- Email: [email protected]
The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study
Seizure 2001; 10: 250–259
doi:10.1053/seiz.2000.0503, available online at http://www.idealibrary.com on
The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study HELGE BJØRNÆS, KIRSTEN STABELL, OLAF HENRIKSEN & YNGVE LØYNING The National Center for Epilepsy, Sandvika, Norway Correspondence to: Helge Bjørnæs, The National Center for Epilepsy, N-1303, Sandvika, Norway. E-mail: [email protected]
The aim of this study was to compare the long-term consequences of refractory epileptic seizures for intellectual functioning in pediatric and adult patients, taking the severity of the epilepsy into consideration. Thirty-four patients, 17 children (mean age 10.2 years) and 17 adults (mean age 24.4 years) were tested twice with the age-appropriate version of Wechsler’s Intelligence Scales. The mean test–retest interval in the two groups was 3.5 and 6.0 years, respectively. There were no statistically significant differences between the groups with respect to severity of the epilepsy at Test 1, as indicated by retrospective assessments of seizure severity, interictal EEG pathology, and number of antiepileptic drugs received per patient. Assessments of changes in these variables during the test–retest interval did not indicate different courses of the disease in the two groups. Despite these similarities, a statistically significant difference was found between the children and the adults regarding changes in intellectual functioning. In the children, there was a decline in mean intelligence quotient (IQ) scores during the test–retest interval, while the IQ scores increased in the adult group. It is concluded that recurrent seizures may represent a considerable risk for intellectual decline in children, while intellectual functioning seem to be less vulnerable in adults with early onset of epilepsy. c 2001 BEA Trading Ltd
Key words: refractory epilepsy; intelligence; children; adults.
INTRODUCTION The effects chronic epilepsy may have on intellectual functioning are of great concern to many patients and their relatives. They are afraid that the illness may lead to a gradual deterioration. However, longitudinal studies of patients who respond to drug treatment, have failed to demonstrate any intellectual decline. This applies both to children1, 2 and adults3–5 . About 20% of patients with epilepsy, however, do not respond to drug therapy6 . Intractability may be due to underlying pathological processes that are more severe than, or different from, those in patients with a good response7, 8 . Yet, adult patients with intractable seizures may not experience intellectual decline4, 9, 10 , while children with poor seizure control seem to be at risk of intellectual retardation11–13 . Thus, it seems that the long-term effects of intractable seizures in children are more serious than in adults. The evidence presented is not conclusive, however, since descriptions of seizure types and seizure frequencies are not given in comparable units. Hence, it is not possible to conclude 1059–1311/01/040250 + 10
$35.00/0
from these studies whether the children may have had a more severe seizure disorder than the adults. It would be preferable to have some common measures of the severity of the seizure condition, and, in follow-up studies, of the changes in the condition. The present study, therefore, was undertaken to explore and compare the consequences of recurrent seizures for intellectual functioning in children and adults, taking both seizure severity and course of the disorder into consideration.
MATERIALS AND METHODS Patient selection Patients were recruited among 162 consecutive patients included in the epilepsy surgery program at the National Center for Epilepsy, Norway, during the previous 10 years14 . Those who had had an IQ testing during earlier referrals to the center were considered for the study. At the time of the first testing (Test 1), c 2001 BEA Trading Ltd
A longitudinal study
the patients were referred to the epilepsy center for an expert evaluation and treatment of their seizures. They were subsequently treated with antiepileptic drugs for an average of 5 years before they were recognized as surgical candidates. The second testing (Test 2) was administered as part of the pre-operative evaluation. Patients were included in the study if they had complete protocols of the Norwegian versions of the Wechsler Intelligence Scale for Children Revised (WISC-R)15 or the Wechsler Adult Intelligence Scale (WAIS)16 , and were at least 18 years of age at Test 1, or less than 18 years of age at Test 2. Seventeen children (6–17 years of age at Test 2) and 17 adults (18–35 years of age at Test 1) fulfilled the criteria. The selected patients may have had a more severe seizure condition than the majority of our surgical candidates. However, this would pertain to both the pediatric and the adult patients and thus not violate the basis for the comparisons between the two groups. One child and one adult patient were considered for corpus callosotomy, while resective surgery was planned in the remainder. In addition to these 34 patients, complete test protocols were available in 15 patients who were pediatric patients at Test 1 but adults at Test 2. Because it was not possible to decide whether changes in IQ scores in these patients had taken place during childhood or later, their data were not included in the statistical comparisons between children and adults, but were evaluated separately to give some additional information on long-term consequences of recurrent seizures. Demographic data from this additional group are presented in the last column of Table 1, under the heading ‘Children retested as adults’.
Methodological issues pertaining to the intelligence tests The psychometric properties of the Norwegian standardization of WISC-R and WAIS, as described in the handbooks, fulfil standard criteria15, 16 . Practice effects
Upon retesting with intelligence tests under stable conditions, with test–retest intervals between 2 and 20 weeks, the results obtained in normal persons will usually improve by an average of 5–7 points in Full Scale IQ scores. Such practice effects are observed to the same extent in normal children and normal adults17–19 . However, studies of patients with uncontrolled seizures fail to find appreciable test– retest improvements3, 4, 20, 21 . Consequently, the practice effects might be negligible in the present study of patients with frequent seizures, who were retested
251
after relatively long intervals. Besides, the test–retest intervals did not differ between the groups. Practice effects, therefore, were not expected to cause any systematic differences between the groups. IQ scores
Verbal, Performance and Full Scale IQ (FSIQ) scores were computed for each patient using Norwegian norm tables15, 16, 22 . IQ scores are age adjusted in both children and adults. Difference scores (Test 2–Test 1) were computed based on the individual Verbal, Performance and FSIQ scores.
Seizure-related factors that may influence the test performance Test performance may be influenced by factors such as seizure activity3, 23–27 , pathological brain activity as seen on the EEG28–31 , and side effects of the antiepileptic drugs (AEDs)32–34 . Systematic differences between the groups with respect to these factors could contribute to differences in the test results. Consequently, attempts were made to estimate these factors retrospectively in the medical files of each patient as described in the following. Seizure severity
Number of seizures during the previous month prior to the two test occasions were recorded, based either on the patients’ own seizure diaries or on the daily records kept by the staff. Many patients had several seizure types, and in some patients the seizure types changed over time. Therefore, a mere count of seizures would not give an adequate evaluation of the seizure severity. The need for expressing the seizure severity rather than the seizure frequency is underlined by several authors, and appropriate scales are available35–37 . However, these scales cannot be used retrospectively. Thus, in the present study, clinically based assessments of seizure severity were performed. Three experienced epileptologists rated the records of each patient independently and blindly, taking both seizure frequency and seizure types into consideration. The seizure condition before Test 1 was rated along a graded 5-point scale ranging from 0 ‘no seizures’ to 4 ‘severe seizure condition’. The ratings were highly reliable (Kendall coefficient of concordance = 0.90, P< 0.001). Individual changes in the seizure condition between test sessions were rated along a graded 7-point scale ranging from −3 ‘much worse’ to +3 ‘much better’ (Kendall coefficient of concordance = 0.94, P< 0.001). Examples of ratings are given in Table 2.
252
H. Bjørnæs et al.
Table 1: Demographics and seizure-related variables. Variable 1. 2. 3.
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
16. 17.
18.
19.
Number of patients Sex (F/M) Etiology, number of patients Idiopathic or crypt./symptomatic Cerebrovascular lesion Intracranial infection Perinatal encephalopathy Asphyxia Post-traumatic Tumour Age at seizure onset, mean yr (SD) Age at Test 1, mean (SD) Seizure severity at Test 1, median (q-range) EEG pathology prior to Test 1, median (q-range) Number of AEDs received per patient at Test 1, median (q-range) Duration of epilepsy at Test 1, mean yr (SD) Test–retest interval, mean yr (SD) Age at Test 2, mean (SD) Change in seizure severity, median (q-range) Change in EEG pathology, median (q-range) Change in medication, median (q-range) Laterality of epileptogenic region, number of patients Left hemisphere/Right hemisphere Generalized Temporal/extratemporal location of epileptogenic region, number of patients Type of surgery planned, number of patients Resection Callosotomy Seizure types previous month prior to Test 1, number of patients Atypical absences Simple partial seizures Complex partial seizures Generalized tonic–clonic seizures (sec. gen) Misc. (Tonic, atonic, myoclonic, undetermined, etc.) Seizure types previous month prior to Test 2, number of patients Atypical absences Simple partial seizures Complex partial seizures Generalized tonic–clonic seizures (sec. gen) Misc. (Tonic, atonic, myoclonic, undetermined etc.)
Study groups Children Adults
P-value
Children retested as adults
NSa
15 10/5
17 8/9
17 7/10
14/3 1 1 0 1 0 0 5.0 10.2 3.0 2.0
(3.0) (2.6) (2) (1)
13/4 0 2 1 1 0 0 9.5 24.4 2.3 2.0
(7.2) (4.6) (1) (1)
<0.05b
1.0
(1)
2.0
5.2 3.5 13.7 1.0 −1.0 0.0
(2.9) (2.5) (2.6) (6) (2) (1)
14.8 6.0 30.4 1.0 0.0 0.0
8/8 1 11/5
8/7 1 15/1
16 1
16 1
0 4 12 4 (3)
0 2 13 11 (10)
NSa NSa NSa <0.05a
2 3 12 6
2
2
NSa
0
1 3 14 5 (4)
0 3 14 11 (10)
NSa NSa NSa <0.05a
0 4 14 9
2
2
NSa
1
NSd
NSc NSc
9/6 0 1 1 0 2 2 5.3 13.6 1.3 2.0
(3.5) (2.6) (1) (2)
(1)
NSc
1.0
(2)
(9.0) (4.8) (6.0) (5) (1) (1)
<0.001b NSb NSc NSc NSc NSa NSd
8.3 11.0 24.6 −4.0 −1.0 0.0
(4.9) (5.6) (5.1) (6) (1) (1)
6/9 0 10/5
15 0
Abbreviations: F, female; M, male; Crypt, cryptogenic; q-range, quartile range; Yr, years; SD, standard deviation; AEDs, antiepileptic drugs; Sec. gen, secondarily generalized; Misc, miscellaneous. a Chi-squared; b Student’s t-test for independent samples; c Median test; d Fisher exact probability test.
Interictal EEG pathology
For each patient, the interictal EEG records closest in time to the test sessions were selected from the medical files. In most patients, EEG records were taken a few days or weeks ahead of the test sessions. In a few patients, the records were done a couple of months
earlier. However, since the test occasions were separated by several years in most patients, we assumed that these samples of EEGs were, to some degree, representative of the patients’ interictal EEG characteristics at the time of the assessments. Summaries of the EEG records were blindly evaluated by a senior neurophysiologist with respect to degree (not lo-
A longitudinal study
calization) of EEG pathology prior to Test 1, and change in the degree of EEG pathology between Test 1 and Test 2. Both evaluations were made according to graded 5-point scales, and comprised background activity, focal slowing, persistent/paroxysmal activity, sharp components, and spike and wave activity. Number of AEDs
The use of AED monotherapy with drug serum levels within the therapeutic range may have negligible side effects on cognition, while combinations of drugs are more likely to cause adverse effects33, 34 . Thus, the number of AEDs received per patient at the first testing, and changes in these numbers from Test 1 to Test 2, were recorded. Two children and one adult were without medication at Test 1, none were without medication at Test 2.
253
tial testing, with a mean duration of 5.2 years in the children and 14.8 years in the adults. More adults than children had secondarily generalized tonic–clonic seizures during the previous month prior to both test occasions. There were, however, no significant differences between the groups with respect to sex, etiological syndromes or epilepsy syndromes. All patients but one in each group had localization-related partial epilepsy. One child had primarily generalized tonic– clonic, atonic and atypical absence seizures of cryptogenic etiology, while one adult had Lennox–Gastaut syndrome of cryptogenic etiology and with tonic– clonic and atonic seizures. Furthermore, there were no between-group differences in laterality or location of the epileptogenic region, nor in seizure severity, EEG pathology and number of drugs per patient at Test 1. Changes from Test 1 to Test 2 in the latter three variables did not differ between the groups (Table 1).
Data analysis
IQ scores
Categorical data in the two study groups were analysed using the chi-squared test wherever the requirements for this test were met. In 2 × 2 contingency tables with few expected frequencies, the Fisher exact probability test was performed. Ordinal-scale data were compared using the median test. Comparisons of continuous demographic variables were performed by Student’s t-test for independent samples. The IQ scores were submitted to analysis of variance. Post hoc tests were performed using Duncan’s multiple range test. To decide whether the changes from Test 1 to Test 2 were influenced by regression toward the mean, product-moment correlations were computed between the IQ scores at Test 1 and the difference scores. Regression toward the mean is significant if the initial values are significantly inversely correlated with the difference scores38 . P-values less than 0.05 in twotailed tests were considered statistically significant. All statistical analyses were run using ‘Statistica for the Macintosh’39 .
Mean Verbal, Performance, and FSIQ scores of the two groups at the two test occasions are shown in Table 3. Initially the children’s FSIQ scores were in the lower end of the normal range (i.e. 85–115). At Test 2, however, their scores had declined on average by 4.6, 6.1, and 5.9 points in Verbal, Performance, and FSIQ, respectively. Mean FSIQ now fell below the normal range. The adults had IQ scores well below the normal range at Test 1, but in contrast to the changes in the children, their corresponding scores increased by 4.2, 7.9, and 6.4 points. To assess whether these differences between the two groups were significant, we subjected the IQ scores to 2 (group) × 2 (test occasion) analyses of variance (ANOVA). There were no significant differences between the groups with respect to mean IQ scores, both test occasions taken together, nor were there any significant test–retest changes, both groups taken together, as shown by the insignificant main effects. However, significant interaction effects were revealed for Group × Test occasion in all the three types of IQ scores. This indicated that the decrease in IQ scores seen in the pediatric patients and the increase in the adults, represented significantly different courses. Post hoc analyses showed that the changes in Performance and FSIQ were statistically significant within each group. Post hoc between-group comparisons showed significantly higher Verbal, Performance, and FSIQ scores in the children at Test 1, but no significant differences at Test 2. Because the groups differed with respect to age at onset of habitual seizures, we performed similar analyses as shown in Table 2 with age at onset as a co-
RESULTS Demographic and seizure-related variables
Demographic and seizure-related variables of the two study groups are presented in Table 1. All but two patients in the present study had seizure onset in childhood, the children being on average 5 years old at seizure debut, and the adults on average 9.5 years old. This difference was statistically significant. There were also significant differences between the groups with respect to the duration of epilepsy at the ini-
254
H. Bjørnæs et al.
Table 2: Examples of seizure severity ratings. Patients Sz/month prior to Test 1 Abs SPS CPS GTC Misc HH EA HB IT
150 25 1
9
2 1
135
Ratings of sz-severity Rater A B C 4 3 3 1
4 2 3 2
4 2 3 1
Sz/month prior to Test 2 score Abs SPS CPS GTC Misc 12 7 9 4
100 500
25 15
16 20 6
30
Ratings of change Rater A B C 3 −3 −1 −2
3 −3 0 −1
3 −3 0 −3
score 9 −9 −1 −6
Abbreviations: Sz, seizures; Abs, atypical absences; SPS, simple partial seizures; CPS, complex partial seizures; GTC, generalized tonic–clonic seizures; Misc., miscellaneous seizure types: tonic, atonic, myoclonic, undetermined. Frequency of different seizure types prior to Test 1 and Test 2 in four patients, and actual ratings by three epileptologists (rater A, B, and C) of seizure severity at Test 1 and change in seizure severity from Test 1 to Test 2. The ratings of seizure severity were done using a graded 5-point scale ranging from 0, no seizures, to 4, severe seizure condition. Ratings of change were done according to a graded 7-point scale ranging from −3, much worse, to 3, much better. The individual patient’s score was the sum of the ratings. Thus, individual seizure severity ranged from 0 to 12, and individual change in seizure severity from −9 to 9.
variate. The interaction effects were still significant in Performance and FSIQ (F = 10.3, P< 0.01; F = 8.2, P< 0.01, respectively), but not in Verbal IQ (F = 3.8, P> 0.06). No significant main effects were obtained. Although the groups differed also in duration of epilepsy at Test 1, this variable is dependent on the age at onset of seizures and the age at Test 1, which was part of the grouping criterion. Consequently, we did not find it appropriate to use this variable as an additional covariate. The 15 children who were retested as adults (additional group) were on average 5.3 years of age at onset of seizures (Table 1). They were on average 13.6 years of age at Test 1 and 24.6 at Test 2. Mean Verbal, Performance, and FSIQ scores at Test 1 were 83.8, 88.9, and 84.7, respectively. These mean scores were not significantly (NS) different from the corresponding scores of the pediatric group at Test 1 (t-values were 0.5, NS; 0.5, NS; and 0.5, NS, respectively, t-test for independent samples). During the test–retest interval, which was on average 11 years, the additional group declined in Verbal, Performance and FSIQ scores by 6.4, 7.7, and 7.7 points, respectively. The changes were statistically significant (t-values were 3.2, P< 0.01; 2.4, P< 0.05; 3.2, P< 0.01, t-test for correlated samples). The decline was not significantly different from the decline in the pediatric group (t-values were 0.5, NS; 0.4, NS; and 0.5, NS; t-test for independent samples).
DISCUSSION In this retrospective study comparing changes in IQ scores at two test occasions in pediatric and adult patients with refractory seizures, we found significantly different courses in intellectual functioning between the groups: The children’s scores declined, while the adults’ increased. Post hoc analyses showed that the within-group changes were significant in Per-
formance and FSIQ in both groups. These findings are in agreement with previous longitudinal studies in adults9, 10 , and in children11–13 . The differences between children and adults in the current study could not be ascribed to differences in etiology or in epilepsy syndromes. Furthermore, the results could probably not be explained by a more severe epilepsy initially in the children, or a more progressive course of their disease, as indicated by assessments of seizure severity and other seizure-related variables (Table 1). A bias might be introduced, however, if the patients selected for the study initially were referred for neuropsychological assessment in a period when they temporarily had an increased seizure frequency. This could give a test–retest gain in the adults and attenuate the decline in the children. However, the changes in seizure severity gave no indications that this was the case (Z = 0.9, P> 0.3, sign test). Regression toward the mean did not significantly influence the results in either group. We assume that the findings reflect a greater vulnerability of intellectual functioning in children than in adults, due to different stages in the development of intellectual abilities. This development may be conceived as a cumulative process undergoing particularly rapid growth in childhood40 . Roughly, this growth can be operationally defined by the increase in the sum of WISC-R raw scores required from one year to the next for a child to maintain mean scaled scores, i.e. scaled scores of 10, and an average FSIQ15 , as illustrated in Fig. 1. Rate of growth in intellectual abilities, as defined here, decelerates and levels off in young adulthood. This can be illustrated in the same way by adding the raw scores necessary to maintain mean scaled scores on the WAIS for successive age groups41 (see Fig. 1). Normal development, as reflected by the curves in Fig. 1, depends on many factors. Among these, maturational processes of the brain and optimal opportunities for the child to learn from interaction with a normal environment, are among the most essen-
A longitudinal study
255
400 Normal development
Sum of raw scores
WISC-R
WAIS
300
Development of case GA
200
IQ=57
IQ=53
IQ=64 IQ=75
100
0 6
8
10
12
14
16
18
20
22
24
26
28
30
Age Fig. 1: Illustration of normal intellectual development (longer curve) and the slowing of development with gradually diminishing FSIQ of case GA (shorter curve). In the WISC-R norm tables, which are based on the standardization samples of children at different ages, the total number of correct solutions required to obtain a FSIQ of 100 increases by a certain amount from one age level to the next. This amount operationally defines the normal intellectual development of children with average intelligence. By adding the numbers of correct solutions (actually the raw scores) for each age level one can observe the normal growth in intellectual abilities at successive ages (longer curve). Similar sums of raw scores from the WAIS show the intellectual development from age 16 to age 30. (To join the curves from the WISC-R and the WAIS into one smooth curve a parallel displacement had to be done by adding a constant number to the WAIS values.) The actual number of correct solutions (raw scores) given by case GA, who was tested four times with the WISC-R, are shown by the shorter curve, and the FSIQ scores she obtained are shown beneath this curve. She was 9 years and 9 months of age at the first assessment, and was subsequently tested at the age of 11 years and 6 months, 13 years and 7 months, and 15 years and 2 months, respectively. GA had cryptogenic epilepsy with seizure onset at 1 year of age, and underwent right temporal lobe resection when she was 13 years of age. She has been completely seizure free since then, i.e. for more than 15 years. Abbreviations: WISC-R, Wechsler Intelligence Scale for Children-Revised; WAIS, Wechsler Adult Intelligence Scale; FSIQ: Full Scale Intelligence Quotient.
tial40, 42 . Whether epilepsy interferes directly with the maturational processes of the brain is a matter of considerable debate43, 44 . However, it is generally recognized that an uncontrolled epilepsy may interfere with the child’s cognitive functioning1, 12, 13, 45, 46 and with its opportunity to interact with the environment47, 48 . This leads to disadvantages in scholastic as well as social participation42, 49–52 . Thus, the epileptic child’s opportunities to learn from normal experience are disturbed47, 48 . This may interfere with the speed of intellectual growth, and the child’s abilities may develop at a slower rate and gradually lag behind those of its agemates. Because IQ scores are age adjusted, such slowing is reflected by a diminishing IQ. This is illustrated by the results after serial testing with the WISC-R of the case GA in Fig. 1, showing a gradual decrease in FSIQ scores despite a small gradual increase in raw scores (see also the legend). Similar explanations for the long-term consequences for intellectual development in children with early brain injuries have been proposed by Fletcher
et al.53 and Taylor and Alden54 , and in patients with epilepsy by Vargha-Khadem et al.55 and Glosser et al.50 The latter authors related the delayed cognitive development in their patients to the possible interference of seizures with neural maturation during critical periods in childhood. The present authors also assume that disturbances of cognitive functioning and, in some cases, a restricted life caused by frequent seizures, may play a role in retarding intellectual development. Neyens et al.56 argued that cognitive decline in children with epilepsy seem to occur in close temporal proximity to seizure onset, and may be caused by the underlying pathophysiological processes. Our data are not in unequivocal support of this hypothesis, as about 20% of the children who had had seizures for at least 5 years at Test 1 declined by 10 or more IQ points during the test–retest interval. The children had higher IQs than the adults at Test 1. This might be a referral artefact, e.g. if the threshold for being referred for neuropsychological assessment was systematically different with respect to in-
256
H. Bjørnæs et al.
Table 3: Mean IQ scores and standard deviations in pediatric and adult patients. Test 1
Test 2
Main effects group
VIQ PIQ FSIQ
Children Adult Children Adult Children Adult
86.8 (17.3) 80.1 (19.2) 92.5 (15.8) 73.5 (20.7) 88.1 (16.6) 75.1 (20.5)
82.2 (19.4) 84.2 (22.3) 86.4 (15.8) 81.5 (20.7) 82.1 (18.6) 81.5 (22.3)
Interactions test
F
P
F
P
F
P
0.1
NS
0.0
NS
5.3
<0.05a
3.8
NS
0.3
NS
13.2
<0.001a
1.1
NS
0.0
NS
10.9
<0.01a
Abbreviations: VIQ, Verbal IQ; PIQ, Performance IQ; FSIQ, Full Scale IQ; NS, not significant. a Two-way analysis of variance (ANOVA). Post hoc tests: Within-group changes Test1–Test 2: Children: VIQ, NS; PIQ, P< 0.05; FSIQ, P< 0.05. Adults: VIQ, NS; PIQ, P< 0.01; FSIQ, P< 0.05. Between-group differences: Test 1: VIQ, P< 0.05; PIQ, P< 0.001; FSIQ, P< 0.001. Test 2: VIQ, NS; PIQ, NS; FSIQ, NS.
tellectual functioning in children and adults. The differences are, however, more likely due to the relatively short seizure history at Test 1 in the children, namely 5.2 years on average. Children with a shorter seizure history are reported to have higher IQs than children with a longer duration of epilepsy45 . The course of the intellectual functioning in the pediatric patients in the present study corroborates these findings. At Test 2, the children had had seizures for an average of 8.7 years, and their mean FSIQ had declined by 5.8 points and was no longer significantly different from the FSIQ of the adult group. This is in agreement with the ‘retarded growth’ hypothesis. In accordance with a developmental model, we would also expect an intellectual decline in the 15 children who were retested as adults. However, although the decline was statistically significant, the interpretation of the results remains uncertain, because of a significant regression toward the mean. Intellectual decline in children with epilepsy seems to depend on the severity of the epilepsy1, 2 . In a 4-year follow-up of children with epilepsy, Aldenkamp et al.57 could find only insignificant drops in mean IQ scores. The differences with the current findings may be due to differences in the severity of the illness, as 22% became seizure free in the study by Aldenkamp et al.57 , while this was not the case in any of our patients. One may wonder why the adults in the present study, who had recurrent seizures of similar severity as the children, did not decline in IQ scores. It is well known that seizures also disturb cognitive functioning in adults3, 24, 25, 58 , and interfere with their opportunities to participate in a normal way in intellectual activities within education42 , work59, 60 , and social settings61, 62 . In addition, the findings of atrophic and sclerotic brain areas in adults with epilepsy raise the question of whether long-standing seizures may lead to progressive brain lesions63, 64 with concomitant hazards for cognitive functioning65 . In spite of these potential hazards, however, we did
not find appreciable intellectual decline in any of the 17 adult patients, even after follow-up periods of more than 15 years. Only three of them had minor decreases, by 1–4 points, in FSIQ scores. In contrast, the adult patients, as a group, experienced gains in Performance and FSIQ. If practice effects may be disregarded, these gains may be due to the introduction of AEDs with less side effects. All but two adult patients had different AEDs at Test 2 compared to Test 1. (Of the children, 13 had changes in types of drugs received, which may have attenuated the decline to some degree.) On the other hand, the adult patients had lower initial IQ scores than the children, and consequently less room for further decline, as measured by the Wechsler scales. They had a mean seizure history of 15 years at Test 1, and, as all but two had seizure onset before the age of 16, they might already have gone through a retardation in their intellectual development. In addition, more adults than children had generalized tonic–clonic seizures prior to both test occasions, which may have contributed to their lower level of performance. A ‘floor’-effect, therefore, may have limited the potential decline in some of the adults. However, previous longitudinal studies in adult patients with normal or low normal intelligence could not find significant adverse changes in IQ scores4, 9, 10 , in agreement with the present findings. Two of these studies9, 10 found small, but significant, gains in IQ scores which may also have been effects of optimalization of drug treatment. The relative stability of functions in adult patients with recurrent seizures is probably not limited to intelligence. Selva et al.9 found no significant decline in measures of memory in a 2.4-year follow-up study. Kalska4 reported no adverse changes in memory, cognitive flexibility and motor function during a 10-year test–retest interval. In the 10-year follow-up study by Holmes et al.10 , a slight but significant decline was seen in six out of 16 neuropsychological tests, mostly in measures of speed and visuo-spatial relations. How-
A longitudinal study
ever, the percentage of results outside normal limits was not changed after 10 years. The investigators concluded that although no overall adverse changes were seen in intellectual or neuropsychological functioning, slight losses in selected areas could not be ruled out. None of the above studies4, 9, 10 considered patients with onset of epilepsy in adulthood. The long-term cognitive effects of seizures in this group remain to be decided. Mesial temporal sclerosis and slightly impaired memory seem to be present in adults with newly diagnosed temporal lobe seizures, indicating that the lesions may have preceded the seizures65 . Recurrent seizures may exacerbate the lesion66 . The progressive nature of epilepsy is, however, disputed43, 44, 65 . Holmes et al.10 regarded their findings as evidence against a progressive process underlying drugresistant epilepsy in adults. On the other hand, a recent cross-sectional study by Jokeit and Ebner67 indicated that epileptic seizures may have a cumulative adverse effect only discernible in adult intelligence after more than 30 years with seizures. An average decline in FSIQ of 5.9 points in children in the current study may seem trifling. But 40% of them declined by 10 or more points, while only one child experienced a comparable gain. In the additional group of children retested as adults, 47% declined by 10 or more points. We find this noteworthy. Although the patients selected for this study may not have been representative for the population of our surgical candidates in all respects, with the additional group they constituted a substantial part (30%), and represent those with an early onset of refractory seizures, which may give the most serious long-term consequences for intellectual functioning. This is a group of surgical candidates most epilepsy centres will encounter, unless they reject patients with low FSIQ from their surgical program, as advocated by some authors68, 69 .
CONCLUSIONS The present study corroborates the findings from previous studies4, 9–13 : adult patients with an early onset epilepsy may not experience a decline in IQ scores, despite recurrent seizures over several years. The longterm consequences of recurrent seizures in children and adolescents may be more serious, as they seem to be at risk for intellectual decline relative to ageadjusted norms. According to our results, the differences are probably not due to a more severe or progressive seizure disorder in the pediatric patients. As an explanation for the different trends seen in children compared to adults, we proposed a developmental model, suggesting that intellectual abilities may be particularly vulnerable to disturbances related
257
to recurrent seizures during the period of rapid development in childhood. Consequently, drug treatment that does not result in satisfactory seizure control in children should not be unduly prolonged before surgical treatment is considered. Only further investigations can decide whether the present finding apply to patients with seizure onset in adulthood.
ACKNOWLEDGEMENTS This study was supported by a grant from the Research Fund of The National Center for Epilepsy. The authors are very grateful to Drs Ebba Wannag, Einar Kinge, and Renate Dahl for assessing the seizure severity in the patients, and to Drs Rasmus Lossius, Karl Otto Nakken and Robert Kloster for providing information on the patients’ AED records. We also thank Drs Svein I. Johannessen and Karl Otto Nakken for valuable advice during the preparation of the manuscript, and Professor Kenneth Hugdahl, Professor Kjetil Sundet, Dr Petter Mowinckel and Dr Lien My Diep for help and advice in connection with the statistical analyses.
REFERENCES 1. Bourgeois, B. F. D., Prensky, A. L., Palkes, H. S., Talent, B. K. and Busch, S. G. Intelligence in epilepsy: A prospective study in children. Annals of Neurology 1983; 14: 438–444. 2. Ellenberg, J. H., Hirtz, D. G. and Nelson, K. B. Do seizures in children cause intellectual deterioration? New England Journal of Medicine 1986; 314: 1085–1088. 3. Seidenberg, M., O’Leary, D. S., Berent, S. and Boll, T. Changes in seizure frequency and test-retest scores on the Wechsler Adult Intelligence Scale. Epilepsia 1981; 22: 75–83. 4. Kalska, H. Cognitive changes in epilepsy. A ten-year follow-up. Commentationes Scientiarum Socialium 1991; 44: 1–85. Helsinki, The Finnish Society of Sciences and Letters. 5. Dodrill, C. B. and Wilensky, A. J. Neuropsychological abilities before and after 5 years of stable antiepileptic drug therapy. Epilepsia 1992; 33: 327–334. 6. Hopkins, A. and Shorvon, S. Definitions and epidemiology of epilepsy. In: Epilepsy 2nd Edition, (Eds A. Hopkins, S. Shorvon and G. Cascino). London, Chapman and Hall Medical, 1995: pp. 1–24. 7. Lesser, R. P., L¨uders, H., Wyllie, E., Dinner, D. S. and Morris III, H. H. Mental deterioration in epilepsy. Epilepsia 1986; 27 (Suppl. 2): S105–S123. 8. Engel, J. Jr and Shewmon, D. A. Overview: Who should be considered a surgical candidate? In: Surgical Treatment of the Epilepsies (Ed. J. Engel Jr). New York, Raven Press, 1993: pp. 23–34. 9. Selwa, L. M., Berent, S., Giordano, B., Henry, T. R., Buchtel, H. A. and Ross, D. A. Serial cognitive testing in temporal lobe epilepsy: Longitudinal changes with medical and surgical therapies. Epilepsia 1994; 35: 743–749. 10. Holmes, M. D., Dodrill, C. B., Wilkus, R. J., Ojemann, L. M. and Ojemann, G. A. Is partial epilepsy progressive? Ten-year follow-up of EEG and neuropsychological changes in adults with partial seizures. Epilepsia 1998; 39: 1189–1193. 11. Dodrill, C. B. and Troupin, A. S. Seizures and adaptive
258
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
H. Bjørnæs et al. abilities—a case study of identical twins. Archives of Neurology 1976; 33: 604–607. Lindsay, J., Ounsted, C. and Richards, P. Long-term outcome in children with temporal lobe seizures. V: Indications and contra-indications for neurosurgery. Developmental Medicine and Child Neurology 1984; 26: 25–32. Czochanska, J., Langner-Tyszka, B., Losiowski, Z. and Schmidt-Sidor, B. Children who develop epilepsy in the first year of life: A prospective study. Developmental Medicine and Child Neurology 1994; 36: 344–350. Løyning, Y. Comprehensive epilepsy programs—Norway. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven, 1998: pp. 2869–2875. Wechsler, D. Manual for the Wechsler Intelligence Scale for Children-Revised. New York, The Psychological Corporation, 1974. Norwegian translation and standardisation by J. O. Undheim, 1978. Wechsler, D. Manual for the Wechsler Adult Intelligence Scale. New York, The Psychological Corporation, 1955. Norwegian translation and standardisation by H. Engvik, O. Hjerkinn, S. Seim, 1978. Wechsler, D. Manual for the Wechsler Intelligence Scale for Children-Revised. New York, The Psychological Corporation, 1974. Matarazzo, R. G., Wiens, A. N., Matarazzo, J. D. and Manaugh, T. S. Test-retest reliabilty of the WAIS in a normal population. Journal of Clinical Psychology 1973; 29: 194–197. Matarazzo, J. D. and Herman, D. O. Base rate data for the WAIS-R: Test-retest stability and VIQ-PIQ differences. Journal of Clinical Neuropsychology 1984; 6: 351–366. Chelune, G. J., Naugle, R. I., L¨uders, H., Sedlak, J. and Awad, I. A. Individual change after epilepsy surgery: Practice effects and base-rate information. Neuropsychology 1993; 7: 41–52. Rausch, R. and Crandall, P. H. Psychological status related to surgical control of temporal lobe seizures. Epilepsia 1982; 23: 191–202. Engvik, H. and Tambs, K. WAIS. Tillegg Til H˚andbok. The Institute for Social Sciences, The University of Oslo, 1985. (In Norwegian only.) Matthews, W. S. and Kløve, H. Differential psychological performances in major motor, psychomotor and mixed seizure classifications of known and unknown etiology. Epilepsia 1967; 8: 117–128. Dikmen, S. and Matthews, C. G. Effect of major motor seizure frequency upon cognitive-intellectual functions in adults. Epilepsia 1977; 18: 21–29. Dodrill, C. B. Correlates of generalized tonic-clonic seizures with intellectual, neuropsychological, emotional, and social functions in patients with epilepsy. Epilepsia 1986; 27: 399–411. Seidenberg, M., Beck, N., Geisser, M. et al. Academic achievement of children with epilepsy. Epilepsia 1986; 27: 753–759. Singhi, P. D., Bansal, U., Singhi, S. and Pershad, D. Determinants of IQ profile in children with idiopathic generalized epilepsy. Epilepsia 1992; 33: 1106–1114. Kløve, H. and Fitzhugh, C. B. The relationship of differential EEG patterns to the distribution of Wechsler–Bellevue scores in a chronic epileptic population. Journal of Clinical Psychology 1962; 18: 334–337. Wilkus, R. J. and Dodrill, C. B. Neuropsychological correlates of the electroencephalogram in epileptics: I. Topographic distribution and average rate of epileptiform activity. Epilepsia 1976; 17: 89–100. Kasteleijn-Nolst Trenit´e, D. G., Siebelink, B. M., Berends, S. G., van Strien, J. W. and Meinardi, H. Lateralized effects of subclinical epileptiform EEG discharges on scholastic performance in children. Epilepsia 1990; 31: 740–746. Aldenkamp, A. P. Effect of seizures and epileptiform dis-
32.
33.
34.
35.
36.
37.
38.
39. 40. 41. 42.
43. 44. 45.
46.
47.
48.
49.
50.
51.
52.
53.
charges on cognitive function. Epilepsia 1997; 38 (Suppl. 1): S52–S55. Trimble, M. Antiepileptic drugs, cognitive function, and behavior in children: Evidence from recent studies. Epilepsia 1990; 31 (Suppl. 4): S30–S34. Meador, K. J. Cognitive side effects of antiepileptic drugs. Canadian Journal of Neurological Sciences 1994; 21 (Suppl. 3): S12–S16. Sabers, A., Møller, A., Dam, M. et al. Cognitive function and anticonvulsant therapy: effect of monotherapy in epilepsy. Acta Neurologica Scandinavica 1995; 92: 19–27. Baker, G. A., Smith, D. F., Jacoby, A., Hayes, J. A. and Chadwick, D. W. Liverpool seizure severity scale revisited. Seizure 1998; 7: 201–205. Carpay, J. A., Vermuelen, J., Stroink, H. et al. Seizure severity in children with epilepsy: a parent-completed scale compared with clinical data. Epilepsia 1997; 38: 346–352. O’Donoghue, M. F., Duncan, J. S. and Sander, J. W. The National Hospital Seizure Severity Scale: a further development of the Chalfont Seizure Severity Scale. Epilepsia 1996; 37: 563–571. Hermann, B. P., Wyler, A. R., VanderZwagg, R. et al. Predictors of neuropsychological change following anterior temporal lobectomy: role of regression toward the mean. Journal of Epilepsy 1991; 4: 139–148. Statistica for the Macintosh. Vol. I, Tulsa, OK, USA, StatSoft, Inc., 1994. Spreen, O., Risser, A. I. and Edgell, D. Developmental Neuropsychology. New York, Oxford University Press, 1995. Wechsler, D. Manual for the Wechsler Adult Intelligence Scale. New York, The Psychological Corporation, 1955. Seidenberg, M. and Clemmons, D. C. Maximizing school functioning and the school-to-work transition. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel, Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 2203–2209. Wasterlain, C. G. Recurrent seizures in the developing brain are harmful. Epilepsia 1997; 38: 728–734. Camfield, P. R. Recurrent seizures in the developing brain are not harmful. Epilepsia 1997; 38: 735–737. Farwell, J. R., Dodrill, C. B. and Batzel, L. W. Neuropsychological abilities of children with epilepsy. Epilepsia 1985; 36: 395–400. Oxbury, S. Neuropsychological evaluation—children. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 989–999. Fowler, M. G., Johnson, M. P. and Atkinson, S. S. School achievement and absence in children with chronic health conditions. Pediatrics 1985; 106: 683–687. Boyle, C. A., Decoufle, P. and Yeargin-Allsopp, M. Prevalence and health impact of developmental disabilities in US children. Pediatrics 1994; 93: 399–403. Camfield, P. R., Gates, R., Ronen, G., Camfield, C., Ferguson, A. and MacDonald, G. W. Comparison of cognitive ability, personality profile, and school success in epileptic children with pure right versus left temporal lobe EEG foci. Annals of Neurology 1984; 15: 122–126. Glosser, G., Cole, L. C., French, J. A., Saykin, A. J. and Sperling, M. R. Predictors of intellectual performance in adults with intractable temporal lobe epilepsy. Journal of the International Neuropsychological Society 1997; 3: 252–259. Rodin, E. A., Schmaltz, S. and Twitty, G. Intellectual functioning of patients with childhood-onset epilepsy. Developmental Medicine and Child Neurology 1986; 28: 25–33. Mizrahi, E. M., Kellaway, P., Grossman, R. G. et al. Anterior temporal lobectomy and medically refractory temporal lobe epilepsy of childhood. Epilepsia 1990; 31: 302–312. Fletcher, J. M., Miner, M. E. and Ewing-Cobbs, L. Age and recovery from head injury in children: Developmental issues. In:
A longitudinal study
54.
55.
56.
57.
58.
59.
60.
61.
Neurobehavioral Recovery from Head Injury (Eds H. S. Levin, J. Grafman and H. M. Eisenberg). New York, Oxford University Press, 1987: pp. 279–291. Taylor, H. G. and Alden, J. Age-related differences in outcomes following childhood brain insults:An introduction and overview. Journal of the International Neuropsychological Society 1997; 3: 555–567. Vargha-Khadem, F., Isaacs, E. and Muter, V. A review of cognitive outcome after unilateral lesions sustained during childhood. Journal of Child Neurology 1994; 9 (Suppl. 2): 2S76–2S73. Neyens, L. G. J., Aldenkamp, A. P. and Meinardi, H. M. Prospective follow-up of intellectual development in children with a recent onset of epilepsy. Epilepsy Research 1999; 34: 85–90. Aldenkamp, A. P., Alpherts, W. C. J., De Bruine-Seeder, D. and Dekker, M. J. A. Test-retest variability in children with epilepsy—a comparison of WISC-R profiles. Epilepsy Research 1990; 7: 165–172. Rausch, R., Minh-Thu, L. and Langfitt, J. T. Neuropsychological evaluation-adults. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 977–987. Thorbecke, R. and Fraser, R. T. The range of need and services in vocational rehabilitation. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 2211–2225. Chaplin, J. E., Wester, A. and Tomson, T. Factors associated with the employment problems of people with established epilepsy. Seizure 1998; 7: 299–303. Jalava, M., Sillanp¨aa¨ , M., Camfield, C. and Camfield, P. Social
259
62.
63.
64.
65.
66.
67.
68. 69.
adjustment and competence 35 years after onset of childhood epilepsy: a prospective controlled study. Epilepsia 1997; 38: 708–715. Austin, J. K. and deBoer, H. M. Disruptions in social functioning and services facilitating adjustment for the child and adult. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 2191–2201. Dam, A. M. Hippocampal neuron loss in epilepsy and after experimental seizures. Acta Neurologica Scandinavica 1982; 66: 601–642. Saukkonen, A., K¨alvi¨ainen, R., Partanen, K., Vainio, P., Riekkinen, P. and Pitk¨anen, A. Do seizures cause neuronal damage? A MRI study in newly diagnosed and chronic epilepsy. Neuroreport 1994; 6: 219–223. Pitk¨anen, A. Seizure-induced pathologies: Structural changes assosiated with behavioral consequences. Epilepsia 1996; 37 (Suppl. 4):Abstract 123. K¨alvi¨ainen, R., Salmenper¨a, T., Partanen, K., Vainio, P., Riekkinen, P. and Pitk¨anen, A. Recurrent seizures may cause hippocampal damage in temporal lobe epilepsy. Neurology 1998; 50: 1377–1382. Jokeit, H. and Ebner, A. Long term effects of refractory temporal lobe epilepsy on cognitive abilities: a cross sectional study. Journal of Neurology, Neurosurgery and Psychiatry 1999; 67: 44–50. Engel, J. Jr. Seizures and Epilepsy. Philadelphia, F. A. Davis Company, 1989. Andermann, F. Identification of candidates for surgical treatment of epilepsy. In: Surgical Treatment of the Epilepsies (Ed. J. Engel Jr). New York, Raven Press, 1987: pp. 51–70.
doi:10.1053/seiz.2000.0503, available online at http://www.idealibrary.com on
The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study HELGE BJØRNÆS, KIRSTEN STABELL, OLAF HENRIKSEN & YNGVE LØYNING The National Center for Epilepsy, Sandvika, Norway Correspondence to: Helge Bjørnæs, The National Center for Epilepsy, N-1303, Sandvika, Norway. E-mail: [email protected]
The aim of this study was to compare the long-term consequences of refractory epileptic seizures for intellectual functioning in pediatric and adult patients, taking the severity of the epilepsy into consideration. Thirty-four patients, 17 children (mean age 10.2 years) and 17 adults (mean age 24.4 years) were tested twice with the age-appropriate version of Wechsler’s Intelligence Scales. The mean test–retest interval in the two groups was 3.5 and 6.0 years, respectively. There were no statistically significant differences between the groups with respect to severity of the epilepsy at Test 1, as indicated by retrospective assessments of seizure severity, interictal EEG pathology, and number of antiepileptic drugs received per patient. Assessments of changes in these variables during the test–retest interval did not indicate different courses of the disease in the two groups. Despite these similarities, a statistically significant difference was found between the children and the adults regarding changes in intellectual functioning. In the children, there was a decline in mean intelligence quotient (IQ) scores during the test–retest interval, while the IQ scores increased in the adult group. It is concluded that recurrent seizures may represent a considerable risk for intellectual decline in children, while intellectual functioning seem to be less vulnerable in adults with early onset of epilepsy. c 2001 BEA Trading Ltd
Key words: refractory epilepsy; intelligence; children; adults.
INTRODUCTION The effects chronic epilepsy may have on intellectual functioning are of great concern to many patients and their relatives. They are afraid that the illness may lead to a gradual deterioration. However, longitudinal studies of patients who respond to drug treatment, have failed to demonstrate any intellectual decline. This applies both to children1, 2 and adults3–5 . About 20% of patients with epilepsy, however, do not respond to drug therapy6 . Intractability may be due to underlying pathological processes that are more severe than, or different from, those in patients with a good response7, 8 . Yet, adult patients with intractable seizures may not experience intellectual decline4, 9, 10 , while children with poor seizure control seem to be at risk of intellectual retardation11–13 . Thus, it seems that the long-term effects of intractable seizures in children are more serious than in adults. The evidence presented is not conclusive, however, since descriptions of seizure types and seizure frequencies are not given in comparable units. Hence, it is not possible to conclude 1059–1311/01/040250 + 10
$35.00/0
from these studies whether the children may have had a more severe seizure disorder than the adults. It would be preferable to have some common measures of the severity of the seizure condition, and, in follow-up studies, of the changes in the condition. The present study, therefore, was undertaken to explore and compare the consequences of recurrent seizures for intellectual functioning in children and adults, taking both seizure severity and course of the disorder into consideration.
MATERIALS AND METHODS Patient selection Patients were recruited among 162 consecutive patients included in the epilepsy surgery program at the National Center for Epilepsy, Norway, during the previous 10 years14 . Those who had had an IQ testing during earlier referrals to the center were considered for the study. At the time of the first testing (Test 1), c 2001 BEA Trading Ltd
A longitudinal study
the patients were referred to the epilepsy center for an expert evaluation and treatment of their seizures. They were subsequently treated with antiepileptic drugs for an average of 5 years before they were recognized as surgical candidates. The second testing (Test 2) was administered as part of the pre-operative evaluation. Patients were included in the study if they had complete protocols of the Norwegian versions of the Wechsler Intelligence Scale for Children Revised (WISC-R)15 or the Wechsler Adult Intelligence Scale (WAIS)16 , and were at least 18 years of age at Test 1, or less than 18 years of age at Test 2. Seventeen children (6–17 years of age at Test 2) and 17 adults (18–35 years of age at Test 1) fulfilled the criteria. The selected patients may have had a more severe seizure condition than the majority of our surgical candidates. However, this would pertain to both the pediatric and the adult patients and thus not violate the basis for the comparisons between the two groups. One child and one adult patient were considered for corpus callosotomy, while resective surgery was planned in the remainder. In addition to these 34 patients, complete test protocols were available in 15 patients who were pediatric patients at Test 1 but adults at Test 2. Because it was not possible to decide whether changes in IQ scores in these patients had taken place during childhood or later, their data were not included in the statistical comparisons between children and adults, but were evaluated separately to give some additional information on long-term consequences of recurrent seizures. Demographic data from this additional group are presented in the last column of Table 1, under the heading ‘Children retested as adults’.
Methodological issues pertaining to the intelligence tests The psychometric properties of the Norwegian standardization of WISC-R and WAIS, as described in the handbooks, fulfil standard criteria15, 16 . Practice effects
Upon retesting with intelligence tests under stable conditions, with test–retest intervals between 2 and 20 weeks, the results obtained in normal persons will usually improve by an average of 5–7 points in Full Scale IQ scores. Such practice effects are observed to the same extent in normal children and normal adults17–19 . However, studies of patients with uncontrolled seizures fail to find appreciable test– retest improvements3, 4, 20, 21 . Consequently, the practice effects might be negligible in the present study of patients with frequent seizures, who were retested
251
after relatively long intervals. Besides, the test–retest intervals did not differ between the groups. Practice effects, therefore, were not expected to cause any systematic differences between the groups. IQ scores
Verbal, Performance and Full Scale IQ (FSIQ) scores were computed for each patient using Norwegian norm tables15, 16, 22 . IQ scores are age adjusted in both children and adults. Difference scores (Test 2–Test 1) were computed based on the individual Verbal, Performance and FSIQ scores.
Seizure-related factors that may influence the test performance Test performance may be influenced by factors such as seizure activity3, 23–27 , pathological brain activity as seen on the EEG28–31 , and side effects of the antiepileptic drugs (AEDs)32–34 . Systematic differences between the groups with respect to these factors could contribute to differences in the test results. Consequently, attempts were made to estimate these factors retrospectively in the medical files of each patient as described in the following. Seizure severity
Number of seizures during the previous month prior to the two test occasions were recorded, based either on the patients’ own seizure diaries or on the daily records kept by the staff. Many patients had several seizure types, and in some patients the seizure types changed over time. Therefore, a mere count of seizures would not give an adequate evaluation of the seizure severity. The need for expressing the seizure severity rather than the seizure frequency is underlined by several authors, and appropriate scales are available35–37 . However, these scales cannot be used retrospectively. Thus, in the present study, clinically based assessments of seizure severity were performed. Three experienced epileptologists rated the records of each patient independently and blindly, taking both seizure frequency and seizure types into consideration. The seizure condition before Test 1 was rated along a graded 5-point scale ranging from 0 ‘no seizures’ to 4 ‘severe seizure condition’. The ratings were highly reliable (Kendall coefficient of concordance = 0.90, P< 0.001). Individual changes in the seizure condition between test sessions were rated along a graded 7-point scale ranging from −3 ‘much worse’ to +3 ‘much better’ (Kendall coefficient of concordance = 0.94, P< 0.001). Examples of ratings are given in Table 2.
252
H. Bjørnæs et al.
Table 1: Demographics and seizure-related variables. Variable 1. 2. 3.
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
16. 17.
18.
19.
Number of patients Sex (F/M) Etiology, number of patients Idiopathic or crypt./symptomatic Cerebrovascular lesion Intracranial infection Perinatal encephalopathy Asphyxia Post-traumatic Tumour Age at seizure onset, mean yr (SD) Age at Test 1, mean (SD) Seizure severity at Test 1, median (q-range) EEG pathology prior to Test 1, median (q-range) Number of AEDs received per patient at Test 1, median (q-range) Duration of epilepsy at Test 1, mean yr (SD) Test–retest interval, mean yr (SD) Age at Test 2, mean (SD) Change in seizure severity, median (q-range) Change in EEG pathology, median (q-range) Change in medication, median (q-range) Laterality of epileptogenic region, number of patients Left hemisphere/Right hemisphere Generalized Temporal/extratemporal location of epileptogenic region, number of patients Type of surgery planned, number of patients Resection Callosotomy Seizure types previous month prior to Test 1, number of patients Atypical absences Simple partial seizures Complex partial seizures Generalized tonic–clonic seizures (sec. gen) Misc. (Tonic, atonic, myoclonic, undetermined, etc.) Seizure types previous month prior to Test 2, number of patients Atypical absences Simple partial seizures Complex partial seizures Generalized tonic–clonic seizures (sec. gen) Misc. (Tonic, atonic, myoclonic, undetermined etc.)
Study groups Children Adults
P-value
Children retested as adults
NSa
15 10/5
17 8/9
17 7/10
14/3 1 1 0 1 0 0 5.0 10.2 3.0 2.0
(3.0) (2.6) (2) (1)
13/4 0 2 1 1 0 0 9.5 24.4 2.3 2.0
(7.2) (4.6) (1) (1)
<0.05b
1.0
(1)
2.0
5.2 3.5 13.7 1.0 −1.0 0.0
(2.9) (2.5) (2.6) (6) (2) (1)
14.8 6.0 30.4 1.0 0.0 0.0
8/8 1 11/5
8/7 1 15/1
16 1
16 1
0 4 12 4 (3)
0 2 13 11 (10)
NSa NSa NSa <0.05a
2 3 12 6
2
2
NSa
0
1 3 14 5 (4)
0 3 14 11 (10)
NSa NSa NSa <0.05a
0 4 14 9
2
2
NSa
1
NSd
NSc NSc
9/6 0 1 1 0 2 2 5.3 13.6 1.3 2.0
(3.5) (2.6) (1) (2)
(1)
NSc
1.0
(2)
(9.0) (4.8) (6.0) (5) (1) (1)
<0.001b NSb NSc NSc NSc NSa NSd
8.3 11.0 24.6 −4.0 −1.0 0.0
(4.9) (5.6) (5.1) (6) (1) (1)
6/9 0 10/5
15 0
Abbreviations: F, female; M, male; Crypt, cryptogenic; q-range, quartile range; Yr, years; SD, standard deviation; AEDs, antiepileptic drugs; Sec. gen, secondarily generalized; Misc, miscellaneous. a Chi-squared; b Student’s t-test for independent samples; c Median test; d Fisher exact probability test.
Interictal EEG pathology
For each patient, the interictal EEG records closest in time to the test sessions were selected from the medical files. In most patients, EEG records were taken a few days or weeks ahead of the test sessions. In a few patients, the records were done a couple of months
earlier. However, since the test occasions were separated by several years in most patients, we assumed that these samples of EEGs were, to some degree, representative of the patients’ interictal EEG characteristics at the time of the assessments. Summaries of the EEG records were blindly evaluated by a senior neurophysiologist with respect to degree (not lo-
A longitudinal study
calization) of EEG pathology prior to Test 1, and change in the degree of EEG pathology between Test 1 and Test 2. Both evaluations were made according to graded 5-point scales, and comprised background activity, focal slowing, persistent/paroxysmal activity, sharp components, and spike and wave activity. Number of AEDs
The use of AED monotherapy with drug serum levels within the therapeutic range may have negligible side effects on cognition, while combinations of drugs are more likely to cause adverse effects33, 34 . Thus, the number of AEDs received per patient at the first testing, and changes in these numbers from Test 1 to Test 2, were recorded. Two children and one adult were without medication at Test 1, none were without medication at Test 2.
253
tial testing, with a mean duration of 5.2 years in the children and 14.8 years in the adults. More adults than children had secondarily generalized tonic–clonic seizures during the previous month prior to both test occasions. There were, however, no significant differences between the groups with respect to sex, etiological syndromes or epilepsy syndromes. All patients but one in each group had localization-related partial epilepsy. One child had primarily generalized tonic– clonic, atonic and atypical absence seizures of cryptogenic etiology, while one adult had Lennox–Gastaut syndrome of cryptogenic etiology and with tonic– clonic and atonic seizures. Furthermore, there were no between-group differences in laterality or location of the epileptogenic region, nor in seizure severity, EEG pathology and number of drugs per patient at Test 1. Changes from Test 1 to Test 2 in the latter three variables did not differ between the groups (Table 1).
Data analysis
IQ scores
Categorical data in the two study groups were analysed using the chi-squared test wherever the requirements for this test were met. In 2 × 2 contingency tables with few expected frequencies, the Fisher exact probability test was performed. Ordinal-scale data were compared using the median test. Comparisons of continuous demographic variables were performed by Student’s t-test for independent samples. The IQ scores were submitted to analysis of variance. Post hoc tests were performed using Duncan’s multiple range test. To decide whether the changes from Test 1 to Test 2 were influenced by regression toward the mean, product-moment correlations were computed between the IQ scores at Test 1 and the difference scores. Regression toward the mean is significant if the initial values are significantly inversely correlated with the difference scores38 . P-values less than 0.05 in twotailed tests were considered statistically significant. All statistical analyses were run using ‘Statistica for the Macintosh’39 .
Mean Verbal, Performance, and FSIQ scores of the two groups at the two test occasions are shown in Table 3. Initially the children’s FSIQ scores were in the lower end of the normal range (i.e. 85–115). At Test 2, however, their scores had declined on average by 4.6, 6.1, and 5.9 points in Verbal, Performance, and FSIQ, respectively. Mean FSIQ now fell below the normal range. The adults had IQ scores well below the normal range at Test 1, but in contrast to the changes in the children, their corresponding scores increased by 4.2, 7.9, and 6.4 points. To assess whether these differences between the two groups were significant, we subjected the IQ scores to 2 (group) × 2 (test occasion) analyses of variance (ANOVA). There were no significant differences between the groups with respect to mean IQ scores, both test occasions taken together, nor were there any significant test–retest changes, both groups taken together, as shown by the insignificant main effects. However, significant interaction effects were revealed for Group × Test occasion in all the three types of IQ scores. This indicated that the decrease in IQ scores seen in the pediatric patients and the increase in the adults, represented significantly different courses. Post hoc analyses showed that the changes in Performance and FSIQ were statistically significant within each group. Post hoc between-group comparisons showed significantly higher Verbal, Performance, and FSIQ scores in the children at Test 1, but no significant differences at Test 2. Because the groups differed with respect to age at onset of habitual seizures, we performed similar analyses as shown in Table 2 with age at onset as a co-
RESULTS Demographic and seizure-related variables
Demographic and seizure-related variables of the two study groups are presented in Table 1. All but two patients in the present study had seizure onset in childhood, the children being on average 5 years old at seizure debut, and the adults on average 9.5 years old. This difference was statistically significant. There were also significant differences between the groups with respect to the duration of epilepsy at the ini-
254
H. Bjørnæs et al.
Table 2: Examples of seizure severity ratings. Patients Sz/month prior to Test 1 Abs SPS CPS GTC Misc HH EA HB IT
150 25 1
9
2 1
135
Ratings of sz-severity Rater A B C 4 3 3 1
4 2 3 2
4 2 3 1
Sz/month prior to Test 2 score Abs SPS CPS GTC Misc 12 7 9 4
100 500
25 15
16 20 6
30
Ratings of change Rater A B C 3 −3 −1 −2
3 −3 0 −1
3 −3 0 −3
score 9 −9 −1 −6
Abbreviations: Sz, seizures; Abs, atypical absences; SPS, simple partial seizures; CPS, complex partial seizures; GTC, generalized tonic–clonic seizures; Misc., miscellaneous seizure types: tonic, atonic, myoclonic, undetermined. Frequency of different seizure types prior to Test 1 and Test 2 in four patients, and actual ratings by three epileptologists (rater A, B, and C) of seizure severity at Test 1 and change in seizure severity from Test 1 to Test 2. The ratings of seizure severity were done using a graded 5-point scale ranging from 0, no seizures, to 4, severe seizure condition. Ratings of change were done according to a graded 7-point scale ranging from −3, much worse, to 3, much better. The individual patient’s score was the sum of the ratings. Thus, individual seizure severity ranged from 0 to 12, and individual change in seizure severity from −9 to 9.
variate. The interaction effects were still significant in Performance and FSIQ (F = 10.3, P< 0.01; F = 8.2, P< 0.01, respectively), but not in Verbal IQ (F = 3.8, P> 0.06). No significant main effects were obtained. Although the groups differed also in duration of epilepsy at Test 1, this variable is dependent on the age at onset of seizures and the age at Test 1, which was part of the grouping criterion. Consequently, we did not find it appropriate to use this variable as an additional covariate. The 15 children who were retested as adults (additional group) were on average 5.3 years of age at onset of seizures (Table 1). They were on average 13.6 years of age at Test 1 and 24.6 at Test 2. Mean Verbal, Performance, and FSIQ scores at Test 1 were 83.8, 88.9, and 84.7, respectively. These mean scores were not significantly (NS) different from the corresponding scores of the pediatric group at Test 1 (t-values were 0.5, NS; 0.5, NS; and 0.5, NS, respectively, t-test for independent samples). During the test–retest interval, which was on average 11 years, the additional group declined in Verbal, Performance and FSIQ scores by 6.4, 7.7, and 7.7 points, respectively. The changes were statistically significant (t-values were 3.2, P< 0.01; 2.4, P< 0.05; 3.2, P< 0.01, t-test for correlated samples). The decline was not significantly different from the decline in the pediatric group (t-values were 0.5, NS; 0.4, NS; and 0.5, NS; t-test for independent samples).
DISCUSSION In this retrospective study comparing changes in IQ scores at two test occasions in pediatric and adult patients with refractory seizures, we found significantly different courses in intellectual functioning between the groups: The children’s scores declined, while the adults’ increased. Post hoc analyses showed that the within-group changes were significant in Per-
formance and FSIQ in both groups. These findings are in agreement with previous longitudinal studies in adults9, 10 , and in children11–13 . The differences between children and adults in the current study could not be ascribed to differences in etiology or in epilepsy syndromes. Furthermore, the results could probably not be explained by a more severe epilepsy initially in the children, or a more progressive course of their disease, as indicated by assessments of seizure severity and other seizure-related variables (Table 1). A bias might be introduced, however, if the patients selected for the study initially were referred for neuropsychological assessment in a period when they temporarily had an increased seizure frequency. This could give a test–retest gain in the adults and attenuate the decline in the children. However, the changes in seizure severity gave no indications that this was the case (Z = 0.9, P> 0.3, sign test). Regression toward the mean did not significantly influence the results in either group. We assume that the findings reflect a greater vulnerability of intellectual functioning in children than in adults, due to different stages in the development of intellectual abilities. This development may be conceived as a cumulative process undergoing particularly rapid growth in childhood40 . Roughly, this growth can be operationally defined by the increase in the sum of WISC-R raw scores required from one year to the next for a child to maintain mean scaled scores, i.e. scaled scores of 10, and an average FSIQ15 , as illustrated in Fig. 1. Rate of growth in intellectual abilities, as defined here, decelerates and levels off in young adulthood. This can be illustrated in the same way by adding the raw scores necessary to maintain mean scaled scores on the WAIS for successive age groups41 (see Fig. 1). Normal development, as reflected by the curves in Fig. 1, depends on many factors. Among these, maturational processes of the brain and optimal opportunities for the child to learn from interaction with a normal environment, are among the most essen-
A longitudinal study
255
400 Normal development
Sum of raw scores
WISC-R
WAIS
300
Development of case GA
200
IQ=57
IQ=53
IQ=64 IQ=75
100
0 6
8
10
12
14
16
18
20
22
24
26
28
30
Age Fig. 1: Illustration of normal intellectual development (longer curve) and the slowing of development with gradually diminishing FSIQ of case GA (shorter curve). In the WISC-R norm tables, which are based on the standardization samples of children at different ages, the total number of correct solutions required to obtain a FSIQ of 100 increases by a certain amount from one age level to the next. This amount operationally defines the normal intellectual development of children with average intelligence. By adding the numbers of correct solutions (actually the raw scores) for each age level one can observe the normal growth in intellectual abilities at successive ages (longer curve). Similar sums of raw scores from the WAIS show the intellectual development from age 16 to age 30. (To join the curves from the WISC-R and the WAIS into one smooth curve a parallel displacement had to be done by adding a constant number to the WAIS values.) The actual number of correct solutions (raw scores) given by case GA, who was tested four times with the WISC-R, are shown by the shorter curve, and the FSIQ scores she obtained are shown beneath this curve. She was 9 years and 9 months of age at the first assessment, and was subsequently tested at the age of 11 years and 6 months, 13 years and 7 months, and 15 years and 2 months, respectively. GA had cryptogenic epilepsy with seizure onset at 1 year of age, and underwent right temporal lobe resection when she was 13 years of age. She has been completely seizure free since then, i.e. for more than 15 years. Abbreviations: WISC-R, Wechsler Intelligence Scale for Children-Revised; WAIS, Wechsler Adult Intelligence Scale; FSIQ: Full Scale Intelligence Quotient.
tial40, 42 . Whether epilepsy interferes directly with the maturational processes of the brain is a matter of considerable debate43, 44 . However, it is generally recognized that an uncontrolled epilepsy may interfere with the child’s cognitive functioning1, 12, 13, 45, 46 and with its opportunity to interact with the environment47, 48 . This leads to disadvantages in scholastic as well as social participation42, 49–52 . Thus, the epileptic child’s opportunities to learn from normal experience are disturbed47, 48 . This may interfere with the speed of intellectual growth, and the child’s abilities may develop at a slower rate and gradually lag behind those of its agemates. Because IQ scores are age adjusted, such slowing is reflected by a diminishing IQ. This is illustrated by the results after serial testing with the WISC-R of the case GA in Fig. 1, showing a gradual decrease in FSIQ scores despite a small gradual increase in raw scores (see also the legend). Similar explanations for the long-term consequences for intellectual development in children with early brain injuries have been proposed by Fletcher
et al.53 and Taylor and Alden54 , and in patients with epilepsy by Vargha-Khadem et al.55 and Glosser et al.50 The latter authors related the delayed cognitive development in their patients to the possible interference of seizures with neural maturation during critical periods in childhood. The present authors also assume that disturbances of cognitive functioning and, in some cases, a restricted life caused by frequent seizures, may play a role in retarding intellectual development. Neyens et al.56 argued that cognitive decline in children with epilepsy seem to occur in close temporal proximity to seizure onset, and may be caused by the underlying pathophysiological processes. Our data are not in unequivocal support of this hypothesis, as about 20% of the children who had had seizures for at least 5 years at Test 1 declined by 10 or more IQ points during the test–retest interval. The children had higher IQs than the adults at Test 1. This might be a referral artefact, e.g. if the threshold for being referred for neuropsychological assessment was systematically different with respect to in-
256
H. Bjørnæs et al.
Table 3: Mean IQ scores and standard deviations in pediatric and adult patients. Test 1
Test 2
Main effects group
VIQ PIQ FSIQ
Children Adult Children Adult Children Adult
86.8 (17.3) 80.1 (19.2) 92.5 (15.8) 73.5 (20.7) 88.1 (16.6) 75.1 (20.5)
82.2 (19.4) 84.2 (22.3) 86.4 (15.8) 81.5 (20.7) 82.1 (18.6) 81.5 (22.3)
Interactions test
F
P
F
P
F
P
0.1
NS
0.0
NS
5.3
<0.05a
3.8
NS
0.3
NS
13.2
<0.001a
1.1
NS
0.0
NS
10.9
<0.01a
Abbreviations: VIQ, Verbal IQ; PIQ, Performance IQ; FSIQ, Full Scale IQ; NS, not significant. a Two-way analysis of variance (ANOVA). Post hoc tests: Within-group changes Test1–Test 2: Children: VIQ, NS; PIQ, P< 0.05; FSIQ, P< 0.05. Adults: VIQ, NS; PIQ, P< 0.01; FSIQ, P< 0.05. Between-group differences: Test 1: VIQ, P< 0.05; PIQ, P< 0.001; FSIQ, P< 0.001. Test 2: VIQ, NS; PIQ, NS; FSIQ, NS.
tellectual functioning in children and adults. The differences are, however, more likely due to the relatively short seizure history at Test 1 in the children, namely 5.2 years on average. Children with a shorter seizure history are reported to have higher IQs than children with a longer duration of epilepsy45 . The course of the intellectual functioning in the pediatric patients in the present study corroborates these findings. At Test 2, the children had had seizures for an average of 8.7 years, and their mean FSIQ had declined by 5.8 points and was no longer significantly different from the FSIQ of the adult group. This is in agreement with the ‘retarded growth’ hypothesis. In accordance with a developmental model, we would also expect an intellectual decline in the 15 children who were retested as adults. However, although the decline was statistically significant, the interpretation of the results remains uncertain, because of a significant regression toward the mean. Intellectual decline in children with epilepsy seems to depend on the severity of the epilepsy1, 2 . In a 4-year follow-up of children with epilepsy, Aldenkamp et al.57 could find only insignificant drops in mean IQ scores. The differences with the current findings may be due to differences in the severity of the illness, as 22% became seizure free in the study by Aldenkamp et al.57 , while this was not the case in any of our patients. One may wonder why the adults in the present study, who had recurrent seizures of similar severity as the children, did not decline in IQ scores. It is well known that seizures also disturb cognitive functioning in adults3, 24, 25, 58 , and interfere with their opportunities to participate in a normal way in intellectual activities within education42 , work59, 60 , and social settings61, 62 . In addition, the findings of atrophic and sclerotic brain areas in adults with epilepsy raise the question of whether long-standing seizures may lead to progressive brain lesions63, 64 with concomitant hazards for cognitive functioning65 . In spite of these potential hazards, however, we did
not find appreciable intellectual decline in any of the 17 adult patients, even after follow-up periods of more than 15 years. Only three of them had minor decreases, by 1–4 points, in FSIQ scores. In contrast, the adult patients, as a group, experienced gains in Performance and FSIQ. If practice effects may be disregarded, these gains may be due to the introduction of AEDs with less side effects. All but two adult patients had different AEDs at Test 2 compared to Test 1. (Of the children, 13 had changes in types of drugs received, which may have attenuated the decline to some degree.) On the other hand, the adult patients had lower initial IQ scores than the children, and consequently less room for further decline, as measured by the Wechsler scales. They had a mean seizure history of 15 years at Test 1, and, as all but two had seizure onset before the age of 16, they might already have gone through a retardation in their intellectual development. In addition, more adults than children had generalized tonic–clonic seizures prior to both test occasions, which may have contributed to their lower level of performance. A ‘floor’-effect, therefore, may have limited the potential decline in some of the adults. However, previous longitudinal studies in adult patients with normal or low normal intelligence could not find significant adverse changes in IQ scores4, 9, 10 , in agreement with the present findings. Two of these studies9, 10 found small, but significant, gains in IQ scores which may also have been effects of optimalization of drug treatment. The relative stability of functions in adult patients with recurrent seizures is probably not limited to intelligence. Selva et al.9 found no significant decline in measures of memory in a 2.4-year follow-up study. Kalska4 reported no adverse changes in memory, cognitive flexibility and motor function during a 10-year test–retest interval. In the 10-year follow-up study by Holmes et al.10 , a slight but significant decline was seen in six out of 16 neuropsychological tests, mostly in measures of speed and visuo-spatial relations. How-
A longitudinal study
ever, the percentage of results outside normal limits was not changed after 10 years. The investigators concluded that although no overall adverse changes were seen in intellectual or neuropsychological functioning, slight losses in selected areas could not be ruled out. None of the above studies4, 9, 10 considered patients with onset of epilepsy in adulthood. The long-term cognitive effects of seizures in this group remain to be decided. Mesial temporal sclerosis and slightly impaired memory seem to be present in adults with newly diagnosed temporal lobe seizures, indicating that the lesions may have preceded the seizures65 . Recurrent seizures may exacerbate the lesion66 . The progressive nature of epilepsy is, however, disputed43, 44, 65 . Holmes et al.10 regarded their findings as evidence against a progressive process underlying drugresistant epilepsy in adults. On the other hand, a recent cross-sectional study by Jokeit and Ebner67 indicated that epileptic seizures may have a cumulative adverse effect only discernible in adult intelligence after more than 30 years with seizures. An average decline in FSIQ of 5.9 points in children in the current study may seem trifling. But 40% of them declined by 10 or more points, while only one child experienced a comparable gain. In the additional group of children retested as adults, 47% declined by 10 or more points. We find this noteworthy. Although the patients selected for this study may not have been representative for the population of our surgical candidates in all respects, with the additional group they constituted a substantial part (30%), and represent those with an early onset of refractory seizures, which may give the most serious long-term consequences for intellectual functioning. This is a group of surgical candidates most epilepsy centres will encounter, unless they reject patients with low FSIQ from their surgical program, as advocated by some authors68, 69 .
CONCLUSIONS The present study corroborates the findings from previous studies4, 9–13 : adult patients with an early onset epilepsy may not experience a decline in IQ scores, despite recurrent seizures over several years. The longterm consequences of recurrent seizures in children and adolescents may be more serious, as they seem to be at risk for intellectual decline relative to ageadjusted norms. According to our results, the differences are probably not due to a more severe or progressive seizure disorder in the pediatric patients. As an explanation for the different trends seen in children compared to adults, we proposed a developmental model, suggesting that intellectual abilities may be particularly vulnerable to disturbances related
257
to recurrent seizures during the period of rapid development in childhood. Consequently, drug treatment that does not result in satisfactory seizure control in children should not be unduly prolonged before surgical treatment is considered. Only further investigations can decide whether the present finding apply to patients with seizure onset in adulthood.
ACKNOWLEDGEMENTS This study was supported by a grant from the Research Fund of The National Center for Epilepsy. The authors are very grateful to Drs Ebba Wannag, Einar Kinge, and Renate Dahl for assessing the seizure severity in the patients, and to Drs Rasmus Lossius, Karl Otto Nakken and Robert Kloster for providing information on the patients’ AED records. We also thank Drs Svein I. Johannessen and Karl Otto Nakken for valuable advice during the preparation of the manuscript, and Professor Kenneth Hugdahl, Professor Kjetil Sundet, Dr Petter Mowinckel and Dr Lien My Diep for help and advice in connection with the statistical analyses.
REFERENCES 1. Bourgeois, B. F. D., Prensky, A. L., Palkes, H. S., Talent, B. K. and Busch, S. G. Intelligence in epilepsy: A prospective study in children. Annals of Neurology 1983; 14: 438–444. 2. Ellenberg, J. H., Hirtz, D. G. and Nelson, K. B. Do seizures in children cause intellectual deterioration? New England Journal of Medicine 1986; 314: 1085–1088. 3. Seidenberg, M., O’Leary, D. S., Berent, S. and Boll, T. Changes in seizure frequency and test-retest scores on the Wechsler Adult Intelligence Scale. Epilepsia 1981; 22: 75–83. 4. Kalska, H. Cognitive changes in epilepsy. A ten-year follow-up. Commentationes Scientiarum Socialium 1991; 44: 1–85. Helsinki, The Finnish Society of Sciences and Letters. 5. Dodrill, C. B. and Wilensky, A. J. Neuropsychological abilities before and after 5 years of stable antiepileptic drug therapy. Epilepsia 1992; 33: 327–334. 6. Hopkins, A. and Shorvon, S. Definitions and epidemiology of epilepsy. In: Epilepsy 2nd Edition, (Eds A. Hopkins, S. Shorvon and G. Cascino). London, Chapman and Hall Medical, 1995: pp. 1–24. 7. Lesser, R. P., L¨uders, H., Wyllie, E., Dinner, D. S. and Morris III, H. H. Mental deterioration in epilepsy. Epilepsia 1986; 27 (Suppl. 2): S105–S123. 8. Engel, J. Jr and Shewmon, D. A. Overview: Who should be considered a surgical candidate? In: Surgical Treatment of the Epilepsies (Ed. J. Engel Jr). New York, Raven Press, 1993: pp. 23–34. 9. Selwa, L. M., Berent, S., Giordano, B., Henry, T. R., Buchtel, H. A. and Ross, D. A. Serial cognitive testing in temporal lobe epilepsy: Longitudinal changes with medical and surgical therapies. Epilepsia 1994; 35: 743–749. 10. Holmes, M. D., Dodrill, C. B., Wilkus, R. J., Ojemann, L. M. and Ojemann, G. A. Is partial epilepsy progressive? Ten-year follow-up of EEG and neuropsychological changes in adults with partial seizures. Epilepsia 1998; 39: 1189–1193. 11. Dodrill, C. B. and Troupin, A. S. Seizures and adaptive
258
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
H. Bjørnæs et al. abilities—a case study of identical twins. Archives of Neurology 1976; 33: 604–607. Lindsay, J., Ounsted, C. and Richards, P. Long-term outcome in children with temporal lobe seizures. V: Indications and contra-indications for neurosurgery. Developmental Medicine and Child Neurology 1984; 26: 25–32. Czochanska, J., Langner-Tyszka, B., Losiowski, Z. and Schmidt-Sidor, B. Children who develop epilepsy in the first year of life: A prospective study. Developmental Medicine and Child Neurology 1994; 36: 344–350. Løyning, Y. Comprehensive epilepsy programs—Norway. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven, 1998: pp. 2869–2875. Wechsler, D. Manual for the Wechsler Intelligence Scale for Children-Revised. New York, The Psychological Corporation, 1974. Norwegian translation and standardisation by J. O. Undheim, 1978. Wechsler, D. Manual for the Wechsler Adult Intelligence Scale. New York, The Psychological Corporation, 1955. Norwegian translation and standardisation by H. Engvik, O. Hjerkinn, S. Seim, 1978. Wechsler, D. Manual for the Wechsler Intelligence Scale for Children-Revised. New York, The Psychological Corporation, 1974. Matarazzo, R. G., Wiens, A. N., Matarazzo, J. D. and Manaugh, T. S. Test-retest reliabilty of the WAIS in a normal population. Journal of Clinical Psychology 1973; 29: 194–197. Matarazzo, J. D. and Herman, D. O. Base rate data for the WAIS-R: Test-retest stability and VIQ-PIQ differences. Journal of Clinical Neuropsychology 1984; 6: 351–366. Chelune, G. J., Naugle, R. I., L¨uders, H., Sedlak, J. and Awad, I. A. Individual change after epilepsy surgery: Practice effects and base-rate information. Neuropsychology 1993; 7: 41–52. Rausch, R. and Crandall, P. H. Psychological status related to surgical control of temporal lobe seizures. Epilepsia 1982; 23: 191–202. Engvik, H. and Tambs, K. WAIS. Tillegg Til H˚andbok. The Institute for Social Sciences, The University of Oslo, 1985. (In Norwegian only.) Matthews, W. S. and Kløve, H. Differential psychological performances in major motor, psychomotor and mixed seizure classifications of known and unknown etiology. Epilepsia 1967; 8: 117–128. Dikmen, S. and Matthews, C. G. Effect of major motor seizure frequency upon cognitive-intellectual functions in adults. Epilepsia 1977; 18: 21–29. Dodrill, C. B. Correlates of generalized tonic-clonic seizures with intellectual, neuropsychological, emotional, and social functions in patients with epilepsy. Epilepsia 1986; 27: 399–411. Seidenberg, M., Beck, N., Geisser, M. et al. Academic achievement of children with epilepsy. Epilepsia 1986; 27: 753–759. Singhi, P. D., Bansal, U., Singhi, S. and Pershad, D. Determinants of IQ profile in children with idiopathic generalized epilepsy. Epilepsia 1992; 33: 1106–1114. Kløve, H. and Fitzhugh, C. B. The relationship of differential EEG patterns to the distribution of Wechsler–Bellevue scores in a chronic epileptic population. Journal of Clinical Psychology 1962; 18: 334–337. Wilkus, R. J. and Dodrill, C. B. Neuropsychological correlates of the electroencephalogram in epileptics: I. Topographic distribution and average rate of epileptiform activity. Epilepsia 1976; 17: 89–100. Kasteleijn-Nolst Trenit´e, D. G., Siebelink, B. M., Berends, S. G., van Strien, J. W. and Meinardi, H. Lateralized effects of subclinical epileptiform EEG discharges on scholastic performance in children. Epilepsia 1990; 31: 740–746. Aldenkamp, A. P. Effect of seizures and epileptiform dis-
32.
33.
34.
35.
36.
37.
38.
39. 40. 41. 42.
43. 44. 45.
46.
47.
48.
49.
50.
51.
52.
53.
charges on cognitive function. Epilepsia 1997; 38 (Suppl. 1): S52–S55. Trimble, M. Antiepileptic drugs, cognitive function, and behavior in children: Evidence from recent studies. Epilepsia 1990; 31 (Suppl. 4): S30–S34. Meador, K. J. Cognitive side effects of antiepileptic drugs. Canadian Journal of Neurological Sciences 1994; 21 (Suppl. 3): S12–S16. Sabers, A., Møller, A., Dam, M. et al. Cognitive function and anticonvulsant therapy: effect of monotherapy in epilepsy. Acta Neurologica Scandinavica 1995; 92: 19–27. Baker, G. A., Smith, D. F., Jacoby, A., Hayes, J. A. and Chadwick, D. W. Liverpool seizure severity scale revisited. Seizure 1998; 7: 201–205. Carpay, J. A., Vermuelen, J., Stroink, H. et al. Seizure severity in children with epilepsy: a parent-completed scale compared with clinical data. Epilepsia 1997; 38: 346–352. O’Donoghue, M. F., Duncan, J. S. and Sander, J. W. The National Hospital Seizure Severity Scale: a further development of the Chalfont Seizure Severity Scale. Epilepsia 1996; 37: 563–571. Hermann, B. P., Wyler, A. R., VanderZwagg, R. et al. Predictors of neuropsychological change following anterior temporal lobectomy: role of regression toward the mean. Journal of Epilepsy 1991; 4: 139–148. Statistica for the Macintosh. Vol. I, Tulsa, OK, USA, StatSoft, Inc., 1994. Spreen, O., Risser, A. I. and Edgell, D. Developmental Neuropsychology. New York, Oxford University Press, 1995. Wechsler, D. Manual for the Wechsler Adult Intelligence Scale. New York, The Psychological Corporation, 1955. Seidenberg, M. and Clemmons, D. C. Maximizing school functioning and the school-to-work transition. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel, Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 2203–2209. Wasterlain, C. G. Recurrent seizures in the developing brain are harmful. Epilepsia 1997; 38: 728–734. Camfield, P. R. Recurrent seizures in the developing brain are not harmful. Epilepsia 1997; 38: 735–737. Farwell, J. R., Dodrill, C. B. and Batzel, L. W. Neuropsychological abilities of children with epilepsy. Epilepsia 1985; 36: 395–400. Oxbury, S. Neuropsychological evaluation—children. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 989–999. Fowler, M. G., Johnson, M. P. and Atkinson, S. S. School achievement and absence in children with chronic health conditions. Pediatrics 1985; 106: 683–687. Boyle, C. A., Decoufle, P. and Yeargin-Allsopp, M. Prevalence and health impact of developmental disabilities in US children. Pediatrics 1994; 93: 399–403. Camfield, P. R., Gates, R., Ronen, G., Camfield, C., Ferguson, A. and MacDonald, G. W. Comparison of cognitive ability, personality profile, and school success in epileptic children with pure right versus left temporal lobe EEG foci. Annals of Neurology 1984; 15: 122–126. Glosser, G., Cole, L. C., French, J. A., Saykin, A. J. and Sperling, M. R. Predictors of intellectual performance in adults with intractable temporal lobe epilepsy. Journal of the International Neuropsychological Society 1997; 3: 252–259. Rodin, E. A., Schmaltz, S. and Twitty, G. Intellectual functioning of patients with childhood-onset epilepsy. Developmental Medicine and Child Neurology 1986; 28: 25–33. Mizrahi, E. M., Kellaway, P., Grossman, R. G. et al. Anterior temporal lobectomy and medically refractory temporal lobe epilepsy of childhood. Epilepsia 1990; 31: 302–312. Fletcher, J. M., Miner, M. E. and Ewing-Cobbs, L. Age and recovery from head injury in children: Developmental issues. In:
A longitudinal study
54.
55.
56.
57.
58.
59.
60.
61.
Neurobehavioral Recovery from Head Injury (Eds H. S. Levin, J. Grafman and H. M. Eisenberg). New York, Oxford University Press, 1987: pp. 279–291. Taylor, H. G. and Alden, J. Age-related differences in outcomes following childhood brain insults:An introduction and overview. Journal of the International Neuropsychological Society 1997; 3: 555–567. Vargha-Khadem, F., Isaacs, E. and Muter, V. A review of cognitive outcome after unilateral lesions sustained during childhood. Journal of Child Neurology 1994; 9 (Suppl. 2): 2S76–2S73. Neyens, L. G. J., Aldenkamp, A. P. and Meinardi, H. M. Prospective follow-up of intellectual development in children with a recent onset of epilepsy. Epilepsy Research 1999; 34: 85–90. Aldenkamp, A. P., Alpherts, W. C. J., De Bruine-Seeder, D. and Dekker, M. J. A. Test-retest variability in children with epilepsy—a comparison of WISC-R profiles. Epilepsy Research 1990; 7: 165–172. Rausch, R., Minh-Thu, L. and Langfitt, J. T. Neuropsychological evaluation-adults. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 977–987. Thorbecke, R. and Fraser, R. T. The range of need and services in vocational rehabilitation. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 2211–2225. Chaplin, J. E., Wester, A. and Tomson, T. Factors associated with the employment problems of people with established epilepsy. Seizure 1998; 7: 299–303. Jalava, M., Sillanp¨aa¨ , M., Camfield, C. and Camfield, P. Social
259
62.
63.
64.
65.
66.
67.
68. 69.
adjustment and competence 35 years after onset of childhood epilepsy: a prospective controlled study. Epilepsia 1997; 38: 708–715. Austin, J. K. and deBoer, H. M. Disruptions in social functioning and services facilitating adjustment for the child and adult. In: Epilepsy: A Comprehensive Textbook (Eds J. Engel Jr and T. A. Pedley). Philadelphia, New York, Lippincott-Raven Pub., 1998: pp. 2191–2201. Dam, A. M. Hippocampal neuron loss in epilepsy and after experimental seizures. Acta Neurologica Scandinavica 1982; 66: 601–642. Saukkonen, A., K¨alvi¨ainen, R., Partanen, K., Vainio, P., Riekkinen, P. and Pitk¨anen, A. Do seizures cause neuronal damage? A MRI study in newly diagnosed and chronic epilepsy. Neuroreport 1994; 6: 219–223. Pitk¨anen, A. Seizure-induced pathologies: Structural changes assosiated with behavioral consequences. Epilepsia 1996; 37 (Suppl. 4):Abstract 123. K¨alvi¨ainen, R., Salmenper¨a, T., Partanen, K., Vainio, P., Riekkinen, P. and Pitk¨anen, A. Recurrent seizures may cause hippocampal damage in temporal lobe epilepsy. Neurology 1998; 50: 1377–1382. Jokeit, H. and Ebner, A. Long term effects of refractory temporal lobe epilepsy on cognitive abilities: a cross sectional study. Journal of Neurology, Neurosurgery and Psychiatry 1999; 67: 44–50. Engel, J. Jr. Seizures and Epilepsy. Philadelphia, F. A. Davis Company, 1989. Andermann, F. Identification of candidates for surgical treatment of epilepsy. In: Surgical Treatment of the Epilepsies (Ed. J. Engel Jr). New York, Raven Press, 1987: pp. 51–70.