Affiliations: [a]
Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health (DINOGMI), University of Genoa, Genoa, Italy
| [b] IRCCS Ospedale Policlinico San Martino, Genoa, Italy
| [c] Istituto Nazionale di Fisica Nucleare (INFN), Genoa, Italy
| [d]
Department of Health Science (DISSAL), University of Genoa, Genoa, Italy
Correspondence:
[*]
Correspondence to: Dr. Nicola Girtler, Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health (DINOGMI), University of Genoa IRCCS Ospedale Policlinico San Martino, Largo Paolo Daneo, 3, 16132 Genoa, Italy. Tel.: +39 0103537778; Fax +39 0105556893; E-mail: nicola.girtler@unige.it.
Abstract: Background:Neuropsychological assessment is still the basis for the first evaluation of patients with cognitive complaints. The Free and Cued Selective Reminding Test (FCSRT) generates several indices that could have different accuracy in the differential diagnosis between Alzheimer’s disease (AD) and other disorders. Objective:In a consecutive series of naturalistic patients, the accuracy of the FCSRT indices in differentiating patients with either mild cognitive impairment (MCI) due to AD or AD dementia from other competing conditions was evaluated. Methods:We evaluated the accuracy of the seven FCSRT indices in differentiating patients with AD from other competing conditions in 434 consecutive outpatients, either at the MCI or at the early dementia stage. We analyzed these data through the receiver operating characteristics curve, and we then generated the odds-ratio map of the two indices with the best discriminative value between pairs of disorders. Results:The immediate and the delayed free total recall, the immediate total recall, and the index of sensitivity of cueing were the most useful indices and allowed to distinguish AD from dementia with Lewy bodies and psychiatric conditions with very high accuracy. Accuracy was instead moderate in distinguishing AD from behavioral variant frontotemporal dementia, vascular cognitive impairment, and other conditions. Conclusion:By using odd-ratio maps and comparison-customized cut-off scores, we confirmed that the FCSRT represents a useful tool to characterize the memory performance of patients with MCI and thus to assist the clinician in the diagnosis process, though with different accuracy values depending on the clinical hypothesis.
