P1-248: Diagnostic utility of cross-sectional and longitudinal hippocampal volumetry in memory clinic patients

Abstract

Background: Alzheimer’s disease (AD) is characterized by reduced hippocampal volume on MR imaging. Rate of loss may be a more sensitive marker of disease progression than cross-sectional volume. We compared hippocampal baseline volume with atrophy rate, measured using MRI, in their ability to distinguish AD and mild cognitive impairment (MCI) patients from controls, and to predict progression to dementia. Methods: We included 63 patients with AD (age 67 ( 8) years; 43% male), 42 patients with MCI (age 71 ( 6) years; 52% male) and 33 controls (age 67 ( 9) years; 55% male) with two serial 3D T1-weighted MRI sequences. Hippocampi were manually outlined on baseline images and normalized for head size[n1] . Annualized percentage volume change within the baseline hippocampal region was calculated using non-linear (fluid) registration of follow-up scans to baseline. Volume and atrophy rate of left and right side were averaged. Differences between the diagnostic categories were calculated using ANOVA. After dichotomisation according to the median value, Cox proportional hazards models were used to calculate the risk of progression of non-demented patients (controls and MCI) to dementia. Results: Normalized baseline hippocampal volumes differentiated AD (3518 ( 625) mm) and MCI patients (3647 ( 506) mm) from controls (4072 ( 341)mm; p 0.01 for both measures), but did not differentiate AD from MCI patients (p 0.22). Atrophy rate also differentiated the AD (-4.0( 1.2) %/year) and MCI patients (-3.7( 1.3) %/year) from controls (-2.2 ( 1.3) %/year; p 0.001 for both measures), but did not differentiate AD from MCI patients (p 0.39). Adjusted for age and sex, both baseline volume (HR (95% CI): 3.2 (1.3-8.4) and atrophy rate (HR (95% CI): 6.8 (2.6-18.3) predicted progression of non-demented subjects to dementia. After additionally adjusting for MMSE, the predictive effect of baseline volume lost significance (HR (95% CI): 2.2 (0.8-5.9)), but hippocampal atrophy rate remained a predictor of progression (HR (95% CI): 4.9 (1.8-13.7)). Conclusions: Both hippocampal baseline volume and atrophy rate can be used to differentiate patients with MCI and AD from controls. Although both measures predict progression to dementia within non-demented subjects, the predictive effect of atrophy rate may be stronger, and less dependent on MMSE than baseline volume.