Abstract
Age-related changes in the intrinsic circadian period ([.tau]) have been hypothesized to account for sleep symptoms in the elderly such as early morning awakening. The authors sought to determine whether the aging process produced quantifiable differences in the [.tau] of totally blind men who had free-running circadian rhythms. The melatonin onset was used as the indicator of circadian phase. Melatonin rhythms had been characterized about a decade previously when the participants were 38 ± 6 (SD) years old. Both previous and current assessments of [.tau] were derived from at least 3 serial measurements of the 24-h melatonin profile from which the melatonin onset was determined. All 6 participants exhibited a longer [.tau] in the 2nd assessment (mean increase ± SD of 0.13 ± 0.08 h; p < 0.01). Four participants exhibited differences in [.tau] with nonoverlapping 95% confidence intervals. The results do not support the commonly held view that [.tau] shortens during human aging. On the contrary, [.tau] appears to slightly, but significantly, lengthen during at least 1 decade in midlife.
