Epikpsiu, 41( 1):81-84, 2000 Lippincott Williams & Wilkins, Inc.

, Philadelphia 0 International League Against Epilepsy

Clinical Research

Discriminating Between Epileptic and Nonepileptic Events: The Utility of Hypnotic Seizure Induction
John J. Barry, Orit Atzman, and Martha J. Morrell

Summary: Purpose: To determine the validity of the Hypnotic Induction Profile (HIP) followed by seizure induction during continuous video-electroencephalographic (EEG) monitoring to discriminate between epileptic (EE) and nonepileptic events (NEE). Methods: Eighty-two patients admitted to the Stanford Comprehensive Epilepsy Center for differential diagnosis of seizure-like events were evaluated. Exclusion criteria included inability or refusal to complete the HIP, lack of a typical event, an IQ <70, present evidence of psychosis, or a physiological cause for NEE. Sixty-nine patients met these criteria. While undergoing continuous video-EEG monitoring, the patient completed an HIP, an inventory designed to measure the degree of hypnotizability. An attempt was then made to induce the patients typical events under hypnosis by using a splitscreen technique. An event without an EEG correlate was thought to represent an NEE. A diagnosis of NEE was made

independently by the neurology team and was compared with results obtained with the hypnotic evaluation. Results: Results for patients with EE were compared with those with NEE and a group consisting of both EE/NEE. All patients with NEE were then contrasted with the EE group. HIP scores for the EE patients indicated lower hypnotizability than the NEE group and were statistically significant when NEE patients and those with NEEEE were combined. The sensitivity of seizure induction in the diagnosis of NEE was 77%, with a specificity of 95%. Conclusions: The HIP coupled with seizure induction is a useful technique to aid in the diagnosis of patients with NEE. It is sensitive and specific, and it may provide the patient with a useful behavioral tool to control NEEs. It may also furnish a conduit for long-term treatment. Key Words: Hypnotic induction profile-Epileptic events-Nonepileptic eventsInduction-Dissociation.

Distinguishing epileptic (EE) from nonepileptic events (NEEs) is often extremely difficult. Historical, psychometric, and behavioral information is frequently useful, but even trained neurologists using videotape correctly differentiate EEs from NEEs in only 70% of cases (1). The task is even further complicated by the cooccurrence of both EEs and NEEs in 540% of cases (2). Morbidity with misdiagnosis also can be significant (3,4). Three types of suggestion techniques have been described in the literature. Ideally, all take place while the patient is simultaneously monitored with continuous video-EEG. Both saline infusion and the alcohol-patch techniques induce a psychogenic event by the use of an agent to provoke a typical seizure. The former is given intravenously, and the latter applied to the patients neck
Accepted August 5, 1999. Address correspondence and reprint requests to Dr. J. J. Barry at Stanford Department of Psychiatry, Stanford University Medical Center, MC572.3, Stanford, CA 94305, U.S.A. E-mail: jbarry@leland. stanford.edu Preliminary results presented at the American Epilepsy Association Meeting, Baltimore, Maryland, December 5-9, 1995.

(5,6). The patient is told that these agents often stimulate an event and that the information obtained will help in the understanding and treatment of the condition. Ongoing controversy exists concerning the impact of these procedures on the patient-physician therapeutic alliance (2). Hypnosis also has been used to differentiate between EE and NEE and may have distinct advantages. In 1950, Peterson et al. (7) attempted to use hypnosis to discriminate NEE patients from those with EE. They used a recall technique and assumed that patients with epilepsy would not have the capacity to remember the specifics of their seizures, in contrast to NEE patients. He found that only those patients (n = 35) with events of psychogenic origin could recount explicit details of their seizures, whereas all of the EE patients (n = 30) were amnestic for their events (7). In 1955, Schwartz (8) used hypnosis in an attempt to .. induce typical seizures during EEG monitoring in a group 26 patients. We successful in Only the lo NEE patients. None of the EE group had their seizures instigated during a trance state (8). The Hypnotic Induction Profile (HIP) was developed by H. and D. Spiegel to provide a clinically useful scale 81

82

J. J. BARRY ET AL.
was placed on the left, and the patient was then instructed to shift to the right side. The patient was asked to imagine experiencing one of his or her typical events, If necessary, characteristic, idiosyncratic features of the patients seizure were suggested and induction ensued. If successful, the subject would then terminate the event by shifting to the left side of the screen and was encouraged to move back and forth, starting and stopping the event, until confident that they could both self-induce and abolish the seizure. The EEG obtained during the procedure was evaluated for any evidence of epileptic activity. Data were analyzed as follows. A one-way analysis of variance (ANOVA) was used to compare mean HIP scores for patients with EE, NEE, and a combineddiagnosis group composed of patients who displayed NEE with the seizure-induction procedure, but who also were found to have EE during other portions of the evaluation. The HIP scores of patients with NEE and NEE/EE were then pooled and compared with the EEonly group by using a two-tailed t test. Finally, the sensitivity (the probability that a patient with a positive diagnosis of epilepsy will have a positive test) and specificity (the probability that a patient with a negative diagnosis will have a negative test) (I 4) of hypnotic seizure induction to diagnose NEE was ascertained.

of hypnotizability that would predict response to treatment and help distinguish different types of psychopathology (9). It consists of a hypothesized biologic measure of a patients capacity to experience hypnosis, the eye roll. This was a serendipitous observation and remains controversial (10). In addition, there is an ideomotor and subjective-experience component. The summation of selected components of the scale yields an induction (IND) score, which will hereafter be termed the HIP score. In addition, a profile score reflects the patients capacity to be hypnotized and the actualization of this potential (9). The HIP also correlates well with other scales of hypnotizability (1 1 ), although some controversy exists when comparing it with experimental research hypnotic measuring devices (10). HIP scores also have been shown to help in clinical diagnostic evaluation. Groups of patients whose psychopathology is characterized by high dissociability such as posttraumatic stress disorder (PTSD) have significantly higher scores than do patients with, for example, schizophrenia (12). This study evaluated two separate facets of the hypnotic evaluation attempting to discriminate EE from NEE. The first investigated the utility of the HIP score alone. The second involved hypnotic seizure induction using a split-screen technique. Patients were then taught self-hypnosis to induce and terminate events (13).

METHODS
The results of all psychiatric referrals for differential diagnosis of seizure-like events over a 4-year period, from 1992 to 1996, were reviewed. Patients were admitted to the Stanford Comprehensive Epilepsy Center and underwent continuous video-EEG monitoring to capture approximately three typical events. A diagnosis of epilepsy or of pseudoseizures was made by the treating neurology team using information independent of the hypnotic induction or seizure provocation. Eighty-two patients were evaluated by one of the authors (J.J.B). Patients were excluded from the study evaluation because of an inability or unwillingness to complete the HIP, lack of a typical event, an IQ <70, present evidence of psychosis, or symptoms explained by an underlying medical cause, such as insulinoma. All patients were offered the option of using hypnosis to elicit a typical event and to help control those events. An HIP score was obtained with a range from zero, showing no hypnotizability, to a maximal hypnotic potential, a 10. The patient was then hypnotized again and taught relaxation followed by the use of a split-screen technique. The patient was instructed to imagine a blank screen and then to divide it in half, the left side marked with RELAX and the right with SEIZURE. The previous image associated with a feeling of calm and safety

RESULTS
Eighty-two patients (20 man and 62 women) were screened. Thirteen did not meet inclusion criteria. Sixtynine patients were included in the analysis with ages ranging from 19 to 59 years, including 17 men and 52 women. Twenty-two patients were diagnosed with EE, and all had localization-related epilepsy. Two had frontal epilepsy without clear lateralization. The rest of the EE patients had ictal events localized to the temporal lobe: seven right and 12 left. Another patient had bilateral temporal epileptic discharges. Thirty-six patients had NEE alone. Another 11 displayed both EE and NEE. Of these 1 1 patients, seven had localization-related epileptic events, five with left and two with a right temporal lobe focus. Of the four remaining patients in this group, one individual had primary generalized epilepsy, and another had an abnormal interictal EEG. The remaining two patients had a reliable clinical history confirming a diagnosis of epilepsy. Mean HIP scores were compared by using a one-way ANOVA. Patients with epilepsy had a lower mean HIP value (5.18; SD, 3.31) in contrast to those with NEE (6.80; SD, 3.14) and patients in the combined diagnostic group (6.92; SD, 1.99). There was no statistically significant difference, however, comparing these means (p

HYPNOTIC SEIZURE INDUCTION

= 0.117, cli 2). By using a two-tailed t test, mean HIP scores of all patients with NEE (6.83; SD, 2.87) were compared with those with EE (5.18; SD, 3.31), and a significant difference was found ( t = -2.116, dJ 67; p = 0.038; refer to Table 1). The capacity of hypnosis with seizure induction to evoke a typical event was then determined. One patient did have an epileptic seizure (confirmed by video-EEG monitoring) while undergoing hypnotic induction. This appeared to be a chance occurrence, because this patient was not hypnotizable (HIP, 0), had a decrement profile and had many EEs that day. A repeated attempt to induce an EE in this patient with hypnosis was unsuccessful. Hypnotic induction of an EE could not be ruled out, and a specificity of 95% was therefore obtained. Of 47 patients with NEE, 36 had seizures induced during hypnosis, providing a sensitivity of 77% (refer to Table 1).

83

DISCUSSION
This study used hypnosis to differentiate EEs from NEEs. The patients capacity to be hypnotized was measured by using the overall induction score on the HIP. Individual differences, as determined by the HIP, differentiated patients with EE from NEE only when the NEE and NEE/EE groups were combined. However, HIP scores were obtained only once, and some patients may require several tests to reach a baseline (10). In addition, the author (J.J.B.) was the only evaluator, introducing potential bias and possibly compromising reproducibility. After the HIP was completed, an attempt was made to induce a typical event by using hypnosis. Patients with NEE were correctly identified by hypnotic seizure induction 77% of the time. Specificity of the procedure was 95%. In other studies, suggestion techniques have been used as diagnostic procedures. Walczak et al. ( 5 ) used the saline placebo infusion method in 68 patients and found that 82% of the 40 patients with NEE had typical psychogenic events induced by the procedure. In two patients with epilepsy, this technique resulted in the inducTABLE 1. HIP

+ diagnosis

by seizure induction
Diagnosis by seizure induction Sensitivity Specificity

Hypnotic Induction Profile scores Group

No.
22 36
11

Mean

SD
3.31 3.14 1.99 2.87

EE NEE EE/NEE NEE + EE/NEE All patients

47 69

5.18 6.80 6.92 6.83

77%

95%

EE, epileptic events; NEE, nonepileptic events. Means are not different statistically (p = 0.1 17). Means are statistically significant (p = 0.038).

tion of true epileptic seizures. Lancman et al. (6) used the alcohol-patch technique in 93 patients with NEE and in 20 patients with EE and found a diagnostic sensitivity of 77.4% and a specificity of 100%. Ethical concerns regarding the use of these procedures has been discussed by Devinsky and Fisher (15). Walczak et al. (16) found no long-term adverse effects from this method. Notably, psychogenic status epilepticus has been induced by such procedures (17). One of the distinct advantages of hypnosis with seizure induction for differential diagnosis is its safety and that the patient is completely informed of its purpose and technique. It is important to consider how hypnosis may aid in the differentiation of NEE from EE. In general, the HIP is considered a measure of hypnotizability (9). Many authors have equated this ability with overall dissociability (10). Dissociation has been viewed as a defense mechanism maintained and overused since childhood because of early trauma, often in the form of sexual abuse (1 8). NEE patients frequently have similar histories of abuse (19), and this may account for their comparatively elevated scores on the HIP. Hypnosis and suggestion techniques might initiate NEE by heightening the ability of these patients to filter out distracting stimuli and attend, uncritically, to the instructions of the examiner. This possibility has been at least partially confirmed in studies demonstrating hypnotic alteration of physiological states associated with selective attention (20). This effect of hypnosis can be seen particularly in certain psychopathologic conditions (e.g., dissociative disorders, a category that might also include NEE). In this study, there was a subset of patients with epilepsy and NEE. Several studies have shown a higher frequency of temporal lobe epilepsy in patients with dissociative disorders, such as Dissociative Identity Disorder (formally called Multiple Personality Disorder) (2123). However, many of these reports are case studies or are methodologically flawed. Ross et al. (24) used the Dissociative Experiences Scale to measure dissociability and found little difference between patients with complex partial seizures and a neurological control group. Dissociative symptoms in epilepsy patients usually are interictal, and arise later in life. They may have a different pathophysiologic and psychological mechanism, because these patients generally do not have the extensive histories of sexual or physical abuse seen in the NEE group (25,26). This cohort of patients with combined NEE/EE needs further characterization. The HIP with seizure induction is a useful diagnostic tool to differentiate EEs from NEEs. It helps the clinician introduce the diagnosis of NEE, as the patient is provided with a way to understand the occurrence of NEE and a means of controlling the events non-pharmacologically (13). It fosters a therapeutic alliance with the patient, which is so crucial to the treatment process. As this alEpilep&ia, Vul. 41, No. l , 2000

84

J. J. BARRY ET AL.
12. Spiegel H, Hunt T, Dondershine HE. Dissociation and hypnotizability in posttraumatic stress disorder. A m J Psychiatry 1988; 145: 301-5. 13. Bush E, Barry JJ, Spiegel D, Wasserstein P, Morrell MJ. The successful treatment of pseudoseizures with hypnosis. Epilepsia 1992;33: 135. 14. Kreamer HC. Evaluating medical tests. Newhury Park: Sage Puhlications, 1992. IS. Devinsky 0, Fisher R. Ethical use of placebos and provocative testing in diagnosing nonepileptic seizures. Neurology 1996;47: 866-7. 16. Walczak TS, Papacostas S, Williams DT, Scheuer ML, Lebowiz N, Notarfrancesco A. Outcome after diagnosis of psychogenic nonepileptic seizures. Epilepsia 199S;36:113 1-7. 17. Ney GC, Zimmerman C, Schaul N. Psychogenic status epilepticus induced by a provocative technique. Neurology 1996;46:546-7. 18. Van der Kolk BA. The body keeps the score: approaches to the psychobiology of posttraumatic stress disorder. In: van der Kolk BA, McFarkdne AC, Weisaeth L, eds. Traumutic Stress: the overwhelming experience of mind, body, und society. New York: Guilford Press, 1996:214-41. 19. Bowman ES. Etiology and clinical course of pseudoseizures, relationship to trauma, depression and dissociation. Psychosomatics 1993;34:33342. 20. Dixon M, Laurencc J-R. Two hundred years of hypnosis research: questions resolved? questions unanswered! In: Fromm E, Nash MR, eds. Contemporary hypnosis research. New York: Guilford Press, 1992:459-574. 21. Mesulam M-M. Dissociative states with abnormal temporal lobe EEG: multiple personality and the illusion of possession. Arch Ncurol I981;38:176-81. 22. Putnam FW, Guroff JJ, Silberman EK, et al. The clinical phenonenology of multiple personality disorder: review of 100 recent cases. J Clin Psychiatry 1986;47:285-93. 23. Scbenk L, Bear D. Multiple personality and related dissociative phenomena in patients with temporal lobe epilepsy. A m .I Psychiatry 1981;138:1311-6. 24. Ross CA, Heber S, Anderson G, et al. Differentiating multiple personality disorder and complex partial sei7,ures. Gen Hosp Psychiatry 1989;11:54-8. 25. Spiegel D. Neurophysiological correlates of hypnosis and dissociation. J Neuropsychiatry 1991;3:440-5. 26. Alper K, Devinsky 0, Perrine K, et al. Dissociation in epilepsy and conversion nonepileptic seizures. Epilepiu 1997;38:99 1-7.

liance develops, the frequency of NEE often decreases. Somatic language is replaced by verbal reports of distress. Psychotherapy and pharmacotherapy can then be used to full advantage. In the authors experience, patients have generally responded to the hypnotic process with enthusiasm and gratitude. It also is easy for the clinician to learn and may provide a significantly less expensive diagnostic technique than inpatient monitoring. Hypnosis offers a useful addition to the armementarium in the diagnosis of NEE.

REFERENCES
1. French JA. The use of suggestion as a provocative test in the diagnosis of psychogenic non-epileptic seizures. In: Rowan AJ, Gates JR, eds. Non epileptic seizures. Boston: Butterworth, 1993: 101-10. 2. Gates JR, Lucian0 D, Devinsky 0. The classification and treatment of nonepileptic events. In: Devinsky 0, Theodore, eds. Epilepsy und behavior. New York: Wiley-Liss, 1991:251-163. 3. Cohen R, Suter C. Hysterical seizures: suggestion as a provocative EEG test. Ann Neurol 1982;11:391-xx. 4. Howell SJL, Owen L, Chadwick DW. Pseudostatus epilepticus. Q J Med 1989;71:507-19. 5. Walczak TS, Williams DT, Berten W. Utility and reliability of placebo infusion in the evaluation of patients with seizures. Neurology 1994;44:394-9. 6. Lancman ME, Asconape JJ, Craven WJ, Howard G, Penry JK. Predictive value of induction of psychogenic seizures by suggestion. Ann Ncurol 1994;35:359-6 I . 7. Peterson DB, Sumner JW, Jones GA. Role of hypnosis in differentiation of epileptic from convulsive like seizures. A m J Psychiatry 1950;107:428-32, 8. Schwartz BE, Bickford RG, Rasmussen WC. Hypnotic phenomena, including hypnotic activated seizures, studied with electroencephalogram. J N e w Ment Dis 1955; 122:564-74. 9. Spiegel H, Spiegel D. Trance and treatment, clinical uses of hypnosis. New York: Basic Books, 1978. 10. Perry C, Nadon R, Button J. The measurement of hypnotic ability. In: Fromm E, Nash MR, eds. Contemporary hypnosis research. New York Guilford Press, 1992:459-74. 11. Frischholz EJ, Spiegel D, Spiegel H. The hypnotic induction profile and absorption. A m J Clin Hypnosis 1987;30:87-94.

Epilepsia, Vol. 41, No, I , 2000