SEARCH

SEARCH BY CITATION

Keywords:

  • circadian rhythm;
  • Parkinson's disease;
  • sleepiness

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Normal subjects show an increase of sleepiness in the morning, early afternoon and before sleep. In the advanced stages of Parkinson's disease (PD) the mean level of sleepiness is quite high, while with respect to healthy subjects it seems to be unchanged in the early stages. The aim of this study was to evaluate the time–course of the sleepiness level during the wakefulness period in untreated patients with early-stage Parkinson's disease. Eighteen Parkinson's disease patients who had never been treated before with dopaminergic drugs (male = 9, female = 9, age: 68.39 ± 1.89, mean ± standard error) and 18 healthy subjects (male = 9, female = 9, age: 67.22 ± 1.98) were recruited for this study. All subjects underwent continuous actigraphic recording for three consecutive days, during which they also completed the Karolinska Sleepiness Scale (KSS) once an hour throughout wakefulness. Our results showed a higher level of sleepiness in the patients than the controls in the hours following awakening and in the early afternoon, specifically at 08:00 and 14:00 hours (08:00 hours, PD patients, KSS: 3 ± 0.3 versus healthy subjects, KSS: 2 ± 0.2, < 0.05; 14:00 hours, PD patients, KSS: 4.4 ± 0.5 versus healthy subjects, KSS: 3 ± 0.3, < 0.05). We suggest that some daytime hours are sensitive windows showing the first increase of sleepiness which will spread later to the whole daytime.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

A high level of sleepiness (excessive daytime sleepiness, EDS) is a very common feature in advanced stages of Parkinson's disease (PD; for a review see Arnulf, 2005), while the level is much lower in the early stages (Fabbrini et al., 2002). Studies performed in normal subjects using the Karolinska Sleepiness Scale (KSS) showed that sleepiness occurs before and after sleep and in the afternoon in both young and elderly healthy subjects (Zilli et al., 2008).

In PD patients subjective sleepiness levels are usually evaluated by means of the Epworth Sleepiness Scale (ESS; Arnulf, 2005), which does not give information about the subject's level of sleepiness at a particular time and subsequently does not allow observation of daily fluctuations in sleepiness levels (Shen et al., 2006).

The aim of this study was to evaluate the time–course of the sleepiness level during the wakefulness period in untreated patients with early-stage Parkinson's disease (de novo PD patients) to look for changes at specific times of the day. Rest–activity rhythm, diurnal sleep habits, fatigue and unintended sleep episodes were investigated.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Subjects

We enrolled 18 consecutive patients with early PD who had not been treated previously with anti-parkinsonian therapy (female = 9, male = 9; age = 68.39 ± 1.89, mean ± standard error). The clinical diagnosis of probable PD was based on Gelb et al.'s (1999) criteria. A single-photon emission computed tomography (SPECT) investigation with Ioflupane demonstrated a reduction of tracer uptake in the putamen in all de novo PD patients. The Hoehn and Yahr score was ≤2, the motor score of the unified Parkinson's Disease Rating Scale was ≤15/108 and the duration of the disease was ≤15 months.

The control group was composed of 18 age-matched (±3 years) healthy subjects (female = 9, male = 9; age = 67.22 ± 1.98, mean ± standard error).

Exclusion criteria for both groups was: (i) any sleep pathology or medications acting on the central nervous system which could influence sleepiness; (ii) cognitive and mnestic impairment (assessed by means of the Mini-Mental State Examination, score <25 and the Wechsler Memory Scale, score <90); (iii) sleep complaints (assessed through a pre-arranged interview focused on sleep habits and subjective sleep quality; Zilli et al., 2009); and (iv) night-time shift work.

Procedure

All subjects received a full description of the research and gave their written consent to participate in the study. Afterwards, each subject was required to complete an Italian version (Mecacci and Zani, 1983) of the Morningness–Eveningness Questionnaire (MEQ) and instructed to wear (on the non-preferred arm) a wrist activity monitor (Actiwatch-Plus, version 3.24; Cambridge Neurotechnology Ltd, Pampisford, Cambridge, UK) for three consecutive working days. Throughout this period they were also required: (i) to keep to their habitual sleep times; (ii) to complete a sleep log after the morning awakening; and (iii) to evaluate their sleepiness once an hour throughout wakefulness using the KSS, a 9-point Likert-type scale developed by Akerstedt and Gillberg (1990).

Data analysis

Rest–activity rhythm features were evaluated through analysis of motor activity using the Actogram software. Bedtime and getting-up time were obtained from the sleep log completed by the subjects themselves. Sleep times, time in bed, sleep latency, sleep duration and sleep efficiency as well as motor activity features during sleep were evaluated through analysis of the activity level by means of sleep analysis software. A 1-min epoch length was used for the analysis. Data were compared between the two groups using one-way analysis of variance (anova).

The mean sleepiness level (average of all KSS scores reported by each subject) and MEQ scores were compared between the two groups through one-way anova.

The time–course of sleepiness across the wakefulness period common to all subjects (i.e. time intervals from 07.00 to 22:00 hours) was assessed using anova for repeated measures, with ‘group’ as the between factor, ‘hour’ as the within factor, the level of sleepiness as the dependent variable and ‘age’ as the covariate. For overall anova significant effects, differences between subjects were evaluated using Student's t-test for unpaired samples. Differences within subjects were tested separately for each group applying an anova for repeated measures; effects related to the covariate were evaluated using Pearson's r coefficient.

Data on diurnal sleep habits, fatigue and unintended sleep episodes were obtained from the sleep habits interview (Zilli et al., 2009). Answer scores were compared between the two groups using the Mann–Whitney U-test. The significance level was set at < 0.05.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Diurnal preference evidenced by the MEQ did not differ between de novo PD patients and healthy subjects [MEQ score: 63.56 ± 1.46, mean ± standard error and 60.83 ± 1.91, respectively; F1,34 = 1.29, = not significant (NS)]. With regard to the rest–activity rhythm and nocturnal sleep features (Table 1), none of the variables varied as a function of the group. The only exception was the motor activity levels, which were lower in PD patients than healthy subjects during 24-h nocturnal sleep episodes and active periods during sleep episodes (Table 1).

Table 1. Actigraphic parameters in untreated patients with early Parkinson's disease [de novo Parkinson's disease (PD) patients] and in healthy control subjects (healthy controls)
 De novo PD patientsHealthy controls F 1,34 P
  1. Mean value ± standard error of each parameter and results of comparisons between the two groups are reported.

Rest–activity rhythm
Time period (hh:mm)23:57 ± 0:040:00 0:040.19NS
Time of activity peak (hh:mm)13:30 ± 0:1214:10 0:144.1NS
Average of motor activity (score)150.44 ± 17.29224.97 20.157.88< 0.01
Variance of motor activity (score)154.4 ± 13.55217.63 20.766.51< 0.05
Nocturnal sleep features
Bed time (hh:mm)22:58 ± 0:1023:20 ± 0:072.31NS
Get-up time (hh:mm)7:19 ± 0:097:15 ± 0:110.05NS
Sleep onset (hh:mm)23:16 ± 0:1323:39 ± 0:082.11NS
Awakening (hh:mm)7:15 ± 0:087:08 ± 0:110.22NS
Sleep latency (hh:mm)0:16 ± 0:020:19 ± 0:030.32NS
Sleep duration (hh:mm)8:01 ± 0:167:28 ± 0:073.34NS
Sleep efficiency (%)89.31 ± 1.1287.46 ± 1.321.14NS
Total average of motor activity (score)10.28 ± 1.7716.48 ± 1.726.33< 0.05
Average of motor activity in active periods (score)80.54 ± 9.07120.39 ± 7.0512.03< 0.001

The mean level of subjective sleepiness did not differ between the two groups (de novo PD patients: 3.27 ± 0.24, healthy control subjects: 3.09 ± 0.21, F1,34 = 0.34, = NS). Conversely, analysis of the time–course of sleepiness across the wakefulness period revealed that sleepiness levels varied with time of day in a different manner between the two groups (group × hour: F15,495 = 2.83, < 0.001; Fig. 1). Post-hoc analyses underscored that sleepiness peaked early in the morning, during the afternoon and late in the evening in both de novo PD patients (F15,255 = 6.29, < 0.001) and healthy subjects (F15,255 = 10.09, < 0.001); however, the morning and afternoon peaks were higher in de novo PD patients than in healthy subjects (08:00–08:59 hours: = 2.31, < 0.05; 14:00–14:59 hours: = 2.29, < 0.05; Fig. 1).

image

Figure 1. Time–course of subjective sleepiness level across wakefulness. Hourly mean value ± standard error is reported separately for de novo Parkinson's disease patients and healthy subjects. Asterisks indicate statistically significant differences between groups.

Download figure to PowerPoint

Diurnal sleep habits, fatigue and unintended sleep episodes did not differ between the two groups (Table 2).

Table 2. Diurnal sleep and fatigue in untreated patients with early Parkinson's disease [de novo Parkinson's disease (PD) patients] and in healthy control subjects (healthy controls)
 De novo PD patientsHealthy controls U 34 P
  1. Median value (1st–3rd quartiles) of each item (scores: 1–5) and results of comparisons between the two groups are reported.

Feeling of fatigue at the afternoon2 (2–3)2.5 (2–3)149NS
Frequency of need to sleep at the afternoon3 (2–4)2.5 (2–3)127NS
Frequency of afternoon nap4 (2.25–4)3 (2–4)141NS
Frequency of unintended sleep episodes1 (1–3)1.5 (1–2.75)147NS

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We evaluated sleepiness by a subjective measure (KSS) only; nevertheless, according to Horne and Burley (2010), KSS provides a good estimation of sleepiness and, in addition, Kaida et al. (2006) showed a close relationship with objective measures (electroencephalogram: EEG).

Our study shows that at an early stage of the disease in de novo PD patients, the mean level of daytime values of subjective sleepiness does not differ with respect to healthy controls. This is consistent with data from Fabbrini et al. (2002), who measured overall daytime sleepiness in de novo PD patients by means of the ESS. Sleepiness shows a circadian-like trend both in PD patients and in controls, as shown for elderly people (Zilli et al., 2008).

The PD patients rate themselves as sleepier than healthy subjects after awakening and even more during the afternoon, specifically at 08:00 and 14:00 hours.

Early morning post-sleep time and early afternoon are thus sensitive windows showing the increase of sleepiness in the early stages of PD before sleepiness spreads over the whole day when clinical conditions worsen, as shown in patients studied by many authors (Gjerstad et al., 2002; Ondo et al., 2001; Poryazova et al., 2009).

The increase of sleepiness during the afternoon in de novo PD patients is associated neither with increased fatigue nor with higher nap frequency. Also the occurrence of unintended sleep episodes is extremely rare. The increase of sleepiness limited to specific periods of the daytime in de novo PD patients is probably not so prominent as to lead to sleep episodes during the day.

Finally, actigraphic monitoring showed that motor activity levels are noticeably reduced in de novo PD patients with respect to healthy subjects, and hypokinesia is widespread across the whole 24-h period in waking as well as during sleep.

The rest–activity rhythm does not differ between de novo PD patients and healthy controls, indicating that at an early stage of the disease rest–activity rhythm and sleep features are basically maintained with respect to control subjects.

In conclusion, our study shows that the mean level of sleepiness is not increased with respect to healthy subjects, while higher levels of sleepiness are found in well-defined time windows, after awakening and in the early afternoon. It is plausible that the overall increase takes place as the severity of the illness increases.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This research was supported by a grant from ‘Ente Cassa di Risparmio’ of Florence.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References