< Back to previous page

Project

Multimodal monitoring of electrophysiological signals in childhood epilepsy and neonatal encephalopathy

Multimodal monitoring of electrophysiological signals in childhood epilepsy and neonatal encephalopathy.</>
</>Summary 
EEG records electrical activity in the brain generated by pyramidal cells forming large neuronal networks.  EEG monitoring provides a constant assessment of brain function and has 3 major applications in neurological practice.  It can be used to appreciate epileptic activity in patients with epilepsy.  It can help in monitoring cerebral recovery after brain insults.  In preterm and term newborns, it can provide important information on brain maturation.  EEG interpretation is dependent on retrospective visual inspection of theEEG signals by a neurophysiologist.  
By measuring other physiological signals as heart rate and respiration together with the EEG, we can learn more about the functionality of the central autonomic nervous system in patients with epilepsy.  Developing advanced algorithms for EEG analysis can help in automatisation and quantification of EEG analysis.  Using these techniques will be helpful in standardizing the technique and making assessment of long term EEG monitoring easier.
The first aim of this project is to study the central autonomic nervous system in children with epilepsy.
Epilepsy is themost frequent neurological disorder characterized by recurrent and unprovoked seizures.  Seizure incidence is the highest in infancy and childhood.  Patients with epilepsy are prone to dysfunction of the autonomic nervous system.  The acute disturbances arelinked with seizure activity on the EEG and can be measured using heartrate or respiratory frequency.  The chronic autonomic dysfunction appears after time and can be appreciated using heart rate variability (HRV).  The heart is innervated by sympathetic and vagal branches of the autonomic nervous system. The balance of these neural influences on the cardiac pacemaker determines heart rate variability.  
To study the autonomic nervous system in children with epilepsy, we used multimodal long-term measurements and included ECG and respiration in addition to EEG.
Acute changes in heart rate or respiration are linked to seizures.  In our study on peri-ictal heart rate changes in childhood epilepsy, we were able to demonstrate that heart rate changes in focal seizures originating from the frontal or temporal lobe.  Seizure detection based on pre-ictal heart rate changes is possible in 70% of the focal seizures, but we have to take intoaccount a very low sensitivity.  The time lag between the heart rate change and onset of ictal activity on the EEG was much shorter compared to studies conducted in adults.    These findings have implications for future seizure detection systems and the development of closed loop systems.  In childhood epilepsy, identification of these early autonomic manifestations in seizures is possible at least in focal seizures, but the time lag was only 3,5 secondsmaking the time to initiate therapeutic measures very short. 
Patients with long-lasting and refractory epilepsy are prone to chronic dysfunction of the autonomic nervous system.  From our studies,we were able to show that cardiac as well as respiratory autonomic changes are present in childhood epilepsy.  We used heart rate variability to detect chronic autonomic dysfunction in children with epilepsy.   In contrast to previous conducted studies, we took into account normal circadian variations of HRV.  In children with refractory epilepsy, heart rate variability was found to be depressed with a prominent reduction in vagal tone during slow wave sleep.  Normal sleep modulation of autonomic function is lost in this patient population but is again improved after treatment with a vagus nervestimulator.   The same reduction in vagal tone was foundin children with West syndrome after a much shorter disease course compared to the adult population.  Reduction of vagal activity during slow wave sleep was found already after 3 years.  There isgrowing evidence for the association of heart rate variability with morbidity and mortality in patients with epilepsy.  Recognition of chronic changes is important because these abnormalities are thought to play an important role in sudden unexpected death in epilepsy patients(SUDEP).
Respiration is altered already at onset of the epileptic syndrome in patients with West syndrome.   Respiration is slower and contains more irregularities, suggesting a higher risk of apneas.  The abnormalities are present in between the typical seizures, but in the presence of continuous spikes.  If these findings are a risk factor for SUDEP in this population remains uncertain.   
Ictal bradypnea is present in childhood temporal lobe seizures.  The same phenomenon has been observed in adults with temporal lobe epilepsy. Absence seizures do not show a uniform ictal respiration pattern, probably due to the short duration of the seizures.  Interictally respiration is disturbed in absence epilepsy but not in temporal lobe epilepsy.  These findings could not be explained by interictal electrical discharges since both cohorts only had few interictal spikes.  The findings are due to involvement of the thalamocortical network in the epileptogenetic process.  
 
The second aim of this project is to take a first step in automatisation and quantification of EEG in critically ill newborns.
In preterm and term newborns, EEG can be used to assess brain maturation, detect seizure activity, monitor treatment and give information on prognosis of the newborn baby. 
In the first part of this study, we used a model of burst-suppression in newborns with hypoxic-ischemic encephalopathy to take a first step towards automated analysis of EEG.  We show that interburst-interval can be detected automatically with high specificity in newborns with hypoxic ischemic encephalopathy with different degree of discontinuity in their EEG background.  In the second part of this study, we demonstrate that quantitative analysis of bursts in discontinuous EEG patterns in newborns is possible and can discriminate between the pathological burst-suppressionpattern in early infantile epileptic encephalopathy and normal tracé alternant in healthy newborns.  Defining and quantifying these parameters will be very useful in the development of bed-side automated analysis of EEG in newborns in the future.
 
Date:1 Mar 2009 →  16 Oct 2013
Keywords:EEG, seizure detection system, autonomic system, epilepsy
Disciplines:Neurosciences, Biological and physiological psychology, Cognitive science and intelligent systems, Developmental psychology and ageing
Project type:PhD project