Sleep macro‐architecture in sleep‐deprivation–sensitive and sleep‐deprivation–insensitive juvenile myoclonic epilepsy: A polysomnographic study

OBJECTIVE: Sleep deprivation (SD) triggers seizures in only a subset of patients with juvenile myoclonic epilepsy (JME). We compared sleep macro-architecture between SD-sensitive and SD-insensitive JME patients and healthy controls. METHODS: Overnight polysomnography was done in healthy volunteers (n = 30), JME patients without SD-triggered seizures (SD-insensitive, n = 31), and JME patients reporting SD-triggered seizures on a structured interview (SD-sensitive, n = 68). Eleven participants with clinically diagnosed narcolepsy were excluded. We analyzed sleep continuity, stage distribution, rapid eye movement (REM)/non-REM (NREM) stability, and cardiorespiratory indices. RESULTS: SD-insensitive patients had lower sleep efficiency than controls but preserved stage N3 and REM. SD-sensitive patients had significantly lower N3 than SD-insensitive patients (p = .002) and reduced REM vs controls (q = .005). SD-sensitive status independently predicted N3% after adjusting for sex and anti-seizure medication class (β = -8.28, p = .013), and remained significant after further adjustment for generalized tonic-clonic seizure (GTCS) frequency (β = -7.35, p = .035). Valproate use was also independently associated with N3 reduction (β = -8.37, p = .007). Respiratory and cardiac indices did not differ. SIGNIFICANCE: SD-sensitive JME shows a light NREM-dominant sleep profile with decreased N3 vs SD-insensitive patients along with decreased REM when compared to controls. These differences are associative (SD sensitivity was self-reported; no experimental challenge). No polysomnography parameter independently predicted SD sensitivity. Prospective studies with controlled SD challenge are required to determine if lighter sleep is a vulnerability factor or a parallel correlate of seizure risk.