The present study is part of a more extensive investigation dedicated to the study and treatment of age-dependent changes/disturbances in the circadian system in humans. It was performed in the Tyumen Elderly Veteran House and included 97 subjects of both genders, ranging from 63 to 91 yrs of age. They lived a self-chosen sleep-wake regimen to suit their personal convenience. The experiment lasted 3 wks. After 1 control week, part of the group (n=63) received 1.5 mg melatonin (Melaxen) daily at 22:30 h for 2 wks. The other 34 subjects were given placebo. Axillary temperature was measured using calibrated mercury thermometers at 03:00, 08:00, 11:00, 14:00, 17:00, and 23:00 h each of the first and third week. Specially trained personnel took the measurements, avoiding disturbing the sleep of the subjects. To evaluate age-dependent changes, data obtained under similar conditions on 58 young adults (both genders, 17 to 39 yrs of age) were used. Rhythm characteristics were estimated by means of cosinor analyses, and intra- and inter-individual variability by analysis of variance (ANOVA). In both age groups, the body temperature underwent daily changes. The MESOR (36.38+/-0.19 degrees C vs. 36.17+/-0.21 degrees C) and circadian amplitude (0.33+/-0.01 degrees C vs. 0.26+/-0.01 degrees C) were slightly decreased in the elderly compared to the young adult subjects (p<0.001). The mean circadian acrophase was similar in both age groups (17.19+/-1.66 vs. 16.93+/-3.08 h). However, the inter-individual differences were higher in the older group, with individual values varying between 10:00 and 23:00 h. It was mainly this phase variability that caused a decrease in the inter-daily rhythm stability and lower group amplitude. With melatonin treatment, the MESOR was lower by 0.1 degrees C and the amplitude increased to 0.34+/-0.01 degrees C, a similar value to that found in young adults. This was probably due to the increase of the inter-daily rhythm stability. The mean acrophase did not change (16.93 vs. 16.75 h), although the inter-individual variability decreased considerably. The corresponding standard deviations (SD) of the group acrophases were 3.08 and 1.51 h (p<0.01). A highly significant correlation between the acrophase before treatment and the phase change under melatonin treatment indicates that this is due to a synchronizing effect of melatonin. Apart from the difference in MESOR, the body temperature rhythm in the elderly subjects undergoing melatonin treatment was not significantly different from that of young adults. The data clearly show that age-dependent changes mainly concern rhythm stability and synchronization with the 24 h day. A single daily melatonin dose stabilizes/synchronizes the body temperature rhythm, most probably via hypothermic and sleep-improving effects.
