Impairment of spatially directed attention in patients with probable Alzheimer's disease as measured by eye movements.

OBJECTIVE To investigate changes in spatially directed attention in patients with a diagnosis of probable Alzheimer's disease (AD). BACKGROUND Impaired attention in patients with probable AD has not been the subject of extensive research. Yet recent reports suggest that attentional deficits may be an important early feature of the disease in a subset of patients. SETTING University hospital center studying dementia and aging. SUBJECTS Ten mild to moderately impaired patients diagnosed as having probable AD, by National Institute of Neurologic and Communicative Diseases and Stroke criteria, and 11 healthy age- and education-matched controls. MEASURES Eye movements were recorded as subjects participated in two experiments designed to measure spatially directed attention. Subjects were instructed to (1) attend to and fixate a target appearing randomly to the right or left of a central marker and (2) direct attention to and fixate a target appearing randomly in one of four peripheral locations. RESULTS Patients with probable AD exhibited fewer accurate trials and longer saccade latencies in both tasks. As a group, patients performed worse in the second task that placed increased demands on attention. However, the performance of patients in this second experiment varied. Four patients performed significantly worse than all other patients, while three patients performed as well as controls. Errors in the second task were reviewed to identify specific types of attentional deficits. Six empirically derived error patterns were classified into one of two major categories: perseveration and impersistence. Seven of 10 patients made greater than 50% errors of perseveration, and three of 10 made greater than 50% errors of impersistence. CONCLUSIONS Impairment of attention may be an early feature of AD and a prominent clinical characteristic of some patients. The differences observed in error types made by patients may reflect the varied distribution of neuropathologic changes affecting structures that mediate aspects of attention. The architecture of eye movements can be used as a physiologic measure that should provide useful information for the diagnosis and clinicopathologic subtyping of patients with AD.