Relation between stress-precipitated seizures and the stress response in childhood epilepsy

The majority of patients with epilepsy report that seizures are sometimes triggered or provoked. Stress is the most frequently self-reported seizure-precipitant. The mechanisms underlying stress-sensitivity of seizures are currently unresolved. We hypothesized that stress-sensitivity of seizures relates to alteration of the stress response, which could affect neuronal excitability and hence trigger seizures. To study this, children with epilepsy between 6 and 17 years of age and healthy controls, with similar age, sex and intelligence, were exposed to a standardized acute psychosocial stressor (the Trier Social Stress Test for Children), during which salivary cortisol and sympathetic parameters were measured. Beforehand, the relation between stress and seizures in children with epilepsy was assessed by (i) a retrospective questionnaire; and (ii) a prospective 6-week diary on stress and seizure occurrence. Sixty-four children with epilepsy and 40 control subjects were included in the study. Of all children with epilepsy, 49% reported that seizures were precipitated by acute stress. Diary analysis showed a positive association between acute stress and seizures in 62% of children who experienced at least one seizure during the diary period. The acute social stress test was completed by 56 children with epilepsy and 37 control subjects. Children with sensitivity of seizures for acute stress, either determined by the questionnaire or by the prospective diary, showed a blunted cortisol response to stress compared with patients without acute stress-precipitated seizures and healthy controls (questionnaire-based F = 2.74, P = 0.018; diary-based F = 4.40, P = 0.007). No baseline differences in cortisol were observed, nor differences in sympathetic stress response. The relation between acute stress-sensitivity of seizures and the cortisol response to stress remained significant in multivariable analysis (beta = -0.30, P = 0.03). Other variables associated with the acute stress response were the number of anti-epileptic drugs (beta = -0.27, P = 0.05) and sleep quality (beta = 0.30, P = 0.03). In conclusion, we show that children with acute stress-sensitive seizures have a decreased cortisol response to stress. These results support our hypothesis that stress-sensitivity of seizures is associated with alterations of the stress response, thereby providing a first step in unravelling the mechanisms behind the seizure-precipitating effects of stress. Increased knowledge of the relation between stress and seizures in childhood epilepsy might benefit our understanding of the fundamental processes underlying epilepsy and ictogenesis in general, and provide valuable clues to direct the development of new therapeutic strategies for epilepsy.