Early-onset epilepsy, cognition, and behaviour: continuity and challenge


Correspondence to: Francesca.Cormack@mrc-cbu.cam.ac.uk


Please also see the reply to this letter by 964–965.

SIR–Childhood epilepsy, particularly when it begins early in life, is associated with an increased risk of cognitive and behavioural impairment.[1, 2] While poor cognitive outcome is not easily explained by more extensive visible neuropathology,[1] the early years of cognitive development provide a critical foundation for the later emergence of higher-order cognitive skills.[3] This early development can be disrupted by epilepsy, thereby affecting the subsequent developmental trajectory into mid-childhood and beyond.

Although much research has been retrospective or cross-sectional (examining school-aged children or adults), it has been previously noted that behavioural features indicative of cognitive impairment are present in early childhood, often before formal cognitive testing is undertaken,[1] suggesting that early identification of children at risk is possible. The recent paper of Berg et al. builds on these observations, and offers a thorough prospective assessment of the behavioural, academic, and cognitive status of children with early-onset epilepsy.[4] By making use of a large and well characterized cohort of children who appeared to be otherwise neurologically unremarkable, followed prospectively over 9 years, the researchers address many of the outstanding issues in the literature such as the contribution of seizure load and medication. The authors provide clear evidence that behavioural abnormalities in the preschool years can predict cognitive and academic difficulties in mid-childhood. It is particularly interesting to note that the adaptive behaviour scale used predicts academic performance (e.g. reading comprehension) even when the correlation between behaviour and IQ is controlled for. This suggests that the early assessment used (Vineland Adaptive Behavior Scales) is tapping constructs, such as language, which are not well assessed by IQ but are nevertheless important determinants of performance in children with epilepsy.

Such an observation, highlighting continuities in behaviour and development, leads to several interesting research challenges. Firstly, there is the need to examine structural and functional brain developmental in children with epilepsy in more detail. Although the Berg et al. report that the children appeared normal on clinical imaging, this does not preclude subtle changes which may be detected using different imaging methods. Secondly, the study focuses on those children with an IQ higher than 70, who are predominantly in mainstream schooling (albeit with additional educational support). This allows standardized cognitive testing and the sophisticated approach used. However, it is important to note that 13% of the study population from which their sample was taken had an IQ of less than 70, suggestive of severe learning difficulties. Such a percentage is understandably less than that seen in surgical samples,[1] but it nevertheless represents a large number of children whose cognitive development is significantly impaired, and for whom educational and occupational opportunities are likely to be severely limited. There is a need to explore the developmental and neural trajectories of these children.

The Berg et al. data suggest that early behavioural problems tend not to resolve spontaneously, but rather herald continuing cognitive and academic difficulties. Therefore, research on interventions, particularly for those children at risk of the most negative outcomes, is needed. For example, it has been suggested that in early-onset symptomatic focal epilepsy, early surgical resection of the epileptogenic cortex may be beneficial.[5] Indeed in the Berg et al. study, remission and not being on an antiepileptic medication at follow-up were both associated with better cognitive outcome, suggesting that seizure-freedom permits adaptive plasticity to compensate to a certain extent for the early burden.[6] Nevertheless, additional cognitive or educational rehabilitation may be helpful to these children. Such interventions will be expected to be most effective for children with early-onset epilepsy when they take place early in childhood, but currently there is no consensus on their timing, nature, and effectiveness.

In conclusion, the paper by Berg et al. makes an important contribution to our understanding of the antecedents of academic and cognitive difficulties in children with early-onset epilepsy. However, important unanswered questions remain with regards to both the most cognitively impaired children and the best therapeutic approaches for optimizing cognitive outcome.


The preparation of this manuscript was supported by the UK Medical Research Council (MC-A060-5PQ00).