgia” independently predict a diagnosis of psychogenic seizures.Epilepsy Behav. 2005; 6: 264-265
- Psychodynamics and psychiatric diagnoses of pseudoseizure subjects.Am J Psychiatry. 1996; 153: 57-63
- Depression and symptoms affect quality of life in psychogenic nonepileptic seizures.Neurology. 2009; 73: 366-371
- Psychiatric and neurologic predictors of psychogenic pseudoseizure outcome.Neurology. 1999; 53: 933-938
- Electroencephalographic studies of simple partial seizures with subdural electrode recordings.Neurology. 1989; 39: 527-533
- Auras and subclinical seizures: characteristics and prognostic significance.Ann Neurol. 1990; 28: 320-328
- Psychogenic pseudosyncope: an underestimated and provable diagnosis.Epilepsy Behav. 2006; 9: 106-110
- Electroencephalographic findings during presyncope and syncope induced by tilt table testing.Can J Cardiol. 1998; 14: 811-816
- Physiological hypnic myoclonus.Electroencephalogr Clin Neurophysiol. 1988; 70: 172-176
- Coexisting epilepsy and non-epileptic seizures.in: Rowan A.J. Gates J.R. Non-epileptic Seizures. 1st edition. Butterworth-Heinemann, Stoneham (MA)1993: 47-54 (Chapter 6)
- Overintepretation of EEGs and misdiagnosis of epilepsy.J Clin Neurophysiol. 2003; 20: 42-44
- Errors in EEGs and the misdiagnosis of epilepsy: importance, causes, consequences, and proposed remedies.Epilepsy Behav. 2007; 11: 257-262
- Errors in EEG interpretation and misdiagnosis of epilepsy. Which EEG patterns are overread?.Eur Neurol. 2008; 59: 267-271
- Update on treatments of psychological nonepileptic seizures.Epilepsy Behav. 2005; 7: 364-374
- The withdrawal of antiepileptic drugs in patients with non-epileptic seizures: safety considerations.Expert Opin Drug Saf. 2006; 5: 609-613
- The safety of antiepileptic drug withdrawal in patients with non-epileptic seizures.J Neurol Neurosurg Psychiatry. 2005; 76: 1682-1685
Article info
Footnotes
Disclosures: Dr LaFrance has received grants from the NINDS, and the Epilepsy Foundation, American Epilepsy Society and receives editor’s royalties for Schachter SC, LaFrance Jr WC, editors. Gates and Rowan’s Nonepileptic Seizures. 3rd edition. Cambridge (NY): Cambridge University Press; 2010.
Dr Benbadis serves as a consultant or a speaker for Cyberonics, GSK, Lundbeck, Pfizer, Sleepmed, UCB pharma, and XLTEK.
Identification
Copyright
© 2011 Elsevier Inc. Published by Elsevier Inc. All rights reserved.
