Yamaguchi fox-pigeon imitation test (YFPIT) for dementia in clinical practice

In out-patient clinics, having simple procedures to check for signs of dementia is invaluable. In the present study, we evaluated the imitation of hand gestures to detect visuomotor deficits in dementia in clinical practice.

In all, 1219 subjects were enrolled in the present study, including 497 with Alzheimer's disease (AD), 98 with dementia with Lewy bodies (DLB), 71 with other types of dementia diseases, 175 with a Clinical Dementia Rating (CDR) of 0.5, and 378 normal controls. All subjects were aged 65 years or older. Subjects were recruited from 10 clinics and two communities. Visuomotor function was evaluated by the Yamaguchi fox-pigeon imitation test (YFPIT), which consists of a simple one-handed sign for 'fox' and a complex two-handed sign for 'pigeon', a rapid, game-like test with low psychological burden.

The success rate (successful/total) for imitating the 'pigeon' hand gesture was reduced as the severity of the dementia increased: 85.7% in normal controls, 60.6% in CDR 0.5 (mild cognitive impairment), 39.2% in CDR 1 (mild dementia), 21.2% in CDR 2 (moderate dementia), and 5.7% in CDR 3 (severe dementia). The success rate for imitating the 'pigeon' hand gesture was higher in patients with DLB than AD within the CDR 1 group (51.2% vs 35.4%, respectively), but lower for patients with DLB than AD within the CDR 2 group (12.5% vs 24.4%, respectively). The success of imitating the hand gesture for 'fox' was similar for patients with AD and DLB. Of those subjects who failed to imitate the hand gesture for 'pigeon', 49.5% of those with AD showed the palm-palm pattern (both palms facing outward), suggesting deficits of perspective conversion from the first-person to the third-person. Conversely, 52.8% of patients with DLB showed a dorsum-dorsum pattern (both dorsa facing outwards), suggesting deterioration of visual attention and recognition.

In conclusion, the YFPIT is a useful test to detect visuomotor deficits in dementia that can differentiate between AD and DLB.