Interictal sleep recordings during presurgical evaluation: Bidirectional perspectives on sleep related network...

Sleep recordings are an integral part of presurgical evaluation in drug-resistant focal epilepsy. Physiological network functioning is substantially different between wakefulness and sleep and thus may add further complexity to the task of determining the epileptogenic zone (EZ). A thorough understanding of changes in epileptic networks in relation to sleep is therefore essential in order to fully appreciate the added value of sleep recordings. Furthermore, shared expertise in epilepsy and sleep is beneficial for both domains, as intracerebral EEG during presurgical evaluation offers a unique window into physiological networks and their interaction during sleep. This review intends to delineate the way in which sleep modifies interictal epileptic discharges (IEDs), and to summarize which sleep state is the most appropriate for aiding in discerning the EZ. Two approaches will be reviewed. First, classical scalp electroencephalography (EEG) recordings help to localize the EZ, especially during rapid-eye-movement (REM) sleep. REM sleep tends to narrow the field size of IEDs, and thus helps to target the core of the EZ. Second, automated analysis of intracerebral recordings can make use of both IEDs and sleep-related oscillations in combination. Notably, high frequency oscillations and directed connectivity measures can be assessed in a single sleep cycle and are valuable tools to probe epileptogenicity. In this approach, which exploits increased network interactions during sleep, non-REM-sleep is the most suitable sleep stage to extract multiple features of local and distributed neuronal activity in order to predict the EZ. The added value of intracerebral recordings is perfectly bidirectional. From a sleep perspective, invasive EEG recordings are a unique opportunity to unravel local sleep-related network function of superficial and deeply situated brain structures. Intracerebral EEG has thus allowed the dissection of sleep features and oscillations and their anatomical sources. A multicenter effort led by the Montreal Neurological Institute resulted in a detailed open-access atlas on normative EEG activity during sleep (https://mni-open-ieegatlas.research.mcgill.ca/). It contributed to our understanding that the human brain does not sleep uniformly but that specifically deep structures have distinct signatures that are discernable from the rest of the brain. Also, this research direction allowed us to gain insights into our understanding of the important neurocognitive functions of sleep. Finally, this review provides a clinical outlook on the benefit of genuine sleep recordings, i.e. recordings with additional sleep sensors, concomitant to presurgical evaluation, in order to fully discern common sleep disorders as a frequent comorbidity of epilepsy. In conclusion, shared expertise in sleep and epilepsy is of mutual added value for improving the management of patients with epilepsy.