P3‐100: 3‐D mapping of episodic memory and executive function in late‐onset depression

Huali Wang,Na Zhang,Mei‐Yan Zhang,Jing Liao,Daniel Chang,Huishu Yuan,Min‐Ying Su,Xin Yu

DOI: https://doi.org/10.1016/j.jalz.2009.04.1176

2009-01-01

Abstract:Cognitive impairment and late-life depression (LOD) are associated with increased risk for subsequent dementia. This study aimed to explore the neural correlates of memory performance and executive function in LOD from the structural neuroimaging perspective. Nine patients with late-onset depression (ICD-10 criteria, LOD group, age=64.9±9.2, male=4, female=5, MMSE=22.7±5.5), fourteen subjects meeting aMCI criteria proposed by Petersen (aMCI group, age=72.6±4.9, male=8, female=6, MMSE=26.0±1.9), and 16 healthy elderly controls (NC group, age=69.9±2.8, male=6, female=10, MMSE=28.6±2.6) with normal-appearing white matter were enrolled in this prospective study. All subjects were administered a set of memory tests, consisting of common object memory test (immediate memory, 5-min delayed recall, 30-min delayed recall) and logical memory (WMS-IIIR). The composite memory score was calculated as the sum of these four test scores. Category fluency test was used to assess the executive function. High-resolution 3D T1 structural images were acquired with 3D-MPRAGE sequence and were analyzed with SPM5 by following optimized voxel-based morphometry (VBM) protocol. The composite memory score and verbal fluency test score were entered as covariates in a general linear model that predict the gray matter density at each voxel. After controlling age and education level, the composite memory scores in LOD and aMCI groups were significantly lower than in NC group (F(4)=4.79, P=0.004). The verbal fluency score was significantly lower in LOD group relative to NC group (F(4)=5.77, P=0.001). There were no significant difference in composite memory score (P=0.49) or verbal fluency score (P=0.201) between LOD and aMCI group. In LOD group, composite memory score correlated strongly with gray matter density in right medial frontal gyrus (r=0.54, Puncorrected<0.001, Figure 1A). Strong correlation was also observed between verbal fluency score and gray matter density in right superior temporal gyrus (r=0.67, Puncorrected<0.001, Figure 1B). In aMCI and NC groups, there was no significant correlation between gray matter density and either composite memory or verbal fluency score. In late-life depression, right frontal lobe is closely linked with memory and right temporal lobe more associated with executive function. Integration of structural and functional neuroimaging techniques is warranted to explore the neural mechanism of cognitive function in late-life depression. Regression plots of the correction between composite memory score (A) and verbal fluency score (B) with gray matter density in the voxel of highest statistical significance in late-onset depression. (A): corresponding to the medical frontal gyrus of the right frontal lobe (MNI coordinates x,y,z=[22 50 16], r=0.5418, p<0.001; (B):corresponding to the superior temporal gyrus of the right temporal lobe (MNI coordinates x,y,z=[34-54 20], r=0.7196, p<0.001)