P 121 Sleep stage sequence of sleep onset REM periods predicts narcolepsy in patients with excessive daytime sleepiness

Introduction Sleep onset REM periods (SOREMP) and sleep latency on the Multiple Sleep Latency Test (MSLT) are crucial to distinguish narcolepsy from other conditions with excessive daytime sleepiness (EDS) but SOREMP may also be found in behaviorally induced inadequate sleep syndrome (BIIS), sleep deprivation, and circadian rhythm disorders. Sleep stage sequence analysis in MSLT recordings has been reported to be useful in identifying patients with narcolepsy. We aimed to confirm this hypothesis in a large cohort with special regard to HLA haplotype and CSF hypocretin (Hcrt) levels. Patients and methods 458 patients with EDS (and sleep-disordered breathing/SDB excluded) underwent nocturnal polysomnography (PSG) and MSLT. Hcrt was measured in 181 patients, and HLA haplotyping results were available in 254 individuals. A total of 1887 naps was recorded containing 272 SOREMP. Results Mean age was 36.2 (13.5) years, and 57.4% of patients were female. According to ICSD-3 criteria, 47 patients were diagnosed with type 1 narcolepsy (NL1), 30 individuals with type 2 narcolepsy (NL2), 78 patients with idiopathic hypersomnia, 9 with BIIS, 11 with circadian rhythm disorders, and 70 with other sleep disorders such as NREM parasomnia, insomnia, periodic limb movements, or unexpected SDB. Secondary CNS hypersomnia was diagnosed in 16 patients, 67 patients had hypersomnia associated with a psychiatric disorder, and in 102 participants, EDS was unexplained. SOREMP arising from sleep stage N1 (N1-SOREMP) was extremely rare in non-narcoleptic patients ( n  = 2), but was present in 45.5% of patients with narcolepsy. The number of N1-SOREMP was significantly higher in patients with NL1 than with NL2. 42.1% of SOREMP in the NL1 group arose from N1 compared to 15.5% in the NL2 group. In patients with narcolepsy, the number of N1-SOREMP showed negative correlation with Hcrt levels, mean sleep latency on MSLT, and REM latency on PSG. Furthermore, it was significantly associated with HLA DQB1*06:02. N1-SOREMP occurred in 69% patients with Hcrt 110 pg/ml. Conclusion Both occurance and frequency of SOREMP arising from sleep stage N1 have high sensitivity for the diagnosis of narcolepsy and type 1 narcolepsy, in particular. The number of N1-SOREMP appears to be a marker of disease severity or instability of REM sleep, respectively. Sleep stage sequence analysis of SOREMP periods may support phenotyping of patients with central disorders of hypersomnolence.