Usefulness of a cube-copying test in outpatients with dementia

Objective: To clarify whether a quantitatively scored cube-copying test could rapidly assess dementia patients, predicting thier performance in cognitive tests. Methods: Subjects were 171 outpatients with amnesia who were including 92 with Alzheimer's disease (AD); 59 vascular dementia (VD); 17 frontotemporal dementia (FTD); and three lewy body disease (DLB) and 32 normal healthy subjects. Subjects asked to copy a perspective drawing of a cube. Points of connection and plane-orientation errors were scored using Maeshima's method. The Mini-Mental State Examination (MMSE), an auditory verbal learning test (AVLT), a word fluency test and Raven's Coloured Progressive Matrices (RCPM) were administered. Age, clinical disease severity, symptom duration, specific diagnosis and neuropsychological scores were evaluated for relationships with constructional ability. Results: The cube-copying test showed errors in most dementia patients and 11 of 32 normal subjects. Only three patients each with AD and VD copied correctly, but 11 patients with 17 FTD drew the cube correctly. Numbers of connections completed and plane-orientation errors correlated significantly with MMSE, AVLT, word fluency and RCPM scores. Conclusion: The cube-copying test is useful for routine clinical dementia screening, however the test examines only one aspect of cognitive function. While not an alternative to conventional neuropsychological examinations, quantitatively scored cube copying can provide a rough estimate of cognitive dysfunction in dementia patients.