Clinical and demographic predictors of mild cognitive impairment for converting to Alzheimer's disease and reverting to normal cognition

Objective: To identify clinical and demographic predictors for mild cognitive impairment (MCI) conversion to Alzheimer's disease (AD) or reversion to normal cognition, and sustained MCI. Methods: In total, 74 baseline MCI subjects were retrospectively investigated and categorized into three subgroups: conversion to AD, sustained MCI, or reversion to normal cognition during one year. The clinical and demographic characteristics assessed were age, gender, educational attainment, vascular risk factors, white matter lesions (WMLs), and parahippocampal gyrus atrophy (PGA), analyzed by magnetic resonance imaging (MRI) using the voxel-based specific regional analysis system for AD (VSRAD). Results: Of the 74 MCI subjects, 29 (39.2%) were classified as "converters", 39 (52.7%) as "sustained MCI", and 6 (8.1%) as "reverters". Among the three subgroups, there were significant differences in educational attainment (years) (*p = 0.03), baseline mini-mental state examination (MMSE) scores (***p < 0.001), and periventricular and deep white matter hyperintensity grades (*p = 0.02 and *p = 0.03, respectively). Baseline PGA showed a significant increasing trend among the three subgroups (reverters < sustained MCI < converters, P-### < 0.001). MCI subjects with higher educational attainment and low VSRAD Z-scores without WMLs were associated with reversion to normal cognitive function. Conclusions: Risk factors for MCI conversion to AD were low educational attainment, low baseline MMSE scores, high grade WMLs, and high VSRAD Z-scores. High educational attainment, low VSRAD Z-scores, and no WMLs characterized reversion to normal cognition. (C) 2014 Elsevier B.V. All rights reserved.