Aims: The main aims of the study were the translation and the subsequent validation in Italian of the Addenbrooke's Cognitive Examination Revised (ACE-R), and the evaluation of its usefulness in discriminating cognitively normal subjects from patients with mild dementia in an elderly population. Methods: The ACE-R was translated and adapted into Italian. The Italian ACE-R was administered to a group of 179 elderly subjects (72 cognitively healthy and 107 subjects with mild dementia, mean age 75.4 ± 6.4 years). The group was stratified into two subsamples according to age, i.e. a young-old ( < 75 years) and an old-old ( ≥ 75 years) group, in order to evaluate the sensitivity and specificity of the test in detecting dementia in different age strata of elderly subjects. Results: The reliability of the Italian ACE-R was extremely good ( alpha -coefficient = 0.85). Two different cutoffs were identified for young-old (cutoff 79; sensitivity 90% and specificity 80%) and old-old subjects (cutoff 60; sensitivity 82% and specificity 100%). Conclusions: The Italian ACE-R is a valid screening tool to detect dementia, especially in the old-old population, which represents not only the fastest growing age group but also the group at the highest risk of dementia in Western countries.