Objective/background: A nocturnal sleep onset REM period (defined as REM onset latency ≤ 15 min; SOREMP) occurs rarely and research has shown that the phenomenon is specific for type 1 and 2 narcolepsy. However, little is known about the meaningfulness of the phenotype in general sleep clinic patients because those that exhibit the phenomenon often present with few traditional narcolepsy symptoms. As such, this study aimed to (1) evaluate the rate of eventual MSLT testing for those with a SOREMP on routine PSG when the phenomenon occurred in the absence of potential explanatory factors and (2) quantify the stability of the SOREMP phenotype.
Patients/methods: This was a retrospective analysis of a large repository of de-identified PSG and MSLT test results from 2008 to 2015. Patient records were retrieved from a repository of studies completed at a variety of sleep laboratories across the USA. A total of 118,046 baseline polysomnograms were evaluated for a PSG SOREMP (occurred in 0.7% of the sample). Patients were excluded if they indicated working either shift or night work at the time of the SOREMP or if their self-reported habitual weekday time in bed was less than 7 h. A final sample of 391 cases with a SOREMP were sequestered and previous or consecutive studies were searched for each individual.
Results: The vast majority of patients (n = 347/391; 89%) with a PSG SOREMP never received MSLT testing. Patients that were evaluated by MSLT (n = 44; 11%) were typically very sleepy and 82% ended up with a diagnosis of narcolepsy or had MSLTs consistent with current narcolepsy criteria (ie, including the nocturnal SOREMP). Only seven of the 140 patients (n = 5%) that with OSA that first underwent one or more PAP titrations were subsequently seen for an MSLT. Compared to those that eventually received an MSLT, patients that did not receive MSLT testing were older (52 vs. 41 years, p < 0.001), more likely to have moderate to severe OSA (AHI ≥ 15; 39% vs. 16%, p < 0.001), and were generally less likely to report severe sleepiness (ESS ≥ 16; 25% vs. 55%, p < 0.001) and vehicle or workplace accidents or injuries. However, 12% of those that never received an MSLT reported such extreme sleepiness that they endorsed a near-miss car accident due to sleepiness, almost twice as prevalent than that found in a random sample of matched moderate-to-severe OSA patients (p < 0.01). Overall, the reliability of the SOREMP phenotype was low at 9.8%, but was much higher for those diagnosed with type 2 narcolepsy (31%) compared to those without narcolepsy (IH or normal MSLTs; 0%; p < 0.01) or where narcolepsy status was unknown because an MSLT was not conducted (7%; p < 0.01).
Conclusions: The MSLT has been historically underutilized for those exhibiting a SOREMP on diagnostic PSG, a potential marker of narcolepsy. This is presumably because patients with a PSG SOREMP reported variable levels of sleepiness (although some severe) and many had some degree of OSA, which may either be a partial factor in symptomology or even obscure true narcolepsy. Some patients with a PSG SOREMP were very sleepy and most, when an MSLT was conducted, received a diagnosis of type 2 narcolepsy despite few presenting with some of the associated features of narcolepsy. Well-controlled longitudinal studies with high quality data on cataplexy and hypocretin status are needed to understand where the PSG SOREMP phenomenon falls on the hypersomnolence spectrum and to establish which comorbidities share variance with and/or potentially mask narcolepsy. However because untreated narcolepsy can have high social, functional, and financial burden, until such studies are done, physicians should consider a narcolepsy workup when a SOREMP is observed (especially if multiple are seen) as well as close follow-up for symptom resolution when, for example, a patient is treated for sleep apnea.
Keywords: Delay; Narcolepsy; REM.
Copyright © 2017 Elsevier B.V. All rights reserved.