TESTOSTERONE AND SLEEP IN OLDER MEN INTRODUCTION THE CIRCULATING TESTOSTERONE LEVELS OF HEALTHY MEN DECREASE ON AVERAGE BY 1-2% A YEAR STARTING IN THE THIRD TO FOURTH DECADE OF life.1-3 However, some octogenarians

THE CIRCULATING TESTOSTERONE LEVELS OF HEALTHY MEN DECREASE ON AVERAGE BY 1-2% A YEAR STARTING IN THE THIRD TO FOURTH DECADE OF life.1-3 However, some octogenarians continue to maintain androgen concentrations that are usually seen in young adults. Since the age-related decline in circulating androgen levels has been associated with metabolic disorders, decreased bone density, increased frailty and higher risk of falls with fractures, the factors responsible for the large variability of testosterone levels in older men are of considerable biological and clinical interest, but remain poorly understood.4-6 Chronic sleep changes are commonly seen in older adults. In the course of aging sleep architecture becomes increasingly fragmented by frequent arousals and awakenings, which leads to an overall reduction in sleep consolidation and in the total amount of sleep.7 Axelsson et al8 recently demonstrated that a period of either nighttime or daytime sleep in healthy young men is accompanied by an exponential rise in their testosterone concentrations, leading the authors to speculate that sleep plays an important role in determining the concentration of androgens in the circulation. Studies of total sleep deprivation in young male volunteers have reported a rapid decline in the circulating androgen levels of the participants.9,10 Additional data indicate that the experimental fragmentation of sleep in young adults can also disrupt the usual rise of blood testosterone levels during the night.11 Consistent with these observations, nighttime testosterone secretion is decreased in middle-aged men,12 a demographic group already affected by the onset of age-related decrements in sleep quantity and quality.13 Given the similarity between the changes in circulating androgen levels that occur spontaneously in the course of aging and those found in young adults under conditions of experimental sleep disruption or loss, this study tested the hypothesis that the large inter-individual variability in morning testosterone levels of healthy older men is related in part to objective differences in their sleep.