Theme : Dreams

Sleep Stages Preceding Spontaneous Awakenings in the Elderly

Piero Salzarulo1, Igino Fagioli2, Pasquale Lombardo1, Sara Gori3, Carlo Gneri3, Roberta Chiaramonti1 and Luigi Murri3
1Department of Psychology, University of Florence, Firenze 50125 Italy, 2Department of Social Theory, History and Research, University of Trento, Trento 38100 Italy and 3Department of Neuroscience, University of Pisa, Pisa, 56100 Italy
Abstract
Spontaneous awakenings from sleep were studied in a group of 21 elderly subjects (mean age 69.29±3.02 years) free of neurological and somatic diseases. The prevalence of awakenings and the duration of waking bouts were analyzed with regard to the prior sleep state. The results showed an increased frequency of awakenings during Stage 2 NREM in the elderly, who wake out of Stage 2 NREM no less frequently than out of REM sleep. This trend is different from that observed in younger subjects (babies and young adults), where a clear prevalence of REM sleep awakenings has been reported. The duration of Stage 2 interrupted by awakening was shorter than Stage 2 followed by sleep. The duration of waking bouts did not differ according to the preceding sleep stage. It is suggested that the relative inability to sustain Stage 2 may be a mechanism which contributes to the difficulty of sleep maintenance in the elderly.

Current Claim: The increase of awakenings out of Stage 2 contributes to the difficulty of sleep maintenance in the elderly.

Activate the ShortNotes by clicking on this link. Your notes will be stored in this area and automatically retrieved upon your next visit.
Several investigations into the sleep characteristics of the elderly have reported an increase in both the number of awakenings (Agnew et al., 1967; Feinberg et al., 1967; Webb and Campbell, 1980; Webb, 1982; Bixler et al., 1984; Bliwise, 1993; Prinz, 1995) and their duration (Garma et al., 1981). Studies previously carried out on young adults (Langford et al., 1972; Schulz and Zulley, 1980; Weitzman et al., 1980) and on babies (Schulz et al., 1985) have shown that most spontaneous awakenings from sleep take place out of REM sleep, and even with an increased prevalence of REM awakenings in very young babies (Schulz et al., 1985). Data concerning the physiological state prior to awakenings in the elderly are not available at the present time.
The aim of the research described here was therefore to extend our previous investigations to older subjects in order to evaluate the effect of aging on the sleep state condition which precedes awakening. We sought to ascertain whether the increase in the number of awakenings is an exaggeration of the trend to awake from REM sleep previously observed, or whether stages other than REM are involved. The latter case may indicate a fragility of sleep stages which makes the maintenance of sleep in the elderly more difficult. We also investigated the relationship between the duration of intra-night waking episodes and the type of preceding sleep, which may influence the next sleep onset and increase the duration of the waking episode.

Activate the ShortNotes by clicking on this link. Your notes will be stored in this area and automatically retrieved upon your next visit.
Twenty-one elderly subjects, 14 males and 7 females, aged from 65 to 75 (mean age±s.d.: 69.29±3.02 yr.) were selected for the experiment, after they had given their informed consent. The subjects reported on their sleeping habits in a structured interview before the experiment; none of them took naps. All subjects met the criteria for good physical health as verified by an extensive screening process that included medical history and clinical examination. They reported no symptoms of respiratory disorders, no current physical illness requiring treatment, no evidence of current or past psychiatric illness, no cardiocirculatory problems, no history of drug or alcohol abuse, no insomnia or daytime sleepiness. The subjects were instructed to maintain their habitual activity patterns and not to use drugs. They were invited to spend a single night in the Sleep Laboratory (University of Pisa), where their sleep was recorded on a polygraph Era 14-21 using the customary technique (EEG [8 channels: Fp1/Fp2/C3/C4/T3/T4/O1/O2], EOG, EMG, ECG, respiration [thoracic; nasal and oral air flow], O2 saturation). The records were scored using Rechtschaffen and Kales's (1968) rules every 20 seconds; however, the minimum length of each state (including waking) was set at two minutes, as it had been for infants and young adults in previous investigations (Bes et al., 1991). Slow wave sleep (Stages 3 and 4) was scored using 40 µV as an amplitude criterion, as suggested by Webb and Dreblow (1982). According to the criteria previously used (Schulz et al., 1985), awakenings were taken into account when they lasted 2 minutes or more.

Activate the ShortNotes by clicking on this link. Your notes will be stored in this area and automatically retrieved upon your next visit.

Figure 1

Figure 2

The mean total sleep time (TST) was 318.44±92.79 min. Of 103 spontaneous awakenings (mean±s.d. for subject per night : 4.90±2.43), 20 (0.95±1.16) were out of Stage 1 NREM, 55 (2.62±2.06) out of Stage 2 NREM, 17 (0.81±0.75) out of SWS (Stages 3 and 4 NREM), and 11 (0.52±0.60) were out of REM sleep.
Nineteen out of 21 subjects showed 82 intra sleep awakenings (103 minus 21 final awakenings, one for each subject; mean±s.d. 3.90±2.43), whereas two subjects displayed an uninterrupted nocturnal sleep period. Of the 82 intra sleep awakenings, 17 (0.81±1.03) were out of Stage 1, 42 (2.00±2.00) out of Stage 2, 13 (0.62±0.67) out of SWS, and 10 (0.53±0.51) out of REM sleep.

The distribution of the final awakenings according to the sleep stage preceding them was similar for intra-sleep awakenings (i.e., those followed by a re-onset of sleep) and final awakenings (2=1,684, d.o.f. = 3, p=0.640, n.s.: Table 1).

None of the awakenings was immediately preceded by, or associated with, body movements or apneas, except in one case, where the awakening was preceded by an obstructive apnea lasting 7 seconds.

Prevalence of awakenings according to the number of sleep stages and their duration
To determine whether the prevalence of the awakenings from sleep was related to the number of the different sleep stages and to their amount, two different tests were performed on REM sleep, Stage 1 NREM, Stage 2 NREM and SWS.

First, the number of awakenings from a given sleep stage (REM, Stage 1 NREM, Stage 2 NREM, and SWS) was compared with the total number of episodes of this stage during the recording time. A quotient of these two quantities was computed for each stage and for each subject. These data were treated using Friedman two-way analysis of variance by ranks (Systat 5.0 for Windows statistical package, Systat Inc., 1990-92, Evanston, Ill, USA), which showed no significant difference among the probabilities of awakening from each of the sleep stages (Friedman test statistic=4.071; Kendall coefficient of concordance=0.065; p=0.254, n.s. assuming 2 distribution with d.o.f.=3). The values for each stage are reported in Table 2. To be stressed are the very high inter-individual differences among REM sleep values shown by both the largest standard deviation and the fact that the median attains the lowest value (0.000), and the mean the highest one (0.244). We further explored the hypothesis of a different prevalence of awakenings according to number of stages by using the Wilcoxon signed ranks test on all the possible pairings of sleep stages. Only one comparison attained statistical significance: the quotient for the Stage 2 NREM was significantly higher than that for Stage 1 NREM (Z=2.110; p<0.035). All the other comparisons failed to reach statistical significance.

Second, two kinds of analysis were performed to determine whether awakenings were distributed according to the duration of the different sleep stages:

a) The number of awakenings from a given sleep stage (REM, Stage 1 NREM, Stage 2 NREM, and SWS) was compared with the total duration (in minutes) of this stage during the recording time, and a quotient of these two quantities was computed for each stage and for each subject. Once again, the Friedman two-way analysis of variance by ranks did not show significant differences among the probabilities of waking from each of the sleep stages (Friedman test statistic=6.214; Kendall coefficient of concordance=0.099; p=0.102, n.s. assuming 2 distribution with d.o.f.=3). The values for each stage are reported in Table 3. The Wilcoxon signed ranks test comparing all the possible pairings of sleep stages showed a significantly lower quotient for SWS with respect to both Stage 1 NREM (Z=-2.047; p=0.041) and Stage 2 NREM (Z=-2.295; p=0.022). All the other comparisons failed to reach statistical significance.

b) The frequency of spontaneous awakenings from each sleep stage was then compared with the percentage of this stage on total sleep time. In 9 of the 21 subjects the relative number of awakenings out of Stage 1 NREM was higher than the relative proportion of this stage; in 11 of the 21 subjects the relative number of awakenings out of Stage 2 NREM was higher than the relative proportion of this stage; in 4 of the 21 subjects the relative number of awakenings out of SWS was higher than the relative proportion of this stage; in 7 of the 21 subjects the relative number of awakenings out of REM sleep was higher than the relative proportion of this stage. A binomial test with =0.5 (Siegel, 1956) showed that, for both Stage 1 NREM and Stage 2 NREM, there was no significant difference between the expected and the observed number of awakenings, whereas the observed number was lower than expected for SWS (p=0.01); the trend to awake with a lower frequency than expected during REM sleep failed to reach statistical significance (p=0.095).

We then addressed the question of in which part of the stages the awakenings were located. To do so, we compared the frequency distributions of the duration of the phases which were respectively interrupted and not interrupted by awakenings. Very similar values were found for Stage 1 NREM (interrupted phases: median: 3.33; mean±s.d.: 4.13±2.53; phases followed by sleep: median: 2.33; mean±s.d.: 3.46±3.59); SWS (interrupted phases: median: 8.67; mean±s.d.: 12.55±10.60; phases followed by sleep: median: 9.67; mean±s.d.: 13.52±12.33); REM sleep (interrupted phases: median: 13.50; mean±s.d.: 16.10±9.84; phases followed by sleep: median: 15.33; mean±s.d.: 17.26±10.88; see Fig. 1). By contrast, the duration of the Stage 2 episodes interrupted by awakening (median: 4.67; mean±s.d.: 6.87±5.78) was significantly shorter than that of the non interrupted episodes (median: 7.33; mean±s.d.: 9.95±8.92) (U Mann-Whitney: 2=7.247, d.o.f.=1, p=.007) (Fig. 2).

Duration of waking according to the preceding sleep stage
The duration of the 82 intra sleep waking periods was not related to the type of stage preceding it (Median test: 2=0.1823; d.o.f.=3; n.s. Table 4).

Activate the ShortNotes by clicking on this link. Your notes will be stored in this area and automatically retrieved upon your next visit.
We begin with some methodological remarks. It was decided to set the minimum duration interval for awakening computation at 2 min. This clearly corresponded to behavioral awakenings, which differ from the "arousals" coded by other investigators (Mathur and Douglas, 1995). This criterion is quite similar to the one used by Garma et al. (1981) and it is exactly the same as the criterion used by Schulz et al. (1985) with groups of younger subjects, against whom these data were to be compared (given that, as mentioned above, these data are part of a cross-life-span investigation). As in previous similar investigations (Mathur and Douglas, 1995), we used a single night of recording for comparative purposes. This decision was also determined by the aim of the research, which was not to evaluate the global number of awakenings in the elderly, but to ascertain their relationship with the preceding sleep stages. Among the elderly, moreover, the values of awakenings during the second night are not always different from those of the first night (Garma and Bouard, 1981; Wauquier and Van Sweeden, 1992). Finally, the number of awakenings is quite close to the number found by Feinberg et al. (1967) and by Garma et al. (1981), who used similar criteria. It should be pointed out that visual analysis does not take account of the heterogeneity of Stage 2 periods across nocturnal sleep.
Behavioral observation did not reveal any association of awakenings with gross body movements; however, the presence of leg movements cannot be excluded. To ascertain the link with such phenomena, new investigations with ad hoc electrodes on the anterior tibialis muscle should be performed.

If we compare the number of awakenings and the number of periods of each sleep stage, the probability of waking from Stage 2 NREM was higher than from Stage 1 NREM; if the duration of the sleep stages is taken into account, the probability of waking from SWS was lower with respect to that of both Stage 1 and Stage 2 NREM, while the probability of waking from Stage 2 NREM was similar to that from REM sleep. In addition, the probability of waking from both REM sleep and Stage 2 NREM did not differ from that expected by chance, whereas the probability of waking from SWS sleep was significantly lower than that expected by chance. All these results shed light on awakening out of Stage 2 NREM in the elderly: they wake out of Stage 2 NREM with a frequency no lower than their waking from REM sleep. In this they differ from younger subjects, who show a clear prevalence of awakenings out of REM sleep (Schulz et al., 1985; Schulz and Zulley, 1980; Weitzman et al., 1980)

The results also show that awakenings interrupt Stage 2 precociously, before its "habitual" ending, while in another age group, which has been previously studied (Schulz et al., 1985), REM is the stage which is precociously interrupted. In addition, the duration of Stage 2 NREM interrupted by awakenings in the elderly is similar to the duration of REM interrupted by awakenings in younger subjects. By contrast, the durations of REM bouts prior to awakening in the elderly are similar to those followed by sleep. This suggests that awakenings in REM sleep in elderly subjects are not "precocious" interruptions of REM, because they appear at the possible natural end of the NREM-REM cycle (Langford et al., 1972; Schulz et al., 1985); by contrast, Stage 2 NREM in the elderly behaves like REM sleep in younger subjects, since awakenings occur before the end of its habitual duration. This is evidence for the fragility of Stage 2 NREM in the elderly; a similar concept concerning Stage 2 fragility has been suggested by Lorrain et al. (1996) in a study on evoked potentials during sleep. Automatic analysis able to detect subtle heterogeneities in EEG activity should be performed in the future, in order to show what kind of activity preceding the awakening is disturbed, if any.

On the other hand, awakening out of Stage 2 NREM does not increase the duration of the following period of wakefulness. This suggests that the mechanisms which cause the awakening and those which control the duration of the waking episode are different. This possibility has been raised several times, in particular by Garma and Bouard (1981) when discussing the physiopathology of insomnia and the role of aging. According to the theory of the regulation of the sleep-wake cycle proposed by Valatx (1995), it could be interpreted as impaired activity by the anti-wake system (which should inhibit the waking network, giving rise to sleep onset), which is possibly responsible for the absence of sleep re-onset (after a nocturnal awakening) in the elderly.

It is worth pointing out that the relative inability to sustain Stage 2 NREM in the elderly is paralleled by a diminution in the number of NREM-REM cycles (Salzarulo et al., 1997; Lombardo et al., 1998) leading to the disruption of the sleep episode. In this case, sleep interruption would not be gated exclusively by the NREM-REM cycle as in younger subjects (Schulz et al., 1985). In the future, investigation will be required into whether the pattern found in elderly subjects is specific to aged persons or whether it also occurs in younger subjects suffering from multiple night awakenings.

1. Agnew HW, Webb WB, Williams RL. Sleep patterns in late middle age males: An EEG study. Electroenceph Clin Neurophysiol, 1967; 23: 168-71.
2. Bes F, Schulz H, Navelet Y, Salzarulo P. The distribution of slow-wave sleep across the night: a comparison for infants, children and adults. Sleep 1991; 14: 5-12.

3. Bixler EO, Kales A, Jacoby JA, Soldatos CR, Vela-Bueno A. Nocturnal sleep and wakefulness: effects of age and sex in normal sleeper. Int J Neurosci 1984; 23: 33-42.

4. Bliwise DL. Sleep in normal aging and dementia. Sleep 1993; 16: 40-81.

5. Feinberg I, Koresko RL, Heller N. EEG Sleep patterns as a function of normal and pathological aging in man. J Psych Res, 1967; 5: 107-44.

6. Garma L, Bouard G. Les éveils nocturnes chez le sujet insomniaque en fonction de l'age. Rev Neurol 1981; 137: 677-92.

7. Garma L, Bouard G, Benoit O. Age et insomnie: parts respectives du nombre et de la durée des éveils. Rev EEG Neurophysiol 1981; 11: 96-101.

8. Langford GW, Meddis R, Pearson AJD. Spontaneous arousals from sleep in human subjects. Psychon Sci 1972; 28: 228-30.

9. Lombardo P, Gori S, Gneri C, Rossi L, Fagioli I, Murri L, Salzarulo P. Disorganisation of NREM-REM cycles in elderly subjects. J. Sleep Res. 1998; 7 Suppl.2: 157.

10. Lorrain D, Campagna F, Bélisle D, Campbell KB. Responses to auditory stimulation during sleep in aged individuals. J Sleep Res, 1996; 5 Suppl 1: 127.

11. Mathur R, Douglas NJ. Frequency of EEG arousals from nocturnal sleep in normal subjects. Sleep 1995; 18: 330-3.

12. Prinz P. Sleep and sleep disorders in older adults. Clin Neurophysiol 1995; 12: 139-46.

13. Rechtschaffen A, Kales A, eds. A Manual of Standarized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Los Angeles: Brain Information Service/Brain Research Institute, 1968.

14. Salzarulo P, Formicola G, Lombardo P, Gori S, Rossi L, Murri L, Cipolli C. Functional uncertainty, aging and memory processes during sleep. Acta Neurol Belg 1997; 97: 118-22.

15. Schulz H, Massetani R, Fagioli I, Salzarulo P. Spontaneous awakening from sleep in infants. Electroenceph Clin Neurophysiol 1985; 61: 267-71.

16. Schulz H, Zulley J. The position of final awakening within the ultradian REM/NREM sleep cycle. Sleep Res 1980; 9: 124.

17. Siegel S. Nonparametric Statistics for the behavioral sciences. New York: McGraw-Hill, 1956.

18. Valatx JL. Regulation of the sleep-wake cycle: a new theory. Sleep Res 1995; 24A: 7.

19. Wauquier A, Van Sweeden B. Aging of core and optional sleep. Biol Psychiat 1992; 31: 866-80.

20. Webb WB. Sleep in older persons: Sleep of 50 to 60 years old men and women. Gerontol 1982; 5: 581-6.

21. Webb WB, Campbell SS. Awakenings and the return to sleep in an older population. Sleep 1980; 3: 41-6.

22. Webb WB, Dreblow LM. A modified method for scoring slow wave sleep of older subjects. Sleep 1982; 5: 195-9.

23. Weitzman ED, Czeisler CA, Zimmerman JC, Joseph R. The timing of REM sleep and its relation to spontaneous awakening during temporal isolation in man. Sleep Res, 1980; 9: 280.

This work has been partially supported by C.N.R. grant No. 95.03925.CT08. We gratefully acknowledge Dr. Adrian Belton for revising the English manuscript.

Original address of this text :

http://www.sro.org/bin/article.dll?Paper&1616&0&0

Please copy this address to the address bar of your internet browser and press the "enter" key.
(We prefer not to put actual links because often page locations change and then our log files get cluttered with error messages
- if the address does not work try to find it from the homepage of the website in question).