Polysomnographic findings in men over 55 years of age with narcolepsy type 1

Overview

Aim: The aim of this study was to compare polysomnographic parameters of nocturnal sleep in male patients with narcolepsy type 1 aged over 55 years with age-matched controls. Patients and methods: The cohort consisted of 17 men (mean age 66.1 ± 8.3 years) with narcolepsy type 1 whose previously obtained polysomnographic records were compared in terms of basic parameters of sleep macrostructure and its basic disorders with polysomnographic records of age-paired men (mean 65.0 ± 6.6 years). Results: Compared to controls, males with narcolepsy type 1 had a significantly lower percentage of NREM 3 (8 vs. 26%; P = 0.003) and REM sleep stages (12 vs. 19%; P = 0.01) and a higher presentation of NREM 1 stages (17 vs. 10%; P = 0.01). The representation of wakefulness (29 vs. 26%) and NREM 2 stages (33 vs. 30%), mean sleep latency, REM sleep latency, sleep duration, and sleep efficiency (67 ± 14 vs. 72 ± 17%) did not differ between groups. The incidence of obstructive sleep apneas was comparable (apnea-hypopnea index 25 ± 26 vs. 23 ± 18). The index of periodic limb movements during sleep was significantly higher (46 ± 31 vs. 16 ± 22; P = 0.02) and the presence of REM sleep without atonia (47 vs. 8%; P = 0.02) was more frequent in the narcoleptic group. Conclusion: Men with narcolepsy type 1 have significantly less NREM 3 and REM stages and more NREM 1 stages.

Keywords:

obstructive sleep apnea – polysomnography – sleep disorders – narcolepsy with cataplexy – sleep macrostructure

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version.

 

Introduction

Narcolepsy type 1 (NT1, the older name / alternatively narcolepsy with cataplexy) belongs to central hypersomnias [1]. NT1 is a chronic disorder with a usual onset in youth and a prevalence of approximately 20-50/100,000. The pathophysiological basis of NT1 is the loss of function of hypocretin neurons in the lateral hypothalamus, probably on an autoimmune basis [2]. The original characteristic tetrad of symptoms, i.e. excessive daytime sleepiness, cataplexy, hypnagogic hallucinations and sleep paralysis [3], was gradually joined by a fifth symptom, fragmented nocturnal sleep [4].

Patients with NT1 typically describe abundant dream activity and perceive nighttime sleep as poor quality. NT1-specific manifestations of REM (rapid eye movement) sleep dissociation, i.e. sleep paralysis and hypnagogic hallucinations, may play a role in the subjective perception of poor sleep quality. In addition, the sleep of NT1 patients is disturbed by other sleep comorbidities that are not specific to NT1 but are likely to occur more frequently in these patients than in the general population. These include snoring, obstructive sleep apnoea (OSA), restless legs syndrome, periodic limb movements in sleep (PLMS) and rapid eye movement behaviour disorder (RBD). Finally, the quality of sleep in NT1 may be affected by psychoactive medications used to treat daytime sleepiness and cataplexy, and an increased incidence of mood disorders [2].

Polysomnographic sleep studies agree that NT1 is characterized by early onset REM sleep (sleep onset REM; SOREM) - the onset of REM sleep within the first 15 min after falling asleep [5]. SOREM was present in about 35% of the nocturnal records of NT1 patients [5], whereas it is quite rare in the records of healthy people without sleep deprivation and without depression [6]. The recording of at least two SOREMs during night sleep and during the multiple sleep latency test is part of the diagnostic criteria for NT1 [1].

Most studies comparing the macrostructure of sleep in adults with NT1 versus control subjects concluded that wakefulness after falling asleep and NREM stage 1 sleep were more abundant in NT1. In contrast, the proportions of NREM 2 and NREM 3 sleep stages were lower. The proportion of REM sleep was not significantly different [7-12]. OSA is present in up to 25% of NT1 [13]. A periodic limb movement index (PLMI) in sleep greater than 5 was found in 67% of NT1 patients compared with 37% of controls, and increasing PLMI with increasing age was observed in both patients and controls [14]. The polysomnographic diagnostic criterion for RBD is the demonstration of REM sleep without muscle atonia [1,15]. The prevalence of RBD in the NT1 population varies according to the methodology, but is orders of magnitude higher than in the general population - up to 50% according to a recent meta-analysis of polysomnography studies (up to 70% in clinical trials). The prevalence of RBD in NT1 does not change with age [16]. It should be emphasized that most objective sleep studies in narcolepsy have been performed in patients at the time of diagnosis, i.e. in young patients.

Ageing is generally associated with a reduced ability to induce and maintain sleep [17]. Moreover, sleep in old age is negatively affected by more frequent illnesses and increased medication. In the elderly, there is a shift in circadian preference towards the morning chronotype - a tendency to fall asleep earlier and wake up earlier [17].

In the elderly without sleep disorders, polysomnographic evidence of impaired sleep continuity, multiplied wakefulness after falling asleep, and reduced sleep efficiency (the proportion of total sleep time to total polysomnographic recording time expressed as a percentage) was found. Sleep latency was prolonged in the senile group only in women [17]. During life, there is a gradual decrease in the proportion of NREM stage 3 and REM sleep, and conversely a prolongation of NREM stages 1 and 2. These changes in sleep macrostructure are pronounced in childhood and adolescence, with a gradual evolution in adulthood. A further more rapid change in the proportion of NREM sleep stages does not occur until approximately 55 years of age. [18] Total sleep duration (the sum of the duration of all sleep stages not wakefulness) decreases from childhood to about 35 years of age and does not change much further, with a slight decline occurring after age 80 [18]. The prevalence of OSA in the general population over the age of 65 years is reported to be in the wide range of 8-50% [19]. OSA is more common in men [20,21]. PLMS are rare before the age of 40 years, but their incidence increases at later ages. At age 65+ years, the prevalence of PLMS (PLMI > 5) is estimated to be 45% [1]. The prevalence of RBD is estimated to be around 2-5% between the ages of 50 and 60 years and 7-8% in older individuals (ascertained by questionnaires). In studies based on polysomnographic examination, the proportion affected was lower. Isolated RBD at a young age is rare [22].

Given the near absence of objective information on nighttime sleep in elderly NT1 patients, we decided to compare baseline information from nocturnal sleep polysomnography in NT1 patients over 55 years of age with age-matched controls. Because of the aforementioned sex differences, we focused only on sleep in men.

 

Patients and methods

We included previously consecutively obtained polysomnographic records of 17 men aged 55 years and older with a diagnosis of NT1 established at our institution. These were patients who had already been followed up for a long time with a diagnosis of NT1 at our institution. The cutoff age of 55 years was chosen because this is the age at which changes in sleep macrostructure are accentuated in the general population, as described in the introduction. Men were chosen because of known differences in sleep macrostructure and the prevalence of sleep comorbidities between men and women, as described in the introduction. Thus, all patients who met the selected criteria and who underwent a follow-up polysomnographic examination between 2012 and 2017 were included. The examination verified that the patient met the diagnostic criteria of NT1 according to the International Classification of Sleep Disorders, 3rd edition (Table 1) [1] at the time of diagnosis. The level of hypocretin in the liquor was not determined because this method was not available at our department at the time of data acquisition. Thirteen patients were completely free of sleep-influencing medications for 14 days before the examination (either not taking medication or discontinued). In four patients, chronic medication could not be discontinued. One patient was taking stimulants (combination of modafinil and methylphenidate), two were taking modafinil and low-dose antidepressants (clomipramine, citalopram), and one patient was taking sodium oxybate in combination with modafinil.

Table 2 summarises basic information about the subjects in the study. The control group consisted of volunteers who were financially compensated for their time in the study. No healthy volunteer had a history of a sleep disorder or any other illness that may cause sleep disturbance. Lipidemia, uricemia, glucose, and thyroid hormone levels were determined in control subjects, all with normal results. Healthy volunteers were enrolled after neuropsychological exclusion of significant depression or anxiety and dementia and after magnetic resonance imaging studies excluding structural brain lesions. All members of the control group had not taken any medication affecting sleep for at least 14 days before the examination.

Polysomnographic recordings of NT1 patients were obtained as part of a standard protocol involving first nocturnal polysomnography and the following day's Multiple Sleep Latency Test. Control subjects were examined using the same protocol. Patients and healthy controls underwent nocturnal polysomnography in the monitoring unit of our institution. Polysomnographic recordings were made with RemLogic, version 3.4.1 (Embla systems, Thorton, CO, USA) and NicoletOne EEG (Natus Medical Inc, Pleasanton, CA, USA). Registration consisted of two electrooculogram (EOG) leads, six electroencephalogram leads (F3-M2, C3-M2, O1-M2, F4-M1, C4-M1, O2-M1), and surface electromyography of the musculi mentales, flexores digitorum superficiales, and tibiales anteriores, electrocardiogram, intranasal pressure, anterior nasal airflow, chest and abdominal respiratory effort, hemoglobin oxygen saturation, patient sounds and digitally synchronized video. Monitoring was performed in a separate room from 10 pm to 6 am. Polysomnographic recordings were scored according to the American Academy of Sleep Medicine Sleep and Sleep Events Scoring Manual, version 2.2 [15]. Sleep stages were standardly scored in 30s epochs. In terms of sleep macrostructure, the following parameters were compared: sleep latency (time from the start of registration - extinction - to the start of the first sleep stage), total sleep time, REM sleep latency (time from sleep onset to the start of the first epoch scored as REM sleep), sleep efficiency, and the percentage of each sleep stage and wake after sleep onset (WASO).

Among the other parameters, also scored according to the rules of the American Academy of Sleep Medicine [15], the following values related to sleep breathing were included in the statistical processing: apnoea/hypopnoea index (AHI - number of apnoeas and hypopnoeas per hour of sleep), number of desaturations per hour of sleep - oxygen desaturation index (ODI), where we defined desaturations as decreases in haemoglobin oxygen saturation of 3% or more, proportion of sleep time with blood oxygen saturation less than 90% (T90). Furthermore, PLMI, the number of periodic limb movements in sleep per 1 h of sleep, was assessed. Movement was determined by surface electromyography over the musculi tibiales anteriores bilaterally. To be classified as periodic limb movements, the following criteria had to be met in surface electromyography leads over the musculi tibiales anteriores (one or both): an increase in amplitude of at least 8 µV and a duration of 0.5-10 s with a minimum of 4 repetitive movements with 5-90 s intervals. An epoch of REM sleep has been termed an epoch of REM sleep without atonia in the presence of either persistent (tonic) activity in the chin leads (at least 50% of the epoch higher than in NREM) or transiently increased (phasic) activity in the EMG of the chin or limbs (at least 50% of the 3 s mini-epics within the epoch contain 0.1-5 s of discharges of at least 4× increased amplitude compared to background) [15].

Statistical data processing was performed using STATISTICA version 12 software (StatSoft). The source dataset contained nominal, ordinal and scaled variables with different statistical properties, which determined the choice of methods used for hypothesis testing. For nominal variables, Pearson chi-square was used because the number of cases in each subgroup was sufficiently high. For continuous variables, the non-parametric Mann-Whitney U test was used because the distribution of parameters obtained by polysomnography is not normal in principle and the number of subjects in each group is not sufficient for the asymptotic assumption of normality required for the parametric t-test. We considered the critical value of the probability of first-order error to be a significance level of p < 0.05.

 

Results

The monitored sleep macrostructure parameters are shown in Table 3. No significant difference was found between the NT1 and control groups in sleep latency, REM sleep latency, total sleep duration, or sleep efficiency. SOREM did not occur in any control subject but was present in four patients.

The percentage of sleep stages is illustrated in Figure 1. Both groups, have a similar proportion of W and NREM 2 stages. NT1 has significantly shorter NREM stage 3 and REM sleep and longer NREM stage 1.

A significant amount of apneas and hypopneas were found in both groups of subjects. The commonly used OSA threshold of AHI > 5 was exceeded by 16 (94.1%) patients and 14 (82.3%) control subjects (p = 0.56). The threshold for moderate sleep apnoea (AHI > 15) was exceeded by 10 patients and the same number of controls (58.8%). The mean AHI, ODI and T90 were not significantly different (table 4).

A higher PLMI was found in the NT1 group (table 4). PLMI > 5 had 13 (76%) subjects with NT1 and 9 (53%) controls (p = 0.16), and PLMI > 15 had 11 (65%) subjects with NT1 and 5 (29%) controls (p = 0.036).

REM sleep without atonia was observed in one control (6%) compared to 8 (47%) NT1 patients (p = 0.002).

Discussion

This study is the first to compare polysomnographic parameters of nocturnal sleep of men with NT1 and sleep of control subjects in isolation. In terms of sleep macrostructure, NT1 patients had a significantly lower percentage of NREM3 and REM sleep stages and a higher proportion of NREM1 stage compared with controls. Mean sleep latency, REM sleep latency, sleep duration, representation of wakefulness after falling asleep, and sleep efficiency did not differ between the groups. The prevalence of OSA was comparable in both groups. PLMI was significantly higher in the NT1 group and the presence of REM sleep without atonia was more frequent in the NT1 group.

The latency to fall asleep of healthy men aged 50-80 years is close to that of our healthy controls (13.4 ± 11 min) according to a large population-based study (15.3 ± 17 min) [17]. According to this work, sleep latency in men does not change much with age. The latency to fall asleep of our NT1 patients (10.8 ± 20 min) is very similar to the latency to fall asleep of men with NT1 of mean age 37 years from the European Narcolepsy Database (EU-NN database) (11.9 ± 28 min) [5]. Our results are consistent with the findings in the general population, i.e., that the latency to fall asleep in men does not change significantly with age. In the work of the Italian authors, who presented polysomnographic results of NT1 patients in the age groups 0-10, 11-18, 19-44, 45-64, > 64 years after approximately 20 untreated patients, the mean sleep latency of NT1 patients > 64 years was 7.4 ± 6.6 min. This is less than in our study. Yet, according to this study, the latency to fall asleep in narcolepsy increased with age (however, these were both males and females) [23]. Sleep latency and other macrostructural parameters are influenced by registration conditions, so comparison with other results is only indicative.

SOREM occurred in four (24%) of our NT1 patients, and the mean REM sleep latency of our patients was 116.0 ± 89.1 min, which is close to that of the healthy population. According to the EU-NN database, REM sleep latency in adult male NT1 patients is 55.0 ± 67.4 min and SOREM in nighttime sleep of NT1 patients occurred in 35% [5]. The influence of medication can be considered, but we do not find a significant effect of medication in our data. Of the three patients taking stimulants, two patients had SOREM (one of them was additionally taking sodium oxybate). The two patients taking antidepressants did not have SOREM; their REM sleep latency was 74.5 and 67.5 min, respectively. According to our results, it could be inferred that REM sleep latency in NT1 is prolonged in later age and thus approaches that of the general population. However, the number of our study subjects is small. An Italian study did not reveal a significant difference in REM sleep latency between age groups of NT1 patients [23].

Compared to controls, our NT1 patients have a lower percentage of NREM stage 3 (8 vs. 26%) and REM sleep (12 vs. 19%) and a higher percentage of NREM stage 1 (17 vs. 10%). Some authors find a lower proportion of NREM stage 3 when compared with healthy younger NT1 subjects of both sexes [7,11,12], but in other papers the difference is not significant [8-10]. The proportion of NREM 3 also decreases with age in the healthy population [17,18]. A higher proportion of NREM stage 1 in NT1 in general has been reported by a number of papers [7,9-12] and is now considered typical, although very non-specific for NT1. According to our results, this can be confirmed in men with NT1 in the senile stage. However, the prolongation of the NREM1 stage is a phenomenon also typical of sleep aging in healthy humans. Shorter REM sleep duration in NT1 patients compared to control subjects, as in our study, has not been shown by any other work with younger patients [7-12].

Almost all studies that do not consider the age of the patients agree on a higher proportion of wakefulness after falling asleep in NT1 compared to controls [7,9-12], which is not apparent from our comparison in older men because our control subjects also have an increased proportion of WASO. WASO is prolonged with age even in healthy people. Thus, it appears that in the general population, the proportion of WASO may level off with NT1 in old age by impairing sleep continuity. Sleep efficiency, like WASO, is not significantly different in our cohort compared to comparisons in younger subjects [8-10]. The mean sleep efficiency in both our cohorts is well below the guideline normal range (< 85%) for all adulthood [24]. Similar to WASO, sleep efficiency in healthy people decreases with age. Total sleep time is also comparable in both our cohorts, but this is also described by some comparisons of younger NT1 patients with controls [7,8,11,12]. All signs of impaired sleep macrostructure may be partly explained by the high OSA rates in both our study groups.

The proportions of sleep stages in our 17 men with NT1 can be confronted with the results of men with NT1 in the large European Narcolepsy Database (EU-NN) at the age of 36.9 ± 17.1 years [5]. The proportion of NREM stage 1 of 17.4% in our patients is almost identical to the results in the EU-NN, where the value is 16.5%. The proportion of NREM stage 2 is on average 10% shorter in our patients (33.4 vs. 43.0%), which cannot be attributed to a high significance. However, the proportion of NREM stage 3 is essentially half that of the EU-NN average (7.7 vs. 16.4%) and the REM stage is also shorter in our cohort (12.1 vs. 19.3%) [5]. Comparing the values of our NT1 patients with the EU-NN results, one could infer a prolongation of the proportion of wakefulness after falling asleep and a reduction of all sleep stages in NT1 patients in old age. A reduction in REM sleep and NREM stage 2 at age over 64 years was also found by the authors of the Italian study [23]. However, the proportion of NREM stage 3 and NREM stage 1 in older patients increases in this Italian study. The increase in the proportion of NREM stage 1 is consistent with our results and can also be explained as a general phenomenon of sleep ageing [18]. The increased proportion of NREM stage 3 sleep in the oldest group of patients aged >64 years in the Italian study [23] is difficult to explain because it has no theoretical explanation and is not supported by clinical experience.

Obstructive sleep apnoea was abundant in both groups of men, our patients and controls, despite the fact that the men in the control group had no subjective sleep difficulties and had a lower body mass index (BMI). The AHI values of our NT1 patients were higher than in other studies, but in which younger patients were included and women were also included. According to Sasai-Sakuma et al. AHI ≥ 5 was present in 14.4% of NT1 patients with a mean age of 28.2 ± 10.9 years [25], and according to Pizza et al. AHI ≥ 5 in 31% of NT1 patients with a mean age of 40.0 ± 16.0 years [26]. In a cohort of 20 untreated NT1 patients > 64 years of age (BMI > 25 in 90%), the AHI was 17.2 ± 16.8, which is also lower than in our cohort [23]. However, the literature and our results suggest that the prevalence of OSA in narcolepsy also increases with increasing age as in the general population. It can be inferred that a higher proportion of OSA in NT1 patients compared to the general population is present at younger ages. This could be due to the increased BMI in NT1 at younger ages [27] and the fact that the BMI difference in NT1 compared to the general population does not increase later [28].

We found significantly higher PLMI in NT1 patients versus controls, which is consistent with previous publications in adult patients and controls, the results of which are summarized by a meta-analysis [29]. In a group of 20 untreated NT1 patients > 64 years, the PLMI was 26.2 ± 18.1 [23], which is lower than our mean value. According to the EU-NN database, the proportion of patients with PLMI > 5 was 49.4% and the proportion of patients with PLMI > 15 was 30.4% [5]. Thus, it can be assumed that this proportion increases with age, since 76% of our NT1 patients had PLMI > 5 and 53% of our NT1 subjects had PLMI > 15. This is in agreement with results from France, according to which PLMI in NT1 increases with age, especially at age 50+ years [14]. PLM is common in the general population over 60 years of age, and our result of PLMI > 15 in 29% of controls is not different from findings in other larger populations of otherwise healthy people [30].

The frequency of RBD is relatively high in adults with NT1 compared to the healthy population [16]. According to the EU-NN database, the incidence of RBD is slightly lower in men with NT1 (46% in men with NT1 vs. 54% in women with NT1) [5]. Our result of 47% of patients with polysomnographically confirmed REM sleep with atonia disorder is similar to the average of younger patients with NT1. This is in agreement with the previously published view that the proportion of patients with RBD does not further increase in NT1 in the elderly [31] in contrast to the general population [22]. However, the incidence of REM sleep epochs with atonia disorder is also significantly more frequent in NT1 patients over the age of 55 years compared to healthy controls.

The shortcomings of this work include the small number of subjects studied and the inclusion of four patients on medication. However, the results comparing the NT1 set of unmedicated patients were not different. Only generally healthier and more motivated subjects could be included in the study, and thus the study did not include patients or controls who were unable to participate for medical or technical reasons. This is, however, an inherent problem in all studies of the elderly that take place outside the subject's home. The abundance of OSA in controls and patients distorts sleep changes directly related to the underlying pathophysiology of NT1.

 

Conclusion

Macrostructural characteristics of sleep quality in men with NT1 aged over 55 years are only partially worse compared to the control group, whereas according to published studies on younger patients the difference is more significant. This suggests that sleep in NT1 deteriorates in leaps and bounds in the early years of the disease, whereas in healthy people sleep deteriorates continuously with age. This study can also be seen as a confirmation of anamnestic studies that even in a series of NT1 patients, night sleep is worse than in the general population.

 

Ethical aspects

The work was carried out in accordance with the Helsinki Declaration of 1975 and its revisions in 2004 and 2008. The study was approved by the Ethics Committee of the General University Hospital (date of approval: 18 September 2014, approval numbers 1812/14 IS and 85/21 GIP Grant). Participants signed an informed consent before entering the study.

 

Grant support

The study was supported by the grant AZV NU20-04-00088 from the Ministry of Health of the Czech Republic, the Internal grant GIP-21-SL-04-212 from the General University Hospital in Prague and the Cooperatio Neuroscience grant from Charles University

Conflict of interest

The authors declare that they have no conflict of interest in relation to the subject of the study.

 

Tables

Table 1: NT1 diagnostic criteria according to the International Classification of Sleep Disorders, 3rd edition .[1]

NT1 diagnosis criteria are met when A + B (1st or 2nd) .[1]

1. The patient has had daily irresistible periods of somnolence or falling asleep for at least the past 3 months.
2. 1. a history of cataplexy and in the multiple sleep latency test a mean sleep latency ≤ 8 min and at least 2 SOREMs (1 SOREM can be replaced by a SOREM during night sleep)

2. the concentration of hypocretin in the cerebrospinal fluid is ≤ 110/pg/ml or < 1/3 of the normal standardized values by the test method.

 

NTI - narcolepsy type 1; SOREM - onset of REM sleep

 

 

Table 2. Basic information about the subjects in the study.

	NT1	Controls	p
number	17	17
age (years), mean (SD)	66,1 ± 8,3	65.1 ± 6.6 years	n.s.
BMI, mean (SD)	31,8 ± 5,3	28,0 ± 3,7	0,02
disease duration (years), mean (SD)	38,8 ± 17,9	N. A.

BMI - body mass index; n.s. - statistically insignificant value (p > 0.05); N.A. - not applicable; NT1 - patients with narcolepsy type 1; SD - standard deviation

 

 

Table 3. Macrostructural sleep parameters.

	NT1	Controls	p
Sleep latency (min), mean (± SD)	10,8 (±20)	13,4 (±11)	n.s.
REM latency (min), mean (± SD)	116,0 (± 89)	83,4 (±56)	n.s.
number of records with SOREM (%)	4 (24 %)	0	0,04
total sleep duration (min), mean (± SD)	324 (±70)	343 (±80)	n.s.
sleep efficiency (%), mean (± SD)	67,4 (±14)	71,5 (±17)	n. s.
WASO (min), mean (± SD)	146 (±68)	123 (±74)	n.s.

n.s. - statistically insignificant (p> 0.05); NT1 - patients with narcolepsy type 1; REM - rapid eye movement; SD - standard deviation; SOREM - sleep onset REM sleep; WASO - wakefulness after sleep

Table 4. AHI, ODI, T90 and PLMI results.

	NT1	Controls	The value of p
AHI mean (± SD)	25,0 (± 26,0)	22,7 (± 18,4)	n.s.
ODI mean (± SD)	24,0 (± 26,2)	21,2 (± 15,8)	n.s.
T90 diameter (± SD)	7,7 (± 10,7)	4,6 (± 7,1)	n.s.
PLMI mean (± SD)	46,0 (± 31,3)	15,6 (± 22,3)	0, 02

Values rounded to whole numbers.

AHI - apnea hypopnea index; Controls - control subjects; n.s. - statistically insignificant value (p > 0.05); NT1 - patients with narcolepsy type 1; ODI - oxygen desaturation index; PLMI - periodic limb movement index;SD - standard deviation; T90 - time spent in saturation < 90% (value given in percentage)