Introduction Cognitive impairment is common in Parkinson's disease (PD), even in the early stages, and appropriate screening tools which are suitable for use in the general neurology clinic are needed.

Methods We investigated the utility of the Addenbrooke's Cognitive Examination-Revised (ACE-R) as a tool for detecting mild cognitive impairment in PD (PD-MCI), defined using a detailed neuropsychological battery in an incident PD cohort. ACE-R was also validated against the Parkinson's Disease Cognitive Rating Scale (PD-CRS). We investigated the relationship between ACE-R score and behaviour and day-to-day function using the Cambridge Behavioural Inventory—Revised (CBI-R) and Parkinson's Disease Questionnaire-39 (PDQ-39). The ACE-R cognitive profile in non-demented PD was compared with profiles in of early cognitive impairment in Huntington's disease (HD), Progressive Supranuclear Palsy (PSP), amnestic mild cognitive impairment (MCI) and Alzheimer's Disease (AD).

Results 27% of the incident PD cohort were defined as PD-MCI. ROC analysis revealed an AUC for the ACE-R of 0.89 (95% CI 0.81 to 0.96). An ACE-R cut-off of <89 gave specificity 91%, sensitivity 65%, whereas a cut-off of <94 gave a sensitivity of 92% and specificity 64%. ACE-R correlated well with PD-CRS (r=0.801, p<0.0001). Significant correlations were seen between the ACE-R and CBI-R (cc=−0.249, p=<0.0001), and PDQ-39 (cc=−0.246, p=<0.0001) following adjustment for age, disease duration, Unified Parkinson's Disease Rating Scale (UPDRS) motor score and Beck Depression Inventory score (BDI). Across disease groups there was a significant difference between ACE-R profiles.

Conclusion The ACE-R is a valid screening tool for mild cognitive impairment in PD and correlates with behavioural deficits and quality of life. Furthermore it is capable of distinguishing between the cognitive profile in non-demented PD and a range of other neurological conditions.