BACKGROUND: Alzheimer's disease (AD) is assessed by carefully examining a patient's cognitive impairment. However, previous studies reported inadequate diagnostic accuracy for dementia in primary care settings. Many hospitals use the automated quantitative evaluation method known as the Voxel-based Specific Regional Analysis System for Alzheimer's Disease (VSRAD), wherein brain MRI data are used to evaluate brain morphological abnormalities associated with AD. Similarly, an automated quantitative evaluation application called the easy Z-score imaging system (eZIS), which uses brain SPECT data to detect regional cerebral blood flow decreases associated with AD, is widely used. These applications have several indicators, each of which is known to correlate with the degree of AD. However, it is not completely known whether these indicators work better when used in combination in real-world clinical practice. METHODS: We included 112 participants with mild cognitive impairment (MCI) and 128 participants with early AD in this study. All participants underwent MRI, SPECT, and the Mini-Mental State Examination (MMSE). Demographic and clinical characteristics were assessed by univariate analysis, and logistic regression analysis with a combination of MMSE, VSRAD and eZIS indicators was performed to verify whether the diagnostic accuracy in discriminating between MCI and early AD was improved. RESULTS: The area under the receiver operating characteristic curve (AUC) for the MMSE score alone was 0.835. The AUC was significantly improved to 0.870 by combining the MMSE score with two quantitative indicators from the VSRAD and eZIS that assessed the extent of brain abnormalities. CONCLUSION: Compared with the MMSE score alone, the combination of the MMSE score with the VSRAD and eZIS indicators significantly improves the accuracy of discrimination between patients with MCI and early AD. Implementing VSRAD and eZIS does not require professional clinical experience in the treatment of dementia. Therefore, the accuracy of dementia diagnosis by physicians may easily be improved in real-world primary care settings.
BACKGROUND:Alzheimer's disease (AD) is assessed by carefully examining a patient's cognitive impairment. However, previous studies reported inadequate diagnostic accuracy for dementia in primary care settings. Many hospitals use the automated quantitative evaluation method known as the Voxel-based Specific Regional Analysis System for Alzheimer's Disease (VSRAD), wherein brain MRI data are used to evaluate brain morphological abnormalities associated with AD. Similarly, an automated quantitative evaluation application called the easy Z-score imaging system (eZIS), which uses brain SPECT data to detect regional cerebral blood flow decreases associated with AD, is widely used. These applications have several indicators, each of which is known to correlate with the degree of AD. However, it is not completely known whether these indicators work better when used in combination in real-world clinical practice. METHODS: We included 112 participants with mild cognitive impairment (MCI) and 128 participants with early AD in this study. All participants underwent MRI, SPECT, and the Mini-Mental State Examination (MMSE). Demographic and clinical characteristics were assessed by univariate analysis, and logistic regression analysis with a combination of MMSE, VSRAD and eZIS indicators was performed to verify whether the diagnostic accuracy in discriminating between MCI and early AD was improved. RESULTS: The area under the receiver operating characteristic curve (AUC) for the MMSE score alone was 0.835. The AUC was significantly improved to 0.870 by combining the MMSE score with two quantitative indicators from the VSRAD and eZIS that assessed the extent of brain abnormalities. CONCLUSION: Compared with the MMSE score alone, the combination of the MMSE score with the VSRAD and eZIS indicators significantly improves the accuracy of discrimination between patients with MCI and early AD. Implementing VSRAD and eZIS does not require professional clinical experience in the treatment of dementia. Therefore, the accuracy of dementia diagnosis by physicians may easily be improved in real-world primary care settings.
Authors: Jennifer L Whitwell; Dennis W Dickson; Melissa E Murray; Stephen D Weigand; Nirubol Tosakulwong; Matthew L Senjem; David S Knopman; Bradley F Boeve; Joseph E Parisi; Ronald C Petersen; Clifford R Jack; Keith A Josephs Journal: Lancet Neurol Date: 2012-09-03 Impact factor: 44.182
Authors: Heiko Braak; Irina Alafuzoff; Thomas Arzberger; Hans Kretzschmar; Kelly Del Tredici Journal: Acta Neuropathol Date: 2006-08-12 Impact factor: 17.088