You are on page 1of 12

Seizure 22 (2013) 333–344

Contents lists available at SciVerse ScienceDirect

Seizure
journal homepage: www.elsevier.com/locate/yseiz

Review

Early predictors of outcome in newly diagnosed epilepsy
Rajiv Mohanraj a, Martin J. Brodie b,*
a b

Salford Royal Hospital, England, UK Epilepsy Unit, Western Infirmary, Glasgow, Scotland, UK

A R T I C L E I N F O

A B S T R A C T

Article history: Received 27 November 2012 Received in revised form 5 February 2013 Accepted 5 February 2013 Keywords: Epilepsy Prognosis Drug response Outcomes Risk factors

Longitudinal studies of newly diagnosed epilepsy in children and adults have identified prognostic factors that allow early identification of patients whose seizures are likely to remain uncontrolled with antiepileptic medication. Results from outcome studies may be subject to bias, depending on the setting (community versus clinic), design (retrospective versus prospective) and characteristics of the patient cohort studied (age, types of epilepsy, specific comorbidities). Nevertheless, factors such as early response to medication, underlying aetiology, and number of seizures prior to initiation of treatment have consistently been found to be predictive of seizure outcomes. Other variables such as age, electroencephalographic findings and the presence or absence of psychiatric co-morbidities have been correlated with outcomes in some analyses. This review has examined studies of seizure outcomes in adults and children with newly diagnosed epilepsy identifying the risk factors that are associated with subsequent refractory epilepsy. ß 2013 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.

1. Introduction Epilepsy, the tendency to have recurrent unprovoked seizures, is the most common serious neurological disorder. Its prevalence ranges from 0.5 to 1% of the population in developed countries, and is probably higher in developing countries.1 This condition is, however, merely a symptom of a wide variety of neurological disorders, ranging from self-limiting and benign to devastating and fatal. Regardless of the aetiology, the tendency to suffer recurrent seizures exposes persons with epilepsy to a variety of physical, psychological and social morbidities.2 Complete control of seizures can negate these consequences to a large extent.3 The majority of patients diagnosed with epilepsy can expect to achieve good control of seizures with antiepileptic drug (AED) therapy,4,5 but a substantial minority will continue to experience seizures in spite of a range of AEDs used in adequate doses either singly or in combination.6 Some patients whose seizures prove difficult to treat could benefit from non-pharmacological strategies, especially epilepsy surgery, which still remains one of the most underutilised effective treatment modalities worldwide.7,8 Early identification of patients whose seizures are likely to be pharmacoresistant would permit them to be offered referral for epilepsy surgery at the most appropriate juncture.9

2. Methodology This short review will attempt to summarise data from relevant studies of outcomes in newly diagnosed epilepsy in paediatric and adult populations. These were identified from Pubmed using the search terms ‘newly diagnosed epilepsy and outcomes’ and ‘newly diagnosed epilepsy and prognosis’. Search results were reviewed manually to identify relevant publications. All studies with a minimum of 100 patients, who were followed up for at least 2 years, were included in this review (see Tables 1 and 2). Results from studies of prognosis of epilepsy are often conflicting. Some of the variability can be explained on the basis of differences in populations and methodologies. As the underlying cause of epilepsy can be widely varied, data from studies that ‘lump’ together all epilepsy types will be skewed in favour of those most frequent in the population. Studying well-defined epilepsy (electroclinical) syndromes separately can provide better prognostic information, but accurate classification is not always achievable even in patients attending specialist services. Studies based in specialist clinics can be expected to have better characterised patient groups, but may be biased towards the more severe epilepsies. Retrospective studies, especially those from specialist clinics, may not include patients with milder forms of epilepsy. Studies that identify all persons with new onset epilepsy in a defined population over a fixed period of time and follow them up prospectively will have the least recruitment bias. However, identifying all cases can be challenging and the diagnosis may be

* Corresponding author at: Epilepsy Unit, Western Infirmary, Glasgow G11 6NT, Scotland, UK. Tel.: +44 141 211 2534; fax: +44 141 211 2072. E-mail address: mjb2k@clinmed.gla.ac.uk (M.J. Brodie).

1059-1311/$ – see front matter ß 2013 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.seizure.2013.02.002

*24 564 definite epilepsy 228 .13 Recruitment Inclusion period criteria 1935–74 Exclusion criteria Type of study Retrospective Setting n Epilepsy type Mixed Follow up 5–20 years Method of follow-up Chart review n with outcome 457 Outcomes 5 year remission Prognostic factors identified Perinatal insult causing physical and intellectual handicap worst outcomes (46% remission) Post natal acquired epilepsy and idiopathi epilepsy 74% remission > 2 seizures not Febrile and provoked by an other provoked seizures acute cause followed for minimum of 5 years Single seizures Community 618 65% by 10 years 76% by 20 years Elwes et al. higher number by 2 of seizures before treatment Number of seizures in Years.probable epilepsy 9 years Idiopathic seizures 69% Remote symptomatic epilepsy 61% Chart review 550 MacDonald*25 12 years Children lower remission rates than adults Partial seizures worse than generalised ones Stephen et al. Study Annegers et al.1% not SF on monotherapy 5 year remission Age and seizure type no effect on outcomes Cockerell et al.334 Table 1 List of studies of outcomes in adult epilepsy. and 98% by the first year of AED treatment correlated 5 years with risk of refractory epilepsy 3–year remission 92% 5 year remission 78% 22. Mohanraj. Studies from the same population have been grouped together. Studies marked with * (asterix) also inlcuded children. or psychiatric handicap A family history of epilepsy Collaborative group18 1982 Previously untreated afebrile seizures AEDs started >3 Prospective months before enrolment Polytherapy as initial treatment Clinic 280 Mixed 48 (median) 6 monthly review 228 Prophylactic use of AEDs Provoked seizures Cockerell et al.*23 1984–87 Definite epilepsy as judged by panel Expert panel disagree with diagnosis of epilepsy Prospective Community 1091 Mixed 6 years Contact primary care physician 1–year remission Remission less likely 62% at 1 year. social. M.33 1984–97 None stated Newly diagnosed localisation related peilepsy Retrospective Clinic 550 Localisation reltated epilepsy Median 5 (2–15) MTS worse than other No difference causes of LRE between cryptogenic and symptomatic LRE . 92% by 3 years.J.15 Not stated Previously untreated epilepsy None stated Prospective Clinic 106 Mixed 6–106 month Clinical review 106 82% 2 year sf periods by 8 years of fu 35% immediate responders 28% SF throughout Poor prognosis associated with Partial seizures R. 81% with multiple seizure types. Brodie / Seizure 22 (2013) 333–344 High frequency of tonic-clonic seizures before treatment A neurologic.

Brodie / Seizure 22 (2013) 333–344 Kwan & Brodie80 1984–1997 Newly diagnosed epilepsy Hitiris et al. Mohanraj. M. Late remission 23. neurological deficit and mental retardation.*88 1989–99 Newly diagnosed epilepsy None stated Retrospective Clinic 352 Mixed Upto 10 years Chart review 352 335 .J.But 30% were already on treatement Seizure free rates worse for MTS 42% AVM 78% Infarction 63% Primary brain tumour 57% Cortical Gliosis 57% Cerebral atrophy 55% Cortical dysplasia 54% None Prospective Clinic 629 Mixed 5 years (range 2–16) Chart review 525 Refractory epilepsy more likely in Symptomatic / cryptogenic (40%) than idiopathic (26%) More than 20 pretreatment seizures (51% v/s 29%) 47% seizure free on first AED 14% on second or 3rd AED 3% seizure free on combination therapy Poor prognosis associated with Higher number of seizures and poor response to initial monotherapy R.0% achieved associated with early remission (immediate seizure control after start of AED) Del Felice et al.66 Mohanraj & Brodie67 1981–2001 New diagnosis of epilepsy None Retrospective Clinic 890 Mixed 79 months median Chart review 780 59.2% seizure free Uncontrolled epilepsy associated with for at least 12 months More than 10 seizures prior to starting treatment Family history of epilepsy History of febrile seizures Post-traumatic epilepsy Recreational drug use Prior or current psychiatric comorbidity Factors not predicting poorer outcome included gender.

45 1985–87 Diagnosis of epilepsy Provoked seizures Retrospective Clinic 107 Mixed 1–12 years (range) Chart review 107 1 year remission 68% Seizures in the first year after starting AED predictive of never achieving 1 year remission Not predictive of remission: Age at diagnosis.7%.5 Clinical review 478 Seizure free rates dropped with sequential failure of AED 41. M.7 for patients with 2– 5 partial seizures 6. Etiology EEG Schiller and Najjar82 1999–2004 Commencing AED therapy None stated Prospective Clinic 478 Mixed 1.336 Table 1 (Continued ) Study Recruitment Inclusion period criteria Exclusion criteria Type of study Setting n Epilepsy type Follow up Method of follow-up n with outcome Outcomes Prognostic factors identified Increased number of 10. Seizure type. Brodie / Seizure 22 (2013) 333–344 Lindsten et al. EEG.6%.J.5 to 7. aetiology. disease duration. seizure types not associated with outcomes R. neurologic/psychiatric abnormality.7 in patients with >5 partial seizures Age. after 2–5 AED 0% after 6–7 AED Response to past AEDs 3 year remission 64% 5 year remission 58% Type of epilepsy Duration of epilepsy Number of seizures in the 3 months prior to AED initiation . after one AED 16.8% achieved partial seizures late remission (remission after >24 months after start of AED) Odds ratio (OR) of late remission 2. Mohanraj.

R. Aetiology Epilepsies relating to structural brain abnormalities are less likely to enter remission compared that occurring in patients with structurally normal brains. Smaller clinic-based studies. electro-encephalographic findings and early response to AED treatment are some of the variables that have been examined by a number of authors. Analysis of the Glasgow cohort of newly diagnosed epilepsy suggested that outcomes were generally better in patients aged over 65 years at the onset of epilepsy. Studies of prognosis in newly diagnosed epilepsy examine the likelihood of seizure remission with AED therapy. This view was supported by Reynolds and colleagues in the early 1980s and early treatment of seizures was considered key to preventing the emergence of drug-resistant epilepsy. thereby minimising the number of subsequent seizures.23–25 Recent evidence implicates seizure clusters as being associated with poorer prognosis. This is intricately linked to the question of whether drug resistance develops as a result of repeated seizures.16. in part. those with temporal lobe epilepsy were over represented compared to other types of localisation related epilepsy.J.25 Multivariate analyses of prognostic factors in children44 and in adults15. aetiology. and so could be expected to have a substantially worse prognosis than found in similar studies in newly diagnosed populations. Alternatively those run by epilepsy specialists are open to the criticism that they are unlikely to be population-based.47 Possible explanations include lack the neuronal plasticity needed for the development of pharmacoresistance and reduced likelihood of genetic factors adversely affecting prognosis. with remission rates of 23 and 26% in the two studies compared to 55 and 48% for patients with generalised tonic–clonic seizures. prognosis could not be predicted early on. the UK National General Practice Survey of Epilepsy (NGPSE).13 Analysis of data from two Veterans Affairs randomised control studies of AED treatment also showed that patients with complex partial seizures fared worst. 5. It is likely that any differences are a reflection of the epilepsy syndromes that are prevalent among the various age groups. usually defined in terms of the number of AEDs unsuccessfully tried. where epilepsy syndromes are generally better defined.49 A long history and high numbers of pre-treatment seizures were thought to correlate with a poor outcome. Other studies have identified myoclonic jerks.27–31 The nature and location of the underlying structural abnormality can also affect the likelihood of response to treatment. Their conclusions regarding each variable are summarised below.14 These low numbers can be.45 have found no independent correlation of age at onset with prognosis. duration of epilepsy and number of seizures before starting AED treatment. Seizure types Some studies have suggested a significant effect of seizure types on prognosis. In the population based study in Rochester. Duration of epilepsy and number of pre-treatment seizures Many studies have examined the impact of the number of seizures and the time from onset of epilepsy to starting AED treatment on prognosis. Mixed cohorts of various epilepsy syndromes cannot reflect the effect of individual syndromes on prognosis.30. This is manifest as a lower remission rate for symptomatic epilepsies (both partial and generalised) compared to idiopathic epilepsy syndromes in children.50 . Mohanraj.40.10 Such studies are also resource intensive and very few have been conducted. many patients included in these analyses did not have newly diagnosed epilepsy. In their meta-analysis of studies on childhood absence epilepsy. Annegers and colleagues found that patients with generalised tonic–clonic and absence seizures had 80 and 85% probability respectively of achieving remission. Seizure types. history of febrile convulsions. Bouma and colleagues found that significant heterogeneity existed between various cohorts in inclusion criteria and definition of remission. with good control of seizures with AED treatment but high rate of recurrence on withdrawal of medication. In developed countries. compared to absence seizures only (78%).33 However. have also supported this finding. another large community based study. This is more pertinent to paediatric cases.38.32 Analysis of 550 patients with localisation related epilepsy in Glasgow also reported worse remission rates in patient with hippocampal sclerosis than in focal epilepsies due to other causes. Development of tonic–clonic seizures was associated with a reduced chance of remission (35%). such as hippocampal sclerosis. 3. including cerebrovascular disease and neurodegeneration.42 Berg and colleagues found that prognosis improved with increasing age. however. and that this effect was not restricted to the first year of life. can have a good outcome and so may never be reviewed at a specialist service.12 While failure to achieve remission with AED therapy does not always equate to refractory epilepsy. eyelid or perioral myoclonia and presence of atypical EEG patterns as predictive of a poor prognosis in childhood absence epilepsy and subsequent evolution into juvenile myoclonic epilepsy (JME). A detailed analysis of all published predictive factors in each epilepsy syndrome is beyond the scope of this article.43 However. Minnesota. intellectual or psychiatric co-morbidity family history of epilepsy. only a few prominent examples will be touched upon.16 Patients experiencing multiple seizure types also appear to have worse outcomes in adults17–19 and children. however. suggests that the effect of seizure types is less important than other factors including early response to treatment in determining eventual outcome. we feel that both outcome measures represented poor outcomes and have included studies of remission and of refractory epilepsy in this analysis.46 These present with epilepsy as a consequence of a variety of acquired brain disorders. M. The concept that ‘‘seizures beget seizures’’ was introduced by Gowers in the 19th century48 and reinforced by the writings of Rodin in the 1960s.26 Analysis of these data overall. Age at onset Some studies in children suggest that onset of epilepsy before the age of 12 months is a poor prognostic factor. Semah and colleagues found that in 2200 treated patients referred to a specialist epilepsy service in France.39 Analysis of JME cohorts suggest that endophenotypes are important in determining prognosis.34 Many patients with substrates for epilepsy often regarded as invariably being associated with drug resistance.36 The identification of well defined electro-clinical syndromes can allow more accurate assessment of prognosis. studies that included patients of all ages did not demonstrate a consistent effect of age at onset on prognosis. however.11. found only a modest effect of seizure type on prognosis. age of onset. Brodie / Seizure 22 (2013) 333–344 337 less secure in the non-specialist setting.13.1.35. will lead to improved outcomes. the highest incidence of epilepsy is in older patients. explained by the inclusion of patients who had already failed their first treatment schedule.28.41 4. compared to only 65% for patients with complex partial seizures. An important question for physicians is whether treating patients early.15.20–22 However. 4. presence of neurological. or the likelihood of developing refractory epilepsy.37 As a result the reported remission rates ranged from 21% to 89%.

M.J. aetiology.98 1988–92 Diagnosis of epilepsy made in the study period Panel judged not epilepsy Prospective Clinic 494 Mixed 5 years Active follow up at 6 months and 2 years Postal questionnaire 466 345 (76%) in terminal remission >1 year at 5 yrs FU 290 (64%) > 2years Aetiology.64 1977–85 Children with 2 or more unprovoked seizures in the recruitment period Follow up available 2–12 years Myloclonic.69 Recruitment period 1988–92 Inclusion criteria: Newly diagnosed epilepsy Exclusion criteria: None stated Type of study Prospective Setting Clinic n 494 Epilepsy type Mixed Follow up 2 years Method of follow-up Structured review n with outcome 466 Outcomes 31% poor outcome (Did not achieve 6 m remission by 2 years) Prognostic factors Intake variables associated with poor outcome Number of seizures Seizure types Aetiology Arts et al. history of febrile seizures and age (at study entry) Sex. post ictal signs. Study Arts et al. and remission at 6 months At 6 months R. history of febrile seizures.8 years mean Structured review for 5 years 413 Five-year terminal remission (TR) was reached by 71% of the cohort. tonic seizures Prospective Community 479 Mixed 2–12 years (median not stated) Chart review 479 70% seizure free and attempted drug withdrawal No effect of number of pretreatment seizures on prognosis for seizure control. Mohanraj. absence.338 Table 2 List of studies of outcomes in paediatric epilepsy. 27 intractable Geerts et al. up to 10 pre treatment seizures Progressive brain disease Unknown pretreatment seizure number Contact with primary care physician 70% seizure free after drug withdrawal More patients suffering >10 pre treatment seizures likely to have complex partial seizures . Studies from the same population have been grouped together. Brodie / Seizure 22 (2013) 333–344 65 (14%) no seizures since start of follow up 108 (24%) with TR < 1 year at 5 years.22 1988–92 As above As above Prospective Clinic 415 Mixed 14. Course during Associations with failure to achieve terminal remission Postal questionnaires at year 10 and 15 Non-idiopathic etiology Febrile seizures No 3 month remission Early intractability Camfield et al.

>1 sz per month) Symptomatic/ cryptogenic generalised syndromes most likely to develop IE Initial seizure frequency Berg et al.3%) Cryptogenic epilepsy (61.6% with cryptogenic / symptomatic generalised 339 .101 2000–2008 CAE treated with Valproate Brain disorders apart from developmental delay and ADHD Retrospective Clinic 235 CAE treated with valproate Not stated Chart review 180 58. absence. Brodie / Seizure 22 (2013) 333–344 Ma et al.100 2000–2004 Children diagnosed with epilepsy in the study period None stated Retrospective Clinic 520 Mixed 12–75 months (range) Chart review Ollivier et al. tonic seizures Prospective Community 417 Mixed 92 =/À 26 months Chart review 417 61% seizure free on first drug and withdrew medication Presence of neurological deficits and complex partial seizures associated with failure of first AED Progressive brain disease Contact with primary care physician Unknown pretreatment seizure number Only 41% of those failing first AED became seizure free with second AED 29% of those failing first AED developed refractory epilepsy 520 (89%) 344 seizure free with first drug Patients with symptomatic epilepsy (60.8%) R. Mohanraj.31 1993–97 Diagnosis of epilepsy in the recruitment period Diagnosis of epilepsy made prior to study period by non participating physician Panel of neurologists felt diagnosis of epilepsy unlikely Prospective Community 885 Mixed 5 years median Active follow up 613 60 (10%) intractable (failure of 2 AEDs.3% achieved seizure freedom Generalised tonic clonic seizures High frequency of seizures (>10/day) at diagnosis Younger age at diagnosis . M.No difference between patients suffering up to 10 pretreatment seizures and those suffering fewer Camfield et al.J.99 1977–85 Children with 2 or more unprovoked seizures in the recruitment period Follow up available 2–12 years Myloclonic.44 34.5%) Less likely to become seizure free Idiopathic epilepsy (73.associated with failure to respond to VPA Berg et al.

7% with other loc rel 8. Brodie / Seizure 22 (2013) 333–344 Multiple seizure types (hazard ratio 6. median 40 years Structured review 144 67% terminal remission Weekly seizure frequency in the first year on treatment . M.J.9% were intractable (more than one seizure/ month over 1 year with failure of three or more anticonvulsants) Factors predictive of intractability R.21 1991–2000 Recurrent unprovoked seizure Age 2–17 Single seizures Retrospective Clinic 240 Mixed 55 months (mean) Chart review 197 52.8% had a poor outcome (recurrent seizures on therapeutic antiepileptic drug levels within 6 months prior to the final assessment) 6.7% of idiopathic Prognostic factors Focal EEG slowing 10. / febrile status no increased risk Oskoui et al.102 Recruitment period Inclusion criteria: Exclusion criteria: Communication and geographical factors limiting follow up Type of study Setting n Epilepsy type Follow up Method of follow-up n with outcome Outcomes 2.2) Seizure recurrence in the first 6 to 12 months of treatment (hazard ratio 70) Sillanpaa & Schmidt90 1965–2008 Age 15 or less 2 or more seizures Living in defined geographical area None Prospective Community based 150 Mixed 11–42 years.5) Seizures starting out with age range Minimum 2 years follow up Mental retardation at onset (hazard ratio 7.340 Table 2 (Continued ) Study Berg et al. Mohanraj.2% oc unclassified epilepsy Acute symptomatic / neonatal status epileticus bad Age of onset 5–9 years good Total number of seizures.6% were in remission 12.

especially associated with intellectual impairment.28. those that are prolonged.22 9. sustaining 100 or more generalised seizures before coming to medical attention have shown that a similar proportion go into remission as patients treated early after only a few seizures.44. treatment with AEDs after the first unprovoked seizure has been shown not to affect the long-term outcome. Febrile seizures Febrile seizures have an almost uniformly benign prognosis.67 This would imply that the epileptogenic process responsible for the high frequency of complex partial seizures is inherently pharmacoresistant. in spite of preventing seizures in the short term.e.66 Psychiatric problems are more frequent in patients with epilepsy.58–61 Several groups have shown a relationship of high initial seizure frequency with poor outcome.27. Brodie / Seizure 22 (2013) 333–344 341 Repeated seizures have been shown to produce neuronal loss and mossy fibre sprouting in the hippocampus.16 It is possible. 7. manifesting as physical and intellectual disability had only 46% probability of achieving seizure control.69 However. especially in children.27. predictive of a poorer outcome. focal or recur within the same day. studies that include all age groups and predominantly adult patients have not supported such an association.80–82 Patients whose seizures continue despite adequate doses of an appropriate AED have a lower chance of subsequent seizure remission.56 Prevention of repeated seizures in this setting by effective drug treatment could theoretically prevent neuronal apoptosis and synaptic reorganisation. studies in patients who have suffered seizures for several years.82 The recently published International League against Epilepsy definition of drug-resistant epilepsy recommends that failure to gain seizure control for at least one year with the first .29.53.7 and 2. and are tried on an alternative agent.20. Kwan and Brodie reported that only 11% of patient who failed initial monotherapy due to lack of efficacy subsequently became seizure free.79 Studies in adults have not found EEG to be independently predictive of outcome after adjusting for other factors.57 Moreover. which in turn can reinforce their production forming excitatory recurrent circuits. approximately 3% of affected infants will develop epilepsy in later life. Petrovski and colleagues demonstrated that the presence of neuropsychological symptomatology was associated with a worse response to treatment in patients with newly diagnosed epilepsy.68 6. Hitiris and colleagues noted that the presence of prior or current psychiatric comorbidity was significantly associated with failure to achieve remission. which might also be responsible for determining drug response.R.49. 10. which may be responsible for further seizures and thereby the emergence of drug resistance. have demonstrated that the presence of neurological deficit.55. M. is indicative of a poorer prognosis.77 8.75 The role of genetic factors in focal epilepsies is less clear. Family history of epilepsy Studies in children31 and adults15. compared to 41% of those who failed due to intolerable adverse effects and 55% of those who experienced an idiosyncratic adverse effects. i. There is evidence to support a close relationship in the pathogenesis of epilepsy and depression.16.66 have reported an association between family history of epilepsy and poorer prognosis. In an analysis of 780 patients diagnosed and treated for epilepsy in Scotland. those occurring in children with pre-existing neurological deficit.80 If the first AED is unsuccessful. Concomitant morbidity verbal memory impairment74 and that impaired memory performance at the outset was predictive of poor outcome. have found a correlation of background slowing and focal spike and wave activity with a poor outcome. On the other hand. respectively.15. There is.73 Other neuropsychological factors may also be important in determining prognosis. therefore that the presence of a psychiatric co-morbidity early in the course of the epilepsy points to a greater underlying cerebral dysfunction and is. However.72 More recently. the authors found that the response rate to the first. and supports the view that the prognosis of each epilepsy syndrome may be a characteristic of that specific disorder.64. subsequent AED regimes have a declining likelihood of success.70 Association of psychiatric co-morbidity with refractory epilepsy has previously been reported in adult patients.15 In the Glasgow cohort of 780 patients with newly diagnosed epilepsy.44 An electroencephalogram (EEG) performed soon after a seizure is more likely to detect such abnormalities.78 Complex febrile seizures.4. and those with a family history of epilepsy in a first degree relative.51.78 Hitiris and colleagues66 found that the febrile convulsions in infancy predicated a poorer outcome to treatment in newly diagnosed epilepsy as did Geerts and co-workers from the Netherlands. Electroencephalographic findings In their landmark population based study of patients with epilepsy in Rochester Minnesota. Response to first drug Several studies have found the response to the first AED to be the strongest predictor of long-term outlook in adults and children. this could be related to the underlying genetic mechanisms underpinning the epilepsy.13 Early clinic based studies also supported this notion in adults. compared to 74% for patients with idiopathic epilepsy and those with epilepsy due to postnatally acquired lesions.54 The degree of hippocampal volume loss was related to the duration of epilepsy. in adults. In generalised syndromes. the prognostic value of routine interictal EEG examination has not been established. Longitudinal studies employing repeat MRI scans have demonstrated progressive hippocampal and temporal neocortical volume loss and have suggested that neuronal loss can be correlated to number of seizures.J. compared to those who are unable to do tolerate the initial AED. and is likely to have greater prognostic value.12. Annegers and colleagues found that patients who had epilepsy as a result of presumed perinatal cerebral insult.18.7%. of course. mainly in children.45 Thus. therefore. second and third AED was 50.71. detailed analyses of data from observational studies have suggested that this is true only for patients suffering complex partial seizures.15 Subsequent studies. Mohanraj. 10. a well known association between febrile seizures in infancy and the development of hippocampal sclerosis in later life. Genetically determined malformations of cortical development may play a role76 as may pharmacogenetic traits that run in families. The NGPSE and Glasgow cohorts showed no significant effect of neurological or intellectual deficits on prognosis25.62–66 However.66 Similar observations have been made following temporal lobectomy in patients with refractory epilepsy reporting a psychiatric history. all confer a higher risk of developing epilepsy.52 Cross-sectional magnetic resonance imaging (MRI) studies have demonstrated smaller hippocampal volumes ipsilateral to the seizure focus in patients with temporal lobe epilepsy and uncontrolled seizures. One small clinic based study found that patients with newly diagnosed epilepsy had Some studies.65 More recently responses have been demonstrated in a few patients up to the seventh regimen.

such as transcranial magnetic stimulation. and other similar proteins are thought to play a role in drug resistant epilepsy by extruding AEDs from the seizure focus. While most patients who respond to AED treatment will do so early. Banerjee PN. especially epilepsy surgery. there is also a need to target available treatments.84 Luciano and Shorvon found that in a population of patients with refractory epilepsy.342 R. although some of this data is conflicting. UCB Pharma. Mohanraj. serves as a consultant for Eisai.97 Disclosures Rajiv Mohanraj has served on advisory boards for UCB Pharma and Eisai. Hauser WA. 22. a further 217 patients also achieved the same outcome over the rest of the follow up. This field has expanded enormously in parallel to the strides made in genetics technology.90 The most recent analyses from the 1098 patients with newly diagnosed epilepsy included in the Glasgow database divided the patient outcomes into 4 distinct groups with 37. achieving 12 months seizure freedom by one year of treatment.2% entered 2 year remission immediately on starting treatment. Eisai and GlaxoSmithKline. with a mean delay of 9 years from starting treatment. Epilepsy Research 2009. and 5% went on to develop drug resistant epilepsy. Sanofi Aventis. Novartis.68 These patients continue to suffer the physical.12.92 Single nucleotide polymorphisms represent the most frequently studied genetic marker to date. Callaghan and colleagues studied 246 patients who met the above definition of drug resistant epilepsy prospectively for 3 years.7% with a relapsing-remitting pattern and the final 24.86. Epidemiological data over the years have identified several factors that correlate with a poor prognosis in children and adults. It is more likely that responsiveness to AED treatment depends on a complex interplay of genetic and environmental influences.96 Studies to date show that the early response to treatment is a powerful predictor of the longterm outlook of newly diagnosed epilepsy. rising to a cumulative 2 year remission rate of 79% by 10 years of follow up. if appropriate. 15.88 In a long term prospective study of childhood onset epilepsy. They also place a strain on health care resources the world over. GlaxoSmithKline. disease-modifying approaches that will prevent or ameliorate the processes underpinning epileptogensis. Following up a clinic population of 256 patients who had been seizure free for a minimum of 12 months. but a further proportion achieved remission with each successive year of follow up.12 11. but a similar number failed to replicate this association. Conclusions Approximately one in three of patients diagnosed with epilepsy will fail to achieve lasting remission of seizures with AEDs treatment despite the global introduction of more than 12 new AEDs over the past 20 years.89 However. Eisai.12 There is. Patients who do not achieve complete seizure control with the first two appropriate regimens of AEDs in the first one to two years after starting treatment. and reported a 14% 6-month remission rate (approximately 5% per year) with further drug therapy alone.8% being refractory de novo. A review. failure to do so does not indicate a uniformly poor prognosis. Sillanpaa and Schmidt reported that patients who experienced ongoing weekly seizures during the first year on AED treatment were less likely to enter remission. of 780 newly diagnosed epilepsy patients.4% having a delayed response but an excellent outcome.65 Other hospital based series have reported the phenomenon of seizure relapse following an initial period of good response to AEDs and eventual development of drug resistant epilepsy. In the Glasgow series. therefore.4% of patients entering remission had relapse of seizures. Analysis of the Glasgow cohort found that 10. arguably more importantly. Early identification and prediction of patients likely to be unresponsive to initial drug therapy will allow earlier specialist intervention.83 Some authors have suggested that this view might be unduly pessimistic and that further drug trials may still result in worthwhile improvements in seizure control for a proportion of patients. this knowledge is far from complete. will subsequently relapse. More long-term studies are required to assess the prognosis of each epilepsy syndrome. Pharmacogenomic studies employing more sophisticated genotyping and bioinformatics technologies promise greater predictability of response to individual AEDs. an urgent unmet need for new symptomatic treatments with improved efficacy/tolerability profiles and.2% demonstrating longterm seizure freedom within 6 months of initiating treatment.J. but genome wide association studies are underway. psychological and social consequences of intractable seizures and adverse effects from an escalating drug burden.85 Many of these patients however. has received speaker fees from UCB Pharma. has accepted conference hospitality from UCB Pharma and Eisai.95 While there is a clear need for new AEDs with novel mechanisms of action. have a relatively small chance of achieving seizure freedom with further pharmacotherapy and may be considered to have drug-resistant epilepsy.91 A large number of genetic markers have been studied in patients with epilepsy to identify predictors of efficacy and adverse effects.77 Many studies have focussed on genes encoding the multidrug resistance protein MDR1 (ABCB1).93 It seems unlikely that a complex outcome such as responsiveness to AED treatment can be fully explained by variability of a single genetic factor. Other possibilities include the identification of biomarkers for drug-resistance. The basic mechanisms of drug resistance in epilepsy also remain largely unclear. and Lundbeck Inc. and has received research support from Eisai and GlaxoSmithKline. Genetics of drug resistance Pharmacogenetics is the study of genetic influences on the pharmacokinetics and pharmacodynamics of medications. and Eisai. serves on speakers’ bureaus for UCB Pharma. appropriately chosen and used AED schedules (whether as monotherapy or in combination) should prompt referral for specialist review for confirmation of the diagnosis for seizure and/or syndrome classification and.. GlaxoSmithKline. Brodie has served on scientific advisory boards for Pfizer Inc. 245 (31. Filippi D. M. Advances in cerebral imaging and molecular biological techniques have allowed a greater insight into the mechanisms underlying seizure generation and propagation. Valeant Pharmaceuticals.11. . Martin J.85:31–45.65 A retrospective survey of 352 newly diagnosed patients from 2 epilepsy clinics in Italy reported that 56. Brodie / Seizure 22 (2013) 333–344 2 tolerated. has received funding for travel from UCB Pharma and GlaxoSmithKline.94 12.4%) became seizure free immediately after starting AEDs. addition of a new AED resulted in seizure freedom in 16%. However. for consideration of epilepsy surgery. C3435T polymorphism in the ABCB1 gene has been found associated with refractory epilepsy in some studies. References 1.. 50% of patients entered remission late. This. The descriptive epidemiology of epilepsy. Schiller reported that 40% of patients had suffered seizure relapse by 5 years and 23% met the criteria for drug resistant epilepsy.11 The number of AEDs failed due to lack of efficacy and duration of epilepsy for more than 5 years before entering remission were identified as risk factors for relapse and subsequent development of refractory epilepsy.87 Good early response to AED treatment is not always sustained in the long term. more effectively.

Epilepsy in the UK: misdiagnosis. Reynolds EH. Epilepsy and other convulsive disorders. Sillanpa follow-up. Ko TS. Solomon JK.369:919–26. Seidl R. Mody I. Gamble C. Elveback LR. Treatment of first tonic–clonic seizure does not improve the prognosis of epilepsy. Valenca MM.42:357–62. D’Agostino MD. Smith S. Gummitt R. 46. Shevell MI. Elwes RD. 12. Dooley J. Bamagous GA.20:369–75. ´nez-Jua ´ rez IE. 343 35. Neurology 1996. 11. Camfield C. 60.38:31–46. Bhatia S. Drug-resistant epilepsy. Brodie MJ. 32. 54. Zahn C. Falilat S.60:405–9. 39. Berkovic SF. 23. Dooley J. 28. 44. Beghi E. et al. Collaborative Group for the Study of Epilepsy.46:41–4. Brorson LO. Kwan P. Sillanpa seizure outcome and mortality of childhood-onset epilepsy. Kaamugisha J. 68.50(Suppl. Zhang X. Syngeniotis A. Berg AT. Lauder R. Mohanraj R. Rodin EA. Epilepsia 1992.47:802–6. Mattson RH. Oskoui M. Du Ron RM. Peters ACB. Annals of Neurology 2003. 67. Mohanraj R. Epilepsia 2009.365:2007–13. Ribeiro LT. Peeters E.14:514–20. Marson A. Gilli G. et al. Why does epilepsy become intractable? Prevention of chronic epilepsy. Shorvon SD. Antiepileptic drug therapy: the story so far. Balin B. 3. Nicolson A. Epilepsy in later life. Predictors of seizure outcome in newly diagnosed partial epilepsy: memory performance as a prognostic factor.49:991–8. Neurology 1996. Brodie MJ. Shorvon SD.86:1–14. Mathern G. Peters B.14:318–23. Current Opinion in Neurology 2001.66:1233–9. 22. Kwan P. Prognosis of juvenile myoclonic epilepsy is related to endophenotypes. Illinois: Thomas Springfield. 131:938–44. Beghi E. et al. Diagnosing refractory epilepsy: response to sequential treatment schedules. 43. 58. McCagh J. Prognosis of epilepsy: a review and further analysis of the first nine years of the British National General Practice Study of Epilepsy. Gatiti S. 12):29–31. Picot M-C. Kalviainen R.R. Epilepsy. Arzimanoglou A. Schachter SC. Brouwer OF. Epilepsy & Behavior 2006. 56. Epilepsia 1996. Mohanraj.110:1245–51. Camfield C. Annals of Neurology 2002. Hauser WA. The natural history of epilepsy: an epidemiological view. Paolini E. Groh C. Hippocampal atrophy and T2-weighted signal changes in familial mesial temporal lobe epilepsy. Annegers JF.28: 97–106. Sander JW. Brodie MJ. Factors predictive of outcome in childhood epilepsy. Patients with refractory seizures. Predictors of intractable epilepsy in childhood: a case–control study. Epilepsia 1997. 57.20:729–37. Epilepsia 2002. New England Journal of Medicine 1984. Lopes-Cendes I. Annals of Neurology 2000. 63.J. Italy. Devinsky O. A prospective study of the requirement for and the provision of epilepsy surgery in the United Kingdom. Lancet 1991. Course and outcome of childhood epilepsy: a 15-year follow-up of the Dutch Study of Epilepsy in Childhood. Shinnar S. Aikia M. Leach JP. Shorvon SD. Seizure relapse and development of drug resistance following longterm seizure remission. Epilepsia 1993. 40. Patterns of treatment response in newly diagnosed epilepsy. Bailey JN.51:1189–97. Theodore WH. Remission of seizures in a populationbased adult cohort with a newly diagnosed unprovoked epileptic seizure. Goodridge DM. Neurology 2001. 45. Alonso ME. Long-term outcome is unchanged by antiepileptic drug treatment after a first seizure: a 15-year follow-up from a randomized trial in childhood. Anderson VE. Prognosis for total control of complex partial and secondarily generalized tonic–clonic seizures. Broglin D. 51. Epilepsia 2010. Clinical Neurophysiology 1999.40(Suppl. Gowers WR.37:24–30. Jacoby A. Epileptic seizures in a population of 6000. Pirker S. Westendorp RGJ. Webster RI. 25. Juvenile myoclonic epilepsy subsyndromes: family studies and longterm follow-up. Solari A. Progress in Brain Research 2002. 34. 53.19: 650–5. Epilepsy. Kwan P. Schiller Y. Johnson AL. Stephen LJ.43:662–3. 16. Beckerman L. Lopez-Riz M. Kelly K. Levy SR. 10.346:140–4. European Journal of Neurology 2006.37:701–8.8:1019–30. Outcomes in newly diagnosed localization-related epilepsies. Norrie JD. mistreatment. Shinnar S. 5): 1–148. Westbrook E. psychosocial and cognitive functioning. Cockerell OC. Berg AT.8:434–7. Wiebe S. Drexel M. Epilepsia 1988. Wranne L. Prognosis of epilepsy in newly referred patients: a multicenter prospective study of the effects of monotherapy on the long-term cause of epilepsy. Levy SR. Tognoni G. Epilepsy Research 2007. Remission of seizures and predictors of intractability in long-term 27. Geerts A. 64. Brodie MJ. Kwan P. Reynolds EH. Elwes RD. Johnson AL. Shorvon SD.52:132–6. 15. Two year remission and subsequent relapse in children with newly diagnosed epilepsy.11:201–4. Holmes GL. Arzimanoglou A. 14. Seizure 2011. Johnson AL. 52. Gordon K. Kalviainen R. Shorvon SD.99: 329–33. Brodie MJ. M. Brodie MJ. Musicco M.47:68–76. Kobayashi E. Camfield PR. Epilepsia 1987. Granieri E. Neurology 1997.44:673–6.51:1069–77. Collaborative Group for the Study of Epilepsy. Sander JW. 53:413–6. Remission of epilepsy: results from the National General Practice Study of Epilepsy. Dooley J. Outcome of childhood epilepsy: a population-based study with a simple predictive scoring system for those treated with medication. Hippocampal atrophy. Factors predicting prognosis of epilepsy after presentation with seizures. Brodie / Seizure 22 (2013) 333–344 2. Caboclo LO. Childhood absence epilepsy: evolution and prognostic factors. Tola MR. EEG and clinical predictors of medically intractable childhood epilepsy. Bouma PAD. Martı et al. The outcome of absence epilepsy. Epilepsia 2001. 38. Seizure 2005. Mohanraj R. Chadwick D. Cockerell OC. Testa FM. Journal of Pediatrics 1993.122:861–8. Norrie J. Epilepsia 2003. The process of epileptogenesis: a pathophysiological approach. Sander JW. Sander JW. Camfield P. 42.42:1253–62. Medina MT. Seizureassociated hippocampal volume loss: a longitudinal magnetic resonance study of temporal lobe epilepsy. Lancet 2005. Hatta J. Andermann F. 19. Gordon KE. Shorvon SD. Epilepsia 2005.20:42–8. 18. Brodie MJ. Does the number of seizures before treatment influence ease of control or remission of childhood epilepsy? Not if the number is 10 or less. 21. Schmidt D. Seizure 2010. Rich SS.46:1796–801. Neurology 2003. Brodie MJ. Camfield P. Mohanraj R. Galimberti D.78:1548–54. Journal of Neurology Neurosurgery and Psychiatry 2004. 33. The prognosis of patients with epilepsy. Johnson AL. Wirrell EC. 65. Brain 2008. Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology 1998. A meta-analysis. Sills GJ. Berg AT. Watson C. Long-term prognosis in childhood epilepsy: survival and seizure prognosis. Clinical and neuroimaging features of good and poor seizure control with mesial temporal lobe epilepsy and hippocampal atrophy. Neurology 1996.33:45–51. 4. Epilepsy Research 1999. Barry SJ. Lancet 1995. Beckerman B. Prognosis of epilepsy in newly referred patients: a multicenter prospective study. 8. Camfield CS. et al. Semah F. Grosso S.337:406–9. Dooley JM.287:645–7. Yacubian EM. Does antiepileptic drug therapy prevent the development of ‘‘chronic’’ epilepsy? Epilepsia 1996. Casetta I. Jackson GD. 24.42:1025–30. Fisk E. Engel J. Epilepsia 2001. 47. et al. Hippocampal sclerosis is a progressive disorder: a longitudinal volumetric MRI study. Juul-Jensen P. Sander JW. Seizure recurrence after a first unprovoked seizure: an extended follow-up. Feksi AT. Neurology 1996. Brodie MJ. Adam C. Berg AT. 66. Brouwer O. De Carli C. ¨a ¨ M. Smith DF. et al. Lhatoo SD. Monetti VC. French J. Guilhoto LM. 29.340:1565–70. 20. Hitiris N. Smith-Rapaport S. Brain 2006.47:912–8. 5.135:279–95. Shinnar S. Pharmacological outcomes in older people with newly diagnosed epilepsy. McEvoy AW. A clinical and social analysis of 1020 adult patients with epileptic seizures.75:192–6. Neurology 2012.20:898–904. 50. and undertreatment? The Wrexham area epilepsy project. Seizure 2005. Immediate versus deferred antiepileptic drug treatment for early epilepsy and single seizures: a randomised controlled trial. Combining antiepileptic drugs – rational polytherapy? Seizure 2011. Prognosis of childhood epilepsy in newly referred patients. Collins JF. Seizure clustering during drug treatment affects 26. Shah J. Novotny EJ. Epilepsia 2010. Briellmann RS. Mervaala E. Matos ALM. Forsgren L. Freilinger M. Shinnar S. Berg AT. a prospective population-based study. et al. Cramer JA. King MA. Neuronal plasticity in animal models and the epileptic human hippocampus. Brodie MJ. Epilepsia 1979. et al. 61.48:833–41. Camfield C. Levy SR. Johnson A. 36. Acta Neurologica Scandinavica 1964. Kitchen ND. 9. Remission of seizures and relapse in patients with epilepsy.311:944–7. Riekkinen Sr PJ. Annegers JF.51:1256–62. et al. 41.51:641–4. 17. Sperk G. Shorvon SD. Does the cause of localisation-related epilepsy influence the response to antiepileptic drug treatment? Epilepsia 2001. Archives of Neurology 2009. et al. Bookheimer SY.75:1376–81. Kim L. Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Smith-Rapeport S. 1881. et al. Filho GM. Early predictors of intractability in childhood epilepsy: a community-based casecontrol study in Copparo. Comprehensive primary health care antiepileptic drug treatment programme in rural and semi-urban Kenya. Neurology 2003. Lancet 1983.129:1269–80. Smith E. Arts WF. Andermann E. Brodie MJ. Lindsten H. Journal of Child Neurology 2005. 7. Journal of Child Neurology 1996. Kwan P. 48. Epilepsia 2003. Sander JW. 31.2:952–4. and febrile seizures in patients with partial seizures. 13. London: Churchill. The prognosis for seizure control in newly diagnosed epilepsy. Stenlund H.28:324–30. Goodridge DM. Camfield P. Salmenpera T. Epilepsy Research 2009. Baker GA. Gordon K. Velasco TR. Cockerell OC.56:1445–52.60:538–47. . 30. Berkovic SF. British Medical Journal 1983. Acta Neurologica Scandinavica 1999. Smith B. 6. Testa FM. Early development of intractable epilepsy in children: a prospective study. Epilepsia 1987. 59. Hauser WA. Stroink H. 55. Reynolds EH. Do recurrent seizures cause neuronal damage? A series of studies with MRI volumetry in adults with partial epilepsy. Hart YM. Farnetani M. Elder AT. Sales LV. Stephen LJ.13:277–82. Van Dijk JG. New England Journal of Medicine 1999. Hauser WA. Early treatment and prognosis of epilepsy. Brodie MJ. 1968. epilepsy duration. Neurology 1990. Di Bartolo RM. 62. Viani F.40:1163–70.29:236–43. Predictors of pharmacoresistant epilepsy. Hauser E. ¨a ¨ M. Lancet Neurology 2009. 49.34:930–6.37:159–67. MacDonald BK. Neurology 1999.44:807–14. Bazzotti S.14:187–92. Andrade-Valenca LP. Lin K. Sills GJ. Guaranha MS. 37. Practice parameter: temporal lobe and localized neocortical resections for epilepsy. Fuerst D. Shorvon SD. Dalby NO. Long-term prognosis of typical childhood absence epilepsy: remission or progression to juvenile myoclonic epilepsy. Shah A. New England Journal of Medicine 2011. II: Treatment and prognosis. Vezzosi P.

Ma MS. Neuropsychiatric symptomatology predicts seizure recurrence in newly treated patients. Schmidt D. Early seizure frequency and aetiology predict long89. Callaghan BC. Seizure 2009. Sheffield LJ.32:42–50. Mechanisms of human inherited epilepsies. Epilepsia 2012. D’Souza R. 81. Dooley JM. Jackson GD. Kanner AM. Results of treatment changes in patients with apparently drug-resistant chronic epilepsy. Progress in Neurobiology 2009. Neurology 2009.51:37–42. Newly diagnosed epilepsy and pharmacogenomics research: a step in the right direction? Epilepsy & Behavior 2011. Jones NC. the bad and the in-between. Dubois MF. ABC transporters in drug-refractory epilepsy: limited clinical significance of pharmacogenetics? Clinical Pharmacology and Therapeutics 2010. Peters AC. ¨ nen A. Quantifying response to antiepileptic drugs: effect of past treatment history. Ollivier ML. Newton MR. Pitka representative view of the European scientific community.23:1250030. Cortical excitability and refractory epilepsy: a three-year longitudinal transcranial magnetic stimulation study. Krajinovic M. Jackson GD. Sammel MD. Levy SR. 96.57: 2259–64. 87. . Zimprich F. Annals of Neurology 2007. ´ S. Stomb BL.75:1015–21. Epilepsia 2009. l1):21–7.9:27–9. Epileptology of the first seizure presentation: a clinical electroencephalographic and magnetic resonance imaging study of consecutive patients.132:989–98. Epilepsia 2012. Geerts AT. Szoeke CEI. Badawy RAB. Aikia diagnosed partial epilepsy. Epilepsia 2011. Kwan P. Kwan P. ¨a ¨ M. 79. et al. Early identification of refractory epilepsy. Epilepsy Research 2001.87:15–8. Perucca E. Del Felice A. 97. French JA. Defining early seizure outcomes in pediatric epilepsy: the good. 102. ¨a ¨ M. Berkovic SF.129:617–24. Berkovic SF.342:314–9. Pediatric Neurology 2009. et al.127:1774–84. 98. New therapeutic opportunities in epilepsy: a genetic perspective. Brodie / Seizure 22 (2013) 333–344 86. Remission and relapse in a drug-resistant epilepsy population followed prospectively. Schmitz PI. Carmant L.70:54–65. Bogliun G. New England Journal of Medicine 2000. Berg AT. Remission of epilepsy after two drug failures in children: a prospective study. Berkovic SF. Brain and Development 2010. Response to first drug trial predicts outcome in childhood temporal lobe epilepsy. Engel JRJ. If a first antiepileptic drug fails to control a child’s epilepsy.352:1007–11. La Neve A.344 R.62:382–9. Hesdorffer D. Reid CA. Pizzardi G. Callaghan B. Neuronal migration disorders: clinical. 88. Smith-Rapaport S. 73. Byrne R. Parisi P. Riekkinen PJ. Petrov S. Schlesinger M. Epilepsia 1999. Luciano AL.87:41–57. Neurology 2008. Salzberg MR. Najjar Y. 71.62: 375–81. Galanopoulou AS. Lo genetics on the treatment of epilepsy. et al. Testa FM. Silvapulle MJ. 72. Petrov S. Hauser WA. 83. 94. Mitchell A. what are the chances of success with the next drug? Journal of Pediatrics 1997. The clinical impact of pharmaco92. Research priorities in epilepsy for the next decade – a 95. 82. ¨ scher W. De Palo A. Brouwer OF. 80. van Donselaar CA. Boero G. Camfield PR. Ying W. Neurology 2001. Iannetti P. current opinions. Foong J. King MA. et al. Brodie MJ. Depression and epilepsy: a bidirectional relation? Epilepsia 2011. Sillanpa prospective. Chicharro A. Levy SR. Current Opinion in Neurology 2009. Beckerman B. 69.98:421–33. Likelihood of seizure remission in an adult population with refractory epilepsy. Epilepsy Research 1995. Schiller Y. Hesdorffer D. Schmidt D. Brain 2006. Baulac M. Sillanpa term medical outcome in childhood-onset epilepsy. Fang F. Berg AT. Gordon K. Johnson MR. Macdonell RAL. Reid CA. Brain 2004. Kanner AM. 75. Ka ¨ lvia ¨ inen R. 85.131:821–4.50:1–23. et al. Jackson GD. Fox Z. Petrovski S. Pharmacology & Therapeutics 2010.128: 274–80. Huggins RM. 99. Treatment of febrile seizures: historical perspective. Rodemer W.72:793–9. A lifetime psychiatric history predicts a worse seizure outcome following temporal lobectomy. M. Fitt GJ. Predictors of psychiatric and seizure outcome following temporal lobe epilepsy surgery. 76. Zou LP. Risk factors for valproic acid resistance in childhood absence epilepsy. 78. Epilepsia 2010. Anand K. Ding CH. Brouwer OF. Marini C. Klotz U. Cleary RA. 50:571–83. Moshe Identification of new epilepsy treatments: issues in preclinical methodology. Del Balzo F. Acta Paediatrica 2009. Arts WF. Frey M. Stroink H.52:619–26. Annals of Neurology 2009. 100. Stroink H. International Journal of Neural Systems 2003. et al. Brain 2009. 101. Course and prognosis of childhood epilepsy: 5-year follow-up of the Dutch study of epilepsy in childhood.22:3–8. 93. and potential future directions. Hauser WA. Cossette P. The Dutch study of epilepsy in childhood. Shorvon SD. Annals of Neurology 2007. 84. long-term population-based study. Cascorbi I. neuroradiologic and genetics aspects. Neurology 2010. et al. Pollard J. Epilepsia 2009.40: 726–34. Natural history of treated childhood-onset epilepsy: 90. Camfield CS. Testa FM. Brodie MJ. Tan NCK. Thompson PJ. The early prognosis of epilepsy in childhood: the prediction of a poor outcome. Early versus late remission in a cohort of patients with newly diagnosed epilepsy. Arts WF. Lux AL. Ding YX.22:157–64. Verbal learning and memory in newly 74. 70.22:150–6. Peeters EA. Brodie MJ.52(Suppl. Sisodiya SM. Effectiveness of the first antiepileptic drug in the treatment of pediatric epilepsy. Staley KJ. Dlugos DJ. Shinnar S. Buckmaster PS. Farrar JT. Definition of refractory epilepsy: defining the indefinable? Lancet Neurology 2010. ¨ M.J. Mohanraj. Beghi E.41:22–6. 77.18: 690–4. Kwan P. Schmidt D. Lancet 1998. Boudewyn Peters AC. Genetics of antiepileptic drug resistance.53: 1705–12. Wuu J.53:571–82.43:75–84. Spalice A. Nicita F. 91.65:510–9.