RESUMO
OBJECTIVE: To pilot a three-step diagnostic model for young and old patients with Parkinson's disease dementia (PDD). METHODS: Prospective investigator-blinded study. We developed a screening questionnaire for patients with Parkinson's disease (PD) and their caregivers. Further, patients were subjected to three screening instruments (Montreal Cognitive Assessment (MoCA), Frontal Assessment Battery (FAB), Addenbrooke's Cognitive Examination-revised (ACE-R) and a detailed neuropsychological examination (NPE). Based on the NPE, patients were divided in a PD (without dementia) and a PDD-group. RESULTS: Forty-one PD patients, aged 37-94 years, participated in this study. Patients were divided in a young group, < or = 65 (n = 22) and an old group >65 years (n = 19). In the young group (PDD, n = 5) the patient-screening questionnaire predicted PDD with a sensitivity/specificity of 100.0%/94.1%; in the old group (PDD, n = 10) the proxy-screening questionnaire predicted PDD with a sensitivity/specificity of 88.9%/66.7%. In the young group, ACE-R had the largest Area Under the Curve (AUC) 0.88 (0.70-1.00), in the old group MoCA (AUC 1.00). However, the three instruments did not differ significantly. CONCLUSIONS: It seems feasible and efficient to use three consecutive diagnostic steps for PDD: (1) a screening questionnaire, (2) if positive: MoCA, FAB or ACE-R as screening instrument and (3) if positive: a detailed NPE for diagnosing PDD.
