- Email: [email protected]
Incidence of unprovoked seizures and epilepsy in Iceland and assessment of the epilepsy syndrome classification: a prospective study
Articles
Incidence of unprovoked seizures and epilepsy in Iceland and assessment of the epilepsy syndrome classification: a prospective study Elias Olafsson, Petur Ludvigsson, Gunnar Gudmundsson, Dale Hesdorffer, Olafur Kjartansson, W Allen Hauser
Summary Background No population-based incidence studies of epilepsy have studied syndrome classification from the outset. We prospectively studied the incidence of a single unprovoked seizure and epilepsy in the population of Iceland, and applied the syndrome classification endorsed by the International League Against Epilepsy to this population. Methods We used a nationwide surveillance system to prospectively identify all residents of Iceland who presented with a first diagnosis of a single unprovoked seizure or epilepsy between December 1995 and February 1999. All cases were classified by seizure type, cause or risk factors, and epilepsy syndrome.
Lancet Neurol 2005; 4: 627–34 Published online September 6, 2005 DOI:10.1016/S1474-4422(05) 70172-1
Interpretation Findings are consistent with incidence studies from developed countries. Although the epilepsy syndrome classification might be useful in tertiary epilepsy centers, it has limited practicality in population studies and for use by general neurologists.
Department of Neurology (E Olafsson MD), Department of Paediatrics (P Ludvigsson MD), and Department of Radiology (O Kjartansson MD) Landspitalinn University Hospital, Reykjavik, Iceland; Sergievsky Center (W A Hauser MD, E Olafsson, D Hesdorffer PhD), Department of Neurology, College of Physicians and Surgeons (W A Hauser), and the Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY, USA (D Hesdorffer, W A Hauser)
Introduction
Dr G Gudmundsson died in May, 1999
Results The mean annual incidence of first unprovoked seizures was 56·8 per 100 000 person-years, 23·5 per 100 000 person-years for single unprovoked seizures, and 33·3 per 100 000 person-years for epilepsy (recurrent unprovoked seizures). Incidence was similar in males and females. Partial seizures occurred in 40% and a putative cause was identified in 33%. Age-specific incidence was highest in the first year of life (130 per 100 000 person-years) and in those 65 years and older (110·5 per 100 000 person-years). Using strict diagnostic criteria for epilepsy syndromes, 58% of cases fell into non-informative categories. Idiopathic epilepsy syndromes were identified in 14% of all cases.
The epilepsy syndrome classification proposed by the International League Against Epilepsy is now in its third revision.1,2,3,4 The classification uses a combination of seizure semiology, family history, clinical neurophysiology, and clinical characteristics to classify patients. Classification aims to facilitate communication between physicians, to provide consistency in scientific studies and therapeutic trials, to inform patients about their prognosis, and to aid targeted drug therapy. The use of the syndrome classification by professionals who do not specialise in epilepsy, as well as by epilepsy researchers and for epidemiological studies, has been encouraged,2 but identifying criteria are complex and the information needed is commonly unavailable to the practising clinician. Additionally, there is a lack of evidence-based criteria for individual syndromes. As a result, the syndrome classification has yet to receive wide acceptance in the general medical community. A separate classification system has been recommended for use in epidemiological studies.5 In this system, cases are categorised on four independent axes: provoked versus unprovoked seizure, seizure type, aetiology, and number of episodes (single versus recurrent). The categorisation allows some determination of initial treatment strategies and prognosis. No population-based incidence studies of epilepsy have included syndrome classification, in addition to http://neurology.thelancet.com Vol 4 October 2005
more traditional categorisation by seizure type and aetiology, from the outset.5 Total population incidence studies of epilepsy have either retrospectively assigned the syndrome classification6,7 or have reported only syndrome classification without providing other data generally used in epidemiological studies.8 We used the medical care system in Iceland to establish a nationwide surveillance system to prospectively identify all individuals who presented with first diagnosis of a single unprovoked seizure or first diagnosis of epilepsy during a 39 month period. All cases were classified by number of episodes, seizure type, cause or risk factors, and epilepsy syndrome.
Correspondence to: Dr E Olafsson Department of Neurology, National University Hospital (Landspitalinn), 101 Reykjavik, Iceland [email protected]
Methods Patients Index cases were all residents of Iceland who were first diagnosed with an unprovoked seizure or epilepsy between December 1, 1995, and February 28, 1999. Epilepsy was defined in accordance with the 1993 International League Against Epilepsy criteria for epidemiological studies as recurrent unprovoked seizures (seizures occurring more than 24 h apart).5 In patients with newly diagnosed epilepsy, the first seizure might have occurred months or years before the study; these patients were included if there had been no previous medical diagnosis of epilepsy at the time of study presentation. People with a single unprovoked 627
Articles
seizure were included in the study but were separately classified. Only definite diagnoses were included. Patients were excluded if they had only provoked (acute symptomatic) seizures,5,9 including febrile seizures and neonatal seizures. In Iceland the study was approved by the Icelandic Government Ethics Committee and the Ethical Committee of the National University Hospital, and in the USA it was approved by the Institutional Review Board of Columbia University and the National Institutes of Health Ethical Committee (Office for Protection from Research Risks).
Procedures Before study initiation, we made Icelandic physicians and the general population aware of the study. The study was explained in a letter sent to all practising physicians and it was covered in the news. Physicians who participated in the study gave lectures on epilepsy and the purpose of the study to physicians across Iceland. To identify possible new cases, we used the countrywide surveillance system to regularly contact all healthcare facilities: all hospitals, emergency rooms, nursing homes, and health-care centres, the four radiology departments with facilities for CT or MRI scanning, and the two electroencephalography laboratories. When a possible case was identified through the surveillance system, one of the neurologists participating in the study (EO, GG, PL, or WAH) reviewed a synopsis of the case. If there was uncertainty about inclusion in the study, we contacted the treating physician for further information. Each potential case was classified; the “definite” classification applied to patients with onset of first unprovoked seizure or newly diagnosed epilepsy, “possible” to patients with episodic symptoms of unknown cause but with sufficient information to establish the symptoms as epileptic, and “not a case” to episodic symptoms other than epilepsy Total
Males
Females
Population
Number
Incidence*
Cumulative incidence
Population
Number
Incidence* Cumulative incidence
Population
Number
Incidence*
Cumulative incidence
13 828 57 367 72 951 66 032 137 702 132 916 131 855 101 829 66 303 58 185 32 649 10 534 882 151 ··
18 31 48 37 98 48 42 35 32 41 55 16 501 ··
130·2 54·0 65·8 56·0 71·2 36·1 31·9 34·4 48·3 70·5 168·5 151·9 56·8 55·2
0·0013 0·0035 0·0067 0·0095 0·0165 0·0201 0·0232 0·0265 0·0312 0·0380 0·0541 0·0683 ·· ··
7054 29 420 37 502 33 678 70 203 66 746 66 980 52 101 32 753 27 797 13 955 3777 441 966 ··
9 14 29 21 45 21 22 20 15 24 26 9 255 ··
127·6 47·6 77·3 62·4 64·1 31·5 32·8 38·4 45·8 86·3 186·3 238·3 57·7 57·3
6774 27 947 35 449 32 354 67 499 66 170 64 876 49 727 33 550 30 388 18 694 6757 440 185 ··
9 17 19 16 53 27 20 15 17 17 29 7 246 ··
132·9 60·8 53·6 49·5 78·5 40·8 30·8 30·2 50·7 55·9 155·1 103·6 55·9 53·7
0·0013 0·0038 0·0064 0·0089 0·0166 0·0206 0·0236 0·0266 0·0315 0·0369 0·0517 0·0617 ·· ··
Age (years) 1 1–4 5–9 10–14 15–24 25–34 35–44 45–54 55–64 65–74 75–84 85 Total Age adjusted†
(eg, syncope, cardiac arrhythmia, parasomnia, tics). Information for each possible case was further reviewed at least once after 6 months to investigate whether an unprovoked seizure or epilepsy had been diagnosed subsequently. Among index cases, the date of diagnosis was the day the patient first sought medical attention that resulted in a first diagnosis of an unprovoked seizure or epilepsy. Index cases participating in a nested case-control study (84·4% of cases ) were given an Icelandic translation of a structured, validated questionnaire to investigate their seizures.9 All other index cases were classified according to medical records only. All data were reviewed to assess case status, age at onset, age at diagnosis, seizure type, cause or risk factors, and epileptic syndrome. At least two of the study neurologists (EO, GG, PL, or WAH) did these reviews, and a larger group reviewed cases that were not clear. We classified seizure type using seizure semiology and electroencephalography findings in accordance with the 1981 recommendations of the International League Against Epilepsy.10 We only used electroencephalographs done within 12 months of the index date. Patients with generalised tonic-clonic seizures and with focal epileptiform abnormalities on electroencephalographs were regarded as having partial seizures with secondary generalisation; those with a three to four per second generalised spike wave and no known brain pathology were regarded as “primary generalised” or idiopathic. Patients with generalised tonic-clonic seizures without aura and with normal or non-specific electroencephalograph abnormalities were grouped separately, as “generalised tonic-clonic seizures only”. Patients with myoclonic, atonic, tonic, or no seizures, with or without generalised tonic-clonic seizures, were included in the generalised-onset seizures group. People with generalised tonic-clonic seizures only were excluded from this category.
0·0013 0·0032 0·0070 0·0101 0·0164 0·0195 0·0227 0·0265 0·0309 0·0393 0·0570 0·0790 ·· ··
*per 100 000 person-years; †Standard European million.
Table 1: Incidence of all unprovoked seizures by age and sex
628
http://neurology.thelancet.com Vol 4 October 2005
Articles
Incidence per 100 000 person-years
250 Male Female
200 150 100 50 0 0
20
40
Age (years)
60
80
100
80
100
Figure 1: Incidence of all unprovoked seizures by age in Iceland from 1995 to 1999
estimate the total population over the study period, we used the official population on December 1 of each year of the study and for the year 2000, with monthly increments interpolated between census figures. Total person-years of observation was 882 151. Total incidence and sex-specific incidence were age adjusted to the population of the Standard European million.13
Role of the funding source The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Results 501 patients met the inclusion criteria during the study period. 207 of these patients had a first, single, unprovoked seizure, and 294 had a first diagnosis of epilepsy. At least one electroencephalograph was done in 444 (89%) patients studied, and at least one CT or MRI study was done in 428 (85%) patients. Inpatient videomonitoring was available in Iceland but was not done on any cases in the first year after diagnosis. The mean number of patients with unprovoked seizures identified
Incidence per 100 000 person-years
We used broad aetiological categories: unprovoked seizures of unknown cause or symptomatic unprovoked seizures. Unprovoked seizures of unknown cause were further categorised as idiopathic or cryptogenic in accordance with the 1985 and 1989 syndrome classification.2,3 Idiopathic cases included patients with ictal characteristics and neurophysiology findings consistent with specific idiopathic syndrome descriptions. Cryptogenic cases were those presumed to have focal seizures but for whom a specific cause was not identified. Symptomatic cases were divided into remote symptomatic5 (associated with static brain pathology after cerebrovascular disease, mental retardation, cerebral palsy, infection, and trauma), and progressive symptomatic (associated with primary and secondary neoplasm of the brain, degenerative brain disease including dementia of Alzheimer’s type, multiple sclerosis, systemic lupus erythematosus, and metabolic brain disease).5 Definitions for aetiological categories from one of our previous studies,11 which were in accordance with the recommendations of the International League Against Epilepsy Commission on Epidemiology and Prognosis, were used to classify cause for patients in the symptomatic group.5 All cases were classified in accordance with the 1989 International League Against Epilepsy classification of epilepsy syndromes, based on age, seizure semiology, and clinical neurophysiology findings.3 Patients with a single unprovoked seizure were classified as special syndrome, single unprovoked seizure, or status epilepticus, and were also classified separately, according to electroencephalography and clinical features, into the category they would have been assigned if they had recurrent seizures. Patients with only generalised tonic-clonic seizures without other features were given different classifications in the epilepsy syndrome and seizure classifications. Patients with generalised tonic-clonic seizures and a generalised spike-wave pattern on electroencephalography and without other seizure types and aura were classified as “other idiopathic generalised epilepsy” (syndrome classification 2·1). Patients with generalised tonic-clonic seizures and focal epileptiform activity were classified as “cryptogenic localisation related epilepsy” (syndrome category 1·3). Cases with generalised tonic-clonic seizures and a history of neurological insult were classified as “symptomatic localisation related epilepsy” (syndrome category 1·2). When there was no history of neurological insult or epileptiform activity on the electroencephalograph, patients were classified as “epilepsy without unequivocal generalised or focal features” (syndrome category 3·2).
200 Single unprovoked seizure Epilepsy All unprovoked seizures
150 100 50 0 0
20
40
Age (years)
60
Statistical analysis Estimates of the total population were made with data from the Icelandic Bureau of Vital Statistics.12 To http://neurology.thelancet.com Vol 4 October 2005
Figure 2: Incidence of a single unprovoked seizure, epilepsy, and all unprovoked seizures by age in Iceland from 1995 to 1999
629
Articles
45 cerebrovascular
71 idiopathic
33 degenerative 7 infection 29 neoplasm 23 trauma 3 other 21 MR/CP 2 genetic
267 cryptogenic
Figure 3: Distribution of causes of and risk factors for all (501) unprovoked seizures in Iceland from 1995 to 1999 MR/CP=mental retardation/cerebral palsy.
per month was 12 in the first year of the study, 15 in the second year, and 12 in the third year. The incidence of all unprovoked seizures was 56·8 per 100 000 person-years (95% CI 51·8–61·8; table 1); for single unprovoked seizures was 23·5 per 100 000 personyears (20·3–26·7), and for epilepsy was 33·3 per 100 000 person-years (29·5–37·1). After adjustment to the European Standard million population,13 incidence was 55·2 for all unprovoked seizures; 22·8 for single unprovoked seizures, and 32·4 for epilepsy. The index cases included 255 (50·9%) male patients and 246 (49·1%) female patients. Incidence was 57·7 (95% CI 50·6–64·8) for male patients and 55·9 (48·9–62·9) for female patients; table 1 shows the incidence for each age group and sex, and the ageadjusted incidence. 108 male patients and 99 female patients presented with a single unprovoked seizure; 147 male patients and 147 female patients presented with epilepsy. The incidence of single unprovoked seizures in male patients was 24·4 (19·8–29·0) and 24·1 per 100 000 person-years when age adjusted. The incidence of single unprovoked seizures in female
Incidence per 100 000 person-years
120 Symptomatic seizures Idiopathic seizures Cryptogenic seizures
100 80 60 40 20 0 0
20
40
Age (years)
60
Figure 4: Incidence of broad causes of epilepsy by age in Iceland from 1995 to 1999
630
80
100
patients was 22·5 (18·1–26·9) and 21·8 per 100 000 person-years when age adjusted. The incidence of epilepsy was 33·3 (27·9–38·6) in male patients and 33·4 (28·0–38·8) in female patients. The age-adjusted incidence of epilepsy was 33·1 per 100 000 person-years in men and 31·9 per 100 000 person-years in women. The age-specific incidence of all unprovoked seizures was high in young and old individuals: 130·2 per 100 000 person-years in patients younger than 1 year, and 110·5 per 100 000 person-years in patients age 65 years and older. Incidence was highest, at 168·5, in those aged between 75 and 84 years. The age-specific incidence was similar in male and female patients until they were older than 85 years, when the incidence in women was half that in men (figure 1). The incidence of a single unprovoked seizure was similar to that for epilepsy up to 64 years of age, although estimates were based on small numbers (figure 2). In patients older than 64 years the incidence of epilepsy exceeded that of single unprovoked seizures. An antecedent disorder, commonly associated with unprovoked seizures, was identified in 163 (33%) patients (figure 3). The most common of these disorders was cerebrovascular disease, which occurred in 45 (9·0%) patients; the next most common disorders were degenerative disease in 33 (6·6%) patients, and neoplasm in 29 (5·8%) patients. The proportion of all cases of unprovoked seizures with a specific cause varied with age at identification (figure 4). A cause was identified in 13% (18/134) of patients who were younger than 15 years, 31% (79/255) of those aged 15 to 64 years, and 59% (66/112) of those 65 years and older. The most common cause in children was cerebral palsy, seen in 5% (7/134). In adults aged 15 to 64 years, the most common causes were brain trauma (9%; 22/255) and cerebrovascular disease (7%; 19/255). In adults aged 65 years and older the most common causes were degenerative disease (25%; 28/112) and stroke (23%; 26/112). A probable cause was identified in a similar proportion of patients with epilepsy (33% 98/294) and single unprovoked seizure (31%; 65/207). Partial seizures (including secondary generalised seizures) were identified in 40% (202/501) of all cases. The incidence of partial seizures was 22·9 per 100 000 per year. The incidence of partial seizures was high in the youngest and oldest patients; specifically, it is high during the first year of life and then rises progressively after 54 years of age (figure 5). Generalised-onset seizures occurred in 6% (30/501) of all cases; incidence was 5·3 per 100 000 per year. The incidence in this category was greatest in the first year of life, but some cases were identified between age 40 and 50 years. The only ictal sign in 263 (52%) patients was a generalised tonic-clonic seizure; a three to four per second generalised spike-wave discharge occurred in 26 (10%) of these and 35 (13%) patients had focal http://neurology.thelancet.com Vol 4 October 2005
Articles
epileptiform discharges. The incidence of generalised tonic-clonic seizures only was 29·8 per 100 000 personyears. The age-specific incidence was similar to that of partial seizures: high in early childhood, and increasing with increasing age in the old age groups. We could not classify seizure type in six (1%) patients; four of these were older than 70 years of age. 207 (41%) patients were identified after having a single unprovoked seizure. Of these, 60 (29%) had partial seizures, three (1%) had generalised-onset seizure other than generalised tonic-clonic seizure, and 143 (69%) had generalised tonic-clonic seizure only. The seizure type was unknown in one patient. Epilepsy was diagnosed at presentation in 294 (59%) patients; partial seizures were seen in 142 (48%), 27 (9%) had generalised-onset seizures, and 120 (41%) had generalised tonic-clonic only. In five (2%) the seizure type was unknown. Special syndromes (syndrome category 4·2: isolated seizure or status epilepticus) occurred in 41% (207/501) of all patients (incidence 23·5 per 100 000 person-years; table 2). Epilepsy without unequivocal evidence of generalised or focal features (syndrome category 3·2; incidence 8·7 per 100 000 person-years, 77/501), cryptogenic localisation-related epilepsy (syndrome category 1·3; incidence 8·8 per 100 000 person-years, 78/501), and symptomatic localisation-related epilepsy (syndrome category 1·2; incidence 8·4 per 100 000 person-years, 74/501) each accounted for 15% of patients. Idiopathic generalised epilepsy (syndrome category 2·1) occurred in 6% (30/501; incidence 3·4 per 100 000 person-years). Classification of single seizures into the category they would have been assigned to if the seizures had been
Incidence per 100 000 person-years
100 Partial seizures Generalised tonic-clonic seizures Other generalised seizures
80 60 40 20 0 0
20
40
60
Age (years)
80
100
Figure 5: Incidence of types of all unprovoked seizures by age in Iceland from 1995 to 1999
recurrent found almost half fitted category 3·2 (table 3). Epilepsy syndromes that have been clinically well defined were rare. Juvenile myoclonic epilepsy occurred in six patients (1%; incidence 0·7 per 100 000 person-years). Childhood absence epilepsy occurred in seven patients (1%; incidence 0·8 per 100 000 person-years). Benign rolandic epilepsy occurred in 25 patients (5%; incidence 2·8 per 100 000 person-years). West syndrome occurred in six patients (1%; incidence 0·007 per 100 000 personyears). Landau-Kleffner syndrome occurred in two patients (0·4%; incidence 0·2 per 100 000 person-years). Benign familial infantile convulsions occurred in three patients (0·6%; incidence 0·3 per 100 000 person-years). One patient had primary reading epilepsy and another had benign occipital epilepsy (each 0·1%; incidence 0·2 per 100 000 person-years). There were major differences in the proportion of cases classified into non-specific categories after age was taken
1·1 Idiopathic localisation related epilepsies
1·2 Symptomatic localisation related epilepsies
1·3 Cryptogenic localisation related epilepsies
2·1 Idiopathic generalised epilepsy
2·2 Cryptogenic or symptomatic generalised epilepsies
2·3 Symptomatic generalised epilepsies
3·1 3·2 4·2 Unknown Epilepsy with Epilepsy without Isolated both generalised unequivocal unprovoked and focal generalised or seizures or features focal features status epilepticus
Age (years)
P
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
1 1–4 5–9 10–14 15–24 25–34 35–44 45–54 55–64 65–74 75–84 85 All ages Age adjusted
13 828 57 367 72 951 66 032 137 702 132 916 131 855 101 829 66 303 58 185 32 649 10 534 882 151 ··
·· 1 9 6 ·· ·· ·· ·· ·· ·· ·· ·· 16 ··
·· 1·7 12·3 9·1 ·· ·· ·· ·· ·· ·· ·· ·· 1·8 1·6
·· ·· 2 1 12 8 7 8 7 13 11 5 74 ··
·· ·· 2·7 1·5 8·7 6·0 5·3 7·9 10·6 22·3 33·7 47·5 8·4 8·4
3 4 6 5 12 8 10 3 7 6 9 5 78 ··
21·7 7·0 8·2 7·6 8·7 6·0 7·6 2·9 10·6 10·3 27·6 47·5 8·8 8·6
1 ·· 7 8 9 4 1 ·· ·· ·· ·· ·· 30 ··
7·2 ·· 9·6 12·1 6·5 3 0·8 ·· ·· ·· ·· ·· 3·4 3·1
6 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 6 ··
43·4 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0·7 0·7
·· 1 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 1 ··
·· 1·7 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0·1 0·1
·· 1 5 1 1 ·· ·· ·· ·· ·· ·· ·· 8 ··
·· 1·7 6·9 1·5 0·7 ·· ·· ·· ·· ·· ·· ·· 0·9 0·8
1 3 0 1 18 7 5 6 6 10 15 5 77 ··
7·2 5·2 0·0 1·5 13·1 5·3 3·8 5·9 9·0 17·2 45·9 47·5 8·7 8·5
7 21 19 15 46 21 19 18 12 10 18 1 207 ··
50.6 36·6 26·0 22·7 38·5 17·3 15·2 19·6 21·1 20·6 61·3 9·5 23·5 22·8
·· ·· ·· ·· ·· ·· ·· ·· ·· 2 2 ·· 4 ··
·· ·· ·· ·· ·· ·· ·· ·· ·· 3·4 6·1 ·· 0·5 0·4
*Incidence per 100 000 person-years. P=population; N=number; I=incidence.
Table 2: Incidence* of epilepsy syndrome classifications in people with epilepsy (n=290) and people with single unprovoked seizures (n=207)
http://neurology.thelancet.com Vol 4 October 2005
631
Articles
1·1 Idiopathic 1·2 Symptomatic 1·3 Cryptogenic 2·1 Idiopathic localisation related localisation related localisation related generalised epilepsies epilepsies epilepsies epilepsy
2·2 Cryptogenic or symptomatic generalised epilepsies
2·3 Symptomatic 3·1 Epilepsy with 3·2 Epilepsy Unknown generalised both generalised without unequivocal epilepsies and focal features generalised or focal features
Age (years)
P
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
1 1–4 5–9 10–14 15–24 5–34 35–44 45–54 55–64 65–74 75–84 85 All ages
13 828 57 367 72 951 66 032 137 702 132 916 131 855 101 829 66 303 58 185 32 649 10 534 882 151
·· 1 5 5 1 ·· ·· ·· ·· ·· ·· ·· 12
·· 1·7 6·9 7·6 0·7 ·· ·· ·· ·· ·· ·· ·· 1·4
2 5 ·· ·· 7 10 2 8 5 3 9 1 52
14·5 8·7 ·· ·· 5·1 7·5 1·5 7·9 7·5 5·2 27·6 9·5 5·9
1 3 7 3 8 2 2 2 2 2 2 ·· 34
7·2 5·2 9·6 4·5 5·8 1·5 1·5 2·0 3·0 3·4 6·1 ·· 3·9
3 1 1 5 3 ·· ·· ·· ·· ·· ·· ·· 13
21·7 ·· 1·4 7·6 2·2 ·· ·· ·· ·· ·· ·· ·· 1·5
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· 1 ·· ·· ·· ·· ·· ·· ·· 1
·· ·· ·· ·· 0·7 ·· ·· ·· ·· ·· ·· ·· 0·1
1 11 6 2 26 9 14 8 5 5 7 ·· 94
7·2 19·2 8·2 3 18·9 6·8 10·6 7·9 7·5 8·6 21·4 ·· 10·7
·· ·· ·· ·· ·· ·· 1 ·· ·· ·· ·· ·· 1
·· ·· ·· ·· ·· ·· 0·8 ·· ·· ·· ·· ·· 0·8
*Incidence per 100 000 person-years. P=population; N=number; I=incidence.
Table 3: Incidence* of epilepsy syndrome classifications in 207 patients with a single seizure
into account. After reassigning those with a single seizure into more specific syndromes, 17% (23/134) of children with incident unprovoked seizure or epilepsy fell into non-informative categories, whereas, 48% (176/367) of adults fell into non-informative syndrome categories
Discussion The incidence of all unprovoked seizures, epilepsy, and single unprovoked seizures, age-adjusted to the Standard European million population, of 55·2 per 100 000 person years was similar to the overall incidence reported in other total population studies in more developed countries.8,5,14–20 A retrospective cohort study from Rochester, MN, USA, that used the same definitions as this study found the incidence of all unprovoked seizures to be 63 per 100 000 and the incidence of epilepsy to be 48 per 100 000 between 1975 and 1984.21 A study of incidence in 13 general practice settings near London, UK, between January 1995 and June 1996 reported the incidence per 100 000 of all unprovoked seizures to be 57, of epilepsy to be 46, and of a single seizure to be 11.14 Reports of incidence in children15 and adults16 from northern Sweden combined found the incidence of all unprovoked seizures to be about 60 per 100 000. The incidence of all unprovoked seizures, assessed mostly through surveillance of electroencephalography laboratories, in Geneva, Switzerland, was 46 per 100 00017 and in France was 44 per 100 000.8 However, the incidence in these studies might not be comparable with that in our study because 15% of the patients in our study were never referred for electroencephalography. Three total population studies report the incidence of epilepsy. A study from rural Iceland in 1992 found the incidence of epilepsy to be 47 per 100 000.18 The incidence of epilepsy in the Faroe Islands was 42 per 100 000,19 and in Copparo, Italy, it was 33·1.20 632
Our finding that the incidence of unprovoked seizures and epilepsy was 130·2 per 100 000 in those younger than 1 year, and that 27% of all cases are younger than 15 years of age at onset, is similar to the findings of other studies. The study done in Rochester found that incidence during the first year of life was 101 per 100 000 person-years, with 26% of all patients presenting before 15 years of age.21 Similarly, the study from Sweden found the incidence during the first year of life to be 166 per 100 000,15 and the incidence in the first year of life in Italy was 233 per 100 000.20 In our study, the incidence of unprovoked seizures and epilepsy in those aged 65 years, which increased with increasing age, matched the findings in several studies in developed countries. Among people aged 65 years and older, the incidence of all unprovoked seizures was 179 per 100 000 in Rochester,21 139 per 100 000 in northern Sweden15 for all unprovoked seizures, and 107 per 100 000 in Iceland for epilepsy only.18 The increase in incidence with age was not as large in the study from the London area14 and was not seen in the epilepsy studies in the Faroes19 or Copparo.20 The role of time trends in differences between incidence in very young and old individuals is unclear but may explain some of the variation.22 In the present study, the low incidence of a first unprovoked seizure in people older than 85 years of age compared with the high incidence of epilepsy in this age group suggests some diagnostic uncertainty and the need for multiple episodes to occur before a definitive diagnosis can be made. Supporting this notion, a substantial proportion of old patients in this study were “possible cases”. We found no difference in the incidence of unprovoked seizures by sex. Several other incidence studies have found a higher incidence in male patients than in female patients. The study from Rochester21 reported the incidence of all unprovoked seizures in http://neurology.thelancet.com Vol 4 October 2005
Articles
male patients to be 73 per 100 000 compared with 58 per 100 000 in female patients. The 1992 study of epilepsy from Iceland,18 the study from the Faroes,19 and the study from Copparo20 all found a higher incidence among men than women. Methods or definitions preclude comparisons for the other contemporary total population studies in developed countries.14,16–18 40% of the individuals in the current study presented to a physician after a single unprovoked seizure compared with 27% in the Rochester study21 and 19% in the London study.14 This difference could reflect differences in use of medical resources in these communities. We think that people in Iceland seek medical attention at the time of a first seizure and that these seizures are recognised as a first seizure. Only the study from Rochester21 can be compared with our study on the incidence of all unprovoked seizures by seizure type. In the current study, 59% of patients in the all-unprovoked-seizures group had generalised seizures, an incidence of 33 per 100 000. In Rochester, similarly, the incidence of all unprovoked generalised seizures was 30 per 100 000.21 Partial seizures occurred in 40% of the cases in this study, with an incidence of 23 per 100 000. The incidence of partial seizures in our study is comparable to an incidence of 30 per 100 000 in Rochester for all unprovoked seizures. Cause was unknown in 67% of people with unprovoked seizures, compared to 66% in Rochester,21 54% in the combined studies from northern Sweden15,16 46% in the study from Geneva,17 and 69% among those with epilepsy in rural Iceland.18 Comparison cannot be made with other studies. We found that 9% of cases were due to cerebrovascular disease (5·1 per 100 000) compared with 11% in the Rochester study,21 14% in the study from rural Iceland,18 16% in the study from Geneva,17 and 20% (symptomatic % localisation related Syndrome classification All unprovoked seizures Current study Current study with single seizure reclassified NGPSE7 Bordeaux8 Rochester6,15 CAROLE*24 CAROLE with single seizure reclassified* Epilepsy Current study Rochester6 CAROLE24 Single seizure Current study CAROLE24
cerebrovascular disease) in studies from northern Sweden.15,16 In the current study, 4% of newly diagnosed cases were attributable to primary brain tumours (3 per 100 000) compared to 7% in both the rural Iceland study18 and in northern Sweden.15,16 Brain trauma was thought to be an aetiological factor in 5% (3 per 100 000) of patients in the current study compared to 6% in Rochester,21 5% in Geneva,17 and 2% in northern Sweden.15,16 Thus the distribution of risk factors in this study is similar to those in other incidence studies in which comparisons can be made. The syndrome classification combines family history, seizure semiology, and neurophysiology to classify patients into specific groups.3 When the syndrome classification is used, 58% of all incident unprovoked seizures fall into non-specific syndrome categories. These categories—without unequivocal generalised or focal features (3·2), unclassifiable, or isolated seizure (4·2)—provide little information about prognosis or management, which is the rationale behind the syndrome classification. Even after reclassifying single unprovoked seizures as more specific epilepsy syndromes, 38% of patients in the all-unprovokedseizures group and 27% of those in the epilepsy group fall into non-specific categories (table 4). This finding highlights the difficulties with use of the syndrome classification in population-based studies. Few studies have reported the incidence of epileptic syndromes in all age groups; none have provided data with syndrome and epidemiological classifications in one paper. Only the retrospective classification of cases from the Rochester study is roughly comparable, providing syndrome data by age.6 Because reclassification in the Rochester study was done some 10 years after the initial review, information that was not available at the time of the review may have affected
% generalised
% undetermined
% special syndrome
2·2
2·3
3·1
3·2
4·1
4·2
5·99 8·58
1·20 1·20
0·20 0·20
1·60 1·80
15·37 34·13
·· ··
41·32 ··
0·80 0·80
57 57
25·75 15·72 27·14 15·29 20·08
9·70 13·21 4·52 14·32 21·94
0·00 2·52 2·01 2·01 2·01
1·94 0·00 3·02 1·34 2·01
0·00 0·00 2·51 0·21 0·31
33·51 4·40 11·56 8·91 31·62
E E ·· ·· ··
13·93 43·19 21·11 47·68 ··
0·00 0·00 0·00 0·72 0·72
·· 46 63 ·· ··
25·17 33·76 13·48
26·53 34·39 29·23
10·20 5·73 27·36
2·04 2·55 3·84
0·34 3·82 2·56
2·72 3·18 0·39
26·19 14·65 17·03
·· ·· ··
·· ·· ··
1·36 ·· 1·38
33 52 ··
25·12 16·09
16·43 10·04
6·28 15·98
0·00 0·00
0·00 1·40
0·48 0·22
45·41 47·62
·· ··
·· ··
0·48 0·00
24 ··
1·1
1·2
1·3
3·19 5·59
14·77 25·15
15·57 22·36
1·23 3·98 1·01 2·47 6·59
16·93 16·35 26·63 7·05 14·73
5·44 1·27 4·72 5·80 8·64
2·1
% unclassifiable Incidence
*Includes epilepsy and single seizure. NGPSE=National General Practice Study of Epilepsy; E=excluded from analysis; CAROLE=Coordination Active du Reseau Observatoire Longitudinal de l’epilepsie.
Table 4: Comparison of syndrome classification across studies of incidence of all unprovoked seizures, epilepsy, and single seizures
http://neurology.thelancet.com Vol 4 October 2005
633
Articles
classification. The National General Practice Study of Epilepsy7 of new onset cases provides no incidence data and there are no data by age. Also, cases were classified retrospectively and it is unclear how information obtained after the diagnosis was handled.7 In the incidence study from Bordeaux, cases were identified via neurologists and through referrals to electroencephalography laboratories only,8 whereas in the current study they were identified through population surveillance. For this reason, in the Bordeaux study few cases were classified as syndrome 3·2 (undetermined whether generalised or focal epilepsy syndrome).8 The same pattern of classification occurred with the newonset cases included in the CAROLE study, which were cases identified by epileptologists throughout France.23 As in most other population-based studies, the specific syndromes in the current study were rare. The exception was benign rolandic epilepsy, which accounted for 20% of new-onset cases in children, albeit only 5% of all cases. The low occurrence of specific syndromes is not limited to population-based studies. In a survey of all cases seen at a tertiary epilepsy centre, only a small proportion were able to be classified into specific syndromes.24 There is a false perception that syndrome classification has high specificity for localisation. In our study, we could assign a specific anatomical area of onset in only three of 238 people with localisation related epilepsy, despite availability of imaging and electroencephalography findings in more than 80% of cases. Our research suggests that the syndrome classification has some flaws. In studies of children with incident seizures, the syndrome classification has been used successfully, particularly for cases with recurrent seizures only.25–27 The present study confirms the usefulness of the syndrome classification in children, but suggests that it is less useful in adults, who make up 75% of incident cases. Acknowledgments Funding was provided in part by the National Institute of Neurologic Disease and Stroke, USA (R01 NS 32663). Authors’ contributions WAH conceptualised and obtained the funding for this study; WAH, DH, PL, OK, and EO designed the study; GG and EO organised physicians in Iceland for this study and supervised the study; DH collected the data; GG reviewed the data; all authors analysed the data; and WAH, DH, PL, OK, and ES prepared the manuscript.
4
5
6
7
8
9
10
11 12 13
14
15
16
17
18
19 20 21
22
23
24
Conflicts of interest We have no conflicts of interest. References 1 Merlis JK. Proposal for an international classification of the epilepsies. Epilepsia 1970; 11: 114–119. 2 Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for classification of epilepsies and epileptic syndromes. Epilepsia 1985; 26: 268–78. 3 Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989; 30: 389–99.
634
25
26 27
Engel J Jr. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia 2001; 42: 796–803. Commission on Epidemiology and Prognosis, International League against Epilepsy. Guidelines for epidemiologic studies on epilepsy. Epilepsia 1993: 34: 592–96. Zarrelli MM, Beghi E, Rocca WA, Hauser WA. Incidence of epileptic syndromes in Rochester, Minnesota: 1980–1984. Epilepsia 1999; 40: 1708–14. Manford M, Hart YS, Sanders JWAS, Shorvon SD. The National General Practice Study of Epilepsy: the syndromic classification of the International League Against Epilepsy applied to epilepsy in a general population. Arch Neurol 1992; 49: 801–08. Loiseau J, Loiseau P, Guyot M, et al. Survey of seizures in the French Southwest, 1: Incidence of epileptic syndromes. Epilepsia 1990; 31: 391–96. Ottman R, Lee JH, Hauser WA, et al. Reliability of seizure classification using a semistructured interview. Neurology 1993; 43: 2526–30. Commission on classification and terminology of the International League Against Epilepsy. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia 1981; 22: 489–501. Hauser WA, Annegers JF, Kurland LT. Prevalence of epilepsy in Rochester, Minnesota 1940–1980. Epilepsia 1991; 32: 429–44. Statistical Bureau of Iceland. http://www.hagstofa.is (accessed September 1, 2005). National Cancer Institute, US National Institutes of Health. Surveillance, epidemiology, and end results; standard populations—19 age groups. http://seer.cancer.gov/ stdpopulations/stdpop.19ages.html (accessed September 1, 2005). MacDonald BK, Cockerell OC, Sander JW, Shorvon SD. The incidence and lifetime prevalence of neurological disorders in a prospective community-based study in the UK. Brain 2000; 123: 665–76. Sidenvall R, Forsgren L, Blomquist HK, Heijbel J. A communitybased prospective incidence study of epileptic seizures in children. Acta Paediatrica 1993; 82: 60–65 Forsgren L, Bucht G, Eriksson S, Bergmark L. Incidence and clinical characterization of unprovoked seizures in adults: a prospective population-based study. Epilepsia 1996; 37: 224–29. Jallon P, Goumaz M, Haenggeli, Morabia A. Incidence of first epileptic seizures in the canto of Geneva, Switzerland. Epilepsia 1997; 38: 547–52. Olafsson E, Hauser WA, Ludvigsson P, Gudmundsson G. Incidence of epilepsy in rural Iceland: a population-based study. Epilepsia 1996; 37: 951–55. Joensen P. Prevalence, incidence, and classification of epilepsy in the Faroes. Acta Neurol Scand 1986; 74: 150–55. Granieri E, Rosati G, Tola R, et al. A descriptive study of epilepsy in the district of Copparo, Italy, 1964–1978. Epilepsia 1983; 24: 502–14. Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy and unprovoked seizures in Rochester Minnesota: 1935–1984. Epilepsia 1993; 34: 453–68. Annegers JF, Hauser WA, Lee JR, Rocca WA. Secular trends and birth cohort effects in unprovoked seizures: Rochester, Minnesota 1935–1984. Epilepsia 1995; 36: 575–79. Jallon P, Loiseau P, Loiseau J. Newly diagnosed unprovoked epileptic seizures: presentation at diagnosis in CAROLE study. Epilepsia 2001; 42: 464–75. Kellinghaus C, Loddenkemper T, Najm IM, et al. Specific epileptic syndromes are rare even in tertiary epilepsy centers: a patientoriented approach to epilepsy classification. Epilepsia 2004; 45: 268–75. Freitag CM, May TW, Pfafflin M, Konig S, Rating D. Incidence of epilepsies and epileptic syndromes in children and adolescents: a population-based prospective study in Germany. Epilepsia 2001; 42: 979–85. Berg AT, Shinnar S, Levy SR, Testa FM. Newly diagnosed epilepsy in children: presentation at diagnosis. Epilepsia 1999; 40: 445–52. Onsurbe Ramirez I, Hernandez Rodriguez M, Aparicio Meix JM. Carrascosa Romero incidence of epilepsy and epileptic syndromes in children in the province of Albacete. An Esp Pediatr 1999; 51: 154–58.
http://neurology.thelancet.com Vol 4 October 2005
Incidence of unprovoked seizures and epilepsy in Iceland and assessment of the epilepsy syndrome classification: a prospective study Elias Olafsson, Petur Ludvigsson, Gunnar Gudmundsson, Dale Hesdorffer, Olafur Kjartansson, W Allen Hauser
Summary Background No population-based incidence studies of epilepsy have studied syndrome classification from the outset. We prospectively studied the incidence of a single unprovoked seizure and epilepsy in the population of Iceland, and applied the syndrome classification endorsed by the International League Against Epilepsy to this population. Methods We used a nationwide surveillance system to prospectively identify all residents of Iceland who presented with a first diagnosis of a single unprovoked seizure or epilepsy between December 1995 and February 1999. All cases were classified by seizure type, cause or risk factors, and epilepsy syndrome.
Lancet Neurol 2005; 4: 627–34 Published online September 6, 2005 DOI:10.1016/S1474-4422(05) 70172-1
Interpretation Findings are consistent with incidence studies from developed countries. Although the epilepsy syndrome classification might be useful in tertiary epilepsy centers, it has limited practicality in population studies and for use by general neurologists.
Department of Neurology (E Olafsson MD), Department of Paediatrics (P Ludvigsson MD), and Department of Radiology (O Kjartansson MD) Landspitalinn University Hospital, Reykjavik, Iceland; Sergievsky Center (W A Hauser MD, E Olafsson, D Hesdorffer PhD), Department of Neurology, College of Physicians and Surgeons (W A Hauser), and the Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY, USA (D Hesdorffer, W A Hauser)
Introduction
Dr G Gudmundsson died in May, 1999
Results The mean annual incidence of first unprovoked seizures was 56·8 per 100 000 person-years, 23·5 per 100 000 person-years for single unprovoked seizures, and 33·3 per 100 000 person-years for epilepsy (recurrent unprovoked seizures). Incidence was similar in males and females. Partial seizures occurred in 40% and a putative cause was identified in 33%. Age-specific incidence was highest in the first year of life (130 per 100 000 person-years) and in those 65 years and older (110·5 per 100 000 person-years). Using strict diagnostic criteria for epilepsy syndromes, 58% of cases fell into non-informative categories. Idiopathic epilepsy syndromes were identified in 14% of all cases.
The epilepsy syndrome classification proposed by the International League Against Epilepsy is now in its third revision.1,2,3,4 The classification uses a combination of seizure semiology, family history, clinical neurophysiology, and clinical characteristics to classify patients. Classification aims to facilitate communication between physicians, to provide consistency in scientific studies and therapeutic trials, to inform patients about their prognosis, and to aid targeted drug therapy. The use of the syndrome classification by professionals who do not specialise in epilepsy, as well as by epilepsy researchers and for epidemiological studies, has been encouraged,2 but identifying criteria are complex and the information needed is commonly unavailable to the practising clinician. Additionally, there is a lack of evidence-based criteria for individual syndromes. As a result, the syndrome classification has yet to receive wide acceptance in the general medical community. A separate classification system has been recommended for use in epidemiological studies.5 In this system, cases are categorised on four independent axes: provoked versus unprovoked seizure, seizure type, aetiology, and number of episodes (single versus recurrent). The categorisation allows some determination of initial treatment strategies and prognosis. No population-based incidence studies of epilepsy have included syndrome classification, in addition to http://neurology.thelancet.com Vol 4 October 2005
more traditional categorisation by seizure type and aetiology, from the outset.5 Total population incidence studies of epilepsy have either retrospectively assigned the syndrome classification6,7 or have reported only syndrome classification without providing other data generally used in epidemiological studies.8 We used the medical care system in Iceland to establish a nationwide surveillance system to prospectively identify all individuals who presented with first diagnosis of a single unprovoked seizure or first diagnosis of epilepsy during a 39 month period. All cases were classified by number of episodes, seizure type, cause or risk factors, and epilepsy syndrome.
Correspondence to: Dr E Olafsson Department of Neurology, National University Hospital (Landspitalinn), 101 Reykjavik, Iceland [email protected]
Methods Patients Index cases were all residents of Iceland who were first diagnosed with an unprovoked seizure or epilepsy between December 1, 1995, and February 28, 1999. Epilepsy was defined in accordance with the 1993 International League Against Epilepsy criteria for epidemiological studies as recurrent unprovoked seizures (seizures occurring more than 24 h apart).5 In patients with newly diagnosed epilepsy, the first seizure might have occurred months or years before the study; these patients were included if there had been no previous medical diagnosis of epilepsy at the time of study presentation. People with a single unprovoked 627
Articles
seizure were included in the study but were separately classified. Only definite diagnoses were included. Patients were excluded if they had only provoked (acute symptomatic) seizures,5,9 including febrile seizures and neonatal seizures. In Iceland the study was approved by the Icelandic Government Ethics Committee and the Ethical Committee of the National University Hospital, and in the USA it was approved by the Institutional Review Board of Columbia University and the National Institutes of Health Ethical Committee (Office for Protection from Research Risks).
Procedures Before study initiation, we made Icelandic physicians and the general population aware of the study. The study was explained in a letter sent to all practising physicians and it was covered in the news. Physicians who participated in the study gave lectures on epilepsy and the purpose of the study to physicians across Iceland. To identify possible new cases, we used the countrywide surveillance system to regularly contact all healthcare facilities: all hospitals, emergency rooms, nursing homes, and health-care centres, the four radiology departments with facilities for CT or MRI scanning, and the two electroencephalography laboratories. When a possible case was identified through the surveillance system, one of the neurologists participating in the study (EO, GG, PL, or WAH) reviewed a synopsis of the case. If there was uncertainty about inclusion in the study, we contacted the treating physician for further information. Each potential case was classified; the “definite” classification applied to patients with onset of first unprovoked seizure or newly diagnosed epilepsy, “possible” to patients with episodic symptoms of unknown cause but with sufficient information to establish the symptoms as epileptic, and “not a case” to episodic symptoms other than epilepsy Total
Males
Females
Population
Number
Incidence*
Cumulative incidence
Population
Number
Incidence* Cumulative incidence
Population
Number
Incidence*
Cumulative incidence
13 828 57 367 72 951 66 032 137 702 132 916 131 855 101 829 66 303 58 185 32 649 10 534 882 151 ··
18 31 48 37 98 48 42 35 32 41 55 16 501 ··
130·2 54·0 65·8 56·0 71·2 36·1 31·9 34·4 48·3 70·5 168·5 151·9 56·8 55·2
0·0013 0·0035 0·0067 0·0095 0·0165 0·0201 0·0232 0·0265 0·0312 0·0380 0·0541 0·0683 ·· ··
7054 29 420 37 502 33 678 70 203 66 746 66 980 52 101 32 753 27 797 13 955 3777 441 966 ··
9 14 29 21 45 21 22 20 15 24 26 9 255 ··
127·6 47·6 77·3 62·4 64·1 31·5 32·8 38·4 45·8 86·3 186·3 238·3 57·7 57·3
6774 27 947 35 449 32 354 67 499 66 170 64 876 49 727 33 550 30 388 18 694 6757 440 185 ··
9 17 19 16 53 27 20 15 17 17 29 7 246 ··
132·9 60·8 53·6 49·5 78·5 40·8 30·8 30·2 50·7 55·9 155·1 103·6 55·9 53·7
0·0013 0·0038 0·0064 0·0089 0·0166 0·0206 0·0236 0·0266 0·0315 0·0369 0·0517 0·0617 ·· ··
Age (years) 1 1–4 5–9 10–14 15–24 25–34 35–44 45–54 55–64 65–74 75–84 85 Total Age adjusted†
(eg, syncope, cardiac arrhythmia, parasomnia, tics). Information for each possible case was further reviewed at least once after 6 months to investigate whether an unprovoked seizure or epilepsy had been diagnosed subsequently. Among index cases, the date of diagnosis was the day the patient first sought medical attention that resulted in a first diagnosis of an unprovoked seizure or epilepsy. Index cases participating in a nested case-control study (84·4% of cases ) were given an Icelandic translation of a structured, validated questionnaire to investigate their seizures.9 All other index cases were classified according to medical records only. All data were reviewed to assess case status, age at onset, age at diagnosis, seizure type, cause or risk factors, and epileptic syndrome. At least two of the study neurologists (EO, GG, PL, or WAH) did these reviews, and a larger group reviewed cases that were not clear. We classified seizure type using seizure semiology and electroencephalography findings in accordance with the 1981 recommendations of the International League Against Epilepsy.10 We only used electroencephalographs done within 12 months of the index date. Patients with generalised tonic-clonic seizures and with focal epileptiform abnormalities on electroencephalographs were regarded as having partial seizures with secondary generalisation; those with a three to four per second generalised spike wave and no known brain pathology were regarded as “primary generalised” or idiopathic. Patients with generalised tonic-clonic seizures without aura and with normal or non-specific electroencephalograph abnormalities were grouped separately, as “generalised tonic-clonic seizures only”. Patients with myoclonic, atonic, tonic, or no seizures, with or without generalised tonic-clonic seizures, were included in the generalised-onset seizures group. People with generalised tonic-clonic seizures only were excluded from this category.
0·0013 0·0032 0·0070 0·0101 0·0164 0·0195 0·0227 0·0265 0·0309 0·0393 0·0570 0·0790 ·· ··
*per 100 000 person-years; †Standard European million.
Table 1: Incidence of all unprovoked seizures by age and sex
628
http://neurology.thelancet.com Vol 4 October 2005
Articles
Incidence per 100 000 person-years
250 Male Female
200 150 100 50 0 0
20
40
Age (years)
60
80
100
80
100
Figure 1: Incidence of all unprovoked seizures by age in Iceland from 1995 to 1999
estimate the total population over the study period, we used the official population on December 1 of each year of the study and for the year 2000, with monthly increments interpolated between census figures. Total person-years of observation was 882 151. Total incidence and sex-specific incidence were age adjusted to the population of the Standard European million.13
Role of the funding source The sponsor of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Results 501 patients met the inclusion criteria during the study period. 207 of these patients had a first, single, unprovoked seizure, and 294 had a first diagnosis of epilepsy. At least one electroencephalograph was done in 444 (89%) patients studied, and at least one CT or MRI study was done in 428 (85%) patients. Inpatient videomonitoring was available in Iceland but was not done on any cases in the first year after diagnosis. The mean number of patients with unprovoked seizures identified
Incidence per 100 000 person-years
We used broad aetiological categories: unprovoked seizures of unknown cause or symptomatic unprovoked seizures. Unprovoked seizures of unknown cause were further categorised as idiopathic or cryptogenic in accordance with the 1985 and 1989 syndrome classification.2,3 Idiopathic cases included patients with ictal characteristics and neurophysiology findings consistent with specific idiopathic syndrome descriptions. Cryptogenic cases were those presumed to have focal seizures but for whom a specific cause was not identified. Symptomatic cases were divided into remote symptomatic5 (associated with static brain pathology after cerebrovascular disease, mental retardation, cerebral palsy, infection, and trauma), and progressive symptomatic (associated with primary and secondary neoplasm of the brain, degenerative brain disease including dementia of Alzheimer’s type, multiple sclerosis, systemic lupus erythematosus, and metabolic brain disease).5 Definitions for aetiological categories from one of our previous studies,11 which were in accordance with the recommendations of the International League Against Epilepsy Commission on Epidemiology and Prognosis, were used to classify cause for patients in the symptomatic group.5 All cases were classified in accordance with the 1989 International League Against Epilepsy classification of epilepsy syndromes, based on age, seizure semiology, and clinical neurophysiology findings.3 Patients with a single unprovoked seizure were classified as special syndrome, single unprovoked seizure, or status epilepticus, and were also classified separately, according to electroencephalography and clinical features, into the category they would have been assigned if they had recurrent seizures. Patients with only generalised tonic-clonic seizures without other features were given different classifications in the epilepsy syndrome and seizure classifications. Patients with generalised tonic-clonic seizures and a generalised spike-wave pattern on electroencephalography and without other seizure types and aura were classified as “other idiopathic generalised epilepsy” (syndrome classification 2·1). Patients with generalised tonic-clonic seizures and focal epileptiform activity were classified as “cryptogenic localisation related epilepsy” (syndrome category 1·3). Cases with generalised tonic-clonic seizures and a history of neurological insult were classified as “symptomatic localisation related epilepsy” (syndrome category 1·2). When there was no history of neurological insult or epileptiform activity on the electroencephalograph, patients were classified as “epilepsy without unequivocal generalised or focal features” (syndrome category 3·2).
200 Single unprovoked seizure Epilepsy All unprovoked seizures
150 100 50 0 0
20
40
Age (years)
60
Statistical analysis Estimates of the total population were made with data from the Icelandic Bureau of Vital Statistics.12 To http://neurology.thelancet.com Vol 4 October 2005
Figure 2: Incidence of a single unprovoked seizure, epilepsy, and all unprovoked seizures by age in Iceland from 1995 to 1999
629
Articles
45 cerebrovascular
71 idiopathic
33 degenerative 7 infection 29 neoplasm 23 trauma 3 other 21 MR/CP 2 genetic
267 cryptogenic
Figure 3: Distribution of causes of and risk factors for all (501) unprovoked seizures in Iceland from 1995 to 1999 MR/CP=mental retardation/cerebral palsy.
per month was 12 in the first year of the study, 15 in the second year, and 12 in the third year. The incidence of all unprovoked seizures was 56·8 per 100 000 person-years (95% CI 51·8–61·8; table 1); for single unprovoked seizures was 23·5 per 100 000 personyears (20·3–26·7), and for epilepsy was 33·3 per 100 000 person-years (29·5–37·1). After adjustment to the European Standard million population,13 incidence was 55·2 for all unprovoked seizures; 22·8 for single unprovoked seizures, and 32·4 for epilepsy. The index cases included 255 (50·9%) male patients and 246 (49·1%) female patients. Incidence was 57·7 (95% CI 50·6–64·8) for male patients and 55·9 (48·9–62·9) for female patients; table 1 shows the incidence for each age group and sex, and the ageadjusted incidence. 108 male patients and 99 female patients presented with a single unprovoked seizure; 147 male patients and 147 female patients presented with epilepsy. The incidence of single unprovoked seizures in male patients was 24·4 (19·8–29·0) and 24·1 per 100 000 person-years when age adjusted. The incidence of single unprovoked seizures in female
Incidence per 100 000 person-years
120 Symptomatic seizures Idiopathic seizures Cryptogenic seizures
100 80 60 40 20 0 0
20
40
Age (years)
60
Figure 4: Incidence of broad causes of epilepsy by age in Iceland from 1995 to 1999
630
80
100
patients was 22·5 (18·1–26·9) and 21·8 per 100 000 person-years when age adjusted. The incidence of epilepsy was 33·3 (27·9–38·6) in male patients and 33·4 (28·0–38·8) in female patients. The age-adjusted incidence of epilepsy was 33·1 per 100 000 person-years in men and 31·9 per 100 000 person-years in women. The age-specific incidence of all unprovoked seizures was high in young and old individuals: 130·2 per 100 000 person-years in patients younger than 1 year, and 110·5 per 100 000 person-years in patients age 65 years and older. Incidence was highest, at 168·5, in those aged between 75 and 84 years. The age-specific incidence was similar in male and female patients until they were older than 85 years, when the incidence in women was half that in men (figure 1). The incidence of a single unprovoked seizure was similar to that for epilepsy up to 64 years of age, although estimates were based on small numbers (figure 2). In patients older than 64 years the incidence of epilepsy exceeded that of single unprovoked seizures. An antecedent disorder, commonly associated with unprovoked seizures, was identified in 163 (33%) patients (figure 3). The most common of these disorders was cerebrovascular disease, which occurred in 45 (9·0%) patients; the next most common disorders were degenerative disease in 33 (6·6%) patients, and neoplasm in 29 (5·8%) patients. The proportion of all cases of unprovoked seizures with a specific cause varied with age at identification (figure 4). A cause was identified in 13% (18/134) of patients who were younger than 15 years, 31% (79/255) of those aged 15 to 64 years, and 59% (66/112) of those 65 years and older. The most common cause in children was cerebral palsy, seen in 5% (7/134). In adults aged 15 to 64 years, the most common causes were brain trauma (9%; 22/255) and cerebrovascular disease (7%; 19/255). In adults aged 65 years and older the most common causes were degenerative disease (25%; 28/112) and stroke (23%; 26/112). A probable cause was identified in a similar proportion of patients with epilepsy (33% 98/294) and single unprovoked seizure (31%; 65/207). Partial seizures (including secondary generalised seizures) were identified in 40% (202/501) of all cases. The incidence of partial seizures was 22·9 per 100 000 per year. The incidence of partial seizures was high in the youngest and oldest patients; specifically, it is high during the first year of life and then rises progressively after 54 years of age (figure 5). Generalised-onset seizures occurred in 6% (30/501) of all cases; incidence was 5·3 per 100 000 per year. The incidence in this category was greatest in the first year of life, but some cases were identified between age 40 and 50 years. The only ictal sign in 263 (52%) patients was a generalised tonic-clonic seizure; a three to four per second generalised spike-wave discharge occurred in 26 (10%) of these and 35 (13%) patients had focal http://neurology.thelancet.com Vol 4 October 2005
Articles
epileptiform discharges. The incidence of generalised tonic-clonic seizures only was 29·8 per 100 000 personyears. The age-specific incidence was similar to that of partial seizures: high in early childhood, and increasing with increasing age in the old age groups. We could not classify seizure type in six (1%) patients; four of these were older than 70 years of age. 207 (41%) patients were identified after having a single unprovoked seizure. Of these, 60 (29%) had partial seizures, three (1%) had generalised-onset seizure other than generalised tonic-clonic seizure, and 143 (69%) had generalised tonic-clonic seizure only. The seizure type was unknown in one patient. Epilepsy was diagnosed at presentation in 294 (59%) patients; partial seizures were seen in 142 (48%), 27 (9%) had generalised-onset seizures, and 120 (41%) had generalised tonic-clonic only. In five (2%) the seizure type was unknown. Special syndromes (syndrome category 4·2: isolated seizure or status epilepticus) occurred in 41% (207/501) of all patients (incidence 23·5 per 100 000 person-years; table 2). Epilepsy without unequivocal evidence of generalised or focal features (syndrome category 3·2; incidence 8·7 per 100 000 person-years, 77/501), cryptogenic localisation-related epilepsy (syndrome category 1·3; incidence 8·8 per 100 000 person-years, 78/501), and symptomatic localisation-related epilepsy (syndrome category 1·2; incidence 8·4 per 100 000 person-years, 74/501) each accounted for 15% of patients. Idiopathic generalised epilepsy (syndrome category 2·1) occurred in 6% (30/501; incidence 3·4 per 100 000 person-years). Classification of single seizures into the category they would have been assigned to if the seizures had been
Incidence per 100 000 person-years
100 Partial seizures Generalised tonic-clonic seizures Other generalised seizures
80 60 40 20 0 0
20
40
60
Age (years)
80
100
Figure 5: Incidence of types of all unprovoked seizures by age in Iceland from 1995 to 1999
recurrent found almost half fitted category 3·2 (table 3). Epilepsy syndromes that have been clinically well defined were rare. Juvenile myoclonic epilepsy occurred in six patients (1%; incidence 0·7 per 100 000 person-years). Childhood absence epilepsy occurred in seven patients (1%; incidence 0·8 per 100 000 person-years). Benign rolandic epilepsy occurred in 25 patients (5%; incidence 2·8 per 100 000 person-years). West syndrome occurred in six patients (1%; incidence 0·007 per 100 000 personyears). Landau-Kleffner syndrome occurred in two patients (0·4%; incidence 0·2 per 100 000 person-years). Benign familial infantile convulsions occurred in three patients (0·6%; incidence 0·3 per 100 000 person-years). One patient had primary reading epilepsy and another had benign occipital epilepsy (each 0·1%; incidence 0·2 per 100 000 person-years). There were major differences in the proportion of cases classified into non-specific categories after age was taken
1·1 Idiopathic localisation related epilepsies
1·2 Symptomatic localisation related epilepsies
1·3 Cryptogenic localisation related epilepsies
2·1 Idiopathic generalised epilepsy
2·2 Cryptogenic or symptomatic generalised epilepsies
2·3 Symptomatic generalised epilepsies
3·1 3·2 4·2 Unknown Epilepsy with Epilepsy without Isolated both generalised unequivocal unprovoked and focal generalised or seizures or features focal features status epilepticus
Age (years)
P
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
1 1–4 5–9 10–14 15–24 25–34 35–44 45–54 55–64 65–74 75–84 85 All ages Age adjusted
13 828 57 367 72 951 66 032 137 702 132 916 131 855 101 829 66 303 58 185 32 649 10 534 882 151 ··
·· 1 9 6 ·· ·· ·· ·· ·· ·· ·· ·· 16 ··
·· 1·7 12·3 9·1 ·· ·· ·· ·· ·· ·· ·· ·· 1·8 1·6
·· ·· 2 1 12 8 7 8 7 13 11 5 74 ··
·· ·· 2·7 1·5 8·7 6·0 5·3 7·9 10·6 22·3 33·7 47·5 8·4 8·4
3 4 6 5 12 8 10 3 7 6 9 5 78 ··
21·7 7·0 8·2 7·6 8·7 6·0 7·6 2·9 10·6 10·3 27·6 47·5 8·8 8·6
1 ·· 7 8 9 4 1 ·· ·· ·· ·· ·· 30 ··
7·2 ·· 9·6 12·1 6·5 3 0·8 ·· ·· ·· ·· ·· 3·4 3·1
6 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 6 ··
43·4 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0·7 0·7
·· 1 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 1 ··
·· 1·7 ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0·1 0·1
·· 1 5 1 1 ·· ·· ·· ·· ·· ·· ·· 8 ··
·· 1·7 6·9 1·5 0·7 ·· ·· ·· ·· ·· ·· ·· 0·9 0·8
1 3 0 1 18 7 5 6 6 10 15 5 77 ··
7·2 5·2 0·0 1·5 13·1 5·3 3·8 5·9 9·0 17·2 45·9 47·5 8·7 8·5
7 21 19 15 46 21 19 18 12 10 18 1 207 ··
50.6 36·6 26·0 22·7 38·5 17·3 15·2 19·6 21·1 20·6 61·3 9·5 23·5 22·8
·· ·· ·· ·· ·· ·· ·· ·· ·· 2 2 ·· 4 ··
·· ·· ·· ·· ·· ·· ·· ·· ·· 3·4 6·1 ·· 0·5 0·4
*Incidence per 100 000 person-years. P=population; N=number; I=incidence.
Table 2: Incidence* of epilepsy syndrome classifications in people with epilepsy (n=290) and people with single unprovoked seizures (n=207)
http://neurology.thelancet.com Vol 4 October 2005
631
Articles
1·1 Idiopathic 1·2 Symptomatic 1·3 Cryptogenic 2·1 Idiopathic localisation related localisation related localisation related generalised epilepsies epilepsies epilepsies epilepsy
2·2 Cryptogenic or symptomatic generalised epilepsies
2·3 Symptomatic 3·1 Epilepsy with 3·2 Epilepsy Unknown generalised both generalised without unequivocal epilepsies and focal features generalised or focal features
Age (years)
P
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
N
I
1 1–4 5–9 10–14 15–24 5–34 35–44 45–54 55–64 65–74 75–84 85 All ages
13 828 57 367 72 951 66 032 137 702 132 916 131 855 101 829 66 303 58 185 32 649 10 534 882 151
·· 1 5 5 1 ·· ·· ·· ·· ·· ·· ·· 12
·· 1·7 6·9 7·6 0·7 ·· ·· ·· ·· ·· ·· ·· 1·4
2 5 ·· ·· 7 10 2 8 5 3 9 1 52
14·5 8·7 ·· ·· 5·1 7·5 1·5 7·9 7·5 5·2 27·6 9·5 5·9
1 3 7 3 8 2 2 2 2 2 2 ·· 34
7·2 5·2 9·6 4·5 5·8 1·5 1·5 2·0 3·0 3·4 6·1 ·· 3·9
3 1 1 5 3 ·· ·· ·· ·· ·· ·· ·· 13
21·7 ·· 1·4 7·6 2·2 ·· ·· ·· ·· ·· ·· ·· 1·5
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· 0
·· ·· ·· ·· 1 ·· ·· ·· ·· ·· ·· ·· 1
·· ·· ·· ·· 0·7 ·· ·· ·· ·· ·· ·· ·· 0·1
1 11 6 2 26 9 14 8 5 5 7 ·· 94
7·2 19·2 8·2 3 18·9 6·8 10·6 7·9 7·5 8·6 21·4 ·· 10·7
·· ·· ·· ·· ·· ·· 1 ·· ·· ·· ·· ·· 1
·· ·· ·· ·· ·· ·· 0·8 ·· ·· ·· ·· ·· 0·8
*Incidence per 100 000 person-years. P=population; N=number; I=incidence.
Table 3: Incidence* of epilepsy syndrome classifications in 207 patients with a single seizure
into account. After reassigning those with a single seizure into more specific syndromes, 17% (23/134) of children with incident unprovoked seizure or epilepsy fell into non-informative categories, whereas, 48% (176/367) of adults fell into non-informative syndrome categories
Discussion The incidence of all unprovoked seizures, epilepsy, and single unprovoked seizures, age-adjusted to the Standard European million population, of 55·2 per 100 000 person years was similar to the overall incidence reported in other total population studies in more developed countries.8,5,14–20 A retrospective cohort study from Rochester, MN, USA, that used the same definitions as this study found the incidence of all unprovoked seizures to be 63 per 100 000 and the incidence of epilepsy to be 48 per 100 000 between 1975 and 1984.21 A study of incidence in 13 general practice settings near London, UK, between January 1995 and June 1996 reported the incidence per 100 000 of all unprovoked seizures to be 57, of epilepsy to be 46, and of a single seizure to be 11.14 Reports of incidence in children15 and adults16 from northern Sweden combined found the incidence of all unprovoked seizures to be about 60 per 100 000. The incidence of all unprovoked seizures, assessed mostly through surveillance of electroencephalography laboratories, in Geneva, Switzerland, was 46 per 100 00017 and in France was 44 per 100 000.8 However, the incidence in these studies might not be comparable with that in our study because 15% of the patients in our study were never referred for electroencephalography. Three total population studies report the incidence of epilepsy. A study from rural Iceland in 1992 found the incidence of epilepsy to be 47 per 100 000.18 The incidence of epilepsy in the Faroe Islands was 42 per 100 000,19 and in Copparo, Italy, it was 33·1.20 632
Our finding that the incidence of unprovoked seizures and epilepsy was 130·2 per 100 000 in those younger than 1 year, and that 27% of all cases are younger than 15 years of age at onset, is similar to the findings of other studies. The study done in Rochester found that incidence during the first year of life was 101 per 100 000 person-years, with 26% of all patients presenting before 15 years of age.21 Similarly, the study from Sweden found the incidence during the first year of life to be 166 per 100 000,15 and the incidence in the first year of life in Italy was 233 per 100 000.20 In our study, the incidence of unprovoked seizures and epilepsy in those aged 65 years, which increased with increasing age, matched the findings in several studies in developed countries. Among people aged 65 years and older, the incidence of all unprovoked seizures was 179 per 100 000 in Rochester,21 139 per 100 000 in northern Sweden15 for all unprovoked seizures, and 107 per 100 000 in Iceland for epilepsy only.18 The increase in incidence with age was not as large in the study from the London area14 and was not seen in the epilepsy studies in the Faroes19 or Copparo.20 The role of time trends in differences between incidence in very young and old individuals is unclear but may explain some of the variation.22 In the present study, the low incidence of a first unprovoked seizure in people older than 85 years of age compared with the high incidence of epilepsy in this age group suggests some diagnostic uncertainty and the need for multiple episodes to occur before a definitive diagnosis can be made. Supporting this notion, a substantial proportion of old patients in this study were “possible cases”. We found no difference in the incidence of unprovoked seizures by sex. Several other incidence studies have found a higher incidence in male patients than in female patients. The study from Rochester21 reported the incidence of all unprovoked seizures in http://neurology.thelancet.com Vol 4 October 2005
Articles
male patients to be 73 per 100 000 compared with 58 per 100 000 in female patients. The 1992 study of epilepsy from Iceland,18 the study from the Faroes,19 and the study from Copparo20 all found a higher incidence among men than women. Methods or definitions preclude comparisons for the other contemporary total population studies in developed countries.14,16–18 40% of the individuals in the current study presented to a physician after a single unprovoked seizure compared with 27% in the Rochester study21 and 19% in the London study.14 This difference could reflect differences in use of medical resources in these communities. We think that people in Iceland seek medical attention at the time of a first seizure and that these seizures are recognised as a first seizure. Only the study from Rochester21 can be compared with our study on the incidence of all unprovoked seizures by seizure type. In the current study, 59% of patients in the all-unprovoked-seizures group had generalised seizures, an incidence of 33 per 100 000. In Rochester, similarly, the incidence of all unprovoked generalised seizures was 30 per 100 000.21 Partial seizures occurred in 40% of the cases in this study, with an incidence of 23 per 100 000. The incidence of partial seizures in our study is comparable to an incidence of 30 per 100 000 in Rochester for all unprovoked seizures. Cause was unknown in 67% of people with unprovoked seizures, compared to 66% in Rochester,21 54% in the combined studies from northern Sweden15,16 46% in the study from Geneva,17 and 69% among those with epilepsy in rural Iceland.18 Comparison cannot be made with other studies. We found that 9% of cases were due to cerebrovascular disease (5·1 per 100 000) compared with 11% in the Rochester study,21 14% in the study from rural Iceland,18 16% in the study from Geneva,17 and 20% (symptomatic % localisation related Syndrome classification All unprovoked seizures Current study Current study with single seizure reclassified NGPSE7 Bordeaux8 Rochester6,15 CAROLE*24 CAROLE with single seizure reclassified* Epilepsy Current study Rochester6 CAROLE24 Single seizure Current study CAROLE24
cerebrovascular disease) in studies from northern Sweden.15,16 In the current study, 4% of newly diagnosed cases were attributable to primary brain tumours (3 per 100 000) compared to 7% in both the rural Iceland study18 and in northern Sweden.15,16 Brain trauma was thought to be an aetiological factor in 5% (3 per 100 000) of patients in the current study compared to 6% in Rochester,21 5% in Geneva,17 and 2% in northern Sweden.15,16 Thus the distribution of risk factors in this study is similar to those in other incidence studies in which comparisons can be made. The syndrome classification combines family history, seizure semiology, and neurophysiology to classify patients into specific groups.3 When the syndrome classification is used, 58% of all incident unprovoked seizures fall into non-specific syndrome categories. These categories—without unequivocal generalised or focal features (3·2), unclassifiable, or isolated seizure (4·2)—provide little information about prognosis or management, which is the rationale behind the syndrome classification. Even after reclassifying single unprovoked seizures as more specific epilepsy syndromes, 38% of patients in the all-unprovokedseizures group and 27% of those in the epilepsy group fall into non-specific categories (table 4). This finding highlights the difficulties with use of the syndrome classification in population-based studies. Few studies have reported the incidence of epileptic syndromes in all age groups; none have provided data with syndrome and epidemiological classifications in one paper. Only the retrospective classification of cases from the Rochester study is roughly comparable, providing syndrome data by age.6 Because reclassification in the Rochester study was done some 10 years after the initial review, information that was not available at the time of the review may have affected
% generalised
% undetermined
% special syndrome
2·2
2·3
3·1
3·2
4·1
4·2
5·99 8·58
1·20 1·20
0·20 0·20
1·60 1·80
15·37 34·13
·· ··
41·32 ··
0·80 0·80
57 57
25·75 15·72 27·14 15·29 20·08
9·70 13·21 4·52 14·32 21·94
0·00 2·52 2·01 2·01 2·01
1·94 0·00 3·02 1·34 2·01
0·00 0·00 2·51 0·21 0·31
33·51 4·40 11·56 8·91 31·62
E E ·· ·· ··
13·93 43·19 21·11 47·68 ··
0·00 0·00 0·00 0·72 0·72
·· 46 63 ·· ··
25·17 33·76 13·48
26·53 34·39 29·23
10·20 5·73 27·36
2·04 2·55 3·84
0·34 3·82 2·56
2·72 3·18 0·39
26·19 14·65 17·03
·· ·· ··
·· ·· ··
1·36 ·· 1·38
33 52 ··
25·12 16·09
16·43 10·04
6·28 15·98
0·00 0·00
0·00 1·40
0·48 0·22
45·41 47·62
·· ··
·· ··
0·48 0·00
24 ··
1·1
1·2
1·3
3·19 5·59
14·77 25·15
15·57 22·36
1·23 3·98 1·01 2·47 6·59
16·93 16·35 26·63 7·05 14·73
5·44 1·27 4·72 5·80 8·64
2·1
% unclassifiable Incidence
*Includes epilepsy and single seizure. NGPSE=National General Practice Study of Epilepsy; E=excluded from analysis; CAROLE=Coordination Active du Reseau Observatoire Longitudinal de l’epilepsie.
Table 4: Comparison of syndrome classification across studies of incidence of all unprovoked seizures, epilepsy, and single seizures
http://neurology.thelancet.com Vol 4 October 2005
633
Articles
classification. The National General Practice Study of Epilepsy7 of new onset cases provides no incidence data and there are no data by age. Also, cases were classified retrospectively and it is unclear how information obtained after the diagnosis was handled.7 In the incidence study from Bordeaux, cases were identified via neurologists and through referrals to electroencephalography laboratories only,8 whereas in the current study they were identified through population surveillance. For this reason, in the Bordeaux study few cases were classified as syndrome 3·2 (undetermined whether generalised or focal epilepsy syndrome).8 The same pattern of classification occurred with the newonset cases included in the CAROLE study, which were cases identified by epileptologists throughout France.23 As in most other population-based studies, the specific syndromes in the current study were rare. The exception was benign rolandic epilepsy, which accounted for 20% of new-onset cases in children, albeit only 5% of all cases. The low occurrence of specific syndromes is not limited to population-based studies. In a survey of all cases seen at a tertiary epilepsy centre, only a small proportion were able to be classified into specific syndromes.24 There is a false perception that syndrome classification has high specificity for localisation. In our study, we could assign a specific anatomical area of onset in only three of 238 people with localisation related epilepsy, despite availability of imaging and electroencephalography findings in more than 80% of cases. Our research suggests that the syndrome classification has some flaws. In studies of children with incident seizures, the syndrome classification has been used successfully, particularly for cases with recurrent seizures only.25–27 The present study confirms the usefulness of the syndrome classification in children, but suggests that it is less useful in adults, who make up 75% of incident cases. Acknowledgments Funding was provided in part by the National Institute of Neurologic Disease and Stroke, USA (R01 NS 32663). Authors’ contributions WAH conceptualised and obtained the funding for this study; WAH, DH, PL, OK, and EO designed the study; GG and EO organised physicians in Iceland for this study and supervised the study; DH collected the data; GG reviewed the data; all authors analysed the data; and WAH, DH, PL, OK, and ES prepared the manuscript.
4
5
6
7
8
9
10
11 12 13
14
15
16
17
18
19 20 21
22
23
24
Conflicts of interest We have no conflicts of interest. References 1 Merlis JK. Proposal for an international classification of the epilepsies. Epilepsia 1970; 11: 114–119. 2 Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for classification of epilepsies and epileptic syndromes. Epilepsia 1985; 26: 268–78. 3 Commission on Classification and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989; 30: 389–99.
634
25
26 27
Engel J Jr. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE Task Force on Classification and Terminology. Epilepsia 2001; 42: 796–803. Commission on Epidemiology and Prognosis, International League against Epilepsy. Guidelines for epidemiologic studies on epilepsy. Epilepsia 1993: 34: 592–96. Zarrelli MM, Beghi E, Rocca WA, Hauser WA. Incidence of epileptic syndromes in Rochester, Minnesota: 1980–1984. Epilepsia 1999; 40: 1708–14. Manford M, Hart YS, Sanders JWAS, Shorvon SD. The National General Practice Study of Epilepsy: the syndromic classification of the International League Against Epilepsy applied to epilepsy in a general population. Arch Neurol 1992; 49: 801–08. Loiseau J, Loiseau P, Guyot M, et al. Survey of seizures in the French Southwest, 1: Incidence of epileptic syndromes. Epilepsia 1990; 31: 391–96. Ottman R, Lee JH, Hauser WA, et al. Reliability of seizure classification using a semistructured interview. Neurology 1993; 43: 2526–30. Commission on classification and terminology of the International League Against Epilepsy. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia 1981; 22: 489–501. Hauser WA, Annegers JF, Kurland LT. Prevalence of epilepsy in Rochester, Minnesota 1940–1980. Epilepsia 1991; 32: 429–44. Statistical Bureau of Iceland. http://www.hagstofa.is (accessed September 1, 2005). National Cancer Institute, US National Institutes of Health. Surveillance, epidemiology, and end results; standard populations—19 age groups. http://seer.cancer.gov/ stdpopulations/stdpop.19ages.html (accessed September 1, 2005). MacDonald BK, Cockerell OC, Sander JW, Shorvon SD. The incidence and lifetime prevalence of neurological disorders in a prospective community-based study in the UK. Brain 2000; 123: 665–76. Sidenvall R, Forsgren L, Blomquist HK, Heijbel J. A communitybased prospective incidence study of epileptic seizures in children. Acta Paediatrica 1993; 82: 60–65 Forsgren L, Bucht G, Eriksson S, Bergmark L. Incidence and clinical characterization of unprovoked seizures in adults: a prospective population-based study. Epilepsia 1996; 37: 224–29. Jallon P, Goumaz M, Haenggeli, Morabia A. Incidence of first epileptic seizures in the canto of Geneva, Switzerland. Epilepsia 1997; 38: 547–52. Olafsson E, Hauser WA, Ludvigsson P, Gudmundsson G. Incidence of epilepsy in rural Iceland: a population-based study. Epilepsia 1996; 37: 951–55. Joensen P. Prevalence, incidence, and classification of epilepsy in the Faroes. Acta Neurol Scand 1986; 74: 150–55. Granieri E, Rosati G, Tola R, et al. A descriptive study of epilepsy in the district of Copparo, Italy, 1964–1978. Epilepsia 1983; 24: 502–14. Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy and unprovoked seizures in Rochester Minnesota: 1935–1984. Epilepsia 1993; 34: 453–68. Annegers JF, Hauser WA, Lee JR, Rocca WA. Secular trends and birth cohort effects in unprovoked seizures: Rochester, Minnesota 1935–1984. Epilepsia 1995; 36: 575–79. Jallon P, Loiseau P, Loiseau J. Newly diagnosed unprovoked epileptic seizures: presentation at diagnosis in CAROLE study. Epilepsia 2001; 42: 464–75. Kellinghaus C, Loddenkemper T, Najm IM, et al. Specific epileptic syndromes are rare even in tertiary epilepsy centers: a patientoriented approach to epilepsy classification. Epilepsia 2004; 45: 268–75. Freitag CM, May TW, Pfafflin M, Konig S, Rating D. Incidence of epilepsies and epileptic syndromes in children and adolescents: a population-based prospective study in Germany. Epilepsia 2001; 42: 979–85. Berg AT, Shinnar S, Levy SR, Testa FM. Newly diagnosed epilepsy in children: presentation at diagnosis. Epilepsia 1999; 40: 445–52. Onsurbe Ramirez I, Hernandez Rodriguez M, Aparicio Meix JM. Carrascosa Romero incidence of epilepsy and epileptic syndromes in children in the province of Albacete. An Esp Pediatr 1999; 51: 154–58.
http://neurology.thelancet.com Vol 4 October 2005