30% of patients with epilepsy suffer from drug-resistant seizures and are at risk of epilepsy-related complications, from cognitive dysfunctions to premature mortality. Both seizures and their complications are modulated by patients' vigilance states, with a tight and bi-directional interplay between sleep and epilepsy. Several epilepsy complications are associated with sleep, such as sleep-disordered breathing or Sudden and Unexpected Death in Epilepsy (SUDEP). SUDEP is a non-traumatic death, unrelated to a documented status epilepticus, which accounts for up 50% of premature deaths in epilepsy, with a cumulative risk of ≈ 10% at 40 years in patients with childhood-onset epilepsy. SUDEP typically occurs during sleep, after a nocturnal seizure, and primarily results from a postictal central respiratory dysfunction in patients with generalized convulsive seizure (GCS), suggesting that interaction between respiratory dysfunction and sleep state may play a role in its pathophysiology. Most of patients with drug-resistant seizures demonstrate transient peri-ictal apnea and hypoxemia, especially in the aftermath of a GCS. Experimental and clinical data suggest that most SUDEP primarily result from a fatal seizure-related respiratory arrest. In patients whose SUDEP had occurred during long-term video-EEG monitoring, we observed fatal postictal central apnea after a nocturnal GCS in all SUDEP. Accordingly, it is currently hypothesized that in a subgroup of patients, repetition of seizures may contribute to chronic alteration of respiratory regulation which may increase the risk of fatal postictal central respiratory arrest. Finally, post-mortem data in SUDEP patients showed alteration of neuronal populations involved in respiratory control in the medulla. The complex network that regulates arousal and sleep and the respiratory network are strongly interconnected. Impairment of the interaction between central respiratory control and arousal systems has been reported in several clinical situations, including sleep apnea syndrome, sudden infant death syndrome or Prader-Willi Syndrome. In epilepsy, preclinical data in rodents indirectly support a role for 5HT in the impairment of the interactions between the arousal and respiratory systems in the cascade of events leading to SUDEP. However, no direct evidence is available, and the link between alterations of the brainstem networks involved in arousal regulation and respiratory dysfunction has not been characterized in patients with epilepsy yet.
Study Type
INTERVENTIONAL
Allocation
NON_RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
NONE
Enrollment
60
Video-EEG monitoring
heart rate, pulse oximetry (oxygen levels in the blood), nasal airflow, respiratory effort (thoracic and abdominal) and capnography (carbon dioxide (CO2) levels in exhaled air)
The healthy patient/subject breathes through the mouth, using a mouthpiece and a nose clip, through a device fitted with a hermetically sealed bag that measures the various parameters of his/her breathing. At the start of the test, the healthy patient/subject breathes ambient air and his or her breathing is measured. Then, after a few minutes, the healthy patient/subject is connected to the bag, breathing in a closed circuit. This causes a gradual increase in carbon dioxide (CO2) in the inspired air. During this time, breathing parameters will be measured and gas exchanges studied with each breath. The test is stopped when the end-tidal carbon dioxide pressure (PetCO2) reaches 60 mm Hg, or in the event of intolerance.
In the evening, as soon as the doctor detects on the EEG that the patient/subject is in deep sleep, various tests will be carried out to assess reactivity to wakefulness. Two hypercapnic challenges will be carried out during sleep, using the same procedure as for wakefulness. The test will be stopped when the healthy patient/subject wakes up, or when the end-tidal carbon dioxide pressure (PetCO2) reaches 60 mm Hg, or in case of intolerance.
Two auditory stimulus tests will be carried out during the patient's sleep. Using headphones, the investigator will administer auditory stimuli at regular, progressively louder intervals to determine the ability of the patient/ healthy subject to awaken to an auditory stimulus.
Questionnaire to assess caffeine consumption habits Quality of life questionnaires QOLIE-31 Anxiety and depression questionnaires HADS
Hôpital Pierre Wertheimer
Bron, Rhone, France
RECRUITINGEnd tidal CO2 (PETCO2) value at arousal in patients with epilepsy and in healthly subjects
PETCO2 corresponds to the value taken at the end of a breath of the partial pressure of CO2. Two hypercapnic challenges will be performed the same night. The first will start once unequivocal N3 stage of NREM will be observed, as determined online by polysomnography data. After the termination of the first test, the patient/subject will go back to sleep and a second procedure will be performed once she/he will reach N3 stage of NREM again. Data from the two procedures will be averaged for each subject.
Time frame: the night of hospitalization
Value of the HCVR slope during sleep in patients with epilepsy and in healthy subjects
HCVR measures the increase in minute ventilation (VE) induced by an increase of PETCO2. The HCVR slope, expressed in L/min/mmHg, will be calculated from the linear regression of VE and PETCO2. Data from the two procedures per night described aboves will be averaged for each subject. will be measured during hypercapnic sleep challenges during the night of hospitalization.
Time frame: one night of hospitalization
Value of the awake HCVR slope in patients with epilepsy and in healthly subjects
As detailed in section 6.1.3, a hypercapnic challenge will be performed awake after arrival of the patient/subject in the epilepsy monitoring unit. Calculation of the awake HCVR slope wil be performed as described for HCVR slope during sleep
Time frame: will be measured during hypercapnic challenges on awakening on the first day of hospitalization.
Value of the ventilatory recruitment threshold (VRT) in patients with epilepsy and in healthly subjects
VRT reflects the PCO2 value at which the central chemoreflex is initiated and corresponds to the onset of compensatory and progressive rise in VE during hypercapnic challenge. The determination of VRT requires a hyperventilation baseline period during which the measured VE is the one required to maintain hypopnea under resting metabolic conditions; termed basal ventilation or the "wakefulness drive" to drive. To determine VRT, it is therefore required that the subject breathes deeply for few minutes, raising her/his tidal volume such that she/he rapidly achieves and maintaines a PETCO2 of 20-25 mmHg. Then the hypercapnic challenge can be performed. In this context, VRT can not be determined during sleep.
Time frame: will be measured during hypercapnic challenges on awakening on the first day of hospitalization.
Number of focal to bilateral tonic-clonic seizures during the 3 months preceding the polysomnography in patients with epilepsy
Time frame: Will be collected during the first day of hospitalization
Central apnea index and Obstructive Apnea Hypopnea Index
Sleep apneas can classified as obstructive, central or both central sleep apnea, Apnea will be defined as a decrease in peak nasal pressure of \>90% of baseline, lasting at least 10 s. Hypopnea was defined as a decrease of \>30% of the baseline nasal pressure, lasting at least 10 s and associated with ≥4% drop in SpO2. Central apnea will be defined by cessation \> 10 seconds of airflow with simultaneous cessation of respiratory effort. The respiratory events will be scored according to the 2007 American Academy of Sleep Medicine guidelines. Apnea index will correspond to total number of each apnea type divided by total sleep time over a 24-hour period.
Time frame: one night of hospitalization
Sound intensity (dB) required to induced awakening
The paradigm will be the one proposed by Martin et al. 1996 ; Philip et al. 1994 . The stimulation will be manually delivered using earphone inserts and will consist in pure sounds of 1000Hz, from 40dB to 110dB, duration 5s. Stimulation starts after 5 min of unequivocal NREM, as assessed with continuous EEG montoring. The stimulation will start with intensity I1=50dB. If complete behavioral arousal (=awakening): stop stimulation and wait until 2 min of stable sleep (defined by re-apparition of a spindle, a K complex or REM) and next trial with I1-10dB. If arousal but no awakening: wait until 2 min of stable sleep and next trial with sound +10dB. If no Arousal : wait 10s and deliver sound +10dB or longer sound (10s)
Time frame: one night of hospitalization
Current and past mean daily intake of caffeine (mg/day)
Current usual caffeine intake and the one over the last ten years will be assessed by a validated self-survey (Simonin et al., Neuroobiology of Disease 2013) filled-in at home by the patient and, if needed, the help of the caregiver filled between the inclusion visit and the polysomnography. Its reliability, assessed by retest (mean interval: 37.5 ± 5.5 days) in 31 patients, was excellent (intraclass correlation coefficient by Fleiss method: 0.97 \[95% CI: 0.94-0.98\]). The mean daily intake of caffeine containing items (coffee, tea, chocolate, sodas) will be reported, along with any changes of consumption over the 10 year period. All data will be then confirmed by interview perfomed during the stay in epilepsy monitoring unit.
Time frame: Will be collected between inclusion and the first day of hospitalization
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.