Significance of Epileptiform Discharges in Patients without Epilepsy in the Community

Transcription

1 Epilepsia, 42(10): , 2001 Blackwell Science, Inc. International League Against Epilepsy Significance of Epileptiform Discharges in Patients without Epilepsy in the Community Maria C. Sam and Elson L. So Section of Electroencephalography, Mayo Clinic, Rochester, Minnesota, U.S.A. Summary: Purpose: To determine the frequency of recording epileptiform discharges (EDs) in patients without epilepsy in the community and to assess their risk of seizure disorders subsequently developing. Methods: We identified all outpatient and inpatient EEGs that were recorded in persons residing in Rochester, Minnesota, from 1979 to Patients with a history of unprovoked seizure disorders before the index EEG were excluded. Periodic lateralized EDs (PLEDs) were not evaluated, because of their well-established association with seizure disorders. Results: Five hundred twenty-one patients in the community had no history of unprovoked seizure disorders before their EEG. Sixty-four (12.3%) patients had EDs; neither isolated unprovoked seizure nor epilepsy developed during person-years of follow-up. Forty-seven (73.4%) of the 64 patients had acute or progressive cerebral disorders when EEG detected EDs. Seizures that were acutely provoked by the underlying disorder (enlarging brain tumor, cerebral infarct, or bilateral subdural hematoma, in one patient each) subsequently developed in three (6.3%) of the 64 patients. Seizures of any type did not develop in the 17 patients without acute or progressive cerebral disorders. Conclusions: In a community setting, EDs are sometimes observed in patients without epilepsy. However, nearly three fourths have underlying acute or progressive cerebral disorders. Acutely provoked seizures may develop in a small proportion of patients. Although none of our patients developed isolated unprovoked seizures or epilepsy, a longer period of follow-up is needed to determine their risks relative to the general population. Key Words: Electroencephalograms Epilepsy Epileptiform discharges Seizures. The association between epileptiform discharges (EDs) and epilepsy was established primarily on the basis of observations made of patients with epilepsy. The frequency and prognosis of EDs of persons without epilepsy are not well established. Determining the significance of EDs of nonepileptic persons is important in many clinical situations. It is not rare for neurologists to find epileptiform abnormalities from EEGs of patients without a history of seizure disorders. The situation behooves neurologists to discuss the likelihood of epileptic seizures occurring in the future with these patients and their families. Similarly, it is not uncommon for patients with proven psychogenic spells to have EDs observed on their EEG recordings (1). This situation creates lingering uncertainties as to whether these patients have only psychogenic spells or also have epileptic seizures. The unexpected finding of epileptiform abnormalities also complicates the issue of discontinuing antiepileptic drug Revision accepted July 13, Address reprint requests to Dr. Elson L. So, Section of Electroencephalography, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, U.S.A. Present address of Dr. Sam: Department of Neurology, Wake Forest School of Medicine, Winston-Salem, NC, U.S.A. (AED) therapy after a diagnosis of psychogenic spells or a period of prophylactic AED therapy for patients who have had head trauma and other neurosurgical procedures and have not experienced an unprovoked seizure (2). Previous reports of EDs in persons without epilepsy did not exclude many of the patterns that are known to be benign and not associated with epileptic seizure disorders (3). Many studies used observations made before many of the epileptiform patterns were recognized to be benign (Table 1). These benign epileptiform patterns should have been distinguished and excluded from the studies. Conversely, periodic lateralized EDs (PLEDs) were not specifically excluded as they needed to be, because they are widely known to be highly associated with acute epileptic seizure disorders (22). In our review of the literature, only five previous studies (14,16,17,19,20) reported the rate of seizure disorders developing after the detection of EDs in persons without epilepsy who had routine EEGs. However, none of these studies made the distinction between acutely provoked and unprovoked seizures. The risk of epilepsy (recurrent unprovoked seizures), as opposed to a single unprovoked seizure, also was not specifically assessed. 1273

2 1274 M. C. SAM AND E. L. SO TABLE 1. Summary of studies of epileptiform discharges of persons without epilepsy a Wake and drowsy Patients with EDs (%) Duration of follow-up Study, yr (ref) Type of patient Patients, no. Benign transients Unspecified Sleep Photic stimulation Hyperventilation Gibbs et al., Volunteers 1,000 Not excluded 0.4 b Not done Not done None NA 1943 (4) 20 yr Williams, 1944 (5) Healthy aircrew 241 Not excluded 0.8 None NA 682 Not excluded NA 2.6 Not done None NA Buchthal and Lennox, 1953 (6) Herrlin, 1954 (7) Brandt and Brandt, 1955 (8) Corbin and Bickford, 1955 (9) Larsson and Wéden 1958 (10) Kooi et al., 1964 (11) Bennett, 1967 (12) Doose et al., 1968 (13) Zivin and Marsan, 1968 (14) Eeg-Olofsson et al., 1971 (15) Cavazzuti et al., 1980 (16) Iida et al., 1985 (17) Bridgers, 1987 (18) Gregory et al., 1993 (19) Okubo et al., 1993 (20) Jabbari et al., 2000 (21) Current study Predominantly, male air force applicants Nonepileptic Healthy <5 yr Normal 1 10 yr Control subjects 7 yr Predominantly male volunteers yr Predominantly male aviators Neurologically normal Inpatients in tertiary care Normal 1 15 yr Neurologically healthy 6 13 yr Nonepileptic outpatients Nonepileptic psychiatric inpatients yr Aircrew trainees yr Healthy 6 12 yr Healthy male volunteers yr Community outpatients and inpatients Seizure development (%) 70 Not excluded NA 0 Not done None NA 135 Not excluded 0.7 Not done Not done Not done None NA 71 NA 5.6 Not done Not done 0 None NA 120 Not excluded 0.8 Not done Not done None NA 218 Not excluded 6.6 None NA 1,332 Not excluded 0.6 None NA 118 Not excluded 0.8 Not done None NA 6,497 Not excluded c 0 Few months to 10 yr Excluded 14 and 6 NA d 8.3 d 0.3 None NA positive spikes 3,726 Excluded 14 and Not Not 0 Unknown 5.3 positive spikes, recorded recorded high-voltage nonepileptiform synchronous activities, and excessive sensitivity to hyperventilation 10,473 Not excluded 8.1 Unknown 1.5 e 3,143 Excluded 1.0 None NA 13,658 Excluded 6-Hz spike wave and positive spikes yr 2.3 1,057 Excluded, except yr 6.0 small sharp spikes 100 Excluded None NA 521 Excluded personyears 6.3 acutely provoked seizures, no unprovoked seizures a Excluding studies of (a) only relatives of persons with epilepsy, (b) the effect of pharmacologic and photic stimulation in normal subjects, and (c) patients with brain injury (e.g., cerebral palsy, mental retardation, perinatal brain damage, or brain trauma). b Spike wave discharges, 3/s. c Spontaneous discharges: 0.13% with and 0.07% without. d Includes nonepileptiform paroxysmal activities. e Based on 202 patients whose EEGs were abnormal on two separate recordings., no information; EDs, epileptiform discharges; NA, not applicable. Because epileptiform abnormalities can be identified only if an EEG is performed, their frequency cannot be directly verified in the general population. One approach for approximating the baseline frequency in the community is to determine how often EDs appear in the EEGs of community patients without a history of epileptic seizures. The objectives of our study were to determine the frequency of recording EDs in patients without epilepsy

3 SIGNIFICANCE OF EPILEPTIFORM DISCHARGES 1275 in the community and to assess their risk of subsequent seizure disorders. METHODS The 10-year study period was selected from 1979 through 1988 because from 1979, the determination of EDs in our EEG laboratory did not include benign transients (e.g., benign sporadic sleep spikes, six per second spike waves, 14 and six positive spikes, wicket waves and rhythms, and benign temporal theta bursts of drowsiness) (13). The EEG procedure during that period was similar to the current procedure. Moreover, we aimed for a longer follow-up period to assess the long-term risks of epileptic seizures developing in our study patients subsequent to the detection of EDs. Sixteen channels of EEG activity were recorded with wake recording of 30 min. Photic stimulation was performed with eyes open and repeated with eyes closed. The rates of stimulation were 1, 3, 8, 10, 12, 15, 20, and 30 Hz. Each stimulation lasted 10 s, and the interval between stimulation frequencies was 4 s. The EEG recording protocol also included vigorous hyperventilation for 3 min. A sleep recording was performed when the EEG was requested to evaluate paroxysmal disorders, among which epileptic seizure disorder was a clinical consideration. The standard instruction was for the patient to be partially sleep deprived (sleep for only 4 h) the night before the procedure. When sleep failed to occur spontaneously during the EEG recording, the patient was sedated with chloral hydrate, unless it was medically contraindicated. The following inclusion criteria were used for this study. 1. The outpatient or inpatient EEG was performed between January 1, 1979, and December 31, The patient resided in Rochester, Minnesota, when the EEG was recorded. 3. The patient had no history of unprovoked seizure occurrence. 4. The patient had no acute symptomatic or provoked seizures for which the index EEG was done. Patients in this study were identified initially from our database of EEG procedures. Residency status within Rochester was determined by the Residency Verification Unit of our Department of Health Sciences Research. Patients were excluded who had unprovoked seizures before the date of the index EEG. Patients who had acute symptomatic or provoked seizures for which the index EEG was done also were excluded. However, patients with a history of febrile or prior acute symptomatic seizures that were unrelated to the index EEG were included in the study. Subjects who met the inclusion criteria were matched with our database of EEG reports. Our laboratory uses a unique classification of EEG findings in each record (23). The EEG report for each patient contains a classification of the results of the EEG procedure. Nonepilepsy Rochester patients whose EEG showed EDs were identified. EDs were focal or generalized spikes or sharp waves occurring spontaneously or during hyperventilation or photic stimulation (24). Patients with PLEDs were not included, because of the well-established association between PLEDs and acute epileptic seizure occurrence (22). Medical records were reviewed for demographic information and the medical status of the study patients at index EEG. Since the early 1900s, our institution has maintained a unified medical record for each patient seen at Mayo Clinic (25). Each record has all aspects of the patient s medical care, including outpatient visits, hospital visits, laboratory and radiologic results, pathologic results, and medical correspondence. Rochester, Minnesota, is an urban community that had a population of 52,629 in 1980, the year of national census within the study period (26). Approximately 95% of the population received some or all of their care at Mayo Clinic. Each year, more than half of the population was seen at the Mayo Clinic or one of its affiliated facilities. Medical records of the study patients were reviewed to determine whether seizure disorders had occurred after the index EEG detected EDs. The patients were followed up until death, last contact, or June 30, 1996, whichever came first. Seizures were classified according to criteria established by the Commission on Classification and Terminology of the International League Against Epilepsy (27). An acute symptomatic or acutely provoked seizure was defined as one that occurred close in time (within 1 week) to the occurrence of an acute cerebral insult. As such, it was considered an acute manifestation of the cerebral insult. An unprovoked seizure was defined as a seizure that occurred in the absence of an acutely provoking factor or cerebral insult. Epilepsy was defined as the occurrence of two or more unprovoked seizures and diagnosed at the time of the second unprovoked seizure. To avoid selection bias, the determination of seizure occurrence was independent of the clinical diagnoses or impression provided in the medical records by the patients physicians. Because Zivin and Marsan (14) had reported that nonepilepsy patients with EDs are at greater risk of seizures developing when progressive neurologic disorders are present, we divided our study patients into those with (group A) and those without (group B) acute or progressive neurologic disorders. Groups A and B were compared by using the 2 test. The Mann Whitney test was used to compare continuous variables. The difference

4 1276 M. C. SAM AND E. L. SO was considered significant when p < The study was approved by our Institutional Review Board. RESULTS Five hundred twenty-one persons met the inclusion criteria, 64 (12.3%) of whom had EDs on their EEGs. Forty-seven (73.4%) of the 64 patients with EDs had an acute or progressive neurologic disorder at index EEG. Table 2 shows the neurologic disorders of those with (group A) and those without (group B) acute or progressive neurologic disorders. The most common underlying acute or progressive neurologic disorders of patients in group A were metabolic and hypoxic encephalopathies, cerebral infarctions, and progressive dementia. Patients without acute or progressive neurologic disorders had mainly behavior disorders, sensory spells, dizziness, inattentiveness, psychosis, or headaches. Seventeen (26.6%) of the patients with EDs were 17 years or younger, and five (7.8%) were 5 years or younger at the index EEG. Table 3 shows a comparison of the demographic, medical, and EEG findings of patients in groups A and B. Compared with patients in group A, patients in group B were younger at the index EEG recording and more likely to have EDs only during sleep. However, there was no difference in the distribution of focal and generalized EDs or the locations of the focal discharges. Only 2.1% of patients in group A and 11.8% of patients in group B had EDs induced by photic stimulation, but there was no significant difference between these groups. TABLE 2. Neurologic conditions of patients with (group A) and without (group B) acute or progressive neurologic disorders at index electroencephalogram Condition No. of patients Group A (n 47) Cerebral infarction 8 Progressive dementia 8 Hypoxic encephalopathy 8 Enlarging brain tumor 5 Acute multifactorial encephalopathy a 5 Intracerebral hemorrhage 3 Acute head trauma 3 Hepatic encephalopathy 2 Septic encephalopathy 2 Status post brain tumor resection 2 Status post cerebral aneurysm repair 1 Group B (n 17) Behavior disorder 5 Sensory spells 3 Headache 2 Psychosis 2 Inattentiveness 2 Dizziness 1 Poststroke dysphasia 1 Posttraumatic memory disturbance 1 a Due to a combination of metabolic, hypoxic, or toxic encephalopathies. The duration of follow-up after detection of EDs was person-years in group A and person-years in group B. Unprovoked seizures did not develop in either group of patients. No patient was taking antiepileptic medications (AEDs) during follow-up. However, seizures that were acutely provoked by an underlying acute or progressive neurologic condition (enlarging brain tumor, cerebral infarct, or bilateral subdural hematomas, in one patient each) subsequently developed in three (6.3%) patients in group A. No acutely provoked seizures developed in patients in group B after EDs were observed on their EEGs. DISCUSSION Our study shows that it is not rare for patients without prior unprovoked seizure disorders or epilepsy to have EDs, but there is often an underlying acute or progressive cerebral disorder. The type and location of the EDs are indistinguishable in patients with and without underlying acute or progressive cerebral disorders, but those without the disorders are more likely to have EDs that appear only during sleep. The risk of isolated unprovoked seizure or epilepsy developing subsequent to the detection of EDs is low. Isolated unprovoked seizures or epilepsy did not develop in any of the patients in either group during their follow-up. However, the 230 person-year duration of their follow-up is relatively short. From the incidence of unprovoked seizures in the community (28), the minimal duration of follow-up needed to observe one case of an unprovoked seizure is 2,325 person-years. Acute seizures that are provoked by underlying acute or progressive cerebral disorders develop in a small proportion of patients (6%). This low risk of acutely provoked seizures does not warrant routine and prophylactic AED treatment when EDs are observed. It is more appropriate to consider using AED treatment when an unprovoked seizure has occurred (29). For patients with a first unprovoked seizure, a history of acute symptomatic or provoked seizure predisposes them to another unprovoked seizure (i.e., epilepsy) (30,31). Only five previous reports (14,16,17,19,20) provided rates of seizure disorders developing after detection of EDs in nonepileptic persons (Table 1). Unlike our study, none of these reports determined whether subsequent seizures were acutely provoked or unprovoked. Thus it was not possible to ascertain from these studies the magnitude of risk of epilepsy (i.e., recurrent unprovoked seizures) developing in persons whose EEGs incidentally showed EDs. Zivin and Marsan (14) found that seizures subsequently developed in a high proportion (15%) of their patients with EDs. The discrepancy between their study and ours may be because their patients were all hospitalized in a tertiary referral center, whereas our co-

5 SIGNIFICANCE OF EPILEPTIFORM DISCHARGES 1277 TABLE 3. Comparison of demographic, medical, and EEG features of patients with (group A) and without (group B) acute or progressive neurologic disorders at index electroencephalogram Feature Group A (n 47) Group B (n 17) p Value Mean age, yr (range) 58.1 (< ) 36.2 ( ) <0.01 Sex, no. (male/female) 20/25 7/10 NS EDs, a no. (focal, generalized) 32/13 11/6 NS Temporofrontal NS Central 2 0 NS Multifocal 8 0 NS Both focal and generalized 2 0 NS Generalized 13 6 NS Photic-induced EDs 1 2 NS EDs during sleep only 2 4 <0.05 Febrile seizure 1 1 NS Mental retardation 1 3 NS Family history of seizures 0 2 NS a Not including two patients in group A with both focal and generalized discharges. EDs, epileptiform discharges; NS, not significant. hort also included outpatients in the community. Their patients were more likely than ours to have had major neurologic disorders. Seizures subsequently developed in the majority of their patients who had EEGs recorded for strokes and head trauma. These conditions are known to increase the risk of acute symptomatic (provoked) and unprovoked seizures (31 33). Zivin and Marsan (14) also found that patients with progressive neurologic disorders were at greater risk of subsequently experiencing seizures than were those with stable neurologic conditions. The difference in seizure rates may have been because of a bias toward documenting seizure occurrence in patients who remained in the hospital or were followed up more frequently because of their progressive neurologic disorder. The duration of follow-up for each group of patients is not known, and <50% of the patients were available for follow-up at 2 years. Previously reported rates of EDs in persons without epilepsy ranged from 0.4 to 8.1% (17). Reasons for the wide range are the lack of uniformity in the types of subjects studied, the exclusion of benign transients, and the technical aspects of the EEG recording. Several studies reported low rates of epileptiform abnormalities (<1.0%). However, the cranial regions studied and the duration of the recordings in some of these studies were limited. For instance, in the study by Gibbs et al. (4), the minimal requirement was the recording of EEGs from three areas of the head for 15 min. No recording was done during sleep. In the study by Brandt and Brandt (8), only samples of EEGs during wakefulness, rather than the whole recording, were reviewed. In Bennett s study (12), each EEG recording was for only 10 min. The study also involved only spike wave abnormalities. The noninclusion of other types of epileptiform abnormalities may account for the relatively low rate of EDs reported in the study. Eeg-Olofsson et al. (15) observed many types of paroxysmal activities in during wakefulness, sleep, photic stimulation, and hyperventilation. The rates of these activities were as high as 8 to 9%. However, it is clear from the authors definitions of these activities and from the examples provided in their figures that most were nonepileptiform activities. For instance, the bilateral hypersynchrony that the authors reported is a physiologic feature that is especially prominent in during drowsiness. The only abnormalities that could be considered unequivocally as epileptiform activities were focal spikes and sharp waves. Some previous reports studied air force personnel, although several others evaluated school. Our study was designed to assess the rate and significance of EDs in a clinical practice setting where many patients who had EEG recordings did not have a history of unprovoked seizure or epilepsy. Only three other studies evaluated patients rather than healthy volunteers (14,17, 18). However, our study is the only one that minimized referral bias by evaluating all inpatients and outpatients in the community. Despite the fact that Iida et al. (17) confined their study to outpatients, their rate of 8.1% approached our rate of 12.3% for inpatients and outpatients combined; however, their study did not specifically exclude benign sharp transients. In summary, we found that EDs are often observed in patients without epilepsy in the community; however, the majority have underlying acute or progressive cerebral disorders. The risk of subsequently developing unprovoked seizures or epilepsy is low, although acutely provoked seizures may develop in a small proportion of patients with underlying acute or progressive cerebral disorders. Acknowledgment: This study was supported by funds from Mayo Foundation.

6 1278 M. C. SAM AND E. L. SO REFERENCES 1. King DW, Gallagher BB, Murvin AJ, et al. Pseudoseizures: diagnostic evaluation. Neurology 1982;32: Temkin NR, Dikmen SS, Wilensky AJ, et al. A randomized, double-blind study of phenytoin for the prevention of posttraumatic seizures. NEnglJMed1990;323: Klass DW, Westmoreland BF. Nonepileptogenic epileptiform electroencephalographic activity. Ann Neurol 1985;18: Gibbs FA, Gibbs EL, Lennox WG. Electroencephalographic classification of epileptic patients and control subjects. Arch Neurol Psychiatr 1943;50: Williams D. The nature of the transient outbursts in the electroencephalograms of epileptics. Brain 1944;67: Buchthal F, Lennox M. The EEG effect of metrozol and photic stimulation in 682 normal subjects. EEG Clin Neurophysiol 1953; 5: Herrlin K-M. EEG with photic stimulation: a study of with manifest or suspected epilepsy. Electroencephalogr Clin Neurophysiol 1954;6: Brandt S, Brandt H. The electroencephalographic patterns in young healthy from 0 to five years of age. Acta Psychiatr Neurol Scand 1955;30: Corbin H, Bickford R. Studies of the electroencephalogram of normal : comparison of visual and automatic frequency analyses. Electroencephalogr Clin Neurophysiol 1955;7: Larsson L-E, Wéden L. Electroencephalographic investigations. Acta Psychiatr Neurol Scand 1958;33(suppl 126): Kooi KA, Güvener AM, Tupper CJ, et al. Electroencephalographic patterns of the temporal region in normal adults. Neurology 1964; 14: Bennett DR. Spike-wave complexes in normal flying personnel. Aerosp Med 1967;38: Doose H, Gerken H, Volzke E. Genetics of centrencephalic epilepsy in childhood. Epilepsia 1968;9: Zivin L, Marsan CA. Incidence and prognostic significance of epileptiform activity in the EEG of non-epileptic subjects. Brain 1968;91: Eeg-Olofsson O, Petersen I, Sellden U. The development of the electroencephalogram in normal from the age of 1 through 15 years: paroxysmal activity. Neuropadiatrie 1971;2: Cavazzuti GB, Cappella L, Nalin A. Longitudinal study of epileptiform EEG patterns in normal. Epilepsia 1980;21: Iida N, Okada S, Tsuboi T. EEG abnormalities in nonepileptic patients. Folia Psychiatr Neurol Jpn 1985;39: Bridgers SL. Epileptiform abnormalities discovered on electroencephalographic screening of psychiatric inpatients. Arch Neurol 1987;44: Gregory RP, Oates T, Merry RT. Electroencephalogram epileptiform abnormalities in candidates for aircrew training. Electroencephalogr Clin Neurophysiol 1993;86: Okubo Y, Matsuura M, Asai T, et al. A follow-up study of healthy with epileptiform EEG discharges. J Epilepsy 1993;6: Jabbari B, Russo MB, Russo ML. Electroencephalogram of asymptomatic adult subjects. Clin Neurophysiol 2000;111: Pohlmann-Eden B, Hoch DB, Cochius JI, et al. Periodic lateralized epileptiform discharges: a critical review. J Clin Neurophysiol 1996;13: Klass D, Espinosa R. Electroencephalography. In: Clinical examinations in neurology. 5th ed. Philadelphia: WB Saunders, 1981: Chatrian G, Bergamini L, Dondey M, et al. A glossary of terms most commonly used by clinical electroencephalographers. In: International Federation of Societies for Electroencephalographers and Clinical Neurophysiology: recommendations for the practice of clinical neurophysiology. Amsterdam: Elsevier Science Publishers, 1983: Kurland LT, Molgaard CA. The patient record in epidemiology. Sci Am 1981;245: Schroeder D, Offord K. A SAS macro which utilized local and reference population counts appropriate for incidence, prevalence, and mortality rate calculations in Rochester and Olmsted County, Minnesota: Mayo Clinic. August Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia 1981;22: Annegers JF, Hauser WA, Lee JR, et al. Secular trends and birth cohort effects in unprovoked seizures: Rochester, Minnesota, Epilepsia 1995;36: So E. Update on epilepsy. Med Clin North Am 1993;77: Annegers JF, Shirts SB, Hauser WA, et al. Risk of recurrence after an initial unprovoked seizure. Epilepsia 1986;27: So EL, Annegers JF, Hauser WA, et al. Population-based study of seizure disorders after cerebral infarction. Neurology 1996;46: Annegers JF, Grabow JD, Groover RV, et al. Seizures after head trauma: a population study. Neurology 1980;30: Annegers JF, Hauser WA, Lee JR, et al. Incidence of acute symptomatic seizures in Rochester, Minnesota, Epilepsia 1995;36: