LAUSR

ABSTRACT It has been hypothesized that disruption of circadian rhythms affects human health. Shift work and sleep deprivation are thought to disrupt the normal light–dark cycle, although the disruption due to shiftwork may be dependent on sleep deprivation. Both conditions have been suggested to be associated with an increased risk of cardiometabolic disorders. Non-photic environmental factors, such as the timing of eating, are also thought to regulate circadian rhythm and thus, may have effects on health, but the evidence from human studies is scarce. Oxidative stress is a risk factor of cardiometabolic disorders. Some laboratory studies suggest an involvement of circadian clock genes in the regulation of the redox system. The present study aimed to examine the association of sleeping habits, nightshift work, and the timing of meals with urinary levels of 8-isoprostane, a marker of oxidative stress, and 6-sulfatoxymelatonin, the principal metabolite of melatonin. Study subjects were 542 women who had previously attended a breast cancer mass screening in a community in Japan. Information on bedtimes and wake-up times, history of nightshift work, and the timing of meals was obtained by a self-administered questionnaire. The 8-isoprostane and 6-sulfatoxymelatonin were measured using the first morning void of urine and expressed per mg of creatinine. The geometric mean of 8-isoprostane levels was 12.1% higher in women with ≤6 hours of sleep than that in those with >8 hours of sleep on weekdays, and longer sleep duration on weekdays was significantly associated with lower urinary levels of 8-isoprostane after controlling for covariates (p for trend = 0.04). Women who were currently working the nightshift had a 33.3% higher geometric mean of 8-isoprostane levels than those who were not working nightshift (p = 0.03). Urinary 6-sulfatoxymelatonin levels were unrelated to sleep habits or nightshift work. Women who ate breakfast at irregular times had a 19.8% higher geometric mean of 8-isoprostane levels than those who ate breakfast at a regular time or who did not eat (p = 0.02). Women who ate nighttime snacks at irregular times had a 16.2% higher geometric mean of 8-isoprostane levels than those who did not eat nighttime snacks or who ate nighttime snacks at a regular time (p = 0.003). Among women who ate dinner at a regular time, earlier times for dinner were associated with higher 8-isoprostane and 6-sulfatoxymelatonin levels (p values for trends were 0.01 and 0.02, respectively). However, the times of dinner and nighttime snack are overlapping, and the time of last meal of the day was not associated with 8-isoprostane and 6-sulfatoxymelatonin levels. The time of breakfast or lunch was not associated with these biomarkers among women who ate the meal at regular times. Disturbing the rhythmicity of daily life may be associated with oxidative stress.