Sleep spindle characteristics in healthy subjects of different age groups

Abstract

Objectives: Few studies have quantified the various characteristics of sleep spindles (SS) in groups of normal human subjects. The aim of the present study was to look at the effects of age on the number, density, duration, intra-spindle frequency, and periodicity of SS during stage 2 sleep in normal subjects of different age groups.

Methods: Thirty-six healthy subjects participated in the study. They were divided into 6 age groups: 10–19, 20–29, 30–39, 40–49, 50–59 and 60–69 years.

Results: The results show that there is a progressive decrease in SS number, density, duration and a progressive increase in intra-spindle frequency with age. These changes occur mainly in the first 4 decades, except for SS number and density, for which the changes seem to continue until the sixth decade. The present study also reveals a clear periodicity of SS in human sleep. SS occur every 3–6 s, and the modal value of inter-spindle intervals increases with aging.

Conclusions: The progressive decrease in the number of SS and slow-wave sleep time with age suggests that SS are part of sleep promoting mechanisms. The negative correlation found between SS density and sleep efficiency in the present study is congruent with the sleep maintenance role of SS.

Introduction

Sleep spindles (SS) were first described by Hans Berger in 1933 (Berger, 1933), but the term was coined by Loomis et al. in 1935 (Loomis et al., 1935). SS are a hallmark of stage 2 sleep, and have been used in the standard scoring method of Rechtschaffen and Kales (1968). Few studies, however, have quantified the various characteristics of SS in groups of normal human subjects. The SS density has been reported to be drastically decreased in Parkinson's disease (Puca et al., 1973, Mouret, 1975), Alzheimer's disease (Montplaisir et al., 1995, Prinz et al., 1982, Reynolds et al., 1985), Creutzfeldt–Jacob disease (Donnet et al., 1992), progressive supranuclear palsy (Montplaisir et al., 1997) and fatal familial insomnia (Sforza et al., 1995). Therefore, normative data are needed to differentiate normal aging from pathological conditions.

Since scoring SS across the night is a time-consuming task, most of these studies have been carried out using computer-based SS detection (Declerck et al., 1986, Gaillard and Blois, 1981, Principe and Smith, 1982). This approach encounters several difficulties due to the great variability of spindle morphology and of intra-spindle frequency. More recent studies have used pattern-recognition techniques (Dijk et al., 1993, Dijk, 1995) which are more powerful, but which still require further validation with visual detection methods.

Age has been shown to produce a decrease in the number and density (number/h of stage 2 sleep) of SS (Feinberg et al., 1967, Guazzelli et al., 1986, Kubicki et al., 1989, Nicolas et al., 1997, Wei et al., 1999), but the temporal pattern of the aging process is unknown since most of these studies have compared groups of young and elderly subjects. Most of these studies raise methodological concerns. Principe and Smith (1982) restricted their analysis to the first half of the night, while the density of SS increases from the beginning to the end of the night (Guazzelli et al., 1986). Therefore, their results may not be representative of the entire SS population. Other authors have included spindles detected in stages 2, 3 and 4 sleep (Gaillard and Blois, 1981, Declerck et al., 1986). It is virtually impossible to detect all SS in the presence of large amplitude delta waves, and even with computerized methods of detection, it is often impossible to correlate SS detected in stages 3 and 4 sleep with the presence of SS on visual examination. Two studies have looked at the intra-spindle frequency in relation to age, and found a decrease and a increase, respectively, of that inner frequency with advancing age (Feinberg et al., 1967, Principe and Smith, 1982). More recently, Evans and Richardson (1995) showed, during nap recording, the presence of a periodicity of approximately one every 3–5 s for SS. This finding was confirmed by Achermann and Borbély (1997) during nocturnal sleep, but no study was carried out with respect to the effects of age.

The aim of the present study was to look at the effects of age on the number, density, duration, intra-spindle frequency, and periodicity of SS during stage 2 sleep in normal subjects of different age groups. SS were counted for the entire night; they were visually selected since the aim of the study was to establish a standard database for further comparisons with automatic detection techniques and with pathological conditions.