Cut-off score of the Mattis Dementia Rating Scale for screening dementia in Parkinson's disease

The prevalence of dementia in Parkinson's disease (PD) is close to 30%, and its incidence is 4 to 6 times higher than in age-matched general population. PD with dementia (PDD) is mainly characterized by a predominant and progressive frontal- subcortical impairment. The Mattis Dementia Rating Scale (MDRS) is a commonly used screening test that sensitively measures the degree of frontal-subcortical defects. Although the MDRS has been validated as a screening test of cognitive dysfunction in nondemented PD patients (PD-ND), its utility for screening dementia in PD is unknown. In order to validate the MDRS for diagnosis of PDD it was prospectively administered to 92 PD patients (57 PD-ND, 35 PDD) fulfilling UK-PDSBB criteria. Dementia was diagnosed according to DSM-IV-TR and a Clinical Dementia Rating (CDR) scale score ≥1. Univariate, logistic regression, and ROC curve analysis were carried out to measure the discriminative power of MDRS in PDD. Regression analysis showed MDRS total scores to independently differentiate PD-ND from PDD (P < 0.001). Age and education did not predict the presence of dementia. ROC curve analysis showed a cut-off score of ≤123 on the MDRS total scores to yield high sensitivity (92.65%), specificity (91.4%), positive and negative predictive values (PPV 83.3%, NPV 96.4%). A brief version of the MDRS obtained by the addition of the memory, initiation/ perseveration, and conceptualization subscores yielded similar discriminant properties. The MDRS has an excellent discriminant ability to diagnose dementia in PD and provides an objective measure to distinguish PD-ND from PDD. © 2008 Movement Disorder Society.