BACKGROUND AND OBJECTIVES:

Seizure freedom is the optimal response to antiepileptic treatment. In previous studies, it has been shown that between 61% and 71% of children with epilepsy achieve seizure freedom, whereas 7% to 20% have drug-resistant epilepsy. The definition of drug resistance has not been consistent across studies, and there is a lack of contemporary population-based data. We used data from a large nationwide child cohort to provide such information, implementing the current standard definition of drug resistance.

METHODS:

The study was based on the Norwegian Mother and Child Cohort Study. Potential epilepsy cases were identified through registry linkages and parental questionnaires. Medical record reviews and parental interviews were used to collect clinical information and to classify seizures, epilepsies, and etiologies.

RESULTS:

The cohort included 112 745 eligible children aged 3 to 13 years (median age 7 years) at end of follow-up. Of these, 600 were epilepsy cases with at least 1 year of follow-up since epilepsy onset (median follow-up time: 5.8 years). There were 178 (30%) who had developed drug-resistant epilepsy, 353 (59%) who had been seizure free for ≥1 year, and 69 (12%) with intermediate seizure outcomes. Having an identified cause of epilepsy (genetic, structural, metabolic, or infectious) was associated with unsatisfactory seizure outcome (48% drug resistance) and influenced the relative risk associated with other prognostic factors. Sociodemographic characteristics were not associated with short-term seizure outcomes.

CONCLUSIONS:

Drug resistance occurs in 3 out of 10 children with epilepsy, whereas 6 out of 10 become seizure free. Having an identified cause of epilepsy is associated with poor response to treatment.

What’s Known on This Subject:

Between 61% and 71% of children with epilepsy achieve remission, whereas 7% to 20% have drug-resistant epilepsy. Previous studies were conducted decades ago with divergent definitions of remission and drug resistance. It is unknown whether the findings are applicable to current pediatric practice.

What This Study Adds:

In this population-based study of epilepsy, 30% of affected children had drug-resistant epilepsy, whereas 59% achieved ≥1 year of seizure freedom. Having an identified cause of epilepsy (genetic, structural, metabolic, or infectious) was associated with drug resistance.

Freedom from seizures is the optimal response to antiepileptic treatment. In studies, it has been shown that between 61% and 71% of children with epilepsy (CWE) achieve permanent remission from seizures.1,4 Childhood epilepsy often has a fluctuating course, and approximately one-third of CWE experience alternating periods of seizure freedom and seizure relapses.5,6 Between 7% and 20% do not become seizure free, despite adequate treatment.2,4,7 

Early onset of epilepsy is associated with worse outcomes. In a longitudinal study of CWE with onset before age 36 months, it was found that 35% did not become seizure free.8 Early-onset epilepsies frequently have identified structural, metabolic, or genetic causes, and such epilepsies are more likely to have poorer treatment responses than epilepsies with nonidentified causes.3,7,8 However, in a study of epilepsy with onset in infancy (age <1 year), it was found that 58% of affected children were in remission by age 24 months and that age of onset did not affect the outcome when etiology was accounted for.9 

Other characteristics associated with poor treatment responses are abnormal findings on neurological examinations, a history of febrile seizures, and pathologic findings in EEGs.4,8 Children from families with low socioeconomic status have a higher baseline risk of epilepsy,10 but socioeconomic factors do not appear to influence seizure outcomes in countries with universal access to health care.2 

Different terms have been used to describe epilepsies in which seizure freedom is not obtained: drug-resistant, pharmacoresistant, intractable, or refractory. A consensus proposal from the International League Against Epilepsy (ILAE) recommends the term “drug-resistant epilepsy” (DRE) and defines this as failure to achieve sustained seizure freedom despite adequate trials of at least 2 antiepileptic drugs used in monotherapy or in combination.11 

In most studies of childhood epilepsy, researchers include patients who were diagnosed decades ago, when the range of diagnostic procedures and antiepileptic treatments of today were not available. All previous studies have been based on the old classification systems for seizures and epilepsies, and definitions of drug resistance and seizure freedom have varied. Consequently, it is not known whether previous findings are applicable to current pediatric practice.

In this study, we have used information from a large population-based child cohort to provide contemporary data on short-term seizure outcomes in childhood epilepsy. We have used the revised classification systems published by the ILAE in 201712,13 and the current ILAE definition of DRE.11 We investigated seizure outcomes across prognostic factors that are available and relevant to clinicians treating CWE: sociodemographic characteristics, epilepsy risk factors, clinical characteristics, investigation results, seizure types, epilepsy types, and etiological categories. We expected that the cause of epilepsy would be strongly associated with seizure outcomes and partly or fully account for associations observed for other prognostic factors.

The study was based on the Norwegian Mother and Child Cohort Study (MoBa), a nationwide cohort study of children born from 1999 to 2009.14 Within MoBa, we established a substudy of epilepsy, which was the Epilepsy in Young Children Study (EPYC).15 CWE were identified through linkages to the Norwegian Patient Registry and parental report of epilepsy in sequential MoBa questionnaires. Clinical information was collected through medical record reviews and clinical telephone interviews with the parents. Seizures, epilepsies, and etiologies were classified by 2 child epileptologists (K.M.A. and R.C.). Details of the data collection have been described in previous publications.15,16 

DRE was defined as seizures within the last year of follow-up despite adequate trials of at least 2 antiepileptic drugs, in concordance with the ILAE recommendation.11 Seizure freedom was defined as being without seizures for ≥1 year at the end of follow-up (regardless of whether antiepileptic drugs were used). CWE with seizures within the last year of follow-up who had not completed trials of at least 2 drugs were defined as having intermediate seizure outcome. Children with <1 year of follow-up since epilepsy onset were not included in the case definition of epilepsy.

Statistical analyses were conducted by using IBM SPSS Statistics 22 (IBM Corp, Armonk, NY). We tested differences across seizure outcome categories by Pearson’s χ2 tests, with 2-sided P values. Log-binomial regression analyses were used to calculate relative risks (RRs) of DRE associated with each prognostic factor. Because of the large number of comparisons, we used an α of .01 for P values and 99% confidence intervals (CIs) for RR estimates.

MoBa holds a license from the Norwegian Data Protection Authority. Participation is based on informed consent by mothers and fathers. The consent includes permission to registry linkages and medical record reviews. The EPYC has approval from the Regional Committee for Medical and Health Research Ethics for Southeast Norway. Participation in the EPYC telephone interviews was based on an additional informed consent.

At the end of registry follow-up on December 31, 2012, MoBa included 112 745 children who were live born, had valid personal identification numbers, and were still residing in Norway. The age range at end of follow-up was 3 to 13 years with a median age of 7 years. There were 606 epilepsy cases in total, of which, 13 (2%) had died during follow-up. The median age of onset of epilepsy was 3 years. Among the cases, 600 had ≥1 year of follow-up since epilepsy onset (median: 6 years; range: 1–13 years). At the end of follow-up, 178 (30%) of these 600 CWE had DRE, 353 (59%) had been seizure free for ≥1 year, and 69 (12%) had intermediate seizure outcomes. The number of antiepileptic drugs tried out ranged from 0 to 16 per child, with a median of 2. A total of 91 CWE (15%) had never used antiepileptic drugs. Nonmedical treatment had been used by a total of 40 CWE (7%), with vagal nerve stimulator in 11 (2%), epilepsy surgery in 5 (1%), and ketogenic diet in 31 (5%). Of the 40 CWE who had tried nonmedical treatment, 34 (85%) had DRE at end of follow-up.

In Table 1, we compare the distributions of sociodemographic characteristics between CWE and noncases and across seizure outcome categories for CWE. In the comparison of CWE with noncases, the CWE had somewhat higher proportions of parents with <12 years of education, that is, parents who had not completed high school (P = .08 for maternal education and P = .004 for paternal education). There were no differences between CWE and noncases with regards to maternal living status, pregnancy planning, or maternal and paternal age. The comparisons across seizure outcome categories in CWE revealed that none of the sociodemographic characteristics were associated with seizure outcomes.

TABLE 1

Epilepsy Risk and Short-term Seizure Outcome by Sociodemographic Characteristics

Noncases N = 112 145All CWEan = 600PbDREcn = 178 (30%)Intermediate Outcomecn = 69 (12%)≥1 y Seizure Free, n = 353 (59%)P
n%n%n%n%n%
Maternal education, y             
 ≤12 35 078 31 213 36 .08 64 36 29 42 120 34 .67 
 ≥13 64 447 57 322 54 92 52 34 49 196 56 
 Missing data 12 620 11 65 11 22 12 37 10 
Paternal education, y             
 ≤12 49 484 44 303 51 .004 87 49 34 49 182 52 .66 
 ≥13 49 202 44 225 38 65 37 29 42 131 37 
 Missing data 13 459 12 72 12 26 15 40 11 
Maternal living status             
 Married or partner 97 761 87 513 86 .11 147 83 58 84 308 87 .65 
 Single 3390 27 10 13 
 Missing data 10 994 10 60 10 21 12 10 32 
Pregnancy planning             
 Planned 81 021 72 430 72 .85 121 68 46 67 263 75 .43 
 Unplanned 19 420 17 109 18 36 20 16 23 57 16 
 Missing data 11 704 10 61 10 21 12 10 33 
Maternal age, y             
 <25 13 058 12 84 14 .36 26 15 10 14 48 14 .92 
 25–34 79 261 71 412 69 126 71 46 67 240 68 
 ≥35 19 646 18 103 17 26 15 13 19 64 18 
 Missing data 180 0.2 <5 <1 — <5 <1 <5 <1 <5 <1 — 
Paternal age, y             
 <25 5801 31 .35 10 10 14 .15 
 25–39 94 462 84 512 85 155 87 51 74 306 87 
 ≥40 11 377 10 52 12 10 14 30 
 Missing data 505 0.5 0.8 — <5 <1 <5 <1 <5 <1 — 
Noncases N = 112 145All CWEan = 600PbDREcn = 178 (30%)Intermediate Outcomecn = 69 (12%)≥1 y Seizure Free, n = 353 (59%)P
n%n%n%n%n%
Maternal education, y             
 ≤12 35 078 31 213 36 .08 64 36 29 42 120 34 .67 
 ≥13 64 447 57 322 54 92 52 34 49 196 56 
 Missing data 12 620 11 65 11 22 12 37 10 
Paternal education, y             
 ≤12 49 484 44 303 51 .004 87 49 34 49 182 52 .66 
 ≥13 49 202 44 225 38 65 37 29 42 131 37 
 Missing data 13 459 12 72 12 26 15 40 11 
Maternal living status             
 Married or partner 97 761 87 513 86 .11 147 83 58 84 308 87 .65 
 Single 3390 27 10 13 
 Missing data 10 994 10 60 10 21 12 10 32 
Pregnancy planning             
 Planned 81 021 72 430 72 .85 121 68 46 67 263 75 .43 
 Unplanned 19 420 17 109 18 36 20 16 23 57 16 
 Missing data 11 704 10 61 10 21 12 10 33 
Maternal age, y             
 <25 13 058 12 84 14 .36 26 15 10 14 48 14 .92 
 25–34 79 261 71 412 69 126 71 46 67 240 68 
 ≥35 19 646 18 103 17 26 15 13 19 64 18 
 Missing data 180 0.2 <5 <1 — <5 <1 <5 <1 <5 <1 — 
Paternal age, y             
 <25 5801 31 .35 10 10 14 .15 
 25–39 94 462 84 512 85 155 87 51 74 306 87 
 ≥40 11 377 10 52 12 10 14 30 
 Missing data 505 0.5 0.8 — <5 <1 <5 <1 <5 <1 — 

—, not applicable.

a

CWE, according to the new definition (ILAE 2014) and ≥1 y of follow-up since onset.

b

P values calculated with Pearson’s χ2 tests.

c

DRE, defined as failure to achieve ≥1 y of seizure freedom at end of follow-up despite trial of ≥2 antiepileptic drugs in monotherapy or in combination. The intermediate category includes seizure outcomes in between DRE and ≥1 y of seizure freedom.

In Table 2, we have shown the distribution of short-term seizure outcomes across categories of epilepsy risk factors, seizure characteristics, and clinical and/or EEG characteristics, with crude RR estimates for DRE. CWE with a history of preterm birth and low Apgar scores had a somewhat increased risk of DRE, but testing across seizure outcome categories provided P values >.01. The child’s sex, family history of epilepsy, and history of febrile seizures were not associated with DRE. The other characteristics (multiple seizure types, status epilepticus, seizure precipitants, infancy onset, additional neurologic or developmental difficulties, abnormal neurologic examinations, and abnormal EEG findings) were all strongly associated with DRE, with P values ≤.001 for all comparisons across seizure outcome categories.

TABLE 2

Short-term Seizure Outcome by Risk Factors, Clinical Characteristics, and EEG Findings

All CWEan = 600DREbn = 178 (30%)Intermediate Outcomebn = 69 (12%)≥1 y Seizure Free n = 353 (59%)Pc
n%n%RR for DRE (99% CI)n%n%
Epilepsy risk factors 
 Preterm birth (gestational age <37 wk) 
  Yes 76 13 31 17 1.5 (1.0–2.2) 10 14 35 10 .04 
  No 520 87 146 82 1 (reference) 58 84 316 90 
  Missing — — — — — — — — 
 5-min Apgar score <7 
  Yes 34 16 1.6 (1.0–2.7) 16 .07 
  No 564 94 162 91 1 (reference) 67 97 335 95 
  Missing ∼0 — — — — — — — — 
 Sex 
  Male 324 54 85 48 0.8 (0.6–1.1) 40 58 199 56 .13 
  Female 276 46 93 52 1 (reference) 29 42 154 44 
 Family history of epilepsy 
  Yes 154 26 38 21 0.8 (0.5–1.2) 24 35 92 26 .09 
  No 446 74 140 79 1 (reference) 45 65 261 74 
 History of febrile seizures 
  Yes 118 20 30 17 0.8 (0.5–1.3) 16 23 72 20 .46 
  No 482 80 148 83 1 (reference) 53 77 281 80 
 
Seizure characteristics 
 No. seizure types 
  ≥3 152 25 106 60 7.6 (4.5–12.8) 14 20 32 <.001 
  2 187 31 48 27 2.8 (1.5–5.1) 26 38 113 32 
  1 261 44 24 13 1 (reference) 29 42 208 59 
 History of status epilepticus 
  Yes 140 23 68 38 2.0 (1.5–2.8) 14 20 58 16 <.001 
  No 460 77 110 62 1 (reference) 55 80 295 84 
 History of seizure precipitants 
  Yes 289 48 104 58 1.5 (1.1–2.1) 39 57 146 41 <.001 
  No 311 52 74 42 1 (reference) 30 43 207 59 
 
Clinical and/or EEG characteristics 
 Age of epilepsy onset, y 
  <1 161 27 68 38 1.8 (1.2–2.9) 15 22 78 22 .001 
  1–4 272 45 72 40 1.2 (0.7–1.8) 30 43 170 48 
  ≥5 166 28 38 21 1 (reference) 24 35 104 29 
  Missing ∼0 — — — — — — — — 
 Additional neurologic or developmental difficultiesd 
  Yes 321 54 142 80 3.4 (2.2–5.3) 36 52 143 41 <.001 
  No 279 47 36 20 1 (reference) 33 48 210 59 
 Neurologic examination 
  Abnormal 242 40 120 67 3.1 (2.2–4.4) 29 42 93 26 <.001 
  Normal 358 60 58 33 1 (reference) 40 58 260 74 
 EEG: epileptic activity 
  Yes 495 83 169 95 4.0 (1.7–9.2) 51 74 275 78 <.001 
  No 105 18 1 (reference) 18 26 78 22 
 EEG: nonepileptic pathology 
  Yes 197 33 99 56 2.6 (1.7–3.5) 19 28 79 22 <.001 
  No 403 67 79 44 1 (reference) 50 72 274 78 
All CWEan = 600DREbn = 178 (30%)Intermediate Outcomebn = 69 (12%)≥1 y Seizure Free n = 353 (59%)Pc
n%n%RR for DRE (99% CI)n%n%
Epilepsy risk factors 
 Preterm birth (gestational age <37 wk) 
  Yes 76 13 31 17 1.5 (1.0–2.2) 10 14 35 10 .04 
  No 520 87 146 82 1 (reference) 58 84 316 90 
  Missing — — — — — — — — 
 5-min Apgar score <7 
  Yes 34 16 1.6 (1.0–2.7) 16 .07 
  No 564 94 162 91 1 (reference) 67 97 335 95 
  Missing ∼0 — — — — — — — — 
 Sex 
  Male 324 54 85 48 0.8 (0.6–1.1) 40 58 199 56 .13 
  Female 276 46 93 52 1 (reference) 29 42 154 44 
 Family history of epilepsy 
  Yes 154 26 38 21 0.8 (0.5–1.2) 24 35 92 26 .09 
  No 446 74 140 79 1 (reference) 45 65 261 74 
 History of febrile seizures 
  Yes 118 20 30 17 0.8 (0.5–1.3) 16 23 72 20 .46 
  No 482 80 148 83 1 (reference) 53 77 281 80 
 
Seizure characteristics 
 No. seizure types 
  ≥3 152 25 106 60 7.6 (4.5–12.8) 14 20 32 <.001 
  2 187 31 48 27 2.8 (1.5–5.1) 26 38 113 32 
  1 261 44 24 13 1 (reference) 29 42 208 59 
 History of status epilepticus 
  Yes 140 23 68 38 2.0 (1.5–2.8) 14 20 58 16 <.001 
  No 460 77 110 62 1 (reference) 55 80 295 84 
 History of seizure precipitants 
  Yes 289 48 104 58 1.5 (1.1–2.1) 39 57 146 41 <.001 
  No 311 52 74 42 1 (reference) 30 43 207 59 
 
Clinical and/or EEG characteristics 
 Age of epilepsy onset, y 
  <1 161 27 68 38 1.8 (1.2–2.9) 15 22 78 22 .001 
  1–4 272 45 72 40 1.2 (0.7–1.8) 30 43 170 48 
  ≥5 166 28 38 21 1 (reference) 24 35 104 29 
  Missing ∼0 — — — — — — — — 
 Additional neurologic or developmental difficultiesd 
  Yes 321 54 142 80 3.4 (2.2–5.3) 36 52 143 41 <.001 
  No 279 47 36 20 1 (reference) 33 48 210 59 
 Neurologic examination 
  Abnormal 242 40 120 67 3.1 (2.2–4.4) 29 42 93 26 <.001 
  Normal 358 60 58 33 1 (reference) 40 58 260 74 
 EEG: epileptic activity 
  Yes 495 83 169 95 4.0 (1.7–9.2) 51 74 275 78 <.001 
  No 105 18 1 (reference) 18 26 78 22 
 EEG: nonepileptic pathology 
  Yes 197 33 99 56 2.6 (1.7–3.5) 19 28 79 22 <.001 
  No 403 67 79 44 1 (reference) 50 72 274 78 

—, not applicable.

a

CWE, according to the new definition (ILAE 2014) and ≥1 y of follow-up since onset.

b

DRE, defined as failure to achieve ≥1 y of seizure freedom at end of follow-up despite trial of ≥2 antiepileptic drugs in monotherapy or in combination. The intermediate category includes seizure outcomes in between DRE and ≥1 y of seizure freedom.

c

P values calculated with Pearson’s χ2 tests.

d

Diagnosed neurologic and developmental disorders or reported neurologic and developmental problems.

In Table 3, we show the distribution of seizure outcomes by the ILAE 2017 classifications of seizure type, epilepsy type, and epilepsy etiology. The crude RR of DRE was >1 for all seizure types except for typical absence seizures and unclassified seizures. It was particularly high for those with tonic seizures (RR = 3.0), myoclonic seizures (RR = 3.5), atonic seizures (RR = 2.4), and epileptic spasms (RR = 2.4). In the analyses of epilepsy types, we compared mode of onset categories, with focal onset as the reference category. CWE with generalized onset had similar seizure outcomes as those with focal onset (RR = 1.0), whereas those with combined focal and generalized onset had a substantially increased risk of DRE, with a crude RR of 2.9 compared with focal onset.

TABLE 3

Short-term Seizure Outcomes by ILAE 2017 Classification of Seizures, Epilepsies, and Etiologies

All CWEan = 600DREbn = 178 (30%)Intermediate Outcomebn = 69 (12%)≥1 y Seizure Free n = 353 (59%)Pc
n%n%RR for DRE (99% CI)n%n%
Seizure classification 
 Focal seizures, any typed 
  Yes 413 69 143 80 1.8 (1.2–2.8) 51 74 219 62 <.001 
  No 187 31 35 20 1 (reference) 18 26 134 38 
 Generalized seizurese 
  Tonic-clonic seizures 
   Yes 111 19 46 26 1.5 (1.1–2.2) 13 56 16 .009 
   No 489 82 132 74 1 (reference) 60 87 297 84 
  Tonic seizures 
   Yes 80 13 56 31 3.0 (2.3–3.9) 20 <.001 
   No 520 87 122 69 1 (reference) 65 94 333 94 
  Myoclonic seizures 
   Yes 82 14 64 36 3.5 (2.7–4.6) 13 <.001 
   No 518 86 114 64 1 (reference) 64 93 340 96 
  Atonic seizures 
   Yes 39 25 14 2.4 (1.6–3.4) 11 <.001 
   No 561 94 153 86 1 (reference) 66 96 342 97 
  Typical absence seizures 
   Yes 62 10 20 11 1.1 (0.7–1.8) 38 11 .42 
   No 538 90 158 89 1 (reference) 65 94 315 89 
  Atypical or other absence seizures 
   Yes 54 25 14 1.7 (1.1–2.5) 25 .02 
   No 546 91 153 86 1 (reference) 65 94 328 93 
 Unclassifiable seizures 
  Epileptic spasms 
   Yes 60 10 38 21 2.4 (1.8–3.3) 17 <.001 
   No 540 90 140 79 1 (reference) 64 93 336 95 
  Other unclassifiable seizures 
   Yes 87 15 26 15 1.0 (0.6–1.6) 11 16 50 14 .93 
   No 513 86 152 85 1 (reference) 58 84 303 86 
 
Epilepsy mode of onset 
 Focal 304 51 70 39 1 (reference) 46 67 188 53 <.001 
 Generalized 144 24 34 19 1.0 (0.6–1.6) 11 16 99 28 
 Combined generalized and focal 110 18 73 41 2.9 (2.1–4.0) 32 
 Unclassifiable 42 0.1 (0.0–1.3) 10 34 
 
Epilepsy etiology 
 Identified causef 196 33 95 53 2.7 (1.8–3.9) 22 32 79 22 <.001 
 Presumed genetic cause 147 25 36 20 1.3 (0.8–2.2) 13 102 29 
 Unknown cause 257 43 47 26 1 (reference) 38 55 172 49 
All CWEan = 600DREbn = 178 (30%)Intermediate Outcomebn = 69 (12%)≥1 y Seizure Free n = 353 (59%)Pc
n%n%RR for DRE (99% CI)n%n%
Seizure classification 
 Focal seizures, any typed 
  Yes 413 69 143 80 1.8 (1.2–2.8) 51 74 219 62 <.001 
  No 187 31 35 20 1 (reference) 18 26 134 38 
 Generalized seizurese 
  Tonic-clonic seizures 
   Yes 111 19 46 26 1.5 (1.1–2.2) 13 56 16 .009 
   No 489 82 132 74 1 (reference) 60 87 297 84 
  Tonic seizures 
   Yes 80 13 56 31 3.0 (2.3–3.9) 20 <.001 
   No 520 87 122 69 1 (reference) 65 94 333 94 
  Myoclonic seizures 
   Yes 82 14 64 36 3.5 (2.7–4.6) 13 <.001 
   No 518 86 114 64 1 (reference) 64 93 340 96 
  Atonic seizures 
   Yes 39 25 14 2.4 (1.6–3.4) 11 <.001 
   No 561 94 153 86 1 (reference) 66 96 342 97 
  Typical absence seizures 
   Yes 62 10 20 11 1.1 (0.7–1.8) 38 11 .42 
   No 538 90 158 89 1 (reference) 65 94 315 89 
  Atypical or other absence seizures 
   Yes 54 25 14 1.7 (1.1–2.5) 25 .02 
   No 546 91 153 86 1 (reference) 65 94 328 93 
 Unclassifiable seizures 
  Epileptic spasms 
   Yes 60 10 38 21 2.4 (1.8–3.3) 17 <.001 
   No 540 90 140 79 1 (reference) 64 93 336 95 
  Other unclassifiable seizures 
   Yes 87 15 26 15 1.0 (0.6–1.6) 11 16 50 14 .93 
   No 513 86 152 85 1 (reference) 58 84 303 86 
 
Epilepsy mode of onset 
 Focal 304 51 70 39 1 (reference) 46 67 188 53 <.001 
 Generalized 144 24 34 19 1.0 (0.6–1.6) 11 16 99 28 
 Combined generalized and focal 110 18 73 41 2.9 (2.1–4.0) 32 
 Unclassifiable 42 0.1 (0.0–1.3) 10 34 
 
Epilepsy etiology 
 Identified causef 196 33 95 53 2.7 (1.8–3.9) 22 32 79 22 <.001 
 Presumed genetic cause 147 25 36 20 1.3 (0.8–2.2) 13 102 29 
 Unknown cause 257 43 47 26 1 (reference) 38 55 172 49 
a

CWE, according to the new definition (ILAE 2014) and ≥1 y of follow-up since onset.

b

DRE, defined as failure to achieve ≥1 y of seizure freedom at end of follow-up despite trial of ≥2 antiepileptic drugs in monotherapy or in combination. The intermediate category includes seizure outcomes in between DRE and ≥1 y of seizure freedom.

c

P values were calculated with Pearson’s χ2 tests.

d

Analyzed as 1 category because there were no differences in seizure outcomes between subcategories of focal seizures.

e

There were 5 CWE with myoclonic-atonic seizures and none with clonic and myoclonic-tonic-clonic seizures. These seizure categories are not included in the table.

f

Includes structural, genetic, metabolic, and infectious causes. There were large degrees of overlap between these categories. No CWE had immunologic etiology.

In the analyses by etiological category for Table 3, we started by examining individual etiological categories: structural, genetic, metabolic, infectious, and unknown causes. Among CWE with structural causes (n = 156), 47% had DRE, whereas among CWE with identified and presumed genetic causes (n = 204), 33% had DRE. Among the CWE with identified genetic causes (n = 56), the proportion with DRE was 56%, which was much higher than the proportion of 24% with DRE among CWE with presumed genetic causes (n = 147) (P < .001). The numbers of CWE with metabolic (n = 5) and infectious (n = 12) causes were too low to reliably calculate proportions of DRE. Immunologic causes had not been identified for any of the CWE, and few had reports of immunologic investigations. Among CWE with identified etiology, there were many who fit into more than 1 category. Given this etiological overlap, as well as the similar seizure outcomes across categories of identified causes, we merged all identified causes into 1 category in the analyses shown in Table 3. This category was compared with presumed genetic causes and unknown causes, with unknown causes as the reference category. Having an identified cause was positively associated with DRE (RR = 2.7). CWE with presumed genetic cause had a slightly higher risk of DRE than CWE with unknown cause (RR = 1.3), but the difference was not statistically significant.

In absolute terms, the risk of DRE was 48% in CWE with an identified cause and 21% in CWE with a nonidentified cause. These differences between CWE with and without identified causes confirmed our a priori expectation that the cause of epilepsy would be strongly associated with short-term seizure outcomes. We also found that the RR estimates for other prognostic factors often differed between the 2 main etiological categories. Because of these differences, we chose to stratify regression analyses by CWE with identified cause and CWE without identified cause. The results of regression analyses are shown in Tables 4 and 5.

TABLE 4

RR for DRE by Risk Factors, Clinical Characteristics, and EEG Findings

CWEa With Identified Causeb (n = 196) Overall Proportion of DRE: 48%CWE Without Identified Cause (n = 404) Overall Proportion of DRE: 21%
Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)
Epilepsy risk factors 
 Preterm birth (gestational age <37 wk) 
  Yes 1.2 (0.8–1.9) 1.2 (0.8–1.8) 1.0 (0.4–2.5) 1.0 (0.4–2.4) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 5-min apgar score <7 
  Yes 1.3 (0.8–2.0) 1.4 (0.9–2.2) Not estimatede Not estimatede 
  No 1 (reference) 1 (reference) — — 
 Sex 
  Male 1.0 (0.7–1.4) 0.9 (0.7–1.4) 0.6 (0.3–1.0) 0.6 (0.4–1.0) 
  Female 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Family history of epilepsy 
  Yes 0.7 (0.4–1.3) 0.7 (0.4–1.3) 1.0 (0.6–1.8) 1.0 (0.6–1.7) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 History of febrile seizures 
  Yes 1.1 (0.7–1.7) 1.3 (0.8–2.0) 0.6 (0.2–1.2) 0.6 (0.3–1.3) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
     
 Crude RR for DREc Adjustedf RR for DRE  Crude RR for DREc Adjustedf RR for DRE 
 (99% CI) (99% CI) (99% CI) (99% CI) 
Seizure and/or epilepsy characteristics 
 No. of seizure types 
  ≥3 3.6 (2.2–6.0) 2.9 (1.6–5.2) 4.0 (2.5–6.2) 3.5 (2.1–5.9) 
  1–2 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 History of status epilepticus 
  Yes 1.5 (1.0–2 1) 1.1 (0.9–1.5) 2.0 (1.2–3.3) 1.2 (0.7–1.9) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 History of seizure precipitants 
  Yes 1.2 (0.8–1.8) 1.0 (0.8–1.2) 1.9 (1.1–3.1) 1.7 (1.0–2.7) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Mode of seizure onset 
  Combined generalized and focal 2.5 (1.7–3.6) 1.4 (1.0–1.9) 2.6 (1.5–4.3) 1.0 (0.6–1.8) 
  Focal or generalized or unclassifiable 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
     
 Crude RR for DREc Adjustedg RR for DRE Crude RR for DREc Adjustedg RR for DRE 
 (99% CI) (99% CI) (99% CI) (99% CI) 
Clinical and/or EEG characteristics 
 Age of epilepsy onset, y 
  <1 1.4 (1.0–2.0) 1.2 (0.8–1.6) 1.2 (0.7–2.3) 1.0 (0.5–1.8) 
  ≥1 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Additional neurologic or developmental difficultiesh 
  Yes 2.9 (0.7–11.4) 1.2 (0.2–6.1) 2.8 (1.7–4.7) 2.0 (1.1–3.8) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Neurologic examination 
  Abnormal 2.9 (1.2–6.6) 2.3 (0.8–6.5) 2.4 (1.5–3.9) 1.3 (0.7–2.4) 
  Normal 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 EEG: epileptic activity 
  Yes 4.7 (1.1–19.2) 3.9 (1.0–15.8) 3.1 (1.1–8.9) 2.5 (0.9–7.0) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 EEG: nonepileptic pathology 
  Yes 1.9 (1.1–3.0) 1.4 (0.9–2.3) 2.2 (1.3–3.6) 1.4 (0.8–2.4) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
CWEa With Identified Causeb (n = 196) Overall Proportion of DRE: 48%CWE Without Identified Cause (n = 404) Overall Proportion of DRE: 21%
Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)
Epilepsy risk factors 
 Preterm birth (gestational age <37 wk) 
  Yes 1.2 (0.8–1.9) 1.2 (0.8–1.8) 1.0 (0.4–2.5) 1.0 (0.4–2.4) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 5-min apgar score <7 
  Yes 1.3 (0.8–2.0) 1.4 (0.9–2.2) Not estimatede Not estimatede 
  No 1 (reference) 1 (reference) — — 
 Sex 
  Male 1.0 (0.7–1.4) 0.9 (0.7–1.4) 0.6 (0.3–1.0) 0.6 (0.4–1.0) 
  Female 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Family history of epilepsy 
  Yes 0.7 (0.4–1.3) 0.7 (0.4–1.3) 1.0 (0.6–1.8) 1.0 (0.6–1.7) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 History of febrile seizures 
  Yes 1.1 (0.7–1.7) 1.3 (0.8–2.0) 0.6 (0.2–1.2) 0.6 (0.3–1.3) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
     
 Crude RR for DREc Adjustedf RR for DRE  Crude RR for DREc Adjustedf RR for DRE 
 (99% CI) (99% CI) (99% CI) (99% CI) 
Seizure and/or epilepsy characteristics 
 No. of seizure types 
  ≥3 3.6 (2.2–6.0) 2.9 (1.6–5.2) 4.0 (2.5–6.2) 3.5 (2.1–5.9) 
  1–2 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 History of status epilepticus 
  Yes 1.5 (1.0–2 1) 1.1 (0.9–1.5) 2.0 (1.2–3.3) 1.2 (0.7–1.9) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 History of seizure precipitants 
  Yes 1.2 (0.8–1.8) 1.0 (0.8–1.2) 1.9 (1.1–3.1) 1.7 (1.0–2.7) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Mode of seizure onset 
  Combined generalized and focal 2.5 (1.7–3.6) 1.4 (1.0–1.9) 2.6 (1.5–4.3) 1.0 (0.6–1.8) 
  Focal or generalized or unclassifiable 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
     
 Crude RR for DREc Adjustedg RR for DRE Crude RR for DREc Adjustedg RR for DRE 
 (99% CI) (99% CI) (99% CI) (99% CI) 
Clinical and/or EEG characteristics 
 Age of epilepsy onset, y 
  <1 1.4 (1.0–2.0) 1.2 (0.8–1.6) 1.2 (0.7–2.3) 1.0 (0.5–1.8) 
  ≥1 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Additional neurologic or developmental difficultiesh 
  Yes 2.9 (0.7–11.4) 1.2 (0.2–6.1) 2.8 (1.7–4.7) 2.0 (1.1–3.8) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Neurologic examination 
  Abnormal 2.9 (1.2–6.6) 2.3 (0.8–6.5) 2.4 (1.5–3.9) 1.3 (0.7–2.4) 
  Normal 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 EEG: epileptic activity 
  Yes 4.7 (1.1–19.2) 3.9 (1.0–15.8) 3.1 (1.1–8.9) 2.5 (0.9–7.0) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 EEG: nonepileptic pathology 
  Yes 1.9 (1.1–3.0) 1.4 (0.9–2.3) 2.2 (1.3–3.6) 1.4 (0.8–2.4) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 

—, not applicable.

a

CWE, according to the new definition (ILAE 2014) and at least 1 y of follow-up since onset.

b

Includes structural, genetic, metabolic, and infectious causes. There were large degrees of overlap between these categories. No CWE had immunologic etiology.

c

DRE, defined as failure to achieve ≥1 y of seizure freedom at end of follow-up despite trial of ≥2 antiepileptic drugs in monotherapy or in combination. The intermediate category includes seizure outcomes in between DRE and ≥1 y of seizure freedom.

d

Adjusted for all other epilepsy risk factors.

e

Not estimated because no CWE without identified cause had a 5-min Apgar score <7.

f

Adjusted for all other seizure and/or epilepsy characteristics.

g

Adjusted for all other clinical and/or EEG characteristics.

h

Diagnosed neurologic and/or developmental disorders or reported neurologic and/or developmental problems.

TABLE 5

RR for DRE by ILAE 2017 Seizure Classification

CWEa With Identified Causeb (n = 196) Overall Proportion of DRE: 48%CWE Without Identified Cause (n = 404) Overall Proportion of DRE: 21%
Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)
Seizure classification 
 Focal seizures, any type 
  Yes 2.0 (1.0–4.1) 1.0 (0.6–1.9) 1.4 (0.8–2.4) 1.1 (0.7–1.9) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Generalized seizuresd 
  Tonic-clonic seizures 
   Yes 1.6 (1.1–2.3) 1.2 (0.9–1.5) 1.1 (0.6–2.1) 0.8 (0.4–1.5) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Tonic seizures 
   Yes 2.2 (1.6–3.1) 1.3 (1.0–1.7) 2.9 (1.7–4.9) 1.6 (1.0–2.7) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Myoclonic seizures 
   Yes 2.5 (1.8–3.5) 1.5 (1.1–2.0) 3.7 (2.3–5.9) 1.5 (0.8–2.7) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Atonic seizures 
   Yes 1.7 (1.2–2.5) 1.2 (0.9–1.6) 1.9 (0.8–4.9) 1.0 (0.4–2.3) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Typical absence seizures 
   Yes Not estimated Not estimated 1.2 (0.6–2.4) 1.4 (0.8–2.4) 
   No — — 1 (reference) 1 (reference) 
  Atypical or other absence seizures 
   Yes 1.2 (0.7–2.0) 1.1 (0.7–1.6) 2.2 (1.2–4.0) 1.4 (0.8–2.3) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Unclassifiable seizures 
  Epileptic spasms 
   Yes 1.5 (1.0–2.2) 1.0 (0.8–1.4) 3.1 (1.8–5.4) 1.8 (1.2–2.8) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Other unclassifiable seizures 
   Yes 1.2 (0.8–1.9) 0.9 (0.7–1.3) 0.5 (0.2–1.5) 0.5 (0.2–1.2) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
CWEa With Identified Causeb (n = 196) Overall Proportion of DRE: 48%CWE Without Identified Cause (n = 404) Overall Proportion of DRE: 21%
Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)Crude RR for DREc (99% CI)Adjustedd RR for DRE (99% CI)
Seizure classification 
 Focal seizures, any type 
  Yes 2.0 (1.0–4.1) 1.0 (0.6–1.9) 1.4 (0.8–2.4) 1.1 (0.7–1.9) 
  No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Generalized seizuresd 
  Tonic-clonic seizures 
   Yes 1.6 (1.1–2.3) 1.2 (0.9–1.5) 1.1 (0.6–2.1) 0.8 (0.4–1.5) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Tonic seizures 
   Yes 2.2 (1.6–3.1) 1.3 (1.0–1.7) 2.9 (1.7–4.9) 1.6 (1.0–2.7) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Myoclonic seizures 
   Yes 2.5 (1.8–3.5) 1.5 (1.1–2.0) 3.7 (2.3–5.9) 1.5 (0.8–2.7) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Atonic seizures 
   Yes 1.7 (1.2–2.5) 1.2 (0.9–1.6) 1.9 (0.8–4.9) 1.0 (0.4–2.3) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Typical absence seizures 
   Yes Not estimated Not estimated 1.2 (0.6–2.4) 1.4 (0.8–2.4) 
   No — — 1 (reference) 1 (reference) 
  Atypical or other absence seizures 
   Yes 1.2 (0.7–2.0) 1.1 (0.7–1.6) 2.2 (1.2–4.0) 1.4 (0.8–2.3) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
 Unclassifiable seizures 
  Epileptic spasms 
   Yes 1.5 (1.0–2.2) 1.0 (0.8–1.4) 3.1 (1.8–5.4) 1.8 (1.2–2.8) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 
  Other unclassifiable seizures 
   Yes 1.2 (0.8–1.9) 0.9 (0.7–1.3) 0.5 (0.2–1.5) 0.5 (0.2–1.2) 
   No 1 (reference) 1 (reference) 1 (reference) 1 (reference) 

—, not applicable.

a

CWE, according to the new definition (ILAE 2014) and at least 1 y of follow-up since onset.

b

Includes structural, genetic, metabolic, and infectious causes. There were large degrees of overlap between these categories. No CWE had immunologic etiology.

c

DRE, defined as failure to achieve ≥1 y of seizure freedom at end of follow-up despite trial of ≥2 antiepileptic drugs in monotherapy or in combination. The intermediate category includes seizure outcomes in between DRE and ≥1 y of seizure freedom.

d

Adjusted for number of seizure types (1–2 vs ≥3).

In Table 4, we divided the examined characteristics into 3 groups: epilepsy risk factors, seizure and/or epilepsy characteristics, and clinical and/or EEG characteristics. We calculated crude RRs for DRE and then fitted models with mutual adjustments for all prognostic factors within each group. For preterm birth and low Apgar scores, the associations with DRE that were observed for CWE overall became weaker or disappeared in the adjusted regression models. Male sex was associated with a lower risk of DRE in CWE without an identified cause (adjusted RR = 0.6) but not in CWE with an identified cause (adjusted RR = 0.9).

As shown in Table 4, a history of ≥3 seizure types was always strongly associated with DRE, regardless of etiology and other seizure and/or epilepsy characteristics. For status epilepticus, seizure precipitants, and combined focal and generalized onset, the associations with DRE became weaker or disappeared after stratification and adjustment. This was also the case for infancy onset of epilepsy. Additional neurologic and developmental difficulties were associated with DRE in CWE without identified cause (adjusted RR = 2.0), whereas abnormal findings on neurologic examinations were associated with DRE in CWE with identified cause (RR = 2.3). Findings of epileptic activity in EEGs was associated with DRE across the CWE subgroups, but the association between nonepileptic EEG pathology and DRE was substantially weakened after stratification by etiology and adjustment for other clinical and/or EEG characteristics.

In Table 5, we have shown RR estimates for associations between DRE and the different seizure types, stratified by main etiological category. In the adjusted models, the RR estimates were adjusted for the total number of seizure types (1–2 vs ≥3) because having multiple seizure types was so strongly associated with DRE. All the associations between individual seizure types and DRE were attenuated after adjustment, and most were not statistically significant.

In this study, we provide contemporary population-based data on the short-term outcomes of childhood epilepsy. At end of follow-up, 59% of CWE had achieved seizure freedom lasting 1 year or more, whereas 30% had DRE. The remaining 12% had intermediate seizure outcomes.

The proportion with DRE in this study was higher than that of other studies.2,4,7 One reason is our relatively young study sample, with a median age of 3 years at epilepsy onset for CWE. Our proportion of DRE (30%) is comparable to that of the previously mentioned study of epilepsy with onset before 36 months of age, in which researchers found intractable seizures in 35%.8 The CWE in our study sample had a higher proportion with identified causes (33%) than previous studies. Having epilepsy with an identified cause (previously referred to as symptomatic epilepsy) is predictive of a higher risk of DRE.3,7,8 Mean follow-up time was also shorter in our study. Finally, our definition of DRE11 is likely to capture more CWE than the definitions used in previous studies.17 

Our study was conducted in a high-income country with universal health service coverage where all CWE have access to modern diagnostic procedures and antiepileptic treatments. From a global perspective, these children have grown up under optimal circumstances. It was somewhat disappointing to find that 3 out of 10 CWE had not obtained seizure control despite adequate diagnostics and treatment. This suggests that recent diagnostic and therapeutic advances have not yet led to significant improvements in short-term seizure outcomes for CWE and confirms the need for increased understanding and more effective treatments of early-onset epilepsy.

Seizure outcomes were similar across sociodemographic categories, which are in line with previous findings from Canada.2 The overall risk of epilepsy was somewhat higher among children of parents with low levels of education, but the differences were not large. There were no associations between epilepsy risk and other sociodemographic characteristics. In our data, it was indicated that sociodemographic factors have little influence on epilepsy risk in Norwegian children. Some caution is warranted because socially disadvantaged groups such as immigrants and single mothers are underrepresented in the MoBa cohort.18 In nationwide survey data from the United States, it has been shown that childhood epilepsy is more common in low-income families.10 In our study, we may not have fully captured the potential risk associated with social disadvantage. The findings are likely to be generalizable to other high-income countries with universal access to health care but may not be representative of more diverse populations where health care is not easily accessible to all children.

A number of characteristics were associated with increased risk of DRE for CWE overall: preterm birth, low Apgar scores, multiple (≥3) seizure types, history of status epilepticus, history of seizure precipitants, combined generalized and focal mode of onset, infancy onset of epilepsy, additional neurologic and developmental difficulties, abnormal findings on neurologic examinations, and abnormal EEG findings. This is in line with findings from previous studies.4,8 Most of the associations were attenuated or disappeared after stratification by etiology and adjustment for related prognostic factors, indicating that they are partially or fully explained by the underlying cause of epilepsy.

Some prognostic factors were associated with DRE regardless of etiology and other prognostic factors. CWE with multiple (≥3) seizure types had a large increase in risk compared with those with 1 to 2 seizure types. The associations observed between individual seizure types and DRE were largely driven by epilepsies with multiple seizure types. This also applied to the seizure types associated with a particularly high risk of DRE, that is, tonic seizures, myoclonic seizures, atonic seizures, and epileptic spasms. Approximately two-thirds of the CWE with 1 or more of these seizure types had a history of ≥3 seizure types. Epileptic activity in EEGs was also strongly associated with DRE across subgroups, but the majority of CWE had such findings, and it is more correct to say that the absence of such findings was predictive of sustained seizure freedom.

We found that additional neurologic or developmental difficulties were associated with DRE in CWE without identified cause, whereas abnormal findings on neurologic examinations were associated with DRE in CWE with identified cause. These characteristics were often overlapping in CWE, regardless of the underlying cause. The differences in adjusted RR estimates between the 2 main etiological categories may just be artifacts of the statistical modeling. We believe that clinicians should always be aware of an increased risk of DRE whenever CWE have any of these characteristics.

The genetic etiological category of the ILAE 2017 classification includes both epilepsies with identified genetic causes and presumed genetic causes. The identified genetic causes often represent de novo mutations that lead to severe epilepsy syndromes, whereas the presumed genetic causes pertain to genetic generalized or familial epilepsies. Consequently, using 1 single category for genetic epilepsies may not always be clinically meaningful in children. This was confirmed by the large difference in risk of DRE between CWE with identified genetic causes and CWE with presumed genetic causes. On the other hand, CWE with poor treatment responses may have been more likely to undergo genetic testing, and such reverse causation could have contributed to the association found between identified genetic causes and DRE.

The proportion of CWE who had died was 2%. In a previous study with the same mean duration of follow-up (5 years), it was also found that 2% of CWE had died,19 whereas mortality was 3.5% in studies with follow-up until adulthood.20 Thus, the mortality of CWE in our study appears to be similar to that of previous studies.

The main limitation of this study was that we were only able to investigate short-term seizure outcomes. However, these are of interest regardless of the final outcome because repeated seizures and long-term antiepileptic treatment are likely to have independent negative effects on the developing brain. The main strength of the study was the population-based and prospective cohort design with follow-up from pregnancy and onwards. This design enabled us to capture the complete spectrum of epilepsies occurring in the general child population. The combination of questionnaire data, registry data, medical record reviews, and parental interviews provided opportunities to explore a wide range of prognostic factors, including sociodemographic risk factors. We have added to previous knowledge by showing how the prognostic value of individual prognostic factors is often determined by the cause of epilepsy.

In this population-based study of childhood epilepsy, 30% of CWE had DRE, 59% had achieved ≥1 year of seizure freedom, and the remaining 12% had intermediate seizure outcomes. Having an identified cause of epilepsy (structural, genetic, metabolic, or infectious) was associated with DRE, and the cause modified the associations observed for most other prognostic factors.

     
  • CI

    confidence interval

  •  
  • CWE

    children with epilepsy

  •  
  • DRE

    drug-resistant epilepsy

  •  
  • EPYC

    Epilepsy in Young Children Study

  •  
  • ILAE

    International League Against Epilepsy

  •  
  • MoBa

    Norwegian Mother and Child Cohort Study

  •  
  • RR

    relative risk

Drs Aaberg, Chin, and Surén contributed to the conception and design of the study, the collection of data, the analysis and interpretation of data, and the drafting of the manuscript; Drs Bakken, Lossius, Tallur, and Stoltenberg contributed to the analyses and interpretation of data and reviewed and revised the manuscript; Dr Lund Søraas contributed to the data collection and to the analyses and interpretation of data and reviewed and revised the manuscript; and all authors have approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

FUNDING: Funded by grants from the Research Council of Norway (grant 213699) and the Regional Health Authority of South-East Norway (grant 2014057).

We are grateful to all the participants in MoBa and the parents who participated in telephone interviews in the EPYC. We would also like to thank our network of pediatricians, neurologists, neurophysiologists, and radiologists at the hospitals providing data and Therese Wardenær Bakke and Kaja Schau Knatten for their efforts in facilitating the data collection.

1
Sillanpää
M
,
Anttinen
A
,
Rinne
JO
, et al
.
Childhood-onset epilepsy five decades later. A prospective population-based cohort study.
Epilepsia
.
2015
;
56
(
11
):
1774
1783
[PubMed]
2
Camfield
C
,
Camfield
P
,
Smith
B
.
Poor versus rich children with epilepsy have the same clinical course and remission rates but a less favorable social outcome: a population-based study with 25 years of follow-up.
Epilepsia
.
2016
;
57
(
11
):
1826
1833
[PubMed]
3
Geerts
A
,
Arts
WF
,
Stroink
H
, et al
.
Course and outcome of childhood epilepsy: a 15-year follow-up of the Dutch study of epilepsy in childhood.
Epilepsia
.
2010
;
51
(
7
):
1189
1197
[PubMed]
4
Berg
AT
,
Rychlik
K
,
Levy
SR
,
Testa
FM
.
Complete remission of childhood-onset epilepsy: stability and prediction over two decades.
Brain
.
2014
;
137
(
pt 12
):
3213
3222
[PubMed]
5
Arts
WF
,
Brouwer
OF
,
Peters
AC
, et al
.
Course and prognosis of childhood epilepsy: 5-year follow-up of the Dutch study of epilepsy in childhood.
Brain
.
2004
;
127
(
pt 8
):
1774
1784
[PubMed]
6
Berg
AT
,
Rychlik
K
.
The course of childhood-onset epilepsy over the first two decades: a prospective, longitudinal study.
Epilepsia
.
2015
;
56
(
1
):
40
48
[PubMed]
7
Sillanpää
M
,
Schmidt
D
.
Early seizure frequency and aetiology predict long-term medical outcome in childhood-onset epilepsy.
Brain
.
2009
;
132
(
pt 4
):
989
998
[PubMed]
8
Wirrell
E
,
Wong-Kisiel
L
,
Mandrekar
J
,
Nickels
K
.
Predictors and course of medically intractable epilepsy in young children presenting before 36 months of age: a retrospective, population-based study.
Epilepsia
.
2012
;
53
(
9
):
1563
1569
[PubMed]
9
Gaily
E
,
Lommi
M
,
Lapatto
R
,
Lehesjoki
AE
.
Incidence and outcome of epilepsy syndromes with onset in the first year of life: a retrospective population-based study.
Epilepsia
.
2016
;
57
(
10
):
1594
1601
[PubMed]
10
Russ
SA
,
Larson
K
,
Halfon
N
.
A national profile of childhood epilepsy and seizure disorder.
Pediatrics
.
2012
;
129
(
2
):
256
264
[PubMed]
11
Kwan
P
,
Arzimanoglou
A
,
Berg
AT
, et al
.
Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies.
Epilepsia
.
2010
;
51
(
6
):
1069
1077
[PubMed]
12
Scheffer
IE
,
Berkovic
S
,
Capovilla
G
, et al
.
ILAE classification of the epilepsies: position paper of the ILAE Commission for Classification and Terminology.
Epilepsia
.
2017
;
58
(
4
):
512
521
[PubMed]
13
Fisher
RS
,
Cross
JH
,
French
JA
, et al
.
Operational classification of seizure types by the International League Against Epilepsy: position paper of the ILAE Commission for Classification and Terminology.
Epilepsia
.
2017
;
58
(
4
):
522
530
[PubMed]
14
Magnus
P
,
Birke
C
,
Vejrup
K
, et al
.
Cohort profile update: the Norwegian Mother and Child Cohort Study (MoBa).
Int J Epidemiol
.
2016
;
45
(
2
):
382
388
[PubMed]
15
Aaberg
KM
,
Gunnes
N
,
Bakken
IJ
, et al
.
Incidence and prevalence of childhood epilepsy: a Nationwide Cohort Study.
Pediatrics
.
2017
;
139
(
5
):
e20163908
[PubMed]
16
Aaberg
KM
,
Surén
P
,
Søraas
CL
, et al
.
Seizures, syndromes, and etiologies in childhood epilepsy: the International League Against Epilepsy 1981, 1989, and 2017 classifications used in a population-based cohort.
Epilepsia
.
2017
;
58
(
11
):
1880
1891
[PubMed]
17
Téllez-Zenteno
JF
,
Hernández-Ronquillo
L
,
Buckley
S
,
Zahagun
R
,
Rizvi
S
.
A validation of the new definition of drug-resistant epilepsy by the International League Against Epilepsy.
Epilepsia
.
2014
;
55
(
6
):
829
834
[PubMed]
18
Nilsen
RM
,
Vollset
SE
,
Gjessing
HK
, et al
.
Self-selection and bias in a large prospective pregnancy cohort in Norway.
Paediatr Perinat Epidemiol
.
2009
;
23
(
6
):
597
608
[PubMed]
19
Callenbach
PM
,
Westendorp
RG
,
Geerts
AT
, et al
.
Mortality risk in children with epilepsy: the Dutch study of epilepsy in childhood.
Pediatrics
.
2001
;
107
(
6
):
1259
1263
[PubMed]
20
Berg
AT
,
Nickels
K
,
Wirrell
EC
, et al
.
Mortality risks in new-onset childhood epilepsy.
Pediatrics
.
2013
;
132
(
1
):
124
131
[PubMed]

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.