Key points are not available for this paper at this time.
We examined the ability of sleep/wake measures to discriminate 45 control subjects from 44 mild Alzheimer's disease (AD) patients. Sleep fragmentation was observed as indicated by significant increases in time awake (37-52%) and number of awakenings (31-36%) during the night as compared to controls. Further, slow wave sleep (SWS) was significantly reduced (22%) in AD patients relative to controls. These findings are consistent with our earlier observations of increased wakefulness and decreased SWS in mild-moderate, moderate-severe, and severe stage AD patients. However, when we used these sleep/wake stage measures in discriminant analyses to classify the current AD subjects vs control subjects, the analyses failed to confirm our earlier high classification rate (90%). The present groups were discriminated at overall classification rates of only 63-67%. We conclude that while sleep/wake patterns are significantly disturbed in AD, this phenomenon is not diagnostically useful for discrimination of mild stage AD.
Vitiello et al. (Sun,) studied this question.