Abstract: Free AccessStimulantsDisrupted Nighttime Sleep in Narcolepsy Thomas Roth, Ph.D., Yves Dauvilliers, M.D., Emmanuel Mignot, M.D., Ph.D., F.A.A.S.M., Jacques Montplaisir, M.D., Ph.D., Josh Paul, M.A., Todd Swick, M.D., F.A.A.S.M., Phyllis Zee, M.D., Ph.D., F.A.A.S.M. Thomas Roth, Ph.D. Address correspondence to: Thomas Roth, Ph.D., Director, Sleep Disorders and Research Center, Henry Ford Hospital, Detroit, MI 48202(313) 916-5171(313) 916-5167 E-mail Address: [email protected] Sleep Disorders and Research Center, Henry Ford Hospital, Detroit, MI , Yves Dauvilliers, M.D. Sleep Unit, University Hospital, Gui de Chauliac, Montpellier, France , Emmanuel Mignot, M.D., Ph.D., F.A.A.S.M. Stanford Center for Sleep Sciences, Stanford University, Palo Alto, CA , Jacques Montplaisir, M.D., Ph.D. Center for Advanced Studies in Sleep Medicine, Hôpital de Sacré-Cœur de Montréal, Centre d'étude du sommeil, Montréal, Québec, Canada , Josh Paul, M.A. Paul Scientific Communications, New York, NY , Todd Swick, M.D., F.A.A.S.M. The Houston Sleep Center, University of Texas-Houston School of Medicine, Houston, TX , Phyllis Zee, M.D., Ph.D., F.A.A.S.M. Sleep Disorders Center, Northwestern University, Chicago, IL Published Online:September 15, 2013https://doi.org/10.5664/jcsm.3004Cited by:93SectionsAbstractPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTStudy Objectives:Characterize disrupted nighttime sleep (DNS) in narcolepsy, an important symptom of narcolepsy.Methods:A panel of international narcolepsy experts was convened in 2011 to build a consensus characterization of DNS in patients with narcolepsy. A literature search of the Medline (1965 to date), Medline In-Process (latest weeks), Embase (1974 to date), Embase Alert (latest 8 weeks), and Biosis (1965 to date) databases was conducted using the following search terms: narcolepsy and disrupted nighttime sleep, disturbed nighttime sleep, fragmented sleep, consolidated sleep, sleep disruption, and narcolepsy questionnaire. The purpose of the literature search was to identify publications characterizing the nighttime sleep of patients with narcolepsy. The panel reviewed the literature. Nocturnal sleep can also be disturbed by REM sleep abnormalities such as vivid dreaming and REM sleep behavior disorder; however, these were not reviewed in the current paper, as we were evaluating for idiopathic sleep disturbances.Results:The literature reviewed provide a consistent characterization of nighttime sleep in patients with narcolepsy as fragmented, with reports of frequent, brief nightly awakenings with difficulties returning to sleep and associated reports of poor sleep quality. Polysomnographic studies consistently report frequent awakenings/arousals after sleep onset, more stage 1 (S1) sleep, and more frequent shifts to S1 sleep or wake from deeper stages of sleep. The consensus of the International Experts' Panel on Narcolepsy was that DNS can be distressing for patients with narcolepsy and that treatment of DNS warrants consideration.Conclusions:Clinicians involved in the management of patients with narcolepsy should investigate patients' quality of nighttime sleep, give weight and consideration to patient reports of nighttime sleep experience, and consider DNS a target for treatment.Citation:Roth T; Dauvilliers Y; Mignot E; Montplaisir J; Paul J; Swick T; Zee P. Disrupted nighttime sleep in narcolepsy. J Clin Sleep Med 2013;9(9):955-965.INTRODUCTIONNarcolepsy is a chronic neurologic disorder defined by a tetrad of symptoms: excessive daytime sleepiness (EDS), cataplexy, hypnagogic hallucinations, and sleep paralysis.1 It affects approximately 0.05% of the general population, with variation according to geography and ethnicity.2 Narcolepsy is associated with substantial morbidity and an impaired health-related quality of life (QOL).3–7EDS is the most common presenting symptom and is necessary for a diagnosis (Table 1),8 although EDS is not specific to narcolepsy.9–11 Cataplexy is the most specific symptom of narcolepsy. It is defined as a sudden, transient partial or total loss of muscle tone triggered by strong emotions, most typically laughing or joking. Although specific, it is not universal, occurring in approximately 75% of patients with narcolepsy as defined by the ICSD-2.2,10,11 Estimates of the prevalence of hypnagogic hallucinations range from one-third to 80% of patients, whereas sleep paralysis affects 25% to 50% of patients.2 Only a minority of patients (10% to 25%) suffer from the complete tetrad of symptoms.2,12Table 1 ICSD-2 criteria for the diagnosis of narcolepsy with and without cataplexy.8Table 1 ICSD-2 criteria for the diagnosis of narcolepsy with and without cataplexy.8Among patients with narcolepsy, disrupted nighttime sleep (DNS) is a common complaint and frequent finding on polysomnographic (PSG) testing, and it is referred to in the contemporary literature as a disease-related symptom. Patients typically report that DNS is more of a problem than are sleep paralysis and hypnagogic hallucinations, and it appears to be more common, with prevalence estimates ranging from approximately 30% to 95%, depending on the definition and sleep assay used.13–15 Thus, DNS may represent an additional characteristic of narcolepsy to be considered as part of a pentad of symptoms rather than EDS and the triad of auxiliary symptoms.Despite the widespread reference to DNS in narcolepsy, there are few systematic patient-report and PSG data describing specific patterns of DNS or its clinical or QOL impact. Finally, there is no consensus characterization of DNS in narcolepsy that distinguishes it from disturbed sleep in other sleep disorder populations, especially primary insomnia, or that differentiates DNS associated with other sleep disorders commonly seen in patients with narcolepsy (e.g., vivid dreaming, periodic leg movements, rapid eye movement [REM] behavior disorder, and sleep apnea) from DNS as a stand-alone symptom. This is an important distinction: DNS may be secondary to a comorbid sleep disorder including insomnia or sleep apnea, or it may affect patients with narcolepsy independently from such conditions. Many patients with narcolepsy appear to experience DNS as a stand-alone and specific symptom that is not associated with other sleep disorders.1,15–18To better characterize the most commonly patient-reported and PSG-determined disruptions to nighttime sleep and to distinguish DNS as a consistent and frequently observed feature of narcolepsy—one that is distinct from insomnia and other sleep-related comorbidities—as well as a possible target for treatment, this article reviews the available literature on DNS in narcolepsy. In this review, we focused on sleep disruptions and PSG findings, not ancillary symptoms.Another approach to characterizing DNS in narcolepsy is to build a consensus among experts that incorporates both clinical experience as well as the published literature. Toward this end, the International Experts' Panel on Narcolepsy was held on June 9-10, 2011, in Minneapolis, Minnesota, to review and discuss the literature on DNS in patients with narcolepsy and their own extensive clinical expertise to clarify the characterization of sleep disturbance in narcolepsy.The meeting was sponsored by Jazz Pharmaceuticals, Inc., Palo Alto, California. Participants attending the International Experts' Panel on Narcolepsy are listed in the appendix, and portions of the panel's proceedings have been incorporated into this characterization of DNS.METHODSWe reviewed and included studies that provided objectively and subjectively collected data on nighttime sleep in patients with narcolepsy. These studies were of variable size, design, and scientific rigor, but each provided data on characteristics of night sleep in patients with narcolepsy such as the number and frequency of PSG-recorded and patient-reported arousals and awakenings, the duration of sleep stages and wake time after sleep onset (WASO), patient-reported sleep quality, and other parameters. The literature included comparisons between patients with narcolepsy and normal controls, patients with and without sleep-related comorbidities, and patients receiving and not receiving narcolepsy-specific medications. Overall, the patient report and PSG studies indicate a nighttime sleep pattern characterized by frequent arousals and brief awakenings that occur independent of other sleep disorders and treatments.Literature Search MethodologyA search of the Medline (1965 to date), Medline In-Process (latest weeks), Embase (1974 to date), Embase Alert (latest 8 weeks), and Biosis (1965 to date) databases was conducted using the following search terms: narcolepsy and disrupted nighttime sleep, disturbed nighttime sleep, fragmented sleep, consolidated sleep, sleep disruption, and narcolepsy questionnaire. There were no date limits on the search.The purpose of the literature search was to identify publications characterizing the nighttime sleep of patients with narcolepsy. Search results were reviewed by the lead author, with other authors consulted as needed to provide consensus on questions or issues arising during the review process. Studies providing specific PSG and/or patient-reported characterizations of nighttime sleep were selected for inclusion in the current article. Review articles and articles describing sleep-related comorbidities, such as periodic leg movements, obstructive sleep apnea, REM behavior disorder, and nightmares, were excluded so as to be able to identify the nature of disturbed sleep in narcolepsy per se without the contamination of other sleep disorders.A total of 642 English-language citations were returned and reviewed for characterizations of nighttime sleep in patients with narcolepsy. When inclusion criteria could not be identified through an abstract, full-text articles were reviewed. As shown in the flow of citations through the review process (Figure 1), of the 642 citations, 20 relevant studies were identified: 12 reported PSG characteristics of nighttime sleep in patients with narcolepsy, 4 reported patient-reported outcomes, and 4 reported both PSG and patient-reported characterizations of sleeping patterns. Baseline data from a randomized clinical trial evaluating the effect of sodium oxybate on nighttime sleep in patients with narcolepsy was also included (another trial was excluded because it did not provide baseline PSG data on patients' nocturnal sleep). Thus, a total of 21 studies met the inclusion criteria (Table 2).14,19–38Figure 1: Flow of citations during the review processDownload FigureTable 2 Publications characterizing the nighttime sleep of patients with narcolepsyTable 2 Publications characterizing the nighttime sleep of patients with narcolepsyTable 2 Publications characterizing the nighttime sleep of patients with narcolepsyRESULTSDNS in NarcolepsyNighttime Sleep in NarcolepsyNighttime sleep in patients with narcolepsy is characterized by a short sleep latency and a sleep-onset REM period in approximately 50% of cases.10,13,39 Aside from the short sleep latency, it is also characterized by an inability to stay asleep.13,22,23,26,31,33,35 Patients with narcolepsy experience frequent awakenings during sleep. The database characterizing such sleep patterns is limited but consistent, with patients reporting frequent awakenings after sleep onset and poor sleep quality,22,31,33–35,38 although only 3 studies provided a quantitative comparison of nocturnal awakenings (Table 3).22,31,33Table 3 The number of nightly awakenings reported by patients with narcolepsy and by comparator groupsTable 3 The number of nightly awakenings reported by patients with narcolepsy and by comparator groupsPatient-Reported DNSThe literature search identified 4 studies that reported on DNS from the patient perspective and 4 studies that included the patient perspective in addition to PSG. Patient reports and sleep diaries from these studies indicate that between 33% and 81% of patients with narcolepsy awaken multiple times throughout the night.23,38 Zachariev and colleagues reported that 16 of 22 (72.7%) consecutive patients with narcolepsy reported DNS, and 36.4% of these patients described the disturbances as frequent awakenings.38 Rosenthal and colleagues, using a 22-item questionnaire, surveyed 237 consecutive adult patients diagnosed with narcolepsy and found that 81.1% of the 119 responders experienced nocturnal sleep disturbances, also characterized as frequent brief awakenings.33 The prevalence of DNS was comparable to the prevalence of patient-reported cataplexy (79.3%), but significantly higher than the prevalence of hypnagogic hallucinations, sleep paralysis, and memory problems (67%, 64.2%, and 46.3%, respectively; p < 0.001 vs. DNS).In the Rosenthal study, patients with narcolepsy and disturbed sleep had significantly less total sleep time (6 vs. 7 h, respectively; p < 0.001) and reported more frequent nocturnal awakenings (4.6 vs. 1.3 per night, respectively; p < 0.001) than did patients who reported no trouble sleeping. PSG data, recorded 7 days after discontinuation of stimulant medications, confirmed these patient reports. Patients were stratified according to whether they reported the presence or absence of difficulty maintaining nocturnal sleep. According to the PSG studies, patients who reported suffering DNS experienced 13.0 awakenings per night vs. 10.5 among those who did not self-report DNS. PSG also demonstrated that patients with DNS had significantly more stage 1 (S1) sleep (23.0% vs. 17.4%, respectively; p < 0.05), significantly less stage 2 (S2) sleep (49.1% vs. 53.5%, respectively; p < 0.05), and significantly more wake time during sleep (72.5 vs. 45.4 min, respectively; p < 0.01), all characteristics of DNS independent of other sleep-related comorbidities.33Parkes et al. reported data on the sleep-wake habits of 183 patients with narcolepsy, 62 patients with idiopathic hypersomnia (IH), and 10 patients with obstructive sleep apnea.31 Responses were compared with 188 controls with normal sleep-wake habits and no complaints of EDS or insomnia. Patients with narcolepsy reported a mean 4.5 nightly awakenings versus 1.4 among controls (p < 0.01) and poor sleep quality, measured using a visual analog insomnia score (43.1 vs. 27.6, respectively; p < 0.01).Sturzenegger and Bassetti compared 40 normal controls with no history of sleep disorders, EDS, or use of sleeping pills with 57 patients with narcolepsy with cataplexy and 56 patients with "nonnarcoleptic hypersomnia," defined by the presence of subjective EDS and an Epworth Sleepiness Score (ESS) ≥ 10 and other sleep disorders.35 Almost 95% of patients with narcolepsy reported falling asleep rapidly, with sleep latency < 20 minutes. Importantly, patient-reported inability to sleep without awakening was significantly more common among patients with narcolepsy than among normal controls (71% vs. 25%, p value not reported), as were early morning awakenings (83% vs. 43%, respectively; p < 0.001). Among patients with nonnarcoleptic hypersomnia, frequent nocturnal awakenings and early morning awakenings were reported by 67% and 71%, respectively. Among normal controls, these rates were 25% and 43%, respectively (p < 0.001 vs. normal controls for both comparisons).35As compared to normal control groups, patients with narcolepsy consistently report disturbed and fragmented nocturnal sleep. This experience is distinct from the nocturnal sleep patterns reported by patients with other sleep disorders. For example, Bruck and Parkes compared the sleep-wake behaviors of 18 patients with IH and 50 patients with narcolepsy with cataplexy using questionnaires and sleep diaries.22 Patients with narcolepsy reported a mean 3.3 awakenings per night, for an average total wake time of 34.4 minutes. By comparison, patients with IH reported only 1.3 nocturnal awakenings (p < 0.005 vs. narcolepsy), with an average total wake time of 7.7 minutes (p < 0.05 vs. narcolepsy).These patient-report studies present a consistent characterization of fragmented, disrupted, and unsatisfactory sleep in patients with narcolepsy. It is clear that nighttime sleep in patients with narcolepsy has its own patterns distinct from those in normal controls and from those with insomnia. Although DNS is important and may be potentially troublesome for patients, there is a paucity of patient-reported data regarding the consequences of DNS. The fact that narcolepsy is associated with a diminished QOL is well established, but how DNS in patients with narcolepsy affects QOL is not well understood.3,5–7,10 In one study using the World Health Organization QOL (WHOQOLBREF) instrument, Rovere et al. compared QOL among 40 patients with narcolepsy and 40 normal controls with fixed daytime work schedules and no history of chronic diseases or sleep disorders.34 Dissatisfaction with sleep was common among patients with narcolepsy, with 22.5% being very unsatisfied and 45.0% unsatisfied. Among normal controls, the rates were 2.5% and 10.0%, respectively (p < 0.001). The rates of regular, satisfied, and very satisfied sleep among patients with narcolepsy were 17.5%, 15.0%, and 0.0%, respectively. Among normal controls, these were 22.5%, 50.0%, and 15.0%, respectively (p < 0.001 for comparisons). The results of this study do not address the QOL impact or other consequences of DNS in narcolepsy. The study does, however, confirm the PSG and sleep questionnaire data indicating high rates of overall dissatisfaction with sleep among patients with narcolepsy.PSG Findings of DNS in NarcolepsyObjective data from the 12 PSG-specific studies,14,19,20,23,25–30,36,37 the 4 with PSG in addition to patient reports,32,33,35,38 and the 1 randomized clinical trial21 confirm the patient-reported data that nighttime sleep in patients with narcolepsy is generally disrupted and fragmented. PSG comparisons between patients with narcolepsy and normal controls with no history of sleep disorders consistently find that patients with narcolepsy experience frequent brief awakenings and arousals (e.g., excessive shifts to S1 sleep or wake from deeper sleep stages), therefore having more time spent in S1 sleep, an elevated WASO, and reduced sleep efficiency relative to normal controls (Table 4 and Figure 2).20,23,24,26–30,32Table 4 Polysomnographically identified arousals and wake time after sleep onset (WASO) among untreated patients with narcolepsy and normal controlsTable 4 Polysomnographically identified arousals and wake time after sleep onset (WASO) among untreated patients with narcolepsy and normal controlsFigure 2: Comparison of the stages of sleep between patients with narcolepsy and normal controlsS1, stage 1; S2, stage 2; S3, stage 3; S4, stage 4; SWS, slow wave sleep. Frauscher et al.,23 Jiménez-Correa et al.,26 Mukai et al.30Download FigureThese nighttime sleep patterns are present whether or not patients are receiving stimulant medications,20,22,32 including in drug-naïve patients with narcolepsy,30 albeit in a small sample (N = 8), as well as when compared with other sleep disorders.32,36 These observations provide further support for an association between narcolepsy and DNS that is also independent of stimulant use, and although the studies do not identify underlying causes of DNS, several suggest an abnormality in non-REM (NREM) sleep as a possible mediator. However, no studies have directly compared DNS in patients with narcolepsy who are on and off stimulants. This lack of objective data on the effects of stimulants on DNS suggests the need for studies addressing this issue.The largest study to characterize PSG-assayed sleep in patients with narcolepsy is a 530-patient database created for a clinical trial of modafinil (US Modafinil in Narcolepsy Multicenter Study Group, 1998).24 All prescription and over-the-counter stimulant- and sedative-containing medications were discontinued 2 weeks prior to patients undergoing 2 consecutive baseline PSG measurements. On both nights, patients displayed evidence of nocturnal sleep disruption, characterized by an arousal index of 12 and an awakening index of 3.8. Analyses of smaller patient groups with narcolepsy in Asia and Eastern Europe have reported similar DNS arousal rates.14,38Comparisons between patients and normal controls confirm the conclusion that these sleep parameters are deviations from normal sleep. Mukai and colleagues compared PSG data between 8 drug-naïve patients with narcolepsy and 8 healthy normal controls with no history of sleep disorders.30 Patients with narcolepsy had significantly more S1 sleep (17.2% vs. 8.9%, respectively; p < 0.01), more stage shifts (mean, 352.6 vs. 251.8, respectively; p < 0.05), and a higher percentage of WASO (3.9% vs. 1.0%, respectively; p < 0.05) than did normal controls. These investigators suggest that DNS is linked to the higher REM densities seen in patients with narcolepsy.In several studies, DNS in patients with narcolepsy has been differentially associated with NREM sleep. Frauscher and colleagues, using nocturnal video-PSG studies, compared motor events during REM and NREM sleep in 6 stimulant-free patients with narcolepsy/cataplexy with 6 sex- and age-matched normal controls with no history of sleep disorders, ESS < 10, and no central nervous system-active medication at the time of investigation.23 PSG was performed on 2 consecutive nights and video monitoring for 1 night. All visible movements were defined as a motor event and classified according to topography, number of involved body parts, duration, and its association with arousal.Patients with narcolepsy with cataplexy had more fragmented sleep with more S1 sleep (20.0 vs. 9.1 min, respectively; p = 0.071) and less S2 sleep (35.5 vs. 51.5 min, respectively; p = 0.022) than did normal controls. The mean motor activity index, defined as movement episodes per hour of sleep excluding electromyographic-related muscle activity in the chin or tibialis anterior channels, was 59.9/h of sleep in patients with narcolepsy with cataplexy versus 15.4/h of sleep among controls (p = 0.004). Motor event indices were similar between REM and NREM sleep periods in the narcolepsy with cataplexy group, whereas motor event indices were higher during REM sleep periods in controls (p = 0.028). Motor events affected significantly more body parts in patients with narcolepsy with cataplexy than controls, and events lasting > 1 second were significantly more common (p = 0.011 and p < 0.001, respectively). The total arousal index among patients and controls was 21.6 versus 8.7, respectively (p = 0.004), and the motor activity-related index was 17.6/h of sleep versus 5.9/h, respectively (p = 0.002). The investigators hypothesized that motor disturbances during NREM sleep in patients with narcolepsy with cataplexy exacerbate DNS in patients with narcolepsy with cataplexy and may be an underlying cause.23Khatami et al. reported that patients with narcolepsy with cataplexy have decreased NREM sleep intensity and suggested that this may mediate DNS.27 Their analysis compared PSG findings between 11 treatment-naïve patients with narcolepsy with cataplexy with 11 matched controls. As in other studies, patients with narcolepsy with cataplexy exhibited more S1 sleep, lower sleep efficiency, and more frequent awakenings/ arousals than did controls. Patients with narcolepsy exhibited longer NREM-REM cycles than did controls and a more dramatic decline in slow wave sleep across successive NREMREM cycles. This was evidenced by less slow wave activity during the second NREM sleep episode among patients. In parallel, during this second NREM sleep episode, patients with narcolepsy with cataplexy experienced more wake episodes than did controls (p = 0.01).27Based on the hypothesis that insufficient NREM intensity is associated with sleep disruptions, Khatami et al. compared 6 drug-free patients with narcolepsy with cataplexy and 6 matched controls after 40 hours of wakefulness.28 Prior to sleep deprivation, slow wave activity decreased in both groups across successive NREM-REM sleep cycles, but, as in the previous study, slow wave activity was smaller among patients with narcolepsy during the second NREM-REM cycle (p = 0.02). During recovery sleep, initial slow wave activity increased in both groups, and there were no sustained differences between groups across sequential NREM episodes. That is, during recovery sleep, there were fewer awakenings during the first and second NREM sleep episodes (p = 0.01 and p = 0.02, respectively) than at baseline, suggesting that increased NREM intensity associated with sleep deprivation reduces disruptions to nocturnal sleep among patients with narcolepsy with cataplexy.These studies provide a consistent characterization of fragmented nighttime sleep in drug-free patients with narcolepsy. Similar results have also been observed in patients receiving active treatments. One study compared normal controls (n = 50) with patients with narcolepsy with cataplexy (n = 50), 36 of whom were being treated for narcolepsy at the time of the study.20 Patients with narcolepsy had significantly greater WASO (92.7 vs. 51.1 min, respectively; p < 0.01), more total wake time (104.9 vs. 77.3 min, respectively; p < 0.01), and a greater percentage of S1 sleep (12.4% vs. 7.8%, respectively; p < 0.01).Rogers et al. compared PSG and sleep diary data from 25 treated patients with narcolepsy with gender-matched controls.32 All patients were receiving stimulants, and 5 patients were also taking REM-suppressant medication. Treatment response was considered satisfactory by both the patient and the patient's physician. The controls had no history of sleep, neurologic, or psychiatric disorders, alcoholism, use of drugs affecting the central nervous system, or shift work. Patients recorded sleep-wake patterns, activities, and medication use in the diaries and underwent PSG assessment. Total sleep time was comparable between patients and controls, but patients with narcolepsy had more WASO (57.5 vs. 34.5 min, respectively; p ≤ 0.05), S1 sleep (15.3% vs. 10.1%, respectively; p ≤ 0.05), and more frequent awakenings, although this parameter failed to reach statistical significance.Finally, studies have compared the nighttime sleep of patients with narcolepsy with the nighttime sleep of patients with other sleep disorders.19,36 Baker et al. found that patients with narcolepsy had more DNS than did patients with IH, characterized by more S1 sleep, greater WASO, and lower sleep efficiency.19 More recently, Takei and colleagues reported PSG findings among patients with narcolepsy both with and without cataplexy (n = 52 and n = 62, respectively) and patients with IH without long sleep times (n = 50).36 Patients with narcolepsy with cataplexy experienced more arousals per hour than did those without cataplexy and those with IH (17.1 vs. 11.7 vs. 10.1, respectively; p < 0.01). There was no statistically significant difference in the number of arousals between patients with narcolepsy without cataplexy and those with IH. Patients with cataplexy also had a greater percentage of S1 sleep than did those without cataplexy and those with IH (15.6% vs. 9.2% vs. 10.8%, respectively; p < 0.01). Sleep efficiency was lowest (p < 0.05) and WASO highest (p < 0.01) among patients with cataplexy. The data suggest that DNS occurs more frequently among patients with narcolepsy with cataplexy than without.36PSG Findings and EDSNighttime sleep in narcolepsy, as defined by PSG, parallels that of patient reports of fragmented sleep, with the preponderance and consistency of data suggesting that DNS is a characteristic of narcolepsy per se, not a comorbid condition. Interestingly, DNS may be related to lower drive for sleep as evidenced by a greater decline across the night in slow wave sleep among patients with narcolepsy. This is especially true for patients with narcolepsy with cataplexy. However, this is not a consistent finding.27,28,40,41Despite the consistency of reports of DNS in narcolepsy, few studies have attempted to correlate DNS with narcolepsy's daytime symptoms.24,26 Those that have, produced inconsistent results or found suggestive, but not completely persuasive, correlations between DNS and EDS.In the US Modafinil in Narcolepsy Multicenter Study, there was a significant, albeit modest, relation between DNS and daytime sleepiness evaluated with the multiple sleep latency test (MSLT), maintenance of wakefulness test, and ESS. Thus they concluded that whereas DNS may exacerbate EDS, DNS is not the primary cause of EDS in narcolepsy.24Jiménez-Correa and colleagues compared patients with narcolepsy and healthy matched controls with no history of sleep disorders (n = 23 in each group) and found that patients with narcolepsy had a greater percent of S1 sleep (16.60% vs. 10.57%, respectively; p < 0.05) and S2 sleep (58.36% vs. 52.09%, respectively; p < 0.05), a significantly greater number of arousals (78.38 vs. 29.13, respectively; p < 0.05), and more awakenings lasting < 3 minutes (mean, 35.15 vs. 3.77, respectively; p < 0.05).26 Using measures of daytime sleepiness (ESS, MSLT, subjective sleep quality) and other tools, investigators reported moderate correlations between EDS and DNS. Daytime sleepiness was associated with WASO and frequency of awakenings, and lower total sleep time, sleep efficacy, and percent of slow wave sleep.26Although fragmented sleep has generally been associated with EDS, the limited number of studies on the role of DNS in producing daytime sleepiness in patients with narcolepsy, and their contrasting results, do not present a clear picture of this relationship. Consequently, further studies are warranted for clarifying this relationship, and addressing the question of whether dire