Premium
P2–076: Treatment response prediction in people with Alzheimer's disease using EEG
Author(s) -
Jóhannesson Gísli,
Snædal Jón,
Blin Nicolas,
Emilsdóttir Ásdís,
Helgadóttir Halla,
Newmann Paula,
Johannsson Magnus,
Guðmundsson Þorkell,
Johnsen Kristinn
Publication year - 2013
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2013.05.719
Subject(s) - cholinergic , cholinesterase , electroencephalography , medicine , longitudinal study , clinical trial , disease , alzheimer's disease , cognition , psychology , psychiatry , pathology
Background: The majority of individuals referred to a Memory Clinic have subtle symptoms of cognitive impairment and from a clinical point of view most of them are not demented. Eventually, many of these individuals develop a form of dementia that is clinically detectable but some of them may remain stable for years, and the concern may even disappear. The EEG diagnostic tool presented here is designed to help the clinician to evaluate possible progress of the cognitive symptoms in MCI subjects over time. Methods: In an EEG database EEGs from 342 mild AD subjects and 181 MCI individuals have been entered. The EEG data from the MCI individuals was collected over a period of 7 years (20052011). Over the course of 8 years these MCI individuals have been followed up. The MCI individuals were divided into three subgroups according to their clinical status at follow up: prodromal AD (pAD), stable MCI (sMCI), and other. The pAD and sMCI groups were compared by constructing a classifier by applying statistical pattern recognition to a large set of EEG features. Results: Of the 181 MCI individuals 70 were diagnosed with AD 1-7.7 years after the EEG measurement with a mean of 2 years. 79 remainedMCI for at least 3 years, with a mean of 4.75 years and standard deviation of 1.00 year. The remaining 32 individuals had developed other types of dementia at follow up. The pAD-sMCI classifier indicates that using EEG it is possible to predict which of the MCI individuals will at a later time develop AD with accuracy of about 80%. Conclusions: Following amore substantial clinical validation and an easy access to the methodology, we expect this application of clinical EEG in support for differential diagnosis of mild cognitive impairment to become a realistic first step in the full clinical workup of patients who visit a memory clinic. The underlying technology is well known, widely available and inexpensive in relation to other imaging techniques.