Rapid eye movement density is reduced in the normal elderly

While there is general agreement on the age changes in non-rapid eye movement sleep, there is conflicting evidence on whether eye movement density (EMD) in rapid eye movement sleep is affected by aging. We therefore performed computer measurement of EMD in young and elderly normal subjects.

Sleep electroencephalogram and electrooculogram were recorded in each subject on 4 nonconsecutive baseline nights. Eye movement density in the elderly subjects was compared to that in young adults.

A sleep research laboratory with 4 separate bedrooms.

Not applicable.

Subjects were 19 young normal adults and 19 elderly normal adults.

Digitized electrooculograms were analyzed with the extensively validated zero-cross period-amplitude module of PASS PLUS software. The EMD was measured as 0.3 to 2 Hz integrated amplitude per 20-second stage of rapid eye movement sleep. Eye-movement incidence was the number of half waves. Eye-movement amplitude was the sum of peak-trough excursions (curve length) in the average half wave. We also counted visually the number of 2-second epochs with eye movements for 1 baseline night in both groups. The EMD in the elderly subjects was substantially and significantly lower than in the young subjects. Visual scoring of EMD on 1 baseline night confirmed the statistically significant difference between age groups. Period-amplitude analysis revealed that a lower eye-movement incidence rather than reduced amplitude caused the lower EMD in the elderly. The EMD was significantly correlated within subjects across the nonconsecutive baseline nights in both groups; in both, subjects' EMD average across 2 nights provided a correlation greater than .90 with the 4-night mean.

The incidence of eye movements during rapid eye movement sleep is substantially reduced in the elderly. We hypothesize that this reduction is due to degenerative (aging) rather than developmental brain changes. The correlation analysis indicates that EMD is a reasonably stable individual trait in both young and elderly adults. These results encourage normative studies of EMD over a wider age span and continued exploration of the relation of EMD to cognitive function in the elderly.