wenz iD - Proefschrift Jolien van Campen by wenz iD

Stress and childhood epilepsy © Jolien S. van Campen, 2015 Layout and cover design: Wenz iD Printing: CPI - KONINKLIJKE WÖHRMANN Print Support: Proefschrift-AIO ISBN: 978-90-393-6346-1

All rights reserved. No part of this publication may be reproduced or transmitted in any form by any means, without permission in writing from the author. The copyright of the articles that have been published or have been accepted for publication has been transferred to the respective journals. The research in this thesis was financially supported by: Alexandre Suerman MD/PhD Stipendium • Bio Research Center for Children The author gratefully acknowledges financial support for the reproduction of this thesis by: UMC Utrecht Brain Center Rudolf Magnus • UCB Pharma

Stress and childhood epilepsy Stress en epilepsie op de kinderleeftijd (met een samenvatting in het Nederlands)

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof.dr. G.J. van de Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op dinsdag 23 juni 2015 des middags te 4.15 uur door Jolien Suzanne van Campen geboren op 16 december 1986 te Nijmegen

Promotoren:

Prof. dr. K.P.J. Braun Prof. dr. M. JoĂŤls

Copromotoren: Dr. F.E. Jansen Dr. P.N.E. de Graan

Content

General introduction CHAPTER 1

Introduction, aim and outline of this thesis

CHAPTER 2

Early life stress in epilepsy: a seizure precipitant and risk factor for epileptogenesis

PART ONE

Stress sensitivity of seizures in childhood epilepsy

CHAPTER 3

Stress sensitivity of childhood epilepsy is related to experienced negative life events

CHAPTER 4

Does Saint Nicholas provoke seizures? Hints from GoogleTrends

CHAPTER 5

Sensory modulation disorders in childhood epilepsy

PART TWO

Hormonal basis of stress sensitivity of seizures

CHAPTER 6

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

CHAPTER 7

Seizure occurrence and the circadian rhythm of cortisol, a systematic review

123

CHAPTER 8

Stress sensitivity of seizures influences the relation between cortisol fluctuations and interictal epileptiform discharges in people with epilepsy

137

PART THREE

Stress hormones and epileptogenesis

CHAPTER 9

Effects of repetitive mild stress and corticosteroid exposure on epileptogenesis after early life experimental febrile seizures in mice

157

Summary and general discussion CHAPTER 10

Summary and general discussion

181

Addendum References Samenvatting Dankwoord List of publications Curriculum vitae

207 241 245 251 253

General Introduction

CHAPTER

Introduction, aim and outline of this thesis

GENERAL INTRODUCTION | CHAPTER 1

INTRODUCTION TO THIS THESIS Epilepsy is one of the most common chronic neurological disorders, especially in childhood. At any given time, approximately 3.5 to 10 out of every 1000 children have active epilepsy (Shinnar and Pellock, 2002; Berg et al., 2013). The unpredictability of seizure occurrence and the effects of seizures as well as their treatment on cognition and behavior, have a large impact on the quality of life of children with epilepsy and their families. These children and their parents often mention their impression that seizures are precipitated by stress. They report increases in seizure frequency around stressful periods, such as birthdays, festivities, going to a new school, or moving homes. Also, they notice that seizures occur following acute moments of stress, for example when children are very excited, angry, anxious or nervous. Physicians are often asked for advice on how to cope with these stressors in the light of the chronic disease. However, how stress influences seizure occurrence is largely unknown. Increased knowledge on the relation between stress and epilepsy might improve counseling of patients with epilepsy and their caregivers, and provide directions for the development of new treatment strategies. The aim of the studies described in this thesis was to increase our understanding of the effects of stress on childhood epilepsy. This general introduction will first provide some background information on childhood epilepsy and the role of stress, and subsequently describe the aim and outline of this thesis. A more detailed overview of the current insights in stress and epilepsy is provided in Chapter 2.

A short history of epilepsy As some epileptic seizures look very dramatic and frightening, epilepsy has always captured the public imagination. Epileptic seizures were already described by the Babylonians in the world’s oldest medical handbook in the second millennium B.C. (Wilson and Reynolds, 1990). The term epilepsy has its origin in the passive tense of the ancient Greek verb ἐπιλαμβάνειν, meaning “to be seized/possessed”, and although Hippocrates in the 5th century B.C. already argued that epilepsy “has the same nature as other diseases” (Adams, 1939), epilepsy has been viewed as a supernatural or mental disorder throughout the main part of the past millennia (Reynolds and Trimble, 2009). Driven by magical and religious beliefs, patients with epilepsy were considered to be possessed by gods, demons or evil spirits until the 17th and 18th century (Diamantis et al., 2010). Only then, the concept of epilepsy as a brain disease began to be widely accepted, but because seizures often occurred in the absence of brain lesions, epilepsy was considered a psychiatric rather than a neurological disease until well into the 20th century (Reynolds and Trimble, 2009). In the past century, knowledge on epilepsy has increased tremendously. However, the pathophysiological mechanisms of this disease are still not fully understood and many questions remain to be answered.

Introduction, aim and outline

CHILDHOOD EPILEPSY Patients with epilepsy have an enduring predisposition to generate epileptic seizures, caused by aberrant electrical activity in the brain. Although the brain can function normally in between seizures, the recurrent seizure-activity as well as the anti-convulsive treatment affect brain function at a neurobiological level and can have cognitive, psychological and social consequences. Epilepsy is a very heterogeneous disease with respect to seizure semiology, underlying brain pathology and prognosis.

Definition of epilepsy Epilepsy is a disease of the brain defined by any of the following conditions (Fisher et al., 2014): • The experience of two or more unprovoked or reflex seizures >24 hours apart • One unprovoked or reflex seizure and a probability of further seizures similar to the general recurrence risk after two unprovoked seizures (≥60% over the next 10 years) • Diagnosis of an epilepsy syndrome

Etiology and pathophysiology Epileptic seizures are caused by excessive or synchronous neuronal activity in the brain. While anyone can experience a seizure under extreme conditions, patients with epilepsy have an increased seizure susceptibility because of a disrupted balance between neuronal excitation and inhibition. This can result from a variety of genetic, structural or metabolic causes, although in a large percentage of patients the etiology of the pathological lowered seizure threshold is unknown (Berg and Scheffer, 2011). The underlying processes promote neuronal hyperexcitability and hypersynchrony at the cellular level, for example by changes in the receptors for glutamate and γ-aminobutyric acid (GABA), the main excitatory and inhibitory neurotransmitter respectively in the brain, and ion channels, i.e., membrane spanning proteins that selectively gate ions across the cell membrane and are important for resting membrane potential and the generation of action potentials (McCormick and Contreras, 2001; Chang and Lowenstein, 2003). In addition, alterations of neuronal circuits, neuronal structure and glial cell function can (among other factors) influence the balance between excitatory and inhibitory neurotransmission and consequently lead to epilepsy (Rho and Strafstrom, 2006). Epilepsy is especially common in young children, because in the immature and rapidly developing brain the balance between excitation and inhibition more easily shifts towards excitation for several reasons, including the excitatory role of GABA and the increased number of synapses in early life (Rho and Strafstrom, 2006). The cascade of neurobiological processes leading to epilepsy is called epileptogenesis. Importantly, epileptogenesis is not

GENERAL INTRODUCTION | CHAPTER 1

limited to the time before the appearance of spontaneous seizures, but continues during the course of epilepsy, and may contribute to progression of the disease (Pitkanen and Sutula, 2002; Williams et al., 2009).

Epileptic seizures Seizures are considered to originate at some point in the brain and spread through epileptic networks to neighboring or remote brain regions. Based on the mode of seizure onset, they are described as ‘generalized’, meaning that they occur in and rapidly engage bilaterally distributed neuronal networks, or ‘focal’, conceptualized as occurring in networks within one hemisphere (Berg et al., 2010). Although seizures can arise from any site in the brain, seizure foci are most often localized in the neocortex and the limbic system, particularly the hippocampus and amygdala (Berkovic et al., 2006). Depending on the localization of the epileptic focus and the spread of epileptic activity through the brain, seizures can present with a wide range of signs and symptoms, including motor, sensory, behavioral and autonomic manifestations. Seizures vary from nearly undetectable focal muscle jerks (myoclonic seizures) and absences to prolonged periods of unconsciousness with involuntary stretching followed by rhythmic muscle contractions and relaxations (generalized tonic-clonic seizures, previously defined as ‘grand mal’ seizures), that are classified based on seizure semiology and electroencephalographic (EEG) distribution. Together with other epilepsy characteristics, some epilepsies can be classified into specific electroclinical syndromes. These complexes of clinical features, signs and symptoms that together define distinctive clinical disorders are especially prevalent in childhood epilepsies. Once in a while, the classification of seizures and epilepsies needs to be updated to the current level of knowledge. In this thesis, we will follow the most recent classification proposal of the International League Against Epilepsy (Berg et al., 2010). Treatment and prognosis Epilepsy is usually treated with anti-epileptic drugs that aim to suppress epileptic neuronal activity, thereby preventing further seizure occurrence, but not curing the epilepsy. Antiepileptic drugs act mainly via voltage-gated ion channels, or via enhancing GABA-ergic or decreasing glutamatergic transmission (Howard et al., 2011). Anti-epileptic drugs can be applied in mono- or polytherapy. The first anti-epileptic drug results in seizure freedom in on average 62% of patients, but the success rate of add-on drugs decreases rapidly (Schiller and Najjar, 2008). Approximately one quarter of all children with epilepsy is considered pharmacoresistant, as they continue to experience seizures despite the use of multiple antiepileptic drugs (Berg et al., 2006; Kwan et al., 2010). Sometimes, specific drug therapies can be used to target the underlying pathology, for instance immunotherapy for epilepsy syndromes with a considered immunological pathogenesis. Non-pharmalogical treatment options that can be considered are the ketogenic diet, a high-fat low-carb diet that induces a state of ketosis, and the vagus nerve stimulator, an implanted programmable device that provides cycles of pulsed retrograde stimulation of the left vagus nerve. The exact mechanisms of action of these treatment options remain largely unclear. Both treatments

Introduction, aim and outline

rarely lead to complete seizure control, and only 40-50% of children achieve a reduction of more than half of their seizures (Morris et al., 2013; Sharma and Jain, 2014). Epilepsy surgery is the only curative treatment option for children with epilepsy, as it results in seizure freedom in 50-80% of operated children (Spencer and Huh, 2008; Ryvlin et al., 2014). Although previously used only for children with pharmacoresistant epilepsy and a focal structural etiology, the indications for epilepsy surgery are currently increasing (Cross et al., 2006; Wiebe and Jette, 2012; Ryvlin et al., 2014). The prognosis of childhood epilepsy largely depends on the specific epilepsy syndrome. In general, childhood epilepsy has a relatively good outcome with respect to seizure control, as of the children who respond to medication 70% remains seizure free after withdrawal of anti-epileptic drugs (Berg and Shinnar, 1994; Braun and Schmidt, 2014), and several childhood epilepsy syndromes are age-specific and can be outgrown (Shinnar and Pellock, 2002). However, children with epilepsy have an increased risk of sudden death and mortality, especially those with refractory seizures and a structural etiology (Callenbach et al., 2001; Shinnar and Pellock, 2002; Chin et al., 2011). Furthermore, recurrent seizures as well as anti-epileptic drugs can have detrimental effects on cognition and behavior, especially in the immature and developing brain (Oostrom et al., 2003; Hermann et al., 2012). Also, childhood epilepsy is frequently associated with psychiatric comorbidities, such as autism spectrum disorders, attention deficit and hyperactivity disorder (ADHD), depression, anxiety and psychosis (Pellock, 2004). Hence, epilepsy significantly impacts the quality of life of children with epilepsy and their caregivers, not only by the recurrent seizures, but also by the associated cognitive, behavioral, academic and social problems (Geerts et al., 2011; Stevanovic et al., 2011; Taylor et al., 2011).

Seizure precipitants In the majority of patients with epilepsy, certain endogenous or exogenous factors can decrease the seizure threshold and subsequently increase the risk of seizure occurrence (Spector et al., 2000; Kasteleijn-Nolst Trenite, 2012). Seizure precipitants should be distinguished from factors that provoke seizures in the non-epileptic brain, such as a severe metabolic disturbance, head trauma and infection, as non-epileptic individuals with such provoked seizures (also called reactive or acute symptomatic seizures) are not considered to have an enduring tendency to have recurrent seizures (Fisher et al., 2014). Some overlap exists between precipitated seizures and reflex seizures. In the latter, a specific stimulus, most often sensory external, consistently provokes a seizure (Kasteleijn-Nolst Trenite, 2012; Illingworth and Ring, 2013). Even though in reflex epilepsies sometimes all seizures are provoked, they are associated with an enduring abnormal predisposition of the brain to generate seizures and are therefore diagnosed as epilepsy (Fisher et al., 2014). The seizure precipitants most often reported are stress, sleep deprivation, fatigue, illness or fever and the menstrual cycle, see Figure 1. Although stress is the seizure-precipitant most often reported by patients (including children) or their caregivers, the underlying mechanisms of stress sensitivity of seizures are unclear.

GENERAL INTRODUCTION | CHAPTER 1

Figure 1. Seizure-precipitating factors reported by patients with epilepsy Percentage of patients with epilepsy reporting specific seizure precipitants. Values represent the mean percentage reported by different studies (Hayden et al., 1992; Antebi and Bird, 1993; Hart and Shorvon, 1995; Spatt et al., 1998; Frucht et al., 2000; Spector et al., 2000; Nakken et al., 2005; da Silva Sousa et al., 2005; Sperling et al., 2008; Fang et al., 2008; Pinikahana and Dono, 2009; dos Santos Lunardi et al., 2011; Ferlisi and Shorvon, 2014; Wassenaar et al., 2014). Only seizure precipitants reported by five studies or more are displayed.

Stress and childhood epilepsy Children with epilepsy might feel stressed in various situations, similar to non-epileptic children and adults. For example when they have to give a performance, compete in sports, are bullied or exposed to a changed environment such as going to a new school or moving homes. In daily life, the term ‘stress’ is most often used in a negative context and considered to decrease the quality of life. However, any threat can be considered a stressor and be subjectively experienced as stress, and stress itself is not necessarily bad. Stress induces physiological and psychological changes that optimize our response to threatening situations and help us to recover afterwards. The stress response, including the effects of stress on the brain, is mediated through two main stress systems: (1) the sympathetic part of the autonomic nervous system, which through the release of adrenaline and noradrenaline induces an increase in heart rate, blood pressure, respiration rate, alertness and vigilance, resulting in a state that is known as the ‘fight-or-flight response’ (Cannon, 1929); and (2) the hypothalamic-pituitary-adrenal (HPA) axis, resulting in the release of corticotrophin releasing hormone (CRH) from the hypothalamus, which promotes the secretion of adrenocorticotropic hormone (ACTH) from the pituitary gland that stimulates the adrenal cortex to synthesize and release corticosteroids (cortisol in humans and corticosterone in rodents). The HPA-axis is regulated through negative feedback, as corticosteroids down-regulate the release of CRH as well as ACTH, thereby dampening the stress response and restoring homeostasis

Introduction, aim and outline

Definition of stress

The concept of stress was introduced by Hans Selye in 1936, who described it as the “general alarm reaction of the organism”. He defined stress as “the non-specific response of the body to any demand placed upon it” and demonstrated the crucial role of the hypothalamic-pituitaryadrenal axis and the subsequent production of corticosteroids in the stress response. Since 1936, Selye’s definition of stress has been widely debated. Although there is no general consensus on its definition, in current biological and psychological research stress is generally considered a perceived threat of homeostasis, which can be psychological, such as social defeat, or physiological, threatening body or cell homeostasis. The threat itself is called a stressor. In this thesis, we acknowledge this last definition, and focus on the effects of psychological stress.

(Sapolsky et al., 2000; Tsigos and Chrousos, 2002). While the sympathetic activity increases very fast and returns to baseline tens of minutes after the stressful situation has passed, the response of the HPA-axis is somewhat slower in onset and longer lasting. Corticosteroid levels peak at 20-30 minutes after stress exposure, and stay elevated for approximately two hours (de Kloet et al., 2005). Their effects can last even longer via regulation of gene expression (Ringold, 1985; Beato et al., 1996). Corticosteroids affect multiple systems in the body, such as glucose metabolism and immunomodulation. They also influence neuronal functioning in the brain in a time- and region-specific manner (Joëls et al., 2007; de Kloet et al., 2008), resulting in changes in higher order brain functions such as learning and memory (Schwabe et al., 2012). The neuronal responses include alterations in neuronal excitability (Joëls, 2009; Joëls et al., 2012), that are likely to influence seizure susceptibility in patients with epilepsy. When stress endures and becomes chronic, this can disturb regulation of the HPA-axis. Chronic disturbance of HPA-axis regulation affects the brain at a structural and functional level and subsequently increases the risk of various neuropsychiatric diseases (Lupien et al., 2009; Gunnar et al., 2009; Andersen and Teicher, 2009). Chronic stress, especially when experienced in early life, has also been shown to increase the risk of developing epilepsy (Christensen et al., 2007; Moshe et al., 2008; Shang et al., 2010).

Stress versus arousal While arousal is often part of the stress response, there is a difference between the two. The term ‘arousal’ is generally used to describe the fast and short-termed increase in physical and mental alertness that is mainly mediated by adrenaline and noradrenaline after activation of the sympathetic part of the autonomic nervous system. Stress, on the other hand, reflects the more long-lasting effects that also depend on the activation of the hypothalamic-pituitary adrenal axis and is often quantified by measuring levels of the stress hormone cortisol.

GENERAL INTRODUCTION | CHAPTER 1

Although stress is the seizure-precipitant that is most often reported, and stress has been shown to increase epilepsy prevalence, the pathophysiological mechanisms underlying the effects of stress on the occurrence of seizures (ictogenesis) and the development of epilepsy (epileptogenesis) remain largely unclear (van Campen et al., 2014). Increased knowledge on the relation between stress, epilepsy and epileptogenesis can improve counseling of children with epilepsy and their caregivers on how to handle stressors and guide the development of new treatment strategies.

AIM The main aim of the studies described in this thesis was to increase our understanding of the pathophysiological mechanisms that underlie the effects of stress on epilepsy and epileptogenesis in children with epilepsy. Specifically, our objectives were to: I. Improve the knowledge on stress sensitivity of seizures in childhood epilepsy in relation to patient and disease characteristics II. Provide a first step towards elucidating hormonal mechanisms responsible for stress sensitivity of seizures III. Explore the effects of stress on epileptogenesis

OUTLINE As a start, a more detailed introduction on stress and the effects of early life stress on epilepsy is provided in chapter 2, where the role of stress as a seizure precipitant and risk factor for epileptogenesis is discussed in a review of the literature. The following chapters describe the studies that were performed in the light of this thesis. These are organized into three parts that correspond to the objectives mentioned above.

Part I. Stress sensitivity of seizures In the first part of this thesis we investigated the nature of stress sensitivity of seizures in childhood epilepsy and its associations with patient and disease characteristics, as these might provide insight in the underlying mechanisms. In chapter 3, the nature of stress sensitivity of seizures is described, as well as differences between children with and children without stress-precipitated seizures, all based on questionnaire data. In chapter 4, we used a big data approach to assess the effects of stress on seizure occurrence, by analyzing the association between a national stressful event for children, Saint Nicholas, and epilepsy information-seeking behavior on the Internet, as an indirect estimate of seizure incidence. As stress is often combined with sensory overload, in chapter 5 we studied sensory modulation disorders in childhood epilepsy and the association between neuronal thresholds for sensory stimuli and stress sensitivity of seizures.

Introduction, aim and outline

Part II. Hormonal basis of stress sensitivity of seizures The second part of this thesis describes the hormonal basis of stress sensitivity of seizures. In chapter 6, the relation between stress and seizures in children with epilepsy was assessed by retrospective as well as prospective information, and the physiological response to a standardized stressor was compared between children with epilepsy with and without stress-precipitated seizures and healthy controls. As cortisol is released in a circadian pattern, in chapter 7 we evaluated the relation between the diurnal rhythm of cortisol and seizure occurrence by a systematic review of the literature. Next, in chapter 8, the influence of cortisol on epileptiform neuronal activity in epilepsy was determined by assessing the correlation between diurnal or ultradian cortisol fluctuations and epileptiform discharges in adults with epilepsy. Part III. Stress, stress hormones and epileptogenesis Part three of this thesis focuses on the effects of stress and stress hormones on epileptogenesis. In chapter 9, we induced epileptogenesis in young mice with the experimental febrile seizure model, and evaluated the effects of repetitive mild stress or corticosterone exposure on hippocampal morphological and functional measures, as proxy for epileptogenesis. In the final section of this thesis, chapter 10, the experimental findings of these clinical and preclinical studies are summarized, integrated and discussed. Finally, these findings are translated into clinical perspectives and future directions.

PART

ONE

Stress sensitivity of seizures in childhood epilepsy

CHAPTER

Early life stress in epilepsy: a seizure precipitant and risk factor for epileptogenesis Epilepsy & Behavior 2014

Jolien S. van Campen Floor E. Jansen Pierre N.E. de Graan Kees P.J. Braun Marian JoĂŤls

PART ONE | CHAPTER 2

ABSTRACT Stress can influence epilepsy in multiple ways. A relation between stress and seizures is often experienced by patients with epilepsy. Numerous questionnaire and diary studies have shown that stress is the most often reported seizure-precipitating factor in epilepsy. Acute stress can provoke epileptic seizures, and chronic stress increases seizure frequency. In addition to its effects on seizure susceptibility in patients with epilepsy, stress might also increase the risk of epilepsy development, especially when the stressors are severe, prolonged, or experienced early in life. Although the latter has not been fully resolved in humans, various preclinical epilepsy models have shown increased seizure susceptibility in na誰ve rodents after prenatal and early postnatal stress exposure. In the current review we first provide an overview of the effects of stress on the brain. Thereafter, we discuss human as well as preclinical studies evaluating the relation between stress, epileptic seizures and epileptogenesis, focusing on the epileptogenic effects of early life stress. Increased knowledge on the interaction between early life stress, seizures and epileptogenesis could improve patient care and provide a basis for new treatment strategies for epilepsy.

Early life stress in epilepsy

INTRODUCTION Epilepsy is a heterogeneous condition in which different underlying pathologies can cause abnormal excessive synchronous neuronal activity in the brain, resulting in epileptic seizures (Fisher et al., 2005). In the majority of patients with epilepsy, seizures can be triggered or provoked by various endogenous and exogenous factors, such as sleep deprivation, fever, light flashing, hyperventilation or alcohol (Hayden et al., 1992; Loyning, 1993; Hart and Shorvon, 1995; Cull et al., 1996; Spatt et al., 1998; Frucht et al., 2000; Spector et al., 2000; Haut et al., 2003; Nakken et al., 2005; da Silva Sousa et al., 2005; Sperling et al., 2008; Fang et al., 2008; Pinikahana and Dono, 2009; Koutsogiannopoulos et al., 2009; dos Santos Lunardi et al., 2011; van Campen et al., 2012). Some of these precipitating factors are more often found in patients with an epileptic focus localized in specific brain regions, a specific electroclinical syndrome, or certain etiology, but most seizure precipitants are reported in a wide variety of patients. The seizure-precipitating factor most often reported by patients with epilepsy is stress; this pertains to both physical stress, like physical exercise or illness, and psychological stress. The first reports on the relation between epilepsy and stress already date from over half a century ago (Levin, 1950; O’Neill, 1958; Stevens, 1959). In addition to its effects on seizure susceptibility, stress can also increase the risk of the development of epilepsy, especially when the stressors are severe, prolonged or experienced early in life. Stressors can alter (genetic) processes important for brain morphology and function or directly affect pathways involved in epileptogenesis, here defined as the gradual changes in molecular, cellular and network properties by which the normal brain develops the ability to generate recurrent spontaneous seizures (Beck and Gavin, 1976; Frye and Bayon, 1999; Edwards et al., 2002; Edwards et al., 2002; Edwards et al., 2002; Huang et al., 2002; Lai et al., 2006; Salzberg et al., 2007; Gilby et al., 2009; Jones et al., 2009; Sadaghiani and Saboory, 2010; Lai and Huang, 2011; Ahmadzadeh et al., 2011; Kumar et al., 2011; Yum et al., 2012; Qulu et al., 2012). In this paper, we will review the current literature on the effects of stress on epileptogenesis and epilepsy. We will discuss human as well as preclinical studies evaluating the relation between stress, epileptic seizures and epileptogenesis, focusing on the epileptogenic effects of early life stress.

STRESS AND THE BRAIN Stress is something we all experience every now and then and most people have an idea what the term ‘stress’ means. Although there is no general consensus on its definition, from a biological perspective stress is most often defined as the subjective experience of a threat of homeostasis (the threat itself is called a stressor) (Selye, 1950). The response to stress consists of adaptive changes in the organism’s physiology, resulting in a physical and psychological state optimal to reinstate homeostasis. Important in this response is the

PART ONE | CHAPTER 2

activation of the sympathetic nervous system and the hypothalamus-pituitary-adrenal (HPA-) axis, that jointly initiate the changes throughout the body that result in, amongst others, alertness and adaptation to the situation at hand. Although stress is often very adaptive in the short-term, it can potentially have long-term maladaptive consequences by changing the neuro-endocrine stress response as well as brain structure and function. This happens especially when stress is experienced chronically or in early life. Stress-induced changes in stress responsiveness and brain function were shown to be associated with increased vulnerability to diseases, including neurological and psychiatric illnesses. In this section, we will briefly discuss the hormonal stress response and its regulation, as well as the results of acute, chronic and early life stress on brain structure and function.

The stress response and its regulation Exposure to a psychological or physiological stressor activates two main pathways: the autonomic nervous system and the HPA-axis, leading to the secretion of catecholamines and corticosteroids respectively (Figure 1). The hypothalamus, important in regulating body homeostasis, has a key role in the initiation of both systems. Psychological stressors are firstly processed cortically and indirectly reach the hypothalamus via multiple nuclei in the limbic system, with an important role for the amygdala (Herman et al., 2003). Physical stressors like pain, inflammation and hypoglycemia activate the hypothalamus directly through monosynaptic projections from receptor cell populations located in the brain stem (Palkovits et al., 1980; Herman et al., 2003). Sympathetic nervous system Stress exposure activates within seconds the sympathetic part of the autonomic nervous system, while suppressing the parasympathetic innervation. This is initiated by neural control centers in the brain stem and hypothalamus, where sensory information from thoracic and abdominal viscera is integrated with information from the cerebral cortex and the limbic system (McCorry, 2007; Ulrich-Lai and Herman, 2009). From the hypothalamus and brain stem, neuronal projections run to preganglionic neurons in the brain stem and the lateral horn of the thoracic and upper lumber segments of the spinal cord (Baumann and Turpin, 2010). Postganglionic neurons project to the viscera, resulting in the release of the catecholamines adrenaline (mostly from the adrenal medulla) and noradrenaline (mainly from sympathetic nerve terminals), together inducing a state of arousal known as the fight-or-flight response (Cannon, 1929). The activation of the sympathetic nervous system is short-lasting, as parasympathetic activation quickly brings the system back to baseline conditions. Indirectly (via the vagal nerve), adrenaline from the adrenal gland can increase noradrenaline levels in the brain. Hypothalamic-pituitary-adrenal axis Activation of the HPA-axis is somewhat slower and induces a more long-lasting response to stress. Stress exposure stimulates the paraventricular nucleus of the hypothalamus to

Early life stress in epilepsy

Figure 1. Stress response Schematic overview of the stress response mediated by the sympathetic nervous system (left) and the hypothalamic-pituitary-adrenal axis (right). Upper-right inset: all of the hormones involved in the stress response reach the brain via the bloodstream (drawn line) or indirectly via the vagal nerve (striped line), and can therefore influence neuronal excitability. CRH corticotropin releasing hormone; ACTH adrenocorticotropic hormone.

release corticotrophin releasing hormone (CRH) and arginine vasopressine (AVP). Through the pituitary portal veins, these hormones reach the anterior pituitary where CRH stimulates the synthesis and secretion of adrenocorticotrophic hormone (ACTH), a process potentiated by AVP (Gillies et al., 1982). Via the bloodstream, ACTH reaches the adrenals and stimulates the adrenal cortex to produce and secrete corticosteroids (mainly cortisol in humans and corticosterone in rodents). This rise in corticosteroids, reaching peak levels 20-30 minutes after initiation of stress, influences cellular processes throughout the body, including the brain. Negative feedback at the level of the hypothalamus and pituitary causes the system to return to baseline levels in approximately two hours after cessation of the stressor. The HPA-axis is also regulated by input from other neurotransmitter systems and brain regions. The effects of HPA-axis activation on the brain are not exclusively caused by corticosteroids.

PART ONE | CHAPTER 2

Many hormones involved in the stress response, including CRH and AVP, can affect neurons and neuronal excitability locally (Figure 1). Furthermore, several other hormones and peptides play a role in the stress response and stress hormone regulation. For instance, stress rapidly increases the level of neuroactive steroids, i.e. steroid hormones that are synthesized throughout the brain (particularly in myelinating glial cells) and locally influence neurons and neuronal excitability. Moreover, some areas other than the hypothalamus are also capable of CRH production within milliseconds after stress exposure, e.g. the amygdala (Swanson et al., 1983), hippocampus (Chen et al., 2001) and locus coeruleus; in the latter area, CRH is known to stimulate the release of noradrenaline (Swanson et al., 1983; Chen et al., 2001; Valentino and Van Bockstaele, 2008). This extrahypothalamic CRH production is also controlled by negative feedback.

Importance of stress duration and timing The hormonal stress response and the subsequent effects of stress mediators on brain tissue depend on the duration and timing of the stressor, as described in more detail below. In short, when an organism faces an acute stressor occasionally, the stress response results in a temporary increase in stress hormone levels and change in brain function, which are both restored to baseline non-stress conditions by a negative feedback loop. However, if the stress exposure occurs repeatedly and stress hormone levels stay elevated over a longer period of time, this chronic stress can cause more permanent changes in stress hormone regulation and brain processes, and is associated with an increased risk on various chronic diseases. Both acute and prolonged stress in early life can have a severe impact on brain function and morphology, as in this time period the nervous system is still developing. Acute stress In acute stress conditions, production of the hormones described above changes the functioning of various organ systems. The effects of acute stress on the brain result for example in an increased alertness, and improved learning and memory for contextual aspects associated with the stressor. These changes can be largely explained by a regionspecific increase in glutamate transmission, due to stress hormone exposure. Areas with increased neuronal excitability after acute stress exposure include limbic regions and the medial prefrontal cortex (JoĂŤls et al., 2007). These changes help the organism to survive potential threats and to store relevant information for future use. Chronic stress Although the time domains for acute and chronic stress are not well-defined, prolonged activation of the stress system (roughly weeks to months in rodents, months to decades in humans) is a well-documented risk factor for many neuropsychiatric diseases. Animal studies have proven that chronic stress can cause long-lasting alterations in HPA-axis regulation. In turn, these persistent changes in stress hormone levels can have long term effects on the brain. Both increased and decreased levels of corticosteroids have been described as a result of chronic stress. Increased levels, most often associated with depression

Early life stress in epilepsy

and cognitive decline, are thought to result from impaired negative feedback of the HPAaxis. In rodents, chronic stress was shown to cause volume loss in the frontal lobe and hippocampus, dendritic atrophy in these areas and inhibited neurogenesis in the dentate gyrus; interestingly, a volume increase was reported in the amygdala, linked to enhanced dendritic arborization (Magarinos and McEwen, 1995; Gould et al., 1997; Mitra and Sapolsky, 2008; Shansky and Morrison, 2009; Lupien et al., 2009). After cessation of chronic stress, most of these changes were reversible in weeks. Similar brain volume changes were reported in association with decreased levels of corticosteroids in humans, as observed e.g. after severe childhood abuse and in patients with post-traumatic stress disorder. It is hypothesized that alterations in hormone levels and brain morphology in patients with post-traumatic stress disorder reflect a pre-existing risk factor for the development of stress related diseases, caused by genetics or early life adversity, rather than a direct effect of chronic stress itself, as reviewed by Lupien et al. (2009).

Early life stress Exposure to stress early in life can result in profound changes in brain function and the response to stress later in life. The definition of â&#x20AC;&#x2DC;early lifeâ&#x20AC;&#x2122; differs between studies. In this review, we take a life-time perspective and include the entire age range from conception until early adulthood, as this period is associated with significant brain development. Similar to chronic stress later in life, early life stress influences regulation of the HPA-axis. The direction of the effect seems to depend on the type of stressor, the timing and the species (Gunnar and Vazquez, 2001; Gunnar and Quevedo, 2008). For example, prenatal stress usually results in HPA-axis hyperactivity, with increased responsiveness to stressors and elevated baseline levels of cortisol or corticosterone and CRH (Phillips, 2007; Seckl, 2008). Stress during childhood can result in hyperactivity of the HPA-axis when induced by maternal separation or low parental care, but also in hypoactivity when stress is severe (Lupien et al., 2009). Epigenetic regulation of stress hormone receptors was shown to play a role in the lastinginfluence of early life stress on HPA-axis activity (McGowan et al., 2009). Despite the fact that changes in HPA-axis regulation after early life adversity are often long lasting, behavioral effects and hypocortisolism can still be reversed in a matter of months in humans (Gunnar and Quevedo, 2008) and weeks in rodents, e.g. by exposure to a stimulating environment (Bredy et al., 2004; Cui et al., 2006). As corticosteroids influence normal brain maturation, changed stress hormone levels as a consequence of early life events are expected to affect brain development. For instance, in rodents, early life stress has been shown to cause long-lasting neuronal hyperexcitability in some limbic regions, with less efficient inhibition by Îł-aminobutyric acid (GABA), due to conservation of an immature (excitatory) GABA receptor phenotype (Koe et al., 2009), permanently reduced hippocampal neurogenesis (Korosi et al., 2012) and decreased volumes of hippocampus and prefrontal cortex (Cohen et al., 2006; Teicher et al., 2006). The results of stress and stress hormones on brain function and development vary with the timing of exposure. For example, in humans stress exposure might have the largest effects on functional and structural development of the human hippocampus in the first two years

PART ONE | CHAPTER 2

of life, the frontal cortex during adolescence and the amygdala in the period from birth until early adulthood, in pace with the structures that undergo major changes at the time of stress exposure, a hypothesis proposed by Lupien et al. (2009). In humans, the effects of early life stress on brain function, morphology and HPA-axis regulation were shown to last until years after stress exposure, probably up to adulthood (Lyons-Ruth et al., 2000; Gutteling et al., 2005; Oâ&#x20AC;&#x2122;Connor et al., 2005). Severe or prolonged early life stress is thought to influence emotional and cognitive development (Stott, 1973; Glover, 1997), and increase the risk on a wide range of neuropsychiatric diseases, including depression, anxiety disorders, substance abuse and dementia (Andersen and Teicher, 2009; Lupien et al., 2009; Gunnar et al., 2009), but also epilepsy (Koe et al., 2009).

THE EFFECTS OF STRESS ON EPILEPSY Stress influences epilepsy at multiple levels. It is not only reported to affect seizure occurrence and seizure frequency, but can also influence the development of the disease in the first place. In this section we will first describe the results of human studies on stress, stress hormones and epilepsy, and next discuss preclinical research on this topic.

Stress hormones in epilepsy Hormones released after HPA-axis activation in the long term can affect brain function and disease risk, as described above. Considering this, what is known about the levels of HPA-axis hormones in relation to epilepsy? Multiple studies examined stress hormone levels in patients with epilepsy. Directly after the seizure, that is often considered a stressor itself, a consistent increase in cortisol and ACTH was shown; this was observed in the postictal state after generalized tonic-clonic as well as partial seizures (Abbott et al., 1980; Aminoff et al., 1984; Pritchard et al., 1985; Takeshita et al., 1986; Culebras et al., 1987; Rao et al., 1989; Zhang and Liu, 2008). Furthermore, the level of cortisol within 12 hours after a status epilepticus was reported to relate to poor outcome, defined as death or severe complications 6 to 12 days later (Calabrese et al., 1993). Seizure activity can also induce changes in the autonomic nervous system via the stress response or by direct excitation of neurons in the central autonomic network. Autonomic signs and symptoms are common during and after generalized tonic-clonic seizures, and in specific electroclinical syndromes, epilepsy etiologies and localizations, such as Panayiotopoulos syndrome, certain genetic conditions and temporal lobe epilepsy, respectively. The autonomic nervous system is also thought to play a role in sudden unexpected death in epilepsy patients (Moseley et al., 2013). Reports on hormone levels in epilepsy patients at resting conditions have been contradictory (see table 1). Some studies reported increased baseline levels of cortisol (Galimberti et al., 2005; Marek et al., 2010) and ACTH (Gallagher et al., 1984; Gallagher, 1987; Marek et al., 2010) in patients with epilepsy, while others reported a decrease in cortisol (Tuveri et al., 2008; Hill et al., 2011) or no differences (Gallagher et al., 1984). Discrepancies can be

Early life stress in epilepsy

Table 1. Stress hormone levels in epilepsy hormone

epilepsy type

main effect

references

baseline blood DOC

catamenial epilepsy

↓

Tuveri et al., 2009

ACTH

all types

↑

Gallagher et al., 1984, Gallagher, 1987, Marek et al., 2010

cortisol

all types

-/↑/↓

Gallagher et al., 1984, Galimberti et al., 2005, Tuveri et al., 2008, Marek et al., 2010, Hill et al., 2011

CSF CRH

infantile spasms

-/v

ACTH

infantile spasms

↓

cortisol

infantile spasms

v/↓

Baram et al., 1992, Nagamistu et al., 2001 Facchinetti et al., 1985, Nalin et al., 1985, Baram et al., 1992, 1995, Heiskala, 1997, Nagamistu et al., 2001 Baram et al., 1992, 1995

postictal ACTH

all types

↑

Aminoff et al., 1984, Zhang et al., 2008

cortisol

all types

↑

Abbott et al., 1980, Aminoff et al., 1984, Pritchard et al., 1985, Takeshita et al., 1986, Culebras 1987, Rao et al., 1989, Zhang et al., 2008

Overview of studies on the level of stress hormones in epilepsy patients compared with controls. DOC deoxycortisone; ACTH adrenocorticotropic hormone; CRH corticotropin releasing hormone; CSF cerebrospinal fluid; ↑= higher; ↓= lower; v = trend level lower; - no effect.

(partly) explained by differences in study population, the use of hepatic enzyme inducing medication and time interval between the sample collection and last seizure; inclusion criteria differed between studies, with high variation in seizure frequency, which might be relevant, as hormone levels were reported to relate to seizure frequency (Galimberti et al., 2005). With a dexamethasone or dexamethasone-CRH suppression test, a lack of inhibitory control of the HPA-axis was shown in epilepsy patients, which was more severe in patients on hepatic enzyme inducing medication or with co-morbid depression (Robertson et al., 1986; Zobel et al., 2004). These changes in the stress hormone system can be (partly) reversed, for example after epilepsy surgery. Gallagher et al. reported a normalized ACTH concentration and secretory rate, and a partial normalization of mean cortisol level within one year after surgery, unrelated to the use of anti-epileptic drugs or postsurgical seizure control (Gallagher et al., 1984), although a more recent study reported no change in cortisol levels one year after surgery (Bauer et al., 2000). Specific attention was paid to stress hormones in children with infantile spasms, because ACTH and corticosteroids are often used as effective treatment in these patients (see 3.2). In cerebrospinal fluid of untreated children with infantile spasms a decrease in CRH, ACTH

PART ONE | CHAPTER 2

Figure 2. Effects of stress on epilepsy Stress exposure affects neuronal function and thereby influences epilepsy at multiple levels. Especially in early life, the developing brain is vulnerable for stress. Together with the genetic background and other environmental factors, early life stress increases the risk of the development of epilepsy (epileptogenesis). Additionally, in pediatric as well as adult patients with epilepsy, acute and chronic stress directly increase seizure susceptibility.

as well as cortisol was found, compared with controls (Facchinetti et al., 1985; Nalin et al., 1985; Baram et al., 1992; Baram et al., 1995; Heiskala, 1997; Nagamitsu et al., 2001). This probably differs from children with other types of epilepsy, as post-mortem cortex tissue of older children with generalized epilepsy displayed a higher expression of CRH (Wang et al., 2001). Stress hormones play a role in various physiological processes also at lower levels. Therefore, changes in stress hormone levels and stress hormone regulation might have significant adverse effects, for example on immunity and metabolism. More knowledge on the relation between stress hormones and epilepsy could provide directions for the development of medication targeting the HPA-axis or neuronal stress hormone receptors as a new treatment strategy for epilepsy and epileptic seizures.

HPA-axis medication in treating epilepsy Corticosteroids such as hydrocortisone, dexamethasone, methylprednisolone, and ACTH agonists are used in the treatment of epilepsy. The effects of HPA-axis medication in epilepsy are mainly considered to be anti-inflammatory and are mostly used in specific childhood epilepsy syndromes that are assumed to have a (partial) inflammatory etiology, or in which inflammation is considered a result or epiphenomenon of epileptic seizures or epileptiform activity, contributing to further seizure generation (Hancock and Cross, 2009). However, as stress hormones influence neuronal excitability, HPA-axis medication might also exert direct anti-epileptic effects. At this moment, strong evidence for the benefits of corticosteroids and ACTH use in the treatment of epilepsy only exists for infantile spasms (Gayatri et al., 2007), although there are also observational retro- as well as prospective studies suggesting beneficial effects in other epilepsy syndromes (Charuvanij et al., 1992).

Early life stress in epilepsy

Effects of stress on seizures and epilepsy Stress is often mentioned as a precipitating factor for epileptic seizures in patients with epilepsy (see table 2). In questionnaire studies, 8-83% of patients of all ages reported stress as a seizure precipitant (Hayden et al., 1992; Loyning, 1993; Hart and Shorvon, 1995; Cull et al., 1996; Spatt et al., 1998; Frucht et al., 2000; Spector et al., 2000; Haut et al., 2003; Nakken et al., 2005; da Silva Sousa et al., 2005; Sperling et al., 2008; Fang et al., 2008; Pinikahana and Dono, 2009; Koutsogiannopoulos et al., 2009; dos Santos Lunardi et al., 2011; van Campen et al., 2012). This was substantiated in a limited number of prospective diary-based studies in which stress or life events were shown to be associated with an increased seizure frequency in adult epilepsy patients (Temkin and Davis, 1984; Webster and Mawer, 1989; Neugebauer et al., 1994; Haut et al., 2007). The first diary study on stress and epilepsy, performed by Temkin and Davis in 1984 (Temkin and Davis, 1984), included 12 adults with epilepsy who kept a three months diary on seizures and stressful events. They showed that patients experienced significant more seizures on ‘high-stress days’ compared with ‘low-stress days’. This effect remained when analyzing seizure frequency one day after the stress exposure, thereby limiting the effects of seizures on daily stress scores. Webster and Mawer (Webster and Mawer, 1989) studied the effect of life events on seizures by retrospectively correlating monthly seizure frequency in the past one to six years to experienced life events in 18 epilepsy patients. They reported an association between monthly seizure frequency and life events in three patients (17%) and a significant change in seizure frequency after experiencing a life event in six (30%). Neugebauer et al. (1994) focused on short term effects of unpleasant events and showed a significant increase in seizure frequency within 24 hours after an unpleasant event in five out of 37 patients (14%) and a significant decrease in two (5%) after correction for potential confounders. In a more general study on seizure precipitants among 71 epilepsy patients, Haut et al. (2007) also reported an association between daily stress score and increased seizure risk the day after, and showed that perceived stress has an important role in self-predication of epileptic seizures. The relation between stress and seizures remained after correction for sleep deprivation and medication noncompliance (Neugebauer et al., 1994; Haut et al., 2007). Emotions can also trigger seizures (Blanchet and Frommer, 1986; Thapar et al., 2009; Haut et al., 2012). Thapar et al. (2009) performed two consecutive questionnaires on stress and depression in patients with epilepsy, and showed that perceived stress levels over the past month predicted time since last seizure as well as seizure frequency, both mediated by depression score. Blanchet and Frommer (1986) reported a decline in scores for positive as well as negative mood states preceding seizures. Recently, Haut et al. (2012) studied pre-ictal mood states using prospective electronic diaries.

seizure diary

experimental setting

general acute stressors/ life events

stressful interview

stressful video of intake

stressful interviews, hydrocortisone infusion

interview

seizure diary

War

natural disaster

terrorist attack

chronic stress

questionnaire/ interview

method

general acute stressors/ life events

acute stress

stress & seizure frequency

stressor

seizure exacerbation in 12% of all epilepsy patients, up to 50% of patients directly affected by the attack

increased seizure frequency in 26% of patients

increased seizure frequency in 8% of patients

no seizure activity in response to stress or hydrocortisone infusion in 4 selected epilepsy patients with seizures observed to be aggravated by emotional stress

stressors induced spontaneous seizures in 5 epilepsy patients selected on anamnestic stress sensitivity

epileptiform EEG abnormalities provoked by stress in 67% of patients, of which 55% showed no abnormalities on routine EEG

reported stress/life events related to an increased seizure frequency

stress reported as seizure precipitant in 8-83% of epilepsy patients

main effect

Table 2. The effects of stress on epilepsy in humans

Swinkels et al., 1998 Klein and van Passel 2005

Mattson et al., 1970

↑

Feldman and Paul 1976

↑

Stevens 1959

↑

Neufeld et al., 1994

Temkin and Davis 1984, Blanchet and Frommer, 1986, Webster and Mawer 1989, Neugebauer et al., 1994, Haut et al., 2007/2012

↑

Mattson et al., 1991, Hayden et al, 1992, Antebi and Bird 1993, Loyning et al, 1993, Hart and Shorvon 1995, Cull et al., 1996, Spatt et al., 1998, Frucht et al., 2000, Spector et al., 2000, Haut et al., 2003, Da Silva Sousa et al., 2005, Nakken et al., 2005, Fang et al., 2008, Koutsogiannopoulos et al., 2008, Sperling et al., 2008, Pinikahana and Dono 2009, Thapar et al., 2009, dos Santos Lunardi et al., 2011, van Campen et al., 2012

↑

references

PART ONE | CHAPTER 2

questionnaire

life events in childhood

questionnaire

higher prenatal maternal stress levels in mothers of children who were later diagnosed with infantile spasm compared with other epilepsies/normal controls; onset risk increased with degree of prenatal stress.

no increased diagnosis of epilepsy

increased incidence of new onset seizures, no differences in seizure occurrence in patients previously diagnosed with epilepsy

increased risk of being diagnosed with epilepsy

number of experienced negative life events correlated with stress sensitivity of epilepsy

higher frequency of epileptic seizures in children with epilepsy living in war-affected areas

↑

Shang et al., 2010

Li et al, 2008

Moshe et al., 2008

↑

Christensen et al., 2007

Van Campen et al., 2012

↑

Bosjnak et al., 2002

↑

Overview of studies on the effects of stress on epilepsy and epileptogenesis in humans. Studies are arranged by type of stress. Main effect main effect of stress on epilepsy diagnosis or seizure occurrence; ↑= higher; ↓= lower; - = no effect; EEG electroencephalogram

prenatal maternal stress

relative

prenatal maternal loss of a close

population database

comparison of military units

work related stress in military

early life stress

Danish National Hospital Register

parents who lost a child

chronic stress

stress & epilepsy incidence

questionnaire + patient files

war

early life stress

Early life stress in epilepsy

PART ONE | CHAPTER 2

They showed that reported negative mood states at predefined time points were associated with increased odds for seizure occurrence in the following 12 hours, while mood improvement decreased seizure risk. This effect was quite large, with a 25% reduction in seizure risk for every standard deviation improvement in mood. A relation between more chronic types of stress and epilepsy has also been reported. For instance, seizure frequency was increased in adult epilepsy patients exposed to objective major life events such as war and evacuation (Neufeld et al., 1994; Swinkels et al., 1998). Increased incidence of new onset seizures was observed in soldiers working in units with a high degree of work-related stress (i.e., combat units compared with maintenance and administrative units) (Moshe et al., 2008), and an increased risk on being diagnosed with epilepsy was found in parents who lost a child (Christensen et al., 2007). In agreement, an experimental study by Stevens (1959) provided evidence for a relation between stress and seizure susceptibility using electroencephalographic (EEG) registrations in epilepsy patients during a stressful interview. Although no ictal discharges were recorded, interictal epileptiform EEG abnormalities were provoked by stress in 20 out of 30 epilepsy patients, of whom 11 showed no abnormalities without stress. Feldman and Paul (1976) reported five epilepsy patients in whom seizures could be triggered by emotions provoked in an experimental setting. However, hypercortisolemia induced by exogenous ACTH administration did not reveal interictal EEG changes within four to six hours after injection in seven epilepsy patients (Klein and Sahoo, 2005). Clearly, these studies are limited in their scale and would require more substantiation before definite conclusions can be drawn. If stress would indeed precipitate seizures, then behavioral therapy might help to reduce seizure frequency (for reviews, see Mostofsky and Balaschak [1977] and Fenwick [1991]). Several studies have looked at the effects of psychological interventions such as relaxation therapy, cognitive behavioral therapy, bio-feedback and education as add-on treatment in epilepsy, but effects on seizure frequency have not yet been clearly established, as systematically reviewed by Ramaratnam et al. (2008).

Early life stress as a risk factor for epilepsy In children epilepsy incidence is higher than in adults (Ngugi et al., 2011). Early in development the brain is generally considered to be more prone to seizures, probably caused by age-related differences in the balance between excitation and inhibition, which may relate for instance to the initially excitatory effects of the neurotransmitter GABA (Holmes, 1997). As described in 2.2.3, acute and chronic stress exposure early in life have been shown to affect brain development and to be associated with an increased risk of many neuropsychiatric diseases. Therefore, the effects of stress on epilepsy and epileptogenesis are expected to be even more apparent during development. Human studies on early life stress and epilepsy The effects of early life stress on the risk of developing epilepsy have hardly been studied in humans. In a population-based cohort study in Denmark, Li et al. (2008) examined the effects of prenatal maternal stress on the diagnosis of epilepsy in the child. The overall

Early life stress in epilepsy

incidence of epilepsy in the offspring, 0 to 27 years of age, of mothers who had lost a close relative while pregnant or within one year before pregnancy, was not increased compared with other children. The authors concluded that their data did not support a strong association between prenatal stress and the development of epilepsy, although they found a marginally significantly increased risk of epilepsy when prenatal maternal stress occurred in the third trimester of pregnancy. Shang et al. (2010) focused on infantile spasms, and reported a significantly increased level of retrospectively reported maternal prenatal stress in 60 children with infantile spasms compared with children with other types of epilepsy and normal controls. Furthermore, the degree of maternal prenatal stress was related to the onset risk of infantile spasms. Prenatal corticosteroid exposure does not seem to influence the incidence of epilepsy in both preterm and term born children (Eriksson et al., 2009; Eriksson et al., 2012). In children diagnosed with epilepsy, the effects of acute and chronic stress early in life on seizure frequency seem comparable to the effects of stress in adulthood. Bosjnak et al. (2002) reported an increased seizure recurrence in children with epilepsy who lived in areas affected by war compared with children living in non-war areas. As much as 36% of the children who had well-controlled epilepsy before war (seizure freedom for at least one year and no epileptic discharges on EEG) had seizure recurrences thereafter, compared with 0% in the control group. They also evaluated children with a first epileptic seizure during war without subsequent diagnosis of symptomatic epilepsy, and reported half of these first epileptic seizures to be directly linked in time to a stressful event. Furthermore, the children with a stress-related first seizure were more likely to be seizure free after 10year follow-up compared with children with a non stress-related first seizure. Although this is, so far, the only study on the role of severe stress in single isolated seizures and replication is important, these findings suggest that the hyperexcitability caused by severe stress is sufficient to cause epileptic insults without primary epileptogenic pathology or predisposition. We recently showed that stress sensitivity of seizures is reported in half of children with epilepsy, with seizures being precipitated by acute stress in 37%, and an increased seizure frequency in stressful periods in 39% of children, as studied with parental questionnaires. The stress sensitivity of seizures was shown to be more common in children who had previously experienced more negative life events, suggesting a modulatory effect of early life stress (van Campen et al., 2012). Although the four studies mentioned above all suggest a relation between early life stress and seizure frequency or epileptogenesis, the evidence in human literature is sparse and often associated with a certain extent of recall bias, subjectivity, and sample heterogeneity. Animal models provide alternative options to study the relation between early life stress and epilepsy. These models allow stress exposure in a controlled fashion and may help to unravel the underlying mechanisms.

rats

sheep

maternal BETA or hydrocortisone i.p.

maternal DEX i.m.

rats

mice

rats

maternal restrain under bright light (3x/day, 45 min)

maternal restrain 2x/day 120min

maternal i.p. vehicle injection

maternal restrain 3x/day 45 min

maternal restrain 2x/day 45 min

stress

rats

species

maternal DEX or BETA i.m.

hormone

prenatal

hormone/stress

G15

G14-20

G10-12

G4-6/1113/18-20

G5-12/12-20

G103-104

G15-18

age stress

NMDA (P15)

lipopolysaccharide + kainic acid (P14)

audiogenic seizures (P23)

pilocarpine (P19)

hipp kHippindling (P14-15/ 120-130)

EEG recordings (G103-104)

flurothyl/kainic acid (P15)

hipp kindling/ECS (P14/15)

epilepsy model

increased seizure number, decreased latency, normalized by 3 day ACTH pretreatment (no effect of acute ACTH or corticotrophin pre-treatment)

increased seizure severity

increased seizure susceptibility

increased seizure severity after stress in mid/late (not early) gestation

increased kindling rates in infants and adult males after early/late prenatal stress (not in females), transiently lowered ADT

induced epileptiform EEG abnormalities and seizures

BETA decreased susceptibility to flurothyl, but not kainate induced seizures, hydrocortisone no effect

increased kindling threshold after DEX and BETA, increased ECS threshold after BETA

main effect

Table 3. The effects of early life stress on epilepsy and epileptogenesis in animal models

Sadaghiani and Saboory 2010 Beck and Gavin 1976 Qulu et al., 2012 Yum et al., 2012

↑

↑ ↑ ↑

Davidson et al., 2011

↑

Edwards et al., 2002a, 2002b, 2002c

Velíšek, 2011

↓ -

↑

Young et al., 2006

↓

references

PART ONE | CHAPTER 2

rats

DOC s.c.

CRH i.c.v.

ACTH i.p. (+ADX)

ACTH i.p.

DEX s.c.

hormone

P15

P4-5

P15

P11

P5-13

P3,7,12,15, 21, 35, 75

G18

rats

maternal restrain 20 min

postnatal

G15-17

rats

maternal restrain 2x/day 120 min or predator exposure 1x/day 120 min

PTZ/MES, hipp kindling (15 resp. 15/60/120 min after DEC)

Hipp kindling (P13)

PTZ/ECS, hipp kindling (15 resp. 15/60/120 min after ACTH)

NMDA (30 min after ACTH)

NA (after CRH)

PTZ/ECS (15-30 min after DOC), hipp kindling (15/60/120 min after DOC P15)

kainic acid (P65-69)

pilocarpine (P25)

no effect on kindling, seizure duration or latency

no effect on ADT or hipp kindling rates

no effect on kindling, seizure duration or latency

ACTH reduced seizure number and seizure severity, also after the increased seizure susceptibility by ADX

induced seizure activity, not blocked by ACTH

rapid anticonvulsant effects of DOC until puberty, via metabolites

Increased number and duration of seizures, decreased effectiveness of THP to prevent seizures

increased seizure severity, decreased seizure latency, correlated with stress severity

Edwards et al., 2002b/c, 2005, Perez-Cruz et al., 2006 Baram and Schultz, 1991, 1995 Wang et al., 2012

Edwards et al., 2002b

Edwards et al., 2002a Edwards et al., 2002b

↑ ↓

Frye and Bayon, 1999

↑

↓

Ahmadzadeh et al., 2011

↑

Early life stress in epilepsy

P4-5

rats

maternal separation 1h in bedding of strange male

maternal separation 1h/day

maternal separation 180min/day vs. 15min/day

cross-fostering seizure prone and seizure resistant rats

amygdala kindling (~P 94)

WAG/Rij rats (P135+)

amygdala kindling (adult)

Li-Pilo SE (P10/12), PTZ (P100)

PTZ (P10-14)

hipp kindling (P14)

hyperthermic seizure (P10)

increased kindling rates in all cross-fostered rats

decreased SWD number (not duration) in handled and stressed group

increased kindling rate, lowered ADT

increased seizure duration P12, decreased seizure threshold P100

increased duration of seizures

no significant effect on kindling rate or ADT

Edwards et al., 2002a

Huang et al., 2002 Lai et al., 2006, 2009 Salzberg et al., 2007, Jones et al., 2009, Kumar et al., 2011 Schridde et al., 2006

Gilby et al., 2009

↑ ↑ ↑

↓

↑

Desgent et al., 2012

↑ decreased seizure latency, MTLE development in males/adrogenized females correlated to CORT levels after freeze lesion

Overview of studies on the effects of stress and stress hormones on epilepsy and epileptogenesis in animal models. Studies are arranged by method and timing of stress exposure. Main effect main effect of hormone/stress on seizure susceptibility; ↑= higher; ↓= lower; - = no difference; DOC deoxycortisone; ACTH adrenocorticotropic hormone; CRH corticotropin releasing hormone; DEX dexamethasone; BETA betamethasone; ADX adrenalectomy; i.m. intramuscular; i.p. intraperitoneal; h hour; P postnatal day; Hipp hippocampal; ECS electroconvulsive shock; EEG electroencephalogram; NMDA n-methyl-D-aspartate; PTZ pentylenetetrazole; NA not applicable; Li-Pilo lithium-pilocarpine; WAG-Rij Wistar Albino Glaxo/Rij; ADT after discharge threshold; THP 3α,5α- tetrahydroprogesterone; MTLE mesial temporal lobe epilepsy; CORT corticosterone; SWD spike-wav

P1-23

P1-21

P2-14

P2-9

freeze lesion induced rats cortical malformation

stress

Table 3. Continued

PART ONE | CHAPTER 2

Early life stress in epilepsy

Animal studies on early life stress and epilepsy Early life stress, prenatal as well as postnatal, was shown to be proconvulsive in various animal models for epilepsy (see table 3). Prenatal maternal restrain increased seizure number and seizure severity, and decreased after-discharge threshold and seizure latency from infancy up to adulthood in kindling and chemically induced epilepsy models (Frye and Bayon, 1999; Edwards et al., 2002; Edwards et al., 2002; Edwards et al., 2002; Sadaghiani and Saboory, 2010; Ahmadzadeh et al., 2011; Yum et al., 2012). These long-lasting effects were already observed after mild stressors such as vehicle injection (Beck and Gavin, 1976) or a single 20 minute period of restrain (Frye and Bayon, 1999), although a recent study suggested a relation between stress severity and seizure susceptibility by showing that prenatal maternal predator exposure increased seizure susceptibility more than just maternal restrain (Ahmadzadeh et al., 2011). Postnatal early life stress has also consistently been shown to increase seizure susceptibility, from the first postnatal weeks (postnatal day 10-14) up to adulthood (studied up until 6 months of age). This was observed in different seizure models, including hippocampus and amygdala kindling, hyperthermic seizures, administration of PTZ and of lithiumpilocarpine (Huang et al., 2002; Lai et al., 2006; Salzberg et al., 2007; Jones et al., 2009; Lai and Huang, 2011; Kumar et al., 2011; Desgent et al., 2012). Most studies used maternal separation as early life stressor, but the duration, frequency and timing differed. Other stressors showed the same direction of effect (Gilby et al., 2009; Desgent et al., 2012). For instance, when studying amygdala kindling rates in adult rats of seizure-prone and seizureresistant strains, increased kindling rates were noticed in all animals after cross-fostering stress (Gilby et al., 2009). The reverse also seems to be true: environmental enrichment and improved maternal care were shown to reduce seizure susceptibility in rodents (Auvergne et al., 2002; Korbey et al., 2008; Leussis and Heinrichs, 2009; Fares et al., 2013)). Although the results of these kindling- and chemical-induced seizures are consistent, contradictory results have been found in an animal model for genetic absence epilepsy, in which maternal deprivation and neonatal handling resulted in a decreased number of spike-wave discharges (Schridde et al., 2006). The underlying mechanisms of the relation between (early life) stress and epilepsy can also be investigated by exposing the brain to exogenous stress hormones. However, direct peripheral injection with specific hormones early in life did not result in increased seizure susceptibility. Considering prenatal stress hormone exposure, most attention has been paid to dexamethasone and betamethasone, as these hormones are routinely used in pregnant women at risk of preterm delivery because they reduce the risk of a variety of conditions in the neonate (Roberts and Dalziel, 2006). Focusing on epilepsy, prenatal maternal injection with dexamethasone or betamethasone reduced postnatal seizure susceptibility in rodents (Young et al., 2006; Velisek, 2011), whereas in sheep, maternal dexamethasone treatment in the beginning of the third trimester of pregnancy actually provoked electrographic epileptiform activity in the fetus, sometimes associated with clinical signs of an epileptic seizure (Davidson et al., 2011). It is important to realize that sheep as well as human neonates are born relatively mature. Therefore, the developmental brain stage of sheep in

PART ONE | CHAPTER 2

their third trimester is comparable to that of rodents in the early postnatal period. However, in rodents, postnatal dexamethasone exposure did not affect after-discharge threshold or hippocampal kindling rate later (Edwards et al., 2002). The effect of other stress hormones has also been tested in postnatal early life. In this period, the neurosteroid deoxycortisone (DOC) was shown to be anticonvulsive, effects that are lost after puberty (Edwards et al., 2002; Edwards et al., 2005; Perez-Cruz et al., 2006). CRH can induce seizure activity by itself after intraventriculair infusion, an effect that is not blocked by ACTH, suggesting that CRH itself is proconvulsive (Baram and Schultz, 1991; Baram and Schultz, 1995). ACTH appears to have no proconvulsive properties, as one study showed that ACTH had no effect on hippocampal kindling rates, on seizure characteristics for pentylenetetrazole (PTZ), or on electroconvulsive shock induced seizures (Edwards et al., 2002). Another study even showed a reduction in seizures and seizure severity of N-methyl-D-aspartate (NMDA)-induced seizures 30 min after ACTH injection (Wang et al., 2012). However, it is unclear to what extent the reported effects are caused by an elevation in the level of the hormone administered, or by subsequently elevated or reduced levels of other hormones in the stress system through feed forward or negative feedback. Early life stress does not only increase seizure outcome, it also aggravates the effects of epileptic seizures on brain morphology. Although early life stress and corticosterone exposure normally reduce neurogenesis, prenatal maternal restrain and postnatal maternal deprivation enhanced the epilepsy-associated increase in neurogenesis, and also aggravated other epilepsy related phenomena such as pyramidal cell loss and neuronal degradation (Koe et al., 2009; Kumar et al., 2011). However, whether this represents accelerated epileptogenesis or increased damage by more severe seizures (associated with a lower seizure threshold) is unclear. In summary, most animal studies show an accelerated development of epileptic symptoms after seizure induction in animals exposed to early life stress. Whether this really reflects aggravated epileptogenesis is uncertain. Additional non epileptogenesis-related changes in neuronal function and morphology after early life stress exposure could also indirectly change seizure vulnerability and therefore seizure latency, severity, frequency and kindling rate.

DISCUSSION There is a solid amount of evidence suggesting a relation between (early life) stress and epilepsy. This relationship exists for different stressors and various epilepsy subtypes. Although studies on early life stress and seizures in humans are limited, generally proconvulsive effects of stress were reported in humans during or after acute, chronic as well as early life stress. Animal studies quite consistently showed enhanced seizure susceptibility in rodents previously exposed to prenatal or neonatal stress. Whether or not early life stress increases epileptogenesis and epilepsy incidence is not entirely clear yet.

Early life stress in epilepsy

Possible mechanisms The relation between stress and epilepsy could be explained by the effects of altered levels of stress hormones on neuronal function. Multiple stress hormones can directly influence neuronal excitability, as extensively reviewed by Joëls (Joëls, 2009). CRH appears to be mainly proconvulsive, which can be explained by the fact that it promotes excitatory transmission in different hippocampal subareas, for example by inducing depolarization and suppressing hyperpolarization. Neurosteroids such as DOC and its metabolite tetrahydrodeoxycorticosterone (TH-DOC) exert mainly anticonvulsant properties, which agrees with the observation that TH-DOC did not only facilitate inhibitory responses by tonic inhibition (Stell et al., 2003) and GABA potentiation (Reddy and Rogawski, 2002; Ferando and Mody, 2012), but also reduced excitatory transmission in the dentate gyrus (Stell et al., 2003). ACTH and corticosterone can be pro- as well as anticonvulsive, depending on the site of action in the brain and timing of administration. With respect to the latter, it was demonstrated that fast membrane receptor mediated effects increase neuronal excitability, while the slow, genomic, effects are mostly anticonvulsant. Effects of stress hormones on neuronal excitability depend on glutamate and gamma-aminobutyric acid (GABA) receptor expression and subunit composition in neurons and glia cells, as well as regulation of calcium influx. Somewhat paradoxically, treatment of specific pediatric epilepsy syndromes with stress hormones can exert clear anticonvulsive effects. This could firstly be explained by their anti-inflammatory actions, quieting down the inflamed and therefore seizure generating brain tissue. Secondly, as exogenous stress hormone administration decreases the levels of other hormones in the stress system by negative feedback, this reduction in for instance CRH levels might be another mechanism of action, although it would be expected to affect a broader range of epilepsy syndromes. Stress not only involves the release of corticosteroid hormones, CRH and neurosteroids; the sympathetic system also influences excitability. Noradrenaline can exert pro- as well as anticonvulsive effects through β- and α-adrenoreceptors respectively (Weinshenker and Szot, 2002; Jurgens et al., 2005). The main effect of endogenous noradrenaline is probably anticonvulsant, as an intact noradrenaline system reduces neuronal damage in animal models of limbic epilepsy and is essential for the use of vagal nerve stimulation in humans, as reviewed by Giorgi et al. (2004). The sympathetic nervous system and the HPA-axis interact at multiple levels, and increasing attention is paid to the combined effects of the different stress hormones. Corticosterone and noradrenaline exert synergistic effects on cell excitability in a time dependent manner (Joëls et al., 2009). This interaction may also exist with other hormones such as endocannabinoids, dopamine and serotonin, leaving much to be investigated. On a different level, epigenetic processes are thought to play an important role in modulating gene-environment interactions. Epigenetic changes have already been demonstrated after early life stress (Murgatroyd and Spengler, 2011) and are thought to mediate effects of early life stress on depression (Heim and Binder, 2012), as well as in the development of epilepsy (Kobow and Blumcke, 2012; Roopra et al., 2012). The enduring effects of chronic and early

PART ONE | CHAPTER 2

life stress on epilepsy and epileptogenesis could thus well be mediated by epigenetic mechanisms.

Variation between studies Although the results of the various studies appear to be highly consistent, variations might be caused by differences in the definition of stress, the stress model used (mediators involved, frequency of stress exposure), the applied epilepsy model or subtype, the stage of brain maturation, brain region, sex, species, control condition and/or outcome parameter. Human studies Almost all human studies looking into stress and epileptic seizures found positive correlations between the two. However, we should keep in mind that most human studies were performed retrospectively and depended on subjective interpretation of the definition of stress, as well as cooperation of, and recall by, caregivers and patients. This is reflected in the large range of reported stress sensitivity of seizures in the different questionnaire studies (8-83%). Human research is further complicated by the fact that patients, often non-intentionally, try to find cause-consequence relationships to get grip on their otherwise unpredictable disease, which may lead to overestimation of the reported stress sensitivity of seizures, limiting the reliability of the results. Also, a variety of other factors might contribute to or mediate the effects of stress on epilepsy, such as lack of sleep, medication non-compliance, hyperventilation and consumption of alcohol and drugs (Mattson et al., 1970; Mattson, 1991). It is important to take these factors into account when investigating stress sensitivity of epilepsy. Although patients with infantile spasms have been investigated separately, most human studies included a wide variety of epilepsy patients. As epilepsy is a very heterogeneous disease, the relation between stress and epilepsy might differ for different epilepsy subtypes, for example related to epilepsy syndrome, etiology or age. Moreover, specific characteristics of the stressor, the cumulative stress exposure earlier in life and the genetic vulnerability of the individual could account for the variability. Studies in a controlled environment are difficult to perform in humans, for obvious ethical reasons. Another confounder (in human as well as in animal studies) is the stress inherent to epilepsy itself. This includes chronic stress caused by living with the diagnosis of epilepsy and the risk of often unpredictable seizures, as well as repetitive acute stress caused by experiencing the seizures themselves. Especially in childhood epilepsy, the early life stress caused by seizures and seizure-risk could have severe effects on the developing brain. Distinguishing between this epilepsy-related stress and other epilepsy related variables such as disease duration and seizure frequency seems often impossible, and is usually not taken into account. Animal studies In animal studies, most genetic as well as environmental variation can be controlled for. Using a combination of injections with specific stress hormones or exposure to standardized

Early life stress in epilepsy

stressors, and a specific method of seizure induction, most studies have shown increased seizure susceptibility after early life stress. The results of studies on early life stress are very comparable to those of the majority of studies on chronic stress, as chronic aberrations of the stress system are also generally considered to be proconvulsive. Chronic stressors, such as chronic social isolation in adult animals, were shown to lower the seizure threshold for kindling and chemical induced seizures (Matsumoto et al., 2003; Chadda and Devaud, 2004). By contrast, the effects of acute stress on epilepsy and seizure susceptibility largely depend on the type of stressor and epilepsy model used, and the developmental stage of the animals. Although psychological stressors like novelty stress and handling are proconvulsive, physical stressors such as swim stress and cold temperature stress increase seizure threshold in various epilepsy models (De Lima and Rae, 1991). These opposing effects of different types of stress nicely show the advantages and disadvantages of animal models. Because of the high level of controllability and homogeneity within a study, animal models can provide great mechanistic insights, but extrapolation of the results to other epilepsy types, stressors, developmental stages and species should be done with care. Additionally, most animal epilepsy models used are not directly comparable to the human situation and for most human epilepsy syndromes no animal model is available yet. Also, the stressors used are not directly comparable to stressors often experienced by humans, which further limits translation to the human situation.

Future directions How can the knowledge on stress and epilepsy help patients? Understanding of the relation between stress and epilepsy can improve patient education, self-management and the development of treatment strategies focusing on stress coping mechanisms. Obviously, although a relation between early life stress and epilepsy seems quite clear, it is not simple to intervene. Medication targeting the stress system might be a future treatment option for (subgroups of) epilepsy patients. However, to get to this point, more insight in the mechanisms responsible for the relation between (early life) stress and epilepsy, and identification of those patients who will benefit most, is needed. Animal studies on early life stress and epilepsy could clarify stress hormone interactions, as well as dose-dependency and time windows for stress effects. Until now, most animal studies have focused on seizure susceptibility for chemical- or kindling-induced seizures in animals previously exposed to early life stress. As epileptogenesis is often a long term process, it would be interesting to evaluate the effect of stress during epileptogenesis on brain function and morphology, and to study epilepsy in a model with better face validity for the human situation. In humans, large population-based studies, prospectively following children and reporting genetic background, stress exposure, as well as epilepsy outcome, would provide more insight in the very important interplay between genetics and environment in epilepsy and in specific epilepsy syndromes. In addition to this, human studies should aim at testing

PART ONE | CHAPTER 2

specific hypotheses in a controlled experimental setting whenever ethically acceptable in order to translate preclinical results into human research, so that the benefits of both fields will be combined. This would not only increase fundamental knowledge, but could also allow for early intervention, improve patient counseling and (perceived) self-control of seizures, and provide a basis for new epilepsy treatment strategies.

Conclusion In conclusion, a vast amount of evidence indicates a relation between stress and epilepsy, with (1) acute stress provoking epileptic seizures and (2) chronic stress increasing seizure frequency in patients with epilepsy and seizure risk in non-epileptic individuals, especially when stress is experienced early in life. Increased knowledge on the link between early life stress, epileptogenesis and vulnerability for (specific types of) epilepsy could help to understand this phenomenon and benefit the treatment and care of patients with epilepsy.

CHAPTER

Stress sensitivity of childhood epilepsy is related to experienced negative life events Epilepsia 2012

Jolien S. van Campen Floor E. Jansen Laurie C. Steinbusch Marian JoĂŤls Kees P.J. Braun

PART ONE | CHAPTER 3

ABSTRACT Introduction To evaluate the effect of stress on seizure frequency in childhood epilepsy, and to assess possible differences between children in whom seizures are precipitated by stress and those in whom they are not. Methods Parents or caregivers of children with active epilepsy (aged 2 to 16 years) were sent questionnaires on developmental and epilepsy characteristics, life-time stress exposure, and the effect of stressful periods and moments of acute stress on seizure frequency in their child. Further information was extracted from patient files. Results Parents or caregivers of 153 patients with a median age of 8.8 years responded to the questionnaires. Thirty-nine percent reported an increase in seizure frequency during periods of stress, with a median increase of 2.5 times the frequency compared with nonstressful periods. Thirty-seven percent reported that seizures were precipitated by acute stress, with stress being a precipitating factor in 33% (median value) of the seizures. Overall, 51% of the patients reported stress sensitivity of seizures. A higher number of negative life events experienced in total life was related to an increase in seizure frequency in stressful periods (odds ratio [OR] 1.3, p = 0.01) as well as to the precipitation of seizures by acute stress (OR 1.3, p = 0.02). Conclusion Stress sensitivity is reported in half of the children with epilepsy. Results of this study suggest a relation between experienced negative life events and stress sensitivity of childhood epilepsy. One possible explanation could be that experiencing negative life events may cause a larger response to daily stressors, thereby increasing the likelihood to induce epileptic activity.

Stress sensitivity of childhood epilepsy

INTRODUCTION In patients with epilepsy, stress is one of the most frequently self-reported precipitants for seizures (Cull et al., 1996; Spatt et al., 1998; Spector et al., 2000; Frucht et al., 2000; da Silva Sousa et al., 2005; Nakken et al., 2005; Haut et al., 2007; Sperling et al., 2008; Pinikahana and Dono, 2009; dos Santos Lunardi et al., 2011). Emotional stress is the precipitant most often reported (Nakken et al., 2005), but acute physical stressors can exacerbate seizures as well (Arida et al., 2009). The relationship between stress and seizures was shown to be independent of sleep deprivation or medication non-compliance (Haut et al., 2007). Prospective explorative studies revealed a higher seizure frequency in children with epilepsy in areas of war compared with children in non-war areas (Bosnjak et al., 2002), an increase in seizure frequency in patients with epilepsy during natural disasters (Swinkels et al., 1998), and a higher incidence of seizures in soldiers who experience more occupational stress (Moshe et al., 2008). Although stress has been shown to influence seizure susceptibility in patients with epilepsy, the evidence was so far mostly obtained in adults. Other types of seizure precipitants such as fever, hot water and photic stimulation have been shown to be more common in children compared with adults (Satishchandra, 2003; Lu et al., 2008; Reid et al., 2009). However, the prevalence and characteristics of stress-sensitive epilepsy have never been studied in a pediatric population. Therefore, the aim of our study was to assess the effect of stress on seizure frequency in children with epilepsy and to evaluate which characteristics of these children relate to their stress sensitivity.

METHODS Patients In this consecutive cohort study, we retrospectively selected all children, aged 2 to 16 years, with active epilepsy (i.e. a definite clinical diagnosis of epilepsy and seizures within two years prior to data collection), who consulted one of three pediatric epileptologists of the UMC Utrecht between January 2008 and December 2010. We excluded children who were known to have non-epileptic events in addition to seizures, and children who became seizure free after epilepsy surgery. Definition of stress Stress was defined as either emotional stress (i.e., stress caused by events associated with positive or negative emotions) or physical stress (as caused by physical exercise or pain); both may have an effect on stress hormone responses in the brain. Physical stress caused by illness was excluded, as in this situation other variables, for example high body temperature and infection, may also influence seizure occurrence. Stress sensitivity of epilepsy was divided in (1) increased seizure frequency during stressful periods, and (2) seizures provoked by acute stress.

PART ONE | CHAPTER 3

Data collection Data was collected from questionnaires and patient files. Questionnaires A questionnaire was sent to parents or caregivers of all patients fulfilling selection criteria. They were asked to complete the questionnaires together with their child if possible.The questionnaire consisted of three parts: Part I involved demographic, developmental and epilepsy characteristics, namely: age at assessment, age at seizure onset, seizure frequency (seizure diaries were not required because of the explorative nature of this study), seizure precipitating factors, intelligence quotient (IQ), school level, type of treatment, anti-epileptic drugs (AEDs) used, and comorbidities. AEDs used, were grouped based on enhancement of γ-aminobutyric acid activity (GABAergic drugs), and induction of cytochrome P450 enzymes, according to the Dutch Pharmacotherapeutic Compass (College voor Zorgverzekeringen [CVZ], 2011). Part II specifically addressed the relationship between stress and epilepsy. Stress was divided in: (1a) periods of positive stress, with the following examples of stressors: birthday, Sinterklaas (a Dutch holiday comparable to Christmas), party, or other; (1b) periods of negative stress, with the following examples of stressors: arguments/fights at home, being bullied, death of a relative, death of a pet, moving homes, a close friend had moved homes, tests at school, or other; and (2) acute stress (emotional and physical stress with a maximum duration of four hours), with the following examples of stressors: anger, fear, arousal, nervousness, startle, pain, physical exercise or other. If the option “other” was chosen, we asked to specify this stressor. For the distinction between acute stress and periods of stress, the cutoff was set at a duration of four hours, as direct effects of corticosteroids on neuronal function (genomic and non genomic) have been demonstrated for up to four hours after exposure to the initial stressor (Joëls, 2001; Alfarez et al., 2002; Alfarez et al., 2003; de Kloet et al., 2008). Parents were asked if seizure frequency increased during stressful periods, and if so, to estimate the percentages of change in positive and negative stressful periods separately. A decrease in sleep quality or quantity, or an increase in medication noncompliance during stressful periods was questioned as well. Further, parents were asked whether acute stress-precipitated seizures, and if so, to report the percentage of seizures precipitated by acute stress. In case of reported increase of seizure frequency, parents were asked if all or only a certain type of seizure became more frequent after periods of stress or acute stress. In part III, the total number of experienced life events was assessed using a Dutch validated Questionnaire of Life Events for children and adolescents (Veerman et al., 1993), which was adapted from the Dutch version of the Social Readjustment Rating Questionnaire (Coddington, 1972; Veerman et al., 1993). This questionnaire consists of 24 standardized potentially stressful events of which 15 have an a priori negative connotation, three an a priori positive connotation and six an a priori ambiguous connotation. The number of negative events in the year preceding the study and the number of negative events in total life were analyzed. Written informed consent was obtained from parents and from children older than 11 years of age. In case of nonresponse, parents were called by the investigator to encourage response

Stress sensitivity of childhood epilepsy

and ask for the reason of nonparticipation. The study was approved by the institutional ethical committee.

Patient files Demographic, developmental and epilepsy characteristics of nonresponders (i.e. sex, age, age at onset of epilepsy, seizure frequency, presence of mental retardation and number of AED used) as well as documentation of stress being a seizure-precipitating factor, etiology, and results of electroencephalographic (EEG) recordings of responders and nonresponders were collected from patient files. Etiology was classified according to the etiological designation proposed by the ILAE in 2010 as genetic, structural, metabolic or unknown (Berg et al., 2010). Analysis First, characteristics of responders and nonresponders were compared using the Fisherâ&#x20AC;&#x2122;s exact test for binominal data, the chi-square test for non-dichotomous nominaldata, the two-tailed independent samples t-test for continuous data with a normal distribution and similar variance in both groups, and the Mann-Whitney test for continuous data not meeting the assumptions of the t-test. Normality was evaluated with Q-Q plots. Variance was evaluated with box plots. Outcome measures were (1) the presence or absence of an increase in seizure frequency in stressful periods, and (2) the presence or absence of seizures induced by acute stress. The relations between demographic, developmental and epilepsy characteristic as well as the number of negative life events experienced in the past year and in total life, and both outcome measures were analysed using logistic regression analysis. Variables with a univariable association of p < 0.2 were included as candidates into a multivariable logistic regression model to maximize sensitivity. In case of multicollinearity between two or more variables, only the variable with the lowest p-value in univariable analysis was included as a candidate in the model, to improve the validity of the results of individual predictors. Variables were removed by a backward stepwise selection procedure. To confirm the robustness of models, multivariable analysis was repeated using a forward selection procedure. In patients reporting the presence of one of the outcome measures, the relation between characteristics and the factor increase in seizure frequency during stressful periods or the percentage of seizures induced by acute stress was analysed using a multiple linear regression model. Normality of residues was evaluated with Q-Q plots. Variance of residues was evaluated with error plots. Variables that were significant at the p < 0.05 level were retained in the multivariable models. There was no adjustment of significance level for multiple testing because multivariable analysis was preferred. The correlation between an increase of seizure frequency in periods of stress and changes in sleep quality or quantity as well as medication noncompliance was analyzed using the phi correlation coefficient. The Spearman correlation coefficient was calculated to correlate dichotomous and not normally distributed continuous variables. Statistical analysis was performed in SPSS (version 15.0).

PART ONE | CHAPTER 3

RESULTS Patient selection Three hundred and forty five children with epilepsy fulfilled inclusion criteria. Parents or caregivers of 153 (44.3%) of these children responded; this information was included in the study (see Figure 1). Demographic and epilepsy characteristics of responders compared with nonresponders are shown in table 1. Stress related seizures were twice as commonly reported in medical files of responders compared with non-responders (15.7% versus 8.3%, p = 0.04). Mental retardation was less common in responders, and they were taking more anti-epileptic drugs. Characteristics of the study group Mean age at time of assessment was 8.8 years. Almost all children (91.5%) had documented seizures in the past year. Most children (94.8%) were treated with (a number of) antiepileptic drugs, 11.8% were on a ketogenic diet, in 3.9% a vagal nerve stimulator was implanted, and 1.4% were treated with methylprednisolone or ACTH therapy (0.7% each). Precipitating factors other than stress were reported in 88.6% of children. The factors most frequently mentioned were tiredness (44%), fever (33%) and sleep deprivation (33%). Stress sensitivity of epilepsy Over half of the responders (51%) reported either an increase in seizure frequency during stressful periods, or the presence of seizures that were precipitated by acute stress (details below). Presence of both outcome measures was reported in 25% of children. Increase of seizure frequency in stressful periods Sixty children (39%) reported an increase in seizure frequency in stressful periods, with a median factor of 2.5 (range 1.3 to 76) compared with overall seizure frequency. The number of stressful periods that was associated with increased seizure frequency varied from 1 to 22 per year, with one outlier reporting 100 periods per year (median 3 per year), and was significantly correlated to seizure frequency. In 15% of children, stress only increased the frequency of a specific type of seizure semiology. Stress sensitivity of epilepsy in positive periods was reported in 46 children (30%), of which birthday (20%), Sinterklaas (16%) and a party (14%) were most frequent. Stress sensitivity in negative periods was reported in 34 children (22%), of which tests at school (7%), arguments at home (4%) and being bullied (4%) were most frequent. Having a seizure was not considered as a negative stressor. In ten children increased use of rapid acting AED in stressful periods was reported. An increase in seizure frequency during periods of stress was highly correlated with a change in quality or quantity of sleep during stressful periods (rĎ&#x2020; = 0.52, p < 0.01), but showed no correlation with medication non compliance (rĎ&#x2020; = 0.08, p = 0.34). Children with an increase (versus no increase) in seizure frequency in stressful periods had a significantly longer duration of the epilepsy, and had experienced more negative life events in the past year and in total life, as shown in table 2. There was no relation with other

Stress sensitivity of childhood epilepsy

Figure 1. Flow chart of patient selection. After screening of patient files, 345 patients fulfilled selection criteria. Number of patients excluded and reason of exclusion are shown in the right upper box. The parents / caregivers of these 345 patients received a questionnaire, of which 153 responded and formed the study population. Reasons for nonresponse are shown in the right lower box.

variables as epilepsy classification, etiology or type of AED. In multivariable analysis, only the number of experienced negative life events in total life remained significant. An increment of one experienced life event was associated with an odds ratio (OR) of 1.3 (95% confidence interval [CI] 1.1 â&#x20AC;&#x201C; 1.6, p = 0.01) for stress sensitivity. This result remained significant after correction for age in the multivariable model (OR 1.2 [95% CI 1.0 - 2.2], p = 0.05). The total number of life events also showed a correlation with a decrease in sleep quality or quantity during stressful periods (r = 0.20, p = 0.02). Further analysis with a linear regression model looking at the factor increase in seizure frequency did not reveal significant relations with patient characteristics.

Positive versus negative stressful periods When we differentiated, within the group of children with stress sensitivity, between those with sensitivity to positive periods, those with sensitivity to negative periods, and those

PART ONE | CHAPTER 3

with sensitivity to both periods, we found that children sensitive for both periods had experienced more life events than children in the two other groups (median 3.5 compared with 1 for positive and 2 for negative periods) (p < 0.01). Further, in this group the median number of stressful periods per year was higher, i.e. 7.5 (range 1 to 100), compared with 2 (range 0 to 12), and 2 (range 1 to 3) respectively in the two other groups (p < 0.01).

Seizures precipitated by acute stress Parents of fifty-seven children (37%) reported that seizures could be precipitated by acute stress. The percentage of seizures precipitated by acute stress varied from 13 to 100% (median 33%). In one third of these children, only a specific type of seizure semiology was precipitated by acute stress. Acute emotional stressors were more often reported to precipitate seizures than acute physical stressors. Acute stressors most frequently reported were arousal (20%), anger (16%), fear (14%), startle (14%), nervousness (13%), physical exercise (13%) and pain (10%). In 19% of children sensitive to acute stress, a stressor of a certain type and intensity was always followed by a seizure, meaning that this stressor never took place without provoking an epileptic event. There was a large overlap between patients reporting emotional stress to be a precipitating factor (33%) and those who reported physical stress to be a triggering factor (20%). The amount of experienced negative life events was significantly related to precipitation of seizures by acute stress, with an odds ratio of 1.3 (95% CI 1.0 - 1.5, p = 0.02) when multivariable analysis was performed, as shown in table 3. After correction for age this relation remained significant (odds ratio of 1.2; 95% CI 1.0 - 1.5, p = 0.04). No other variable related to the precipitation of seizures by acute stress. Table 1. Characteristics of responders compared with nonresponders. responders (n = 153) sex, % male age, median (range) mental retardation, %

nonresponders (n = 192)

p-value

51.6

56.8

0.38

8.8yr (2.2 - 16.9)

9.2yr (2.1 - 16.9)

0.88

47.7

<0.01

2.3yr (0.0 - 12.7)

1.7yr (0.0 - 14.8)

0.58

- genetic

18.1

18.8

- structural

41.8

44.8

- metabolic

10.5

8.3

- unknown

29.4

28.1

66.0

66.7

48/yr (0/yr â&#x20AC;&#x201C; 100/d)

12/yr (0/yr â&#x20AC;&#x201C; 90/d)

0.27

2 (0 - 5)

1 (0 - 5)

<0.01

15.7

8.3

0.04

age at onset of epilepsy, median (range) etiology, %

focal, % seizure frequency, median (range) number of currently used AED, median (range) stress documented as precipitant in patient files, %

1.00

n number of patients; p p-value; range minimum and maximum value; yr year; d day; AED anti-epileptic drugs.

50.0

2 (0 – 8)

negative life events total life, median (range)

52.7

1 (0 – 7)

0 (0 – 3)

34.4

65.6

26.9

65.6

26.9

10.8

45.2

17.2

20/yr (0/yr – 96/d)

3.9 (0.3 – 14.6)

1.5yr (0.0 – 12.7)

5.4

64 (7 – 120)

8.6yr (2.2 – 16.9)

univariable

1.3 (1.1-1.6)

1.5 (1.0-2.2)

0.9 (0.4-1.8)

1.2 (0.6-2.6)

0.8 (0.4-1.8

1.1 (0.5-2.2)

1.1 (0.4-2.8)

0.8 (0.2-2.8)

0.7 (0.3-1.7)

ref

1.0 (1.0-1.0)

1.1 (1.0-1.2)

1.0 (0.9-1.1)

2.0 (0.6-6.7)

1.0 (1.0-1.0)

1.1 (1.0-1.2)

1.1 (0.6-2.1)

multicoll2 1.3 (1.1-1.6)

0.04 0.01

0.67

0.61

0.62

0.85

0.76

0.03

0.99

0.29

0.68

multicoll1

0.75 0.05

0.01

p-value

multivariable OR

p-value

n number of patients; OR odds ratio; p p-value; range minimum and maximum value; yr year; d day; multicoll excluded from multivariable analysis because of multicollinearity with 1duration of epilepsy or 2negative life events total life; ns excluded in multivariable model by backward as well as forward procedure because of p>0.05; ref reference category; GABAergic one or more of the anti-epileptic drugs used enhances γ-aminobutyric acid activity; CYP450 inducing one or more of the anti-epileptic drugs used induces cytochrome P450 enzymes; bold p < 0.2 in univariable analysis; Italic p < 0.05 in multivariable analysis.

1 (0 – 3)

26.1

- CYP450 inducing

negative life events past year, median (range)

58.0

- GABAergic

anti-epileptic drugs %

23.3

33.3

- unknown 66.7

10.0

- metabolic

temporal seizure onset, %

36.7

- structural

focal, %

20.0

- genetic

etiology, %

92/yr (0/yr – 100/d)

5.7 (0.6 – 14.5)

duration of epilepsy, median (range)

seizure frequency, median (range)

3.5yr (0.0 – 9.4)

10.0

67.5 (17 - 125)

9.4yr (3.3 – 16.3)

age at onset of epilepsy, median (range)

autistic spectrum disorder, %

intelligence quotient, median (range)

age, median (range)

sex, % male

no (n = 93)

yes (n = 69)

stressful periods increase seizure frequency

Table 2. Clinical characteristics of children with and children without an increase in seizure frequency in periods of stress.

Stress sensitivity of childhood epilepsy

58 156/yr (0/yr – 100/d)

seizure frequency, median (range)

2 (0 – 8)

negative life events total life, median (range)

1 (0 – 7)

0 (0 – 3)

35.5

67.7

70.2

30.2

11.5

43.8

14.6

12/yr (0/yr – 96/d)

4.2 (0.3 – 14.6)

1.9yr (0.0 – 12.7)

4.2

71 (7 – 125)

8.7yr (2.2 – 16.9)

univariable

1.2 (1.0-1.5)

1.1 (0.7-1.6)

0.8 (0.4-1.7)

1.1 (0.5-2.4)

0.8 (0.4-1.7)

0.7 (0.3-1.3)

0.5 (0.2-1.4)

0.5 (0.1-1.7)

0.5 (0.2-1.3)

ref

1.0 (1.0-1.0)

1.0 (1.0-1.1)

1.0 (0.9-1.1)

3.2 (0.9-11.5)

1.0 (1.0-1.0)

1.0 (0.9-1.1)

1.2 (0.6-2.3)

0.03

0.68

0.60

0.77

0.56

0.27

0.15

0.34

0.90

0.07

0.53

0.63

0.60

p-value

1.3 (1.0-1.5)

0.02

p-value

Multivariable OR

n number of patients; OR odds ratio; p p-value; range minimum and maximum value; yr year; d day; ns excluded in multivariable model by backward as well as forward procedure because p>0.05; ref reference category; GABAergic one or more of the anti-epileptic drugs used enhances γ-aminobutyric acid activity; CYP450 inducing one or more of the anti-epileptic drugs used induces cytochrome P450 enzymes; bold p < 0.2 in univariable analysis; Italic p < 0.5 in multivariable analysis.

0 (0 – 3)

24.6

-CYP450 inducing

negative life events past year, median (range)

40.6

- GABAergic

anti-epileptic drugs %

22.8

28.1

- unknown 61.4

8.8

- metabolic

temporal seizure onset, %

38.6

- structural

focal, %

24.6

- genetic

etiology, %

5.7 (0.4 – 14.5)

duration of epilepsy, median (range)

12.3 2.6yr (0.0 – 10.1)

age at onset of epilepsy, median (range)

autistic spectrum disorder, %

63.5 (17 – 115)

intelligence quotient, median (range)

54.4 9.1yr (2.2 – 16.8)

age, median (range)

sex, % male

no (n = 96)

yes (n = 57)

acute stress precipitates seizures

Table 3. Clinical characteristics of children with and children without seizures precipitated by acute stress

PART ONE | CHAPTER 3

Stress sensitivity of childhood epilepsy

DISCUSSION This explorative observational study revealed that in 51% of children with epilepsy, seizures may be provoked by stress. In 39% of patients an increase in seizure frequency was reported during periods of stress, and in 37% acute stress was noted to be a seizure precipitating factor. Stress sensitivity was significantly related to the amount of experienced negative life events in total life. No association was found between stress sensitivity and other characteristics. The results of this study, focussing on children with epilepsy, are in agreement with the results of previous questionnaire as well as diary based studies on precipitants of epilepsy in generally adult populations, reporting stress as a possible triggering factor in 14 to 66% of patients with epilepsy (Temkin and Davis, 1984; Hayden et al., 1992; Neugebauer et al., 1994; Hart and Shorvon, 1995; Cull et al., 1996; Spatt et al., 1998; Frucht et al., 2000; Spector et al., 2000; da Silva Sousa et al., 2005; Nakken et al., 2005; Haut et al., 2007; Sperling et al., 2008; Pinikahana and Dono, 2009; dos Santos Lunardi et al., 2011). Frucht et al. (2000) found that stress sensitivity was more common in females and differed between epilepsy syndromes. Both findings were not confirmed in the present study, which might be explained by a difference in study population, as our study included children exclusively. Stress susceptibility of seizures is more likely to be sex dependent in adults, as (young) adult females in the follicular phase of the menstrual cycle, and women taking oral contraceptives, show diminished free cortisol responses compared with men (Kudielka et al., 2004; Kudielka and Kirschbaum, 2005; Kajantie and Phillips, 2006). In the present study, only nine girls were aged above 13.15 years, the mean age of menarche in the Netherlands (Fredriks et al., 2000) and none of the study participants used oral contraceptives. The reported effect of previously experienced life events on stress sensitivity of epilepsy has, to our knowledge, never been described before. However, there are several studies reporting an effect of experienced life events on other aspects of epilepsy. In an exploratory study, Webster and Mawer (1989) reported major life events to be related to seizure frequency in 6 of 18 patients with epilepsy. Their results suggest an increase in seizure frequency after negative life events, and a decrease after positive life events. In the present study, only increases in seizure frequency were reported, and these were reported even more often in positive than in negative periods of stress. In addition, prenatal life events, experienced by the mother, have been shown to be associated with an increased risk of infantile spasms (Shang et al., 2010). The power of these studies, however, was very limited compared with the present study.

Study design and limitations Due to the retrospective study design with use of questionnaires, a certain amount of recall and selection bias could not be prevented. To minimize recall-bias, we included only patients with active epilepsy. To address selection bias, which could be caused by the high nonresponse rate of 56.7%, characteristics of responders and nonresponders were compared. Stress as a precipitating factor for seizures was more often documented in

PART ONE | CHAPTER 3

responders compared with nonresponders. This difference may have given rise to an overestimation of the prevalence of stress sensitivity among responders, although information about stress sensitivity in patient files was just available in 11.6% of children, and was only reported when positive. Moreover, the information about stress sensitivity obtained from patient files did not correspond well to the answers given in this questionnaire. In addition, our results confirm previous findings in adults. Furthermore, this study was performed in a tertiary referral center for children with (often intractable) epilepsy, so that the disease severity is probably higher than in general practice. For these reasons, caution should be taken when translating our results automatically to all children with epilepsy. Because questionnaires were completed by parents (if possible with their children), recall of stress sensitivity as well as life events may well be influenced by parentsâ&#x20AC;&#x2122; stress perception (rather than the stress experienced by the child), their coping capabilities and their ability to recognise which situations are stressful for their child. As in every study in which data are obtained by self or parental report, this impairs reliability of the results. The higher odds for stress-sensitive epilepsy in children who experienced more life events could be (partially) explained by this group of children and their parents paying more attention to the occurrence of stressful events in general. Less subjective data could be obtained using a prospective study design in which the occurrence of seizures as well as specific events, known to be stressful for a child of a certain age, could be reported without adding an interpretation. A prospective diary study is currently being undertaken. The reported increase in seizure frequency during stressful periods may be partially explained by sleep deprivation, often also associated with stressful periods. Indeed, a high correlation was observed between reported sensitivity for periods of stress and a decrease in sleep quality or quantity during these periods, although not all children who were sensitive to periods of stress reported a decrease in quality or quantity of sleep during stressful periods. We cannot exclude, though, that a decrease in sleep quality or quantity in stressful periods may partially mediate the relation between life events and stress sensitivity of epilepsy. Obviously, precipitation of seizures by acute stress cannot be attributed to sleep deprivation; the same holds for the relation between life events and epileptic seizures precipitated by acute stressors.

Possible mechanisms How might acute stressors or periods of stress affect seizure frequency, especially in individuals exposed to multiple life events? In animal studies, the stress mediator corticotrophin-releasing hormone (CRH) has a proconvulsive effect, whereas neurosteroids have an anticonvulsive effect (Biagini et al., 2010); other stress hormones, such as ACTH, corticosterone (which is the equivalent of cortisol in rodents) and noradrenaline, show both proconvulsive and anticonvulsive properties (JoĂŤls, 2009; Sawyer and Escayg, 2010). Accordingly, exposure to a brief period of stress can increase as well as decrease seizure susceptibility depending on the method of seizure induction and the type of stressor used. Stress hormones are thought to affect seizure susceptibility by influencing hippocampal

Stress sensitivity of childhood epilepsy

N-methyl-D-aspartate (NMDA) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors, involved in glutamatergic excitation, as well as by changing calcium influx into neurons (Joëls, 2009). In contrast to the mixed effects of a brief period of stress, chronic stress exposure and stress experienced early in life were consistently found to be proconvulsive (Joëls, 2009; Sawyer and Escayg, 2010). This may relate to long-lasting changes in the properties of the hypothalamic-pituitary-adrenal axis reported for individuals with a history of chronic or early life stress, generally causing a higher exposure to e.g. corticosteroid hormones after each stressful event, although hyporesponsiveness has also been described, depending on the developmental stage during exposure to stress and the type of stressor used (Meaney et al., 1996; Sanchez et al., 2001; Fries et al., 2005; McEwen, 2007). Moreover, glutamatedependent processes in the hippocampus are differently affected by corticosteroid hormones in organisms with a history of chronic stress or early life events (Karst and Joëls, 2003; Joëls et al., 2004; Oomen et al., 2010). It is very well possible that in humans too the cumulative exposure to life events causes an enhanced corticosteroid release and altered effects of stress hormones on glutamatergic processes, as observed in rodent models. The evidence for this assumption, though, is presently only circumstantial. Thus, human studies investigating the link between stress and epilepsy did demonstrate changes in stress hormone levels in patients with epilepsy. Most studies have shown increased baseline levels of ACTH and cortisol, a prolonged increase of cortisol levels after a stressor and a slow return to baseline levels (Rao et al., 1989; Zobel et al., 2004), supposedly due to impaired negative feedback of the HPA-axis. However, there are also studies with contradictory results, demonstrating a decrease in baseline ACTH (Motta et al., 1994) or cortisol (Hill et al., 2011) in patients with epilepsy (for extensive reviews on the hormonal stress response in epilepsy, see Joëls, 2009 and Sawyer and Escayg, 2010). To what extent these altered neuroendocrine responses in epileptic patients relate to their cumulative life event exposure was not examined. Yet, altered neuroendocrine responses in association with increased disease vulnerability as a function of the number of experienced stressful life events is certainly not unprecedented. This was demonstrated for psychiatric conditions such as depression and anxiety disorders (Tennant, 2002; Nemeroff, 2004; Hulbert-Williams and Hastings, 2008; Nugent et al., 2011; Heim and Binder, 2012), but also cardiovascular disease, asthma and exacerbations of multiple sclerosis have been shown to be more common in patients who have experienced more life events (Tosevski and Milovancevic, 2006; Mitsonis et al., 2009). An interesting aspect of cumulative life event exposure is that epileptic seizures themselves can also be a cause of stress and hence influence regulation of the HPA-axis (Joëls, 2009; Sawyer and Escayg, 2010). However, analysing the effects of stress on seizure frequency, including the experience of an epileptic seizure as a stressful event, or living with a diagnosis of epilepsy (i.e. repetitive and uncontrollable seizures) as chronic stressor, raises methodological problems. It would be impossible to differentiate the effect of stress caused by experiencing a seizure on the next seizure, from other, combined, seizure precipitating factors, and the effect of stress caused by coping with epilepsy from duration of epilepsy

PART ONE | CHAPTER 3

and stress coping style. In the current study design, stress caused by epileptic seizures was therefore not taken into account.

Implications The high reported prevalence of stress sensitivity in childhood epilepsy and its relation to the reported number of previous stressful life events in responders to this questionnaire, suggests that stress is an important cofactor of seizure occurrence in children with epilepsy. This underlines the counselling needed for stress sensitivity of epilepsy, because early psychosocial support could improve coping with stressful situations. This may not only prevent an increase of seizure frequency after stress, but also cause fewer alterations in the HPA-axis, putatively resulting in a better prognosis of seizure outcome as well as decreasing the effect on functional networks and cognition. In conclusion, this study shows that stress sensitivity of childhood epilepsy is very often reported. We found that reported stress sensitivity of childhood epilepsy is related to the reported number of negative stressful events experienced earlier in life. The underlying mechanism, that links stress hormones and life events to childhood epilepsy, requires further study. Ultimately, better insight in the relation between stress hormones and epilepsy may provide a basis for new treatment strategies.

CHAPTER

Does Saint Nicholas provoke seizures? Hints from GoogleTrends Epilepsy & Behavior 2014

Jolien S. van Campen Eric van Diessen Willem M. Otte Marian JoĂŤls Floor E. Jansen Kees P.J. Braun

PART ONE | CHAPTER 4

ABSTRACT Introduction Stress is the most often reported seizure-precipitant in epilepsy. As most evidence for the relation between stress and epilepsy is derived from human self-reports, observational studies including a larger part of the population could provide additional proof. A stressor often reported to increase seizure frequency in children with epilepsy in the Netherlands is the national celebration of Saint Nicholas’ eve (December 5) and the weeks before; this is the main period of festivities for children in this country. To study the relation between stress and epilepsy, we analyzed epilepsy information-seeking behavior on the Internet, an indirect measure of seizure frequency, around this national children’s celebration. Methods Google Trends was used to extract relative search percentages for ‘epilepsy’ on Google in the Netherlands, the United States and the United Kingdom, between 2004 and 2013. Relative search percentages during the Saint Nicholas period were compared with baseline. Results Epilepsy searches increased with 14% in the Saint Nicholas period compared with baseline (p < 0.001). This effect was not found for searches performed in the same period in the United States or the United Kingdom, countries where this holiday is not celebrated. Conclusions The increase in epilepsy information-seeking behavior in the Saint Nicholas period is possibly caused by an increased occurrence of epileptic seizures. This underscores the potential of health information-seeking behavior on the Internet to answer clinically relevant research questions and provides circumstantial evidence for a relation between stress and the occurrence of epileptic seizures.

Saint Nicholas and epileptic seizures

INTRODUCTION Stress is the most often reported seizure-precipitant in epilepsy (van Campen et al., 2014). Results from self-report questionnaires or diaries still need to be confirmed by larger scale studies. Although people may experience stressors of various nature at different times, some events induce stress in a large part of the population. A national celebration in the Netherlands that is often considered to be very stressful for children is Saint Nicholas (‘Sinterklaas’ in Dutch). Saint Nicholas, a historical figure closely related to Santa Claus, and his helpers, the ‘Black Petes’ (Figure 1), arrive by boat in the Netherlands around mid-November. This arrival is broadcasted on national television and demarks the beginning of a three-to-four-week stay. The myth says that Saint Nicholas and his helpers leave well-behaved children presents but punish those who have been naughty. Children sing songs to please the Saint, hoping to receive presents, which are left in their shoes at night. The celebration of Saint Nicholas’ anniversary on the 5th of December marks the end of his stay. The Saint Nicholas period is generally experienced as exciting and stressful, especially for children but also for adults (Helsloot, 2011). In a recent questionnaire study, the Saint Nicholas period was reported to increase seizure frequency in 16% of children with epilepsy (van Campen et al., 2012). Patients increasingly use search engines on the Internet to retrieve medical information (Dickerson et al., 2004), including information on epileptic seizures (Brigo et al., 2014). Therefore, Google queries can provide indirect, but accurate, estimates of incidences of diseases, such as epilepsy and epileptic seizures (Polgreen et al., 2008; Ginsberg et al., 2009; Carneiro and Mylonakis, 2009). In Figure 1. this study, we relate the stressful national celebration of Saint Nicholas and Black Pete People dressed like Saint Nicholas Saint Nicholas to epilepsy using epilepsy information- and Black Pete, photographed by DeEchteSint.nl. seeking behavior.

METHODS Data collection Epilepsy information-seeking behavior on the worldwide web was quantified using Google TrendsTM, a web tool of Google Inc., freely available at www.google.com/trends/. Google TrendsTM provides the search volume for entered queries per week, relative to the total number of queries performed on Google. On September 22, 2013, the relative search volume of the term ‘epilepsy’ or its appropriate translation (‘epilepsie’ in Dutch) were extracted for the Netherlands, the United States of America (USA) and the United Kingdom (UK)

PART ONE | CHAPTER 4

between June 2004 and September 2013,without the use of a category filter. The term ‘seizure(s)’ was not used because its Dutch translation is not specific to epilepsy. The Saint Nicholas period was defined as the period between the arrival of Saint Nicholas (midNovember) and the 5th of December. Two control conditions were used for data verification. First, epilepsy information-seeking behavior in the same period was quantified in the United Kingdom and the United States. Both countries do not celebrate Saint Nicholas. Second, school holidays may also influence the frequency of Internet searches and thereby epilepsy information-seeking behavior. Therefore, data were separately analyzed using a baseline consisting of all weeks in the year outside of the Saint Nicholas period and using a baseline also excluding the summer (JulySeptember; week 27-35) and Christmas/New Year (week 52-1) holiday periods.

Statistical analysis Data of multiple years were combined. To correct for changes in search behavior over the years, data were converted to search percentages over 12 months around the Saint Nicholas period, every year ranging from the month June prior to Saint Nicholas to the month May after, correcting for the number of weeks in the year. For every week, the percentage increase in search volume compared with the mean of the corresponding year was computed. Search percentages during the total period of Saint Nicholas’ stay were compared with baseline (defined above) with an independent samples t-test. Search results in the individual weeks around Saint Nicholas were compared with baseline values with a Mann-Whitney U-test. Results were corrected for multiple comparison (Bonferroni). Significance level was set at p = 0.05. Statistical analysis was performed in SPSS (version 20.0).

RESULTS Mean searches for ‘epilepsy’ in the Netherlands increased by 14% in the Saint Nicholas period compared with baseline (p < 0.001), while no significant increase was found in the USA (p = 0.14) or the UK (p = 1.00) after correction for multiple comparison (Figure 2). The increase in ‘epilepsy’ searches in the Saint Nicholas period in the Netherlands was present for every year, except for the Saint Nicholas period in 2010 (supplementary Figure 1). The school holidays were associated with a decrease in ‘epilepsy’ searches in all three countries ranging from 11 to 22% compared with non-holiday weeks (p < 0.001 for all countries). Compared with the non-holiday baseline, ‘epilepsy’ searches in the Netherlands still increased by 7% in the Saint Nicholas period (p = 0.03), while no differences were found in the USA (p = 1.00) or the UK (p = 1.00). Looking into weekly searches in the Netherlands around Saint Nicholas, the relative number of ‘epilepsy’ searches was significantly increased at the arrival of Saint Nicholas (p = 0.007), and the week after arrival (p = 0.049) compared with baseline (Figure 3). The small peak in ‘epilepsy’ searches in the week of December 5, the actual anniversary of Saint Nicholas and the end of his stay, was not significant (p = 0.48).

Saint Nicholas and epileptic seizures

Figure 2. Epilepsy information-seeking behavior in the Saint Nicholas period in different countries Percentage increase in ‘epilepsy’ search queries during the stay of Saint Nicholas in the Netherlands compared with baseline. A. Saint Nicholas period compared with the overall baseline (i.e., all year except the Saint Nicholas period), B. Saint Nicholas period compared with the non-holidays baseline (i.e., all year except the Saint Nicholas and holiday periods). NL the Netherlands; USA United States of America; UK United Kingdom. Data are presented as mean ± 95% confidence interval, *p < 0.05.

DISCUSSION During Saint Nicholas’ stay, the relative number of Google searches for ‘epilepsy’ in the Netherlands was significantly increased compared with baseline, in contrast to countries that do not have this national celebration. This specific increase in epilepsy information-seeking behavior might be caused by an increased occurrence of epileptic seizures, as search behavior has previously been shown to closely relate to disease prevalence (Polgreen et al., 2008; Pelat et al., 2009; Carneiro and Mylonakis, 2009; Ginsberg et al., 2009; Seifter et al., 2010). The relation between Saint Nicholas and ‘epilepsy’ search behavior could well be mediated by stress, as the Saint Nicholas holiday period is reported to be stressful by a large percentage of children and adults (Helsloot, 2011; van Campen et al., 2012). The relation between stress and epilepsy is a complex one, as many psychological and physiological factors are involved. One explanation could be found in the effects of altered stress hormones on neuronal functioning, both directly and indirectly. Multiple stress hormones can influence the excitability of the brain and, subsequently, increase the risk of seizures in patients with epilepsy (Joëls, 2009). The absence of a similar pattern in epilepsy information- seeking behavior during the same period in other countries supports our hypothesis that the Saint Nicholas’ celebration is a seizure-precipitating factor. Our results, based on the search behavior of millions of Internet users, are in line with the positive association between stress and epilepsy that has been reported in previous studies using self-report questionnaires and seizure diaries in humans, as well as animal experiments, as recently reviewed (van Campen et al., 2014). The increase in epilepsy information-seeking behavior during (inter)national celebrations is probably not specific for Saint Nicholas. However, Google TrendsTM might not be specific enough to analyze effects of one-day feasts like Halloween in the United Kingdom and United States, as search queries are only provided per week. Interestingly, we have no indication that

PART ONE | CHAPTER 4

‘epilepsy’ search behavior is increased during the stay of Santa Claus in the UK or USA. Rather, the period of the Christmas holiday – similar to the summer holidays – was characterized by a significant drop in search volumes in all three countries examined. The time off from work or school might result in less healthcare contacts or Internet access. In addition, people likely experience less stress during holidays, which can affect disease characteristics. As our results are based on aggregated Internet search queries, we can only speculate on the cause of the relation between Saint Nicholas and epilepsy information-seeking behavior. Besides the stress caused by Saint Nicholas and the Black Petes, other factors that may play a role in epilepsy information-seeking behavior during Saint Nicholas’ stay could include (stress-related) sleep deprivation and more parental attention. Also, other events influencing ‘epilepsy’ search behavior during the Saint Nicholas period as well as during baseline might affect the results, although the influence of non-annual events is limited by combining data of multiple years. We are unaware of a comparable nationally celebrated period of festivities during another time of the year, equally non-associated with time off from work or school. Therefore, causality of the reported association can still be debated, even though it is supported by the course of the search volume over individual weeks of the Saint Nicholas period (Figure 3).

Figure 3. Epilepsy information-seeking behavior during the Saint Nicholas period per week Percentage increase in ‘epilepsy’ search queries during the stay of Saint Nicholas in the Netherlands (black bar), and the weeks before and after, compared with baseline. After the arrival of Saint Nicholas and the Black Petes by boat, children receive small presents (typically left in their shoes) until the final celebration on December 5th. Data are presented as mean ± 95% confidence interval, *p < 0.05 compared with baseline.

Saint Nicholas and epileptic seizures

Another limitation of this study is that the algorithms used by Google for data preparation are not fully disclosed. Although Google has made relative search query data freely available to the public, detailed specifics on algorithm normalization steps and updates are lacking. As a result, extracted data could slightly differ depending on the day of data retrieval. Although results should therefore be interpreted with caution, they can provide valuable clues and circumstantial evidence. In conclusion, the current study illustrates the potential of Internet acquired data, such as health information-seeking behavior, to answer clinically relevant research questions, and it provides circumstantial evidence for a relation between stress and the occurrence of epileptic seizures.

PART ONE | CHAPTER 4

Supplement

Supplementary Figure 1. Epilepsy information-seeking behavior in the Saint Nicholas period in the Netherlands per year Percentage increase in â&#x20AC;&#x2DC;epilepsyâ&#x20AC;&#x2122; search queries during the stay of Saint Nicholas in the Netherlands compared with baseline per year (every year ranging from June to May).

CHAPTER

Sensory modulation disorders in childhood epilepsy submitted

Jolien S. van Campen Floor E. Jansen Nienke J. Kleinrensink Marian JoĂŤls Kees P.J. Braun Hilgo Bruining

PART ONE | CHAPTER 5

ABSTRACT Introduction It is a common notion that seizures in childhood epilepsy may be provoked by stress. The subjective experience of stress preceding seizures is often associated with excessive exposure to sensory stimuli of different modalities. This may indicate that children with epilepsy have difficulties in modulating their response to sensory inputs. To test this hypothesis, we assessed the prevalence of sensory modulation disorders among children with epilepsy and their relation with stress sensitivity of seizures. Methods We used the Sensory Profile to assess behavioral responses to sensory stimuli and categorize sensory modulation disorders in children with active epilepsy (aged 4-17 years). We related these outcomes to epilepsy characteristics, including stress sensitivity of seizures, and tested their association with comorbid symptoms of developmental psychopathology. Results Sensory modulation disorders were reported in 49% of the 158 children. Children with epilepsy reported increased behavioral responses associated with ‘sensory sensitivity’, ‘sensory avoidance’ and ‘poor registration’, but not ‘sensory seeking’. Sensitivity of seizures to acute stress situations was related to behavior of sensory overload. Comorbidity of developmental psychopathology was associated with more severe sensory modulation problems, although 27% of normally developing children with epilepsy also reported a sensory modulation disorder. Conclusions Sensory modulation disorders are an under recognized problem in children with epilepsy and seem to mediate acute stress sensitivity of seizures. The extent of the modulation difficulties indicates a substantial burden on daily functioning and may explain an important part of the behavioral distress associated with childhood epilepsy.

Sensory modulation in childhood epilepsy

INTRODUCTION Epilepsy is among the most common chronic diseases in childhood, affecting 0.5-1% of children (Shinnar and Pellock, 2002). Seizures cause a major disease burden and greatly affect cognitive and behavioral development (Ferro et al., 2014). In the majority of children with epilepsy, seizures can be provoked or precipitated. Stress is the seizure precipitant most frequently reported (Cull et al., 1996; Frucht et al., 2000; da Silva Sousa et al., 2005; Nakken et al., 2005; Pinikahana and Dono, 2009; van Campen et al., 2012), but the mechanistic underpinnings of stress sensitivity of epilepsy are unresolved. The subjective experience of stress is often associated with exposure to an excess of sensory stimuli of different modalities, both during positive events, such as birthdays, and during negative events, such as fights. Although seizures that are repetitively provoked by sensory stimuli (i.e., reflex seizures) only occur in a minority of patients (Symonds, 1959; Kasteleijn-Nolst Trenite, 1989; Mani et al., 1998), subclinically, children with epilepsy often exhibit altered neuronal responses to sensory stimulation compared with controls (Eeg-Olofsson et al., 1971; Doose and Gerken, 1973; Wolf and Goosses, 1986; Seri et al., 1998; Abubakr and Wambacq, 2003; Grant, 2005; Fiedler et al., 2006; Wang et al., 2011; Chipaux et al., 2013). Also in the developmental psychopathologies frequently co-occurring with childhood epilepsy (Ratey and Johnson, 1997; Charman, 2008), symptoms can be aggravated by stress or an excess of sensory stimulation (referred to as ‘sensory overload’) (Ratey and Johnson, 1997). In these disorders, the capacity to regulate responses to sensory input in a graded and adaptive manner is generally referred to as sensory modulation (Mulligan, 2002). Dunn proposed a model in which sensory modulation is characterized by four behavioral patterns: (1) sensory sensitivity—discomfort and distractibility caused by intense stimuli, (2) sensory avoiding—controlling or limiting the amount and type of sensations, (3) poor registration— lack of, or low awareness of sensations, and (4) sensory seeking—enjoyment of sensations and interest in increasing them (Dunn and Brown, 1997). Abnormal patterns of sensory modulation in children are known to interfere with effective learning, daily functioning (Dunn and Brown, 1997), and interactions (Ornitz et al., 1978; Talay-Ongan and Wood, 2000), and are referred to as sensory modulation disorders (SMDs) (Ornitz et al., 1978; Dunn and Brown, 1997; Talay-Ongan and Wood, 2000; Miller et al., 2007; Miller et al., 2009). SMDs are common in other neurodevelopmental disorders such as such as autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) (Dunn, 2001; Mangeot et al., 2001; Kern et al., 2006; Miller et al., 2012; Dunn, 2013; Ludlow et al., 2014), both frequently comorbid with epilepsy. Furthermore, SMDs have been linked to an imbalance between neuronal excitation and inhibition (Davies and Gavin, 2007; Davies et al., 2009; Davies et al., 2010; Gavin et al., 2011; Zhang and Sun, 2011), which is a key feature of epilepsy, being present both during and in between seizures (Demont-Guignard et al., 2012). Taken together, we assumed that disordered sensory modulation might be a frequent problem in children with epilepsy which might contribute to stress sensitivity of seizures. As a first step to investigate the presence and role of sensory modulation in childhood epilepsy, we investigated (1) the prevalence of atypical behavioral responses to

PART ONE | CHAPTER 5

sensory stimuli and SMDs in children with epilepsy and their association with developmental psychopathology, and (2) the relation between atypical behavioral responses to sensory stimuli and stress sensitivity of epilepsy.

METHODS Patients We retrospectively selected all children, aged 4 to 17 years, with active epilepsy (i.e., a definitive clinical diagnosis of epilepsy and seizures within one year prior to data collection), who consulted the pediatric neurology outpatient clinic of the University Medical Center Utrecht between September 2012 and October 2013. Children of whom caregivers could not recognize seizures, those who were well-controlled with AEDs (seizure-free for over a year) or since epilepsy surgery, and children with non-Dutch speaking caregivers were excluded. The study was approved by the institutional ethical committee. Questionnaires were sent to all caregivers. In case of non-response, caregivers were contacted by phone, asked for the reason of non-response and motivated to complete the questionnaires. Parents or legal guardians and children aged ≥ 12 years provided informed consent. Children were retrospectively excluded when additional parental information revealed that children no longer fulfilled inclusion criteria or had a developmental age < 2 years. Methods General demographics and epilepsy characteristics Information on demographic and epilepsy characteristics, intellectual disability and comorbidities (including ASD and ADHD) were extracted from patient files. Epilepsy characteristics were classified according to the terminology proposed by the ILAE (Berg et al., 2010). Epilepsy localization was classified as generalized or focal based on the diagnosis made by the treating pediatric neurologist. Focal epilepsies were further specified with respect to the hemisphere and lobe of origin. Characteristics of responders and nonresponders were compared to estimate selection bias. Children and their caregivers were questioned about the age of seizure onset, seizure frequency, current use of medication, comorbidities (including ASD and ADHD), psychiatric family history, school type and intelligence or developmental quotient (IQ resp. DQ) (both referred to as ‘intelligence’). When diagnostic reports of ASD or ADHD diagnoses and IQ test results could not be extracted from patient files, these were provided by parents. Anti-epileptic drugs used were classified according to their main known mechanism of action as γ-aminobutyric acid (GABA) enhancers, glutamate receptor antagonists, sodium channel blockers, calcium channels blockers or other, according to Stafstrom (2010). In addition, information on reflex seizures and self-induction was obtained from patient files, and the response to photostimulation was extracted from EEG reports.

Sensory modulation in childhood epilepsy

Sensory modulation Behavioral responses to sensory stimuli were quantified with the Sensory Profile-method, consisting of well-validated questionnaires for different age groups (Brown and Dunn, 2007; Dunn, 2013). Behavioral responses to sensory stimuli are measured on four quadrants based on the interaction between the sensory threshold (i.e. the threshold for neurons to get activated by sensory stimuli, ranging from low to high) and the amount of self-regulating behavior (i.e. behavior used to regulate the sensory input, ranging from passive to active), see Figure (Brown and Dunn, 2007; Gere et al., 2009; Dunn, 2013). Subscores also exist for sensory processing modalities, modulation and behavior and emotional responses (Figure 2). Results correlate well with diagnoses of SMD made by occupational therapists and physiological responses to sensory stimuli (Davies and Gavin, 2007; Davies et al., 2010; Gavin et al., 2011; Marco et al., 2012; Miller et al., 2012; Owen et al., 2013; Ludlow et al., 2014). Caregivers of children aged 4 to 12 years completed the parental Dutch version of the Sensory Profile (SP-NL; (Dunn, 2013). Children aged 13 to 17 years completed the selfcompleted version of the Sensory Profile (Adolescent/Adult Sensory Profile [AASP]; (Brown and Dunn, 2007). Normed parental scores of children 4 to 12 years and self-scores of children 13-16 years were used for analysis. Z-scores were based on the standard deviations (SD) of the standardization samples published in the instruction manuals, available for SP-NL quadrants and subsections and for AASP quadrant self-scores (Brown and Dunn, 2007; Dunn, 2013). As results of SP-NL and AASP are indicated in the opposite direction, SP-NL z-scores were transposed so that higher scores indicated more symptoms.

Figure 1. Sensory modulation quadrants Sensory modulation quadrants: relationship between sensory threshold and self-regulation (Brown and Dunn, 2007; Gere et al., 2009; Dunn, 2013). Figure adapted from Dunn, 1997.

PART ONE | CHAPTER 5

In this study, SMDs were defined as a score â&#x2030;Ľ2 SD above the mean on one or more of the quadrant scores (Taal et al., 2013). To investigate the prevalence and nature of SMDs in children with epilepsy, quadrant scores (for all children) and subsections (for children 4 to 12 years) were compared with the norm populations. In addition, associations between quadrant scores and demographic, epilepsy and developmental characteristics were examined. To acquire more insight in the relation between sensory modulation and seizure precipitation, a question was added at the end of the subsections auditory, visual, vestibular, touch and oral sensory processing, asking whether one or more of the items stated in that subsection could precipitate epileptic seizures.

Figure 2. Sensory modulation scores in children with epilepsy Sensory Profile quadrant scores of all children 4-17 years of age (n = 158), and subsection scores of children 4-12 years of age (n = 120). Data are presented as mean z-scores (based on age specific norm scores (Brown and Dunn, 2007; Dunn, 2013) and 95% confidence interval (CI), *p-value <0.001 compared with the norm population.

Sensory modulation in childhood epilepsy

Stress sensitivity of epilepsy The relation between stress and epileptic seizures was quantified with a questionnaire previously designed to obtain information on the relation between stress and epileptic seizures (van Campen et al., 2012). In this questionnaire, parents and children together, report whether or not seizures could be precipitated by acute stress (defined as stress lasting minutes to several hours), and whether seizure frequency increased during periods of stress (defined as stress lasting days to weeks). Overall stress sensitivity of seizures was defined as a positive answer to one or both questions. To assess the relation between stress sensitivity of epilepsy and sensory modulation, we compared the prevalence of SMDs in children with and without stress-sensitive epilepsy. Symptoms of developmental psychopathology Autism traits were measured using the Dutch Social Responsiveness Scale (SRS), a well validated 65-item parental questionnaire (Constantino and Gruber, 2011). Raw total scores were converted into age- and sex-specific T-scores using norm scores derived from the Dutch standardization sample. ADHD symptoms were measured using the Dutch parental Strength and Difficulties Questionnaire (SDQ), 25-item version (Goodman, 1997; van Widenfelt et al., 2003). Scores on the hyperactivity-attention scale of this well validated questionnaire were used as measure of ADHD symptoms (Goodman et al., 2000; Goodman et al., 2000). Normal development was defined as having no existing clinical diagnosis of ADHD or ASD, SRS score for ASD symptoms < 75, SDQ score for ADHD symptoms < 7, and (estimated) IQ ≥ 70. Data analysis Group comparisons were performed with Fisher’s exact tests for binominal data, chi-square tests for non-dichotomous nominal data, t-tests for continuous data and Mann-Whitney tests for continuous data not meeting the assumptions of the t-test. Unequal variances t-tests were used to compare the study group with the combined norm group of the Sensory Profile (as a large difference in sample size exists and a larger variance is expected in the epilepsy group), followed by Bonferroni correction for multiple comparison with adjustment for correlation between outcome parameters (Sankoh et al., 1997). Correlations between continuous variables were tested with Pearson correlation coefficient (r) or Spearman correlation coefficient (ρ) based on compliance to assumptions. Associations between patient characteristics and sensory profile quadrant scores were examined with linear regression analysis. To simultaneously analyze outcomes on all four sensory profile quadrants, multivariate regression analysis was used. Patient characteristics with a univariable association of p < 0.1 were included as candidates into a multivariable regression model to maximize sensitivity. Variables were checked for multicollinearity using variance inflation factor (VIF) and tolerance, where a VIF ≥ 5 and tolerance ≤ 0.10 were considered to indicate a multicollinearity problem. Independent variables were removed by a backward stepwise selection procedure (threshold for removal p ≥ 0.05). To confirm

PART ONE | CHAPTER 5

robustness of the models, multivariable analysis was repeated using a forward selection procedure. To determine whether certain patient characteristics were related to scores on individual quadrants, post-hoc univariate linear regression analyses were performed per quadrant, including all characteristics with an association of p < 0.05 in the multivariate multivariable regression analysis as dependent variables and a single quadrant as dependent variable. Associations between SMDs and stress sensitivity of epilepsy were tested with Fisherâ&#x20AC;&#x2122;s exact test. Posthoc, differences in sensory profile quadrant z-scores were compared between groups with an independent samples t-test. All post-hoc tests were Bonferroni corrected for multiple comparisons with adjustment for correlation between the quadrant scores (Sankoh et al., 1997). Normality was visually inspected using Q-Q plots. Homogeneity of variance was evaluated with box plots. Results with a two-tailed p-value of < 0.05 were considered significant. Data were analyzed using SPSS, version 20.0.

RESULTS Patient selection Questionnaires were sent to 375 patients. Sixty-six children were excluded retrospectively (exclusion criteria specified in supplementary Table 1). Of the remaining 309 children, 158 (51%) responded. Compared with non-responders, responders consisted of less males (51% versus 65%, this difference was most clear in the adolescent subgroup), and those with focal epilepsies showed more frontal and temporal seizure foci and less epilepsies with unknown localization (supplementary Table 2). Median age of responders at completion of the questionnaires was 9.6 years. Seizure frequency in the past three months varied between 0 per three months and 15 per day (median two per week). Most children (94.7%) were treated with anti-epileptic drugs at time of assessment. Demographics and epilepsy characteristics of responders are shown in Table 1. Sensory modulation SMDs were reported in nearly half (49%) of the study cohort (Figure 3). Although less frequent, SMDs were also reported in 27% of the children with epilepsy who had a normal development. Compared with the norm group, mean Sensory Profile scores were significantly increased for all quadrants except for sensory seeking (age 4 to 17 years), and for all subsections (age 4 to 12 years) (Figure 2). Quadrant scores were significantly correlated, but did not meet criteria for multicollinearity. In univariable analysis, sensory profile quadrant scores were significantly associated with age, intelligence, age at seizure onset, epilepsy localization, number of AEDs, and ASD as well as ADHD symptom scores. No relation existed with etiology, epilepsy classification, or seizure frequency (Table 2). In multivariable analysis, age, intelligence, ASD symptoms and ADHD symptoms remained significant. Posthoc tests for the individual quadrants

Figure 3. Sensory modulation in subgroups of children with epilepsy. A. Percentage of sensory modulation disorders. B. Sensory Profile quadrant scores. n, number of patients; ASD, autism spectrum disorder; ADHD, attention deficit hyperactivity disorder. Data are presented as mean z-scores and 95% confidence interval, based on age specific norm scores (Brown and Dunn, 2007; Dunn, 2013).

Sensory modulation in childhood epilepsy

PART ONE | CHAPTER 5

revealed that age was negatively correlated to all quadrants except ‘sensory sensitivity’, intelligence was negatively related to ‘sensory seeking’, ASD symptoms were positively correlated to all quadrants except ‘sensory seeking’ and ADHD symptoms were positively correlated to ‘sensory seeking’ and ‘sensory sensitivity’.

Sensory seizure precipitants Items in subsections of the sensory profile were reported to precipitate epileptic seizures in 27% of children. This was most often reported for items of auditory (16%) and visual (12%) processing, but also for vestibular (8%), touch (6%) and oral (5%) processing items. With EEG, photosensitivity was demonstrated in four children (3%), one of which had visual reflex seizures with self-induction, and in the other three a photoparoxysmal EEG response was described. Auditory reflex seizures were demonstrated in one child. Sensory profile scores per subsection were positively correlated to the self-reported likelihood of these items to provoke epileptic seizures (ρ = 0.25 - 0.32, p = 0.001 - 0.006), except for oral processing (ρ = 0.04, p = 0.64). Stress sensitivity of seizures Stress sensitivity of epilepsy was reported in half of children with epilepsy (50%), of which 34% reported seizures precipitated by acute stress and 37% an increase in seizure frequency during stressful periods (21% reported both). SMDs tended to be more frequently reported in children with stress sensitivity of epilepsy compared with children without (57 vs. 43%, p = 0.07). When analyzing acute stress and periods of stress separately, SMDs were significantly more often reported in children with seizures precipitated by acute stress compared with children without (63 vs. 43%, p = 0.02), while this was not the case for children reporting an increase in seizure frequency during periods of stress (49 vs. 50%, p = 0.53) (Figure 4A). Quadrant analysis showed significantly higher scores on both lowthreshold quadrants in children reporting seizures precipitated by acute stress compared with other children (p = 0.03 and p = 0.04 respectively). Although scores on the other quadrants were also increased, group differences did not reach significance (Figure 4B).

DISCUSSION This study shows that children with epilepsy report substantial problems in regulating their behavioral responses to sensory input, generally referred to as SMDs. Sensory modulation problems are related to stress sensitivity of seizures, as low thresholds for sensory stimuli are more common in children reporting epileptic seizures precipitated by acute stress. SMDs are known to perturb behavioral functioning and development, which is confirmed by their correlation with ASD and ADHD morbidity in our sample. Interestingly, SMDs are also reported by 27% of the normal developing children with epilepsy. Together, these findings indicate that SMDs are a substantial yet under recognized problem in childhood epilepsy.

Sensory modulation in childhood epilepsy

Table 1. Characteristics of study subjects characteristics

study participants (n = 158)

general sex, % male age, median (range)

seizure frequency, median (range)

9.6 yr (4.1 yr - 16.7 yr)

intelligence - IQ or DQ, median (range) - intellectual disability (IQ < 70), %

80 (23 - 148) 41

ASD 3.2 yr (0 d â&#x20AC;&#x201C; 15.0 yr) 2/wk (0/3 mo - 15/day)

etiology, % - genetic

- structural

- metabolic

- unknown

localisation, % - focal

study participants (n = 158)

neurodevelopmental 51

epilepsy age at onset of epilepsy, median (range)

characteristics

- frontal

- temporal

- parietal

- occipital

- unknown

- multifocal

- generalized

- unknown

- diagnosis, %

- ASD symptoms (SRS), % - severe ASD symptoms

- ASD symptoms

- mild/moderate ASD symptoms

- no ASD symptoms

ADHD - diagnosis, %

- ADHD symptoms (SDQ), % - abnormal

- borderline

- normal

anti-epileptic treatment, % - AED - GABA enhancers

95 50

- glutamate receptor antagonists

- sodium channel blockers

- calcium channel blockers

- other

- vagal nerve stimulator

- ketogenic diet

n number of patients; AED anti-epileptic drugs; IQ intelligence quotient; DQ developmental quotient; ASD autism spectrum disorder; SRS Social Responsiveness Scale; ADHD attention deficit hyperactivity disorder; SDQ Strength and Difficulties Questionnaire; h hour; d day; wk week; yr year.

The atypical behavioral responses to sensory stimuli in children with epilepsy might be associated with the neuronal hyperexcitability. Previous electrophysiological and imaging studies have shown increased sensitivity to sensory stimuli and overactive sensory processing networks in patients with epilepsy compared with controls (Seri et al., 1998; Abubakr and Wambacq, 2003; Grant, 2005; Fiedler et al., 2006; Wang et al., 2011; Chipaux

PART ONE | CHAPTER 5

et al., 2013). Our findings add that atypical behavioral responses to sensory stimuli are frequent in children with epilepsy and indirectly support the hypothesized role of the balance between excitation and inhibition in sensory modulation (Zhang and Sun, 2011). SMDs play a major role in ASD and ADHD (Dunn, 2001; Mangeot et al., 2001; Kern et al., 2006; Miller et al., 2009; Miller et al., 2012; Dunn, 2013; Ludlow et al., 2014). Also in our cohort we found a clear association between symptoms of these neurodevelopmental disorders and sensory modulation, suggesting that SMDs may explain an important part of the behavioral problems in childhood epilepsy. Since epilepsy and seizures are related to neuronal hyperexcitability, we expected childhood epilepsy to be associated with low

Figure 4. Sensory modulation and stress sensitivity of seizures in children with epilepsy A. Percentage of sensory modulation disorders in children with and without stress sensitivity of seizures. B. Mean sensory profile quadrant z-scores in children with and without stress sensitivity of seizures for acute stress. *p-value < 0.05.

Sensory modulation in childhood epilepsy

Table 2. Relation between patient characteristics and sensory modulation. characteristics

analysis (F)

univar

sex, male

posthoc tests (Beta)

multivara

poor registration

sensory seeking

sensory sensitivity

sensory avoiding

1.8

age

24.8***

24.6***

-0.41***

-0.13

-0.46***

intelligence (IQ or DQ)

14.2***

5.6***

-0.16

-0.21*

-0.12

0.15

age at onset of epilepsy

8.7***

2.0^

- genetic

0.6

- structural

1.3

- metabolic

1.0

- unknown

ref

- left

3.9**

- right

2.1*

- both

ref

- frontal

0.7

- temporal

1.5

- parietal

0.1

seizure frequency etiology

distribution, hemisphere

distribution, lobe

- occipital

1.3

4.6**

- GABA-enhancer

3.3*

- glutamate antagonist

1.9^

- sodium channel blocker

0.5

number of AEDs AED type

0.4

ASD symptoms (SRS)

- calcium channel blocker

13.9***

6.3***

0.29**

0.01

0.22*

0.45***

ADHD symptoms (SDQ)

18.3***

7.3***

0.10

0.33***

0.24**

-0.06

univar multivariate univariable regression analysis; multivar multivariate multivariable regression analysis; Posthoc tests univariate multiple regression analysis per quadrant, Bonferroni corrected for multiple comparison; IQ intelligence quotient; DQ developmental quotient; AED anti-epileptic drug; GABA Îł-aminobutyric acid, - not included in model, ns not significant. aAdjusted R2 poor registration = 0.453, sensory seeking = 0.463, sensory sensitivity = 0.269, sensory avoiding = 0.340. P-values: ^<0.1 (only indicated for multivariate univariable regression analysis), *<0.05, **<0.01, ***<0.001.

PART ONE | CHAPTER 5

sensory thresholds and the experience of sensory overload. However, atypical sensory responses patterns in our sample also included ‘poor registration’, generally attributed to a high sensory threshold (Dunn and Brown, 1997). A possible explanation might be found in compensatory inhibitory mechanisms (Badawy et al., 2009), inducing a paradoxal hypoexcitability. The effect of AEDs on sensory modulation is likely to be minimal, as AEDs did not relate to sensory modulation patterns in multivariable analysis. Additionally, previous studies have shown differences in sensory processing of evoked potentials in patients with epilepsy irrespective of AEDs (Lucking et al., 1970; Campanella et al., 1978; Knecht et al., 1996; Grant, 2005). Another possibility is that comorbidity with ASD or ADHD influences the pattern of SMDs. In children without comorbid ASD or ADHD, SMDs were also present, but they were overall less severe, which was particularly due to fewer behaviors related to ‘poor registration’, shifting the overall pattern of sensory modulation towards more sensory hypersensitivity (Figure 3). Together, our findings indicate an important relation between epilepsy, excitability and sensory behaviors, but this relationship is not straightforward and different processes might interact. Additionally, we found differences in sensory modulation between children with and those without stress sensitivity of seizures. As self-reports on stress sensitivity of epilepsy may be subject to recall bias, the validity of this problem is sometimes questioned, although the association between stress and seizures is supported by prospective studies as well as a large amount of preclinical research, as recently reviewed (van Campen et al., 2014). The reported differences in sensory modulation between children with and those without stress sensitivity of seizures provides further evidence for the seizure precipitating effects of acute stress in some children with epilepsy and may add to our understanding of a possible underlying mechanism. The association between sensory modulation and stress sensitivity of seizures differed between acute stress and periods of stress, suggesting that different mechanisms underlie their seizure precipitating effects. It should be noted that acute stress sensitivity was also often reported in the absence of SMDs, indicating that additional factors will be involved, e.g. stress hormone levels which can also influence neuronal excitability (van Campen et al., 2014). The increased prevalence of SMDs in children with seizures precipitated by acute stress, however, supports our hypothesis that an inapt response to sensory stimuli is associated with (acute) stress sensitivity of seizures.

Study design and limitations Due to data collection by questionnaires, a certain amount of recall and selection bias could not be prevented. To minimize recall bias, only patients with active epilepsy were included and responses were obtained concerning the current situation. To address selection bias, we compared characteristics of responders to non-responders. The decreased prevalence of males in responders, especially in the adolescent subgroup, is in line with the sex difference in survey response behavior that is well reported in adults (Moore and Tarnai, 2002), and results in an even distribution of males and females in the study cohort. The increased frequency of frontal epilepsies might suggest an overestimated prevalence of

Sensory modulation in childhood epilepsy

behavioral disorders, although effects on sensory modulation are probably limited, as regression analysis revealed no relation between quadrant scores and epilepsy localization. This study was performed in a tertiary referral center for children with epilepsy, where disease severity is expected to be higher than in general practice. Although we found no relation between atypical sensory processing and measures of disease severity in multivariable analysis, caution should be taken when extrapolating the results to all children with epilepsy. The influence of age found in this cohort, with more atypical scores in younger children, might be explained by the developmental trends in some sensory modalities in the first years of life (Dunn, 2001; Eeles et al., 2013), in combination with high prevalence of developmental delay. As for all cross-sectional studies, no evidence is provided on causality of the reported associations. Although altered sensory modulation may affect seizure susceptibility, the changes in brain function caused by recurrent seizure activity might also influence sensory modulation. Additionally, sensory modulation is quantified using questionnaires for behavioral responses. Although these correlate well to SMD diagnoses and physiological responses to sensory stimuli (Davies and Gavin, 2007; Davies et al., 2010; Gavin et al., 2011; Marco et al., 2012; Miller et al., 2012; Owen et al., 2013; Ludlow et al., 2014), unravelling differential effects of specific sensory modulation profiles is difficult as scores on different quadrants are correlated. Furthermore, it is unclear to which extent atypical sensory responses of children with epilepsy are caused by active seeking or avoidance of seizure triggers. Although scores on specific subsections of the sensory profile were significantly correlated to the likelihood of these items to provoke seizures, the higher scores on the quadrants representing a passive behavioral response compared with the active ones, might suggest that the SMDs in children with epilepsy are less likely to be caused by active coping mechanisms.

Implications The reported changes in sensory thresholds might pose a large daily burden on children with epilepsy and their caregivers and play a causal role in (part of) the behavioral problems associated with epilepsy. Behavioral manifestations of a low sensory threshold are more common in children with seizures precipitated by acute stress, suggesting that sensory overload plays a mediating role in stress sensitivity of seizures. Interventions that promote self-regulation have been shown helpful for patients with difficulties in sensory modulation in other neurodevelopmental disorders (Case-Smith et al., 2014), and might also be beneficial for children with epilepsy. Further studies on SMDs in childhood epilepsy are needed to unravel the relation between sensory processing, neuronal excitability and epileptic seizures. Conclusion Atypical behavioral responses to sensory stimuli are highly frequent in childhood epilepsy and relate to stress sensitivity of seizures. Increased attention for SMDs in children with epilepsy could benefit treatment and care.

PART ONE | CHAPTER 5

Supplement Supplementary Table 1. Children excluded after receiving questionnaires exclusion criteria

developmental age <2 years

last seizure >12 months ago

caregivers cannot recognize epileptic seizures

no understanding of Dutch language

age at completion of questionnaires > 17 years

seizure freedom after epilepsy surgery

death

total

n number of children excluded

Sensory modulation in childhood epilepsy

Supplementary Table 2. Characteristics of responders compared with non-responders characteristics

responders (n = 158)

non-responders (n = 151)

p-value

0.01

9.6 (4.1 - 16.7) yr

10.0 (4.1 - 16.6) yr

0.45

general sex, % male age, median (range) epilepsy etiology, %

0.78

- genetic

- structural

- metabolic

- unknown

1/wk (0/yr - 20/h)

1/wk (0/yr - 12/h)

0.64

- focal

0.91

- generalized

0.17

seizure frequency, median (range) seizure classification, % of patients

- epileptic spasms

0.62

- unknown

0.76

localisation, distribution, %

0.06

- focal

- multifocal

- generalized

- unknown

- frontal

<0.001

- temporal

0.006

- parietal

0.16

- occipital

0.60

- unknown

0.07

intellectual disability (IQ or DQ <70), %

0.36

ADHD diagnosis in patient file, %

0.10

ASD diagnosis in patient file, %

0.86

localisation, lobe, %

developmental

Data for both responders and non-responders based on information from patient files (differences compared with Table 1 are explained by the latter including additional information provided by responders). IQ intelligence quotient; DQ developmental quotient; ASD autism spectrum disorder; ADHD attention deficit hyperactivity disorder; h hour; wk week; yr year.

PART

TWO

Hormonal basis of stress sensitivity of seizures

CHAPTER

Relation between stress-precipitated seizures and the stress response in childhood epilepsy Brain 2015

Jolien S. van Campen Floor E. Jansen Milou A. Pet Willem M. Otte Manon H.J. Hillegers Marian JoĂŤls* Kees P.J. Braun* *authors share seniority

PART TWO | CHAPTER 6

ABSTRACT Introduction 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. Methods 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 (1) a retrospective questionnaire and (2) a prospective sixweek diary on stress and seizure occurrence. Results Sixty-four children with epilepsy and 40 controls 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 controls. 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 (Î˛ = -0.30, p = 0.03). Other variables associated with the acute stress response were the number of antiepileptic drugs (Î˛ = -0.27, p = 0.05) and sleep quality (Î˛ = 0.30, p = 0.03). 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 unraveling 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.

Stress response in childhood epilepsy

INTRODUCTION The majority of patients with epilepsy report that seizures are sometimes precipitated by specific factors, among which stress is the most frequent, reported by approximately half of children as well as adults (Frucht et al.,2000; Spector et al.,2000; Nakken et al., 2005; Sperling et al., 2008; van Campen et al., 2012). These self-report data are confirmed by diary-based studies in adults (Temkin and Davis, 1984; Blanchet and Frommer, 1986; Webster and Mawer, 1989; Neugebauer et al., 1994; Swinkels et al., 1998; Klein and van Passel, 2005; Haut et al., 2007). Despite the high prevalence of self-reported stress sensitivity of seizures, many believe this reflects the subjectively biased recall of patients who try to get a grip on their unpredictable seizures, rather than a biological process, as the underlying mechanisms are unknown (Spatt et al., 1998; Sperling et al., 2008; Rajna et al., 2008). The effects of stress on epilepsy might well be mediated by stress hormones. Exposure to a stressful event activates the sympathetic nervous system as well as the HPA-axis. Activation of the sympathetic nervous system initiates a fight-or-flight response through the quick and short-lasting release of monoamines such as the catecholamines adrenaline and noradrenaline. This results for example in increases in heart rate, blood pressure and vigilance. The slower activation of the HPA-axis causes a more long-lasting release of the stress hormone cortisol, as well as other stress mediators such as CRH, ACTH and neurosteroids. Both catecholamines and cortisol induce changes in the brain and other parts of the body that help the individual to cope with the stressor, while cortisol also has a role in restoring body homeostasis afterwards (Ulrich-Lai and Herman, 2009). In animal studies, noradrenaline, and cortisol, among other stress mediators like CRH, have been shown to affect neuronal functioning, including neuronal excitability (Weinshenker and Szot, 2002; JoĂŤls, 2009; Sawyer and Escayg, 2010; Moseley et al., 2013). Therefore, changes in stress hormone regulation in patients with epilepsy might mediate stress sensitivity of seizures. Previous studies have shown altered baseline stress hormone levels in patients with epilepsy, although the direction of the effect differs between studies (Gallagher et al., 1984, Nalin et al., 1985, Facchineti et al., 1985, Gallagher, 1987, Rao et al., 1989; Baram et al., 1992, 1995, Heiskala et al., 1997, Nagamitsu et al., 2001, Wang et al, 2001, Zobel et al., 2004; Galimberti et al. 2005, Tuveri et al., 2008, Marek et al.,2010, Hill et al., 2011), and a recent pilot study suggested an increased hormonal, and decreased functional, stress response in adults with left temporal lobe epilepsy (Allendorfer et al., 2014). Hormone levels in response to stress have not been studied thoroughly, let alone their relation with stress sensitivity of seizures. In order to increase our understanding of the pathophysiological mechanisms underlying stress sensitivity of seizures in childhood epilepsy, we assessed the sympathetic and HPAaxis response to acute stress in children with epilepsy with and without stress-sensitive seizures, and in healthy controls. We hypothesized that stress sensitivity of seizures relates to alterations in the stress response, which could affect neuronal excitatory transmission and hence trigger seizures.

PART TWO | CHAPTER 6

METHODS Subjects Children aged 6 to 17 years with active epilepsy (i.e., a definitive clinical diagnosis of epilepsy and having had at least one seizure per year, including one in the past 6 months) were recruited from the pediatric neurology departments of four hospitals and one epilepsy center in the Netherlands (listed in the acknowledgements), and through the website of the Dutch Epilepsy Association. Children with enduring seizure freedom after anti-epileptic drug (AED) treatment or epilepsy surgery, and those with chronic (psychiatric) comorbidities (with the exception of attention deficit hyperactivity disorder [ADHD]) or usage of stress hormone medication or oral contraceptives, were excluded, as were patients with intellectual disabilities (intelligence quotient [IQ] < 70), a mean seizure frequency of more than 20 seizures per day, and children whose parents reported that a status epilepticus had occurred in response to acute stress. Healthy controles with a similar distribution in age, sex and intelligence were recruited through the included children with epilepsy, as well as through schools in the neighborhood of the test center. Controls were screened on the same exclusion criteria. During the process of inclusion, age, sex and gender of included children with epilepsy and healthy controls was regularly compared and recruitment of healthy controls was adjusted to obtain groups that were comparable on these characteristics. The study protocol was approved by the institutional ethical review board. All legal guardians and children older than 11 years of age provided written informed consent. General characteristics Information on intelligence or school level, comorbidities and use of medication was collected by a phone interview with parents or caregivers. Parents were asked whether an official IQ test had been performed on their child and if so, to provide a copy of the test results. In children in whom no official IQ tests were performed, this was estimated based on level of schooling, i.e., using the mean IQ of children following the same level of schooling in the Netherlands (van Dijk and Tellegen, 2004). Epilepsy characteristics were obtained from medical records and classified according to the terminology proposed by the International League Against Epilepsy (Berg et al., 2010). To evaluate symptoms of undiagnosed psychopathology, both parents completed the Child Behavior Checklist (Verhulst et al., 1996). Scores of both parents were averaged and the percentage of children scoring in the pathological range of the domain-specific syndrome scales was compared between groups. The total number of experienced negative life events was assessed using the validated Dutch Questionnaire of Life Events for children and adolescents (Veerman et al., 2003). Definition of stress sensitivity of seizures Stress sensitivity of seizures was determined by two different measures, a retrospective questionnaire and a prospective diary.

Stress response in childhood epilepsy

Questionnaire Children with epilepsy and their parents together completed a questionnaire previously designed to obtain information on the relation between stress and epileptic seizures (van Campen et al., 2012). Stress sensitivity of seizures for acute stress was defined as the subjective perception that seizures could be precipitated by acute stress (defined as stress lasting minutes to several hours). This dichotomous measure was used for further analysis. In addition, stress sensitivity for periods of stress was defined as an increase in seizure frequency during positive or negative periods of stress (defined as stress lasting for days to weeks). Overall stress sensitivity of seizures was defined as a positive answer to one or both questions. Diary Children with epilepsy and controls were asked to complete a diary for six weeks in which they reported on the occurrence of epileptic seizures and the experience of stress on a daily basis. Parents were allowed to help if needed. For all epileptic seizures, the time of onset, seizure semiology and subjective seizureprovoking factors were reported. Also, all acute stressful events (stress lasting minutes to several hours) that children experienced were reported during the day. Children and parents were provided with a list of possible stressful events based on the Kiddie-Life Events and Difficulties Schedule (K-LEDS, Monck and Dobbs, 1985; Hillegers et al., 2004), and asked to report all events on that list (whether experienced as stressful or not), as well as all stressful events that were not on the list, excluding stress caused by seizure occurrence itself (as otherwise the increased seizure risk after a first seizure would bias the effects of stress on seizure occurrence). For every event, participants reported the nature, starting time, duration and level of experienced stress (reported on a visual analogue scale [VAS] ranging from “no stress” [0] to “most severe stress I have ever experienced” [10]). Reported events with a VAS score ≥5 were regarded acute stressors and used for analysis. Children also reported the number and nature of stressful periods (stress lasting days to weeks), including stress caused by an event that happened in the recent past or was expected to happen in the near future. Furthermore, at the end of the day, children provided an overall daily stress score (VAS ranging from “no stress” [0] to “most severe stress I have ever experienced” [10]). To assess the influence of possible confounders, daily diary reports also included sleep duration, sleep quality (VAS score), use of medication (drug, dosage), medication noncompliance (time), fever (yes/no), alcohol consumption (yes/no) and period in the menstrual cycle (menstrual period yes/no, recoded into peri-menstrual, mid-follicular, peri-ovulatory or mid-luteal). Children and parents reported on the amount of parental help in completing the diary. They also reported which of the parents helped in diary completion. To correct for possible influences of parental stress evaluation, both parents completed a validated questionnaire on coping (Utrecht Coping List, Schreurs et al., 1993) and personality traits (Dutch NEO Five Factor Inventory, Hoekstra et al., 1996). To assess stress sensitivity of seizures for acute stress, the relation between the number of acute stressful events and the number of seizures per day was determined for every child

PART TWO | CHAPTER 6

that reported one or more seizures in the diary. Stress sensitivity of seizures for acute stress was defined as a positive association between acute stressors and seizures after correction for possible confounders (exponential beta coefficient (β) > 1.1), while the absence of stress-sensitive seizures was defined as a negative association (β < 0.9). For more information on the analysis, see Statistics. This presence or absence of stress-sensitive seizures for acute stress was used as the dichotomous outcome measure for further analysis. Also, the relation between the number of periods of stress or the daily stress scores versus the number of seizures per day was analyzed for descriptive purposes using the same cut-off values. Additionally, the time between start of acute stress and the first seizure following stress was analyzed. Seizure occurrence within the first two hours after stress exposure was compared between children with and without stress-sensitive seizures, as many effects of stress hormones on neuronal excitability occur in this time window (de Kloet et al., 2005). Data of children completing less than two diary weeks was discarded for diary analysis.

Experimental stress manipulation Stress manipulation The acute stress response was assessed with the Trier Social Stress Test for Children (TSST-C) (Buske-Kirschbaum et al., 1997). The total test paradigm, with a duration of approximately 3.5 hours, was always performed between 1.00 and 5.00 PM, to mitigate the effect of circadian variations in cortisol level. In preparation, children received a list of items not to eat or drink from 11.00 AM onwards, containing foods and beverages that might interfere with blood pressure regulation or noradrenaline reactivity, and parents were asked to limit the amount of fluids their children consumed at lunchtime to a maximum of 250 ml to avoid interruption of the test paradigm for toilet visits. Using a clinical interview, information was assessed on current epilepsy status and potential confounders of cortisol regulation, namely medication use, pubertal stage (Dutch translation of the Pubertal Development Scale, Petersen et al., 1988), sleep duration and sleep quality in the previous night (Groninger Sleep Quality Scale, Meijman et al., 1990), where the sleep quality score (0-15) was recoded into ‘good’ (13-15), ‘intermediate’ (9-12) or ‘poor’ (0-8). Weight, length and body mass index (BMI) were determined. The TSST-C is a standardized stress-provoking procedure that has been shown to effectively induce a multidimensional stress response (Dickerson and Kemeny, 2004; Campbell and Ehlert, 2012). Children 6 to 12 years of age were exposed to the original TSST-C protocol (Buske-Kirschbaum et al, 1997), while children 13 to 17 years of age were exposed to the slightly modified version (Gunnar et al., 2009), to provoke a similar stress response. In brief, the TSST-C consists of a 30-minute relaxation period in front of a video with neutral content, a 5-minute preparation period, a 5-minute public speech task and a 5-minute age-appropriate mental arithmetic task. The speech and arithmetic tasks were performed in front of a highly visible video camera and a live non-responsive audience consisting of a male and a female, wearing a white lab coat, who were introduced as behavioral experts. After this 10-minute stress period, there was a 5-minute period during which the child was praised for its excellent performance, followed by a video with neutral content.

100

Stress response in childhood epilepsy

Figure 1. Experimental timeline Timeline of the testing day, time respective to start of the Trier Social Stress Test for Children (stress). p preparation. Saliva samples were not collected at t= -55.

Children remained seated during the whole experiment to avoid effects on blood pressure, heart rate, and the risk of vasovagal syncope. At predefined time points (t) during the TSST-C (t= -35, -5, 0, 10, 20, 30, 40, 60, 120 in minutes, where t=0 reflects the start of the TSST-C, see Figure 1) saliva was collected with a Salivette速 (Starstedt, Etten-Leur, the Netherlands) sterile cotton swab on which children were instructed to chew gently for two minutes. At the same time points plus t= -55, blood pressure was measured with a fully automated blood pressure monitor (Omron 705IT) and children were asked for a subjective VAS stress score. The protocol also included a number of memory tasks that were outside the scope of the current research question. Heart rate and heart rate variability were measured continuously with a portable transmission device (Polar RS800CX). Interbeat intervals were imported to Kubios HRV Package version 2.0 (Tarvainen et al., 2009). Data were visually inspected and artefacts were corrected using medium level artefact correction. Heart rate variability was analysed in 5-minute intervals ending at the time points described above. Overall heart rate variability was estimated by the RMSSD (root mean square of successive differences, Thayer et al., 2012). Physiological responses to acute stress and VAS scores were compared between children with stress-sensitive seizures, children without stress-sensitive seizures and healthy controls. The analyses were performed separately for questionnaire- and diary-based determination of stress sensitivity. In addition, associations were assessed between patient/epilepsy characteristics and the magnitude of the cortisol response (a measure of HPA-axis activation) or alpha amylase response (generally considered a measure of sympathetic activity [Granger et al., 2007; Nater and Rohleder, 2009; Schumacher et al., 2013]).

Diurnal cortisol variability Throughout the day, cortisol is released in a circadian pattern, with the highest levels 30 minutes after awakening, and a gradual decrease throughout the day. To assess diurnal cortisol variability, participants were instructed to collect saliva samples at home 30 minutes after awakening and in the evening immediately before going to bed (corresponding to the diurnal cortisol peak and trough respectively) using the Salivette速 device. As cortisol is released in ultradian pulses with amplitudes varying with the circadian pattern (Lightman et al., 2008), morning samples were collected on three consecutive days, of which the sample with highest value was included in further analysis, to increase reliability of the morningpeak measurement. Children were instructed not to brush their teeth, eat or drink in the

101

PART TWO | CHAPTER 6

30 minutes before and during sampling, and to store saliva samples in the home freezer until returning them by mail. Morning and evening cortisol levels were compared between children with and without stress-sensitive seizures and healthy controls.

Analysis of cortisol and alpha-amylase Salivettes® collected during the TSST-C and at home were centrifuged for 10 minutes at 3000 rpm. Saliva was stored at -20°C until further analysis. Cortisol levels were measured without extraction with an in house competitive radio-immunoassay using a polyclonal anticortisol-antibody (K7348) and [1,2- 3H(N)]-hydrocortisone (PerkinElmer NET396250UC) tracer. Alpha-amylase levels were determined on a Beckman-Coulter AU5811 chemistry analyzer (Beckman-Coulter Inc., Brea, CA) after 1000x dilution with 0.2% bovine serum albumin in 0.01 M phosphate buffer pH 7.0 (van den Bos et al., 2013). All samples of the child were analyzed in the same batch, all batches consisted of samples from children of different groups. Statistics Group comparisons were performed with Fisher’s exact tests for binominal data, Pearson chi square tests for non-dichotomous nominal data, one-way independent ANOVA for continuous data with a normal distribution and similar variance in all groups, and a Kruskall-Wallis test for continuous data not meeting these assumptions. Correlation between binominal data was assessed with Phi correlation. Diary analysis For all patients reporting seizures in the diary, the relation between stress and seizure occurrence was analyzed per patient with a maximum-likelihood Poisson regression analysis, where the log of the number of seizures per day was assumed to be independently distributed as a Poisson variable (for more details, see Neugebauer et al., 1994). Subsequently, a multivariable Poisson regression analysis was performed, adjusting for possible confounders including sleep (sleep quality multiplied by the duration of sleep in the preceding night), AED noncompliance, alcohol intake, fever, and menstrual stage (the last four were not applicable in most children). The β of the multivariable Poisson analysis for acute stress, representing the strength of the association between acute stress and seizures regardless of statistical significance, was used as a measure for acute stress sensitivity of seizures (Neugebauer et al., 1994). The time between the start of acute stress and the occurrence of the first epileptic seizure thereafter was analyzed with survival statistics. Data of all patients were combined. Patients entered the model at the start of every stressor that was followed by an epileptic seizure without the interference of a second stressor. Time-to-event, i.e., time between start of the acute stressor and seizure occurrence, was analyzed with Kaplan-Meier statistics (Kaplan, 1958).

102

Stress response in childhood epilepsy

Experimental stress manipulation Differences in the stress response were compared between controls and children with epilepsy with and without acute stress sensitivity of seizures. Groups were compared using a repeated measures ANOVA with the within-subject factor ‘time’ and between-subjects factor ‘group’. Age-dependent measures (cortisol, alpha-amylase, blood pressure, heart rate and heart rate variability) were adjusted for age by including age as a covariate in the analysis. When the assumption of sphericity was violated, Greenhouse-Geisser corrections were used. Significant group differences were followed up by analysis of group differences per time point with a one-way ANOVA. Associations between patient characteristics and the magnitude of the stress response were examined with linear regression analysis. In order to obtain a single value reflecting the cortisol and the alpha-amylase response per child to use as determinant of these regression analysis, an area under the curve index with respect to increase (AUCi) was calculated (Grice and Jackson, 2004) out of the nine measurements between baseline (t= -5) and the end of the stress protocol (t = 120). Patient characteristics with a univariable association of p < 0.10 were included as candidates into a multivariable regression model to maximize sensitivity. Included variables were checked for multicollinearity using VIF (variance inflation factor) and tolerance, where a VIF ≥ 5 and tolerance ≤ 0.10 were considered to indicate a multicollinearity problem. Independent variables were removed by a backward stepwise selection procedure. Variables that were significant at the 0.05 level were retained in the multivariable models. Normality was evaluated with Q-Q (quantile-quantile) plots. Variance was evaluated with box plots. Data showing a skewed distribution were transformed to meet assumptions of normality when possible and further analysis was performed on the transformed data. Analysis was performed on complete cases and interpreted with a two-tailed alpha of 0.05. Posthoc analyses were Bonferroni corrected for multiple comparison. Data processing and statistical analysis were performed using SPSS (Statistical Packages for Social Sciences; version 20.0).

RESULTS Subject characteristics Sixty-four children with epilepsy and 41 controls were included in the study (see flow chart in Figure 2). One control was retrospectively excluded because of diagnosis of Treacher Collins syndrome, explaining highly aberrant cortisol and alpha-amylase values (> 3 standard deviations [SD] from the group mean at every time point). Children with epilepsy and controls reported the same number of previously experienced negative life events and none of the included children scored in the clinical range of psychopathology symptoms (Table 1).

103

PART TWO | CHAPTER 6

Table 1. Demographic and epilepsy characteristics of the total study group characteristics general sex, % male

control (n = 40)

epilepsy (n = 64)

p-value

45%

41%

0.68

11.6 (3.0)

11.6 (3.3)

1.00

100 (79-129)

99 (70-129)

0.14

intelligence (median, range)

0.53

attention deficit hyperactivity disorder, %

4.7 (4.2)

5.6 (4.1)

0.28

age at onset, year (mean, SD)

5.6 (3.8)

epilepsy duration, year (median, range)

5.5 (0.4-16.5)

seizure frequency (median, range)

1/wk (1/yr-17.5/d)

- seizure type, %

- focal

- generalized

- focal + generalized

- Rolandic epilepsy

10.9

- childhood absence epilepsy

7.8

- juvenile absence epilepsy

1.6

- GEFS+

1.6

- Panayiotopoulos syndrome

1.6

- none

73.4

- genetic

- structural

- metabolic

- unknown

- none

- anti-epileptic drugs

- anti-epileptic drugs+ketogenic diet

age, years (mean, SD)

psychopathology symptoms, % in clinical range experienced life events (mean, SD) epilepsy

localisation, % focal

electroclinical syndrome, %

etiology, %

epilepsy treatment, %

Demographic and epilepsy characteristics of study participants. Intelligence intelligence quotient or developmental quotient; SD standard deviation; yr year; d day; GEFS+ generalized epilepsy with febrile seizures plus; â&#x20AC;&#x201C; not applicable.

104

Stress response in childhood epilepsy

Figure 2. Flow-chart Flow chart of the children included in this study. Children with epilepsy completed a questionnaire on stress sensitivity of seizures, followed by six-week diary on stress and seizures. Controls completed a similar diary on stress only. In children with epilepsy, acute stress sensitivity of seizures was based on (1) self-report of acute stress-precipitated seizures in a retrospective questionnaire (yes versus no), and (2) association between daily acute stress and seizures in the diary (Î˛, exponential beta coefficient of the multivariable analysis, representing the strength of the association between acute stress and seizures, >1.1 considered stresssensitive, <0.9 considered not stress-sensitive). Afterwards, the acute stress response was measured and compared between children with epilepsy with and without acute stress-sensitive seizures (for questionnaireand diary-based determination of stress sensitivity) and controls.

105

PART TWO | CHAPTER 6

Figure 3. Stress sensitivity of seizures for acute stress Percentage of patients with stress sensitivity of seizures for acute stress based on self-report (questionnaire-based) and a positive association between acute stressors and seizures reported in the diary (diary-based) in all included patients (n = 64); patients reporting one or more seizures in the diary with an adequate distribution of reported seizures and stressors for statistical analysis (n = 37); and patients reporting six or more seizures in the diary (n = 27).

Experienced stress and stress sensitivity of seizures Experienced stress Diaries of 63 children with epilepsy and 38 controls were completed for over two weeks and included in the analysis. Out of these children, 79% completed the total six weeks. The median amount of missing data was 2.9% per child. The median number of reported acute stressful events per child was 0.6 (range 0.0 - 2.0) per day, for periods of stress this was 0.5 (range 0.0 - 2.0) per day and the mean daily stress score was 5.4 (SD 1.4). The diary was completed with parental help in 72% of children. Parental neuroticism and coping style did not relate to the reported number of acute stressors, periods of stress, nor to daily stress scores. Neither the reported stress, nor the amount of parental help significantly differed between children with epilepsy and controls. Stress sensitivity of seizures Based on the questionnaire, 61% of all children with epilepsy reported stress-precipitated seizures: 49% reported seizures precipitated by acute stress, 42% an increased seizure frequency during periods of stress, with an overlap in 30% of children. Only 44 children (70%) reported epileptic seizures during the diary period, the total number of seizures in these children ranged from one to 212 per child. The following seizure precipitating factors were self-reported for seizures experienced during the diary period: stress (32% of children), fatigue (30%), waking up (23%), sleep deprivation (18%), physical exercise (16%), watching television (16%), computer games (14%), startle (14%), light flashing (7%), sound/noise (5%), heat (5%), temperature change (5%), humidity (2%), reading (2%) and fever (2%). Questionnaire and diary self-reports of stress as a seizure-precipitant were significantly correlated (Ď&#x2020; = 0.44, p = 0.003), as of all children prospectively reporting stress as a seizureprecipitant (n = 14), only one did not report this retrospectively.

106

Stress response in childhood epilepsy

Figure 4. Time duration between acute stress and seizures in diary reports A. Number of reported seizures per time interval after reported start of an acute stressor for 15 minute time intervals. Top: all reported seizures within 24 hours after start of an acute stressor. Middle: only seizures reported by children with an overall positive association between acute stress and seizures reported in diary. Bottom: only seizures reported by children with an overall negative association between acute stress and seizures reported in diary. B. Kaplan-Meier curve of the time-to-event (time to seizure), time starts at the reported start of an acute stressor. Top: all seizures reported after start of an acute stressor. Bottom: all seizures reported within two hours of start of an acute stressor by children with an overall positive versus negative association between acute stress and seizures reported in diary.

We next determined the association between stress and the occurrence of seizures during the diary period in the 44 children reporting one or more seizures in the diary. We focused on the results for acute stress, as diary-based acute stress sensitivity of seizures is linked to the acute experimental stress response in the next section. The association between acute stress and seizures could not be calculated in seven children because of rarity of the studied events. A positive association between acute stress and seizures (Î˛ > 1.1) in multivariable

107

PART TWO | CHAPTER 6

analysis was found in 62% of children, while a negative association (β < 0.9) was found in 22%, and a neutral association (0.9 ≤ β ≤ 1.1) in 16%. Restriction of the analysis to only children reporting six or more seizures during their diary period (n = 27) yielded similar results (Figure 3). The percentages of children with a positive association between acute stress and seizures in the diary were comparable to the percentages of self-reported stress sensitivity on the questionnaire in this subgroup of patients. However, questionnaire- and diary association-based determination of acute stress sensitivity of seizures resulted in a different selection of children (correlation φ = -0.05, p = 0.78). Considering the other possible seizure precipitants that were included in the multivariable model as potential confounders, a relation between seizure occurrence and sleep was observed in only 5% of children. Other potential confounders (i.e. AED noncompliance, alcohol intake, fever and menstrual stage) were only reported in three to seven children and if so, on a limited number of days (besides menstrual stage). Therefore, the associations between these variables and seizure occurrence could not be examined reliably on a group level. For results on periods of stress and daily stress scores, see supplementary Figure 1. Of the 983 seizures reported in the diaries, 333 occurred after an acute stressor without interference of a second stressor or seizure, and these were included for time analysis. Time interval between start of acute stress and the next reported seizure ranged between one minute and 10.8 days, with 50% of seizures occurring in the first 6.6 hours (Figure 4). Of all seizures occurring within two hours after start of stressor, 50% occurred within the first 32 minutes. In children with acute stress-sensitive seizures (based on the diary association), seizure occurrence was increased directly after stress exposure and showed a gradual decline during the first hours, while seizure occurrence in children without stress-sensitive seizures was more evenly distributed throughout the day.

Figure 5. Physiological stress response in children with and without acute stress sensitivity of seizures Physiological stress response measures in children with and without acute stress-sensitive seizures and healthy controls. Stress sensitivity of seizures for acute stress was based on self-reported presence versus absence of seizure-precipitation by acute stress on a questionnaire (A-H) as well as a positive versus negative association between diary reports of acute stressors and seizures (E-H). A/E Cortisol; B/F Alpha-amylase; C/G Heart rate; D/H Heart rate variability (root mean square of successive differences). * + ^ Group differences per time-point between children with compared with those without stress-sensitive seizures (+), children with stress-sensitive seizures compared with controls (*), or children without stress-sensitive seizures compared with controls (^) p < 0.05 after correction for multiple comparison. Data are presented as mean ± standard error of the mean.

108

Stress response in childhood epilepsy

109

110 0.31 0.03 0.05 0.05 0.12 0.01 0.09

1.16 2.18 2.08 1.92 1.62 2.50 1.77

self-report

diary

self-report

diary

self-report

diary

0.63

0.83

0.11

0.09

diary

1.51

0.05

0.06

self-report

1.54

diary

1.76

dairy

self-report

0.84

self-report

4.40

diary

1.20

4.04

1.72

1.55

0.00

0.23

1.07

0.68

0.93

0.10

0.31

0.02

0.19

0.22

1.00

0.80

0.35

0.51

0.40

0.90

0.50

0.05

3.13 0.70

0.04

0.003

p-value

3.49

6.29

F-value

group

0.002

0.05 28.66

<0.001

0.02

0.006

0.008

0.34

0.25

0.36

0.60

0.10

0.17

<0.001

2.40 97.56

posthoc analysis of group differences per timepoint

p-value stress-sens vs. stress-sens vs. not stressnot stresscontrols sens vs. sens controls

2.68

3.90

3.55

1.14

1.27

1.10

0.79

1.79

1.53

8.56

5.54

F-value

time

stress-sens children with acute stress sensitivity of seizures; not stress-sens children without acute stress sensitivity of seizures. + significant group differences in posthoc analysis on one or more time points; - no significant group differences in posthoc analysis on one or more time points; bold p < 0.05.

subjective stress scores

heart rate variability

heart rate

diastolic blood pressure

systolic blood pressure

alpha-amylase

0.02 0.007

2.74

self-report

cortisol

p-value

group Ă&#x2014; time interaction F-value

stress sensitivity measure

stress response measure

Table 2. Group differences in stress response: statistics

PART TWO | CHAPTER 6

Stress response in childhood epilepsy

Experimental stress response After diary completion, 10 children withdrew from the study, leaving 56 children with epilepsy and 38 controls to complete the TSST-C. Baseline characteristics of the remaining children were similar to the total study cohort (supplementary Table 1). Children with epilepsy did not differ from controls regarding pubertal stage, the number of hours of sleep or sleep quality in the night previous to the stress test, but had a higher body mass index (mean 20.1 versus 18.5 kg/m2, p = 0.04). Six children experienced one or more seizures during the testing day, these were all focal seizures or absences. Five of these children were known to have a seizure frequency of multiple seizures per day. In total, six seizures occurred before stress induction, five seizures between 15 and 55 minutes after stressinduction and two at the end of the testing protocol. Physiological stress measures did not differ between children with and children without seizures during the testing day. Alpha-amylase results of one control were excluded because of very low levels (< 3 SD from the group mean at every time point) with unknown cause. Due to technical problems heart rate measurements could not be reliable determined in some children, and the overall analysis was confined to only 43 children with epilepsy and 29 controls. The response to acute psychosocial stress was significantly different between patients with and those without acute stress-sensitive seizures, for both questionnaire- and diary-based determination of stress sensitivity (Figure 5, Table 2). Based on questionnaire results, children with acute stress-sensitive seizures showed a significantly lower cortisol response to experimental stress compared with patients without stress-sensitive seizures and controls, while no differences in baseline levels (i.e., after relaxing) were observed (Figure 5A). Overall alpha-amylase levels, but not the alphaamylase increase in response to stress, were significantly lower in children with compared with those without acute stress-sensitive seizures (Figure 5B). The two epilepsy groups did not differ in systolic or diastolic blood pressure, heart rate or heart rate variability, although children with epilepsy showed significantly lower heart rate variability at baseline compared with controls (Figure 5C/D). Children with acute stress-sensitive seizures reported less subjective stress compared with patients without acute stress-sensitive seizures and controls. Organizing acute stress sensitivity of seizures based on diary association revealed exactly the same patterns, although variation was larger, which presumably results from the smaller group sizes (5E-H). The magnitude of the cortisol response (AUCi) was not only negatively associated with self-reported acute stress sensitivity of seizures, but also with the number of AEDs and seizure frequency, while it was positively associated with poor sleep quality in the previous night and intelligence in univariable analysis. In multivariable analysis only the association with acute stress sensitivity of seizures, sleep quality and the number of AEDs remained significant (Table 3). None of the studied variables were significantly associated with magnitude of the sympathetic stress response (i.e., alpha-amylase AUCi, see supplementary Table 2). Cortisol levels in samples collected at home during the diurnal peak and trough did not differ between groups (questionnaire-based group division: peak F(2,92) = 1.34, p = 0.27; trough F(2,92) = 1.04, p = 0.36; diary-based group division: peak F(2,61) = 1.36, p = 0.26; trough F(2,61) = 0.37, p = 0.69), see supplementary Figure 2. 111

PART TWO | CHAPTER 6

DISCUSSION This study aimed to elucidate the association between stress sensitivity of epileptic seizures and the stress response, by exposing children with or without acute stress-precipitated seizures and healthy controls to acute stress in a controlled laboratory environment. We found a relation between the stress-induced cortisol response and acute stress sensitivity of seizures, the latter both determined by a questionnaire and a prospective diary. Children with acute stress-sensitive seizures showed a markedly blunted cortisol response to the TSST-C and overall decreased alpha-amylase levels compared with epileptic children without acute stress-sensitive seizures and healthy controls. No differences were observed in baseline cortisol, measured just prior to the TSST-C as well as at home during the diurnal peak and trough, nor sympathetic parameters of the stress response. Stress sensitivity of seizures was related to cortisol response irrespective of other patient and epilepsy characteristics, such as the number of AEDs and seizure frequency. The relation between stress and seizures was previously reported in numerous questionnaire studies investigating seizure-precipitants in patients with epilepsy. According to these studies, stress was reported to precipitate seizures in 8-83% of patients (reviewed by van Campen et al., 2014). These results have been confirmed by a number of diary studies in adults, and remain after correction for possible confounders such as sleep deprivation and medication noncompliance (Neugebauer et al., 1994; Haut et al., 2007). Also stress hormone medication is used in the treatment of specific (pediatric) epilepsy syndromes, and although their anti-inflammatory effects might be one of their mechanisms of action (Hancock and Cross, 2009), animal research also shows direct effects of stress hormones on neuronal excitability (JoĂŤls et al., 2009). Although none of these studies investigated the mechanisms behind stress sensitivity of seizures, associations have been described with the number of previously experienced life events (van Campen et al., 2012), and with anxiety symptoms (Privitera et al., 2014). The relation between stress and seizures based on questionnaire and diary results that we report in children with epilepsy is in line with these previous studies in adults. Importantly, the combination with an experimental stressor provides unique insight into the underlying hormonal mechanisms. The reported association between cortisol response and sleep is in accordance with previous literature in healthy populations (RĂ¤ikkĂśnen et al., 2010). As sleep quality in the night before the TSST did not differ between subgroups, the relation between sleep and cortisol cannot explain the group differences in cortisol response. Previous studies in adult populations have shown that BMI can also influence the hormonal stress response (Dockray et al., 2009; Francis et al., 2013; Ruttle et al., 2014). In our sample, no differences in BMI existed between children with epilepsy with and those without stress-sensitive seizures. Therefore, BMI is not expected to influence our main results. Furthermore, we did not find an association between BMI and cortisol response (Table 3). The clear distinction in cortisol response after psychosocial stress in a controlled laboratory setting between children with and without stress-sensitive seizures, was observed both when children were subdivided based on retrospective self-report in a questionnaire

112

Stress response in childhood epilepsy

completed prior to entering the experiment, and based on the association between stress and seizures as determined from the analysis of prospective diaries during a period that the children actually reported seizures, even though the two methods resulted in different selections of children. These substantial differences in stress-induced cortisol levels between children with and without stress-sensitive seizures strongly support the idea that interindividual variation in stress sensitivity of seizures has a biological base.

Table 3. Determinants of cortisol response to stress in children with epilepsy cortisol (AUCi) characteristics

univariable

multivariable

Î˛

p-value

0.06

0.64

Î˛

p-value

general sex, male age, years

0.14

0.30

intelligence

0.29

0.03

ref

-0.08

0.59

0.14

0.32

ref

- intermediate (9-12)

0.07

0.61

- poor (<=8)

0.25

0.07

0.21

0.12

age at onset, year

0.18

0.19

epilepsy duration

-0.07

0.60

seizure frequency (ln)

-0.24

0.09

localisation, focal

-0.11

0.44

- genetic

0.17

0.23

- structural/metabolic

0.07

0.61

ref

-0.23

- acute - periods

puberty stage, - prepubertal - not prepubertal body mass index sleep quality night before stress test - good (13-15)

negative life events

6 0.30

0.03

0.09

-0.27

0.05

-0.28

0.04

-0.30

0.03

0.07

0.61

epilepsy-related

etiology,

- unknown anti-epileptic drugs, number self-reported stress sensitivity of seizures

ref reference group; ln log transformed to obtain normality; bold p < 0.05.

113

PART TWO | CHAPTER 6

Possible mechanisms At first glance, the decreased cortisol response to stress in children with acute stresssensitive seizures in combination with their lower subjective stress scores may suggest that these children were less stressed by the experiment. However, the effects of stress on heart rate, blood pressure and heart rate variability did not differ between children with and without acute stress-sensitive seizures, indicating that the sympathetic effect of the stressor was similar between groups. The selective decrease in cortisol reactivity to experimental stress in children with acute stress-sensitive seizures could imply a down-regulation of the HPA-axis, which might be driven by the increased experience of stressful events in the past (van Campen et al., 2012): some studies indeed report down-regulation of HPA-axis activity in association with early life stress (Faravelli et al., 2012; StrĂźber et al., 2014). This may also apply to our cohort of children with epilepsy, as the association between the number of experienced negative life events and acute stress sensitivity of seizures was replicated in the current cohort (data not shown), although the number of negative life events did not significantly relate to the magnitude of the cortisol response. To prove this, a prospective design determining stress responses prior to epilepsy onset and during the course of epileptogenesis would be necessary, which clearly is beyond the scope of the current study. The blunted cortisol response in patients with stress-sensitive seizures may also signify hyposensitivity of the adrenals to adrenocorticotropic hormone (ACTH). If so, the decreased levels of cortisol after stress â&#x20AC;&#x201C;and hence reduced capacity for negative feedbackmight be insufficient to constrain the central stress response. Increased levels of central stress hormones, such as corticotrophin-releasing hormone (CRH) and ACTH, that are known to increase excitability in rodents (Aldenhoff et al., 1983; Baram and Schultz, 1991; Hollrigel et al., 1998), might then alter the excitation-inhibition balance during a prolonged period of time. This notion is in line with previous studies showing a reduced negative feedback of the HPA-axis in patients with epilepsy by performing a dexamethasone suppression test without or with exogenous CRH stimulation respectively (Robertson et al., 1986; Zobel et al., 2004), as well as studies showing increased levels of ACTH (Gallagher et al., 1984; Gallagher, 1987; Marek et al., 2010) and CRH (Wang et al., 2001) in patients with epilepsy. Indirectly, our diary data too suggests an important role of early stress mediators, like CRH, in stress-precipitation of seizures. CRH production is not limited to the hypothalamus, but also occurs at several extrahypothalamic brain sites, including limbic and cortical regions (Nemeroff, 1996). CRH exerts proconvulsive effecs especially in the immature brain (Baram et al., 1991). Many extrahypothalamic regions exert an inhibitory control over the HPA-axis (Ulrich-Lai and Herman, 2009). The blunted cortisol response to stress might therefore reflect a indirectly suppressed HPA axis. This hyperactivity of stress hormones at the location of the seizure onset zone could be one explanation of stress sensitivity of seizures in children with a blunted cortisol response. Interestingly, a recent pilot study in adults with left temporal lobe epilepsy showed a decreased fMRI activation to stress in the temporal and medial prefrontal regions, and an increased cortisol response compared to healthy controls (Allendorfer et al., 2014). Although the self-report diary data used for time-analysis should be interpreted with caution, these suggest that seizure

114

Stress response in childhood epilepsy

occurrence after stress peaks shortly after start of the stressor, similar to the levels of noradrenaline and CRH that increase almost instantly after stress exposure and normalize within 30-60 minutes, while ACTH and cortisol peak only after 10-15 and 20-30 minutes respectively. Although central levels of CRH and noradrenaline are difficult to measure in humans, future studies could assess ACTH levels in response to stress, ACTH and cortisol levels in response to CRH injection, or study the negative feedback loop by exposing patients to a dexamethasone suppression test. Furthermore, continuous measurements of stress hormones in relation to the occurrence of electroencephalographic epileptiform activity would improve insight in the effects of endogenous stress hormones on neuronal excitability.

Study design and limitations In this study, a stress response was provoked in children with epilepsy and controls using the TSST-C. This standardized procedure has extensively been shown to effectively induce a multidimensional stress response and has great comparability to everyday stressful situations (as reviewed by Dickerson and Kemeny, 2004; Campbell and Ehlert, 2012). To be able to apply this procedure reliably, strict inclusion criteria were necessary regarding age, intelligence and comorbidities. Therefore, caution is warranted when automatically extrapolating the results to all children with epilepsy. Within these criteria, the study population was very heterogeneous with respect to epilepsy characteristics, age and pubertal status. Because our cohort was selected on age, normal intelligence and the absence of comorbidities, our results do not provide evidence for infants, children with mental retardation or with comorbidities. Moreover, given the heterogeneity of epilepsy types among the participants, we cannot exclude that some types were more vulnerable to dysregulation of the HPA-axis than others, although this notion was not supported by our preliminary analysis. Furthermore, as we were interested in the stress response in children with active epilepsy, children experienced seizures at different moments respective to the stress test and almost all children used AEDs. Especially AEDs are known to influence cortisol levels. However, we showed the effects of stress sensitivity of seizures on stress response to be independent of seizure frequency and the number of AEDs. This study focused on acute stress, both in terms of stress sensitivity of seizures and the stress response. The interaction between stress hormone regulation in epilepsy and chronic stress requires further study. To examine the effects of stress sensitivity of seizures on the acute stress response, we categorized children with epilepsy into a group of children with and a group without acute stress-sensitive seizures. As retrospective self-report might be influenced by recall bias and subjective interpretation, all children were asked to complete a six week diary on stressors and seizures. Although the prospective data collection benefits reliability of the results, diary-based stress sensitivity also has several limitations. Firstly, the broad age range of included children resulted in different degrees of parental help. However, the degree of parental help, as well as parental coping and neuroticism, were found not to influence reported stress. Secondly, the limited duration of the diary in combination with the large

115

PART TWO | CHAPTER 6

variation in seizure frequency resulted in reported seizures in only 70% of children with epilepsy. . Importantly, the children who did not report seizures in the diary cannot be regarded to be non stress-sensitive even though reported stress levels were comparable to children with seizures.. Besides stress, several other variables influence neuronal excitability, such as anti-epileptic drugs and sleep stages. Seizure occurrence is likely to be the result of the neuronal excitability, influenced by all these different variables, exceeding a certain threshold. Therefore, whether a certain level of stress precipitates a seizure might differ between different time periods within a patient and certainly between patients. As in the majority of children that did report seizures, the number of reported seizures was low, the relation between stress and seizures was determined based on the direction of the association instead of on statistical significance. Although this method has been used before (Neugebauer et al., 1994), it calls for careful interpretation of results. We cannot exclude that some children with stress-sensitive seizures did not have enough seizures in the diary period to reliably determine their p-value. Also, time interval analysis depended on the reported start of stressors and seizures, while the start of a stressor might not reflect the start of the stress response, as patients might not report the anticipated stress beforehand. Additionally, diary associations did not completely match with self-reports of stress sensitivity, suggesting that stress sensitivity is a continuous rather than a dichotomous feature. Furthermore, as only children with seizures during the diary period were included in diary analysis and in subsequent analyses using diary-based stress sensitivity, these analyses are somewhat biased towards subjects with more frequent seizures. Because of these limitations, diary results should be interpreted with some caution. Stress sensitivity of seizures was therefore also categorized based on the subjective perception of having stress-sensitive seizures as reported in retrospective questionnaires separately. Interestingly, a striking similarity was found in group differences in stress response measures between these two methods of assessing stress sensitivity of seizures, supporting the validity of our results. The lower stress hormone levels in both epilepsy groups at the start of the experiment might reflect that children with epilepsy, compared to healthy controls, are more used to going to a hospital environment, and therefore experience less anticipatory stress at the start of the experiment. This might explain why stress hormone levels at the first measurement differ from those after experimental stress, which is designed to evoke a stress response in all populations.

Conclusion and implications Epileptic seizures cause a major disease burden in children with epilepsy and greatly affect cognitive and behavioral development (Ferro et al., 2014). Although the effect of stress on seizures is reported by half of children with epilepsy (van Campen et al., 2012), this is often attributed to sleep deprivation and fatigue by their physicians. Currently, a vast amount of evidence shows that stress is an independent precipitant of seizure occurrence in adults. Our results confirm these findings in a pediatric population and, for the first time, offer a biological basis for stress sensitivity of seizures. The reported differences in stress response

116

Stress response in childhood epilepsy

between children with and those without acute stress-sensitive seizures suggest that the seizure-precipitating effects of stress are associated with dysregulation of the HPA-axis and hence â&#x20AC;&#x201C;potentiallyâ&#x20AC;&#x201C; alterations in neurotransmitter balance. Based on these results, and the knowledge that current AEDs fail to control seizures in 20-40% of patients with epilepsy (Picot et al., 2008; Laxer et al., 2014) and have substantial side-effects, patients with stresssensitive seizures might be actively advised to avoid stress or use stress reduction strategies such as meditation, yoga, relaxation therapy and biofeedback training (Rousseau et al., 1985; Dahl et al., 1987, Puskarich et al., 1992; Deepak et al., 1994; Panjwani et al., 1996; Rajesh et al., 2006; Lundgren et al., 2008, Sathyaprabha et al., 2008). Increased knowledge of the relation between stress and seizures in childhood epilepsy will benefit our understanding of the fundamental processes underlying epilepsy and ictogenesis in general, and may provide valuable clues to direct the development of new treatment regimens for epilepsy.

ACKNOWLDEGEMENTS We would like to thank T. Kliest, M.M.J. Linszen and H. ter Riet for their contributions to data collection and/or processing. We also gratefully acknowledge M. Willemsen, Radboud University Medical Center; H. Stroink, Canisius Wilhelmina Ziekenhuis; H.M. Schippers, St. Antonius Ziekenhuis; P. Augustijn and E. Hagebeuk, the Epilepsy Institutes of the Netherlands Foundation (SEIN), for their cooperation and contributions regarding patient recruitment; and E.W.G. Lentjes and I. Maitimu for salivary cortisol and alpha-amylase analysis. This work was supported by a grant of the University Medical Center Utrecht Alexandre Suerman Stipendium and Bio Research Center for Children.

117

PART TWO | CHAPTER 6

Supplement

Supplementary Figure 1. Stress sensitivity of seizures for periods of stress and daily stress scores Percentage of children with stress sensitivity of seizures for periods of stress (based on self-report of an increase in seizure-frequency during periods of stress on the questionnaire and the association between periods of stress and seizures in the diary), and daily stress scores (based on the association between daily stress scores and seizures in the diary). A. Percentages of all patients reporting one or more seizures in the diary with an adequate distribution of reported seizures and stressors for statistical analysis. For periods of stress, the relation between stress and seizures could be determined in 33 children. The association was positive in 52% of children and negative in 27%. The diary-association between periods of stress and seizures in the diary (positive versus negative) was significantly correlated with stress sensitivity for periods of stress on the questionnaire (yes versus no) (Ď&#x2020; = 0.46, p = 0.02). For daily stress scores, the relation between stress and seizures could be determined in 40 children. The association was positive in 43% of children and negative in 25%. B. Restriction of the analysis to only children reporting six or more seizures during their diary period (n = 27) resulted in similar results.

118

Stress response in childhood epilepsy

Supplementary Figure 2. Diurnal cortisol variability Cortisol concentration in saliva samples collected at home 30 minutes after awakening, during the circadian peak (morning), and before going to bed, during the circadian through (evening) in children with and without stress sensitivity of seizures for acute stress and healthy controls. Stress sensitivity of seizures for acute stress was based on: A. self-report of seizures precipitated by acute of stress on the questionnaire; and B. a positive association between acute stressors and seizures in the diary.

119

PART TWO | CHAPTER 6

Supplementary Table 1. Demographic and epilepsy characteristics of children exposed to the stress test characteristics

control (n = 37)

epilepsy (n = 56)

p-value

41%

38%

0.83

general sex, % male age, years (mean, SD)

11.6 (3.0)

11.5 (3.3)

0.82

101 (79-129)

97 (71-129)

0.14

attention deficit hyperactivity disorder, %

3.6

0.52

psychopathology symptoms, % clinical range

intelligence (median, range)

experienced life events (median, range)

4 (0-19)

5 (0-18)

0.21

puberty stage, % prepubertal

0.53

body mass index (mean, SD)

18.5 (3.1)

20.1 (4.2)

0.04

10.3 (4.3-12.0)

10.4 (7.3-12.3)

0.64

- good (13-15)

0.15

- intermediate (9-12)

- poor (<=8)

sleep in the night before the stress test - hours (median, range) - quality

epilepsy age at onset, year (mean, SD)

5.9 (4.0)

epilepsy duration, year (median, range)

5.1 (0.4-16.0)

seizure frequency (median, range)

1/wk (1/yr-7.5/d)

localisation, % focal

- genetic

- structural/metabolic

- unknown

- none

- anti-epileptic drugs

- anti-epileptic drugs+ketogenic diet

etiology, %

epilepsy treatment, %

Intelligence intelligence quotient or developmental quotient; SD standard deviation; yr year; d day; â&#x20AC;&#x201C; not applicable.

120

Stress response in childhood epilepsy

Supplementary Table 2. Determinants of alpha-amylase response to stress in children with epilepsy alpha-amylase (AUCi) characteristics

univariable

multivariable

Î˛

p-value

sex, male

0.02

0.86

age, years

0.09

0.53

intelligence

0.05

0.72

Î˛

p-value

general

- prepubertal - not prepubertal body mass index

ref

0.12

0.41

-0.03

0.80

sleep quality night before stress test - good

ref

- intermediate (9-12)

0.05

0.69

- poor (<=8)

0.002

0.99

-0.24

0.08

age at onset, year

0.24

0.07

epilepsy duration

-0.18

0.81

seizure frequency, ln

0.40

0.53

localisation, focal

-0.05

0.71

- genetic

0.03

0.85

- structural/metabolic

0.09

0.55

negative life events epilepsy-related

etiology,

- unknown

ref

-0.14

0.31

- acute

-0.24

0.08

- periods

-0.25

0.07

anti-epileptic drugs, number self-reported stress sensitivity of seizures

ref reference group; ln log transformed to obtain normality.

121

CHAPTER

Seizure occurrence and the circadian rhythm of cortisol, a systematic review Epilepsy & Behavior 2015

Jolien S. van Campen Floris A. Valentijn Floor E. Jansen Marian JoĂŤls Kees P.J. Braun

PART TWO | CHAPTER 7

ABSTRACT Introduction Stress is the seizure precipitant most often reported by patients with epilepsy or their caregivers. The relation between stress and seizures is presumably mediated by stress hormones such as cortisol, affecting neuronal excitability. Endogenous cortisol is released in a circadian pattern. To gain insight into the relation between the circadian rhythm of cortisol and seizure occurrence, we systematically reviewed studies on the diurnal distribution of epileptic seizures in children and adults and linked the results to the circadian rhythm of cortisol. Methods A structured literature search was conducted to identify relevant articles, combining the terms ‘epilepsy’ and ‘circadian seizure distribution’, plus synonyms. Articles were screened using predefined selection criteria. Data on 24-hour seizure occurrence were extracted, combined, and related to a standard circadian rhythm of cortisol. Results Fifteen relevant articles were identified of which twelve could be used for data aggregation. Overall, seizure occurrence showed a sharp rise in the early morning, followed by a gradual decline, similar to cortisol rhythmicity. The occurrence of generalized seizures and focal seizures originating from the parietal lobe in particular followed the circadian rhythm of cortisol. Conclusion The diurnal occurrence of epileptic seizures shows similarities to the circadian rhythm of cortisol. These results support the hypothesis that circadian fluctuations in stress hormone level influence the occurrence of epileptic seizures.

124

Circadian rhythms of seizures and cortisol

INTRODUCTION The majority of patients with epilepsy report seizures triggered or provoked by endogenous or exogenous factors. The seizure precipitant most often reported is stress (Frucht et al., 2000; Spector et al., 2000; Nakken et al., 2005; Haut et al., 2007; Sperling et al., 2008; van Campen et al., 2012; Wassenaar et al., 2014). The relation between stress and seizures is expected to be mediated by stress hormones like corticotrophin-releasing hormone (CRH), adrenocorticotropic hormone (ACTH) and cortisol, affecting neuronal excitability and seizure threshold (Joëls, 2009; van Campen et al., 2014). Under physiological conditions, cortisol – the ‘end product’ of the hypothalamic-pituitaryadrenal axis –, is released in a distinct circadian pattern (Krieger, 1970; Dickmeis, 2009). These circadian changes in cortisol concentration are likely to affect many important homeostatic processes in the body, including the balance between neuronal excitability and inhibition (Young et al., 2004; Lightman, 2008; Sarabdjitsingh et al., 2014). However, the relationship between circadian fluctuations in cortisol levels and seizure susceptibility is unknown. To provide a first step in exploring this possible relationship, we systematically reviewed the current literature on circadian seizure occurrence in children as well as adults, and compared this to the rhythm of cortisol release.

METHODS Search strategy To identify studies describing the full 24-hours’ seizure distribution of seizures in children as well as adults, a literature search was conducted in PubMed and EMBASE on the 30th of July 2014, combining the term ‘epilepsy’ plus synonyms with a term describing the occurrence of seizures during the day (see Supplementary Table 1). Article selection After the exclusion of double publications, titles or abstracts were screened and excluded when no English abstract was available, the study did not report on humans, epilepsy or the circadian occurrence of epileptic seizures, the study did not present original patient data, or when no full text was available. Full texts of the remaining articles were screened and excluded when additional information revealed a conflict with these criteria or seizure occurrence was reported for less than 24-hours. References of the remaining articles and articles citing the remaining articles were identified using Web of Science and screened using the same criteria. Only studies reporting results of (video-) scalp electroencephalographic (EEG) or electrocorticographic (ECoG) monitoring were included in data aggregation to avoid inaccuracy caused by reporting bias in diary studies. In case of overlapping patient populations, multiple inclusions of data from a single seizure were avoided by excluding one of the two overlapping studies, in favor of the one with the largest patient population.

125

PART TWO | CHAPTER 7

Data analysis Seizure data (i.e., time of occurrence, seizure classification and localization of the epileptogenic focus) were extracted from the selected articles. Secondary generalized seizures were classified as focal. Data was analyzed for the overall population, as well as for children and adults separately when possible, where the threshold for â&#x20AC;&#x2DC;childrenâ&#x20AC;&#x2122; corresponded with the threshold used in each individual publication. Data of different studies were combined by calculating the total number of seizures observed in a specific time window. As the time bins in which seizures were measured varied between studies, all results were recalculated into the number of seizures per hour, by dividing the number of seizures observed within a certain time bin by the number of hours in that time window (if time bins comprised two or more hours, seizure frequency in these hours was kept constant). None of the selected studies measured cortisol in study subjects. Therefore, circadian distribution of epileptic seizures could only be visually compared to a standard circadian cortisol concentration curve in humans. The circadian cortisol rhythm in children develops between one month and two years of age (Franks, 1967; Mills, 1975; Zurbrugg, 1976; Price et al., 1983; Onishi et al., 1983; Spangler, 1991; Kiess et al., 1995; Santiago et al., 1996; Van Cauter et al., 1996; Pruessner et al., 1997; Groschl et al., 2003; de Weerth et al., 2003). Although inter-individual variability in cortisol levels has been related to age or pubertal status (Jonetz-Mentzel and Wiedemann, 1993; Kiess et al., 1995; Shirtcliff et al., 2012), the overall circadian pattern is very robust (Knutsson et al., 1997; Pruessner et al., 1997). Therefore, data of children and adults was compared to the same cortisol curve, obtained from Weitzman et al. (1971).

RESULTS Initially, our search resulted in 2533 titles and abstracts, of which 15 articles fulfilled selection criteria. Reference screening identified one additional article, which did not lead to modification of the query because a term describing circadian seizure distribution in title and abstract was lacking (for a flow chart, see supplementary Figure 1). Results of the selected studies are described in Table 2. One study was excluded from data aggregation because seizure occurrence was only reported in a seizure diary (Le et al., 2011), as opposed to EEG of ECoG in the other studies. Of the 14 studies remaining, seven focused on children, one on adults, two described results for children and adults separately, and four provided results without this distinction. All studies that provided separate data for adults focused on focal seizures. For two of the studies, only data for specific generalized seizure types could be included in our analysis, because of possible overlap in patient populations with other studies of the same research group (Zarowski et al., 2011; Sanchez Fernandez et al., 2013). Epileptic seizures were monitored by scalp EEG in 12 studies and intracranial electrocorticography in two. The number of recorded seizures per study varied between 80 and 1350, within a period ranging

126

Circadian rhythms of seizures and cortisol

from 24 hours to 16 days. Information on the number of seizures per patient was only reported in eight of the original articles. Combined data resulted in a total of 5700 seizures, of which 2074 could specifically be attributed to children, 1393 to adults, and the other 2233 were reported in studies not distinguishing between adults and children.

Circadian seizure distribution Circadian seizure distribution varied between studies (Table 1). Most of the original studies described an increased seizure occurrence in the (early) morning compared with the rest of the day, both in epilepsy populations of all ages (Karafin et al., 2010; Le et al., 2011; Pavlova et al., 2012), and in cohorts including children (Kaleyias et al., 2011; Loddenkemper et al., 2011; Zarowski et al., 2011; Ramgopal et al., 2012; Ramgopal et al., 2012; Sanchez Fernandez et al., 2013) or adults separately (Durazzo et al., 2008). Many also reported a peak at varying times in the afternoon or the evening (Plouin et al., 1987; Quigg et al., 1998; Pavlova et al., 2004; Durazzo et al., 2008; Hofstra et al., 2009; Hofstra et al., 2009; Karafin et al., 2010; Kaleyias et al., 2011; Loddenkemper et al., 2011; Zarowski et al., 2011; Ramgopal et al., 2012; Pavlova et al., 2012; Sanchez Fernandez et al., 2013). The aggregated seizure data showed a steep increase in seizure occurrence in the early morning, starting around 4h (military time), following the rise in plasma cortisol with a time lag of roughly one to two hours (see Figure 1). The seizure occurrence reached a plateau level around 6h, approximately the moment in time when plasma cortisol showed its awakening response. During the next three hours, seizure occurrence and cortisol both stayed high. During the day, additional peaks were observed in seizure occurrence (11-12h and 15-17h), that are not observed in cortisol level. At the end of the day, both seizure occurrence and cortisol levels dropped to reach a nightly quiescence. Few differences were observed in seizure occurrence between children and adults (Figure 2).

7 Figure 1. Circadian seizure distribution and cortisol rhythmicity Combined seizure occurrence over a 24-hoursâ&#x20AC;&#x2122; period, displayed on top of a standard curve of the plasma cortisol concentration (adapted from Weitzman et al., 1971) to enable visual comparison. time military time (0-24 hours); total seizure data for children, adults and unspecified age groups; h hours.

127

128

90 / 26

1350 / 96

Pavlova et al. (2004)

Quigg et al. (1998)

866 / 215

219 / 51

223 / 71

259 / 66

Sanchez Fernandez et al. (2013)2

Ramgopal et al. (2012a)

Ramgopal et al. (2012b)

Kaleyias et al. (2011)

children

694 / 60

25720 / 1877

129 / 44

Karafin et al. (2010)

Le et al. (2011)1

Pavlova et al. (2012)

general population

n (seizures / patients)

1-77

1-13

10.5±7.9

1-8

seizures per patient

2-18

0-21

30.9±10.2

35.5±11

35.9±9.2

29.9±16.0

1-92

age (years)

Table 1. Studies describing 24 hours’ seizure occurrence

focal

(secondary) generalized tonic-clonic

epileptic spasms

seizures with multiple semiology phases

focal

all

type of seizures

focal (lesional)

all

focal

focal (TLE vs. XTLE)

MTLE

all

localization

EEG: 3-10d

EEG: 1-10d

EEG: 1-30d

EEG: ≥24h

EEG: 3-14d

EEG: ≥24h

EEG: 2-16d

seizure diary

EEG: 1-3d

method of seizure monitoring

time bin (h)

Temporal seizures peaked at 9-12h and 15-18h, extratemporal seizures at 6-9h.

Peaks for focal seizures with tonic-clonic evolution at 00–03h and 6–9h. Primarily generalized tonic-clonic seizures peaked at 9-12h.

Seizures peaked at 9–12h and 15–18h. For children <3 years peaks at 9– 12h and 15–18h, for children >3 years a peak at 6-9h and a non-significant peak at 15–18h.

Clonic seizures peaked at 0–3h and 6–9h; tonic seizures at 21-12h.

Random seizure patterns in LTLE and XTLE. Peak at 15h for MTLE.

Differences in the patterns between TLE and XTLE, with significant peaks for TLE at 15-19h; and for XTLE at 19-23h.

Bimodal pattern with peaks at 6-8h and 15-17h.

Significantly different distribution in timing of focal and generalized seizures. Generalized seizures mainly occurred in the morning, (34% 5-11h), whereas focal seizures were fairly evenly distributed throughout the day and evening.

Frontal seizures occurred mainly at 05:15–07:30h, temporal lobe seizures at 18:45–23:56h.

time of reported peaks in seizure occurrence

PART TWO | CHAPTER 7

138 / 7

80 / 16

Hofstra et al. (2009b)

Plouin et al. (1987)

412 / 100

312 / 26

Hofstra et al. (2009a)

Hofstra et al. (2009b)

-3

2-8

-3

1-13

1-12

generalized

all

focal focal

infantile spasms

focal

focal + secondary generalized focal

0-21

1-15 5-15

- (infants)

29.6±10.9

16-65 17-45

focal

focal (TLE vs. XTLE)

focal

all

focal

focal (TLE vs. XTLE)

all

generalized

ECoG: 3-9d

EEG: 1-7d

ECoG: 9-10d

EEG: 24h

ECoG: 2-8d

EEG: 22h - 7d

EEG: 1-10d

EEG: 1-8d

Temporal seizures peaked at 11–17h, frontal seizures at 23-05h and parietal seizures at 17-23h.

Overall seizures, temporal seizures peaked at daytime (11–17h).

Frontal and parietal lobe seizures peaked at 4-7h, occipital seizures at 16-19h, mesial temporal seizures at 7-10h and 16-19h, neocortical temporal seizures at 13-16h.

Seizure clusters were most frequently observed at 14-20h, with a peak at 19-20h. Isolated spasms were most frequent at 18-24h with a peak at 20-22h.

Only frontal seizures could be analyzed reliably. Fewer frontal seizures occurred between 5–11h compared with the rest of the day.

Overall seizures and extratemporal seizures peaked at daytime (11-17h).

Generalized seizures peaked at 6–12h, temporal seizures at 21–9h, frontal seizures at 0–6h, parietal seizures at 6–9h, and occipital seizures at 9–12h and 15–18h.

Peaks for clonic seizures at 6-9h and 12-15h, absence seizures at 9-12h and 18-0h, atonic seizures at 12-18h and epileptic spasms at 6-9h and 15-18h.

Excluded from data aggregation because seizure occurrence was measured by seizure diary only; ²Excluded from overall data aggregation because of possible overlap in patient populations; 3Maximum of 15 seizures per patient per time bin; n number; EEG scalp electroencephalography; ECoG intracranial electrocorticography; h hours (time windows are displayed in military time [0-24h]); d days; TLE temporal lobe epilepsy; XTLE extratemporal lobe epilepsy; MTLE mesiotemporal lobe epilepsy; general population no distinction between results for children and adults; - not specified.

669 / 131

Durazzo et al. (2008)

adults

396 / 76

1008 / 225

316 / 77

Hofstra et al. (2009a)

Loddenkemper et al. (2011)

Zarowski et al. (2011)2

Circadian rhythms of seizures and cortisol

129

PART TWO | CHAPTER 7

Seizure occurrence in children follows the same pattern as overall seizure occurrence, with an additional increase in the evening (21-22h). In adults, from whom only data on focal seizures was available, an additional increase in seizure occurrence can be observed between 13-17h, followed by a steady decrease during the evening and early night.

Specific localizations and seizure types Circadian seizure distribution varied with localization of the epileptic focus in patients with focal seizures, and between seizure types (Figure 2). Both seizures with a focal (n = 3783) and primarily generalized (n = 575) seizure onset showed a clear increase in seizure occurrence in the early morning (4-7h resp. 6-8h), after a nightly quiescence. Where focal seizures showed an additional peak in the afternoon, occurrence of generalized seizures gradually declined throughout the day, following the circadian rhythm of cortisol (Figure 2A).

Figure 2. Circadian seizure distribution for focal and generalized seizures separately Combined seizure occurrence over a 24-hour period, displayed on top of a standard concentration curve of the plasma cortisol concentration (grey) to enable visual comparison. A. Seizures with a focal versus generalized seizure onset; B. Focal seizures per lobe of origin. C1/2. Various generalized seizures and epileptic spasms. time military time (0-24 hours [h]).

130

Circadian rhythms of seizures and cortisol

Although no EEG based data on seizures with a generalized onset were available for adults, a diary study by Le et al. (2011) also reported these seizures to mainly occur in the morning. Subdivision of focal seizures based on the localization of the epileptic focus showed that associations with the circadian rhythm of cortisol existed especially for seizures with a parietal onset (n = 250) (Figure 2B). These seizures showed a clear peak in occurrence in the early morning and an afternoon decline, although an additional peak was observed in the evening. While the occurrence of seizures with a frontal (n = 752) and seizures with a temporal (n = 2008) onset showed the least circadian variation, both showed an increase in seizure frequency in the early morning, with an additional peak in the afternoon. Seizures from the occipital lobe showed a strong afternoon preference. Some studies suggested differences in circadian seizure occurrence between neocortical and mesial temporal lobe epilepsy, with a morning peak especially for mesial temporal lobe seizures (Durazzo et al., 2008; Karafin et al., 2010). Subdivision of generalized seizures based on seizure type showed that seizure occurrence increased in the early morning (resembling the rising phase in the circadian cortisol rhythm) for myoclonic (n = 89), clonic (n = 297), tonic (n = 201), and atonic (n = 55) seizures (Figure 3C). Most of these seizure types showed a morning peak, except for atonic seizures that peaked in the early afternoon. Although tonic-clonic seizures (n = 230) also often occurred in the early morning, similarity to the cortisol rhythm was less striking because their occurrence was already high during the night and showed a downward trend during the day. Absences (n = 47) and epileptic spasms (n = 299) showed a different rhythm.

DISCUSSION This review summarized the circadian rhythmicity of seizure occurrence by pooling the data of previous studies, and enabled visual comparison with circadian cortisol rhythmicity. The overall circadian occurrence of epileptic seizures resembles the circadian rhythm of cortisol, especially the increase in seizure occurrence in the early morning and the state of quiescence at night. This similarity was observed both in children and adults, but varied between different seizure types and localizations of the epileptic focus. The similarity between the circadian rhythms of seizures and cortisol could be explained by the proconvulsive effects of stress hormones, that were suggested by human studies reporting stress as a seizure precipitant (reviewed by van Campen et al., 2014), and animal studies showing that stress hormones, such as CRH (Ehlers et al., 1983; Marrosu et al., 1987; Marrosu et al., 1988; Baram and Schultz, 1991; Baram and Schultz, 1995) and corticosteroids (Roberts and Keith, 1994; Krugers et al., 1999; Krugers et al., 2000; Schridde and van Luijtelaar, 2004), can influence neuronal excitability and seizure threshold. Besides absolute levels of cortisol, its rise and decline might also play a role. Cortisol is released in ultradian pulses, occurring once every one to two hours. In humans, pulse amplitudes are high early in the morning and gradually decline during the day, thus contributing to the overall circadian pattern (Young et al., 2004; Lightman, 2008). When averaging cortisol

131

PART TWO | CHAPTER 7

concentration curves of multiple patients, individual pulses are no longer discernable, but the rapid changes in cortisol concentration associated especially with the high morning peaks might have large effects on signal transduction and seizure susceptibility (Young et al., 2004; Lightman, 2008; Sarabdjitsingh et al., 2014). The reported similarity between the morning rise in cortisol and seizure occurrence is striking and the time delay, where an increase in cortisol levels precedes the increase in seizure occurrence, supports the possibility of a causal relation. When interpreting the presented data, it is important to keep in mind that circadian seizure occurrence is also influenced by other variables than cortisol. Various types of seizures are differentially affected by sleep patterns. For example in juvenile myoclonic epilepsy, myoclonic seizure frequency increases in the first hours after awakening (Mendez and Radtke, 2001; Dinner, 2002). The studies reviewed here do report differences in seizure occurrence between sleep and wake, but no information is provided on seizure occurrence with respect to the timing of awakening. Although awakening might be considered to confound the relation between cortisol and seizure occurrence, the pathophysiological mechanisms behind the seizure-precipitating effects of awakening are yet unresolved and it could well be hypothesized that the peak levels in cortisol just before and after awakening play a causal role in this. In addition to the sleep-wake cycle, fluctuations in the concentration of anti-epileptic drugs (AEDs), related to dosage intervals (Riva et al., 1984; Hartley et al., 1991; Nielsen et al., 2008), could contribute to the increased seizure occurrence in the early morning. Both factors might also explain some of the differences observed between children and adults, as daytime naps in young children can influence sleep-related seizure occurrence, and for many AEDs, children show an increased clearance and therefore higher fluctuations in blood levels compared with adults (Battino et al., 1995; Battino et al., 1995; Pellock, 2008). In addition, children below the age of two might not yet have developed circadian cortisol rhythmicity, while they are included in the majority of pediatric studies. Especially epileptic spasms mainly occur in infants, which might explain the dissimilarity between their circadian occurrence and the displayed cortisol curve. For these reasons, the chronobiological distribution of epileptic seizures can never be expected to exclusively depend on cortisol concentration, but is rather a result of the interaction of many circadian and exogenous processes influencing neuronal excitability, including cortisol and other stress hormones. This is emphasized by the large differences in circadian seizure occurrence between different seizure types and localizations of the seizure focus, where generalized seizures and focal seizures originating from the parietal lobe in particular followed the circadian rhythm of cortisol. Theoretically, this might be explained by (1) differences in stress sensitivity of the specific brain regions, determined by for example the density of stress hormone receptors and connectivity to other brain regions; or (2) sensitivity of the specific brain regions to other seizure precipitating variables with a circadian rhythm, that might blunt the association with stress hormone levels. As stress hormone receptors are especially abundant in limbic regions and prefrontal cortex (Reul and de Kloet, 1985; JoĂŤls and de Kloet 1992; Watzka et al., 2000), the high resemblance of parietal and generalized seizure occurrence with cortisol levels points towards the second hypothesis. Further studies

132

Circadian rhythms of seizures and cortisol

need to resolve the mechanisms behind the differences in circadian seizure occurrence between seizure types and localizations. In this systematic literature review, data of multiple studies on 24-hoursâ&#x20AC;&#x2122; occurrence of epileptic seizures were pooled. Data subdivisions were restricted to those shown in the original articles with respect to age limits, time bins, seizure classification or localization, as individual patient data was not available. Also, the exact number of patients in which seizures of specific localizations or seizure types was recorded was not reported in the majority of articles. Minimal overlap in seizure data between different studies cannot be completely excluded, but was mitigated as much as possible by excluding data from studies describing the same patient population (Zarowski et al., 2011; Sanchez Fernandez et al., 2013). Furthermore, none of the selected studies measured cortisol during seizure registration. Also, temporal distribution of seizure occurrence, as well as cortisol levels, in inpatient epilepsy monitoring units might differ from the daily life situation. Therefore, this review can only provide indirect evidence for a relation between cortisol concentration and seizure occurrence.

Conclusion and implications We conclude that the circadian occurrence of epileptic seizures shows similarities to the circadian rhythm of cortisol. These results are compatible with the hypothesis that stress hormones influence the occurrence of epileptic seizures. Prospective studies, measuring endogenous cortisol levels or applying exogenous cortisol in patients with epilepsy during EEG â&#x20AC;&#x201C;preferably with high time resolutionâ&#x20AC;&#x201C; are needed to further unravel the effects of the circadian and ultradian cortisol variability on epileptic activity. Increased knowledge on the relation between stress, stress hormones and epilepsy may provide insight in the mechanisms underlying stress sensitivity of seizures and ictogenesis in general, and contribute to improvements of the treatment and care of patients with epilepsy.

133

PART TWO | CHAPTER 7

Supplement

Supplementary Figure 1. Flow chart of article screening and selection ยนArticles were excluded following the criteria in this particular order (number of articles excluded)

134

Supplementary Table 1. Query database

query

PubMed

(epilepsy [Tiab] OR epileptic [Tiab] OR epilepticus [Tiab] OR seizure [Tiab] OR seizures [Tiab]) AND (circadian [Tiab] OR diurnal [Tiab] OR chronobiology [Tiab] OR “sleep/wake” [Tiab] OR “wake/sleep” [Tiab] OR “sleep/wakefulness” [Tiab] OR “wakefulness/sleep” [Tiab] OR rhythm [Tiab] OR rhythmicity [Tiab])

EMBASE

((epilepsy:ti:ab OR epileptic:ti:ab OR epilepticus:ti:ab OR seizure:ti:ab OR seizures:ti:ab) AND (circadian:ti:ab OR diurnal:ti:ab OR chronobiology:ti:ab OR “sleep/wake”:ti:ab OR “wake/sleep”:ti:ab OR “sleep/wakefulness”:ti:ab OR “wakefulness/sleep”:ti:ab OR rhythm:ti:ab OR rhythmicity:ti:ab))

Search performed on 30/07/2014

135

CHAPTER

Stress sensitivity of seizures influences the relation between cortisol fluctuations and interictal epileptiform discharges in people with epilepsy in preparationâ&#x20AC;&#x192;

Jolien S. van Campen E. Lorraine Hompe Demetrios N. Velis Willem M. Otte Fia van der Berg Floor E. Jansen Kees P.J. Braun Gerhard H. Visser Josemir W. Sander Marian JoĂŤls Maeike Zijlmans

PART TWO | CHAPTER 8

ABSTRACT Introduction People with epilepsy often report seizures precipitated by stress. This is considered to be due to the effects of stress hormones on neuronal excitability and seizure threshold. Stress hormone levels vary under baseline conditions. Stress hormones are released in a circadian rhythm with additional hourly pulses. The effect of these ultradian peaks in stress hormone levels on epileptic activity is yet unknown. We expected that (1) stress hormone levels correlate with the incidence of epileptiform abnormalities in the electroencephalogram (EEG) of people with epilepsy, and (2) this relationship is stronger in people with stressprecipitated seizures. Methods We included people with pharmacoresistant localization-related epilepsy who were admitted for 24-hour or multiple day video-EEG monitoring. Morning cortisol levels were measured in saliva samples obtained every 15 minutes for five hours, starting directly after awakening, on one or two consecutive days. In the same time periods, the incidence of interictal epileptiform discharges (IEDs) was determined. We investigated the relationship between cortisol levels and the incidence of IEDs per patient and on a group-level. Additionally, we analyzed the effect of individual, epilepsy and recording characteristics, including selfreported stress sensitivity of seizures, on the strength and direction of this relation. Results Twenty-nine recordings were performed in 21 people. Overall, cortisol levels showed a significant positive relationship with the incidence of IEDs (Î˛ = 0.11, p = 0.046). On an individual basis there was large variability, with both positive and negative correlations occurring. In multivariable analysis, the relationship between cortisol and IEDs was positively associated with the self-reported stress sensitivity of seizures (Î˛ = 0.32, p = 0.005), and not with other individual, epilepsy or recording characteristics. Conclusion The relationship between cortisol levels and the incidence of IEDs suggests that stress hormones influence disease activity in epilepsy, also under basal conditions. Self-reported stress sensitivity of seizures partly explains individual differences in this relationship, which provides indirect proof for the existence of a pathophysiological basis for this subjective phenomenon.

138

Cortisol and epileptiform discharges

INTRODUCTION Stress is the seizure-precipitant most often reported by people with epilepsy (Sperling et al., 2008; van Campen et al., 2012; Wassenaar et al., 2014). The effects of stress on seizure susceptibility are considered to be mediated by stress hormones, as stress and stress hormones have been shown to influence neuronal excitability and seizure susceptibility in preclinical epilepsy models (reviewed by (Joëls, 2009; van Campen et al., 2014). The principal neuroendocrine system activated by stress is the hypothalamic-pituitary-adrenal (HPA-) axis. The stress hormone cortisol is the ‘end product’ of this axis. The stress-induced release of cortisol comes on top of endogenous daily fluctuations in hormonal concentrations. More specifically, cortisol is released in ultradian pulses, occurring every one to two hours (Hellman et al., 1970; Weitzman et al., 1971; Veldhuis et al., 1990; Young et al., 2004; Lightman and Conway-Campbell, 2010). The amplitude of these pulses is high around awakening, and low at the end of the active period, giving rise to an overarching diurnal pattern (Dickmeis, 2009). Previous animal studies have shown large effects of corticosteroids pulses on hippocampal neurotransmission (Sarabdjitsingh et al., 2014). Therefore, the rapid changes in cortisol concentration associated especially with the high amplitude morning peaks in humans might have large effects on neuronal excitability and seizure susceptibility. Currently, the potential relationship between diurnal and ultradian fluctuations in cortisol levels and epileptiform activity is unknown. In 80-90% of people with epilepsy, interictal epileptiform discharges (IEDs) can be demonstrated in the electroencephalogram (EEG) after repeated recordings (Ajmone Marsan and Zivin, 1970; Salinsky et al., 1987; Walczak et al., 1993). These IEDs have a high specificity for epilepsy, as they are only seen in approximately 0.0-0.5% of asymptomatic healthy adults (Gibbs and Gibbs, 1964; O’Connor, 1964; Bennett, 1967; LeTourneau and Merren, 1973; Gregory et al., 1993; Jabbari et al., 2000). They are therefore invaluable to support the clinical diagnosis of epilepsy, the classification of the electroclinical syndrome and to localise the epileptic focus. IEDs have been suggested as a measure of epileptic activity and seizure risk (Pillai and Sperling, 2006), although it has also been suggested that IEDs control rather than promote seizure activity (Avoli, 2001). Either way, IEDs enable EEG interpretation of epilepsy-related neuronal activity, also in the absence of (more sporadic) seizures. We investigated the effect of diurnal and ultradian fluctuations in cortisol levels on interictal epileptiform activity by simultaneously measuring salivary cortisol levels and IEDs in people with localization-related epilepsy. We hypothesized that (1) the incidence of IEDs in people with epilepsy is related to the cortisol concentration, and (2) this relationship is stronger in people with known stress-precipitated seizures.

139

PART TWO | CHAPTER 8

METHODS Subjects The study cohort consisted of people with epilepsy who were admitted for long term EEG recording (≥24 hour) for diagnostic or pre-surgical evaluation to the epilepsy center “Stichting Epilepsie Instellingen Nederland” [SEIN], Heemstede, the Netherlands. Only adults, legally capable of providing informed consent and not taking stress hormone medication or oral contraceptives (as this blunts cortisol peaks [Pruessner et al., 1997]) were eligible for inclusion. Individuals with an expected seizure frequency of more than one seizure per five hours, or IED incidence of less than one IED per 15 minutes, were excluded. To determine the latter, IED incidence during the initial afternoon of the videoEEG registration was examined by an experienced epileptologist (MZ). Only individuals fulfilling these inclusion criteria were asked to be study participants. Individuals in whom fewer than five cortisol samples could be obtained were retrospectively excluded. The study was approved by the ethical review board of the University Medical Center Utrecht and informed consent was obtained from all individuals. Individual and epilepsy characteristics Information on general characteristics (age, sex, comorbidities, use of co-medication), epilepsy characteristics (age at onset, seizure semiology, seizure frequency, underlying etiology, localization of the epileptic focus, current use of anti-epileptic drugs [AEDs]), and the habitual interventions related to the long-term video-EEG monitoring (i.e., tapering of AEDs, hyperventilation and sleep deprivation provocations) aimed at increasing the seizure yield of the recordings, was obtained from medical records. Subjects reported time of awakening, which was verified by the video-EEG recording. Experimental procedures Morning cortisol levels were measured every 15 minutes for five hours, starting directly after awakening (see below). This period was selected because ultradian fluctuations are then most pronounced (Lightman and Conway-Campbell, 2010). During the same time periods, the number of IEDs was documented. Whenever possible, measurements were performed on two consecutive days. In individuals in whom medication was tapered before or during a presurgical 5-day EEG, the experiment was performed on the third and fourth days of the recording to minimize interference of change in AED dosage. During the time of the experiment, activities were documented every 15 minutes and individuals provided a subjective stress score on a visual analogue scale (VAS) ranging from “no stress” (0) to “the most severe stress I have ever experienced” (10). Stress sensitivity of seizures was assessed using a previously designed questionnaire (van Campen et al., 2012). In this questionnaire, people report on seizures precipitated by acute stress and an increase in seizure frequency in periods of stress. The number of hours of sleep and sleep quality in the previous night were assessed with the Groninger Sleep Quality Scale (Meijman et al., 1990).

140

Cortisol and epileptiform discharges

EEG registration EEG was recorded using 32 electrodes (Micromed, System PLUS EVOLUTION, Mogliano Veneto, TV, Italy), placed according to the international 10â&#x20AC;&#x201C;20 system, referred to a G2 electrode (placed between Cz and Fz). Electrodes were placed at electrode descriptor positions Fp1, Fp2, Fz, F3, F4, F7, F8, F9, Cz, C3, C4, F10, Pz, P3, P4, P7, P8, T7, T8, O1 and O2. In addition to the above, the 10% system (Revised Combinatorial Nomenclature of EEG Descriptors) was used for electrode descriptor positions F9 and F10. In EEGs recorded for pre-surgical evaluation, additional 10% electrodes were placed at FT11, FT12, T9 and T10 and two sphenoidal electrodes were sometimes placed to improve registration of the temporal lobe. Registrations included electrocardiography (ECG). Signals were sampled at 256 Hz. Identification of seizures and IEDs Time and duration of clinical and electrophysiological seizure activity was deduced from the clinical reports and checked in the video-EEG recordings. To identify IEDs, EEG channels were high-pass filtered on 0.3 and 5.3 Hz and referenced to a common average montage. IEDs were defined as spikes or sharp waves (including single spikes, polyspikes, spike-and-slow-wave complexes, sharp-and-slow-wave complexes and single sharp waves) with an evident epileptiform pattern (Noachtar et al., 1999; Pillai and Sperling, 2006). IEDs were counted from 15 minutes before the first to 15 minutes after the last saliva sampling. IEDs were first identified and marked by one of the authors (LH) under supervision of an experienced EEG technician (FvdB) and the whole files were reviewed by an experienced epileptologist (MZ). Both observers were blinded for the cortisol levels. In case of multifocality, IED analysis was restricted to the focus with most frequent discharges. Time windows in which artefacts interfered with reliable IED detection (e.g., movement, muscle activity, eye movement and eye blinks) were excluded from the analysis, as was the time when individuals 1) were asleep (which could occur in the 15 minutes before the first saliva sampling), 2) were hyperventilating or 3) had an electroencephalographic seizure (including the first two minutes after hyperventilation or ictal activity). We investigated the relationship between cortisol levels and the incidence of IEDs in each 15 minute time interval. Time-windows in which less than one minute out of the 15 minutes was available for determination of IED incidence were excluded from analysis. Cortisol measurements Cortisol was measured in saliva, a noninvasive method with the advantage of measuring the biologically active, free fraction of cortisol, avoiding the stress of blood sampling and enabling repetitive sampling over short time periods (Kirschbaum and Hellhammer, 2000; Gozansky et al., 2005). Peak concentrations of cortisol in saliva lag by less than two to three minutes after plasma levels (Kirschbaum and Hellhammer, 2000). Fluctuations in free peripheral corticosteroid levels correlate well with levels in the brain, at least in rodents (Qian et al., 2012).

141

PART TWO | CHAPTER 8

Saliva sampling started immediately after awakening and was repeated every 15 minutes for five hours, resulting in a total of 20 samples per individual. Saliva was collected using Salivettes® (Starstedt, Etten-Leur, the Netherlands), i.e., sterile cotton swabs, on which participants were instructed to chew gently for two minutes. They refrained from liquid or food intake in the five minutes before and during sampling, and rinsed their mouths with water five minutes before sampling after any food intake, to avoid contamination of saliva (Groschl et al., 2003; Chiappin et al., 2007). Lunch was consumed after finishing the experimental protocol for that day. Collected saliva samples were stored at -9 °C for one to four days, after which Salivettes® were centrifuged for 10 minutes at 3000 rpm. Saliva was stored at -20 °C until further analysis. Cortisol levels were measured without extraction with an in-house competitive radio-immunoassay using a polyclonal anticortisol-antibody (K7348) and [1,2-3H(N)]hydrocortisone (PerkinElmer NET396250UC) tracer (van den Bos et al., 2014). All samples from one individual were analyzed in the same batch.

Statistical analysis For each recording day, the mean IED incidence per 15 minute time window was calculated by dividing the number of IEDs by the artefact-free proportion of the time window in minutes. The correlation between cortisol and the mean IED incidence was analyzed per recording day with Spearman cross-correlation to assess a possible time lag in the relationship between cortisol and IEDs. In cross-correlation analysis, a sliding time lag was used with 7.5 minute shifts, where cortisol was related to the incidence of IEDs in the time windows ranging from 15 to 0 minutes before, to 105 to 120 minutes after the specific cortisol measurement, as the main effects of cortisol on neuronal excitability are expected to occur in this time window (de Kloet et al., 2005). The absolute correlation coefficients per time lag were averaged over all recording days and the time lag with the highest correlation coefficient over all recordings was used for further analysis (preferred time lag). Cortisol levels were correlated to IED rate per 15 minute time intervals with the preferred time lag using a Spearman correlation coefficient. The relationship between cortisol and IED incidence on group-level was assessed with a linear mixed model, with each individual as the random identity factor and time point and testing day as repeated measures (to correct for some people having one and other two days of measurements), with covariance type ‘autoregressive order 1’ to correct for autocorrelation in the data. To obtain normal distribution residuals (evaluated with Q-Q plots), windows without IEDs were excluded from the analysis, and cortisol levels and IED incidence were log transformed prior to model fitting. Next, we examined the effects of individual, epilepsy and recording characteristics, including stress sensitivity, on the relationship between cortisol and IED by including them as covariate, additional to cortisol, in the group-level model. Variables that showed a significant main effect or cortisol-interaction effect were included in a multivariable model. In case of high collinearity between variables (variance inflation factor [VIF] ≥ 5 or tolerance ≤ 0.10), only the variable with the lowest p-value in univariable analysis was included in the model. For nominal variables showing a significant interaction

142

Cortisol and epileptiform discharges

with cortisol on IED incidence, posthoc tests were performed per group. To get an indication of the relationship between cortisol levels and seizures, cortisol levels before seizures were compared with cortisol levels at all other time-points, using the time lag previously determined for the relationship between cortisol and IEDs. This analysis was performed for all seizures combined, as well as separately for recording days with a positive versus a negative direction of the correlation between cortisol and IEDs, with an independent samples two-sided t-test for unequal variances (which is preferred for very small sample sizes; (de Winter, 2013). A two-sided p value of < 0.05 was considered statistically significant. Posthoc tests were Bonferroni corrected for multiple comparisons. Analysis was performed using the Statistical Package for the Social Sciences version (SPSS 22.0) and R (version 3.0.3).

RESULTS Twenty-three individuals were included, of whom two withdrew after a generalized tonicclonic seizure, resulting in fewer than five saliva samples. Data of 29 recording days in 21 individuals (57% male) were available for analysis (Table 1). Individuals had an average age of 40 years (range 21 â&#x20AC;&#x201C; 69 years), and a median seizure frequency of five seizures per month (ranging from one per year to 4.5 per day). On two recording days, fewer than 20 samples were collected: one individual quit after collection of 16 samples on the first recording day, and the other provided only nine samples on the second recording day, before the protocol was stopped because the prolonged postictal status after a generalized seizure prevented further study participation. The localization of the epileptic focus and underlying etiology varied. AEDs were used on 25 days of videoEEG telemetry recording (17 individuals); on 21 of them (14 individuals) AEDs were tapered in the days preceding the recording. Eight recording days (eight individuals) were preceded by sleep deprivation provocation and on twelve days (nine individuals) short periods of hyperventilation were performed during morning hours. Overall, 53% of the time was artefact-free and available for identification of IEDs. Fewer than 5% of time windows were excluded because artefact-free time was less than one minute. In 26% of time windows, no IEDs occurred. A total of 8025 IEDs were marked and the median number of IEDs per artefact-free 15 minutes was 6 (range 0-36). On five of the 29 recording days, individuals experienced one or two seizures during the testing protocol.

Time lag analysis Over all recording days the highest absolute correlation existed between cortisol and IEDs in the 15 minutes following the saliva sampling for cortisol measurement (median Ď =0.38). Exclusion of the recordings during which seizures occurred resulted in the same preferred time lag (median Ď =0.37). This time lag was used for further analysis.

143

144

3-4

1-2

3-4

0.4

150

1.3

0.1

0.4

R+L

unknown

frontal

temporal

frontal

temporal

occipital

temporal

frontal

temporal

frontal

temporal

cortical dysplasia

MTS

polymicrogyria + heterotopia

unknown

MTS

low grade neoplasma

unknown

MTS

unknown

MTS

hippocampal developmental disorder MTS

24 h

5 day

24 h

5 day

24 h

5 day

24 h

5 day

24 h

5 day

24 h

5 day

24 h

5 day

stressEEG sensitive duration seizures

MTS

unknown

hemorrhagic infarct

unknown

vascular malformation

etiology

CBZ, PHT

LTG, OCB

LTG, CBZ

none

LTG

CBZ

none

PHT, PB, LTG

CBZ, LEV, PGB

TPM, OCB

LEV

LEV, LSM

CLB, LTG, LEV

none

VPA, CBZ

LEV, LTG

ZNS, CBZ

LTG, LEV, CLB

none

GBP, CBZ

AED use

day 1

day 2

day 1

day 2

AED seizures tapering

yr years; stress-sensitive seizures self-reported increase in seizure frequency in periods of stress; EEG electro-encephalogram; AED anti-epileptic drug; m male; f female; L left; R right; MTS mesiotemporal sclerosis; + yes; - no; h hour; CBZ carbamazepine; CLB clobazam; GBP gabapentin; LEV levetiracetam; LTG lamotrigine; LSM lacosamide; OCB oxcarbazepine; PB phenobarbital; PGB pregabalin; PHT phenytoin; TPM topiramate; VPA valproic acid; ZNS zonisamide; day 1/2: seizure on the first/second registration day.

subject sex age age at seizure hemisphere localisation (yr) onset frequency (yr) (per month)

Table 1. Individual characteristics

PART TWO | CHAPTER 8

Cortisol and epileptiform discharges

Relation between cortisol levels and IED incidence Analysis of the correlation between cortisol and IED incidence per recording day revealed a significant relationship in 10 (34%) of the recording days (9 [43%] of patients). There were similar numbers of positive and negative correlations (Table 2; Figure 1). In seven of the eight individuals with two recording days, the direction of the correlation between cortisol and IEDs on these two days was the same (Table 2). Autocorrelation existed for most patients in cortisol samples and IED incidence measures obtained within 30-45 minutes after each other. When data of all individuals were combined and corrected for auto-correlation, a significant positive relation between cortisol levels and IED incidence was observed (Î˛ = 0.11, p = 0.046). Relationship with individual, epilepsy and recording characteristics Next, we analyzed the possible interaction of cortisol with different individual, epilepsy and recording characteristics with respect to their effects on IED incidence. The relationship between cortisol and IED incidence was significantly higher in people reporting stresssensitive seizures, specifically during periods of stress. Additionally, the relationship between cortisol and IED incidence was positively associated with subjective stress-scores during the recording day and negatively associated with the occurrence of seizures during the recording. In multivariable analysis, only stress sensitivity of seizures for periods of stress remained significant (Table 2). Posthoc analysis of the relationship between cortisol and IED incidence suggested that in people reporting an increase in seizure frequency in periods of stress, cortisol was significantly and positively related to IED incidence (Î˛ = 0.27, p = 0.02), while this was not the case in those not reporting this stress sensitivity of seizures (Î˛ = -0.07, p = 0.64) (Figure 2). Cortisol levels before seizures To get an indication of the cortisol levels in relation to seizures (n = 7), cortisol levels in samples collected in the 0 to 15 minutes before seizure onset were compared with cortisol levels in all other samples. Overall, mean cortisol levels before seizures were slightly higher than at other time points, but this difference was not statistically significant (log transformed cortisol 2.7 vs. 2.6, p = 0.28).

145

PART TWO | CHAPTER 8

Figure 1. Examples of the correlation between cortisol and IED incidence Cortisol and interictal epileptiform discharges (IEDs) in the time in three representative recordings, during which no seizures occurred. In subject one day two (upper panel) and subject 15 (middle panel), cortisol levels and IED incidence follow the same pattern, while in subject 19 (lower panel), an inverse relationship is present.

146

Cortisol and epileptiform discharges

Table 2. Relationship between cortisol and IEDs per recording no.

correlation cortisol-IEDs

subject

day

0.28

0.25

0.51

0.023

0.40

0.08

0.74

<0.001

0.28

0.23

-0.77

<0.001

-0.32

0.24

-0.53

0.016

0.51

0.023

0.49

0.027

0.19

0.42

0.02

0.93

-0.05

0.85

-0.13

0.58

-0.20

0.60

0.37

0.12

0.26

0.39

-0.44

0.09

-0.52

0.019

0.40

0.08

-0.38

0.13

-0.41

0.08

-0.57

0.021

-0.70

0.001

0.17

0.48

-0.23

0.33

0.30

0.20

0.45

0.048

-0.04

0.87

no. recording number; IED interictal epileptiform discharge; r correlation coefficient; p p-value; adata available for16 time points; b data available for 9 time points; bold significant correlation; italic significant positive correlation.

147

148 0.00 0.00 0.00 0.00

age at onset of epilepsy, years

epilepsy duration, years

seizure frequency per month

periods of stress

0.01

hyperventilation

AED tapering

0.96

0.66

0.80 0.16

0.21

0.08

-0.89

-0.04

-0.63

-0.33

-0.81

0.21

0.38

-0.54

-0.64

-0.01

-0.02

-0.01

-0.02

0.54

0.71

0.55

0.81

0.03

0.49

0.13

0.44

0.07

0.60

0.39

0.23

0.16

0.39

0.26

0.63

0.17

0.19

p-value

variable

0.11

-0.12

0.14

0.02

-0.06

0.28

-0.04

0.24

-0.10

0.10

0.20

0.22

0.08

0.07

0.13

-0.06

0.17

0.11

0.31

0.17

0.77

0.49

<0.001

0.67

<0.001

0.05

0.14

0.05

0.01

0.18

0.49

0.20

0.72

0.03

p-value

cortisol

0.17

0.03

0.32

0.25

0.24

0.005

p-value

interaction (variable x cortisol)

-0.24

0.02

-0.40

0.59

0.76

0.36

p-value

variable

multivariable analysis

-0.10

0.33

p-value

cortisol

The effect of various individual, epilepsy and recording characteristics on the relationship between cortisol and IED incidence (interaction variable x cortisol), as well as the specifics of the other variables in the model (i.e., the effect of the variable and cortisol itself on IED incidence). Variables with a significant interaction with cortisol in the analysis per variable, were included in the multivariable analysis. AED anti-epileptic drug; aexcluded from multivariable analysis because of high collinearity with stress sensitivity for periods of stress; bold interaction p < 0.05

0.04

sleep deprivation

0.01

0.37 -0.03

seizures during registration

0.03

0.05

subjective stress-scores (per 15 min)

recording specific

0.06 0.001

0.23 0.37

acute stress

0.001

0.42

stress sensitivity overalla

0.74 0.40

-0.04

- temporal

0.30

0.15

0.09

0.13

- frontal

0.21

0.32

0.95

0.27

0.23

p-value

- mesiotemporal

0.17

- hemisphere, left

localization

-0.13

age, years

interaction (variable x cortisol)

sex, male

individual or epilepsy specific

characteristic

analysis per characteristic

Table 3. Effect of individual, epilepsy and recording characteristics on the relationship between cortisol and IEDs

PART TWO | CHAPTER 8

Cortisol and epileptiform discharges

Figure 2. Cortisol and IED incidence in relation to stress sensitivity of seizures Visualization of the interaction between stress sensitivity for periods of stress and cortisol on IED incidence. The interaction was significant with p = 0.001.

DISCUSSION Cortisol levels relate to the incidence of IEDs. On a group level, cortisol was significantly related to IEDs incidence. Large individual differences existed and on an individual basis, positive as well as negative associations existed. The relationship between cortisol and IED incidence was higher in people with a self-reported increase in seizure frequency during periods of stress. The relationship between cortisol levels and IED incidence suggests that stress hormones –within 15 minutes– affect neuronal synchronization. The overall positive relationship is in line with previous human studies reporting a relationship between stress and epilepsy, either by retrospective self-report (Hayden et al., 1992; Antebi and Bird, 1993; Hart and Shorvon, 1995; Spatt et al., 1998; Frucht et al., 2000; Spector et al., 2000; da Silva Sousa et al., 2005; Nakken et al., 2005; Fang et al., 2008; Sperling et al., 2008; Pinikahana and Dono, 2009; dos Santos Lunardi et al., 2011; van Campen et al., 2012; Ferlisi and Shorvon, 2014; Wassenaar et al., 2014; Privitera et al., 2014), or based on diary association (Temkin and Davis, 1984; Blanchet and Frommer, 1986; Webster and Mawer, 1989; Neugebauer et al., 1994; Swinkels et al., 1998; Klein and van Passel, 2005; Haut et al., 2007)). In animal studies, stress hormones such as corticotrophic hormone (CRH) (Ehlers et al., 1983; Marrosu et al., 1987; Marrosu et al., 1988; Baram and Schultz, 1991; Baram and Schultz, 1995) and corticosterone (Roberts and Keith, 1994; Smith-Swintosky et al., 1996; Krugers et al., 1999; Krugers et al., 2000; Schridde and van Luijtelaar, 2004) have been shown to influence neuronal excitability and seizure threshold. In humans, the effect of ultradian fluctuations in stress hormone levels on neuronal excitability and seizure susceptibility has not previously been studied. However, in animals single or repetitive corticosteroid pulses –without exposure to other stress mediators– have been consistently reported to change hippocampal or amygdalar excitability (Karst and Joëls, 2005; Groc et al., 2008; Sarabdjitsingh et al., 2014). T h e increased relationship between cortisol and IED incidence in individuals with stresssensitive seizures, provides indirect proof for the existence of a pathophysiological basis for this subjective phenomenon.

149

PART TWO | CHAPTER 8

Possible mechanisms The individual differences in the relationship between cortisol and IED incidence might be explained by variation in (the balance of) stress hormone receptors. A relationship between stress hormone regulation and the effects of stress hormones on seizure susceptibility (i.e., stress sensitivity of seizures) was previously shown in a pediatric epilepsy sample (chapter 6), and suggested before by a relationship between early life stress and stress-precipitated seizures (van Campen et al., 2012). In other stress-related diseases, the effects of genes and environment on HPA-axis dysregulation and subsequent disease vulnerability have been hypothesized to be mediated by an altered balance between the two corticosteroid receptor types (de Kloet et al., 1991; De Kloet et al., 1998; Holsboer, 2000; de Kloet et al., 2005; de Kloet, 2014). Both the mineralocorticoid receptor and the glucocorticoid receptor mediate the effect of corticosteroids on HPA-axis activity and neuronal excitability. Therefore, variations in the balance between these (and maybe also other) stress hormone receptors might influence the effects of stress hormones on neuronal excitability and epileptiform activity. The effects of stress hormones on neuronal functioning depend on the timing of the measurement after exposure. In the current study, salivary cortisol levels showed the strongest correlation with IED incidence immediately afterwards. This is consistent with the fast, non-genomic effects of corticosteroids on neuronal excitability –which are mainly contributed to the membrane-associated mineralocorticoid receptor–, rather than the slow gene-mediated glucocorticoid actions which typically develop with a delay of one to two hours (Tasker et al., 2006; Joëls et al., 2008). The effects of single corticosteroid pulses on neuronal excitability are widely studied in rodents, but the impact of the ultradian cortisol variability on brain functioning is largely unresolved. Recently, it was suggested –based on rodent studies– that the effects of the repetitive ultradian corticosteroid pulses on neuronal transmission and synaptic plasticity differ from those of single cortisol peaks (Sarabdjitsingh et al., 2014). Besides the repetition, the release of other stress mediators that can also affect neuronal excitability themselves and in interaction with cortisol (e.g., noradrenaline; (Joëls and Baram, 2009), might also result in differential effects of an increase in cortisol concentration as part of the total stress response compared with circadian and ultradian fluctuations in cortisol levels. Furthermore, ultradian variations in HPA-activity are not only reflected in cortisol levels, but also in other hormones involved in the system, e.g. ACTH (Lightman and Conway-Campbell, 2010). These peptidergic hormones cannot reliably be measured in saliva. We can therefore not exclude the possibility that ultradian fluctuations in stress hormones other than cortisol contribute to the observed correlations. Study design and limitations We chose to study IEDs in scalp video-EEG telemetry recordings as a marker for epileptiform activity. IEDs, however, probably result from combined inhibitory and excitatory synchronous neuronal firing (Kohling et al., 1998; Cohen et al., 2002; de Curtis et al., 2012), and do not necessarily mirror seizure susceptibility (Engel and Ackermann,

150

Cortisol and epileptiform discharges

1980; Gotman and Marciani, 1985; Gotman and Koffler, 1989; Avoli, 2001; de Curtis and Avanzini, 2001; Spencer et al., 2008). Seizures themselves might have been a better marker, but would require sampling during a much more prolonged period to yield enough statistical power which, with the current method of cortisol sample collection, would greatly increase the burden for study participants. Automated sample collection systems, which are currently being developed (Bhake et al., 2013), could facilitate this type of research. Alternatively, high frequency oscillations or spike characteristics might be more specific biomarkers for epileptic disease activity (Zijlmans et al., 2009). This study was performed in a clinical sample of people with active epilepsy and EEGs were recorded for diagnostic purposes, e.g. assessing epileptic origin of suspected events and localisation of the epileptogenic zone in people considered candidates for epilepsy surgery. To increase the likelihood of capturing an epileptic seizure, AEDs were often tapered during the recording, and on some recording days individuals were sleep deprived or performed hyperventilation provocation tasks. These interventions are likely to influence the number of IEDs and might also affect cortisol concentrations. Although IEDs occurring during or in the direct aftermath of seizures or hyperventilation provocation were excluded from the analysis, the occurrence of seizures during the registration was associated with the incidence of IEDs and the relationship between cortisol and IED incidence. Hyperventilation, seizure occurrence, and other potential confounders such as AED tapering and sleep deprivation did not, however, interact with the relationship between cortisol and IED incidence in multivariable analysis. Data were analyzed with linear mixed model analysis to correct for autocorrelation and to pool results of all subjects. To enable this analysis, time windows in which no IEDs occurred had to be excluded, resulting in data loss that impaired reliable individual analysis of recording days. The relationship between cortisol and IED incidence per recording day was therefore analyzed by calculating correlation coefficients, which could not be corrected for autocorrelation, resulting in a possible overestimation of the reported correlations. Hence, results of both methods are complementary and should be interpreted with these limitations in mind. Furthermore, as for all observational studies, no evidence is provided on causality of the reported associations. Controlled experimental trials are needed to unravel further the effects of stress and stress hormones on neuronal excitability or seizure susceptibility in people with epilepsy. Particularly studies in which an experimental stressor, or controlled stress hormone application, combined with long-term EEG recordings and cortisol monitoring, could give insight in the effects of stress hormones on epileptiform activity and seizure susceptibility, and provide knowledge on similarities, differences and possible interactions between the effects of stress hormone release in the light of circadian and ultradian variability versus the stress response.

Conclusion We showed that fluctuations in cortisol levels relate to interictal epileptiform EEG activity in people with localization-related epilepsy. Self-reported stress sensitivity of seizures can partly explain individual differences in this relationship. Increased knowledge on the

151

PART TWO | CHAPTER 8

relationship between stress hormones and epileptic or epileptiform activity can improve counseling of people with epilepsy on how to deal with stressful situations, and may provide directions for the development of new treatment strategies.

ACKNOWLEDGEMENTS The authors would like to thank Dr. E.W.G. Lentjes and I. Maitimu for salivary cortisol and alpha-amylase analysis and all EEG technicians and nurses from the Stichting Epilepsie Instellingen Nederland (SEIN) Heemstede for their helpful suggestions and collaboration. JvC was supported by the UMC Utrecht Alexandre Suerman Stipendium, MZ was supported by the Rudolf Magnus Young Talent Fellowship and ZonMw Veni grant no. 91615149.

152

153

PART

THREE

Stress hormones and epileptogenesis

CHAPTER

Effects of repetitive mild stress and corticosteroid exposure on epileptogenesis after early life experimental febrile seizures in mice submittedâ&#x20AC;&#x192;

Jolien S. van Campen Ellen V.S. Hessel Kirsten Bohmbach Giorgio Rizzi Paul J. Lucassen Sada Lakshmi Turimella Eduardo H.L. Umeoka Gideon F. Meerhoff Kees P.J. Braun Pierre N.E. de Graan* Marian JoĂŤls* *authors share seniority

PART THREE | CHAPTER 9

ABSTRACT Introduction Stress is the most frequently self-reported seizure precipitant in patients with epilepsy. Moreover, a relation between early life stress and epilepsy has been suggested. Although early life stress and stress hormones are known to influence seizure threshold in rodents, effects on the development of epilepsy (epileptogenesis) are still unclear. Therefore, we studied the consequences of early life corticosteroid exposure for epileptogenesis, under highly controlled conditions in an animal model. Methods Experimental febrile seizures (eFS) were elicited in 10-days-old mice by warm-air induced hyperthermia, while a control group was exposed to a normothermic condition. In the following two weeks, mice received either seven corticosterone or vehicle injections or were left undisturbed. Specific measures indicative for epileptogenesis were examined at 25-days of age and compared with vehicle injected or untreated mice. We examined structural (neurogenesis, dendritic morphology and mossy fiber sprouting) and functional (glutamatergic postsynaptic currents and long-term potentiation) plasticity in the dentate gyrus (DG). Results We found that differences in DG morphology induced by eFS were aggravated by repetitive (mildly stressful) vehicle injections and corticosterone exposure. In the injected groups, eFS were associated with decreases in neurogenesis, and increases in cell proliferation, dendritic length and spine density. No group differences were found in mossy fiber sprouting. Despite these changes in dentate gyrus morphology, no effects of eFS were found on functional plasticity. Conclusion Corticosterone exposure during early epileptogenesis, elicited by eFS, aggravates morphological, but not functional, changes in the DG, which partly supports the hypothesis that early life stress stimulates epileptogenesis.

158

Corticosteroids and epileptogenesis

INTRODUCTION Epilepsy is a common neurological disorder, especially in childhood where its prevalence is as high as 0.5-1.0% (Shinnar et al., 2002). An important factor influencing epilepsy and epileptic seizures is stress, which is the most frequently self-reported seizure precipitant in patients with epilepsy (reviewed by van Campen et al., 2014). The seizure precipitating effects of stress are also confirmed by prospective studies (Feldman and Paul,1976; Neufeld et al.,1994; Swinkels et al.,1998; Bosjnak et al., 2002; Klein and van Passel 2005). Besides direct effects on seizure susceptibility, stress can also influence the risk of being diagnosed with epilepsy later in life (Christensen et al., 2007; Shang et al., 2010; Li et al., 2008). Thus, associations between stress and epilepsy exist on multiple levels. However, the mechanisms behind these relations are so far poorly understood. Animal models can provide more insight into the exact mechanisms by which stress influences epilepsy. In various preclinical epilepsy models, stress has been shown to lower the threshold for the induction of epileptic seizures and to increase seizure severity (Huang et al., 2002; Lai et al., 2006, 2009; Salzberg et al., 2007; Jones et al., 2009; Kumar et al., 2011). The effects of stress on epilepsy are largely attributed to neuronal exposure to stress hormones. Especially stress hormone exposure early in life can have profound effects on later brain morphology and function and predispose to the development of brain diseases (as reviewed by Lupien et al., 2009; Lai and Huang, 2011; Lucassen et al., 2013; Carr et al., 2013; Loi et al., 2014). Stress hormones have been shown to directly affect neuronal excitability (reviewed by JoĂŤls, 2009; Maguire and Salpekar, 2013). Despite these effects of stress on seizures on the one hand, and on brain development on the other, effects of stress and stress hormones on the development of epilepsy (i.e. epileptogenesis) are currently unknown. To improve insight into the impact of stress hormones on epileptogenesis, we studied the effects of corticosterone, an important stress hormone, during early life epileptogenesis on neuronal morphology and functional plasticity in the rodent brain. Using a controlled design, we elicited experimental febrile seizures (eFS) in young mouse pups by warm-air induced hyperthermia (HT) and subsequently exposed them to repetitive (1) high concentrations of corticosterone, (2) vehicle injections (a control condition that is also a mild stressor) or (3) no injections. We next examined alterations in morphological and functional parameters in the dentate gyrus (DG), a hippocampal subarea that is affected by eFS (Bender et al., 2003; Lemmens et al., 2005; Kwak et al., 2008; Notenboom et al., 2010) as well as stress hormones (JoĂŤls, 2007). To assess morphological changes we investigated neurogenesis, cell proliferation, dendritic morphology, spine density and mossy fiber sprouting. Functional plasticity was assessed measuring glutamatergic transmission and long-term potentiation in the DG. We hypothesize that corticosterone aggravates the epileptogenic changes after eFS.

159

PART THREE | CHAPTER 9

METHODS Animals Breeding pairs of C57BL6/J mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA) and subsequently bred in-house. Litters used in this experiment were derived from multiple breeding pairs (n = 99). On postnatal day (P) 1, litters were culled to four to six pups consisting of both males and females. The pups were not weaned during the experiment. Animals were kept in a controlled 12-h light-dark cycle (light on 7 AM - 7 PM) with a temperature of 22 ± 1 °C. Food and water were available ad libitum (2111 RMH-TM diet; Hope Farms, Woerden, the Netherlands). All animals were housed in transparent Plexiglas cages (Macrolon type II) with sawdust bedding and paper tissues for nest building. Cages were cleaned at P7 and at P17/18 (in between injection days) by replacing half of the sawdust bedding. All experimental procedures were performed according to the institutional guidelines of the University Medical Center Utrecht and approved by the committee on ethical considerations in animal experiments of Utrecht University (DEC Utrecht, permit number 2012.I.03.047). All efforts were made to minimize suffering of the animals. Pups were assigned to multiple treatment groups per litter. A maximum of two pups per litter was used per treatment group per method of analysis to minimize litter effects on outcome measures. All animal experiments were performed within a period of six months and animals in all treatment groups were tested across the whole period to control for environmental or seasonal variation. For the purpose of this study, experiments were only performed on male animals. Corticosterone levels after injection To evaluate corticosterone levels after injection with corticosterone or vehicle, naïve P12 mice (n = 4-6 per time point per treatment group) were injected intraperitoneally with corticosterone (corticosterone-HBC complex 3 mg/kg dissolved in saline, total injection volume of 10 microliter/gram body weight) or vehicle (both obtained from Sigma-Aldrich, the Netherlands) between 8.30 and 9.00 AM. Immediately before, or at 15, 30, 60, 120, 180 or 240 minutes after injection, mice were decapitated and trunk blood was collected. Between injection and decapitation, pups were returned to their home cage and left undisturbed. A separate group of non-injected mice was decapitated at the same time points to control for diurnal corticosterone variability. Epileptogenesis Epileptogenesis was induced using the HT-induced eFS model, a very subtle epilepsy model with close resemblance to the human situation in which children who experience complex febrile seizures are at increased risk to develop temporal lobe epilepsy later in life (Dubé et al., 2007, 2009). A unique aspect of this model is the relatively long-lasting latent phase of epileptogenesis, making it easier to study effects of additional risk factors prior to the actual onset of epilepsy, and irrespective of the damage and compensatory mechanisms induced by spontaneous seizures.

160

Corticosteroids and epileptogenesis

Prolonged eFS were induced in P10/11 mice by heated-air induced HT using a previously described paradigm in rats (Baram et al.,1997), which we adapted to mice (van Gassen et al., 2008). A temperature-sensitive transponder (IPTT-300 BioMedic Data Systems, Plexx BV, Elst, the Netherlands) was implanted subcutaneously on P9/10 in pups with a bodyweight between 5.0 and 6.5 g. One day after transponder implantation, body weight was determined and mice were placed in a preheated cylindrical chamber and exposed to a warm air stream of 41-48°C. To prevent skin burn and adverse effects on behavior, the temperature of the chamber floor was maintained at 39°C. Core body temperature was measured at least every 2.5 minute period using a wireless temperature reader (WRS6007; Plexx BV). To provoke prolonged seizures, air temperature was adjusted to maintain the core body temperature between 41.5 and 42°C. The presence of tonic–clonic convulsions was monitored by observation. These behavioral seizures correlate closely with electroencephalographic seizures, i.e., spike-wave discharges in the hippocampus, as shown by previous experiments in our lab (Hessel et al., 2009; Notenboom et al., 2010). After 30 minutes of HT (defined as core temperature ≥ 39°C) pups were rapidly cooled in a water bath at room temperature, gently dried with paper and returned to the dam. This procedure is known to induce epileptogenesis, as spontaneous (encephalographic) seizures are observed after a latent period of approximately three months in 35-68% of animals (Dubé et al., 2006; Kwak et al., 2008; Koyama et al., 2012), and epileptiform interictal discharges in 88% (Dubé et al., 2006). Normothermia (NT) controls were treated as HT pups, except that the temperature of the air stream was kept at 30-32°C, resulting in a constant body temperature. All eFS experiments were performed between 10.00 AM and 3.00 PM.

Injections and decapitation Animals were weighed at 2, 4, 6, 8, 10, 12 and 14 days after exposure to HT/NT and injected intraperitoneally with corticosterone or vehicle between 8.00 and 9.30 AM, during the circadian trough. A separate group of animals was left undisturbed after HT/NT. One day (~24 hours) after the last injection, mice were weighed and decapitated or perfused (see Figure 1). Decapitation was performed between 8.00 and 9.30 AM. Endocrinology Trunk blood samples were collected immediately after decapitation at P25/26 from all animals subjected to HT/NT except those receiving perfusion fixation. Directly after decapitation, thymus and both adrenals were resected and weighed. Blood samples were centrifuged for 10 minutes at 4000 rpm at 4 °C. Plasma was stored at -80 °C until assayed with an I125-corticosterone radioimmunoassay for mice (MP Biomedicals, Inc., Aberdeen, UK) according to the manufacturer’s instructions. All samples were processed in the same assay to exclude inter-assay variability.

161

PART THREE | CHAPTER 9

Figure 1. Experimental design Animals were exposed to hyperthermia (HT) or normothernia (NT) at postnatal day (P) 10 or 11 and subsequently received injections with corticosterone, vehicle or no injection every other day in the latent phase of epileptogenesis. Brains were dissected 24 hours after the last injection, before the onset of spontaneous seizure activity.

Morphology and functional plasticity All morphological and functional outcome parameters were examined in the DG. This hippocampal area was selected based on its functional and anatomical characteristics. Firstly, the (epileptogenic) changes induced by eFS are located in the hippocampus, including the DG (Bender et al., 2003; Lemmens et al., 2005; Kwak et al., 2008; Notenboom et al., 2010; Koyama et al., 2012). Secondly, the DG has an important function in filtering excitation and seizure propagation to the other parts of the hippocampus (Behr et al.,1998, Hsu, 2007). Thirdly, it is one of the few sites where neurogenesis continues in later life, a process that can be stimulated by seizures and has been implicated in epileptogenesis (Cameron and McKay, 2001, Parent et al.,1997, 2002; Bielefeld et al., 2014; Danzer et al., 2012). Finally, the DG exhibits abundant receptors for corticosterone and therefore its morphology and function are largely influenced by early life stress (Joëls, 2007). All following experimental procedures were performed by experimenters unaware of the treatment groups. Per outcome measure, tissue of animals belonging to different treatment groups was ordered in a semi-randomized way to control for environmental or experimenter effects on recording, staining and/or quantification over the total experimental period. Neurogenesis Male mice (n = 6 per group) were decapitated between 9.00 and 10.30 AM. Brains were dissected, post fixed at 4°C in 4% formaldehyde for 4 hours, transferred to 30% sucrose for 24 hours, frozen on powdered dry ice and stored at -80°C. Cryostat sections (20 µm) were cut in the coronal plane, collected in series of 15 on Superfrost slides and stored at -80°C until further use. As a measure of neurogenesis, tissue was stained for Doublecortin (DCX), a marker for neuronal precursor cells and immature neurons, and Ki67, a marker for proliferating cells using modified protocols. Mounted sections were postfixed in acetone-methanol (1:1) at -20°C for 7.5 minutes, washed in 0.05M Tris HCl 0.9% saline (TBS) pH 7.6 and heated in

162

Corticosteroids and epileptogenesis

0.01 M citrate buffer (pH 6.0) in a microwave oven for 10 minutes at 800 W followed by 5 minutes at 480 W and 5 minutes at 260 W. After a cool down period of 20 minutes, sections were washed in TBS. Endogenous peroxidase activity was blocked with 0.5% (for DCX) or 1.5% (for Ki67) H2O2 in TBS for 15 minutes. Sections were washed and, after incubation in 2% milk powder in TBS for 30 minutes, incubated with the primary antibody (Doublecortin [DCX] [polyclonal goat anti-DCX, SantaCruz; 1:800] or Ki67 [polyclonal rabbit anti-Ki67p, Novocastra, 1:5000]), diluted in supermix (0.25% gelatine and 0.1% Triton in TBS) at room temperature for 1 h and then incubated overnight at 4°C. The next morning, sections were washed and incubated for 2 h with donkey anti-goat biotinylated (for DCX, Jackson; 1:500) or goat anti-rabbit biotinylated (for Ki67, Vector; 1:200) secondary antibody diluted in supermix at room temperature. Sections were washed, incubated in avidin-biotin complex (ABC) (ABC Elite, Vector Laboratories; 1:800 in TBS) for 2 h (for DCX) or 1.5 h (for Ki67) at room temperature, washed again and incubated with biotinylated tyramide (1:500) in 0.01% H2O2 in TBS for 30 minutes. Sections were washed and incubated in ABC (1:800 in TBS) for 1.5 h at room temperature and washed in TBS. After washing in 0.05M Tris HCl pH 7.6 (TB), chromogen development was performed with diaminobenzidine (DAB; 50 mg/100mL Tris buffer, pH 7.6, 0.01% H2O2, 0.05% Nickel) for 40 minutes. Sections were washed in TB and stored overnight at 4°C. The next day, sections were washed in distilled water, counterstained with Haematoxylin and shortly rinsed in distilled water. After washing with running tap water, sections were dehydrated using a grading series of ethanol, cleared in xylene and coverslipped using Entallan. DCX+ and Ki67+ cells in the granule cell layer and the subgranular zone of the DG were quantified unilaterally in every 15th section in a total of five sections per animal within Bregma range -1.46 to -2.80 (coronal) without a left/right preference within or between animals. DCX+ cells were quantified stereologically using a StereoInvestigator system (Microbright field, USA) with a ×100 oil-immersion objective of a Zeiss Axiophot microscope and StereoInvestigator software, according to the optical fractionator method. The number of DCX+ cells was estimated using a 25µm×25µm counting frame, with a grid size of 70 × 80 µm. Section thickness was 10.5µm. The estimated total of DCX+ cells within the studied range was determined using the optical fractionator method and multiplied by two to correct for unilateral counting. Ki67+ cells were counted manually using a light microscope (Olympus BH-2) with ×40 magnification and multiplied by the inverse of the sampling fraction and by two to correct for unilateral counting. The total estimated number of DCX+ or Ki67+ cells per animal was used for analysis.

Dendritic morphology Male mice (n = 6 per group) were decapitated between 9.00 and 10.30 AM. Directly after decapitation, brains were dissected. Rapid Golgi staining (FD rapid-Golgi staining, Neurotechnologies) was performed according to the manufacturer’s instructions with an impregnation time of 9 days. Vibratome sections (200 µm) were cut in the transversal plane (Leica VT 1000S; Leica Microsystems, Nussloch, Germany). Images were obtained using Zen 2011 (Carl Zeiss) in combination with an automated stage and focus control connected

163

PART THREE | CHAPTER 9

to the microscope. Golgi-impregnated dentate granule cells, fulfilling the following criteria, were randomly selected: (1) localization in the middle part (relative to the DG curvature and start of the CA3 region) of the suprapyramidal blade of the DG, within Bregma range -2.16 and -3.16 (transversal), (2) consistent and dark impregnation along the entire extent for all dendrites, (3) relative isolation from neighboring impregnated neurons to avoid interference with analysis. For morphological quantification, 8 neurons from each animal in each treatment group were traced. Image stacks of 0.5 µm thickness were automatically acquired and combined. Neurons were traced using NeuroLucida software (MicroBrightField, Inc. Colchester, Vt, USA) to obtain a 3D representation of each cell. Numerical analysis and graphical processing were performed with NeuroExplorer (MicroBrightField). Traced dendritic trees were evaluated by two investigators unaware of the treatment, on completeness of staining/ tracing and on whether they belonged to a single cell, followed (if required) by a consensus meeting. Dendritic trees that were considered not completely stained/traced or belonging to multiple neurons were excluded from analysis (n = 50 [17%], 5 - 10 neurons per group, evenly distributed over the groups). Spines were counted in two segments of ±20 µm per neuron, located on different dendrites. Segments were randomly chosen based on the following criteria: (1) localization at approx. 100 µm (80-120 µm) radial distance from the cell soma; (2) secondary or higher order dendritic branches; and (3) straight and remaining in a single focal-plane. Of each cell, dendritic properties were evaluated by analyzing the total dendritic length and maximum dendritic reach (radial distance), and the dendritic complexity index ([Σ branch tip orders + # branch tips] × [total dendritic length/total number of primary dendrites] [Pillai et al., 2012]). If dendritic length significantly differed between treatment groups, post-hoc Sholl plots (Sholl,1953) were constructed by plotting the dendritic length as a function of radial distance from the soma center in 18 µm intervals. To normalize the distribution of the dendritic complexity index, neurons with a dendritic complexity index >2 standard deviations (SD) from the group mean were considered outliers and removed from the respective analysis (n = 12 [5%], one to three neurons per group, evenly distributed over the groups). The remaining cells were subdivided into cells located in the inner-most part of the granule cell layer versus cells located in the middle or outer part of the granule cell layer (see Results). This subdivision was performed independently by two investigators blind to the treatment groups and compared afterwards. In 96% of the cases the subdivision made by the two investigators was in agreement. In the remaining cases consensus was reached after in-depth investigation of the location.

Mossy fiber sprouting Male mice (n = 6 per group) were killed between 9 and 12 AM under deep pentobarbital anesthesia (200 mg⁄kg body weight, i.p.) by transcardial perfusion with 0.1% sodium sulfide for 5 minutes, followed by 4% formaldehyde for 5 minutes (each in 0.01 phosphate-buffered saline, pH 7.4). Brains were removed from the skull, postfixed at 4°C in 4% formaldehyde/15% sucrose overnight, immersed at 4°C in 30% sucrose in phosphate-buffered saline until they sank and frozen on powered dry ice. Cryostat sections (30 µm) were cut in the coronal

164

Corticosteroids and epileptogenesis

plane, mounted on superfrost slides in series of 15 and stored at -80°C until further use. Mossy fibers were stained with Timm histochemistry, according to Danscher (1981). Staining was performed in two batches, each containing one slide of every animal to avoid staining-based variation between treatment groups. Sections were developed in the dark for 180 minutes in a freshly prepared 90/45/15/0.75 (volume/volume) solution of 50% arabic gum, 51% hydroquinone, 25.5% citric acid / 23.5% sodium citrate, and 17% silver nitrate. After washing with running tap water, sections were dehydrated using a grading series of ethanol, cleared in xylene and coverslipped using malinol. Mossy fiber staining in the hippocampal CA3 area and infrapyramidal blade of the DG (the main areas of possible mossy fiber sprouting) was scored according to the scoring system described by Holmes et al. (1999), ranging from 0 (no staining) to 5 (maximum staining). Eight sections of the septal hippocampus per animal (four sections for both staining batches), within Bregma range -1.46 to -2.80 (coronal), were scored by two independent observers, followed by a consensus meeting. Consensus scores of these eight sections were pooled per animal and the average was used for statistical analysis.

Slice preparation for electrophysiology Male mice (n = 6-9 per group) were decapitated between 8.15 and 9.15 AM, a few minutes after taking the animal out of its home cage. Only the first two mice of each litter were used for electrophysiological analysis to avoid confounding effects of a rise in plasma corticosterone due to acute stress. After rapid dissection, the brain was chilled in ice-cold, carbogenated (95% O2:5% CO2) artificial cerebrospinal fluid (aCSF) containing in mM: NaCl 120, KCl 3.5, MgSO4 5.0, NaH2PO4 1.25, CaCl2 0.2, NaHCO3 25.0 and D-glucose 10.0. After removing frontal lobes and cerebellum, 350 µm transversal hippocampal sections were prepared using a vibratome (Leica VT 1000S; Leica Microsystems, Nussloch, Germany). Both hemispheres were separated and all sections were incubated at room temperature in continuously carbogenated aCSF for at least one hour. Patch-clamp recording of spontaneous excitatory postsynaptic currents (EPSCs) Patching of DG neurons was performed using an upright microscope (Nicon Eclipse E600FN) with differential interference contrast and a water immersion objective (x40) to visually identify the cells. The sections were continuously perfused with carbogenated aCSF containing in mM NaCl 120, KCl 3.5, MgSO4 5.0, NaH2PO4 1.25, CaCl2 0.2, NaHCO3 25.0 and D-glucose 10 and 20 µM biccuculine to block γ-aminobutyric acid (GABA)A-receptor mediated transmission (flow rate 2.0 ml⁄min, temperature 32°C, pH 7.4). Cell patching was performed as described by Pasricha et al. (2011). Briefly, patch electrodes were pulled from a Sutter Instruments Micropipette puller and had a tip resistance of 4-6 MΩ when filled with the pipette (intracellular) solution containing in mM; Cs-methane sulfonate 120, CsCl 17.5, HEPES 10, BAPTA 5, MgATP 2, NaGTP 0.1, pH 7.3 (adjusted with CsOH). An Axopatch 200B amplifier (Axon Instruments, Foster City, CA, USA) was used for whole cell recordings, operating in the voltage-clamp mode. The patch-clamp amplifier was interfaced to a computer via a Digidata (type 1322A; Axon Instruments) analog-to-digital

165

PART THREE | CHAPTER 9

converter. Data acquisition was performed with Clampex, version 8.2 (Axon Instruments) at a sampling rate of 50µs and a 5 kHz Bessel filter. The surface of the section was cleaned to have better vision of the cells in the deeper layers of DG. After establishing a gigaseal, the membrane patch was ruptured and the cell was clamped at a holding potential of -70 mV to allow measurement of currents mediated by the α-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid (AMPA) receptors, as at this potential, the N-Methyl-D-aspartic acid or N-Methyl-D-aspartate [NMDA] receptor is blocked by Mg2+. Spontaneous EPSCs (sEPSCs) were recorded for 5 minutes, starting 5-10 minutes after membrane rupture. After administration of tetrodotoxin (TTX, 0.5µM, Bioconnect services) for 5 minutes to block voltage-gated Na channels and consequently also the development of action potentials, miniature EPSCs (mEPSCs) were measured for 5 minutes. Only one cell was recorded per slice and no more than two recordings were obtained per animal. Only recordings with an uncompensated series resistance of <2.5 times the pipette resistance, <20% variation during the recording period, and with frequencies <2 SD from the mean, were accepted for analysis. For each event, the area under the curve and inter-event interval was analyzed using Clampfit version 9.2 (Axon Instruments). The median area under the curve and inter-event interval per cell were used for statistical analysis.

Field potential recordings Sections were transferred to a submersion type recording chamber and continuously perfused with carbogenated aCSF containing in mM: NaCl 124, KCl 2.5, MgSO4 4.0, NaH2PO4 1.2, CaCl2 4.0, NaHCO3 26.0, D-glucose 10.0 and 20 µM bicuculline (flow rate 2 ml/min, temperature 32°C, pH 7.4). Field excitatory postsynaptic potentials (fEPSPs) were recorded in the DG, using glass microelectrodes filled with aCSF, positioned in the medial perforant pathway over the suprapyramidal blade of the DG, as confirmed by paired pulse stimulation elicited paired pulse inhibition. Minimum and maximum stimulation intensities were identified. After an incubation period of 20 minutes, an input-output response curve was generated by gradually increasing the stimulus intensity to define the stimulus intensity that generated the half-maximal response in peak-amplitude; this intensity was used for the remainder of the experiment. To measure paired pulse depression, paired pulses were delivered at an inter stimulus interval of 50, 100 and 200 ms. After 15-20 minutes of stable baseline recordings, sections were tetanized with 4 trains of 50 pulses of supra-maximal intensity at 100Hz (30 seconds inter train interval) to induce long term potentiation (LTP). Field potentials were recorded for one hour post-tetanus at 30s intervals. At the end of this period, the post-tetanic paired pulse depression and final input-output curve were determined. LTP was quantified by calculating the ratio between fEPSP slopes recorded post tetanization and pre tetanization. Data were acquired and analyzed with Signal 2.0 software (Cambridge Electronic Design, United Kingdom). Statistical analysis The effect of eFS and injection type on outcome measures was analyzed with a general linear mixed model in a 2x3 design, examining the main effects of HT (HT versus NT) and

166

Corticosteroids and epileptogenesis

injection type (corticosterone versus vehicle versus none), as well as their interaction. In case of a significant main effect of injection type, post-hoc tests were performed with Bonferroni correction for multiple comparison. Additionally, differences between HT and NT animals were analyzed per injection group with an independent samples t-test, Bonferroni corrected for multiple comparison. When data from multiple neurons per animal, or multiple segments per neuron, were analyzed, a linear mixed model was used, including the animal (for dendritic complexity and glutamatergic transmission) or neuron (for spine density) as subject variables. Effects of HT, injection type and their interaction on Sholl distribution were tested with a repeated measure general linear model. Correlation between parameters was assessed with Pearson correlation coefficient. Mean fEPSP slopes pre- and post-tetanization were compared with a paired samples t-test. Normality of residues was evaluated with Q-Q plots, variance of residues was evaluated with error plots. Differences were considered statistically significant at p < 0.05 (two-tailed) after correction for multiple comparison. Differences that were significant before, but not after correction for multiple comparison (0.05 ≤ p < 0.15), were considered trends. Data were analyzed using SPSS 20.0 (SPSS, Inc., Chicago, IL, USA). Unless stated otherwise, data are presented as mean ± standard error of the mean.

RESULTS Epileptic seizures and neuroendocrine changes Experimental seizures were elicited in all mice exposed to the HT protocol and the average seizure duration was 26.4 ± 2.0 minutes. Corticosterone injection resulted in a high peak concentration of corticosterone (1043.3 ± 28.0 ng/ml at 15 minutes after injection) with a fast return to baseline in approx. 180 min, while vehicle injection elicited a much smaller (94.6 ± 8.2 ng/ml) and more short-termed increase in corticosterone levels (Figure 2A). Weight increase between HT/NT treatment and decapitation, a 15 days interval, was significantly lower in HT animals compared with controls (main effect HT, F(1,280)=10.73, p = 0.001), but was not influenced by subsequent injections (Figure 2B). No group differences were observed in adrenal and thymus weights (data not shown). Although there were less than 2 minutes between successive decapitations of animals belonging to the same litter, later decapitation was associated with increased corticosterone levels (Figure 2C, left panel). While no group differences were observed in corticosterone levels in animals decapitated first of their litter (Figure 2C, middle panel), injection type did significantly influence corticosterone levels in animals decapitated as third to sixth of their litter (Fig 2C, right panel, main effect injection, F(2,68) = 8.7, p < 0.001; corticosterone versus vehicle p = 0.009, corticosterone versus no injection p < 0.001, vehicle versus no injection p = 0.36), indicating that repetitive corticosterone injection indeed affected stress hormone regulation, which confirms successful manipulation.

167

PART THREE | CHAPTER 9

Figure 2. Endocrinology A. Corticosterone (cort) levels at several time points after single intraperitoneal injection with corticosterone, vehicle or no injection at 9.00 AM. Reference cort levels: basal < ± 50 ng/ml, circadian peak or mild stress ± 50 - 200 ng/ml, severe stress ± 200 - 600 ng/ml, supraphysiological > ± 600 ng/ml (Barriga et al., 2001; Malisch et al., 2007; Conboy et al., 2011). B. Weight increase was significantly lower in hyperthermia (HT) compared with normothermia (NT) treated animals overall, while analysis per injection type only showed a trend difference in the not injected groups. C. Corticosterone levels after decapitation increased with the order of decapitation within the litter (left panel). In animals decapitated as first of their litter (middle panel), corticosterone levels did not differ between treatment group, while in animals decapitated third to sixth of their litter (right panel), corticosterone levels differed between injection types –with significantly lower levels in the corticosterone injected animals–, but not between HT and NT treated animals. Data are presented as mean ± standard error of the mean. ***p < 0.001, **p < 0.01, ^p < 0.05 before correction for multiple comparison (trend).

168

Corticosteroids and epileptogenesis

Morphology Neurogenesis Neurogenesis significantly differed between HT and NT animals. HT was associated with a decrease in the number of immature DCX-positive neurons (main effect HT, F(1,30) = 6.21, p = 0.019) (Figure 3A), and an increase in the number of proliferating cells in the DG (main effect HT, F(1,30) = 4.94, p = 0.034) compared with NT (Figure 3B). Accordingly, the number of immature neurons and proliferating cells per animal were negatively correlated (r = -0.55, p = 0.001). Analysis per injection type revealed that the effects of HT on both DCXand Ki67-staining were significant in the vehicle injected animals (F(1,10) = 12.16, p = 0.018, resp. F(1,10) = 15.89, p = 0.009), while a similar effect was observed at trend level after corticosterone injection (F(1,10) = 5.71, p = 0.12, resp. F(1,10) = 6.88, p = 0.09). Non-injected HT and NT groups did not differ at all.

9 Figure 3. Neurogenesis A. Left: the number of immature neurons in the dentate gyrus was decreased in male animals exposed to hyperthermia (HT) compared with normothermia (NT) after injection. Middle/right: representative example of DCX staining at lower and higher magnification. B. Left: the number of proliferating cells in the dentate gyrus was increased in males exposed to HT. Middle/right: representative example of Ki67 staining at lower and higher magnification. cort corticosterone. Data are presented as mean Âą standard error of the mean. *p < 0.05, ^p < 0.05 before correction for multiple comparison (trend).

169

PART THREE | CHAPTER 9

Figure 4. Dendritic morphology A. Dendritic complexity. A1. Dendritic complexity differed between dentate granule cells with a cell body located in the granular cell layer or inner part of the molecular layer (inner layer) versus the middle or outer part of the molecular cell layer (outer layer), representative examples of corticosterone (cort) injected animals after hyperthermia (HT) and normothermia (NT). A2. Left: dendritic complexity index (DCI) in inner layer neurons was increased after HT compared with NT. Right: total dendritic length of inner layer neurons was increased after HT, an effect that increased with injection type. A3. HT significantly influenced Sholl distribution. B. Spine density. B1. Representative example of HT-cort and NT-cort animal. B2. Spine density was significantly higher in animals exposed to HT compared with NT, an effect that increased with injection type. cort corticosterone. Data are presented as mean Âą standard error of the mean. *p < 0.05, ^p < 0.05 before correction for multiple comparison (trend).

170

Corticosteroids and epileptogenesis

Dendritic morphology To assess the effects of eFS and injection type on dendritic morphology, we evaluated total dendritic length, dendritic reach, dendritic complexity and spine density. As dendritic complexity differed between dentate granule cells with a cell body located in the inner part of the granular cell layer, bordering the subgranular layer (referred to as ‘inner layer’, n = 15-27 per group), versus the middle or outer part of the granule cell layer (referred to as ‘outer layer’, n = 11-25 per group) (Figure 4A1), these cells were analyzed separately. In inner layer neurons, dendritic complexity index was increased in HT animals compared with NT controls (main effect HT, F(1,111) = 6.23, p = 0.014), an effect that was only significant in animals receiving no subsequent injections (F(1,7) = 6.64, p = 0.04) and was less pronounced after vehicle or corticosterone injection (Figure 4A2, left panel). Also the total dendritic length of inner layer neurons was increased in HT animals (main effect HT, F(1,29) = 4.47, p = 0.04), an effect that aggravated with injection type, although analysis per injection type only revealed a difference in the corticosterone-injected group at trend level (Figure 4A2, right panel). No main nor interaction effects were observed on total dendritic reach. Sholl analysis per neuron revealed a main effect of HT (F(1,120) = 6.18, p = 0.01) and radial distance from the soma (F(3,358) = 560.71, p < 0.001) on dendrite length, as well as a HT × injection type × radial distance interaction (F(6,358) = 3.18, p = 0.005) (Figure 4A3). In outer layer neurons, no main or interaction effects of HT or injection type were observed on dendritic complexity (data not shown). Spine density was significantly higher in animals exposed to HT compared with NT (main effect HT, F(1,165) = 7.28, p = 0.008). This effect was significant after corticosterone injection (F(1,58) = 8.80, p = 0.01), at trend level after vehicle injection (F(1,55) = 4.67, p = 0.11), while it was not observed in animals receiving no injections (F(1,52) = 0.00, p = 1.00) (Figure 4B1/2). Mossy fiber sprouting Hippocampal mossy fiber sprouting was assessed in the DG and the CA3 area of the hippocampus. As expected, intense Timm staining was observed in the hilus of the DG and in the stratum lucidum of the CA3 area, the main projection sites of mossy fibers. The amount of infrapyramidal Timm staining, characteristic for mossy fiber sprouting, was very low in all treatment groups (mean mossy fiber sprouting score DG 0.76 ± 0.09, CA3 1.07 ± 0.08, on a scale ranging from 0 to 5). HT or injection type did not significantly affect mossy fiber sprouting score in both areas, although there was a trend towards an increased mossy fiber sprouting score after HT in animals receiving corticosterone injections (F(1,10) = 7.56, p = 0.06, data not shown). Functional plasticity The eFS-induced changes in morphology and particularly the increased spine density may affect glutamatergic transmission in the dentate gyrus, which could be reflected at the level of single cells as well as field potentials. To test this, we examined effects of HT and corticosterone injection on single cell spontaneous synaptic events (which are among other things determined by the number of synaptic contacts) and field excitatory potentials.

171

PART THREE | CHAPTER 9

Single cell glutamatergic transmission Whole-cell voltage-clamp recordings of AMPA receptor mediated EPSCs in the dentate gyrus were analyzed in 9 to 11 cells per treatment group. Input resistance and capacitance did not differ between groups (see Table 1). For both sEPSCs and mEPSCs, no HT × injection interaction or main effect of HT was observed. Interestingly, injection type did influence sEPSC and mEPSC properties (Figure 5A/B). A significant main effect of injection type existed on the interval between consecutive events for both sEPSC (main effect injection, F(2,32) = 3.48, p = 0.03; corticosterone versus vehicle p = 0.63; corticosterone versus no injections p = 0.06, vehicle versus no injections p = 0.01) (Figure 5A, right panel) and mEPSC (main effect injection, F(2,32) = 3.95, p = 0.03; corticosterone versus vehicle p = 0.65; corticosterone versus no injections p = 0.07, vehicle versus no injections p = 0.01) (Figure 5B, right panel), with a lower inter-event interval in the vehicle injection group compared with the non-injection group; a similar difference, although only at trend level, was observed for the corticosterone-injection versus non-injection groups. Also the area under the curve of the sEPSCs was influenced by injection type (F(2,36) = 4.39, p = 0.02; corticosterone versus vehicle p = 0.04, vehicle versus no injections p = 0.004, corticosterone versus no injections not significant) with a smaller area under the curve after vehicle injection compared with the other groups (Figure 5A, middle panel), while the area under the curve of mEPSC did not significantly differ between groups (Figure 5B, middle panel). Synaptic plasticity Changes in glutamatergic transmission at the single cell level may alter circuit properties and the ability to induce synaptic plasticity. This was tested with field potential recording and application of high-frequency stimulation. Treatment groups did not differ with respect to baseline slopes or stimulation intensity necessary to produce the half-maximal fEPSP, suggesting that HT and injection type did not influence baseline evoked transmission (see Table 1). A significant increase in fEPSP slope post tetanization (LTP) was elicited in 71% of all animals (38 - 88% per group). Synaptic plasticity at half-maximal or maximal stimulation was not significantly influenced by the HT × injection interaction or by HT,

Figure 5. Functional plasticity A. Left: example showing a spontaneous excitatory postsynaptic current (sEPSC) recording. Middle and right: a main effect of injection type, but not hyperthermia (HT) vs. normothermia (NT), is observed on sEPSC area under the curve (AUC) and interevent interval. B. Left: example showing a miniature excitatory postsynaptic current (mEPSC) recording. Middle: no group differences are observed in mEPSC AUC. Right: a main effect of injection type, but not HT, is observed on mEPSC interevent interval. C. Left: recordings of field evoked postsynaptic potentials (fEPSP) in HT vs. NT animals after repetitive corticosterone injection (tetanization started at t=0). Inset: example showing increased slope and amplitude of fEPSP after tetanization. Right: injection type, but not HT, influenced fEPSP long term potentiation (LTP). cort corticosterone. Data are presented as mean ± standard error of the mean **p < 0.01,*p < 0.05, ^p < 0.05 before correction for multiple comparison (trend).

172

Corticosteroids and epileptogenesis

173

PART THREE | CHAPTER 9

but only by injection type (main effect injection, F(2,38) = 4.20, p = 0.02; corticosterone versus vehicle p = 0.02, corticosterone versus no injections p = 0.44, vehicle versus no injections p = 0.01, other group comparisons not significant) (Figure 5C). At maximum stimulation intensities, no group differences were observed (data not shown).

Table 1. Baseline measurements of single cell glutamatergic transmission and synaptic plasticity treatment group

single cell transmission

synaptic plasticity

input resistance (mΩ)

capacitance (pF)

baseline slope (V/s)

stimulation intensity (mA)

HT-cort

336 ± 49

9.8 ± 0.7

-0.13 ± 0.02

0.63 ± 0.05

NT-cort

338 ± 38

9.0 ± 0.8

-0.20 ± 0.02

0.71 ± 0.03

HT-vehicle

303 ± 33

9.0 ± 0.7

-0.19 ± 0.02

0.66 ± 0.02

NT-vehicle

320 ± 38

9.2 ± 0.7

-0.14 ± 0.03

0.77 ± 0.08

HT-none

285 ± 35

9.4 ± 0.8

-0.22 ± 0.02

0.80 ± 0.12

NT-none

302 ± 37

9.6 ± 1.1

-0.22 ± 0.04

0.84 ± 0.14

Baseline measurements of functional plasticity were not significantly influenced by the HT × injection type interaction, HT or injection type. HT hyperthermia, NT normothermia, cort corticosterone. Data are presented as mean ± standard error of the mean.

DISCUSSION To increase understanding of the effects of stress and stress hormones on early life epileptogenesis, we induced epileptogenesis in young mice using the HT-induced eFS model and subsequently exposed them to (1) corticosterone, (2) vehicle injection (a mild stressor) or (3) no injections, and evaluated morphological and functional parameters in the DG. In this latent phase of early epileptogenesis, few effects of eFS were observed in animals that were not subsequently injected. However, in mice that received repetitive corticosterone or vehicle injection (mild stress), eFS induced epileptogenesis was associated with changes in dentate gyrus morphology, namely a lower number of immature cells, an increased cell proliferation and an increased dendritic length and spine density. These morphological changes associated with epileptogenesis did not translate into differences in glutamatergic functional plasticity.

Experimental model The eFS model is a model of early life prolonged febrile seizures with close resemblance to the human situation (Dubé et al., 2007, 2009). The relatively long-lasting latent phase of epileptogenesis in this model provides the unique opportunity to study epileptogenesis in the absence of spontaneous seizures and relatively apart from the damage and compensatory mechanisms that may arise due to recurrent electrographical or clinical seizure activity.

174

Corticosteroids and epileptogenesis

However, non-epileptogenic effects of the initial eFS cannot be fully excluded. As we were interested in epileptogenic changes, outcome measures were assessed before the onset of spontaneous seizures. Based on earlier literature, a subset of animals exposed to eFS is expected not to develop epilepsy (DubĂŠ et al., 2006; Kwak et al., 2008; Koyama et al., 2012). However, the normal distribution of our outcome measures suggests a gradual distribution of epileptogenesis among animals. Further studies measuring long term (seizure) outcome are required to determine whether the morphological and functional measures assessed are stable during the course of epileptogenesis, as well as their relevance for epilepsy outcome. We hypothesized the mild epileptogenic changes in the latent phase after eFS to be aggravated by stress (hormone) exposure. As different stressors might differentially affect seizure susceptibility (reviewed by Sawyer and Escayg, 2010; Maguire and Salpekar, 2013; van Campen et al., 2014), we decided to use a clean design of injections with corticosterone -the end product of the stress response, exerting large effects on brain structure and function, including neuronal excitability (JoĂŤls, 2009; JoĂŤls et al., 2009)-, and control groups of mild injection stress and undisturbed animals. Stress paradigms in rodents are usually associated with a reduction in bodyweight and effects on adrenals and thymus. In our experiment we did not observe such changes, possibly due to the mild nature of the injection stress. However, the reduced corticosterone levels in corticosterone-injected animals that were decapitated after prior handling of littermates, which can be considered an acute stressor, indicate that corticosterone injections did down-regulate responsiveness of the hypothalamic-pituitary-adrenal axis. The effects of these supraphysiological doses of corticosterone were smaller than expected, which might relate to the young age of the animals, just after the stress hyporesponsive period. Also, the transient increases in corticosterone levels in the morning may not have interfered with the ultradian pulsatility of endogenous hormone levels (Lightman et al., 2008), benefiting its comparability to real life stress exposure.

Effects of corticosteroids on structural plasticity during epileptogenesis In animals that were not subsequently injected, HT only affected dendritic complexity, but none of the other outcome measures. The negative results during this early phase of epileptogenesis are in line with previous studies that also reported no differences in mossy fiber sprouting (Notenboom et al., 2010; Bender et al., 2003) or DG neurogenesis (Bender et al., 2003) around this age. Similarly, in non-injected mice we observed no effects of HT on DG spine density. Differences between animals exposed to HT and NT only became manifest after repetitive corticosterone or vehicle injections. The increase of HT-associated changes after stress (hormone) exposure is in line with the vast amount of previous studies reporting stress exposure before seizure induction to increase seizure susceptibility and seizure-severity (reviewed by Jones et al.,2014). The effects of HT combined with corticosteroids or mild stress on proliferation are similar to those reported at a later stage of epileptogenesis after HT only (Lemmens et al., 2005; Kwak et al., 2008), suggesting that corticosterone and stress accelerate epileptogenesis-related structural plasticity. We observed an inverse relation between the

175

PART THREE | CHAPTER 9

number of proliferating cells and the amount of immature neurons, suggesting that the increased proliferation mainly occurs in non-neuronal (e.g. glial) cells, consistent with the gliosis prominent in many epilepsy models (Represa et al., 1995; Aronica et al., 2012). Although dendritic morphology has not previously been evaluated in the eFS model, dendritic length, complexity and spine density are generally reported to decrease after seizures (Ward, 1961; Belichencko et al., 1992; Multani et al., 1994; Nishimura et al., 2008; Gibbs et al., 2011; Singh et al., 2013), which is considered a compensatory mechanism in response to the excess of excitatory input. The enhanced dendritic length and spine density that we found during the latent phase of early epileptogenesis in combination with corticosterone exposure and mild stress might therefore well be part of the epileptogenic process. Clearly, the parameters that we investigated might not only be altered by epileptogenesis, but also by corticosterone and stress themselves. This is shown for instance by the increased number of DCX+ cells in the vehicle injected compared with non-injected NT group, that is consistent with the literature (Oomen et al.,2009; Loi et al.,2014).

Effects of corticosteroids on functional plasticity during epileptogenesis In non-injected animals, glutamatergic transmission and LTP were unaffected by HT. The latter is somewhat surprising, as around this age HT was shown to affect GABAergic transmission in DG (Swijsen et al., 2012) and both glutamateric and GABA-ergic transmission in CA1 (Chen et al., 1999; Dubé et al., 2000; Kamal et al., 2006; Notenboom et al., 2010; Ouardouz et al., 2010). Since we did not record GABAergic signals, we cannot exclude that in our model GABAergic transmission in the DG might have been altered. In view of the increased dendritic length and spine density observed after HT when followed by corticosterone (and to a lesser degree vehicle) injection, the latent phase of early epileptogenesis may be accompanied by an expanded postsynaptic ‘potential’ for synaptic transfer of signals. We tested whether this translated to the functional level; i.e., a higher spine density may result in increased mEPSC frequency and this, in turn, may enhance the ability to induce LTP (Harms and Dunaevsky, 2007). However, this appeared not to be the case. Rather than HT, the condition of mild stress related to vehicle injection resulted in a higher frequency of both sEPSCs and mEPSCs, and a smaller area under the curve of sEPSCs; the latter could explain the reduced ability to induce LTP in these groups. The discrepancy between structural changes in dendrites and spines versus spontaneous EPSC frequency is unexpected, but emphasizes that presynaptic changes, e.g. after mild injection stress, are important for the overall outcome in functional terms. Presynaptic effects of stress or corticosterone are indeed well-documented (Karst et al., 2005, 2010), also in the mouse DG (Pasricha et al., 2011), although these were never investigated with this particular paradigm and at this age. The data furthermore illustrate that the functional effect of mild injection stress cannot be extrapolated to effects of a high dose of corticosterone. This may relate to a bell-shaped dose-dependency for corticosterone, as indeed often observed (Diamond et al., 1992; Joëls, 2006) or, importantly, the fact that mild (injection) stress causes the release of several stress-related hormones in addition to corticosterone.

176

Corticosteroids and epileptogenesis

Potential implications for epileptogenesis The increase in morphological differences after HT when followed by corticosteroid and mild stress exposure, suggests that stress (hormones) aggravate or accelerate epileptogenesis. Epileptogenesis, i.e., the neurobiological processes leading to epilepsy, is not limited to the time before the onset of spontaneous seizures, but continues during the course of epilepsy, contributing to the progression of the disease (Pitkänen and Sutula, 2002; Williams et al., 2009). As none of the currently available anti-epileptic drugs can favorably modify the disease process, prevention or reduction of epileptogenesis remains one of the main challenges in the field of epilepsy (Baulac and Pitkänen, 2008, Kelley et al., 2009; Pitkänen, 2010) and could be beneficial for all patients, irrespective of the underlying pathology. Children with complex febrile seizures are of special interest, as they have a high risk to develop epilepsy –ranging from 21% after prolonged febrile seizures (Verity and Golding, 1991) to as much as 49% after prolonged seizures with focal semiology that re-occur within 24h (Annegers etal., 1987)– that is already identified at the very beginning of epileptogenesis, making them very suitable for early intervention. Large population-based studies, prospectively following children after complex febrile seizures and systematically documenting stress exposure, as well as epilepsy outcome, can provide more insight into the effects of stress on epileptogenesis in this human population. In conclusion, our results suggest that stress and stress hormones modulate epileptogenesis, indicating that stress reduction strategies and possibly even medication targeting the stress system may have a potential role in reducing epileptogenesis. As studying epileptogenesis apart from seizure frequency is difficult, especially in humans, animal studies could provide valuable information on the effects of stress reduction on epileptogenesis, for example by studying effects of an enriched environment during the latent phase of epileptogenesis on seizure outcome.

ACKNOWLEGDEMENTS We acknowledge the expert help and technical assistance of H. Karst, R.A. Sarabdjitsingh and B. Jongbloets. Moreover, we thank the practical assistance of B. Visser, V. Bonapersona and T. Kliest in neuronal reconstructions for dendritic complexity or spine analysis, S. Vugts in optimizing neurogenesis stainings, M. Sep in Ki67 staining and quantification, and M.A. Pet in DCX quantification. This work was supported by a grant of the UMC Utrecht Alexandre Suerman Stipendium.

177

Summary and general discussion

CHAPTER

Summary and general discussion

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

SUMMARY Epilepsy is one of the most common chronic diseases in childhood, characterized by the enduring predisposition to generate epileptic seizures. Children with epilepsy and their parents often report seizures precipitated by stress. Previous studies in humans as well as rodents have shown that acute stress can provoke seizures and chronic stress can increase seizure frequency in some patients with epilepsy. Furthermore, chronic stress and stress experienced early in life, have been shown to change brain morphology and function in rodents and increase seizure risk in individuals without epilepsy (chapter 2). In order to increase our understanding of the pathophysiological mechanisms that underlie the effects of stress on seizures and epileptogenesis in childhood epilepsy, we performed a variety of studies, which are described in this thesis. In Part I we evaluated the extent of stress sensitivity of seizures in childhood epilepsy and its associations with patient and disease characteristics. We showed that stress sensitivity of seizures was reported in half of children with epilepsy and pertained to seizures precipitated by acute stress and an increase in seizure frequency during periods of stress. Children and their caregivers reported seizures precipitated by stress caused by negative events, such as tests at school, arguments at home and being bullied as well as positive events, such as birthday parties and the celebration of Saint Nicholas. To get a clue on the pathophysiological mechanisms underlying stress sensitivity of seizures, we assessed the relation between stress sensitivity of seizures in childhood epilepsy and several patient and disease characteristics. While we did not find differences in demographic or epilepsy characteristics between children with and children without stress-sensitive seizures, stress sensitivity of seizures was more common in children who reported a larger number of experienced negative life events. As early life stress-exposure is known to have long-term effects on hypothalamic-pituitary-adrenal (HPA-) axis regulation, these results suggested a mediating role of stress hormones in stress sensitivity of seizures (chapter 3). We found additional support for the self-reported relation between stress and childhood epilepsy using a big data approach, where we analyzed the relation between epilepsy search behavior on the Internet – an indirect estimate of seizure incidence – and the stressful period around the celebration of Saint Nicholas’ birthday; a historical Dutch children’s festivity. We found a country-specific increase in search queries for epilepsy during the Saint Nicholas period in the Netherlands, which provided circumstantial evidence for the relation between stress and seizure occurrence (chapter 4). Next, we assessed the prevalence of sensory modulation disorders among children with epilepsy and their relation with stress sensitivity of seizures, as stressors can be associated with excessive exposure to sensory stimuli of different modalities. We showed sensory modulation disorders to be highly frequent in childhood epilepsy, as they were reported in half of all children. Furthermore, acute stress-precipitated seizures were associated with reported lowered thresholds for sensory stimuli, suggesting that sensory overload might also contribute to the seizure-precipitating effects of acute stress (chapter 5).

182

Summary and general discussion

Part II focused on the hormonal basis of stress sensitivity of seizures. To assess the relation between stress sensitivity of seizures and the biological stress response, children with epilepsy and healthy controls were exposed to a standardized acute psychosocial stressor. We demonstrated an altered regulation of the hormonal stress response in children with stress-sensitive seizures, with a decreased cortisol response to stress compared with children without stress-sensitive seizures and healthy controls, both when stress sensitivity was based on a retrospective self-report and when based on prospective diary data (chapter 6). These results confirmed the hypothesized relation between stress hormone regulation and stress sensitivity of seizures. Stress hormone levels do not only increase in response to stress, they also vary throughout the day. Therefore, we wondered whether circadian and ultradian variability affects epileptic activity and seizure susceptibility. In a systematic review of the literature, we found that diurnal seizure occurrence shows similarities to the circadian rhythm of cortisol, with a sharp rise in the early morning, a gradual decline during the day and a state of quiescence at night (chapter 7). Next, we assessed the relation between basal fluctuations in cortisol and epileptiform activity by simultaneous electroencephalographic and cortisol monitoring in adult patients with focal epilepsy. We showed that cortisol levels are positively associated with interictal epileptiform EEG activity, although large individual differences exist. The relation between cortisol and epileptiform discharges was higher in patients with self-reported stress sensitivity of seizures. These results are compatible with the idea that stress hormones influence epileptiform EEG activity in humans, also in non-stress conditions (chapter 8). In Part III, the effects of stress and stress hormones on epileptogenesis were evaluated in an animal model. To initiate epileptogenesis, experimental febrile seizures were induced in young mice. In the following weeks, mice received repetitive corticosterone or vehicle injections (mild stress), or were left undisturbed, after which alterations in morphological and functional parameters were examined in the dentate gyrus, a subarea of the hippocampus. In this latent phase of early epileptogenesis, little morphological alterations were observed in animals that were not injected. However, after corticosteroid or mild stress exposure, epileptogenesis was associated with a lower number of immature cells, increased cell proliferation, and increased dendritic length and spine density, while functional glutamatergic transmission was not affected. We concluded that corticosterone and mild stress during epileptogenesis increased morphological changes in the dentate gyrus, that did not translate to the functional level, partly supporting the hypothesis that early life stress stimulates epileptogenesis (chapter 9). In conclusion, we investigated the effects of stress on epilepsy with a variety of approaches. We showed that the effect of stress on seizures relates to early life stress exposure, sensory modulation and hormonal regulation, and that stress and stress hormones aggravate epileptogenesis. These studies provide a first step towards elucidating the mechanisms underlying the relation between stress and epilepsy.

183

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

GENERAL DISCUSSION In this thesis, we aimed to increase our understanding of the pathophysiological mechanisms that underlie the effects of stress on epilepsy. More specifically, our objectives were to (1) improve knowledge on stress sensitivity of seizures in childhood epilepsy in relation to patient and disease characteristics; (2) provide a first step towards elucidating hormonal mechanisms related to stress sensitivity of seizures; and (3) explore the effects of stress on epileptogenesis. In this discussion, an integrative perspective on the effects of stress on epilepsy is provided. Next, methodological considerations are critically discussed. To conclude, the findings are translated into clinical perspectives and future directions.

Stress and epilepsy, what have we learned? Part I. Which children have stress-sensitive seizures? For centuries, it has been known that in patients with epilepsy, seizures can be triggered by endogenous or exogenous factors (Gowers, 1964). Although many different seizure precipitants have been reported, stress is the precipitant most often reported by patients with epilepsy (Hayden et al., 1992; Antebi and Bird, 1993; Hart and Shorvon, 1995; Spatt et al., 1998; Spector et al., 2000; Frucht et al., 2000; Nakken et al., 2005; da Silva Sousa et al., 2005; Sperling et al., 2008; Fang et al., 2008; dos Santos Lunardi et al., 2011; Ferlisi and Shorvon, 2014; Wassenaar et al., 2014). This self-reported relation between stress and seizures has been confirmed by prospective diary studies in adults that investigated seizure occurrence in relation to self-reported stress levels (Temkin and Davis, 1984; Blanchet and Frommer, 1986; Webster and Mawer, 1989; Neugebauer et al., 1994; Haut et al., 2007), or in relation to general stressors, such as a natural disasters or a terrorist attack (Swinkels et al., 1998; Klein and van Passel, 2005). Even more objectively, animal studies have shown that stress early in life can lower the threshold for the induction of epileptic seizures, and can increase seizure severity (see chapter 2). To get a grasp of the mechanisms underlying stress sensitivity of seizures, we investigated stress sensitivity of seizures in childhood epilepsy in relation to patient and disease characteristics. We showed that stress sensitivity of seizures in childhood epilepsy is common among all subtypes of patients with epilepsy, as it was not related to any demographic or epilepsy characteristic such as age, epilepsy duration or epilepsy type in multivariable analysis. Interestingly, stress sensitivity of seizures was more common in children who had experienced a larger number of negative life events (chapter 3). Early life stress can cause hyper- or hypoactivity of the stress hormone axis, thereby changing stress hormone levels, both basal and in response to stress. The direction of the effect can differ between different types of stressors, developmental stages and species (see chapter 2). Stress hormone levels have been shown to influence neuronal excitability and seizure susceptibility in rodents (JoĂŤls, 2009). Therefore, the relation between stress sensitivity of seizures and

184

Summary and general discussion

previous negative life events is in line with a mediating role of the stress hormone regulation. This hypothesis was tested in part II. Nonetheless, there are clearly many other variables that influence neuronal excitability and seizure susceptibility. Some of these are also associated with stress exposure, such as an excess of sensory stimulation. As stress sensitivity of seizures is related to susceptibility for sensory overload (chapter 5), the latter might provide an additional pathway by which stress can influence seizure susceptibility. However, the relation between sensory modulation disorders and the effect of sensory stimuli on neuronal excitability in patients with epilepsy needs to be studied more extensively, e.g. in an experimental setting.

CONCLUSION I Stress sensitivity of seizures is common among all subtypes of children with epilepsy, especially in those with difficulties in sensory modulation and a history of early life stress.

Part II. What hormonal mechanisms underlie stress sensitivity of seizure? To provide a first step towards elucidating hormonal mechanisms responsible for stress sensitivity of seizures, we investigated differences in stress hormone regulation between children with and those without stress-sensitive seizures. The blunted cortisol response to stress in children with acute stress-sensitive seizures, compared with those without stresssensitive seizures and healthy controls, revealed an association between the seizureprecipitating effects of stress and HPA-axis activity (chapter 6). This supports the hypothesis that HPA-axis regulation modulates stress sensitivity of seizures. In addition, (adult) patients with self-reported stress-precipitated seizures showed a stronger relation between cortisol levels and epileptiform EEG activity (chapter 8). In general, increased stress hormone levels are associated with enhanced neuronal excitability and seizure susceptibility (but not always so, see chapter 2). Therefore, further studies are needed to unravel whether the blunted cortisol response to stress (1) diminishes the likelihood of stress-precipitated seizures in a stress-sensitive population and thus represents a compensatory mechanism; or (2) inadequately constrains the central stress response (including CRH activation), which subsequently drives the relation between stress and seizures. Stress hormones can be expected to influence neuronal excitability to some extent in all individuals, patients with epilepsy as well as healthy controls. Interestingly, the association between cortisol levels and epileptiform EEG activity not only differs between individuals, but both positive as well as negative correlations exist (chapter 8). Electrophysiological studies on the effects of corticosteroid application on neurotransmission in rodent brain slices consistently showed that the fast, nongenomic effects of corticosteroids result in increased excitability, while the delayed, genomic effects normalize this (JoĂŤls et al., 2012). Recent studies have shown that when brain tissue is exposed to repetitive peaks of corticosteroids, such as the ultradian pulses, different peaks can have differential effects on HPA-axis responsiveness and neuronal excitability, depending on amplitude and timing

185

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

after the previous pulse (Sarabdjitsingh et al., 2014). The relation between circadian and ultradian stress hormone fluctuations, stress and neuronal excitability is likely to be subject to such ‘metaplasticity’. Also, different stress mediators involved in the stress response interact (Joëls and Baram, 2009; Hermans et al., 2014). This suggests that in the complex system of the human body, the timing after stress exposure and the balance between the different stress mediators and receptors largely determine seizure susceptibility. As seizure occurrence reflects the sum of all different variables influencing neuronal excitability, the effect of stress on seizures should be considered in interaction with other seizure precipitants, seizure suppressors and pre-existing neuronal hyper-excitability, as well as variability in stress responsiveness based on earlier experiences and genetic variation. The effects of stress on epilepsy and the interaction with other variables are visualized in Figure 1. Although the studies described in this thesis provided a first step towards elucidating hormonal mechanisms underlying stress sensitivity of seizures, follow-up studies are needed to further unravel where exactly the HPA-axis dysregulation takes place, and how different stress hormones interact in relation to seizure susceptibility.

CONCLUSION II Patients with stress-sensitive seizures have an: • increased correlation between cortisol and epileptiform activity • altered stress hormone regulation, with a decreased cortisol response to stress

Part III. Does stress influence epileptogenesis? Besides the acute effects of stress on seizure susceptibility, stress can also influence the risk of the development of epilepsy. From studies in non-epileptic populations, we have learned that when stress is severe, chronic or experienced early in life, it can have long term effects on HPA-axis regulation, resulting in altered stress hormone levels in basal as well as stress conditions (Gunnar and Vazquez, 2001; Phillips, 2007; Seckl, 2008; Gunnar and Quevedo, 2008; Lupien et al., 2009). Stress hormones influence neuronal function and brain development. Therefore, when stress hormone levels are altered for a prolonged period of time, this affects brain morphology and function, especially in childhood when the brain is rapidly developing. A relation between stress exposure early in life, HPA-axis regulation and pathology has been shown for a wide variety of stress-related diseases such as depression, anxiety disorders, metabolic disorders and cardiovascular diseases (Phillips et al., 2006; Bellinger et al., 2008; Heim et al., 2008; Shively et al., 2009; Faravelli et al., 2012). The relation between chronic stress and many psychiatric disorders is well acknowledged even outside the medical field (for example, the relation between negative experiences and depression). Regarding epilepsy, animal studies consistently showed enhanced seizure susceptibility and kindling rates in rodents previously exposed to prenatal or neonatal stress (Beck and Gavin, 1976; Frye and Bayon, 1999; Huang et al., 2002; Lai et al., 2006; Salzberg et al., 2007; Lai et al., 2009; Jones et al., 2009; Gilby et al., 2009; Kumar et al., 2011;

186

Summary and general discussion

Figure 1. Effects of stress on epilepsy

Ahmadzadeh et al., 2011; Desgent et al., 2012; Qulu et al., 2012; Yum et al., 2012). These effects on seizure outcome can be understood through the neuronal and neuroendocrine changes caused by early life stress exposure. The alterations in neuronal excitability, neurogenesis and neuronal morphology that exist before induction of epileptogenesis, can change seizure vulnerability and, therefore, influence seizure latency, severity, frequency, and kindling rate (reviewed in chapter 2). In patients with hippocampal sclerosis, a frequent subtype of temporal lobe epilepsy, the process of epileptogenesis is initiated by frequent or prolonged febrile seizures early in life. It is unknown whether exposure to stress during childhood, after the possible induction of epileptogenesis by febrile seizures, alters the susceptibility to develop epilepsy. In the febrile status-induced model of temporal lobe epileptogenesis, we showed that stress directly affects pathways involved in epileptogenesis, as mild stress and stress hormone exposure during the latent phase following status epilepticus aggravated the morphological changes induced by epileptogenesis in a highly controlled animal model (chapter 9).

187

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

Figure 2. Effects of stress on epileptogenesis

As epileptogenesis is often a long term process that continues during the course of the disease, intervening in epileptogenesis is currently one of the main challenges in the field of epilepsy (Baulac and Pitkanen, 2008; Kelley et al., 2009; Pitkanen, 2010). As we showed epileptogenesis to be influenced by stress, intervention in this pathway might help to reduce the chance of developing epilepsy or its severity, thereby favourably influencing seizure outcome. However, to be able to intervene, the contributions of the different stress mediators and their interactions on epileptogenesis need to be resolved. Furthermore, it should be emphasized that, obviously, stress is not the only factor influencing epileptogenesis. For example, variability in genes coding for stress hormone receptors affect HPA-axis regulation and stress hormone levels (reviewed by (DeRijk and de Kloet, 2008; DeRijk, 2009; DeRijk et al., 2011) and also multiple childhood epilepsy syndromes have a genetic etiology (Berg et al., 2010). In addition, other environmental factors besides stress can also influence HPA-axis regulation and stress hormone levels as well as epilepsy risk (Figure 2).

CONCLUSION III Stress does not only increase seizure susceptibility, it also directly affects pathways involved in epileptogenesis.

An evolutionary perspective From an evolutionary perspective, the stress response serves to accomplish one goal: to maximize the likelihood of survival of a life-threatening experience. To reach that goal, stress induces several changes in the body. Increased neuronal excitability might for example enable faster processing of information and facilitate synaptic plasticity to generate a rapid enhancement in memory formation (Farmer et al., 2014). These changes are likely to increase the odds for survival this time (e.g., by activating the fight-or-flight response) and in the future (e.g., by influencing memory formation), making the enhanced neuronal

188

Summary and general discussion

excitability in response to acute stress evolutionary beneficial. However, this increase in excitability also lowers the threshold for seizures. In patients with epilepsy this additional increase in neuronal excitability in response to stress significantly increases the likelihood of seizure occurrence. When stress endures, or is experienced early in life, activation of the stress system can cause more permanent changes in stress hormone regulation and brain function. The altered basal levels of stress hormones increase the risk for various psychiatric and neurologic diseases, including epilepsy (McEwen, 1998; Lupien et al., 2009; Gunnar et al., 2009; Koe et al., 2009; Andersen and Teicher, 2009), especially in the developing brain. Therefore, chronic stress is often called maladaptive (Yehuda et al., 1993), although it is also argued that this depends on the future environment. In the developmental match-mismatch hypothesis, plasticity is considered to evolve to match an organism to its predicted environment. It is the mismatch between the phenotype and the later-life environment that increases the risk on diseases (Champagne et al., 2009; Gluckman et al., 2009). Therefore, the changes associated with early life stress might have negative consequences for brain functioning in a non-stressful situation, but potentially make the individual better equipped to deal with future stressors. The latter, however, may no longer apply when the organism is exposed to additional challenges, as in the case of diseases like epilepsy.

GENERAL CONCLUSIONS Stress affects epilepsy in multiple ways. Stress experienced early in life and chronic stress can lastingly alter brain morphology and function, especially in the developing brain. Stress also aggravates epileptogenesis by inducing alterations in specific pathways involved in epileptogenesis. Thirdly, stress exerts acute effects on seizure susceptibility (Figure 3). The studies reported in this thesis reveal part of the mechanisms underlying the effects of stress on the occurrence of seizures and on epileptogenesis, thereby increasing insight in the pathophysiological mechanisms underlying the effects of stress on (childhood) epilepsy. Further research is needed to resolve the complicated interaction between stress and epilepsy.

Figure 3. Stress influences epilepsy at multiple levels (adapted from chapter 2)

189

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

Methodological considerations To increase insight in the effects of stress on childhood epilepsy, we approached this topic from different viewpoints and performed a wide variety of studies in humans as well as animals. In this section, some considerations on the methodological decisions that were made are discussed.

General considerations Stress and epilepsy, a bidirectional relation In this thesis, we focused on the effects of stress on epilepsy. However, epilepsy itself can also be a cause of stress. Seizures are stressful per definition, as they are a threat to physiological homeostasis. Indeed, a postictal rise in stress hormone levels has been shown very consistently (as reviewed in chapter 2). Additionally, epilepsy is associated with psychological stress caused by the unpredictable occurrence of seizures as well as the social impact of a chronic disease. In childhood epilepsy, the disease is not only stressful for the patients themselves, but also for their caregivers (Hoare and Kerley, 1991; Carlton-Ford et al., 1997; Aytch et al., 2001; Wirrell et al., 2008; Wu et al., 2014). The early life stress caused by the recurrent seizures and the seizure risk in children with epilepsy is likely to influence HPA-axis regulation. Therefore, life events associated with epilepsy (i.e., hospitalization) were included in the determination of early life stress exposure (chapter 3 and 6). Unfortunately, when determining stress sensitivity of seizures it is impossible to differentiate the effect of stress caused by experiencing a seizure on the next seizure, from other, combined, seizure-precipitating factors, and the effect of stress caused by coping with epilepsy from duration of epilepsy and stress coping style. When determining stress sensitivity of seizures (chapter 3, 5, 6 and 8), stress caused by epileptic seizures was not taken into account. Manipulating and measuring stress When manipulating and measuring stress, the (anticipatory) stress induced by the experiment itself should always be considered. This anticipatory effect was clearly shown in our experiments by the increase in cortisol levels and measures of sympathetic activation in the first (pre-stress induction) measurement in children (chapter 6), as well as the increased corticosterone levels with the order of decapitation in mice (chapter 9). We accounted for this in our human stress experiment by including a specific relaxing period into the protocol (chapter 6) and taking saliva instead of blood samples (as the latter would be more stressful) to determine cortisol levels (chapter 6 and 8), and in our animal studies by determining measures that were most likely to be influenced by acute stress in the animals that were decapitated first (chapter 9). However, the acute and longer lasting (genomic) effects of anticipatory stress might still have influenced some of our outcome parameters, such as the diurnal cortisol rhythm (chapter 8), and stress associated with the hyperthermia / normothermia experiment (chapter 9). Stress associated with experimental procedures is present in the majority of studies involving living subjects (e.g., humans and

190

Summary and general discussion

rodents). Generally, this receives little attention. As the stress response interacts with various systems in the body, the influence of stress on outcome measures should always be considered in any study design and interpretation of the results, especially when studying general processes, time differences and ‘baseline’ or ‘resting state’ measures. As mentioned in the General Introduction, stress is a broad term that involves many different processes, including activation of the HPA-axis and the sympathetic nervous system. Since we were interested in the effects of real life stressors on epilepsy, we evaluated stress exposure in children by having them report stressful events in daily life, as well as by exposing them to a standardized stressor (chapter 6), both capturing the multidimensional stress response. As stress is characterized by HPA-axis activity, and corticosteroids are the end product of the HPA-axis, we focused on corticosteroids both when measuring HPAaxis activity (chapter 6 and 8) and when applying exogenous stress hormones (chapter 9). However, it should be mentioned that many other neuromodulators are involved in the stress response, such as noradrenaline, corticotrophin releasing hormone (CRH), adrenocorticotropic hormone (ACTH), dopamine and serotonin, and the interaction between different stress-sensitive neuromodulators is essential for the brain response to stress (Joëls and Baram, 2009; Hermans et al., 2014). As the effects of stress mediators on the brain, as well as their interaction, are region and time specific, the exact mechanisms by which (1) the interplay between all the different neuromodulators involved in the stress response influences different cell types in different brain regions; (2) different cells and brain areas interact; and (3) these processes influence total brain functioning in response to stress; are not easy to unravel (Figure 4). Understanding the coordination of the different stress mediators across multiple levels will improve our understanding of individual and context specific differences in responses to stressors.

Study populations: heterogeneity When studying humans with or without epilepsy, a large variability in genetic as well as environmental factors exists between study subjects. On top of that, epilepsy is a very heterogeneous disease, for example with respect to underlying etiology, seizure semiology, seizure frequency, comorbidities and given treatments. The study populations described in chapter 3 to 9 may be considered heterogeneous with respect to patient as well as disease characteristics. Interestingly, this heterogeneity in study population enabled us to investigate the effects of a wide variety of patient and disease characteristics on outcome measures such as stress sensitivity of seizures (chapter 3, 5, 6 and 8), sensory modulation (chapter 5) and stress-responsiveness (chapter 6), which provided valuable clues in our search for the pathophysiological mechanisms underlying stress sensitivity of seizures. Furthermore, results obtained by studying this heterogeneous population of patients might be considered to be more representative of the general population of patients with epilepsy and thus increase the likelihood of reproducibility in other epilepsy cohorts. However, this large variation also comes with limitations. First, although we did not find differences in the relation between stress and epilepsy between epilepsy subtypes in multivariable analysis in most of the studies (chapter 3, 5, 6 and 8), not all subtypes were

191

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

well represented in our studies and we cannot exclude that some differences exist. Second, as we studied subjects with active epilepsy, most of them used (different combinations of) anti-epileptic drugs. Although these drugs are expected to interfere with many brain processes, treatment type did not significantly relate to outcome measures in multivariable analysis. Third, the large inter-individual variation can also confound the results. To mitigate this, we controlled for variables that are known to affect outcome measures, such as the effects of age on physiological measures (chapter 6), and the effects of developmental psychopathology on sensory modulation (chapter 5), using multivariable analysis. Fourth, the considerable heterogeneity in genetic and environmental factors between subjects, poses the risk that reported associations are actually driven by variables that were not included in the analysis. Finally, we assessed this risk by comparison of baseline characteristics, and used multivariable analysis to determine whether or not associations existed irrespective of other variables (chapter 3, 5, 6 and 8), but it is important to realize that, in observational studies like these, no claims can be made considering causality.

Study populations: selection We started with the clinical observation that stress sensitivity of seizures in children with epilepsy is often reported by parents, and therefore decided to focus most studies described in this thesis on childhood epilepsy. As children and adults differ in physiology, research questions in children can often not be answered by studying an adult population. For example, epilepsies in children differ from those in adults regarding seizure types,

Figure 4. Effects of stress on outcome measures are influenced by many different mediators

192

Summary and general discussion

syndromes, etiology and seizure frequency (Forsgren et al., 2005). Furthermore, the effects of stress hormones on brain morphology and function are age-dependent, in humans and rodents (Veldhuis et al., 2013; Klein and Romeo, 2013; Loi et al., 2014). However, the age limit of childhood at the age of 16 or 17 years is arbitrary and, in our studies, mainly reflects the population that is treated within the child neurology department, or the age range for which questionnaires were validated, rather than a biologically based rationale. As the acute effects of stress and stress hormones on seizure susceptibility that we reported in children (chapter 3 to 6) are consistent with results previously obtained in adults (chapter 2), and studies in adults are easier to perform from an ethical and pragmatic perspective, we decided to study the effects of cortisol on epileptiform EEG activity in an adult population (chapter 8). In all studies where study subjects can actively opt-in or opt-out (chapter 3, 5, 6 and 8), a certain amount of selection bias cannot be prevented. Subjects who recognize themselves in the problem that is studied might be more willing to participate (Hoeymans et al., 1998), and also personality characteristics, academic background and financial status (in case of a monetary incentive [chapter 6]) have been shown to play a role in behavior regarding study participation (Porter and Whitcomb, 2005). Comparison of responders with nonresponders can indicate the amount of selection bias, especially in well characterized patient populations where detailed information is available for both (chapter 3 and 5). However, in most studies, especially when studying healthy controls (chapter 6), detailed information on non-responders is usually limited or even absent. When including healthy controls in our studies (chapter 6), we preferably recruited them as peers of included children with epilepsy to limit this selection bias.

The validity of self-report In most of the studies presented in this thesis, questionnaires were used to estimate stress sensitivity of seizures, as well as some other patient and disease characteristics (chapter 3, 5, 6 and 8). Although questionnaires are a time-efficient way to gather information, answers provided might be influenced by individual factors such as interpretation, personality, emotional state, and recall by both patient and caregivers. Previous research showed that, overall, study subjects give quite accurate information on questionnaires (Walsh, 1967). To improve the validity of self-report we used validated questionnaires wherever possible, and to minimize recall bias, only children with active epilepsy were included so that questions could be answered about the current situation. As stress is a subjective experience, the validity of self-reported stress and stress sensitivity of seizures can be debated. However, while stress might mean different things to different people in different situations, the resulting stress response, i.e., the activation of the sympathetic nervous system and the HPA-axis, is very consistent, as are the results of this stress response on processes throughout the body (Selye, 1936). Previous studies showed stress self-report to relate well to biomarkers of stress and stress-related diseases (Borders et al., 2010; Masood et al., 2012; Klopp et al., 2012). Also in our studies, reported stress scores in response to psychosocial stress were associated with cortisol levels (chapter 6), and self-reported stress sensitivity of seizures

193

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

related to the effects of cortisol on epileptiform EEG activity (chapter 8) as well. In healthcare, patient self-report, or parental-report in children, is generally considered an extremely valuable source of information, not only because it is often the only source, but also because the human brain is very good at integrating complex information (Tononi et al., 1998), and recognizing patterns and meaningful events in noisy background data, which is also the main reason why expert assessment is often valued above automated detection (as in chapter 8).

Animal models Because of the large inter-individual variability, the low level of controllability of the environment, the long life span, and ethical limitations to invasiveness, some hypotheses are hard to study in humans and reproducibility is limited. In these situations, animal models can provide valuable insights (chapter 9). In epilepsy research, animal models provide the opportunity to study the time-line in (early) epileptogenesis, even before the onset of spontaneous seizures. This helps to unravel the causes and consequences of epilepsy, and thus to identify critical processes that are potential candidates for intervention. However, extrapolation of the results to other epilepsy types, stressors, developmental stages, brain areas and species should be done with care. Some strengths and weaknesses of the animal research described in this thesis are discussed below. Study populations Although physiology of most mammals is quite comparable, rodents are obviously not little humans. That is exactly one of the major benefits of animal research, as the shorter life span makes it possible to perform prospective research within a limited amount of time, and rodents can be housed in a research facility (although the degree to which the current animal housing facilities in research labs meet their needs is debatable). Furthermore, inter-individual variability can be severely reduced by the use of inbred strains with nearly identical genotype, and the ability to control environmental circumstances during the total life-span of the animal (with the exception of social hierarchy in group-housed animals), resulting in â&#x20AC;&#x2DC;cleanerâ&#x20AC;&#x2122; data. In addition, choices were made regarding the age and sex of the animals: we focused on male animals as, in general, effects of stress and epileptogenesis are larger in males (Jones et al., 2014), and tested them at 25 days of age (before the reproductive age) to avoid stressful effects of weaning and enable comparison with previous studies (Notenboom et al., 2010). These decisions are likely to influence outcomes, as the effects of seizures and stress hormones on the brain are age- and sex dependent (Veldhuis et al., 2013; Loi et al., 2014; Jones et al., 2014). All these decisions that were made to decrease variability, lower generalizability of the results. Selection of a model To study stress and epilepsy in animals in a controlled way, different models exist with varying similarity to the human situation. As we aimed to study the effects of stress and stress hormones on epileptogenesis during childhood, we used a pediatric epilepsy model

194

Summary and general discussion

in which the period of epileptogenesis can be studied relatively apart from the damage and compensatory mechanisms induced by spontaneous seizure activity (Bender et al., 2004). The experimental febrile seizure model is quite unique in this aspect and has also good face validity to the human situation (Dube et al., 2007; Dube et al., 2009), in which children with complex febrile seizures have an increased risk to develop temporal lobe epilepsy (Annegers et al., 1987; Verity and Golding, 1991), especially in comparison to (1) adult epilepsy models where the chemically or electrically induced status epilepticus that initiates the process of epileptogenesis is by itself associated with severe brain damage (Loscher, 2011), and (2) genetic models in which epileptic seizures are often accompanied with abnormal (brain) morphology (Burgess, 2006). Also, a wide range of experimental stress models, and models altering stress hormone levels, has been described (for an overview, see (Heinrichs and Koob, 2006)). What should be considered in these stress models is the extent to which their effects are caused by variables other than stress, e.g., pain, deprivation of food and maternal care, or the release of endorphins. These factors may play a role in the foot shock, maternal separation/deprivation and acute physical stress models, respectively. These aspects might explain some of the differential effects that have been shown of different acute stressors on seizure susceptibility (reviewed in chapter 2). Therefore, we decided to use a model of repetitive exogenous corticosterone administration. This pharmacological model for the effects of corticosteroids can obviously not capture all the effects of the complex stress response and does not necessarily mimic the role of corticosteroids in the stress response, as it is not accompanied by similarly high levels of all the other stress mediators. However, it can provide cleaner mechanistic insights, which was the aim of this study. Interestingly, we found that the effects of high dosages of injected corticosteroids and of mild stress induced by vehicle injection, were similar on many outcome parameters (chapter 9), which leaves the exact stress mediator(s) responsible for these effects open for debate and further study.

Selection of outcome measures Another opportunity of animal research is that data can be obtained with more invasive techniques. By examining brain slices or individual cells of animals after exposure to certain treatments, specific mechanisms can be studied and provide valuable insights into isolated processes. We selected outcome measures of neuronal morphology as well as functioning that have shown to be altered in epilepsy (Ward, 1961; Belichencko et al., 1992; Multani et al., 1994; Lemmens et al., 2005; Kwak et al., 2008; Nishimura et al., 2008; Notenboom et al., 2010; Ouardouz et al., 2010; Gibbs et al., 2011; Singh et al., 2013), as well as after stress (reviewed by Lupien et al., 2009), and focused on a brain area affected by experimental febrile seizures (Bender et al., 2003; Lemmens et al., 2005; Kwak et al., 2008; Notenboom et al., 2010) as well as by stress hormones (JoĂŤls, 2007): the hippocampus. We decided to assess multiple different morphological and functional parameters in the exact same model, instead of focusing on one parameter and studying this in various areas, hoping that the combination of outcome measures would provide unique insights into the underlying mechanism. However, there are still multiple brain areas (e.g., CA1-3) and mechanisms

195

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

(e.g., GABA-ergic transmission) that we did not study. Therefore, even this combination of outcome measures does not capture all aspects of epileptogenesis. Furthermore, we studied the latent phase of early epileptogenesis to mitigate the damage and compensatory mechanisms caused by spontaneous seizures. However, not all animals in the model will develop epilepsy (Dube et al., 2006; Kwak et al., 2008; Koyama et al., 2012). In this latent phase, the epileptic fate of the animals is unknown, which is a source for inter-individual variability. Long term prospective studies are needed to assess the consequences for seizure outcome.

Clinical implications Although patients with epilepsy and their parents often report the impression that seizures are precipitated by stress, until now many physicians believed that this reflected the subjectively biased recall of patients trying to get a grip on their unpredictable seizures, rather than a biological process. Currently, evidence supporting the relation between stress and seizure susceptibility is accumulating, including objective data obtained from experimental human and animal studies. Therefore, we would like to emphasize that seizure susceptibility can be influenced by stress, at least in a subset of patients.

Counseling and stress reduction strategies The increased knowledge on the relation between stress and epilepsy can improve counseling of patients with epilepsy and their caregivers. Especially in children with therapy-resistant seizures, adequate information and psychosocial guidance to children and their caregivers can (1) help them to self-manage or cope with the chronic disease and often unpredictable occurrence of seizures, (2) benefit their adherence to treatment and restrictions, and (3) improve their quality of life (Aytch et al., 2001; Galletti and Sturniolo, 2004). In addition, proper parental counseling has been shown to reduce family stress (Dean, 2011). However, based on the existing literature, so far no conclusions can be drawn on the efficacy of stressreduction in the treatment of epilepsy. Most patients with epilepsy are treated with anti-epileptic drugs. However, approximately 25% of patients with epilepsy are drug-resistant (Shinnar and Pellock, 2002; Berg et al., 2006; Kwan et al., 2010), and anti-epileptic drugs have many side-effects, including detrimental effects on cognitive performance and behavior, especially in the immature and developing brain (Oostrom et al., 2003; Hermann et al., 2012). Stress reduction strategies might be a safe and low-cost additive treatment strategy for children with stress-sensitive seizures, to improve seizure-control with limited side-effects (Polak et al., 2012). In potential, these behavioral interventions might prevent an increase of seizures after stress, and be beneficial for the prognosis, not only by influencing seizure frequency and subsequently anti-epileptic drug load, but also by influencing HPA-axis regulation. Without being advised to do so by their treating physicians, over half of patients with self-reported stress-sensitive epileptic seizures have tried stress reduction methods, with very high self-

196

Summary and general discussion

reported success rates on seizure frequency (88%) (Privitera et al., 2014). Beneficial effects of stress-reduction on seizure frequency have even been described by patients not reporting stress sensitivity of seizures (Privitera et al., 2014). In observational studies, positive effects on seizure frequency have been reported of mindfulness, yoga, cognitive behavioral therapy and biofeedback training (Rousseau et al., 1985; Dahl et al., 1987; Puskarich et al., 1992; Deepak et al., 1994; Panjwani et al., 1996; Rajesh et al., 2006; Lundgren et al., 2008; Sathyaprabha et al., 2008). However, the effects of stress reduction on seizure frequency have not systematically been studied and clinical trials are warranted (Polak et al., 2012). Stress reduction might be tried as additional treatment strategy, especially in patients who recognize stress as a seizure precipitant themselves. However, we are awaiting further evidence, before it can be widely implemented as general â&#x20AC;&#x2DC;add-onâ&#x20AC;&#x2122; therapy in epilepsy.

Development of new anti-epileptic drugs Increased knowledge on the fundamental mechanisms underlying the effects of stress on neuronal excitability might also provide new targets for anti-epileptic drugs. Currently, corticosteroids and ACTH agonists are already used in the treatment of specific childhood epilepsy syndromes that are assumed to have a (partial) inflammatory etiology or in which inflammation is considered a result or epiphenomenon of epileptic seizures or epileptiform activity, contributing to further seizure generation (reviewed in chapter 2). Some observational, retrospective as well as prospective, studies have also shown beneficial effects in various other types of epilepsy, both structural and genetic (Chutorian et al., 1968; Lagenstein et al., 1978; Willig and Lagenstein, 1980; Pentella et al., 1982; Dooley et al., 1989; Charuvanij et al., 1992; Oguni et al., 2002; Kalra et al., 2003; Mall et al., 2003). Together with the reported effects of stress and stress hormones on neuronal excitability and seizure susceptibility, this suggests that HPA-axis directed medication might also exert direct anticonvulsive effects. Prevention or reduction of epileptogenesis is currently one of the main challenges in the field of epilepsy (Baulac and Pitkanen, 2008; Kelley et al., 2009; Pitkanen et al., 2010). As our animal studies suggest that stress and stress hormones aggravate epileptogenesis (chapter 9), medication influencing the stress hormone regulation might hypothetically also be anti-epileptogenic, although possibly in a selected group of patients only. The dysregulation of the HPA-axis in stress-sensitive epilepsy might be another target for treatment. Dysregulation of the HPA-axis has been shown to be important in the pathophysiology of many other brain disorders, such as depression and anxiety (reviewed by de Kloet et al., 2005; Lupien et al., 2009; Faravelli et al., 2012). It is even considered the main mediator between stress experienced early in life or chronic stress and an increased risk on a variety of psychopathology, in interaction with genetic background (Cicchetti and Toth, 2009). Psychosocial interventions or drugs intervening with the HPA-axis have been shown to reduce HPA-axis dysfunction as well as disease symptoms after early life stress and in diseases such as depression and post-traumatic stress disorder (PTSD) (Schatzberg et al., 1985; Zobel et al., 2000; Belanoff et al., 2002; HĂ¸yberg et al., 2002; Flores et al., 2006; Oomen et al., 2007; Anacker et al., 2011; Slopen et al., 2014). Mifepristone,

197

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

a potent glucocorticoid receptor (GR) antagonist, has received special interest, as it has been suggested to be able to restore or â&#x20AC;&#x2DC;resetâ&#x20AC;&#x2122; HPA-axis activity and GR/MR balance to normal ranges, resulting in persistent beneficial effects on HPA-axis regulation as well as outcome (DeBattista, 2006; DeBattista and Belanoff, 2006; Flores et al., 2006). These manipulations of HPA-axis regulation might also be beneficial in epilepsy. However, as stress hormones play a role in various physiological processes, changes in stress hormone levels and stress hormone regulation can have significant adverse effects in both the short and the long term, for example on immunity and metabolism. Possibly, this explains why clinical trials on the use of HPA-axis modulators for the treatment of major depressive disorder and anxiety were unsuccessful so far (Nihalani and Schwartz, 2007; Koob and Zorrilla, 2012). Additionally, it has been argued that higher doses of these modulators need to be administered, to overcome the consequences of the blood-brain barrier (Schatzberg and Lindley, 2008).

Future directions The studies described in this thesis provide a first step towards elucidating the pathophysiological mechanisms behind the effects of stress on seizure susceptibility and epileptogenesis. These results raise several questions, which are partly addressed in previous chapters of this thesis and previous sections of this discussion. This section will provide broader recommendations for future research topics that need to be resolved to elucidate the relation between stress and epilepsy.

Endogenous stress hormone levels and seizure susceptibility In the studies presented in this thesis, evidence for a relation between stress hormone levels and seizure susceptibility was obtained by focussing on cortisol as a measure of HPA-axis activity. As stated in the previous sections of this discussion, many other stress hormones and mediators are involved in the stress response, and whether the effects depend on cortisol, on other stress mediators, or their interaction, still needs to be resolved. A next step to unravel HPA-axis responsiveness in relation to stress sensitivity of seizures would be to measure endogenous levels of ACTH in response to stress, in addition to cortisol and sympathetic responses, and to perform pharmacological dexamethasone-suppression CRHstimulation tests in patients with and without stress-sensitive seizures. Simultaneous EEG registration or transcranial magnetic stimulation (TMS) could provide insights into the acute effect of stress and stress hormones on epileptiform EEG activity and cortical excitability, and might provide a more objective measure of excitability in response to stress. Further, continuous monitoring of stress mediators (HPA-axis, e.g., by using automatic collection systems [Bhake et al., 2013], as well as sympathetic measures) and EEG monitoring over multiple days, could provide more insight into the interaction between the circadian and ultradian variation in stress hormone levels, stress, and the development of seizures.

198

Summary and general discussion

As epilepsy is not only characterized by neuronal hyperexcitability, but also by excessive and abnormal synchronization, stress might not only affect epilepsy by changing excitability at the cellular level, but also by increasing interneuronal communication, and on a larger scale, by influencing functional network organization. Changes in functional brain networks have been found both in individuals exposed to stress (reviewed by Hermans et al., 2014) and patients with epilepsy (reviewed by Kramer and Cash, 2012; van Diessen et al., 2013; van Mierlo et al., 2014). Therefore, it would be interesting to see whether the shift in network organization during stress influences seizure susceptibility by assessing stress and stress hormone levels, functional network measures and epileptiform EEG activity simultaneously. While the effects of (chronic) stress on HPA-axis regulation clearly differ between children and adults, the effects of acute stress on seizure susceptibility are probably more comparable. Studying the latter in adult populations might therefore be preferred, first of all because of ethical issues with respect to providing consent for invasive procedures with limited direct benefit, but also because adults can be better instructed, which might reduce unintentional experimental stress and study withdrawal.

The interplay between different stress hormones It has become increasingly clear that the interactions between many stress mediators are crucial for the brain response to stress. Although the effects of different hormones and neuropeptides involved in the stress response are region and time specific, substantial overlap allows for significant interactions (Joëls and Baram, 2009; Hermans et al., 2014). Especially for corticosteroids and noradrenaline synergistic effects on cell excitability have been shown in a time dependent manner (Joëls and Baram, 2009; Krugers et al., 2012). Similar interactions probably exist between other stress mediators. Further studies are needed to unravel the specific contributions of various stress hormones or their receptors in the effects of stress on neuronal excitability and seizure susceptibility. These studies could for example expose patients with epilepsy and healthy controls to (1) stress, (2) administering selective agonists instead of stress, or (3) stress in combination with receptor antagonists, while measuring (1) interictal epileptiform discharges or high frequency oscillations on EEG recordings in patients with epilepsy, or (2) cortical excitability with TMS in patients with epilepsy as well as healthy controls. Also animal research can be extremely valuable in unraveling the exact mechanisms underlying the effects of stress on epilepsy from a cellular to a complex-system level. Animal studies, both in- and ex-vivo, could clarify stress hormone interactions, and also dosedependency and time windows for stress effects in a controlled environment. To optimize the benefits of translational research in general, this should be performed in a continuing cycle, where studies on animals are used for questions that can’t be answered in humans and studies on brain slices or cells to answer questions that can’t be answered by studying the total system. The results of these experiments should guide further research in larger and more complex systems and in the real life human situation, where the cycle should start over again.

199

SUMMARY AND GENERAL DISCUSSION | CHAPTER 10

The interplay between stress and other variables The effects of stress on brain functioning does not only depend on the timing and concentration of the various stress mediators, but also on other variables that influence the stress response and neuronal excitability. These comprise both individual specific ‘background’ variables (i.e., genetics, epigenetics, developmental stage), and factors that are timing or context specific (i.e., presence of other seizure precipitants and seizure suppressors). Regarding the first, gene-environment interactions are well known to be important in the stress response and in the development of stress-related diseases. For example, several haplotypes of genes coding for stress hormone receptors have been associated with HPA-axis regulation and vulnerability for or resilience to a wide variety of neuropsychiatric disorders (reviewed by DeRijk and de Kloet, 2008). The sustained effects of early life stress are thought to be (partly) mediated by epigenetic regulation (de Kloet et al., 2005; Meaney and Szyf, 2005). The genetic and environmental influences on disease risk are probably not disease-specific. Prospective cohort studies in high risk populations (e.g., defined by Doyle et al., 2014), starting prenatally, determining haplotypes of genes for stress hormone receptors and their known regulators, prospectively assessing DNA methylation of the same genes, (early life) stress exposure, stress response regulation, and disease onset, might reveal a generic risk profile for stress-related diseases. Regarding the second, neuronal excitability is influenced by many variables, including stress hormones. Although the definition of epilepsy as the experience of unprovoked (i.e., spontaneous) seizures (Fisher et al., 2014), might suggest that seizure occurrence is unpredictable, increased knowledge of the variety of factors that affect neuronal excitability and synchronicity might lead to a model that can predict seizure occurrence. Ultimately, perfect seizure prediction for individual patients could change treatment regimens and have a major impact on quality of life. Because of all these variables and their interactions, the effects of stress on total brain functioning are not easy to predict. Large ‘brain’ initiatives such as the U.S. Brain Research through Advancing Innovative Neurotechnologies (BRAIN, www.nih.gov/science/brain) and the Human Brain Project (www.humanbrainproject.eu/), aim to integrate the findings from different research groups, experimental paradigms, models, and research fields. Although these large projects encounter many technical and conceptual difficulties (Rose, 2014; Fregnac and Laurent, 2014; The Lancet, 2014), such integration of knowledge is essential to provide breakthroughs in understanding of the brain in its total complexity.

200

201

Addendum

References Samenvatting Dankwoord List of publications Curriculum vitae

206

References

REFERENCES Abbott RJ, Browning MC, Davidson DL (1980) Serum prolactin and cortisol concentrations after grand mal seizures. J Neurol Neurosurg Psychiatry 43:163-167. Abubakr A,Wambacq I (2003) The localizing value of auditory event-related potentials (P300) in patients with medically intractable temporal lobe epilepsy. Epilepsy Behav 4:692-701. Adams F (1939) The genuine works of Hippocrates. Baltimore: Williams & Wilkins. Ahmadzadeh R, Saboory E, Roshan-Milani S, Pilehvarian AA (2011) Predator and restraint stress during gestation facilitates pilocarpine-induced seizures in prepubertal rats. Dev Psychobiol 53:806812. Ajmone Marsan C, Zivin LS (1970) Factors related to the occurrence of typical paroxysmal abnormalities in the EEG records of epileptic patients. Epilepsia 11:361-381. Alfarez DN, Joëls M, Krugers HJ (2003) Chronic unpredictable stress impairs long-term potentiation in rat hippocampal CA1 area and dentate gyrus in vitro. Eur J Neurosci 17:1928-1934. Alfarez DN, Wiegert O, Joëls M, Krugers HJ (2002) Corticosterone and stress reduce synaptic potentiation in mouse hippocampal slices with mild stimulation. Neuroscience 115:1119-1126. Allendorfer JB, Heyse H, Mendoza L, Nelson EB, Eliassen JC, Storrs JM, Szaflarski JP. Physiological and cortical response to acute psychosocial stress in left temporal lobe epilepsy – A pilot crosssectional fMRI study. Epilepsy Behav 26: 115-123. Aminoff MJ, Simon RP, Wiedemann E (1984) The hormonal responses to generalized tonic-clonic seizures. Brain 107 ( Pt 2):569-578. Anacker C, Zunszain PA, Cattaneo A, Carvalho LA, Garabedian MJ, Thuret S, Price J, Pariante CM (2011) Antidepressants increase human hippocampal neurogenesis by activating the glucocorticoid receptor. Mol Psychiatry 16:738-750. Andersen SL, Teicher MH (2009) Desperately driven and no brakes: developmental stress exposure and subsequent risk for substance abuse. Neurosci Biobehav Rev 33:516-524. Annegers JF, Hauser WA, Shirts SB, Kurland LT (1987) Factors prognostic of unprovoked seizures after febrile convulsions. N Engl J Med 316:493-498. Antebi D, Bird J (1993) The facilitation and evocation of seizures. A questionnaire study of awareness and control. Br J Psychiatry 162:759-764. Arida RM, Scorza FA, Terra VC, Scorza CA, de Almeida AC, Cavalheiro EA (2009) Physical exercise in epilepsy: what kind of stressor is it? Epilepsy Behav 16:381-387. Auvergne R, Lere C, El Bahh B, Arthaud S, Lespinet V, Rougier A,Le Gal La Salle G (2002) Delayed kindling epileptogenesis and increased neurogenesis in adult rats housed in an enriched environment. Brain Res 954:277-285.

207

ADDENDUM

Avoli MÂ (2001) Do interictal discharges promote or control seizures? Experimental evidence from an in vitro model of epileptiform discharge. Epilepsia 42 Suppl 3:2-4. Aytch LS, Hammond R, White C (2001) Seizures in infants and young children: an exploratory study of family experiences and needs for information and support. J Neurosci Nurs 33:278-285. Badawy R, Macdonell R, Jackson G, Berkovic S (2009) The peri-ictal state: cortical excitability changes within 24 h of a seizure. Brain 132:1013-1021. Baram TZ, Mitchell WG, Hanson RA, Snead OC 3rd, Horton EJ (1995) Cerebrospinal fluid corticotropin and cortisol are reduced in infantile spasms. Pediatr Neurol 13:108-110. Baram TZ, Mitchell WG, Snead OC 3rd, Horton EJ, Saito M (1992) Brain-adrenal axis hormones are altered in the CSF of infants with massive infantile spasms. Neurology 42:1171-1175. Baram TZ, Schultz L (1995) ACTH does not control neonatal seizures induced by administration of exogenous corticotropin-releasing hormone. Epilepsia 36:174-178. Baram TZ, Schultz L (1991) Corticotropin-releasing hormone is a rapid and potent convulsant in the infant rat. Brain Res Dev Brain Res 61:97-101. Battino D, Estienne M, Avanzini G (1995) Clinical pharmacokinetics of antiepileptic drugs in paediatric patients. Part I: Phenobarbital, primidone, valproic acid, ethosuximide and mesuximide. Clin Pharmacokinet 29:257-286. Battino D, Estienne M, Avanzini G (1995) Clinical pharmacokinetics of antiepileptic drugs in paediatric patients. Part II. Phenytoin, carbamazepine, sulthiame, lamotrigine, vigabatrin, oxcarbazepine and felbamate. Clin Pharmacokinet 29:341-369. Bauer J, Stoffel-Wagner B, Flugel D, Kluge M, Schramm J, Bidlingmaier F, Elger CE (2000) Serum androgens return to normal after temporal lobe epilepsy surgery in men. Neurology 55:820-824. Baulac M, Pitkanen A (2008) Research Priorities in Epilepsy for the Next Decade-A Representative View of the European Scientific Community. Epilepsia Baumann N, Turpin JC (2010) Neurochemistry of stress. An overview. Neurochem Res 35:1875-1879. Beato M, Truss M, Chavez S (1996) Control of transcription by steroid hormones. Ann N Y Acad Sci 784:93-123. Beck SL, Gavin DL (1976) Suceptibility of mice to audiogenic seizures is increased by handling their dams during gestation. Science 193:427-428. Belanoff JK, Rothschild AJ, Cassidy F, DeBattista C, Baulieu EE, Schold C, Schatzberg AF (2002) An open label trial of C-1073 (mifepristone) for psychotic major depression. Biol Psychiatry 52:386-392. Belichencko P, Dahlstrom A, von Essen C, Lindstrom S, Nordborg C, Sourander P (1992) Atypical pyramidal cells in epileptic human cortex: CFLS and 3-D reconstructions. Neuroreport 3:765-768.

208

References

Bellinger DL, Lubahn C, Lorton D (2008) Maternal and early life stress effects on immune function: relevance to immunotoxicology. J Immunotoxicol 5:419-444. Bender RA, Dube C, Baram TZ (2004) Febrile seizures and mechanisms of epileptogenesis: insights from an animal model. Adv Exp Med Biol 548:213-225. Bender RA, Soleymani SV, Brewster AL, Nguyen ST, Beck H, Mathern GW, Baram TZ (2003) Enhanced expression of a specific hyperpolarization-activated cyclic nucleotide-gated cation channel (HCN) in surviving dentate gyrus granule cells of human and experimental epileptic hippocampus. J Neurosci 23:6826-6836. Bennett DRÂ (1967) Spike-wave complexes in "normal" flying personnel. Aerosp Med 38:1276-1282. Berg AT, Jallon P, Preux PMÂ (2013) Chapter 43: The epidemiology of seizure disorders in infancy and childhood: definitions and classifications. In: Handbook of Clinical Neurology, Vol. III (3rd series), Pediatric Neurology Part I, 1st edition (Dulac O, Lassonde M, Samat HB eds), pp 391-398. Amsterdam: Elsevier B.V. Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross JH, van Emde Boas W, Engel J, French J, Glauser TA, Mathern GW, Moshe SL, Nordli D, Plouin P, Scheffer IE (2010) Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005-2009. Epilepsia 51:676-685. Berg AT, Scheffer IE (2011) New concepts in classification of the epilepsies: entering the 21st century. Epilepsia 52:1058-1062. Berg AT, Shinnar S (1994) Relapse following discontinuation of antiepileptic drugs: a meta-analysis. Neurology 44:601-608. Berg AT, Vickrey BG, Testa FM, Levy SR, Shinnar S, DiMario F, Smith S (2006) How long does it take for epilepsy to become intractable? A prospective investigation. Ann Neurol 60:73-79. Berkovic SF, Mulley JC, Scheffer IE, Petrou S (2006) Human epilepsies: interaction of genetic and acquired factors. Trends Neurosci 29:391-397. Bhake RC, Leendertz JA, Linthorst AC, Lightman SL (2013) Automated 24-hours sampling of subcutaneous tissue free cortisol in humans. J Med Eng Technol 37:180-184. Biagini G, Panuccio G, Avoli M (2010) Neurosteroids and epilepsy. Curr Opin Neurol 23:170-176. Blanchet P, Frommer GP (1986) Mood change preceding epileptic seizures. J Nerv Ment Dis 174:471476. Borders AE, Grobman WA, Amsden LB, McDade TW, Sharp LK, Holl JL (2010) The relationship between self-report and biomarkers of stress in low-income reproductive-age women. Am J Obstet Gynecol 203:577.e1-577.e8. Bosnjak J, Vukovic-Bobic M, Mejaski-Bosnjak V (2002) Effect of war on the occurrence of epileptic seizures in children. Epilepsy Behav 3:502-509.

209

ADDENDUM

Braun KP, Schmidt D (2014) Stopping antiepileptic drugs in seizure-free patients. Curr Opin Neurol 27:219-226. Bredy TW, Zhang TY, Grant RJ, Diorio J, Meaney MJ (2004) Peripubertal environmental enrichment reverses the effects of maternal care on hippocampal development and glutamate receptor subunit expression. Eur J Neurosci 20:1355-1362. Brigo F, Igwe SC, Ausserer H, Nardone R, Tezzon F, Bongiovanni LG, Trinka E (2014) Why do people Google epilepsy? An infodemiological study of online behavior for epilepsy-related search terms. Epilepsy Behav 31:67-70. Brown CE, Dunn W (2007) Adolescent/Adult Sensory Profile-NL (manual). (Rietman A ed),Amsterdam: Pearson Assessment and Information B.V. Burgess DL (2006) Chapter 16: Transgenic and Gene Replacement Models of Epilepsy. In: Models of Seizures and Epilepsy, (Pitkänen A, Schwarzkroin PA, Moshé SL eds),Burlington, San Diego, USA: Elsevier Inc. Calabrese VP, Gruemer HD, Tripathi HL, Dewey W, Fortner CA,DeLorenzo RJ (1993) Serum cortisol and cerebrospinal fluid beta-endorphins in status epilepticus. Their possible relation to prognosis. Arch Neurol 50:689-693. Callenbach PM, Westendorp RG, Geerts AT, Arts WF, Peeters EA, van Donselaar CA, Peters AC, Stroink H,Brouwer OF (2001) Mortality risk in children with epilepsy: the Dutch study of epilepsy in childhood. Pediatrics 107:1259-1263. Campanella G, Filla A, De Michele G (1978) Smell and taste acuity in epileptic syndromes. Eur Neurol 17:136-141. Cannon WB (1929) Bodily changes in pain, hunger, fear, and rage. New York: Appleton-CenturyCrofts. Carlton-Ford S, Miller R, Nealeigh N, Sanchez N (1997) The effects of perceived stigma and psychological over-control on the behavioural problems of children with epilepsy. Seizure 6:383-391. Carneiro HA, Mylonakis E (2009) Google trends: a web-based tool for real-time surveillance of disease outbreaks. Clin Infect Dis 49:1557-1564. Case-Smith J, Weaver LL, Fristad MA (2014) A systematic review of sensory processing interventions for children with autism spectrum disorders. Autism Chadda R, Devaud LL (2004) Sex differences in effects of mild chronic stress on seizure risk and GABAA receptors in rats. Pharmacol Biochem Behav 78:495-504. Champagne DL, de Kloet ER, Joëls M (2009) Fundamental aspects of the impact of glucocorticoids on the (immature) brain. Semin Fetal Neonatal Med 14:136-142. Chang BS, Lowenstein DH (2003) Epilepsy. N Engl J Med 349:1257-1266.

210

References

Charman T (2008) Autism spectrum disorders. Psychiatry 7:331-334. Charuvanij A, Ouvrier RA, Procopis PG, Antony JH, Fagan ER (1992) ACTH treatment in intractable seizures of childhood. Brain Dev 14:102-106. Chen Y, Bender RA, Frotscher M, Baram TZ (2001) Novel and transient populations of corticotropinreleasing hormone-expressing neurons in developing hippocampus suggest unique functional roles: a quantitative spatiotemporal analysis. J Neurosci 21:7171-7181. Chiappin S, Antonelli G, Gatti R, De Palo EF (2007) Saliva specimen: a new laboratory tool for diagnostic and basic investigation. Clin Chim Acta 383:30-40. Chin RF, Cumberland PM, Pujar SS, Peckham C, Ross EM, Scott RC (2011) Outcomes of childhood epilepsy at age 33 years: a population-based birth-cohort study. Epilepsia 52:1513-1521. Chipaux M, Vercueil L, Kaminska A, Mahon S, Charpier S (2013) Persistence of cortical sensory processing during absence seizures in human and an animal model: evidence from EEG and intracellular recordings. PLoS One 8:e58180. Christensen J, Li J, Vestergaard M, Olsen J (2007) Stress and epilepsy: a population-based cohort study of epilepsy in parents who lost a child. Epilepsy Behav 11:324-328. Chutorian AM, Gold AP, Low NL (1968) Steroid therapy of non-infantile (childhood) myoclonic epilepsy. Neurology 18:304-305. Cicchetti D, Toth SL (2009) The past achievements and future promises of developmental psychopathology: the coming of age of a discipline. J Child Psychol Psychiatry 50:16-25. Coddington RD (1972) The significance of life events as etiologic factors in the diseases of children. II. A study of a normal population. J Psychosom Res 16:205-213. Cohen I, Navarro V, Clemenceau S, Baulac M, Miles R (2002) On the origin of interictal activity in human temporal lobe epilepsy in vitro. Science 298:1418-1421. Cohen RA, Grieve S, Hoth KF, Paul RH, Sweet L, Tate D, Gunstad J, Stroud L, McCaffery J, Hitsman B, Niaura R, Clark CR, McFarlane A, Bryant R, Gordon E, Williams LM (2006) Early life stress and morphometry of the adult anterior cingulate cortex and caudate nuclei. Biol Psychiatry 59:975-982. Constantino JN, Gruber CP (2011) SRS: Screeningslijst voor autismespectrumstoornissen (manual). Amsterdam: Hogreve Uitgevers. Cross JH, Jayakar P, Nordli D, Delalande O, Duchowny M, Wieser HG, Guerrini R, Mathern GW, International League against Epilepsy, Subcommission for Paediatric Epilepsy Surgery, Commissions of Neurosurgery and Paediatrics (2006) Proposed criteria for referral and evaluation of children for epilepsy surgery: recommendations of the Subcommission for Pediatric Epilepsy Surgery. Epilepsia 47:952-959. Cui M, Yang Y, Yang J, Zhang J, Han H, Ma W, Li H, Mao R, Xu L, Hao W, Cao J (2006) Enriched environment experience overcomes the memory deficits and depressive-like behavior induced by early life stress. Neurosci Lett 404:208-212.

211

ADDENDUM

Culebras A, Miller M, Bertram L, Koch J (1987) Differential response of growth hormone, cortisol, and prolactin to seizures and to stress. Epilepsia 28:564-570. Cull CA, Fowler M, Brown SW (1996) Perceived self-control of seizures in young people with epilepsy. Seizure 5:131-138. da Silva Sousa P, Lin K, Garzon E, Sakamoto AC, Yacubian EM (2005) Self-perception of factors that precipitate or inhibit seizures in juvenile myoclonic epilepsy. Seizure 14:340-346. Dahl J, Melin L, Lund L (1987) Effects of a contingent relaxation treatment program on adults with refractory epileptic seizures. Epilepsia 28:125-132. Davidson JO, Quaedackers JS, George SA, Gunn AJ, Bennet L (2011) Maternal dexamethasone and EEG hyperactivity in preterm fetal sheep. J Physiol 589:3823-3835. Davies PL, Chang WP, Gavin WJ (2010) Middle and Late Latency ERP Components Discriminate between Adults, Typical Children, and Children with Sensory Processing Disorders. Front Integr Neurosci 4:16. Davies PL, Chang WP, Gavin WJ (2009) Maturation of sensory gating performance in children with and without sensory processing disorders. Int J Psychophysiol 72:187-197. Davies PL, Gavin WJ (2007) Validating the diagnosis of sensory processing disorders using EEG technology. Am J Occup Ther 61:176-189. De Curtis M, Avanzini G (2001) Interictal spikes in focal epileptogenesis. Prog Neurobiol 63:541-567. De Curtis M, Jefferys JGR, Avoli M (2012) Interictal Epileptiform Discharges in Partial Epilepsy: Complex Neurobiological Mechanisms Based on Experimental and Clinical Evidence. In: Jasper's Basic Mechanisms of the Epilepsies, 4th edition (Noebels JL, Avoli M, Rogawski MA, Olsen RW, Delgado-Escueta AV eds),Bethesda (MD): Michael A Rogawski, Antonio V Delgado-Escueta, Jeffrey L Noebels, Massimo Avoli and Richard W Olsen. De Kloet ER (2014) From receptor balance to rational glucocorticoid therapy. Endocrinology 155:2754-2769. De Kloet ER, Joëls M, Holsboer F (2005) Stress and the brain: from adaptation to disease. Nat Rev Neurosci 6:463-475. De Kloet ER, Joëls M, Oitzl M, Sutanto W (1991) Implication of brain corticosteroid receptor diversity for the adaptation syndrome concept. Methods Achiev Exp Pathol 14:104-132. De Kloet ER, Karst H, Joëls M (2008) Corticosteroid hormones in the central stress response: quickand-slow. Front Neuroendocrinol 29:268-272. De Kloet ER, Vreugdenhil E, Oitzl MS, Joëls M (1998) Brain corticosteroid receptor balance in health and disease. Endocr Rev 19:269-301.

212

References

De Lima TC, Rae GA (1991) Effects of cold-restraint and swim stress on convulsions induced by pentylenetetrazol and electroshock: influence of naloxone pretreatment. Pharmacol Biochem Behav 40:297-300. De Weerth C, Zijl RH, Buitelaar JK (2003) Development of cortisol circadian rhythm in infancy. Early Hum Dev 73:39-52. De Winter JCF (2013) Using the Student's t-test with extremely small sample sized. Practical Assessment, Reseach & Evaluation 18: Dean P (2011) Counseling of parents of a child with uncontrollable seizures. Semin Pediatr Neurol 18:209-212. DeBattista C (2006) Augmentation and combination strategies for depression. J Psychopharmacol 20:11-18. DeBattista C, Belanoff J (2006) The use of mifepristone in the treatment of neuropsychiatric disorders. Trends Endocrinol Metab 17:117-121. Deepak KK, Manchanda SK, Maheshwari MC (1994) Meditation improves clinicoelectroencephalographic measures in drug-resistant epileptics. Biofeedback Self Regul 19:25-40. Demont-Guignard S, Benquet P, Gerber U, Biraben A, Martin B,Wendling F (2012) Distinct hyperexcitability mechanisms underlie fast ripples and epileptic spikes. Ann Neurol 71:342-352. DeRijk RH (2009) Single nucleotide polymorphisms related to HPA axis reactivity. Neuroimmunomodulation 16:340-352. DeRijk RH, de Kloet ER (2008) Corticosteroid receptor polymorphisms: determinants of vulnerability and resilience. Eur J Pharmacol 583:303-311. DeRijk RH, de Kloet ER, Zitman FG, van Leeuwen N (2011) Mineralocorticoid receptor gene variants as determinants of HPA axis regulation and behavior. Endocr Dev 20:137-148. Desgent S, Duss S, Sanon NT, Lema P, Levesque M, Hebert D, Rebillard RM, Bibeau K, Brochu M, Carmant L (2012) Early-life stress is associated with gender-based vulnerability to epileptogenesis in rat pups. PLoS One 7:e42622. Diamantis A, Sidiropoulou K, Magiorkinis E (2010) Epilepsy during the Middle Ages, the Renaissance and the Enlightenment. J Neurol 257:691-698. Dickerson S, Reinhart AM, Feeley TH, Bidani R, Rich E, Garg VK, Hershey CO (2004) Patient Internet use for health information at three urban primary care clinics. J Am Med Inform Assoc 11:499-504. Dickmeis T (2009) Glucocorticoids and the circadian clock. J Endocrinol 200:3-22. Dinner DS (2002) Effect of sleep on epilepsy. J Clin Neurophysiol 19:504-513.

213

ADDENDUM

Dooley JM, Camfield PR, Goulden KJ,Macken SR (1989) Low-dose alternate-day corticotropin therapy in the treatment of childhood seizures. Am J Dis Child 143:1263-1265. Doose H, Gerken H (1973) On the genetics of EEG-anomalies in childhood. IV. Photoconvulsive reaction. Neuropadiatrie 4:162-171. Dos Santos Lunardi M, Sukys-Claudino L, Guarnieri R, Walz R, Lin K (2011) Seizure precipitants and inhibiting factors in mesial temporal lobe epilepsy. J Neurol Sci 308:21-24. Doyle LW, Anderson PJ, Battin M, Bowen JR, Brown N, Callanan C, Campbell C, Chandler S, Cheong J, Darlow B, Davis PG, DePaoli T, French N, McPhee A, Morris S, O'Callaghan M, Rieger I, Roberts G, Spittle AJ, Wolke D,Woodward LJ (2014) Long term follow up of high risk children: who, why and how? BMC Pediatr 14:279-2431-14-279. Dube C, Richichi C, Bender RA, Chung G, Litt B,Baram TZ (2006) Temporal lobe epilepsy after experimental prolonged febrile seizures: prospective analysis. Brain 129:911-922. Dube CM, Brewster AL,Baram TZ (2009) Febrile seizures: mechanisms and relationship to epilepsy. Brain Dev 31:366-371. Dube CM, Brewster AL, Richichi C, Zha Q,Baram TZ (2007) Fever, febrile seizures and epilepsy. Trends Neurosci 30:490-496. Dunn W (2013) Sensory Profile-NL: handleiding (manual). (Rietman A ed),Amsterdam: Pearson Assessment and Information B.V. Dunn W (2001) The sensations of everyday life: empirical, theoretical, and pragmatic considerations. Am J Occup Ther 55:608-620. Dunn W (1997) The impact of sensory processing abilities on the daily lives of young children and their families: a conceptual model. Infants and young children 9:23-24 25. Dunn W, Brown C (1997) Factor analysis on the Sensory Profile from a national sample of children without disabilities. Am J Occup Ther 51:490-5; discussion 496-9. Durazzo TS, Spencer SS, Duckrow RB, Novotny EJ, Spencer DD,Zaveri HP (2008) Temporal distributions of seizure occurrence from various epileptogenic regions. Neurology 70:1265-1271. Edwards HE, Dortok D, Tam J, Won D,Burnham WM (2002) Prenatal stress alters seizure thresholds and the development of kindled seizures in infant and adult rats. Horm Behav 42:437-447. Edwards HE, Vimal S,Burnham WM (2005) The acute anticonvulsant effects of deoxycorticosterone in developing rats: role of metabolites and mineralocorticoid-receptor responses. Epilepsia 46:18881897. Edwards HE, Vimal S,Burnham WM (2002) Dose-, time-, age-, and sex-response profiles for the anticonvulsant effects of deoxycorticosterone in 15-day-old rats. Exp Neurol 176:364-370. Edwards HE, Vimal S, Burnham WM (2002) The effects of ACTH and adrenocorticosteroids on seizure susceptibility in 15-day-old male rats. Exp Neurol 175:182-190.

214

References

Eeg-Olofsson O, Petersen I, Sellden U (1971) The development of the electroencephalogram in normal children from the age of 1 through 15 years. Paroxysmal activity. Neuropadiatrie 2:375-404. Eeles AL, Anderson PJ, Brown NC, Lee KJ, Boyd RN, Spittle AJ, Doyle LW (2013) Sensory profiles obtained from parental reports correlate with independent assessments of development in very preterm children at 2 years of age. Early Hum Dev 89:1075-1080. Ehlers CL, Henriksen SJ, Wang M, Rivier J, Vale W, Bloom FE (1983) Corticotropin releasing factor produces increases in brain excitability and convulsive seizures in rats. Brain Res 278:332-336. Engel J Jr, Ackermann RF (1980) Interictal EEG spikes correlate with decreased, rather than increased, epileptogenicity in amygdaloid kindled rats. Brain Res 190:543-548. Eriksson L, Haglund B, Ewald U, Odlind V, Kieler H (2012) Health consequences of prophylactic exposure to antenatal corticosteroids among children born late preterm or term. Acta Obstet Gynecol Scand 91:1415-1421. Eriksson L, Haglund B, Ewald U, Odlind V, Kieler H (2009) Short and long-term effects of antenatal corticosteroids assessed in a cohort of 7,827 children born preterm. Acta Obstet Gynecol Scand 88:933-938. Facchinetti F, Nalin A, Petraglia F, Galli V, Genazzani AR (1985) Reduced ACTH, while normal beta-endorphin CSF levels in early epileptic encephalopathies. Peptides 6:31-33. Fang PC, Chen YJ, Lee IC (2008) Seizure precipitants in children with intractable epilepsy. Brain Dev 30:527-532. Faravelli C, Lo Sauro C, Lelli L, Pietrini F, Lazzeretti L, Godini L, Benni L, Fioravanti G, Talamba GA, Castellini G, Ricca V (2012) The role of life events and HPA axis in anxiety disorders: a review. Curr Pharm Des 18:5663-5674. Fares RP, Belmeguenai A, Sanchez PE, Kouchi HY, Bodennec J, Morales A, Georges B, Bonnet C, Bouvard S, Sloviter RS, Bezin L (2013) Standardized environmental enrichment supports enhanced brain plasticity in healthy rats and prevents cognitive impairment in epileptic rats. PLoS One 8:e53888. Farmer GE, Park CR, Laura A. Bullard LA, Diamond DM (2014) Evolutionary, Historical and Mechanistic Perspectives on How Stress Affects Memory and Hippocampal Synaptic Plasticity. In: Synaptic Stress and Pathogenesis of Neuropsychiatric Disorders, (Popoli M, Diamond D., Gerard Sanacora G eds), pp 167-182. New York Heidelberg Dordrecht London: Springer. Feldman RG, Paul NL (1976) Identity of emotional triggers in epilepsy. J Nerv Ment Dis 162:345-353. Fenwick PÂ (1991) Evocation and inhibition of seizures. Behavioral treatment. Adv Neurol 55:163-183. Ferando I, Mody I (2012) GABAA receptor modulation by neurosteroids in models of temporal lobe epilepsies. Epilepsia 53 Suppl 9:89-101. Ferlisi M, Shorvon S (2014) Seizure precipitants (triggering factors) in patients with epilepsy. Epilepsy Behav 33:101-105.

215

ADDENDUM

Ferro MA, Chin RF, Camfield CS, Wiebe S, Levin SD, Speechley KN (2014) Convulsive status epilepticus and health-related quality of life in children with epilepsy. Neurology Fiedler BJ, Debus OM, Neubauer BA, Kienle M, Kurlemann G (2006) P50 sensory gating deficit in children with centrotemporal spikes and sharp waves in the EEG. Neurosci Lett 393:206-210. Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, Engel J Jr, Forsgren L, French JA, Glynn M, Hesdorffer DC, Lee BI, Mathern GW, Moshe SL, Perucca E, Scheffer IE, Tomson T, Watanabe M, Wiebe S (2014) ILAE official report: a practical clinical definition of epilepsy. Epilepsia 55:475-482. Fisher RS, van Emde Boas W, Blume W, Elger C, Genton P, Lee P, Engel J Jr (2005) Epileptic seizures and epilepsy: definitions proposed by the International League Against Epilepsy (ILAE) and the International Bureau for Epilepsy (IBE). Epilepsia 46:470-472. Flores BH, Kenna H, Keller J, Solvason HB, Schatzberg AF (2006) Clinical and biological effects of mifepristone treatment for psychotic depression. Neuropsychopharmacology 31:628-636. Forsgren L, Beghi E, Oun A, Sillanpaa M (2005) The epidemiology of epilepsy in Europe - a systematic review. Eur J Neurol 12:245-253. Franks RCÂ (1967) Diurnal variation of plasma 17-hydroxycorticosteroids in children. J Clin Endocrinol Metab 27:75-78. Fredriks AM, van Buuren S, Burgmeijer RJ, Meulmeester JF, Beuker RJ, Brugman E, Roede MJ, Verloove-Vanhorick SP, Wit JM (2000) Continuing positive secular growth change in The Netherlands 1955-1997. Pediatr Res 47:316-323. Fregnac Y, Laurent G (2014) Neuroscience: Where is the brain in the Human Brain Project? Nature 513:27-29. Fries E, Hesse J, Hellhammer J,Hellhammer DH (2005) A new view on hypocortisolism. Psychoneuroendocrinology 30:1010-1016. Frucht MM, Quigg M, Schwaner C, Fountain NB (2000) Distribution of seizure precipitants among epilepsy syndromes. Epilepsia 41:1534-1539. Frye CA, Bayon LE (1999) Prenatal stress reduces the effectiveness of the neurosteroid 3 alpha,5 alpha-THP to block kainic-acid-induced seizures. Dev Psychobiol 34:227-234. Galimberti CA, Magri F, Copello F, Arbasino C, Cravello L, Casu M, Patrone V, Murialdo G (2005) Seizure frequency and cortisol and dehydroepiandrosterone sulfate (DHEAS) levels in women with epilepsy receiving antiepileptic drug treatment. Epilepsia 46:517-523. Gallagher BBÂ (1987) Endocrine abnormalities in human temporal lobe epilepsy. Yale J Biol Med 60:93-97. Gallagher BB, Murvin A, Flanigin HF, King DW, Luney D (1984) Pituitary and adrenal function in epileptic patients. Epilepsia 25:683-689.

216

References

Galletti F, Sturniolo MG (2004) Counseling children and parents about epilepsy. Patient Educ Couns 55:422-425. Gavin WJ, Dotseth A, Roush KK, Smith CA, Spain HD, Davies PL (2011) Electroencephalography in children with and without sensory processing disorders during auditory perception. Am J Occup Ther 65:370-377. Gayatri NA, Ferrie CD, Cross H (2007) Corticosteroids including ACTH for childhood epilepsy other than epileptic spasms. Cochrane Database Syst Rev (1):CD005222. Geerts A, Brouwer O, van Donselaar C, Stroink H, Peters B, Peeters E, Arts WF (2011) Health perception and socioeconomic status following childhood-onset epilepsy: the Dutch study of epilepsy in childhood. Epilepsia 52:2192-2202. Gere DR, Capps SC, Mitchell DW, Grubbs E (2009) Sensory sensitivities of gifted children. Am J Occup Ther 63:288-95; discussion 296-300. Gibbs FA, Gibbs EL (1964) Atlas of electroencephalography. Vol III. Cambridge Mass: Addiso-Wesley Press. Gibbs S, Chattopadhyaya B, Desgent S, Awad PN, Clerk-Lamalice O, Levesque M, Vianna RM, Rebillard RM, Delsemme AA, Hebert D, Tremblay L, Lepage M, Descarries L, Di Cristo G, Carmant L (2011) Long-term consequences of a prolonged febrile seizure in a dual pathology model. Neurobiol Dis 43:312-321. Gilby KL, Sydserff S, Patey AM, Thorne V, St-Onge V, Jans J, McIntyre DC (2009) Postnatal epigenetic influences on seizure susceptibility in seizure-prone versus seizure-resistant rat strains. Behav Neurosci 123:337-346. Gillies GE, Linton EA, Lowry PJ (1982) Corticotropin releasing activity of the new CRF is potentiated several times by vasopressin. Nature 299:355-357. Ginsberg J, Mohebbi MH, Patel RS, Brammer L, Smolinski MS, Brilliant L (2009) Detecting influenza epidemics using search engine query data. Nature 457:1012-1014. Giorgi FS, Pizzanelli C, Biagioni F, Murri L, Fornai F (2004) The role of norepinephrine in epilepsy: from the bench to the bedside. Neurosci Biobehav Rev 28:507-524. Glover V (1997) Maternal stress or anxiety in pregnancy and emotional development of the child. Br J Psychiatry 171:105-106. Gluckman PD, Hanson MA, Buklijas T, Low FM, Beedle AS (2009) Epigenetic mechanisms that underpin metabolic and cardiovascular diseases. Nat Rev Endocrinol 5:401-408. Goodman R (1997) The Strengths and Difficulties Questionnaire: a research note. J Child Psychol Psychiatry 38:581-586. Goodman R, Ford T, Simmons H, Gatward R, Meltzer H (2000) Using the Strengths and Difficulties Questionnaire (SDQ) to screen for child psychiatric disorders in a community sample. Br J Psychiatry 177:534-539.

217

ADDENDUM

Goodman R, Renfrew D, Mullick M (2000) Predicting type of psychiatric disorder from Strengths and Difficulties Questionnaire (SDQ) scores in child mental health clinics in London and Dhaka. Eur Child Adolesc Psychiatry 9:129-134. Gotman J, Koffler DJ (1989) Interictal spiking increases after seizures but does not after decrease in medication. Electroencephalogr Clin Neurophysiol 72:7-15. Gotman J, Marciani MG (1985) Electroencephalographic spiking activity, drug levels, and seizure occurrence in epileptic patients. Ann Neurol 17:597-603. Gould E, McEwen BS, Tanapat P, Galea LA, Fuchs E (1997) Neurogenesis in the dentate gyrus of the adult tree shrew is regulated by psychosocial stress and NMDA receptor activation. J Neurosci 17:2492-2498. Gozansky WS, Lynn JS, Laudenslager ML, Kohrt WM (2005) Salivary cortisol determined by enzyme immunoassay is preferable to serum total cortisol for assessment of dynamic hypothalamic--pituitary-adrenal axis activity. Clin Endocrinol (Oxf) 63:336-341. Granger DA, Kivlighan KT, el-Sheikh M, Gordis EB, Stroud LR (2007). Salivary alpha-amylase in biobehavioral research: recent developments and applications. Ann N Y Acad Sci 1098: 122-144. Grant ACÂ (2005) Interictal perceptual function in epilepsy. Epilepsy Behav 6:511-519. Gregory RP, Oates T, Merry RT (1993) Electroencephalogram epileptiform abnormalities in candidates for aircrew training. Electroencephalogr Clin Neurophysiol 86:75-77. Groc L, Choquet D, Chaouloff F (2008) The stress hormone corticosterone conditions AMPAR surface trafficking and synaptic potentiation. Nat Neurosci 11:868-870. Groschl M, Rauh M, Dorr HG (2003) Circadian rhythm of salivary cortisol, 17alphahydroxyprogesterone, and progesterone in healthy children. Clin Chem 49:1688-1691. Gunnar MR, Quevedo KM (2008) Early care experiences and HPA axis regulation in children: a mechanism for later trauma vulnerability. Prog Brain Res 167:137-149. Gunnar MR, Vazquez DM (2001) Low cortisol and a flattening of expected daytime rhythm: potential indices of risk in human development. Dev Psychopathol 13:515-538. Gunnar MR, Wewerka S, Frenn K, Long JD, Griggs C (2009) Developmental changes in hypothalamuspituitary-adrenal activity over the transition to adolescence: normative changes and associations with puberty. Dev Psychopathol 21:69-85. Gutteling BM, de Weerth C, Buitelaar JK (2005) Prenatal stress and children's cortisol reaction to the first day of school. Psychoneuroendocrinology 30:541-549. Hancock EC, Cross HH (2009) Treatment of Lennox-Gastaut syndrome. Cochrane Database Syst Rev (3):CD003277. doi:CD003277. Hart YM, Shorvon SD (1995) The nature of epilepsy in the general population. I. Characteristics of patients receiving medication for epilepsy. Epilepsy Res 21:43-49.

218

References

Hartley R, Forsythe WI, McLain B, Ng PC, Lucock MD (1991) Daily variations in steady-state plasma concentrations of carbamazepine and its metabolites in epileptic children. Clin Pharmacokinet 20:237246. Haut SR, Hall CB, Borkowski T, Tennen H, Lipton RB (2012) Clinical features of the pre-ictal state: mood changes and premonitory symptoms. Epilepsy Behav 23:415-421. Haut SR, Hall CB, Masur J, Lipton RB (2007) Seizure occurrence: precipitants and prediction. Neurology 69:1905-1910. Haut SR, Vouyiouklis M, Shinnar S (2003) Stress and epilepsy: a patient perception survey. Epilepsy Behav 4:511-514. Hayden M, Penna C, Buchanan N (1992) Epilepsy: patient perceptions of their condition. Seizure 1:191-197. Heim C, Binder EB (2012) Current research trends in early life stress and depression: review of human studies on sensitive periods, gene-environment interactions, and epigenetics. Exp Neurol 233:102111. Heim C, Newport DJ, Mletzko T, Miller AH, Nemeroff CB (2008) The link between childhood trauma and depression: insights from HPA axis studies in humans. Psychoneuroendocrinology 33:693-710. Heinrichs SC, Koob GF (2006) Application of experimental stressors in laboratory rodents. Curr Protoc Neurosci Chapter 8:Unit8.4. Heiskala H (1997) CSF ACTH and beta-endorphin in infants with West syndrome and ACTH therapy. Brain Dev 19:339-342. Hellman L, Nakada F, Curti J, Weitzman ED, Kream J, Roffwarg H, Ellman S, Fukushima DK, Gallagher TF (1970) Cortisol is secreted episodically by normal man. J Clin Endocrinol Metab 30:411422. Helsloot JIA (2011) Stress and ritual. December family traditions in the Netherlands of today. Lietuvos Etnologija 11:143-157. Herman JP, Figueiredo H, Mueller NK, Ulrich-Lai Y, Ostrander MM, Choi DC, Cullinan WE (2003) Central mechanisms of stress integration: hierarchical circuitry controlling hypothalamo-pituitaryadrenocortical responsiveness. Front Neuroendocrinol 24:151-180. Hermann BP, Jones JE, Jackson DC, Seidenberg M (2012) Starting at the beginning: the neuropsychological status of children with new-onset epilepsies. Epileptic Disord 14:12-21. Hermans EJ, Henckens MJ, Joëls M, Fernandez G (2014) Dynamic adaptation of large-scale brain networks in response to acute stressors. Trends Neurosci 37:304-314. Hill M, Vrbikova J, Zarubova J, Kancheva R, Velikova M, Kancheva L, Kubatova J, Duskova M, Marusic P, Parizek A, Starka L (2011) The steroid metabolome in lamotrigine-treated women with epilepsy. Steroids 76:1351-1357.

219

ADDENDUM

Hoare P, Kerley S (1991) Psychosocial adjustment of children with chronic epilepsy and their families. Dev Med Child Neurol 33:201-215. Hoeymans N, Feskens EJ, Van Den Bos GA, Kromhout D (1998) Non-response bias in a study of cardiovascular diseases, functional status and self-rated health among elderly men. Age Ageing 27:3540. Hofstra WA, Grootemarsink BE, Dieker R, van der Palen J, de Weerd AW (2009) Temporal distribution of clinical seizures over the 24-h day: a retrospective observational study in a tertiary epilepsy clinic. Epilepsia 50:2019-2026. Hofstra WA, Spetgens WP, Leijten FS, van Rijen PC, Gosselaar P, van der Palen J, de Weerd AW (2009) Diurnal rhythms in seizures detected by intracranial electrocorticographic monitoring: an observational study. Epilepsy Behav 14:617-621. Holmes GL (1997) Epilepsy in the developing brain: lessons from the laboratory and clinic. Epilepsia 38:12-30. Holsboer F (2000) The corticosteroid receptor hypothesis of depression. Neuropsychopharmacology 23:477-501. Howard P, Twycross R, Shuster J, Mihalyo M, Remi J, Wilcock A (2011) Anti-epileptic drugs. J Pain Symptom Manage 42:788-804. Høyberg ØJ, Wik G, Mehtonen OP, Peeters BWMM, Sennef C (2002) ORG 34517, a selective glucocorticoid receptor antagonist with potent antidepressant activity: first clinical results. Int J Neuropsychopharmacol 5:S148. Huang LT, Holmes GL, Lai MC, Hung PL, Wang CL, Wang TJ, Yang CH, Liou CW, Yang SN (2002) Maternal deprivation stress exacerbates cognitive deficits in immature rats with recurrent seizures. Epilepsia 43:1141-1148. Hulbert-Williams L, Hastings RP (2008) Life events as a risk factor for psychological problems in individuals with intellectual disabilities: a critical review. J Intellect Disabil Res 52:883-895. Illingworth JL, Ring H (2013) Conceptual distinctions between reflex and nonreflex precipitated seizures in the epilepsies: a systematic review of definitions employed in the research literature. Epilepsia 54:2036-2047. Jabbari B, Russo MB, Russo ML (2000) Electroencephalogram of asymptomatic adult subjects. Clin Neurophysiol 111:102-105. Joëls M (2009) Stress, the hippocampus, and epilepsy. Epilepsia 50:586-597. Joëls M (2007) Role of corticosteroid hormones in the dentate gyrus. Prog Brain Res 163:355-370. Joëls M (2001) Corticosteroid actions in the hippocampus. J Neuroendocrinol 13:657-669. Joëls M, Baram TZ (2009) The neuro-symphony of stress. Nat Rev Neurosci 10:459-466.

220

References

Joels M, de Kloet ER (1992). Control of neuronal excitability by corticosteroid hormones. Trends Neurosci 15:25-30. Joëls M, Karst H, Alfarez D, Heine VM, Qin Y, van Riel E, Verkuyl M, Lucassen PJ, Krugers HJ (2004) Effects of chronic stress on structure and cell function in rat hippocampus and hypothalamus. Stress 7:221-231. Joëls M, Karst H, DeRijk R, de Kloet ER (2008) The coming out of the brain mineralocorticoid receptor. Trends Neurosci 31:1-7. Joëls M, Karst H, Krugers HJ, Lucassen PJ (2007) Chronic stress: implications for neuronal morphology, function and neurogenesis. Front Neuroendocrinol 28:72-96. Joëls M, Krugers HJ, Lucassen PJ, Karst H (2009) Corticosteroid effects on cellular physiology of limbic cells. Brain Res 1293:91-100. Joëls M, Sarabdjitsingh RA, Karst H (2012) Unraveling the time domains of corticosteroid hormone influences on brain activity: rapid, slow, and chronic modes. Pharmacol Rev 64:901-938. Jones NC, Kumar G, O'Brien TJ, Morris MJ, Rees SM, Salzberg MR (2009) Anxiolytic effects of rapid amygdala kindling, and the influence of early life experience in rats. Behav Brain Res 203:81-87. Jones NC, O'Brien TJ, Carmant L (2014) Interaction between sex and early-life stress: Influence on epileptogenesis and epilepsy comorbidities. Neurobiol Dis Jonetz-Mentzel L, Wiedemann G (1993) Establishment of reference ranges for cortisol in neonates, infants, children and adolescents. Eur J Clin Chem Clin Biochem 31:525-529. Jurgens CW, Boese SJ, King JD, Pyle SJ, Porter JE, Doze VA (2005) Adrenergic receptor modulation of hippocampal CA3 network activity. Epilepsy Res 66:117-128. Kajantie E, Phillips DI (2006) The effects of sex and hormonal status on the physiological response to acute psychosocial stress. Psychoneuroendocrinology 31:151-178. Kaleyias J, Loddenkemper T, Vendrame M, Das R, Syed TU, Alexopoulos AV, Wyllie E, Kothare SV (2011) Sleep-wake patterns of seizures in children with lesional epilepsy. Pediatr Neurol 45:109-113. Kalra V, Dua T, Kumar V (2003) Efficacy of albendazole and short-course dexamethasone treatment in children with 1 or 2 ring-enhancing lesions of neurocysticercosis: a randomized controlled trial. J Pediatr 143:111-114. Karafin M, St Louis EK, Zimmerman MB, Sparks JD, Granner MA (2010) Bimodal ultradian seizure periodicity in human mesial temporal lobe epilepsy. Seizure 19:347-351. Karst H, Joëls M (2005) Corticosterone slowly enhances miniature excitatory postsynaptic current amplitude in mice CA1 hippocampal cells. J Neurophysiol 94:3479-3486. Karst H, Joëls M (2003) Effect of chronic stress on synaptic currents in rat hippocampal dentate gyrus neurons. J Neurophysiol 89:625-633.

221

ADDENDUM

Kasteleijn-Nolst Trenite DG (2012) Provoked and reflex seizures: surprising or common? Epilepsia 53 Suppl 4:105-113. Kasteleijn-Nolst Trenite DG (1989) Photosensitivity in epilepsy. Electrophysiological and clinical correlates. Acta Neurol Scand Suppl 125:3-149. Kelley MS, Jacobs MP, Lowenstein DH, NINDS Epilepsy Benchmark Stewards (2009) The NINDS epilepsy research benchmarks. Epilepsia 50:579-582. Kern JK, Trivedi MH, Garver CR, Grannemann BD, Andrews AA, Savla JS, Johnson DG, Mehta JA, Schroeder JL (2006) The pattern of sensory processing abnormalities in autism. Autism 10:480-494. Kiess W, Meidert A, Dressendorfer RA, Schriever K, Kessler U, Konig A, Schwarz HP, Strasburger CJ (1995) Salivary cortisol levels throughout childhood and adolescence: relation with age, pubertal stage, and weight. Pediatr Res 37:502-506. Kirschbaum C, Hellhammer DH (2000) Salivary cortisol. In: Encyclopedia of stress. Vol. 3, (Fink G ed), pp 379-383. New York: Academic Press. Klein P, Sahoo S (2005) Effect of ACTH-induced hypercortisolemia on the EEG in patients with stress-related epilepsy. Epilepsy Behav 6:187-190. Klein P, van Passel L (2005) Effect of stress related to the 9/11/2001 terror attack on seizures in patients with epilepsy. Neurology 64:1815-1816. Klein ZA, Romeo RD (2013) Changes in hypothalamic-pituitary-adrenal stress responsiveness before and after puberty in rats. Horm Behav 64:357-363. Klopp C, Garcia C, Schulman AH, Ward CP, Tartar JL (2012) Acute social stress increases biochemical and self report markers of stress without altering spatial learning in humans. Neuro Endocrinol Lett 33:425-430. Knecht S, Henningsen H, Deppe M, Osinska L, Diehl B, Stodieck S, Ringelstein EB (1996) Persistent unihemispheric perceptual impairments in humans following focal seizures. Neurosci Lett 217:66-68. Knutsson U, Dahlgren J, Marcus C, Rosberg S, Bronnegard M, Stierna P,Albertsson-Wikland K (1997) Circadian cortisol rhythms in healthy boys and girls: relationship with age, growth, body composition, and pubertal development. J Clin Endocrinol Metab 82:536-540. Kobow K, Blumcke I (2012) The emerging role of DNA methylation in epileptogenesis. Epilepsia 53 Suppl 9:11-20. Koe AS, Jones NC, Salzberg MR (2009) Early life stress as an influence on limbic epilepsy: an hypothesis whose time has come? Front Behav Neurosci 3:24. Kohling R, Lucke A, Straub H, Speckmann EJ, Tuxhorn I, Wolf P, Pannek H, Oppel F (1998) Spontaneous sharp waves in human neocortical slices excised from epileptic patients. Brain 121 ( Pt 6):1073-1087.

222

References

Koob GF, Zorrilla EP (2012) Update on corticotropin-releasing factor pharmacotherapy for psychiatric disorders: a revisionist view. Neuropsychopharmacology 37:308-309. Korbey SM, Heinrichs SC, Leussis MP (2008) Seizure susceptibility and locus ceruleus activation are reduced following environmental enrichment in an animal model of epilepsy. Epilepsy Behav 12:30-38. Korosi A, Naninck EF, Oomen CA, Schouten M, Krugers H, Fitzsimons C, Lucassen PJ (2012) Earlylife stress mediated modulation of adult neurogenesis and behavior. Behav Brain Res 227:400-409. Koutsogiannopoulos S, Adelson F, Lee V, Andermann F (2009) Stressors at the onset of adult epilepsy: implications for practice. Epileptic Disord 11:42-47. Koyama R, Tao K, Sasaki T, Ichikawa J, Miyamoto D, Muramatsu R, Matsuki N, Ikegaya Y (2012) GABAergic excitation after febrile seizures induces ectopic granule cells and adult epilepsy. Nat Med 18:1271-1278. Kramer MA, Cash SS (2012) Epilepsy as a disorder of cortical network organization. Neuroscientist 18:360-372. Krieger DT (1970) Factors influencing the circadian periodicity of adrenal steroid levels. Trans N Y Acad Sci 32:316-329. Krugers HJ, Karst H, Joëls M (2012) Interactions between noradrenaline and corticosteroids in the brain: from electrical activity to cognitive performance. Front Cell Neurosci 6:15. Krugers HJ, Maslam S, Korf J, Joëls M, Holsboer F (2000) The corticosterone synthesis inhibitor metyrapone prevents hypoxia/ischemia-induced loss of synaptic function in the rat hippocampus. Stroke 31:1162-1172. Krugers HJ, Maslam S, Van Vuuren SM, Korf J, Joëls M (1999) Postischemic steroid modulation: effects on hippocampal neuronal integrity and synaptic plasticity. J Cereb Blood Flow Metab 19:10721082. Kudielka BM, Buske-Kirschbaum A, Hellhammer DH, Kirschbaum C (2004) Differential heart rate reactivity and recovery after psychosocial stress (TSST) in healthy children, younger adults, and elderly adults: the impact of age and gender. Int J Behav Med 11:116-121. Kudielka BM, Kirschbaum C (2005) Sex differences in HPA axis responses to stress: a review. Biol Psychol 69:113-132. Kumar G, Jones NC, Morris MJ, Rees S, O'Brien TJ, Salzberg MR (2011) Early life stress enhancement of limbic epileptogenesis in adult rats: mechanistic insights. PLoS One 6:e24033. Kwak SE, Kim JE, Kim SC, Kwon OS, Choi SY, Kang TC (2008) Hyperthermic seizure induces persistent alteration in excitability of the dentate gyrus in immature rats. Brain Res 1216:1-15. Kwan P, Arzimanoglou A, Berg AT, Brodie MJ, Allen Hauser W, Mathern G, Moshe SL, Perucca E, Wiebe S, French J (2010) Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Epilepsia 51:1069-1077.

223

ADDENDUM

Lagenstein I, Willig RP, Iffland E (1978) Standardised ACTH and dexamethasone therapy of convulsions in early childhood. I. Clinical results (author's transl). Monatsschr Kinderheilkd 126:492-499. Lai MC, Holmes GL, Lee KH, Yang SN, Wang CA, Wu CL, Tiao MM, Hsieh CS, Lee CH, Huang LT (2006) Effect of neonatal isolation on outcome following neonatal seizures in rats--the role of corticosterone. Epilepsy Res 68:123-136. Lai MC, Huang LT (2011) Effects of early life stress on neuroendocrine and neurobehavior: mechanisms and implications. Pediatr Neonatol 52:122-129. Lai MC, Lui CC, Yang SN, Wang JY, Huang LT (2009) Epileptogenesis is increased in rats with neonatal isolation and early-life seizure and ameliorated by MK-801: a long-term MRI and histological study. Pediatr Res 66:441-447. Le S, Shafer PO, Bartfeld E, Fisher RS (2011) An online diary for tracking epilepsy. Epilepsy Behav 22:705-709. Lemmens EM, Lubbers T, Schijns OE, Beuls EA, Hoogland G (2005) Gender differences in febrile seizure-induced proliferation and survival in the rat dentate gyrus. Epilepsia 46:1603-1612. LeTourneau DJ, Merren MD (1973) Experience with electroencephalography in student naval aviation personnel, 1961-1971: a preliminary report. Aerosp Med 44:1302-1304. Leussis MP, Heinrichs SC (2009) Quality of Rearing Guides Expression of Behavioral and Neural Seizure Phenotypes in El Mice. Brain Res Levin M (1950) The pathogenesis of narcoleptic and epileptic seizures occurring under acute emotional stress. J Nerv Ment Dis 111:101-108. Li J, Vestergaard M, Obel C, Precht DH, Christensen J, Lu M, Olsen J (2008) Prenatal stress and epilepsy in later life: a nationwide follow-up study in Denmark. Epilepsy Res 81:52-57. Lightman SL (2008) The neuroendocrinology of stress: a never ending story. J Neuroendocrinol 20:880-884. Lightman SL, Conway-Campbell BL (2010) The crucial role of pulsatile activity of the HPA axis for continuous dynamic equilibration. Nat Rev Neurosci 11:710-718. Loddenkemper T, Vendrame M, Zarowski M, Gregas M, Alexopoulos AV, Wyllie E, Kothare SV (2011) Circadian patterns of pediatric seizures. Neurology 76:145-153. Loi M, Koricka S, Lucassen PJ, Joëls M (2014) Age- and sex-dependent effects of early life stress on hippocampal neurogenesis. Front Endocrinol (Lausanne) 5:13. Loscher W (2011) Critical review of current animal models of seizures and epilepsy used in the discovery and development of new antiepileptic drugs. Seizure 20:359-368. Loyning Y (1993) Seizures and traffic risks. Tidsskr Nor Laegeforen 113:2231.

224

References

Lu Y, Waltz S, Stenzel K, Muhle H, Stephani U (2008) Photosensitivity in epileptic syndromes of childhood and adolescence. Epileptic Disord 10:136-143. Lucking CH, Creutzfeldt OD, Heinemann U (1970) Visual evoked potentials of patients with epilepsy and of a control group. Electroencephalogr Clin Neurophysiol 29:557-566. Ludlow A, Mohr B, Whitmore A, Garagnani M, Pulvermuller F, Gutierrez R (2014) Auditory processing and sensory behaviours in children with autism spectrum disorders as revealed by mismatch negativity. Brain Cogn 86:55-63. Lundgren T, Dahl J, Yardi N, Melin L (2008) Acceptance and Commitment Therapy and yoga for drug-refractory epilepsy: a randomized controlled trial. Epilepsy Behav 13:102-108. Lupien SJ, McEwen BS, Gunnar MR, Heim C (2009) Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nat Rev Neurosci 10:434-445. Lyons-Ruth K, Wolfe R, Lyubchik A (2000) Depression and the parenting of young children: making the case for early preventive mental health services. Harv Rev Psychiatry 8:148-153. Magarinos AM, McEwen BS (1995) Stress-induced atrophy of apical dendrites of hippocampal CA3c neurons: involvement of glucocorticoid secretion and excitatory amino acid receptors. Neuroscience 69:89-98. Mall RK, Agarwal A, Garg RK, Kar AM, Shukla R (2003) Short course of prednisolone in Indian patients with solitary cysticercus granuloma and new-onset seizures. Epilepsia 44:1397-1401. Mangeot SD, Miller LJ, McIntosh DN, McGrath-Clarke J, Simon J, Hagerman RJ, Goldson E (2001) Sensory modulation dysfunction in children with attention-deficit-hyperactivity disorder. Dev Med Child Neurol 43:399-406. Mani KS, Rangan G, Srinivas HV, Kalyanasundaram S, Narendran S, Reddy AK (1998) The Yelandur study: a community-based approach to epilepsy in rural South India--epidemiological aspects. Seizure 7:281-288. Marco EJ, Khatibi K, Hill SS, Siegel B, Arroyo MS, Dowling AF, Neuhaus JM, Sherr EH, Hinkley LN, Nagarajan SS (2012) Children with autism show reduced somatosensory response: an MEG study. Autism Res 5:340-351. Marek B, Kajdaniuk D, Kos-Kudla B, Kapustecki J, Swietochowska E, Ostrowska Z, Sieminska L, Nowak M, Glogowska-Szelag J, Borgiel-Marek H, Ciesielska-Kopacz N, Foltyn W, Pierzchala K, Krysiak R, Bienek R (2010) Mean daily plasma concentrations of beta-endorphin, leu-enkephalin, ACTH, cortisol, and DHEAS in epileptic patients with complex partial seizures evolving to generalized tonic-clonic seizures. Endokrynol Pol 61:103-110. Marrosu F, Fratta W, Carcangiu P, Giagheddu M, Gessa GL (1988) Localized epileptiform activity induced by murine CRF in rats. Epilepsia 29:369-373. Marrosu F, Mereu G, Fratta W, Carcangiu P, Camarri F, Gessa GL (1987) Different epileptogenic activities of murine and ovine corticotropin-releasing factor. Brain Res 408:394-398.

225

ADDENDUM

Masood K, Ahmed B, Choi J, Gutierrez-Osuna R (2012) Consistency and validity of self-reporting scores in stress measurement surveys. Conf Proc IEEE Eng Med Biol Soc 2012:4895-4898. Matsumoto K, Nomura H, Murakami Y, Taki K, Takahata H, Watanabe H (2003) Long-term social isolation enhances picrotoxin seizure susceptibility in mice: up-regulatory role of endogenous brain allopregnanolone in GABAergic systems. Pharmacol Biochem Behav 75:831-835. Mattson RH (1991) Emotional effects on seizure occurrence. Adv Neurol 55:453-460. Mattson RH, Heninger GR, Gallagher BB, Glaser GH (1970) Psychophysiologic precipitants of seizures in epileptics. Neurology 20:407. McCormick DA, Contreras D (2001) On the cellular and network bases of epileptic seizures. Annu Rev Physiol 63:815-846. McCorry LK (2007) Physiology of the autonomic nervous system. Am J Pharm Educ 71:78. McEwen BS (2007) Physiology and neurobiology of stress and adaptation: central role of the brain. Physiol Rev 87:873-904. McEwen BS (1998) Stress, adaptation, and disease. Allostasis and allostatic load. Ann N Y Acad Sci 840:33-44. McGowan PO, Sasaki A, D'Alessio AC, Dymov S, Labonte B, Szyf M, Turecki G, Meaney MJ (2009) Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nat Neurosci 12:342-348. Meaney MJ, Diorio J, Francis D, Widdowson J, LaPlante P, Caldji C, Sharma S, Seckl JR, Plotsky PM (1996) Early environmental regulation of forebrain glucocorticoid receptor gene expression: implications for adrenocortical responses to stress. Dev Neurosci 18:49-72. Meaney MJ, Szyf M (2005) Environmental programming of stress responses through DNA methylation: life at the interface between a dynamic environment and a fixed genome. Dialogues Clin Neurosci 7:103-123. Meijman TFT, Thunnissen MJ, de Vries-Griever AG (1990) The after-effects of a prolonged period of day-sleep on subjective sleep quality. Work Stress 4:65-70. Mendez M, Radtke RA (2001) Interactions between sleep and epilepsy. J Clin Neurophysiol 18:106127. Miller LJ, Anzalone ME, Lane SJ, Cermak SA, Osten ET (2007) Concept evolution in sensory integration: a proposed nosology for diagnosis. Am J Occup Ther 61:135-140. Miller LJ, Nielsen DM, Schoen SA (2012) Attention deficit hyperactivity disorder and sensory modulation disorder: a comparison of behavior and physiology. Res Dev Disabil 33:804-818. Miller LJ, Nielsen DM, Schoen SA, Brett-Green BA (2009) Perspectives on sensory processing disorder: a call for translational research. Front Integr Neurosci 3:22.

226

References

Mills JNÂ (1975) Developement of circadian rhythms in infancy. Chronobiologia 2:363-371. Mitra R, Sapolsky RM (2008) Acute corticosterone treatment is sufficient to induce anxiety and amygdaloid dendritic hypertrophy. Proc Natl Acad Sci U S A 105:5573-5578. Mitsonis CI, Potagas C, Zervas I, Sfagos K (2009) The effects of stressful life events on the course of multiple sclerosis: a review. Int J Neurosci 119:315-335. Moore DL, Tarnai J (2002). Evaluating nonresponse error in mail surveys. In Survey Nonresponse. Edited by Groves RM, Dillman DA, Eltinge JL, Little RJA. . John Wiley & Sons; 197-211. Morris GL 3rd, Gloss D, Buchhalter J, Mack KJ, Nickels K, Harden C (2013) Evidence-based guideline update: vagus nerve stimulation for the treatment of epilepsy: report of the Guideline Development Subcommittee of the American Academy of Neurology. Neurology 81:1453-1459. Moseley B, Bateman L, Millichap JJ, Wirrell E, Panayiotopoulos CP (2013) Autonomic epileptic seizures, autonomic effects of seizures, and SUDEP. Epilepsy Behav 26:375-385. Moshe S, Shilo M, Chodick G, Yagev Y, Blatt I, Korczyn AD, Neufeld MY (2008) Occurrence of seizures in association with work-related stress in young male army recruits. Epilepsia 49:1451-1456. Mostofsky DI, Balaschak BA (1977) Psychobiological control of seizures. Psychol Bull 84:723-750. Motta E, Rosciszewska D, Buntner B (1994) Serum ACTH levels in patients treated for epilepsy. Neurol Neurochir Pol 28:317-323. Mulligan SÂ (2002) Advances in sensory integration research. In: Sensory integration: Theory and practice, 2nd edition (Bundy AC, Lane S, Murray EA eds), pp 397-411. Philadelphia: F.A. Davis. Multani P, Myers RH, Blume HW, Schomer DL, Sotrel A (1994) Neocortical dendritic pathology in human partial epilepsy: a quantitative Golgi study. Epilepsia 35:728-736. Murgatroyd C, Spengler D (2011) Epigenetic programming of the HPA axis: early life decides. Stress 14:581-589. Nagamitsu S, Matsuishi T, Yamashita Y, Shimizu T, Iwanaga R, Murakami Y, Miyazaki M, Hashimoto T, Kato H (2001) Decreased cerebrospinal fluid levels of beta-endorphin and ACTH in children with infantile spasms. J Neural Transm 108:363-371. Nakken KO, Solaas MH, Kjeldsen MJ, Friis ML, Pellock JM, Corey LA (2005) Which seizureprecipitating factors do patients with epilepsy most frequently report? Epilepsy Behav 6:85-89. Nalin A, Facchinetti F, Galli V, Petraglia F, Storchi R, Genazzani AR (1985) Reduced ACTH content in cerebrospinal fluid of children affected by cryptogenic infantile spasms with hypsarrhythmia. Epilepsia 26:446-449. Nater UM, Rohleder N (2009). Salivary alpha-amylase as a non-invasive biomarker for the sympathethic nervous system: current state of research. Psychoneuroendocrinology 34: 486-496.

227

ADDENDUM

Nemeroff CB (2004) Neurobiological consequences of childhood trauma. J Clin Psychiatry 65 Suppl 1:18-28. Neufeld MY, Sadeh M, Cohn DF, Korczyn AD (1994) Stress and epilepsy: the Gulf war experience. Seizure 3:135-139. Neugebauer R, Paik M, Hauser WA, Nadel E, Leppik I, Susser M (1994) Stressful life events and seizure frequency in patients with epilepsy. Epilepsia 35:336-343. Ngugi AK, Kariuki SM, Bottomley C, Kleinschmidt I, Sander JW, Newton CR (2011) Incidence of epilepsy: a systematic review and meta-analysis. Neurology 77:1005-1012. Nielsen KA, Dahl M, Tommerup E, Hansen B, Erdal J, Wolf P (2008) Diurnal lamotrigine plasma level fluctuations: clinical significance and indication of shorter half-life with chronic administration. Epilepsy Behav 13:470-473. Nihalani ND, Schwartz TL (2007) Mifepristone, a glucocorticoid antagonist for the potential treatment of psychotic major depression. Curr Opin Investig Drugs 8:563-569. Nishimura A, Honda N, Sugioka N, Takada K, Shibata N (2008) Evaluation of carbamazepine pharmacokinetic profiles in mice with kainic acid-induced acute seizures. Biol Pharm Bull 31:23022308. Noachtar S, Binnie C, Ebersole J, Mauguiere F, Sakamoto A, Westmoreland B (1999) A glossary of terms most commonly used by clinical electroencephalographers and proposal for the report form for the EEG findings. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 52:21-41. Notenboom RG, Ramakers GM, Kamal A, Spruijt BM, de Graan PN (2010) Long-lasting modulation of synaptic plasticity in rat hippocampus after early-life complex febrile seizures. Eur J Neurosci 32:749-758. Nugent NR, Tyrka AR, Carpenter LL, Price LH (2011) Gene-environment interactions: early life stress and risk for depressive and anxiety disorders. Psychopharmacology (Berl) 214:175-196. O'Connor P (1964) Analysis of 500 routine EEG's of R.A.F. aircrew cadets. Electroenceph clin Neurophysiol 17:341. O'Connor TG, Ben-Shlomo Y, Heron J, Golding J, Adams D, Glover V (2005) Prenatal anxiety predicts individual differences in cortisol in pre-adolescent children. Biol Psychiatry 58:211-217. Oguni H, Tanaka T, Hayashi K, Funatsuka M, Sakauchi M, Shirakawa S, Osawa M (2002) Treatment and long-term prognosis of myoclonic-astatic epilepsy of early childhood. Neuropediatrics 33:122132. O'Neill D (1958) Stress and epilepsy. Med World 89:425-428. Onishi S, Miyazawa G, Nishimura Y, Sugiyama S, Yamakawa T, Inagaki H, Katoh T, Itoh S, Isobe K (1983) Postnatal development of circadian rhythm in serum cortisol levels in children. Pediatrics 72:399-404.

228

References

Oomen CA, Mayer JL, de Kloet ER, Joëls M, Lucassen PJ (2007) Brief treatment with the glucocorticoid receptor antagonist mifepristone normalizes the reduction in neurogenesis after chronic stress. Eur J Neurosci 26:3395-3401. Oomen CA, Soeters H, Audureau N, Vermunt L, van Hasselt FN, Manders EM, Joëls M, Lucassen PJ, Krugers H (2010) Severe early life stress hampers spatial learning and neurogenesis, but improves hippocampal synaptic plasticity and emotional learning under high-stress conditions in adulthood. J Neurosci 30:6635-6645. Oostrom KJ, Smeets-Schouten A, Kruitwagen CL, Peters AC, Jennekens-Schinkel A, Dutch Study Group of Epilepsy in Childhood (2003) Not only a matter of epilepsy: early problems of cognition and behavior in children with "epilepsy only"--a prospective, longitudinal, controlled study starting at diagnosis. Pediatrics 112:1338-1344. Ornitz E, Guthrie D, Farley AJ (1978) The early symptoms of childhood autism. In: Cognitive defects in the development of mental illness, (Sherban G ed), pp 207-229. New York: Brunner/Mazel. Ouardouz M, Lema P, Awad PN, Di Cristo G, Carmant L (2010) N-methyl-D-aspartate, hyperpolarization-activated cation current (Ih) and gamma-aminobutyric acid conductances govern the risk of epileptogenesis following febrile seizures in rat hippocampus. Eur J Neurosci 31:1252-1260. Owen JP, Marco EJ, Desai S, Fourie E, Harris J, Hill SS, Arnett AB, Mukherjee P (2013) Abnormal white matter microstructure in children with sensory processing disorders. Neuroimage Clin 2:844-853. Palkovits M, Zaborszky L, Feminger A, Mezey E, Fekete MI, Herman JP, Kanyicska B, Szabo D (1980) Noradrenergic innervation of the rat hypothalamus:experimental biochemical and electron microscopic studies. Brain Res 191:161-171. Panjwani U, Selvamurthy W, Singh SH, Gupta HL, Thakur L, Rai UC (1996) Effect of Sahaja yoga practice on seizure control & EEG changes in patients of epilepsy. Indian J Med Res 103:165-172. Pavlova MK, Shea SA, Bromfield EB (2004) Day/night patterns of focal seizures. Epilepsy Behav 5:44-49. Pavlova MK, Woo Lee J, Yilmaz F, Dworetzky BA (2012) Diurnal pattern of seizures outside the hospital: is there a time of circadian vulnerability? Neurology 78:1488-1492. Pelat C, Turbelin C, Bar-Hen A, Flahault A, Valleron A (2009) More diseases tracked by using Google Trends. Emerg Infect Dis 15:1327-1328. Pellock JM (2008) Chapter III: Drug Treatment in Children. In: Epilepsy: A Comprehensive Textbook, 2nd edition (Engel Jr J, Pedley TA, Aicardi J eds), pp 1249-1258. Philadelphia: Lippincott Williams & Wilkins. Pellock JM (2004) Defining the problem: psychiatric and behavioral comorbidity in children and adolescents with epilepsy. Epilepsy Behav 5 Suppl 3:S3-9. Pentella K, Bachman DS, Sandman CA (1982) Trial of an ACTH4-9 Analogue (ORG 2766) in children with intractable seizures. Neuropediatrics 13:59-62.

229

ADDENDUM

Perez-Cruz C, Likhodii S, Burnham WM (2006) Deoxycorticosterone's anticonvulsant effects in infant rats are blocked by finasteride, but not by indomethacin. Exp Neurol 200:283-289. Phillips DIÂ (2007) Programming of the stress response: a fundamental mechanism underlying the long-term effects of the fetal environment? J Intern Med 261:453-460. Phillips LJ, McGorry PD, Garner B, Thompson KN, Pantelis C, Wood SJ,Berger G (2006) Stress, the hippocampus and the hypothalamic-pituitary-adrenal axis: implications for the development of psychotic disorders. Aust N Z J Psychiatry 40:725-741. Pillai J, Sperling MR (2006) Interictal EEG and the diagnosis of epilepsy. Epilepsia 47 Suppl 1:14-22. Pinikahana J, Dono J (2009) The lived experience of initial symptoms of and factors triggering epileptic seizures. Epilepsy Behav 15:513-520. Pitkanen AÂ (2010) Therapeutic approaches to epileptogenesis--hope on the horizon. Epilepsia 51 Suppl 3:2-17. Pitkanen A, Pietila J, Makinen P, Huhtala A,Uusitalo H (2010) Comparison of laser in situ keratomileusis reoperation outcomes with the Moria M2 head 90 and 130 following previous photorefractive keratectomy or laser in situ keratomileusis. Acta Ophthalmol 88:352-357. Pitkanen A, Sutula TP (2002) Is epilepsy a progressive disorder? Prospects for new therapeutic approaches in temporal-lobe epilepsy. Lancet Neurol 1:173-181. Plouin P, Jalin C, Dulac O, Chiron C (1987) Ambulatory 24-hour EEG recording in epileptic infantile spasms. Rev Electroencephalogr Neurophysiol Clin 17:309-318. Polak EL, Privitera MD, Lipton RB, Haut SR (2012) Behavioral intervention as an add-on therapy in epilepsy: designing a clinical trial. Epilepsy Behav 25:505-510. Polgreen PM, Chen Y, Pennock DM, Nelson FD (2008) Using internet searches for influenza surveillance. Clin Infect Dis 47:1443-1448. Porter SR, Whitcomb ME (2005) Non-Response in Student Surveys: The Role of Demographics, Engagement and Personality. Research in Higher Education 46:127-152. Price DA, Close GC, Fielding BA (1983) Age of appearance of circadian rhythm in salivary cortisol values in infancy. Arch Dis Child 58:454-456. Pritchard PB 3rd, Wannamaker BB, Sagel J, Daniel CM (1985) Serum prolactin and cortisol levels in evaluation of pseudoepileptic seizures. Ann Neurol 18:87-89. Privitera M, Walters M, Lee I, Polak E, Fleck A, Schwieterman D, Haut SR (2014) Characteristics of people with self-reported stress-precipitated seizures. Epilepsy Behav 41C:74-77. Pruessner JC, Wolf OT, Hellhammer DH, Buske-Kirschbaum A, von Auer K, Jobst S, Kaspers F, Kirschbaum C (1997) Free cortisol levels after awakening: a reliable biological marker for the assessment of adrenocortical activity. Life Sci 61:2539-2549.

230

References

Puskarich CA, Whitman S, Dell J, Hughes JR, Rosen AJ, Hermann BP (1992) Controlled examination of effects of progressive relaxation training on seizure reduction. Epilepsia 33:675-680. Qian X, Droste SK, Lightman SL, Reul JM, Linthorst AC (2012) Circadian and ultradian rhythms of free glucocorticoid hormone are highly synchronized between the blood, the subcutaneous tissue, and the brain. Endocrinology 153:4346-4353. Quigg M, Straume M, Menaker M, Bertram EH 3rd (1998) Temporal distribution of partial seizures: comparison of an animal model with human partial epilepsy. Ann Neurol 43:748-755. Qulu L, Daniels WM, Mabandla MV (2012) Exposure to prenatal stress enhances the development of seizures in young rats. Metab Brain Dis 27:399-404. Rajesh B, Jayachandran D, Mohandas G, Radhakrishnan K (2006) A pilot study of a yoga meditation protocol for patients with medically refractory epilepsy. J Altern Complement Med 12:367-371. Ramaratnam S, Baker GA, Goldstein LH (2008) Psychological treatments for epilepsy. Cochrane Database Syst Rev (3):CD002029. doi:CD002029. Ramgopal S, Shah A, Zarowski M, Vendrame M, Gregas M, Alexopoulos AV, Loddenkemper T, Kothare SV (2012) Diurnal and sleep/wake patterns of epileptic spasms in different age groups. Epilepsia 53:1170-1177. Ramgopal S, Vendrame M, Shah A, Gregas M, Zarowski M, Rotenberg A, Alexopoulos AV, Wyllie E, Kothare SV, Loddenkemper T (2012) Circadian patterns of generalized tonic-clonic evolutions in pediatric epilepsy patients. Seizure 21:535-539. Rao ML, Stefan H, Bauer J (1989) Epileptic but not psychogenic seizures are accompanied by simultaneous elevation of serum pituitary hormones and cortisol levels. Neuroendocrinology 49:33-39. Ratey JJ, Johnson C (1997) Shadow Syndromes. New York: Pantheon Books. Reddy DS, Rogawski MA (2002) Stress-induced deoxycorticosterone-derived neurosteroids modulate GABA(A) receptor function and seizure susceptibility. J Neurosci 22:3795-3805. Reid AY, Galic MA, Teskey GC, Pittman QJ (2009) Febrile seizures: current views and investigations. Can J Neurol Sci 36:679-686. Reul JM, de Kloet ER (1985). Two receptor systems for corticosterone in rat brain: Microdistribution and differential occupation. Endocrinology 117:2505-2511. Reynolds EH, Trimble MR (2009) Epilepsy, psychiatry, and neurology. Epilepsia 50 Suppl 3:50-55. Rho JM, Strafstrom CE (2006) Neurophysiology of epilepsy. In: Pediatric Neurology: Principles and Practice, 4th edition (Swaiman KF, Ashwal S, Ferreiro DM eds), pp 991-1007. Philadelphia: Mosby Elsevier. Ringold GM (1985) Steroid hormone regulation of gene expression. Annu Rev Pharmacol Toxicol 25:529-566.

231

ADDENDUM

Riva R, Contin M, Albani F, Perucca E, Ambrosetto G, Gobbi G, Cortelli P, Procaccianti G, Baruzzi A (1984) Free and total plasma concentrations of carbamazepine and carbamazepine-10,11-epoxide in epileptic patients: diurnal fluctuations and relationship with side effects. Ther Drug Monit 6:408413. Roberts AJ, Keith LD (1994) Mineralocorticoid receptors mediate the enhancing effects of corticosterone on convulsion susceptibility in mice. J Pharmacol Exp Ther 270:505-511. Roberts D, Dalziel S (2006) Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. Cochrane Database Syst Rev (3):CD004454. Robertson MM, Coppen A, Trimble MR (1986) The dexamethasone suppression test in medicated epileptic patients. Biol Psychiatry 21:225-228. Roopra A, Dingledine R, Hsieh J (2012) Epigenetics and epilepsy. Epilepsia 53 Suppl 9:2-10. Rose NÂ (2014) The Human Brain Project: social and ethical challenges. Neuron 82:1212-1215. Rousseau A, Hermann B, Whitman S (1985) Effects of progressive relaxation on epilepsy: analysis of a series of cases. Psychol Rep 57:1203-1212. Ryvlin P, Cross JH, Rheims S (2014) Epilepsy surgery in children and adults. Lancet Neurol 13:11141126. Sadaghiani MM, Saboory E (2010) Prenatal stress potentiates pilocarpine-induced epileptic behaviors in infant rats both time and sex dependently. Epilepsy Behav 18:166-170. Salinsky M, Kanter R, Dasheiff RM (1987) Effectiveness of multiple EEGs in supporting the diagnosis of epilepsy: an operational curve. Epilepsia 28:331-334. Salzberg M, Kumar G, Supit L, Jones NC, Morris MJ, Rees S, O'Brien TJ (2007) Early postnatal stress confers enduring vulnerability to limbic epileptogenesis. Epilepsia 48:2079-2085. Sanchez Fernandez I, Ramgopal S, Powell C, Gregas M, Zarowski M, Shah A, Vendrame M, Alexopoulos AV, Kothare SV, Loddenkemper T (2013) Clinical evolution of seizures: distribution across time of day and sleep/wakefulness cycle. J Neurol 260:549-557. Sanchez MM, Ladd CO, Plotsky PM (2001) Early adverse experience as a developmental risk factor for later psychopathology: evidence from rodent and primate models. Dev Psychopathol 13:419-449. Sankoh AJ, Huque MF, Dubey SD (1997) Some comments on frequently used multiple endpoint adjustment methods in clinical trials. Stat Med 16:2529-2542. Santiago LB, Jorge SM, Moreira AC (1996) Longitudinal evaluation of the development of salivary cortisol circadian rhythm in infancy. Clin Endocrinol (Oxf) 44:157-161. Sapolsky RM, Romero LM,Munck AU (2000) How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocr Rev 21:55-89. Sarabdjitsingh RA, Jezequel J, Pasricha N, Mikasova L, Kerkhofs A, Karst H, Groc L, JoĂŤls M (2014)

232

References

Ultradian corticosterone pulses balance glutamatergic transmission and synaptic plasticity. Proc Natl Acad Sci U S A 111:14265-14270. Sathyaprabha TN, Satishchandra P, Pradhan C, Sinha S, Kaveri B, Thennarasu K, Murthy BT, Raju TR (2008) Modulation of cardiac autonomic balance with adjuvant yoga therapy in patients with refractory epilepsy. Epilepsy Behav 12:245-252. Satishchandra P (2003) Hot-water epilepsy. Epilepsia 44 Suppl 1:29-32. Sawyer NT, Escayg A (2010) Stress and epilepsy: multiple models, multiple outcomes. J Clin Neurophysiol 27:445-452. Schatzberg AF, Lindley S (2008) Glucocorticoid antagonists in neuropsychiatric [corrected] disorders. Eur J Pharmacol 583:358-364. Schatzberg AF, Rothschild AJ, Langlais PJ, Bird ED, Cole JO (1985) A corticosteroid/dopamine hypothesis for psychotic depression and related states. J Psychiatr Res 19:57-64. Schiller Y, Najjar Y (2008) Quantifying the response to antiepileptic drugs: effect of past treatment history. Neurology 70:54-65. Schridde U, Strauss U, Brauer AU, van Luijtelaar G (2006) Environmental manipulations early in development alter seizure activity, Ih and HCN1 protein expression later in life. Eur J Neurosci 23:3346-3358. Schridde U, van Luijtelaar G (2004) Corticosterone increases spike-wave discharges in a dose- and time-dependent manner in WAG/Rij rats. Pharmacol Biochem Behav 78:369-375. Schumacher S, Kirschbaum C, Fydrich T, Strohle A (2013). Is salivary alpha-amylase an indicator of autonomic nervous system dysregulations in mental disorders?–a review of preliminary findings and interactions with cortisol. Psychoneuroendocrinology 38: 729-743. Schwabe L, Joëls M, Roozendaal B, Wolf OT, Oitzl MS (2012) Stress effects on memory: an update and integration. Neurosci Biobehav Rev 36:1740-1749. Seckl JR (2008) Glucocorticoids, developmental 'programming' and the risk of affective dysfunction. Prog Brain Res 167:17-34. Seifter A, Schwarzwalder A, Geis K, Aucott J (2010) The utility of "Google Trends" for epidemiological research: Lyme disease as an example. Geospat Health 4:135-137. Selye H (1936) A syndrome produced by diverse nocuous agents. Nature 138:32. Selye H (1950) Stress and the general adaptation syndrome. Br Med J 1:1383-1392. Seri S, Cerquiglini A, Pisani F (1998) Spike-induced interference in auditory sensory processing in Landau-Kleffner syndrome. Electroencephalogr Clin Neurophysiol 108:506-510. Shang NX, Zou LP, Zhao JB, Zhang F, Li H (2010) Association between prenatal stress and infantile spasms: a case-control study in China. Pediatr Neurol 42:181-186.

233

ADDENDUM

Shansky RM, Morrison JH (2009) Stress-induced dendritic remodeling in the medial prefrontal cortex: effects of circuit, hormones and rest. Brain Res 1293:108-113. Sharma S, Jain P (2014) The ketogenic diet and other dietary treatments for refractory epilepsy in children. Ann Indian Acad Neurol 17:253-258. Shinnar S, Pellock JM (2002) Update on the epidemiology and prognosis of pediatric epilepsy. J Child Neurol 17 Suppl 1:S4-17. Shirtcliff EA, Allison AL, Armstrong JM, Slattery MJ, Kalin NH, Essex MJ (2012) Longitudinal stability and developmental properties of salivary cortisol levels and circadian rhythms from childhood to adolescence. Dev Psychobiol 54:493-502. Shively CA, Musselman DL, Willard SL (2009) Stress, depression, and coronary artery disease: modeling comorbidity in female primates. Neurosci Biobehav Rev 33:133-144. Singh SP, He X, McNamara JO, Danzer SC (2013) Morphological changes among hippocampal dentate granule cells exposed to early kindling-epileptogenesis. Hippocampus 23:1309-1320. Slopen N, McLaughlin KA, Shonkoff JP (2014) Interventions to improve cortisol regulation in children: a systematic review. Pediatrics 133:312-326. Smith-Swintosky VL, Pettigrew LC, Sapolsky RM, Phares C, Craddock SD, Brooke SM, Mattson MP (1996) Metyrapone, an inhibitor of glucocorticoid production, reduces brain injury induced by focal and global ischemia and seizures. J Cereb Blood Flow Metab 16:585-598. Spangler GÂ (1991) The emergence of adrenocortical circadian function in newborns and infants and its relationship to sleep, feeding and maternal adrenocortical activity. Early Hum Dev 25:197-208. Spatt J, Langbauer G, Mamoli B (1998) Subjective perception of seizure precipitants: results of a questionnaire study. Seizure 7:391-395. Spector S, Cull C, Goldstein LH (2000) Seizure precipitants and perceived self-control of seizures in adults with poorly-controlled epilepsy. Epilepsy Res 38:207-216. Spencer S, Huh L (2008) Outcomes of epilepsy surgery in adults and children. Lancet Neurol 7:525537. Spencer SS, Goncharova II, Duckrow RB, Novotny EJ, Zaveri HP (2008) Interictal spikes on intracranial recording: behavior, physiology, and implications. Epilepsia 49:1881-1892. Sperling MR, Schilling CA, Glosser D, Tracy JI, Asadi-Pooya AA (2008) Self-perception of seizure precipitants and their relation to anxiety level, depression, and health locus of control in epilepsy. Seizure 17:302-307. Stell BM, Brickley SG, Tang CY, Farrant M, Mody I (2003) Neuroactive steroids reduce neuronal excitability by selectively enhancing tonic inhibition mediated by delta subunit-containing GABAA receptors. Proc Natl Acad Sci U S A 100:14439-14444.

234

References

Stevanovic D, Jancic J, Lakic A (2011) The impact of depression and anxiety disorder symptoms on the health-related quality of life of children and adolescents with epilepsy. Epilepsia 52:e75-8. Stevens JR (1959) Emotional activation of the electroencephalogram in patients with convulsive disorders. J Nerv Ment Dis 128:339-351. Stott DH (1973) Follow-up study from birth of the effects of prenatal stresses. Dev Med Child Neurol 15:770-787. Swanson LW, Sawchenko PE, Rivier J, Vale WW (1983) Organization of ovine corticotropin-releasing factor immunoreactive cells and fibers in the rat brain: an immunohistochemical study. Neuroendocrinology 36:165-186. Swinkels WA, Engelsman M, Kasteleijn-Nolst Trenite DG, Baal MG, de Haan GJ, Oosting J (1998) Influence of an evacuation in February 1995 in The Netherlands on the seizure frequency in patients with epilepsy: a controlled study. Epilepsia 39:1203-1207. Symonds C (1959) Excitation and inhibition in epilepsy. Proc R Soc Med 52:395-402. Taal MN, Rietman AB, Meulen SV, Schipper M, Dejonckere PH (2013) Children with specific language impairment show difficulties in sensory modulation. Logoped Phoniatr Vocol 38:70-78. Takeshita H, Kawahara R, Nagabuchi T, Mizukawa R, Hazama H (1986) Serum prolactin, cortisol and growth hormone concentrations after various epileptic seizures. Jpn J Psychiatry Neurol 40:617623. Talay-Ongan A, Wood K (2000) Unusual sensory sensitivities in autism: A possible cross-roads. International Journal of Disability, Development and Education 47:201-212. Tasker JG, Di S, Malcher-Lopes R (2006) Minireview: rapid glucocorticoid signaling via membraneassociated receptors. Endocrinology 147:5549-5556. Taylor RS, Sander JW, Taylor RJ, Baker GA (2011) Predictors of health-related quality of life and costs in adults with epilepsy: a systematic review. Epilepsia 52:2168-2180. Teicher MH, Tomoda A, Andersen SL (2006) Neurobiological consequences of early stress and childhood maltreatment: are results from human and animal studies comparable? Ann N Y Acad Sci 1071:313-323. Temkin NR, Davis GR (1984) Stress as a risk factor for seizures among adults with epilepsy. Epilepsia 25:450-456. Tennant C (2002) Life events, stress and depression: a review of recent findings. Aust N Z J Psychiatry 36:173-182. Thapar A, Kerr M, Harold G (2009) Stress, anxiety, depression, and epilepsy: investigating the relationship between psychological factors and seizures. Epilepsy Behav 14:134-140. The Lancet N (2014) The Human Brain Project: mutiny on the flagship. Lancet Neurol 13:8554422(14)70181-4.

235

ADDENDUM

Tononi G, Edelman GM, Sporns O (1998) Complexity and coherency: integrating information in the brain. Trends Cogn Sci 2:474-484. Tosevski DL, Milovancevic MP (2006) Stressful life events and physical health. Curr Opin Psychiatry 19:184-189. Tsigos C, Chrousos GP (2002) Hypothalamic-pituitary-adrenal axis, neuroendocrine factors and stress. J Psychosom Res 53:865-871. Tuveri A, Paoletti AM, Orru M, Melis GB, Marotto MF, Zedda P, Marrosu F, Sogliano C, Marra C, Biggio G,Concas A (2008) Reduced serum level of THDOC, an anticonvulsant steroid, in women with perimenstrual catamenial epilepsy. Epilepsia 49:1221-1229. Ulrich-Lai YM, Herman JP (2009) Neural regulation of endocrine and autonomic stress responses. Nat Rev Neurosci 10:397-409. Valentino RJ, Van Bockstaele E (2008) Convergent regulation of locus coeruleus activity as an adaptive response to stress. Eur J Pharmacol 583:194-203. Van Campen JS, Jansen FE, de Graan PN, Braun KP, Joëls M (2014) Early life stress in epilepsy: A seizure precipitant and risk factor for epileptogenesis. Epilepsy Behav 38C:160-171. Van Campen JS, Jansen FE, Steinbusch LC, Joëls M, Braun KP (2012) Stress sensitivity of childhood epilepsy is related to experienced negative life events. Epilepsia 53:1554-1562. Van Cauter E, Leproult R, Kupfer DJ (1996) Effects of gender and age on the levels and circadian rhythmicity of plasma cortisol. J Clin Endocrinol Metab 81:2468-2473. Van den Bos R, Taris R, Scheppink B, de Haan L, Verster JC (2014) Salivary cortisol and alpha-amylase levels during an assessment procedure correlate differently with risk-taking measures in male and female police recruits. Front Behav Neurosci 7:219. Van Diessen E, Diederen SJ, Braun KP, Jansen FE, Stam CJ (2013) Functional and structural brain networks in epilepsy: what have we learned? Epilepsia 54:1855-1865. van Dijk H, Tellegen PJ (2004(. Nederlandse intelligentietest voor onderwijsniveau, handleiding en verantwoording (Dutch Intelligence test for schooling level, manual and justification). Amsterdam: Boom testuitgevers. Van Mierlo P, Papadopoulou M, Carrette E, Boon P, Vandenberghe S, Vonck K,Marinazzo D (2014) Functional brain connectivity from EEG in epilepsy: seizure prediction and epileptogenic focus localization. Prog Neurobiol 121:19-35. Van Widenfelt BM, Goedhart AW, Treffers PD, Goodman R (2003) Dutch version of the Strengths and Difficulties Questionnaire (SDQ). Eur Child Adolesc Psychiatry 12:281-289. Veerman JW, ten Brink LT, van der Horst H, Koedoot P (1993) Vragenlijst Meegemaakte Gebeurtenissen (VMG). Voorlopige handleiding 3de herziening (Dutch instruction manual). Amsterdam: ACWS/Praktikon.

236

References

Veldhuis JD, Iranmanesh A, Johnson ML, Lizarralde G (1990) Amplitude, but not frequency, modulation of adrenocorticotropin secretory bursts gives rise to the nyctohemeral rhythm of the corticotropic axis in man. J Clin Endocrinol Metab 71:452-463. Veldhuis JD, Sharma A, Roelfsema F (2013) Age-dependent and gender-dependent regulation of hypothalamic-adrenocorticotropic-adrenal axis. Endocrinol Metab Clin North Am 42:201-225. Velisek L (2011) Prenatal corticosteroid exposure alters early developmental seizures and behavior. Epilepsy Res 95:9-19. Verity CM, Golding J (1991) Risk of epilepsy after febrile convulsions: a national cohort study. BMJ 303:1373-1376. Walczak TS, Scheuer ML, Resor S, Pedley TA (1993) Prevalence and features of epilepsy without interictal epileptiform discharges. Neurology 43:287-288. Walsh WB (1967) Validity of Self-Report. Journal of Counseling Psychology 14:18-23. Wang W, Dow KE, Fraser DD (2001) Elevated corticotropin releasing hormone/corticotropin releasing hormone-R1 expression in postmortem brain obtained from children with generalized epilepsy. Ann Neurol 50:404-409. Wang YJ, Zhang Y, Liang XH, Yang G, Zou LP (2012) Effects of adrenal dysfunction and high-dose adrenocorticotropic hormone on NMDA-induced spasm seizures in young Wistar rats. Epilepsy Res 100:125-131. Wang Z, Lu G, Zhang Z, Zhong Y, Jiao Q, Zhang Z, Tan Q, Tian L, Chen G, Liao W, Li K, Liu Y (2011) Altered resting state networks in epileptic patients with generalized tonic-clonic seizures. Brain Res 1374:134-141. Ward AA Jr (1961) The epileptic neurone. Epilepsia 2:70-80. Wassenaar M, Kasteleijn-Nolst Trenite DG, de Haan GJ, Carpay JA, Leijten FS (2014) Seizure precipitants in a community-based epilepsy cohort. J Neurol 261:717-724. Watzka M, Bidlingmaier F, Beyenburg S, Henke RT, Clusmann H, Elger CE, Schramm J, Klingmüller D, Stoffel-Wagner B (2000). Corticosteroid receptor mRNA expression in the brains of patients with epilepsy. Steroids 65:895-901. Webster A, Mawer GE (1989) Seizure frequency and major life events in epilepsy. Epilepsia 30:162167. Weinshenker D, Szot P (2002) The role of catecholamines in seizure susceptibility: new results using genetically engineered mice. Pharmacol Ther 94:213-233. Weitzman ED, Fukushima D, Nogeire C, Roffwarg H, Gallagher TF, Hellman L (1971) Twenty-four hour pattern of the episodic secretion of cortisol in normal subjects. J Clin Endocrinol Metab 33:1422.

237

ADDENDUM

Wiebe S, Jette N (2012) Epilepsy surgery utilization: who, when, where, and why? Curr Opin Neurol 25:187-193. Williams PA, White AM, Clark S, Ferraro DJ, Swiercz W, Staley KJ, Dudek FE (2009) Development of spontaneous recurrent seizures after kainate-induced status epilepticus. J Neurosci 29:2103-2112. Willig RP, Lagenstein I (1980) Therapeutic trial with a fragment of ACTH (ACTH 4-10) in early childhood epilepsy (author's transl). Monatsschr Kinderheilkd 128:100-103. Wilson JV, Reynolds EH (1990) Texts and documents. Translation and analysis of a cuneiform text forming part of a Babylonian treatise on epilepsy. Med Hist 34:185-198. Wirrell EC, Wood L, Hamiwka LD, Sherman EM (2008) Parenting stress in mothers of children with intractable epilepsy. Epilepsy Behav 13:169-173. Wolf P, Goosses R (1986) Relation of photosensitivity to epileptic syndromes. J Neurol Neurosurg Psychiatry 49:1386-1391. Wu YP, Follansbee-Junger K, Rausch J, Modi A (2014) Parent and family stress factors predict healthrelated quality in pediatric patients with new-onset epilepsy. Epilepsia 55:866-877. Yehuda R, Resnick H, Kahana B, Giller EL (1993) Long-lasting hormonal alterations to extreme stress in humans: normative or maladaptive? Psychosom Med 55:287-297. Young EA, Abelson J, Lightman SL (2004) Cortisol pulsatility and its role in stress regulation and health. Front Neuroendocrinol 25:69-76. Young NA, Teskey GC, Henry LC, Edwards HE (2006) Exogenous antenatal glucocorticoid treatment reduces susceptibility for hippocampal kindled and maximal electroconvulsive seizures in infant rats. Exp Neurol 198:303-312. Yum MS, Chachua T, Veliskova J, Velisek L (2012) Prenatal stress promotes development of spasms in infant rats. Epilepsia 53:e46-9. Zarowski M, Loddenkemper T, Vendrame M, Alexopoulos AV, Wyllie E, Kothare SV (2011) Circadian distribution and sleep/wake patterns of generalized seizures in children. Epilepsia 52:1076-1083. Zhang SW, Liu YX (2008) Changes of serum adrenocorticotropic hormone and cortisol levels during sleep seizures. Neurosci Bull 24:84-88. Zhang Z, Sun QQ (2011) The balance between excitation and inhibition and functional sensory processing in the somatosensory cortex. Int Rev Neurobiol 97:305-333.

238

References

Zijlmans M, Jacobs J, Zelmann R, Dubeau F, Gotman J (2009) High-frequency oscillations mirror disease activity in patients with epilepsy. Neurology 72:979-986. Zobel A, Wellmer J, Schulze-Rauschenbach S, Pfeiffer U, Schnell S, Elger C, Maier W (2004) Impairment of inhibitory control of the hypothalamic pituitary adrenocortical system in epilepsy. Eur Arch Psychiatry Clin Neurosci 254:303-311. Zobel AW, Nickel T, Kunzel HE, Ackl N, Sonntag A, Ising M, Holsboer F (2000) Effects of the highaffinity corticotropin-releasing hormone receptor 1 antagonist R121919 in major depression: the first 20 patients treated. J Psychiatr Res 34:171-181. Zurbrugg RPÂ (1976) Hypothalamic-pituitary-adrenocortical regulation. A contribution to its assessment, development and disorders in infancy and childhood with special reference to plasma cortisol circadian rhythm. Monogr Paediatr 7:1-83.

239

240

Samenvatting

SAMENVATTING Epilepsie is een van de meest voorkomende chronische neurologische aandoeningen bij kinderen. De onvoorspelbare epileptische aanvallen en de behandeling ervan kunnen een grote invloed hebben op hun cognitie, gedrag en kwaliteit van leven. Kinderen met epilepsie en hun ouders vertellen vaak dat epileptische aanvallen uitgelokt kunnen worden door stress en vragen hun artsen hoe dit kan en hoe zij hiermee om moeten gaan (hoofdstuk 1). Voorafgaand aan dit onderzoek was echter niet goed bekend hoe het kan dat stress de kans op epileptische aanvallen beïnvloedt. De beschikbare kennis was vooral gebaseerd op proefdieronderzoek en niet op onderzoek bij mensen met epilepsie (hoofdstuk 2). Het onderzoek dat in dit proefschrift staat beschreven, heeft tot doel de kennis over de effecten van stress op epilepsie op de kinderleeftijd en de mechanismen die hieraan ten grondslag liggen te vergroten. Om dit doel te bereiken, zijn de effecten van stress op epilepsie onderzocht vanuit een verscheidenheid aan invalshoeken en gebruikmakend van een breed scala aan onderzoeksmethodieken, variërend van literatuurstudies tot experimentele onderzoeken bij patiënten met epilepsie en dierexperimentele studies. Ten eerste is stressgevoeligheid van aanvallen bij kinderen met epilepsie in kaart gebracht en gerelateerd aan demografische en ziektegerelateerde factoren (deel I). Daarna zijn de hormonale mechanismen bestudeerd die ten grondslag zouden kunnen liggen aan stressgevoeligheid van aanvallen (deel II). Tot slot is in een diermodel gekeken naar de effecten van stress en stresshormonen op het ontstaan van epilepsie, de zogenaamde epileptogenese (deel III).

Deel I. Stressgevoeligheid van aanvallen bij kinderen met epilepsie Om te beginnen bestudeerden we stressgevoeligheid van aanvallen bij kinderen met epilepsie in relatie tot patiënt- en ziektegebonden kenmerken. Vragenlijstonderzoek liet zien dat stressgevoeligheid van aanvallen wordt gerapporteerd bij de helft van de kinderen met epilepsie. Dit omvat zowel aanvallen uitgelokt door acute stress, als een toename in de aanvalsfrequentie gedurende stressvolle perioden. Kinderen en hun ouders rapporteerden aanvallen uitgelokt door stress gerelateerd aan negatieve gebeurtenissen, zoals toetsen op school, ruzie thuis en gepest worden, maar ook aan positieve gebeurtenissen, zoals verjaardagsfeestjes en Sinterklaas. Om een eerste inzicht te krijgen in de mechanismen die leiden tot stressgevoeligheid van aanvallen, onderzochten we de relatie tussen stressgevoeligheid van aanvallen en verschillende patiënt- en ziektekenmerken. We vonden geen verschillen in demografische of epilepsiekenmerken tussen kinderen met en kinderen zonder stressgevoelige epilepsie. Wel kwam stressgevoelige epilepsie vaker voor bij kinderen die meer negatieve levensgebeurtenissen hebben meegemaakt. Het is bekend dat deze, wanneer zij vroeg in het leven optreden, langdurige effecten kunnen hebben op de werking van het hormonale stresssysteem. De resultaten suggereren daarom dat deze stresshormonen een belangrijke rol spelen in stressgevoeligheid van aanvallen (hoofdstuk 3).

241

ADDENDUM

Een ‘big-data’ onderzoek leverde aanvullend bewijs voor de zelf gerapporteerde relatie tussen stress en epilepsie op de kinderleeftijd. In dit onderzoek bestudeerden we de relatie tussen zoekopdrachten op Google naar informatie over epilepsie (een indirecte maat voor het voorkomen van epileptische aanvallen) en de stressvolle sinterklaasperiode over een periode van tien jaar met behulp van Google Trends. In Nederland wordt tijdens de sinterklaasperiode meer gezocht naar informatie over epilepsie op Google dan in de rest van het jaar, terwijl dit niet het geval is in landen waar geen Sinterklaas wordt gevierd. Deze resultaten leveren aanvullend bewijs voor de relatie tussen stress en het optreden van epileptische aanvallen (hoofdstuk 4). Stressoren gaan vaak gepaard met blootstelling aan veel verschillende zintuiglijke prikkels, bijvoorbeeld drukke beelden of harde geluiden. Om een indruk te krijgen van de zintuiglijke prikkelverwerking bij kinderen met epilepsie en de eventuele rol hiervan in stressgevoeligheid van epileptische aanvallen, brachten we de zintuiglijke prikkelverwerking in kaart bij kinderen met epilepsie. Problemen in de zintuiglijke prikkelverwerking werden gerapporteerd bij de helft van de kinderen met epilepsie. Vooral kinderen met aanvallen uitgelokt door acute stress gaven aan dat zij zintuiglijke prikkels sneller opmerken, wat kan leiden tot een teveel aan zintuiglijke informatie (‘overload’). Dit zou kunnen bijdragen aan de toegenomen kans op epileptische aanvallen na stress (hoofdstuk 5).

Deel II. Hormonale basis van stressgevoeligheid van epilepsie Het tweede deel van dit proefschrift richt zich op de hormonale basis van stressgevoeligheid van epilepsie. Om de relatie tussen stressgevoelige aanvallen en de biologische reactie op stress te onderzoeken, hebben we kinderen met epilepsie en gezonde controles blootgesteld aan een acute psychosociale stressor, de ‘Trier Social Stress Test’. Hieruit bleek dat de hormonale stressrespons bij kinderen met stressgevoelige aanvallen anders was afgesteld dan bij kinderen zonder stressgevoelige aanvallen en bij gezonde controles. Na blootstelling aan dezelfde soort stress, werd bij kinderen met stressgevoelige aanvallen minder cortisol (stresshormoon) afgegeven uit de bijnier. Dit gold zowel wanneer het vaststellen van stressgevoeligheid was gebaseerd op zelf ingevulde vragenlijsten, als wanneer dit was gebaseerd op een bijgehouden dagboek (hoofdstuk 6). Deze resultaten bevestigen dat de stresshormonen een belangrijke rol kunnen spelen bij stressgevoeligheid van epileptische aanvallen. Stresshormonen worden niet alleen geproduceerd en afgegeven in reactie op stress. Ook onder niet stressvolle omstandigheden varieert de hoeveelheid stresshormonen in het lichaam gedurende de dag. We vroegen ons af of deze variatie gedurende de dag ook invloed heeft op de epileptische activiteit en op de kans op epileptische aanvallen. Uit een systematisch literatuuronderzoek bleek dat het voorkomen van aanvallen over de dag overeenkomsten vertoonde met het 24-uurs ritme van het cortisolniveau. Beide kennen een snelle toename in de vroege ochtend en een geleidelijke afname gedurende de dag (hoofdstuk 7).

242

Samenvatting

Hierna hebben we onderzocht of er een relatie was tussen de schommelingen in cortisol en afwijkingen passend bij epilepsie op het electroencephalogram (EEG). Bij volwassenen met epilepsie vonden we dat het optreden van deze afwijkingen op het EEG (in de afwezigheid van aanvallen) inderdaad gerelateerd was aan het cortisolniveau, hoewel er grote verschillen bestonden tussen patiënten. De relatie tussen cortisol en EEG afwijkingen was groter bij patiënten die zelf aangaven stressgevoelige aanvallen te hebben. Deze resultaten ondersteunen het idee dat stresshormonen invloed hebben op epileptische activiteit, ook in niet-stressvolle omstandigheden (hoofdstuk 8).

Deel III. Stresshormonen en epileptogenese Tot slot onderzochten we de effecten van stress en stresshormonen op het ontstaan van epilepsie (epileptogenese) in een diermodel. In jonge muizen werden koortsstuipen opgewekt door middel van opwarming (hyperthermie), om zo de epileptogenese op gang te brengen. In de weken hierna werden de muizen herhaaldelijk geïnjecteerd met het stresshormoon corticosteron of met de dragerstof hiervan (milde stress), of werden zij met rust gelaten. Hierna bestudeerden we het uiterlijk en de functie van de gyrus dentatus, een onderdeel van de hippocampus (een specifiek deel van de hersenen, waar in dit model de epileptogenese plaatsvindt). Er veranderde weinig in het uiterlijk van de gyrus dentatus in muizen die koortsstuipen hadden, maar daarna geen injecties kregen, ten opzichte van dieren die geen koortsstuipen hadden meegemaakt. In muizen die waren geïnjecteerd met stresshormonen of waren blootgesteld aan milde stress, ging de epileptogenese echter gepaard met minder jonge zenuwcellen, meer celdelingen en een ander uiterlijk van de zenuwcellen, namelijk een grotere lengte van de dendrieten en meer dendritische spines, zonder dat dit invloed had op functionele eigenschappen (in het bijzonder glutamaterge neurotransmissie). Dit toont aan dat corticosteroïden en milde stress tijdens de epileptogenese de uiterlijke veranderingen in de gyrus dentatus doen toenemen, zonder dat er functionele veranderingen zijn, wat de hypothese deels ondersteunt dat stress in het vroege leven de epileptogenese stimuleert (hoofdstuk 9).

Conclusie De studies die staan beschreven in dit proefschrift geven nieuwe inzichten in de effecten van stress op epilepsie. Zij laten onder andere zien dat stressgevoeligheid van aanvallen veel voorkomt en gerelateerd is aan het aantal eerder doorgemaakte levensgebeurtenissen en aan gevoeligheid voor sensorische prikkels. Ook tonen ze aan dat de hormonale stressrespons bij kinderen met stressgevoelige epilepsie afwijkend is, en dat stresshormonen op jonge leeftijd het proces van ontstaan van epilepsie (epileptogenese) in een diermodel bevorderen. Deze resultaten kunnen in de toekomst consequenties hebben voor de voorlichting aan patiënten met epilepsie en aanknopingspunten bieden voor de ontwikkeling van nieuwe behandelstrategieën voor epilepsie (hoofdstuk 10). 243

244

Dankwoord

DANKWOORD Het is af! Naast de resultaten die in dit boekje beschreven zijn, hebben de afgelopen jaren van promotieonderzoek mij erg veel leuke en leerzame (en af en toe ook stressvolle) ervaringen gebracht. Alle collega’s, vrienden, familie en kennissen die geholpen hebben bij de totstandkoming van dit proefschrift of juist voor de nodige afleiding hebben gezorgd, bedankt! Een aantal mensen wil ik hier in het bijzonder bedanken voor hun belangrijke directe of indirecte bijdrage: Studiedeelnemers Ten eerste, alle patiënten en controles die hebben deelgenomen aan dit onderzoek en hun ouders of verzorgers. De meeste resultaten van dit proefschrift zijn gebaseerd op de stapels met vragenlijsten en dagboekjes die jullie hebben ingevuld en de testdagen waar jullie aan hebben deelgenomen. Hartelijk bedankt voor jullie inzet! Promotoren en copromotoren Kees Braun, het idee van jou en Marian om onderzoek te doen naar de relatie tussen stress en epilepsie bij kinderen stond aan de basis van dit proefschrift. Toen ik aan het einde van mijn studie bij je kwam met het plan een onderzoeksbeurs aan te vragen, was je direct enthousiast. Binnen korte tijd heb je een team samengesteld van begeleiders met verschillende expertises en hebben we het plan geschreven dat de blauwdruk werd voor dit promotietraject. Bedankt voor je bevlogenheid, je enorme schat aan kennis op het gebied van epilepsie, je altijd nauwkeurige, kritische en snelle commentaar op manuscripten en presentaties en de grote vrijheid die je je mij hebt geboden in dit onderzoek. Marian Joëls, bedankt voor je enorm betrokken begeleiding. Als stressexpert was jouw kennis essentieel zowel bij het humane als het dierexperimentele werk. Ik ben onder de indruk van jouw efficiëntie en je kwaliteit in het stellen van prioriteiten en heb veel geleerd van jouw oog voor zowel de details als de grote lijnen. Gedurende het gehele traject was geen moeite je te veel, van het uren samen achter de computer neuronen beoordelen tot hoofdbrekens over de interpretatie van de resultaten en het meeschrijven aan manuscripten. Jouw enthousiasme voor onderzoek werkte zeer motiverend en inspirerend. Ik hoop dat we in de toekomst nog vaker samen mogen werken. Floor Jansen, als student ben ik begonnen met onderzoek onder jouw hoede. Je begeleiding, met veel vertrouwen en ruimte voor zelfstandigheid, gecombineerd met praktische adviezen in de veilige sfeer die jij weet te creëren, heb ik vanaf het begin als zeer prettig ervaren. Deze eerste stages bij jou hebben mij verder geënthousiasmeerd voor de wetenschap en gemotiveerd voor een promotietraject in jullie onderzoeksgroep. Bedankt voor je laagdrempelige begeleiding en de altijd aanwezige bereidheid om mee te denken en oplossingen te zoeken. Pierre de Graan, bedankt voor je begeleiding bij de uitvoer van de dierexperimenten. Toen we besloten om ‘jouw’ muizenmodel voor koortsstuipen te gebruiken voor het induceren

245

ADDENDUM

van epileptogenese was het duidelijk dat jij een grote rol zou spelen bij het dierexperimentele deel van dit proefschrift. Met name jouw kritische afweging van de tijdsinvestering van alle betrokkenen ten opzichte van de meerwaarde van resultaten, zou ik graag met me meenemen. Leescommissie Geachte professor De Vries, professor Nieuwenhuis, professor Brouwer, professor Roelofs en professor Burbach, bedankt dat u zitting hebt willen nemen in mijn leescommissie. Collega’s (kinder)neurologie Lieve collega-onderzoekers van het WKZ en UMCU, bedankt voor alle gezelligheid, inspiratie, steun en afleiding. Eric, bedankt voor je altijd (gevraagd of ongevraagd) eerlijke en kritische commentaar, voor je enthousiaste bereidheid mee te denken en ideeën te spuien en voor alle rondleidingen in steden over de hele wereld. Bart, bedankt voor de gezelligheid op onze kamer en het initiatief tot leuke kamer-uitjes. Hanna, leuk dat jij de onderzoeksgroep in het WKZ kwam versterken, bedankt voor je gezelligheid en de thee-breaks. Maryse en Judith, bedankt voor jullie creativiteit, out-of-the-box denken en de gezellige knutsel- en breiprojecten. Wim, bedankt voor alle statistische adviezen en de gezelligheid in Edinburgh. Maeike, bedankt voor je stroom aan ideeën en je zeer aanstekelijke enthousiasme, ik vond het erg leuk om tegen het einde van mijn project nog met jou te mogen samenwerken en hoop dat dit nog een vervolg krijgt! Suzanne, bedankt voor de brede introductie in wetenschappelijk onderzoek tijdens mijn stage bij jou, inclusief internationaal congres en de road trip in Texas. Kim, bedankt dat jij me op sleeptouw nam toen ik als verlegen nieuwe collega bij jullie begon. Frans, Geertjan, Nicole, Herm, Willemiek, Jonas en alle andere epilepsie-onderzoekers, bedankt voor de kritische vragen tijdens research meetings en journal clubs en de interesse in de wandelgangen. Degenen die nog druk bezig zijn met onderzoeken en schrijven: heel veel succes! Professor Onno van Nieuwenhuizen, bedankt voor mijn start als student-onderzoeker bij de kinderneurologie en voor uw rol in de samenwerking met Stichting Bio Kinderrevalidatie. Irene Jonkers, bedankt voor je hulp bij ingewikkelde METC zaken. Hélène, bedankt voor je ondersteuning met betrekking tot afspraken, bestellingen en handtekeningen. Eltje, bedankt voor het bewaken van de SOPs, archivering en andere regeldingen. Johanna en Sladjana, bedankt voor alle hulp bij problemen met printen, enveloppen en stickers. Arts-assistenten neurologie, bedankt voor de gezelligheid tijdens assistentenweekenden, lunches, borrels en andere activiteiten. Nico en alle andere KNF’ers, bedankt voor de interesse en gezelligheid op de KNF gang. Collega’s translationele neurowetenschappen All stress group colleagues: Henk Karst, bedankt voor je electrofysiologische kennis, de begeleiding van masterstudenten en je eeuwige geduld bij het uitleggen van ingewikkelde theorie. Paul en Gideon, bedankt voor jullie expertise in de uitvoer van neurogenese

246

Dankwoord

kleuringen en analyses, jullie hulp heeft deze experimenten gered. Angela, bedankt voor het meedenken met experimenten, de corticosteron analyses en je electrofysiologische expertise. Anup, thanks for your patch clamp expertise. Christiaan Vinkers en Lotte Houtepen, bedankt dat ik mocht meekijken met jullie stresstesten, het uitlenen van jullie protocollen bij volwassenen en de bereidheid om mee te denken bij de opzet en de interpretatie van de resultaten. Sandra Cornelisse, bedankt voor het meedenken met de opzet van de stresstesten. Lotte Kok en Remmelt, bedankt voor de gezelligheid tijdens stress group meetings en cursusbezoek. Manila, thanks for your sociability. All other stress group colleagues from Utrecht and Amsterdam, thanks for sharing your ideas and expertise! All PhD students and postdocs of the preclinical epilepsy group: Ellen, bedankt dat je me wilde inwerken in alle muizenzaken en je nauwkeurige administratie. Bart, bedankt voor je zeer laagdrempelige hulp bij het aanleren van tal van technieken en het oplossen van problemen. Eduardo, great to have a colleague from Brazil also working in the stress ánd epilepsy field, thanks for the vivid discussions on experimental planning and interpretation of the results. Good luck with finishing your thesis! Sada, thanks for your help with the Prox1 stainings and analyses. Marina, bedankt voor je hulp op het lab. Vamshi, thanks for sharing your ideas during lab meetings. Alle collega’s op het lab, bedankt voor de gezelligheid en voor jullie vanzelfsprekende bereidheid dingen uit te leggen en mee te kijken. Voor mij als dokter op het lab waren jullie adviezen en tips broodnodig bij het aanleren van basisvaardigheden (pipet? Milli-Q?) en bij de cultuurswitch (gewoon proberen!). Marjolein, bedankt dat je een oogje in het zeil hield op mijn proefdieren, als jij een dagje vrij was maakte ik me altijd extra zorgen over het stressniveau van ‘mijn’ muizen. Leo en Henk Spierenburg, bedankt voor de praktische tips en goede zorgen. Sandra en Joke, bedankt voor alle hulp bij de bestellingen. Vicki en Ria, bedankt dat jullie altijd bereid waren om vragen te beantwoorden en mee te denken met het plannen van afspraken in onmogelijk veel drukbezette agenda’s. Studenten Ik vond het een genot om met studenten samen te werken en heb het geluk gehad er veel te hebben mogen begeleiden. Zonder jullie was dit boekje een heel stuk dunner geweest! Op chronologische volgorde: Kirsten and Giorgio, you were ‘my’ first students, and with your biomedical neuroscience background you were more familiar around the lab than I was and performed the complicated electrophysiology experiments very independently. By sharing your ideas and by helping me out with handling the mice in the early morning and weekends whenever I needed an extra pair of hands, you were not only of great practical support but also made my first months in the lab a lot more enjoyable! Sietske, bedankt voor de hulp bij het perfuseren en je niet aflatende enthousiasme. Milou Sep en Simone, bedankt voor jullie geduld en doorzettingsvermogen met de kleuringen. Milou, mooi dat je nu met je eigen PhD project begint, veel succes en plezier! Bas, Vivian en Joris, bedankt voor de vele uren die jullie hebben doorgebracht in de donkere

247

ADDENDUM

microscopiekamer, jullie inzet en precisie. Mascha, bedankt voor je geduld bij de screening van patiëntenstatussen en je enthousiasme in het uitzoeken van ingewikkelde statistiek. Christel, bedankt voor je hulp met het opzetten van de geheugentesten en het samen oefenen van de stresstesten. Hanneke, bedankt voor je inzet en je roze presentatie, misschien komen we elkaar in Nijmegen nog eens tegen. Monisa, bedankt voor je hulp met de administratie. Nienke, bedankt voor je hulp bij het opzetten en uitvoeren van de prikkelgevoeligheidsstudie en de avonden vragenlijsten invoeren (met pizza). Floris, als bachelor honoursstudent was je de jongste student die ik heb begeleid en ik vond het mooi om te zien hoe open, onbevangen, nieuwsgierig en vooral leergierig je je in het onderzoek stortte. Milou Pet en Tessa, bedankt voor jullie toewijding, multi-inzetbaarheid en niet te vergeten de gezelligheid. Fijn om de grootste bulk van de testmiddagen samen met jullie te mogen doen! Lorraine, de drijvende kracht achter de SPIKE studie, de ochtenden waarop jij voor dag en dauw opstond om (na een trein- en busreis) in Heemstede te zijn nog voordat de patiënten wakker werden, hebben geresulteerd in unieke data. Dat er nog vele publicaties mogen volgen! Overige collega’s Hilgo Bruining en Manon Hillegers, bedankt voor jullie kinderpsychiatrische expertise, waardoor het mogelijk was studies te doen op het grensvlak tussen de kinderneurologie en de kinderpsychiatrie. Collega’s van de neuropsychologie: Monique, Olga en Lennie, bedankt voor het meedenken bij het opzetten en uitwerken van de geheugentesten. Het laboratorium endocrinologie, Inge Maitimu en Eef Lentjes, bedankt voor de cortisol en alpha-amylase analyses. De afdeling research genetica, Bobby Koeleman en Ruben van ‘t Slot bedankt voor het meedenken over de genetische analyses. Alexandre Suerman programma, professor Paul Coffer, Lucette Teurlings en de selectiecommissie, bedankt voor het genereuze stipendium. Alle Alexandre Suerman collega's, bedankt voor het delen van ervaringen en de leerzame en gezellige masterclasses. Stichting Bio Kinderrevalidatie, bedankt voor de financiële ondersteuning en de prettige samenwerking, waardoor het mogelijk was ons onderzoek uit te voeren op jullie prachtige en stressarme locatie. Ad Groen, bedankt voor het vertrouwen. Liesbeth, bedankt voor al je hulp bij praktische zaken. Bertus, bedankt voor je enthousiasme als ons vaste jurylid bij de stresstesten. De heer Rogier van Nieuwenhuizen, bedankt dat we uw grote kring van vrienden en kennissen mochten benaderen voor deelname in de jury en voor uw eigen optreden. De overige juryleden: de heer Van Diehl, mevrouw Agricola, de heer en mevrouw Van der Schans, de heer en mevrouw Provó-Kluit, de heer en mevrouw Van de Broek, de heer en mevrouw Van Dis, de heer en mevrouw Hoefakker, de heer en mevrouw Van Iersel, de heer en mevrouw Baart de la Faille, mevrouw Ullman-ter Haar Romeny, de heer Gevers Deynoot, mevrouw Onderwater, mevrouw Van de Loo, mevrouw Hertzberger, mevrouw Laman Trip, de heer Nettl en de heer Van Blokland, allen hartelijk bedankt voor uw hulp.

248

Dankwoord

Alle co-auteurs die ik nog niet bij naam genoemd heb, hartelijk bedankt voor uw bijdrage aan de manuscripten. Vrienden en familie Mijn paranimfen Anne en Charine, bedankt dat jullie op deze dag naast mij willen staan. Als studiegenootjes, alle drie inmiddels werkzaam in de kliniek en het onderzoek, hebben we de afgelopen jaren lief en leed gedeeld. Anne, bedankt voor alle attenties, je motiverende woorden en jouw zelfverkozen en met optimisme gedragen driedubbele werklast die mij altijd doen inzien dat het nog veel erger kan ;-). Charine, als vriendinnen vanaf het eerste mentorgroepje hebben we de afgelopen tien jaar veel leuke ervaringen, maar ook gedeelde twijfels en onzekerheden besproken: over studie, co-schappen en daarna promotieonderzoek, persoonlijke ontwikkelingen en dé perfecte keuzes voor de toekomst. Ik hoop dat er nog veel high tea’s en high beers, spelletjesavonden en creatieve workshops met zijn drieën en met de aanhang zullen volgen! Josien, wat een luxe om een vriendin te hebben die al zolang meegaat. Soms zien we elkaar wat meer, soms wat minder, maar als het nodig is, weten we elkaar altijd te vinden. Fijn dat we nu weer dichter bij elkaar in de buurt wonen en ook zonder lang van tevoren een afspraak te maken even kunnen bijkletsen. Elly, Peter, Esther en Peter, Frans en Anna, Anne en Niels, Jan-Peter, 'oma' Gesina en ook Bruno, Jelmer en Bas, bedankt voor de gezelligheid in jullie warme familie. Oma, het is altijd fijn om te merken hoe trots je bent op mij en je andere kleinkinderen. Bedankt voor je betrokkenheid en waardering. Lennart, bedankt dat je mij steeds weer laat zien dat het leven toch vooral mooi is als je ‘gewoon’ doet wat je leuk vindt en voor jouw vertrouwen in de expertise van je grote zus (al moeten we die Nobelprijs nog even afwachten :-p). Pap en mam, ik prijs me gelukkig met een kindertijd met zeer weinig early life stress en heel veel liefde en vertrouwen. Pap, bedankt dat je me van jongs af aan gestimuleerd hebt om dingen te willen begrijpen. Mam, jouw enthousiaste verhalen over werken met kinderen met psychische problemen hebben zeker mijn richting bepaald. Bedankt dat jullie er altijd voor me zijn. Koos, bedankt voor alles. Bedankt voor je liefde en steun, maar ook voor jouw aandacht voor het stellen van prioriteiten en het bijstellen van de werk-privé balans. Alles is leuker met jou. Ik hou van je!

249

250

List of publications

LIST OF PUBLICATIONS Koudijs SM, van Campen JS, Braams OB, Leemans A, van Nieuwenhuizen O, Jansen FE en Braun KPJ, Witte stof afwijkingen en intelligentie bij kinderen met tubereuze sclerose, Epilepsie 2010; 3: 19-21. van Campen JS, Jansen FE, Steinbusch LC, Joëls M, Braun KPJ, Stress sensitivity of childhood epilepsy is related to experienced negative life events, Epilepsia 2012; 53: 1554-1562. van Campen JS, Jansen FE, Brouwer OF, Nicolai J, Braun KPJ, Interobserver agreement of the old and the newly proposed ILAE epilepsy classification in children, Epilepsia 2013, 54: 726-32. van Campen JS, van Diessen E, Otte WM, Joels M, Jansen FE, Braun KPJ, Does Saint Nicholas provoke seizures? Hints from Google Trends, Epilepsy&Behavior, 2014, 32: 132-4. van Campen JS, Jansen FE, de Graan PNE, Braun KPJ, Joëls M, Early life stress in epilepsy: a seizure-precipitant and risk factor for epileptogenesis, Epilepsy&Behavior, 2014, 38: 160-71. van Campen JS, Jansen FE, Pet MA, Otte WM, Hillegers MHJ, Joëls M*, Braun KPJ*, Relation between stress-precipitated seizures and the stress response in childhood epilepsy, Brain, in press van Campen JS, Valentijn FA, Jansen FE, Joels M, Braun KPJ, Seizure occurrence and the circadian rhythm of cortisol, a systematic review, Epilepsy&Behavior, in press Verbeek NE, Wassenaar M*, van Campen JS*, Sonsma A, Gunning B, Knoers N, Lindhout D, Jansen FE, Leijten F, Brilstra EH, Kasteleijn- Nolst Trenité D, Seizure precipitants in Dravet syndrome: What events and activities are specifically provocative, compared to other epilepsies?, Epilepsy&Behavior, in press Koudijs SM*, van Campen JS*, Braams OB, Leemans A, van Nieuwenhuizen O, Jansen FE en Braun KPJ, submitted van Campen JS, Kliest T, Jansen FE, van Schooneveld MMH, Braun KPJ and Joels M, Effects of stress on working memory in children with epilepsy and controls, submitted van Campen JS, Jansen FE, Kleinrensink NJ, Joels M, Braun KPJ, Bruining H, Sensory modulation disorders in childhood epilepsy, submitted

251

ADDENDUM

van Campen JS, Hessel EVS, Bohmbach K, Rizzi G, Lucassen PJ, Lakshmi Turimella S, Umeoka EHL, Meerhoff GF, Braun KPJ, de Graan PNE*, Joels M*, Effects of repetitive mild stress and corticosteroid exposure on epileptogenesis after early life experimental febrile seizures in mice, submitted Verbeek NE, Wassenaar M, van Campen JS, Sonsma A, Gunning B, de Weerd, AW, Knoers N, Lindhout D, Spetgens WPJ, Gutter T, Jansen FE, Leijten F, Brilstra EH, Kasteleijn- Nolst TrenitĂŠ D, Photosensitivity in children with SCN1A-related Dravet syndrome, submitted * These authors contributed equally to this work

252

Curriculum vitae

CURRICULUM VITAE Jolien Suzanne van Campen was born on the 16th of December 1986 in Nijmegen, the Netherlands. She graduated cum laude at the Stedelijk Gymnasium Nijmegen in 2005 and started medical school at the Utrecht University in the same year. Jolien has always been interested in brain functioning, especially during development. Therefore, after having been selected for the honours program of the Faculty of Medicine in 2008, she started a research project on childhood epilepsy at the department of Pediatric Neurology of the University Medical Center (UMC) Utrecht under supervision of Floor Jansen and Kees Braun. Jolien graduated from medical school cum laude in 2011 and was awarded the Alexandre Suerman stipendium: a personal grant to facilitate an MD/PhD project. This MD/PhD project on the effects of stress on childhood epilepsy was performed at the departments of Pediatric Neurology and Translational Neuroscience of the UMC Utrecht Brain Center Rudolf Magnus under supervision of Kees Braun, Marian Joëls, Floor Jansen and Pierre de Graan. The results of this project are presented in this thesis. In March 2015, Jolien started her training in psychiatry at the Radboud university medical center (Radboudumc) in Nijmegen to become a child and adolescent psychiatrist.

Jolien Suzanne van Campen werd geboren op 16 december 1986 in Nijmegen. In 2005 behaalde zij haar gymnasium diploma cum laude aan het Stedelijk Gymnasium Nijmegen. In datzelfde jaar begon ze met de studie Geneeskunde aan de Universiteit Utrecht. Jolien is altijd geïnteresseerd geweest in de werking en de ontwikkeling van de hersenen. In 2008 startte zij met onderzoek naar epilepsie op de kinderleeftijd bij de afdeling Kinderneurologie van het Universitair Medisch Centrum (UMC) Utrecht, onder begeleiding van dr. Floor Jansen en prof. dr. Kees Braun. Dit deed ze onder andere in het kader van het Excellent Tracé voor talentvolle geneeskundestudenten. In augustus 2011 behaalde Jolien haar artsexamen cum laude. Zij ontving een Alexandre Suerman stipendium, een persoonlijke beurs voor een MD/PhD project, om haar onderzoek voort te zetten. Dit promotietraject naar de effecten van stress op epilepsie voerde zij uit bij de afdelingen Kinderneurologie en Translationele Neurowetenschappen van het UMC Utrecht Hersencentrum Rudolf Magnus onder begeleiding van prof. dr. Kees Braun, prof. dr. Marian Joëls, dr. Floor Jansen en dr. Pierre de Graan. De resultaten van haar onderzoek staan beschreven in dit proefschrift. In maart 2015 is Jolien met veel plezier begonnen aan de opleiding tot psychiater in het Radboud universitair medisch centrum (Radboudumc) in Nijmegen, met het doel zich te specialiseren in de kinderen jeugdpsychiatrie.

253