Symptomatic or cryptogenic partial epilepsy of childhood onset: fourteen-year follow-up

Original Articles

Symptomatic or Cryptogenic Partial Epilepsy of Childhood Onset: Fourteen-Year Follow-Up ´ kos Szabo´, MD*, A. David Rothner, MD†, Prakash Kotagal, MD†, C. A Gerald Erenberg, MD†, Dudley S. Dinner, MD†, and Elaine Wyllie, MD† This study reports on the seizure and psychosocial outcome of 29 patients with electroclinically welldefined childhood-onset symptomatogenic or cryptogenic partial epilepsy with complex partial seizures who were followed prospectively over 14 years. Many were refractory at the time of enrollment. At 14-year follow-up, we acquired information on seizure type and frequency, psychiatric history, substance abuse, criminal activity, in addition to educational, vocational, and marital status through chart reviews and/or structured telephone interviews. Sixteen patients were only treated medically. They were divided by their following responses to medications: eight patients with less than one seizure per month were in the medically responsive group and eight patients with at least one seizure per month constituted the medically refractory group. Thirteen patients underwent focal resection for medically refractory epilepsy. Medically refractory patients displayed worse educational, vocational, social, and behavioral outcomes than medically responsive patients. Behavioral abnormalities persisted or evolved in five medically refractory patients when they became seizure free. Other studies have indicated that patients with medically refractory complex partial seizures have poor psychosocial outcomes. Although behavioral problems can occur even when seizures are well controlled, their early detection and treatment may be essential to the improvement of psychosocial outcomes. © 2001 by Elsevier Science Inc. All rights reserved.

Introduction Childhood-onset complex partial seizures (CPS) have a poorer prognosis for remission than other seizure types. Studies demonstrate that remission occurs in only 10-65% of patients with CPS [1-6], whereas the remission rate for other seizure types is 65-70% [3,6]. Although CPS is particularly refractory to antiepileptic medications, childhood-onset partial epilepsy (COPE) with CPS has been associated with a particularly poor seizure and psychosocial outcome [1,2,6-8]. Social maladjustment, severe behavioral abnormalities, and poor educational and vocational performance are prevalent among patients with symptomatogenic or cryptogenic COPE [1,2,5-8]. Behavioral abnormalities, such as affective, psychotic, anxiety, and personality disorders, are frequent among patients with medically refractory COPE [1,2,7,9,10]. Some authors have proposed that behavioral problems, independent of seizure severity or frequency, can impact psychosocial outcomes [2,10,11]. In patients undergoing temporal-lobe resection for medically refractory COPE, the improvement of some behaviors has been described [2,10,11] whereas other behaviors persisted or worsened despite a seizure-free outcomes [12-18]. This study reports the psychosocial outcomes of 29 patients with COPE and CPS 14 years after their enrollment. This group has been presented previously at enrollment [19] and 5-year follow-up [8]. Patients and Methods

Szabo´ CA, Rothner AD, Kotagal P, Erenberg G, Dinner DS, Wyllie E. Symptomatic or cryptogenic partial epilepsy of childhood onset: Fourteen-year follow-up. Pediatr Neurol 2001;24:264-269.

This study was presented as a poster at the Child Neurology Meeting in Baltimore, MD, in October 1995 (Ann Neurol 1995;38:529). From the *Division of Neurology; University of Texas Health Science Center; San Antonio, Texas; and †Department of Neurology; The Cleveland Clinic Foundation; Cleveland, Ohio.

264

PEDIATRIC NEUROLOGY

Vol. 24 No. 4

Patients. From a group of 599 patients seen for seizures by pediatric neurologists at The Cleveland Clinic Foundation in 1980, 52 patients were identified with an onset of symptomatic or cryptogenic epilepsy with CPS beginning at 16 years old or younger. Thirty patients received a definite diagnosis of symptomatic or cryptogenic COPE based on

Communications should be addressed to: Dr. Szabo´; Division of Neurology; University of Texas Health Science Center At San Antonio; 7703 Floyd Curl Drive; San Antonio, TX 78284-7883. Received March 20, 2000; accepted January 15, 2001.

© 2001 by Elsevier Science Inc. All rights reserved. PII S0887-8994(01)00246-6 ● 0887-8994/01/$—see front matter

clinical and electroencephalographic criteria according to the definition proposed by the International League Against Epilepsy (ILAE) [20]. CPS was defined as episodic change in behavior associated with an alteration of awareness or memory. In addition, all of the patients had generalized tonic-clonic seizures and/or simple partial seizures. Using strict electroencephalogram (EEG) criteria, only patients with focal epileptiform activity, such as sharp waves or spikes or spike-and-wave complexes on a routine EEG, were included [21,22]. Patients with only benign focal discharges of childhood, multifocal independent and generalized discharges, focal slowing, psychomotor variant, 14- and 6-Hz positive spikes, and other nonepileptiform sharp transients were excluded [21,22]. Twenty-two (76%) patients lived in the greater Cleveland or northern Ohio area, and six were referred from northern or western Pennsylvania. Follow-up was obtained in 29 patients in 1994, including one patient who died 2 years earlier. The only patient lost to follow-up had undergone a left-temporal lobectomy for medically refractory seizures. The patients consisted of 15 women and 14 men. Their mean age was 15 years (range ⫽ 6-24 years) at the time of enrollment, whereas the onset of their epilepsy was 8 years (range ⫽ 8 months to 16 years). They were followed by two of the investigators (D.R. and G.E.) for 1-11 years before enrollment. All patients had been treated with antiepileptic drugs for at least 6 months at the time the study was originally published. One third received monotherapy and the rest were on a combination therapy. Surgery had been performed in four patients before they were enrolled. All four had reduced but persistent CPS postoperatively. Surgery subsequently was performed in nine patients. The mean age of the 13 patients at the time of surgery was 18 years (range ⫽ 11-29 years). Focal epileptiform EEG abnormalities were right temporal in 14, left temporal in nine, and left posterior temporal, left parieto-occipital, left frontal and temporal, right frontal, biparietal, and bitemporal in individual patients. Hence by electroclinical criteria, 25 of 30 patients (83%) were diagnosed with temporal-lobe epilepsy. All but one patient undergoing epilepsy surgery was diagnosed with temporal-lobe epilepsy. Computed tomography (CT) scans were abnormal in 11 of 29 patients (38%), revealing structural lesions in four and nonspecific changes, such as atrophy, ventricular dilatation, or diffuse calcification, in seven patients. Magnetic resonance imaging (MRI) studies in five of these 11 patients revealed no further abnormalities. However, all MRI scans were performed before 1985. Methods. Charts were reviewed for all of the patients for information concerning previous medical and neurologic evaluations, seizure and medication history, education, and psychiatric diagnoses. Twenty-three patients (79%) were interviewed personally. In a structured telephone interview, they were asked about seizure frequency, antiepileptic drugs, education, employment, marital status and living situation, and psychiatric and criminal history. Family histories of epilepsy or mental illness were also obtained. Seizure frequencies were averaged over a 12-month period at the time of enrollment and at 5- and 14-year follow-up. Of the remaining six patients, two died from causes unrelated to their epilepsy (one in 1992 and the other in 1994). In the first patient, information was acquired from the father, with whom the patient had lived, and, in the second patient, from his last primary-care physician. Of the four remaining patients, immediate family members were interviewed (n ⫽ 2) or information was obtained from physicians’ records (n ⫽ 2).

Results The patients treated only medically were divided into medically responsive (n ⫽ 8) and medically refractory (n ⫽ 8) groups according to seizure frequency. The frequency of at least one CPS per month was chosen based on the stability of both groups over the course of the follow-up. Six of eight medically responsive patients became subsequently seizure free on antiepileptic drugs, whereas the remaining two patients continued to have less than one seizure per month. Seven of eight patients in the

medically refractory, nonsurgical group continued to have more than one seizure per month. The only patient to become seizure free was a young man who had been diagnosed with autism as a child and with bipolar disorder as an adult. He remained on carbamazepine for the treatment of his behavioral problems. Hence, only one patient would have changed groups over 14 years. The third group consisted of the medically intractable patients who underwent epilepsy surgery (n ⫽ 13). Eight of 13 surgical patients became seizure free (although auras persisted in five patients), two patients had greater than 75% seizure reduction (one of whom had been seizure free for 5 years), and three patients continued to have seizures. There was no significant difference between the medically responsive (n ⫽ 8) and refractory (n ⫽ 21) groups regarding age of onset of the epilepsy (two-tailed Student t test for unequal variances) or gender (Fisher’s Exact Test). The results regarding, educational, vocational, marital, and behavioral status are described below and summarized in Table 1. Education The medically responsive group exhibited superior educational achievement compared with the medically refractory groups. All eight patients in this group completed high school compared with 17 of 21 patients in the medically refractory group. One patient in the medically responsive group and seven patients in the combined medically refractory group had attended special education classes. Three patients of the medically responsive group obtained college degrees, and one patient was enrolled in college. Five of eight patients in the medically refractory, nonsurgical group had completed high school, and only one had attended college. Four of eight patients who had attended special education classes were in this group. In the surgical group, 12 of 13 patients completed high school. However, of four patients attending college, only one graduated with a degree. Employment In the medically responsive group, five of eight patients were employed full-time and one part-time, all in skilled professions, compared with six of 21 patients in the medically refractory group. In the medically refractory group, only one of eight patients was employed full-time. In the medically responsive group, all of the patients had skilled professions. Although the medically refractory patients pursued employment in the past, working mainly in service industries, they were unable to maintain their jobs as a result of recurrent seizures and/or behavioral problems. Although the highest rate of employment occurred in the surgical group (92%), four patients were in sheltered positions, two worked part-time, and two frequently changed jobs.

Szabo´ et al: Fourteen-Year Follow-Up of Complex Partial Seizures 265

Table 1.

Psychosocial outcome of patients grouped based on seizure frequency at the time of enrollment Number of Patients

Age of Onset of Epilepsy

Medically responsivea

8

Medically refractory, nonsurgicalb

Patient Group

Medically refractory, surgicalc

EEG Findings

Education

10 years

Right temporal in six Left temporal in two†

8

9 years

13

7 years

Right temporal in one Left temporal in two Bitemporal in three Extratemporal in two Right temporal in seven* Left temporal in four† Bitemporal in one Extratemporal in two

Eight HS graduates (one in SE) Three college degrees One enrolled in college Five graduates (three in SE) One SE

12 HS graduates (two in SE) One SE One college degree

Employment

Marital Status

Psychopathology

Five full-time (all skilled) One part-time

Four married Three living independently

Depression in one AD in one ADHD in one

One full-time (unskilled)

One married One living independently

Six full-time (4 skilled) Two part-time Four in sheltered jobs

Two married Six living independently

SA in three (depression in one, psychoses in two) MR in three (ADHD in two, bipolar disease in one) SA in five (depression in one) Depression in four (AN/B in one, AD/OCD/ADHD in one) MR in two (ADHD in one) Bipolar disorder in one Delinquency in two (SA in one)

⬍1 CPS per month, treated medically. At least one CPS per month, treated medically. c Treated surgically, all with at least one CPS per month preoperatively. † Second extratemporal interictal focus. * Right frontal lobe focus resected in 1. a

b

Abbreviations: AD ⫽ Anxiety disorder ADHD ⫽ Attention-deficit–hyperactivity disorder AN/B ⫽ Anorexia nervosa/bulimia EEG ⫽ Electroencephalogram HS ⫽ High school

MR OCD SA SE

⫽ ⫽ ⫽ ⫽

Mental retardation Obsessive-compulsive disorder Substance abuse Special education

Marital Status

Behavioral Problems

In the medically responsive group, seven of eight patients lived independently or were married compared with 11 of 21 patients of the medically refractory group. Only a few patients were married at long-term follow-up. Most patients lived with their parents, in group homes, or independently. All four men who were married at longterm follow-up were in the medically responsive group. They had all completed at least high school and were employed. Two of them had children, whereas none of the men in the medically refractory or surgical groups had families. Among medically refractory patients, only women were married. Furthermore, only one of eight patients with a seizure-free outcome after epilepsy surgery was married.

Psychopathology was diagnosed by a psychiatrist, psychologist, or pediatric neurologist by DSM-III criteria in 21 of 29 (72%) patients [23]. The diagnoses included mental retardation, learning disorders, attention-deficit– hyperactivity disorder (ADHD), autistic disorder, substance-related disorders, schizophrenia, and other psychotic disorders, in addition to mood, anxiety, eating, and personality disorders. Behavioral problems were present at or before enrollment in ten patients (34%), including five (17%) with ADHD, two with sociopathic behaviors, one with schizophrenia, one with bipolar disease, and one with anxiety disorder. The prevalence increased to 89% by 5-year follow-up. Although three of eight patients (38%) in the medically responsive group had behavioral abnor-

266

PEDIATRIC NEUROLOGY

Vol. 24 No. 4

malities, six of eight (75%) medically refractory patients and 12 of 13 (92%) surgical patients exhibited psychopathology. Hence, there was a significant difference in the prevalence of behavioral abnormalities between the medically responsive and refractory patients (P ⫽ 0.008, Fisher’s Exact Test, P ⬍ 0.05). The manifestations of behavioral abnormalities were age dependent. Autistic symptoms were present in one patient with infantile onset epilepsy, whereas in four of five patients with attention problems and hyperactivity, the onset of epilepsy was before 5 years of age and the diagnosis of ADHD was made at or soon after the onset of their epilepsy. In the remaining patient the diagnosis of ADHD preceded the onset of epilepsy by several years. Substance abuse became prominent during adolescence and early adulthood, occurring in eight (28%) patients, all with intractable CPS. Two patients admitted to delinquent and criminal acts during adolescence. Affective and psychotic behaviors presented during adolescence and young adulthood in eight patients. A history of depression was documented in six patients (20%), five of whom were females. Three patients had a family history of depression, two of whom developed depression postoperatively. Of four patients diagnosed with chronic psychoses, two had underlying bipolar disease, and one had schizophrenia and could not be specified in the last case. Only one patient, who became seizure free after a right temporal lobe resection, had a history of postictal psychosis. Personality disorders included anorexia nervosa in one patient with a history of depression, obsessive-compulsive disorder in two patients (one of whom was diagnosed with depression, the other with bipolar disorder), and anxiety disorder in two patients. A history of a severe psychiatric illness in a first- or second-degree relative was found in five patients (17%), all of whom had medically refractory epilepsy at enrollment. Two of these patients also had a family history of epilepsy. In four patients with a familial predisposition for psychiatric illness, the behavioral disturbance evolved in the setting of a spontaneous remission (one patient) or a seizure-free outcome after epilepsy surgery (three patients). Discussion This study examined the educational, vocational, marital, and behavioral status of patients with symptomatic or cryptogenic COPE. All patients had CPS. Educational, vocational, and marital status varied according to whether patients were in the medically responsive or medically refractory group at the time of enrollment. Whereas medically responsive patients tended to complete high school and college, medically refractory patients were less likely to complete high school and attend college. Fewer medically refractory patients achieved full-time employment, lived independently, or were married compared with medically responsive patients, despite a seizure-free out-

come in most surgical cases. Nonetheless, the outcome of those patients undergoing epilepsy surgery appeared more favorable than of the medically refractory patients who did not undergo surgery. Twelve of 13 children had high school degrees in the surgical group, which is in contrast to four of eight patients in the medically refractory, nonsurgical group. However, because most of the surgeries were performed around the age of 18 years, it is unlikely that seizure control alone would account for this difference. Although the age of onset of epilepsy did not appear to differ significantly between the surgical and nonsurgical cases, patients in the surgical group less often required special education classes. The ability to work and live independently was probably due to a combination of the better educational outcome and postoperative seizure control in the surgical group. Other studies identified postoperative seizure control, preoperative employment or education, low age at surgery, and postoperative improvements in cognitive functioning as predictors for postoperative employment, whereas persistent postoperative seizures predicted a poor vocational outcome [24-27]. Unfortunately, we did not have systematically collected neuropsychologic data on the medically and surgically treated patients for comparison. To understand the effects of surgery it would have been necessary to control for baseline cognitive functioning, social background, and behavioral factors. In addition, we did not have a nonepileptic control population to better assess occupational outcome. Few patients married in the medically refractory groups, and this finding was most pronounced in males. This is consistent with previous studies [27,28]. In the recent decade, there has been increased interest regarding the interaction of epilepsy and antiepileptic drugs with fertility and sexual function [29,30]. Although marriage rates among medically responsive males, all of whom remained on antiepileptic drugs, were similar to men who had become seizure free after epilepsy surgery, most of whom were still on antiepileptic drugs, it is likely that psychosocial and not biologic factors play a more important role [28-30]. Negative social stigmas include medically refractory epileptic seizures, the inability to gain independence, and the ability to provide the emotional or financial basis for a family. There is a need to identify and compare biologic and psychosocial factors among well-controlled and poorly controlled epileptic patients to understand the importance of their roles. Finally, behavioral problems may affect other aspects of psychosocial outcome as profoundly as medical intractability [2,7,10,12,13]. The relationship between behavioral problems and medically refractory COPE has been demonstrated by other authors [1,2,4,5,7], especially in association with temporal-lobe epilepsy. The most common behavioral problems include depression and anxiety disorders [1,2,7,9,13]. The prevalence and types of behavioral problems observed in our group were similar to those of other studies, except for a relative decrease of anxiety

Szabo´ et al: Fourteen-Year Follow-Up of Complex Partial Seizures 267

disorders compared with psychotic disorders [7,10,13,15]. Anxiety disorders may have been underdiagnosed in our study because patients did not routinely undergo psychiatric evaluation, and these symptoms usually are less apparent or disabling than depression or psychosis. The impact of behavioral problems was best exemplified by a patient who became seizure free after a right-temporal lobectomy. Although he was the only patient in the medically refractory surgical group to have graduated from college, the evolution of a bipolar disorder postoperatively led him to return to live with his parents and allowed him to work only part-time. Behavioral problems can persist or even evolve in a seizure-free period brought about by medication change or surgery [12-18,31]. This phenomenon has been noted in children with autistic regression [18,32] and with affective or psychotic disorders in adulthood [12-17]. Hence the emergence of behavioral abnormalities does not always appear to be related to seizure severity. A family history for depression or bipolar disease may have played a role in the evolution or persistence of psychiatric symptoms in some of our patients. Other studies have linked the site of resection [14,15], the persistence of postoperative seizures [7,10-15,17], and the underlying pathology [10,11,17] to postoperative affective and psychotic disorders. The role of a family history of psychiatric illness in the manifestation of depression and bipolar disorder after successful epilepsy surgery has not been formally studied so far. Nonetheless, depression in adult and childhood epileptic patients appears, in part, to be dependent on familial predisposition [9,33], and it is not unlikely that hereditary factors could play a role in the emergence of postoperative behavioral problems [18]. Although postoperative behavioral disturbances are rarely chronically debilitating, identification of risk factors for affective or psychotic disorders could help to counsel surgical candidates and their families. Because the number of patients was relatively small in this study and most of the patients were refractory to antiepileptic drugs at the time of enrollment, our focus was not on seizure outcome. Some of the patients had been followed before enrollment, and a few had already undergone epilepsy surgery. Most tertiary referral centers offer the most advanced medical or surgical treatments and their patients tend to be more refractory than those seen in population-based studies. This is a likely explanation for the relatively high proportion of our patients with temporal lobe epilepsy. Hence, it can be expected that remission rates are lower in these centers than in population-based studies [1,2,4,5]. However, tertiary referral centers tend to base their diagnoses on more detailed investigation. The results of this study are similar to those conducted at tertiary referral centers and relying on stricter electroclinical criteria for the diagnosis of epilepsy [4,5]. Nonetheless, more recent population-based studies implementing similar criteria may be in a better position to evaluate the

268

PEDIATRIC NEUROLOGY

Vol. 24 No. 4

interaction of psychologic or behavioral factors regarding educational, vocational, and social outcome [6]. Another shortcoming of this study was the lack of more current neuroimaging modalities. All of our patients were initially evaluated in the era of CT, with only five patients undergoing a subsequent MRI scan. However, before 1985, MRI was still inadequate in detecting epileptogenic lesions, specifically focal cortical dysplasia or hippocampal atrophy, and hence, added little to the CT scans. In summary, this study confirms that symptomatogenic and cryptogenic COPE is associated with a poor seizure and psychosocial outcome. Although our patient group was small and highly selected, it appears that patients with at least one CPS per month after instigation of treatment are at an increased risk for continued seizures and behavioral abnormalities. If patients with COPE do not respond to medications, it is necessary to focus on the educational and behavioral status. Early surgical intervention may improve educational, vocational, social, and behavioral outcomes; however, behavioral problems may persist or even worsen in some patients after epilepsy surgery.

References [1] Currie S, Heathfield WG, Henson RA, Scott DF. Clinical course and prognosis of temporal lobe epilepsy: A survey of 666 patients. Brain 1971;94:173-90. [2] Lindsay J, Ounsted C, Richards P. Long-term outcome in children with temporal lobe seizures. I: Social outcome and childhood factors. Dev Med Child Neurol 1979;21:285-98. [3] Annegers JF, Hauser WA, Elveback LR. Remission of seizures and relapse in patients with epilepsy. Epilepsia 1979;20:729-37. [4] Huttenlocher PR, Hapke RJ. A follow-up study of intractable seizures in childhood. Ann Neurol 1990;28:699-705. [5] Harbord MG, Manson JI. Temporal lobe epilepsy in childhood: Reappraisal of etiology and outcome. Pediatr Neurol 1987;3:263-8. [6] Sillanpa¨a¨ M, Jalava M, Kaleva O, Shinnar S. Long-term prognosis of seizures with onset in childhood. N Engl J Med 1998;338:171522. [7] Taylor DC, Falconer MA. Clinical, socio-economic, and psychological changes after temporal lobectomy for epilepsy. Br J Psychiatry 1968;114:1247-61. [8] Kotagal P, Rothner AD, Erenberg G, Cruse RP, Wyllie E. Complex partial seizures of childhood onset: A five-year follow-up study. Arch Neurol 1987;44:1177-80. [9] Robertson MM, Trimble MR, Townsend HRA. Phenomenology of depression in epilepsy. Epilepsia 1987;28:364-72. [10] Taylor DC. Mental state and temporal epilepsy: A correlative account of 100 patients treated surgically. Epilepsia 1972;13:727-65. [11] Davidson S, Falconer MA. Outcome of surgery in 40 children with temporal lobe epilepsy. Lancet 1975;1:1260-3. [12] Stevens JR. Psychiatric consequences of temporal lobectomy for intractable seizures: A 20-30 year follow-up of 14 cases. Psychol Med 1990;20:529-45. [13] Bladin PF. Psychosocial difficulties and outcome after temporal lobectomy. Epilepsia 1992;33:898-907. [14] Mace CJ, Trimble MR. Psychosis following temporal lobe surgery: A report of six cases. J Neur Neurosurg Psychiatry 1991;54: 639-44. [15] Trimble MR. Behavior changes following temporal lobectomy, with special reference to psychosis. J Neurol Neurosurg Psychiatry 1992;55:89-91. [16] Leinonen E, Tuunainen A, Lepola U. Postoperative psychoses

in epileptic patients after temporal lobectomy. Acta Neurol Scand 1994;90:394-9. [17] Andermann LF, Savard G, Meencke HJ, McLachlan R, Moshe S, Andermann F. Psychosis after resection of ganglioglioma or DNET: Evidence for an association. Epilepsia 1999;40:82-7. [18] Szabo´ CA, Wyllie E, Dolske M, Stanford LD, Kotagal P, Comair YG. Epilepsy surgery in children with pervasive developmental disorders. Pediatr Neurol 1999;20:349-53. [19] Dinner DS, Lu¨ders H, Rothner AD, Erenberg G. Complex partial seizures of childhood onset: A clinical and electroencephalographic study. Clevel Clin Q 1984;51:287-91. [20] Gastaut H. Clinical and electroencephalographical classification of epileptic seizures. Epilepsia 1970;11:102-13. [21] Jasper H. Report of committee on methods of clinical examination in electroencephalography. Electroencephalogr Clin Neurophysiol 1958;10:370-5. [22] Chartrian G, Bergamini L, Dondey M, Klass EW, LennoxBuchtal M, Peterse´n I. A glossary of terms most commonly used by clinical electroencephalographers. Electroencephalogr Clin Neurophysiol 1974;37:538-48. [23] American Psychiatric Association. Diagnostic and statistical manual of mental disorders, third ed. Washington, DC: Am Psychiatric Association, 1980. [24] Ziegler RG. Impairments of control and competence in epileptic children and their families. Epilepsia 1981;22:339-46. [25] Augustine EA, Novelly RA, Mattson RH, et al. Occupational

adjustment following neurosurgical treatment of epilepsy. Ann Neurol 1984;15:68-72. [26] Sperling MR, Saykin AJ, Roberts FD, French JA, O’Connor MJ. Occupational outcome after temporal lobectomy for refractory epilepsy. Neurology 1995;45:970-7. [27] Lendt M, Helmstaedter C, Elger CE. Pre- and postoperative socioeconomic development of 151 patients with focal epilepsies. Epilepsia 1997;38:1330-7. [28] Lindsay J, Ounsted C, Richards P. Long-term outcome in children with temporal lobe seizures. II: Marriage, parenthood, and sexual indifference. Dev Med Child Neurol 1979;21:433-40. [29] Jalava M, Sillanpa¨a¨ M. Reproductive activity of offspring health of young adults with childhood-onset epilepsy: A controlled study. Epilepsia 1997;38:532-40. [30] Morrell MJ. Sexuality in epilepsy. In: Engel J, Pedley TA, eds. Epilepsy: A comprehensive textbook. Philadelphia: Lippincott-Raven Publishers, 1997:2021-6. [31] Wolf P. Acute behavioral symptomatology at disappearance of epileptiform EEG abnormality: Paradoxical or “forced” normalization. In: Smith DB, Treiman DM, Trimble MR, eds. Neurobehavioral problems in epilepsy. Advances in Neurology, Vol. 55. New York: Raven Press, 1991:127-42. [32] Amir N, Gross-Tsur V. Paradoxical normalization in childhood epilepsy. Epilepsia 1994;35:1060-4. [33] Brent DA, Crumrine PK, Varma RR, Allan M, Allman C. Phenobarbital treatment and major depressive disorder in children with epilepsy. J Pediatrics 1987;80:909-17.

Szabo´ et al: Fourteen-Year Follow-Up of Complex Partial Seizures 269