What is a ‘benign’ childhood epilepsy?
There are several childhood epilepsy syndromes that were reported as benign. The term ‘benign’ was used to characterize epilepsy syndromes of childhood with an expected remission without significant neurological sequelae in almost all affected patients. These syndromes also include childhood absence epilepsy (CAE), benign childhood epilepsy with centrotemporal spikes (BECTS), and Panayiotopoulos syndrome. Some other syndromes were also called benign, but these syndromes are less frequent (e.g. benign myoclonic epilepsy of infancy, now called myoclonic epilepsy of infancy).
The Report of the International League Against Epilepsy (ILAE) Commission on Classification and Terminology, 2005–2009, proposed to avoid using the term ‘benign’ to characterize any epilepsy syndrome. This is based on the better understanding and the data from research on comorbidities (cognition, psychiatric, and behavior comorbidities) showing that patients with these syndromes also exhibit a higher risk for cognitive and/or psychiatric involvement (Berg et al. 2010). This is also based on the long-term outcome. The proposal from the ILAE to define the epilepsy syndromes that tend to resolve spontaneously with time is to use the term “self-limited.”
Even if an ILAE report strongly suggests avoiding the word ‘benign’ to define any group of epilepsy, the same published report include the name ‘benign’ epilepsy with centrotemporal spikes for the rolandic epilepsy syndrome. However, the work on terminology and classification of the ILAE is an ongoing process; this point might be clarified when a new classification of seizures and epilepsy syndrome will be adapted by ILAE.
Beyond the discussion around terminology to define epilepsy syndrome, it might be helpful in clinical practice to avoid the term ‘benign.’ The health care professional might then pay more attention to finding patients with cognitive or psychiatric comorbidities leading to early recognition and management. This might have a significant impact on the risk of academic underachievement.
BENIGN EPILEPSY WITH CENTROTEMPORAL SPIKES
BECTS or rolandic epilepsy is the most frequent epilepsy syndrome in children (Shinnar et al. 1999; Zarrelli et al. 1999). These are sensorimotor seizures (warming/paresthesia perioral; clonic involvement of the face/upper limb; speech arrest) that occur most frequently after falling asleep or before awakening. In some patients, a bilateral involvement happens shortly after the start of the seizure. Most of the patients experience very few seizures. The ictal recording of a seizure in this syndrome is very rare. Interictal EEG shows spikes of the centrotemporal area (Wirrell 1998).
BECTS should be considered as a self-limited epilepsy syndrome because in mid-adolescence, the seizures disappear in 100% of cases (Bouma et al. 1997; Peters et al. 2001). The good outcome regarding the remission of seizure does should not imply that BECTS is ‘benign;’ treatment with antiepileptic drugs (AEDs) is not usual. In a review of the literature on the treatment of BECTS by AEDs, Hughes (2010) reported that two-thirds of 96 studies generally favored and one-third generally did not favor AED treatment for BECTS. Only two studies favored treatment for all patients with BECTS; many other investigations were in favor, with some restrictions such as treating only patients with early onset, multiple seizures at onset, or large numbers of seizures, especially generalized tonic–clonic seizures, and limiting treatment to 1 year (Hughes 2010).
In BECTS, the evaluation of IQ is usually normal (Pinton et al. 2006; Riva et al. 2007; Piccinelli et al. 2008; Tedrus et al. 2009). However, specific cognitive deficits are prevalent in BECTS. These deficits are the major concerns because they result in a risk of academic underachievement explaining why this syndrome is not always ‘benign.’ Indeed, detailed studies have reported the problems in specific cognitive domains, particularly in language and verbal memory (Piccinelli et al. 2008; Danielsson and Petermann 2009; Overvliet et al. 2011). Some studies report language deficits and normal nonverbal abilities (Riva et al. 2007; Goldberg-Stern et al. 2010), but more frequently, the nonverbal performance is also found to be lower in patients with BECTS than in controls (Piccinelli et al. 2008; Danielsson and Petermann 2009).
An attention deficit as well as inhibition problems are also frequently reported (Kim et al. 2014). These have been shown in several studies with poorer performance on auditory attention and inhibition subtests (Deltour et al. 2007; Verrotti et al. 2013). These types of cognitive/behavior problems are prevalent in BECTS. They are also common in childhood epilepsy more generally (Hermann et al. 2006).
These specific cognitive deficits are subtle differences shown by comparing a group of patients matched to controls. Children and adolescents with BECTS are also at a higher risk of poorer academic performance. In a study comparing 20 patients with BECTS to 21 controls, 45% of children with epilepsy had specific difficulties in reading or writing compared with 9% of controls, and 31% of patients had specific difficulties in mathematics compared with 6% of controls (Piccinelli et al. 2008). Overvliet et al. (2011) noted that 23% of 48 children with BECTS had a history of speech therapy.
Finally, the global psychosocial outcome of BECTS raises more concern about the benignity of this epilepsy syndrome. Using adverse social outcomes, Camfield and Camfield (2014) report mitigated long-term outcome (37±3.4y follow-up). They used the following criteria to evaluate the social outcomes: failure to complete high school, pregnancy outside of a stable relationship (<6mo), depression or another psychiatric diagnosis, unemployment, living alone, never in a romantic relationship for more than 3 months, and poverty. In BECTS, they reported 41% with one criterion or more and 22% with two criteria or more (Camfield and Camfield 2014).
CHILDHOOD ABSENCE EPILEPSY
Childhood absence epilepsy begins usually between 4 and 12 years of age with a peak onset at 6 years. This syndrome represents up to 10% of the pediatric epilepsies (Shinnar et al. 1999; Zarrelli et al. 1999). In this syndrome, only one type of seizure is observed: typical absence seizure. A history of febrile seizures is possible, but any other type of seizures rules out the diagnosis. Absences occur very frequently, often 20 to 100 times per day. Absence seizures are brief seizures (mean duration of 8s) clinically characterized by impairment of consciousness with automatisms in some patients, concomitant to a typical EEG pattern of generalized, bilateral, synchronous, symmetrical 3Hz spike–wave discharges (Matricardi et al. 2014).
Recently, a double-blind, randomized, controlled clinical trial (RCT) has compared ethosuximide, valproate, and lamotrigine in children with newly diagnosed CAE. Four hundred and fifty-three children were randomly assigned to treatment with ethosuximide (ESM; n = 156), lamotrigine (LTG; n = 149), or valproic acid (VPA; n = 148) (Glauser et al. 2010). After 16 weeks of therapy, the freedom-from-failure rates for ESM and VPA were similar (53% and 58%, respectively; p = 0.35) and were higher than the rate for LTG (29%; p < 0.001 for comparisons of ESM to LTG and VPA to LTG). During the RCT, there was no significant difference among the three drugs with regard to discontinuation because of adverse events. Pretreatment evaluation demonstrated that 36% of the included patients exhibited attention deficits. Notably, attention deficits persisted at 16 to 20 weeks even if seizure freedom was achieved (Masur et al. 2013). However, attention dysfunction was more common after the treatment with VPA than with ESM (49% of the children vs 33%; p = 0.03) (Glauser et al. 2010; 2013). A recent prospective study reports the long-term prognosis of CAE (Berg et al. 2014). The epilepsy remission seems to be linked to the initial treatment (Berg et al. 2014). Seventy-three children were initially diagnosed with CAE. The initial treatment was ESM in 41 and VPA in 18 children. Initial success rates were identical in these two groups (59% for ESM and 56% for VPA). Using a multivariate analysis, only the initial treatment predicts the long-term outcome. At the time of last follow-up contact, 38 (64%) of 59 participants were in complete remission (5y seizure free and 5y off medication), 31 (76%) of those first treated with ESM and 7 (39%) of those treated with VPA (p = 0.007). In 53 children followed for 10 years or more, 10-year remission was also higher in the ESM (76%) versus VPA (44%) group (p = 0.06) (Berg et al. 2014).
These data are in line with our view that CAE should be considered as possibly self-limited because only a proportion of the patients (about 60%) will have complete remission (Matricardi et al. 2014). Some patients seem to develop a lifelong epilepsy syndrome such as juvenile myoclonic epilepsy (Wirrell et al. 1996). In an additional study reporting the long-term outcome of the patients included in the large RCT conducted by Glauser et al. (median of follow-up: 7y), the occurrence of at least generalized tonic–clonic seizure was reported in 12% (n = 53) at a median age of 13.1 years (Shinnar et al. 2015).
Cognitive concerns in CAE have been raised by the very few available studies (Caplan et al. 2008). Most of the studies have evaluated a group of patients with various childhood epilepsy syndromes, and some of them focused on idiopathic generalized epilepsy. In a large group of children with epilepsy including 12 patients with CAE, Jambaque et al. (1993) reported that children with idiopathic generalized epilepsy performed poorly relative to controls on visual memory, while verbal memory was unaffected. The finding that general intellectual functioning is lowered in children with generalized idiopathic epilepsy syndromes is consistent with the results of later work (Nolan et al. 2003). In most of the studies, CAE was not a selectively studied group but was studied in the group of idiopathic generalized epilepsy syndromes, thereby restricting the ability to draw any conclusion restricted to only one syndrome. In a case–control study with CAE patients matched for sex and socioeconomic status, it has been shown that four-fifths of the patients had normal IQ scores (Pavone et al. 2001). However, patients had lower IQ than matched controls. In a more detailed evaluation, a lower level of general cognition, visual spatial skills, nonverbal memory, and delayed recall have been identified in children with CAE. While investigating the particular cases of early-onset CAE (onset before 3y of age), Chaix et al. (2003) suggested that the overall prognosis for children with early-onset absence epilepsy is poorer than a more typical CAE.
The pretreatment cognitive evaluation of 450 patients from the large RCT found 36% of children with attention troubles, whereas global cognition as well as specific cognitive domain were normal (Masur et al. 2013). This attention disturbance results in memory and executive function involvement and academic underachievement (Masur et al. 2013).
Comparing patients with CAE (n = 56) to patients with juvenile arthritis (n = 61), it has been shown that the patients with prior CAE were more likely to require special educational help (p < 0.02), to have below average academic performance (p < 0.01), or to repeat a grade (p < 0.005) (Wirrell et al. 1997).
Finally, the psychosocial outcome further suggests that CAE should not be considered as a ‘benign’ epilepsy. Comparing patients with CAE (n = 56) to patients with juvenile arthritis (n = 61), it has been shown that the patients with prior CAE were more likely to have an unplanned pregnancy (p < 0.001) and were less likely to graduate from high school (p < 0.005) or attend college/university (p < 0.001) (Wirrell et al. 1997).
Why do some patients present with learning problems?
There are multiple factors that might contribute to the learning problems reported in childhood self-limited epilepsy syndromes (Fig 9.1