Evaluation of visual atrophy scales in cognitive impairment continuum

Introduction and Objective: Progressive cognitive impairment is a clinical symptom particularly caused by neurodegenerative dementias, mostly Alzheimer's disease (AD). It is becoming increasingly important to recognize and differentiate the neurodegenerative diseases which lead cognitive impairment in the pre-dementia state, to apply disease modifying therapies and to predict the prognosis. Precisely for this reason, various biomarkers and their combinations are used to anticipate the underlying pathology and the risk of progression of dementia in the pre-dementia stage of cognitive impairment. Structural magnetic resonance imaging (MRI) plays a substantial role among these biomarkers. In this study, our first aim was to assess the reliability of the visual scales evaluated in the different patient groups by four neurologists with different years of clinical experience. Our second aim was to investigate the patterns of atrophy in the cognitive impairment continuum, to differentiate pre-dementia stages, different causes of dementia, and to show the change in the atrophy with the severity of cognitive impairment. Methods: We included 135 patients with cognitive impairment and 38 cognitively normal individuals as the control group. Besides the visual rating scales have defined in the literature so far, new visual rating scales have been defined to evaluate the atrophy more precisely in the temporal lobe and analyse the development of temporal lobe atrophy throughout the cognitive impairment continuum. In the first cohort, visual assessment scales were applied to the groups of mild cognitive impairment (MCI)-due to AD and MCI- unlikely due to AD (based on cerebrospinal fluid biomarkers), mild Alzheimer's disease dementia (ADD), and moderate-severe ADD to determine the patterns of atrophy according to cognitive impairment severity and possible underlying pathology. In the second cohort, non-Alzheimer's dementias were grouped in accordance with the clinical diagnostic criteria. The scores of the visual assessment scales were defined in these groups, and comparisons were made between behavioural variant frontotemporal dementia (dvFTD) and ADD. In addition, the sensitivity/specificity values of the scales were evaluated according to different cut-off scores. Results: Inter-rater reliability was good in all of the scales have already defined in the literature, and also good in newly defined scales except for the left inferior temporal sulcus (0.56 in the left inferior temporal sulcus,  0.71 in the others). The visual assessment of the scales was demonstrated similar pattern in the MCI- unlikely due to AD patients and in the control group. However, in the MCI- due to AD patients scores in the all temporal and frontal scales were significantly higher than the control group and were similar to the mild ADD patients. Posterior atrophy was evident from the moderate-severe ADD stage and therefore was not a useful scale for identifying mild dementia and pre-dementia stages. It was observed that the fronto-insula, collateral sulcus, and superior temporal sulcus, and the total score of the all scales, could identify moderate-severe ADD from MCI-due to AD, unlike other scales. Among these scales, the sensitivity of MTA and collateral sulcus together at a cut-off value of ≥1 was 80.8% and the specificity was 90.1%, and evaluating these scales together was much more sensitive than the MTA score alone in determining the prodromal AD. It has been shown that white matter lesions become more pronounced in the dementia stage with the progression of cognitive impairment. In the comparison between ADD and dvFTD, orbito-frontal atrophy was more prominent in dvFTD and was the best scale in the differentiation of the two dementias. Discussion: This study has demonstrated that the visual rating scales, which can be evaluated simply and reliably on routinely acquired magnetic resonance images, can provide us valuable information regarding the aetiology and severity of cognitive impairment, as well as the differential clinical diagnosis of dementia. Atrophy in the frontal and temporal lobar regions on structural MRI in MCI patients should be considered as an indicator of an underlying Alzheimer's disease pathology. These patients may be candidates for disease modifying therapies, therefore investigations in this group should be reinforced.