Prognosis of epilepsy is influenced strongly by many factors. Most individuals who develop epilepsy have a high likelihood of achieving remission. Eventually, many patients successfully discontinue the use of medications.56–60 Frequency of seizures, type of seizure, and number of seizure types are important predictors of outcome. Children with mental retardation (MR), cerebral palsy (CP), or both have lower remission rates than do children in the general population.57,58 An overall 5-year seizure-free remission rate of 56% was observed by age 22 years in patients with epilepsy who had MR but no CP.5 Remission rates were lower (47%) in patients with both MR and CP.
Sillanpaa et al.60 followed a population-based cohort of 245 patients with childhood-onset epilepsy. Of the 176 survivors available for follow-up in 1992, 30 years or more after the onset of their epilepsy, 143 (81%) had attained 5-year remission at some point, and 112 (64%) were in terminal 5-year remission on or off medications. However, only 45% of the remote symptomatic cases (the majority of whom had MR, CP, or both) were in 5-year terminal remission, as compared with 68% of those with epilepsy of cryptogenic etiology and 92% of those with idiopathic epilepsy. If remission status is limited to remission free of AED use, however, the differences are even more striking. Of the cohort, 47% were in 5-year terminal remission off medications, but only 19% of those with a remote symptomatic etiology met this criterion, as compared with 57% of those with epilepsy of cryptogenic etiology and 86% of those with idiopathic epilepsy. Remission could be achieved even many years after seizure onset.
To determine the prognosis of seizures in children with CP, children referred to a neuropediatric center for seizure disorder were divided in two groups according to the presence or absence of delayed motor development. 42 AEDs were discontinued if no seizures had occurred for at least 2 years. Seizure freedom without medication occurred more commonly in patients with epilepsy without CP:
|End point||Epilepsy with CP||Epilepsy without CP|
|Seizure freedom without antiepileptic drugs||14%||52%|
|Seizure relapse after drug withdrawal||62.5%||24.6%|
Similar findings were reported by Hosking and colleagues:46 Only 5% of children with CP or intellectual impairment reached “ultimate freedom” from seizures, as compared with 44% of children with epilepsy only.
Zafeiriou et al.34 determined the rates of relapse after discontinuation of AEDs following a 3-year seizure-free period in patients with epilepsy with and without CP. The relapse rate in patients with epilepsy without CP was 4.7%, as compared to 24.7% in the CP group. Interpretation of the results of this study is difficult, as the relapse rate in children without CP was lower than in any other series of medication withdrawal in children or adults.57,61,62
In studies focusing on medication withdrawal, the presence of MR or CP has consistently been associated with an increased risk of relapse after medication withdrawal.57,61,62 Shinnar et al.62 withdrew medications from 264 children who were seizure-free for 2 or more years. Recurrence risk was higher (rate ratio, 1.81) in those with epilepsy of remote symptomatic etiology (almost all of whom had MR or CP) than in those with cryptogenic or idiopathic epilepsy. Etiology remained a significant predictor of relapse on multivariable analysis. Although children with MR or CP had an increased risk of recurrence after medication withdrawal, approximately 50% remained seizure-free, indicating that medication withdrawal is feasible even in this group if they become seizure-free. In children with remote symptomatic epilepsy, presence of moderate to severe MR, but not mild MR or CP, was an additional risk factor for recurrence.
In a meta-analysis of the studies of medication withdrawal in both children and adults available up to 1993, Berg and Shinnar57 examined the effect of MR or CP on relapse rate. The presence of a remote symptomatic etiology (pooled relative risk, 1.55), of MR (pooled relative risk, 1.66), or of CP (pooled relative risk, 1.79) increased the relapse rate after medication withdrawal, as compared with patients with cryptogenic or idiopathic epilepsy.
Children with developmental disabilities who have epilepsy are not only less likely to achieve remission but also are more likely to develop medically refractory, intractable epilepsy.63–65 In a general hospital population of children with epilepsy,38 23% were considered to have intractable epilepsy, defined as the occurrence of more than one seizure per month despite adequate AED plasma levels. Among those with therapy-resistant epilepsy, 72% were children with neurologic disability (MR and CP). Fois et al.66 observed that 82% of children with epilepsy who did not respond to treatment were mentally retarded. While being treated with three major AEDs, these children experienced at least one seizure per month, up to several seizures per day. Conversely, of those who achieved complete clinical control of seizures on medications, only 38% had MR.
Independent predictive factors for intractability in children with MR and epilepsy are:63–65
Early age at onset of epilepsy and the occurrence of infantile spasms have also been associated with intractability in this population.63
People with epilepsy have an increased mortality when compared with the general population.67,68 The increased mortality is related to the etiology of the epilepsy.60,67,68 Both seizure frequency and the presence of another neurologic impairment have a strong influence on mortality risk.60,69
Few studies have evaluated mortality in patients with epilepsy and MR, CP, or both. In a population-based study of children with epilepsy,70 no increase in mortality was noted in children who showed no signs of neurologic impairment, but mortality was increased in children with MR or CP. Because children with MR have increased mortality even without epilepsy, interpreting these findings is difficult. Significantly increased mortality as compared to the general population has been observed in persons with MR (standardized mortality ratio [SMR], 1.6),71 and it was even higher when MR was associated with epilepsy (SMR, 5.0) and with both epilepsy and CP (SMR, 5.8) . The increase was more pronounced in the youngest in all groups (MR, MR plus epilepsy, MR plus epilepsy plus CP). In individuals with MR and epilepsy, mortality was increased for all seizure types but was more pronounced in persons with generalized seizures. Mortality increased with increasing number of seizures per year (SMR, 4.7 in those with 1 to 50 seizures per year, and 16.8 in those with more than 50 seizures per year). Whether mortality was directly correlated to the presence of epilepsy was impossible to determine. A detailed analysis of the circumstances at the time of death of patients with epilepsy showed that none died of a seizure.
Sillanpaa et al.60 examined the mortality in a cohort of 245 subjects with childhood-onset epilepsy who were followed up for 30 years. Among these patients, 44 deaths occurred, all but 5 in patients who either were not in seizure remission or had remote symptomatic epilepsy. Of 123 subjects with remote symptomatic epilepsy, 33 died, as compared with 11 deaths in the 122 subjects with cryptogenic or idiopathic epilepsy (p <.001). In the remote symptomatic group, death was definitely or probably related to seizures in 14 cases and to the underlying neurologic disorder in 19, including pneumonia in 12.
Among children with CP and epilepsy, 78% of those who died during a 5-year follow-up period (12% of all children) had severe or profound MR.29 It has been suggested that prognosis, including intellectual impairment, is determined largely by the underlying origin of seizures and not by the seizures per se.72,73
Reviewed and revised May 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.
© 2014 Epilepsy.com. All rights reserved.