RESUMO
INTRODUCTION: Fifteen percent of people with mild cognitive impairment (MCI) will progress to dementia within 2 years. There is increasing focus on the evaluation of biomarkers which point towards the underlying pathology. This enables better prediction of clinical outcomes. Early diagnosis of the dementia subtype is crucial for appropriate management and accurate prognosis. The aim of this study was to compare MRI measures in stable mild cognitive impairment patients (stable-MCI), prodromal Alzheimer's disease (pro-AD), and prodromal dementia with Lewy bodies (pro-DLB). METHODS: Out of 1,814 patients assessed in Essex memory clinic between 2002 and 2017, 424 had MCI at baseline with follow-up data. All patients underwent comprehensive clinical and cognitive assessment at each assessment. MRI scans were acquired at patients' baseline assessment, corresponding to the time of initial MCI clinical diagnosis. Patients were grouped according to their diagnosis at the end of follow-up. All baseline scans were visually rated according to established rating scales for medial temporal atrophy (MTA), global cortical atrophy (GCA), and white matter lesions (WMLs). RESULTS: MRI scans were available for 28 pro-DLB patients and were matched against 27 pro-AD and 28 stable-MCI patients for age, sex, and education. The mean follow-up duration was 34 months for the pro-AD group, 27 months for the pro-DLB group, and 21 months for the stable-MCI group. MTA scores were significantly greater in pro-AD patients compared to pro-DLB (p = 0.047) and stable-MCI patients (p = 0.012). There was no difference on GCA or WMLs between pro-AD, pro-DLB, and stable-MCI. CONCLUSIONS: This study indicates that a simple visual rating of MTA at the stage of MCI already differs at a group level between patients that progress to AD, DLB, or continue to be stable-MCI. This could aid clinicians to differentiate between MCI patients who are likely to develop AD, versus those who might progress to DLB or remain stable.