The research described in the present thesis focuses on frequent problems in dementia which all put a high burden on the quality of life of both, the demented patients themselves and their caregivers. These problems not only consist of cognitive disabilities, but also of disturbances in behaviour, mood and the circadian regulation of sleep, hormones and temperature. The general hypothesis studied in this thesis is that deterioration of the suprachiasmatic nucleus (SCN) is for a large part responsible for disturbed functionality of the circadian timing system (CTS), and contributes to disturbances in sleep, mood, behaviour and cognition in demented elderly. Moreover, we hypothesize that reactivation of the biological clock by the proper stimuli will improve the functionality of the biological clock, or prevent its deterioration. According to this hypothesis, enhanced exposure to the stimuli to which the biological clock is most sensitive, i.e. bright light and melatonin, should be a rational method to ameliorate sleep disturbances and to improve mood, behavioural and cognitive disturbances, at least as far as they are under the influence of the CTS. Chapter 2 reviews how age-related degeneration in the core of the CTS, the suprachiasmatic nucleus, and the resulting weak or altered output of the system, have negative implications for health, sleep, mood, cognitive performance and overall well-being of healthy and demented elderly. Given these negative consequences, it is of importance to investigate which mechanisms underlie the degeneration of the CTS in order to be able to develop rational treatment strategies. We demonstrate how aging can be paralleled by a diminishing input of environmental light to the CTS. This condition of chronically decreased activation could contribute to neuronal shrinkage and consequently a poor expression of rhythms, paraphrased as ‘use it or lose it’. We review the studies that generally confirm that bright light contributes to the expression of rhythms in core body temperature, melatonin and sleep-wakefulness in healthy and demented elderly. Chapter 3 reviews age-related changes in the major internal stimulus to the CTS, the pineal hormone melatonin. The circadian rhythm in melatonin production is, in a feedback system, regulated by the SCN itself, resulting in low daytime circulating levels of melatonin and an increase of circulating melatonin after darkness onset. The relation between the age-related changes in melatonin levels and sleep is discussed in this chapter. Based on the hypothesis that there is a relationship between the increased prevalence of sleep-disturbances and decreased night-time melatonin levels in elderly, several studies have been performed to investigate the effect of exogenous melatonin supplementation on sleep in elderly and demented subjects. An overview of these findings is presented and the various results are discussed. We concluded that there is the need for larger studies before widespread use of melatonin in sleep disorders can be advocated. These studies should have the statistical power to elucidate which subgroups of patients can be expected to show a positive effect of melatonin and which cannot. Chapter 4 focuses on the functional neurobiology of the central pacemaker in major depression, an other condition in which the CTS is disturbed. Data of a post-mortem study are presented, comparing vasopressin (AVP) immunoreactivity and AVP m-RNA in the SCN between depressed and non-depressed subjects. We found that in depressed subjects, the number of AVP-IR neurons in the SCN was more than one and one-half times higher than in controls, while the total masked area of silver grains, as an estimate of the amount of AVP-mRNA, was about one half that of controls. These findings suggest that in depressed patients the transport of AVP is even more reduced than 177 SUMMARY its synthesis in the SCN, resulting in an impaired functional ability of the CTS. Chapter 5 presents a study investigating the relation of two types of measurement of sleep disturbances; i.e. actigraphy and a questionnaire, the Circadian Sleep Inventory for Normal and Pathological States (CSINAPS). We found good correlations between the questionnaire items about habitual timing of sleep and wakefulness and their actigraphic counterparts. However, caregivers overestimated the actual time between sleep onset and offset by more than 1.5 hours. The assessment of sleep and wake disturbances in demented elderly by the CSINAPS can be best performed by the parallel use of actigraphy. Though, if sleep-disordered breathing and leg movements are a focus of interest, additional assessments are needed. Chapter 6 describes the association between actigraphic estimates of the sleep-wake rhythm and a range of functional domains that contribute to wellbeing in demented elderly patients. Cognitive, functional, behavioural and emotional states showed moderately strong correlations with multiple rhythm variables. Partial correlations indicated that this could not only be attributed to a uniform worsening with advancing cognitive decline. Stepwise regressions indicated three most distinctive rhythm variables: (1) the interdaily stability of the 24-hour rhythm was most strongly, negatively, related to cognitive decline and depression; (2) the median level of daytime activity was most strongly, negatively, related to impairments of function, of activities of daily living, and of social interaction; (3) nocturnal restlessness was secondarily, positively, related to impairments of function and of social interaction. This raises the possibility that treatments that enhance daytime activity and the stability of the rest-activity rhythm may improve these components of well-being. Chapter 7 presents the results of longterm (up to 3.5 years) daily supplementation of the circadian synchronisers light (± 1000 lux) and/or melatonin (2.5 mg) on cognition and actigraphic estimates of sleep-wake rhythms of 189 mostly demented elderly residents of group facilities in 12 nursing homes. This first study on longterm stimulation of the human CTS showed that prolonged administration of combined melatonin and whole-day bright light treatment in mostly demented elderly residents of group facilities, induces effects that in combination can be regarded as clinically relevant. A further important finding is that the improvement of the sleep-wake rhythm contributed to attenuation of cognitive decline, with an effect size comparable to the effect that acetylcholinesterase inhibitors have been reported to bring about in a younger, less affected and more homogeneous group of Alzheimer patients. Chapter 8 reports on the non-cognitive outcomes of long-term application of bright light and/or melatonin in mostly demented elderly residents of group facilities. Light ameliorated depressive symptoms by 19% and limitations of activities of daily living by 11%. Melatonin ameliorated the severity and distress of psychiatric symptoms in subjects about to drop out of the study due to nursing home placement or death by 26% per year and 45% per year respectively. However, melatonin adversely influenced the ratings for positive and negative observed affect, and for withdrawn behaviour. These adverse effects were partly counteracted if melatonin was given in combination with light. The combined treatment furthermore attenuated aggressive behaviour by 9%. Chapter 9 discusses the findings of the studies described in the thesis and the answered and unanswered research questions. The primary question of the present thesis was whether chronic treatment with light and/or melatonin 178 SUMMARY remains effective in the long term. We can finally conclude that this is the case and that most effects even increase with the duration of treatment. These findings strongly suggest that continued stimulation of the biological clock of demented elderly is needed to keep it ticking loud enough to be heard by the brain systems involved in the regulation of sleep, mood and cognition. Future research should address the question whether a lower dose of melatonin would be as effective on the long-term, but without the adverse effect on mood found with a daily dose of 2.5 mg. For the time being, however, the best long-term effect can be obtained with the combination of 1000 lux light and 2.5 mg melatonin per day.