SU29 - GENETIC DISSECTION OF SLEEP DISTURBANCE IN BIPOLAR DISORDER

Abstract

Background Bipolar Disorder (BD) is often associated with sleep disturbances such as insomnia or excessive sleep, even when mood is stable. Sleep loss may trigger the first episode of BD or subsequent episodes. Sleep disturbances have attracted much attention in the field of BD research, but also in clinical practice, demonstrated in the recent development of therapies manipulating sleep-wake rhythms. Sleep has been proposed as a biomarker of BD and its disturbance has been implicated in the biology of mood disorders. However, the mechanisms that link mood and sleep are not fully understood. For example, sleep loss may be a cause for BD, but the converse is also true. Alternatively, sleep disorders and BD may co-occur because they have risk factors in common, with no strong causal relationship between them. Genomics has already provided important insights into the relationship between psychiatric disorders, and may help disentangle the intricate relationship between sleep and mood. Methods This research project will take advantage of an existing dataset, the Bipolar Disorder Research Network (BDRN) collection of clinical and genetic data (available for over 5700 participants with BD) to evaluate: (1) the methods researchers and clinicians use to measure sleep in BD, (2) whether sleep characteristics can help identify biologically valid subgroups of individuals with BD, using genomic approaches.We will evaluate a range of sleep measures, from low intensity, easy to fill, one-off questionnaires to more intensive daily sleep and mood diaries. We will also include objective measures gathered through actigraphs. Actigraphs are devices worn on the wrist that record movements and can be used to estimate sleep characteristics. For each person with BD, we will calculate scores for their genetic risk for sleep disturbances based on the findings of large genetic studies of sleep in the UK general population. We will then compare the scores in individuals with BD with and without sleep disturbances to test the underpinning biological validity of measured sleep disturbance in BD. Results Participants (N=5,728) have been (and continue to be) recruited across the United Kingdom. To date, 5,549 participants have been genotyped using the Affymetrix (n=1,868), PsychChip (n=1,104) and Omni-Express (n=2,577) chips. All are at least 18 years old, provide written informed consent, and have a DSM-IV diagnosis of BD (I, II, not otherwise specified) or schizoaffective disorder, bipolar subtype. To date, over 1,500 individuals with BD have completed questionnaires on circadian preference and sleep disorders. Over 100 individuals with BD have completed 6-weeks of high-intensity sleep and mood monitoring. Preliminary analyses on 36 individuals with BD have shown moderate agreement (0.40 | rs | 0.59) between the low-intensity measures (e.g. sleep questions of the Beck Depression Inventory) and high-intensity measures such as actigraphy parameters. Discussion The proposed research will be the first to systematically evaluate a range of sleep measures in the largest study of individuals with BD in the world to date. The results will inform the choice and use of sleep measures in large-scale BD studies. The genetic analyses planned are well powered to provide insights into the mood-sleep relationship, and might pave the way to personalized diagnosis and treatment.