Objectives

To elucidate the unique sleep and waking characteristics in progressive supranuclear palsy (PSP), a neurodegenerative disease associated with motor deficits and dementia that largely affects the brainstem and thalamic regions.

Methods

A total of 20 PSP and 16 healthy older adult controls participated in this study. The participants underwent an overnight polysomnography and multiple sleep latency test (MSLT) the following day. Prior to the MSLT last trial, they were asked to complete the Stanford Sleepiness Scale. Data were assessed for measures of latency to sleep onset, sleep duration, waking, and sleep staging during the night. Mean sleep latency, a measure of daytime sleepiness, sleep onset rapid eye movement (REM) periods, and microsleeps were studied with the MSLT. Spectral analysis of wake electroencephalogram (EEG) was performed for 30-second periods at the start of each MSLT trial.

Results

PSP took significantly longer time to fall asleep (p < .001), slept less during the night (p ≤ .001), and had more wake after sleep onset than controls (p ≤ .001). PSP had less N2 sleep (p < .05) and N3 sleep (p < .05), and REM sleep (p < .001) than controls. During the MSLT, PSP took significantly longer to fall asleep (p < .001), did not have microsleeps when they remained awake throughout the assessment periods, but were subjectively sleepier than controls (p < .05). Gamma power was increased during wake EEG in PSP (p < .01).

Conclusions

Sleep/waking regulation and REM sleep regulation are disrupted in PSP, leading to profound sleep deprivation without recuperation. Our findings suggest a diminished homeostatic sleep drive in PSP. This hyperaroused state is unique and is a severely disabling feature of PSP.

Progressive supranuclear palsy (PSP) and corticobasal syndrome (CBS) are both 4 microtubule binding repeat tauopathy related disorders. Clinical trials need new biomarkers to assess the effectiveness of tau-directed therapies. This study investigated the regional distribution of longitudinal diffusion tensor imaging changes, measured by fractional anisotropy, radial and axial diffusivity over 6 months median interval, in 23 normal control subjects, 35 patients with PSP, and 25 patients with CBS. A mixed-effects framework was used to test longitudinal changes within and between groups. Correlations between changes in diffusion variables and clinical progression were also tested. The study found that over a 6 month period and compared to controls, the most prominent changes in PSP were up to 3±1% higher rates of FA reduction predominantly in superior cerebellar peduncles, and up to 18±6% higher rates of diffusivity increases in caudate nuclei. The most prominent changes in CBS compared to controls were up to 4±1% higher rates of anisotropy reduction and 18±6% higher rates of diffusivity increase in basal ganglia and widespread white matter regions. Compared to PSP, CBS was mainly associated with up to 3±1% greater rates of anisotropy reduction around the central sulci, and 11±3% greater rates of diffusivity increase in superior fronto-occipital fascicules. Rates of diffusivity increases in the superior cerebellar peduncle correlated with rates of ocular motor decline in PSP patients. This study demonstrated that longitudinal diffusion tensor imaging measurement is a promising surrogate marker of disease progression in PSP and CBS over a relatively short period.