P1‐312: Differentiating Amnestic Mild Cognitive Impairment Converting to Probable Alzheimer's Disease from Stable Amnestic Mild Cognitive Impairment Using FDG‐PET and an Alzheimer's Disease‐related Hypometabolism Overlap Index

Napatkamon Ayutyanont,Kewei Chen,Xiaofen Liu,Cole Reschke,Wendy Lee,Dan Bandy,Gene E. Alexander,William J. Jagust,Robert A. Koeppe,Norman L. Foster,Eric M. Reiman
DOI: https://doi.org/10.1016/j.jalz.2008.05.903
2008-01-01
Alzheimer s & Dementia
Abstract:Alzheimer's disease (AD) is characterized by a characteristic pattern of reductions in the cerebral metabolic rate for glucose (CMRgl). We recently introduced use of an AD-Related Hypometabolism Overlap Index to determine the percentage of previously characterized AD-related hypometabolic brain voxels that are hypometabolic in an individual subject's fluorodeoxyglucose positron emission tomography (FDG-PET) image and help distinguish between probable AD patients, amnestic mild cognitive impairment (MCI) patients and normal controls (NC). The objective of sthis study is to evaluate the ability of our proposed Overlap Index to distinguish between probable AD patients, MCI patients who convert to probable AD within the next 18 months (MCIc), MCI patients who have so far remained stable (MCIs) and NC from the AD Neuroimaging Initiative (ADNI). The Overlap Index was computed in 74 AD patients, 14 MCIc, 124 MCIs patients, and 82 NC as follows: First, statistical parametric mapping (SPM5) was used to generate individual cerebral hypometabolism z-score maps, comparing each person to a database of 82 NC. Next, the individual hypometabolism map was superimposed on an AD-related hypometabolic map generated from a previous study (Alexander et al, 2002). Finally, the Overlap Index was computed as VADI /VAD×100%, such that VADI is the hypometabolic volume in both previous AD-NC group comparison and the subsequent individual-NC comparison, and VAD is the hypometabolic volume in only the previous AD-NC group comparison. The Overlap Index significantly distinguished the four subject groups (ANOVA P=8.8e-15). As predicted, the overlap index had the following rank order: ADMCIcMCIsNC (ANOVA with linear trend P=3.9e-16). Moreover, the Overlap Index significantly distinguished between the MCIc group and the MCIs group (t-test P=0.013). Our preliminary findings suggest that the AD Overlap Index can help identify hypometabolic patterns consistent with AD and help differentiate MCI patients converting to AD from stable MCI patients without the need to correct for multiple comparisons.
What problem does this paper attempt to address?