High-resolution sleep fragmentation assessment in narcolepsy type 1 and their non-narcoleptic siblings: a 5-s mini-epoch study

STUDY OBJECTIVES: Narcolepsy type 1 (NT1) is characterized by fragmented sleep, yet brief transitions may be concealed in traditional 30-s epochs. We therefore aimed to assess sleep fragmentation in NT1 patients and their non-narcoleptic siblings using 5-s mini-epochs. METHODS: We analyzed sleep stages from human-scored 30-s epochs and automatically (U-Sleep) scored 5-s mini-epochs in polysomnographies from 125 NT1 patients and their 100 non-narcoleptic siblings. Sleep fragmentation was quantified using transition indices and probabilities in epochs and mini-epochs, furthermore, stratified by 1st and 2nd night-half. Predictors (H1N1-vaccination, CSF hypocretin-1 deficiency severity (low [40-150 pg/mL]; undetectable [<40 pg/mL]), HLA-DQB1*06:02-positivity, narcolepsy core symptom severity) for sleep transitions were explored. RESULTS: NT1 patients had significantly higher all-stages and sleep-wake transition indices compared to siblings in epochs and mini-epochs. Both epoch and mini-epoch transition indices varied significantly from 1st to 2nd night-half. Only in mini-epochs, sleep-wake fragmentation was (a) higher in patients with severe core symptoms, and (b) increased in the 2nd night-half in NT1 patients with severe hypocretin-1 deficiency (<40 pg/mL) and/or severe core symptoms. CONCLUSIONS: Our findings suggest that sleep fragmentation, a core feature of NT1, is associated with disease severity and hypocretin deficiency severity when high-resolution sleep stages are analyzed. Five-second sleep staging and split-night analyses enhance detection of sleep instability and reveal new specific patterns and clinical associations. Automated mini-epoch analyses may improve future NT1 phenotyping and possibly its borderland by supplementing with clinically relevant high-resolution features otherwise hidden in traditional full-night 30-s epoch analyses. Statement of Significance This study demonstrates that high-resolution 5-s sleep staging provides more detailed sleep characterization than epochs and additionally uncovers clinically relevant sleep fragmentation patterns in narcolepsy type 1 which remain hidden in traditional 30-s analyses. By combining 5-s mini-epoch scoring with split-night analyses, we identified specific associations between disease severity and increased sleep-wake fragmentation. In mini-epochs, but not in epochs, sleep-wake fragmentation was significantly higher in patients with all core narcolepsy symptoms and increased significantly from the 1st to the 2nd night-half in patients with undetectable hypocretin-1 levels (<40 pg/mL) and/or all core narcolepsy symptoms. These findings highlight the clinical value of high-resolution, temporally sensitive sleep stage analyses for detailed phenotyping and detection of future disease biomarkers.