Phenotypic and metabolic imaging characterization of posterior cognitive dysfunctions.

In the last decade, the differential diagnosis of dementia has become very challenging. Different clinical syndromes can in fact be associated with the same underlying pathology and different pathologies can be associated with the same clinical phenotype, making it very difficult to distinguish them in vivo. In the present project we focused on the differential diagnosis of cognitive syndromes affecting, at onset, posterior cerebral networks with particular reference to PCA, CBD and LBD, respectively. We enrolled 70 consecutive patients, referred to the memory clinic of the Neurology Department of S. Gerardo Hospital, with a cognitive profile characterized primarily by impairment of posterior cognitive functions and a relative spare of language and memory functions. Each patient underwent, an extensive neuropsychological battery, a neurological examination and a [18F]FDG-PET in close proximity to the neuropsychological assessment. Applying Principal Components Analysis to all the tasks of posterior cognitive functions, we highlighted three cognitive sub-syndromes that we interpret as anatomical-based and label as “left parietal”, “left temporo-occipital” and “right parietal” and we classify patients on the basis of these three components. Cerebral metabolism measured with [18F]FDG-PET confirm this interpretation of the three sub-syndromes. We then compared empirical patients classification and the diagnoses made with the current diagnostic criteria for Posterior Cortical Atrophy (PCA), Corticobasal degeneration (CBD), Lewy body dementia (LBD) and Alzheimer’s disease (AD). One of the most important evidence was that all patients classified as “left temporo-occipital” were diagnosed as AD or PCA (only one of these patients met both criteria of PCA and CBD). Two additional quite interesting associations emerged, although less strong: the first between “left parietal” and CBD diagnosis and the second between “right parietal” and PCA diagnosis. Finally a large group of patients could not be classified only on the basis of their very specific posterior cognitive profile. In conclusion, this new empirical approach seems to be quite useful in the differential diagnosis of dementia syndromes.