Working memory assessment with mini-mental state examination

Introduction. Delayed recall items of the Mini-Mental State Examination (MMSE) and the Mini Cognitive Examination are based on a simultaneous processing of the storage of the material to be remembered while other test's items are executed. This function is probably mediated by the working memory (WM) system. Objective. A new protocol of WM assessment during the Mini Cognitive Examination application is presented. Patients and methods. The standard Mini Cognitive Examination is modified with two new items of registration and recall. Criteria to introduce three new worn's in registration 2 and 3 is a complete recall of the three previous words. The score of the three recall trials, until a 27 points maximum, represents WM. A sample of 44 patients with probable Alzheimer's disease (AD), 19 with possible AD, 39 patients with cognitive impairment without dementia and 14 control subjects were evaluated. Results. Patients with probable AD and possible AD showed the minimum WM scores and the greatest percentage of words repetitions (WM 3.2 +/- 3.6 and 7 +/- 6, with 91 and 83% of repetitions, respectively); control subjects scored a WM of 22 +/- 6 with 14% of repetitions; patients with cognitive impairment without dementia scored a WM of 11 +/- 4 and 50% of repetitions. Conclusions. We propose that delayed recall items of the MMSE and the Mini Cognitive Examination are an index of WM capacity. This new method reinforces the assessment of delayed recall during Mini Cognitive Examination or MMSE application, and can facilitate an early detection of anterograde amnesia.