The role of hippocampal subfield volumes in subjective cognitive decline

281 

Abstract Submission 

Linda MAH¹, Susan Vandermorris², Nicolaas Verhoeff³, Nathan Herrmann⁴

¹Rotman Research Institute, Baycrest, University of Toronto, Toronto, Ontario, ²Baycrest, Toronto, Ontario, ³Baycrest, University of Toronto, Toronto, Ontario, ⁴Sunnybrook Health Sciences, University of Toronto, Toronto, Ontario

Linda MAH, MD, MHS, FRCPC  
Rotman Research Institute, Baycrest, University of Toronto
Toronto, Ontario

Susan Vandermorris  
Baycrest
Toronto, Ontario

Nicolaas Verhoeff  
Baycrest, University of Toronto
Toronto, Ontario

Nathan Herrmann  
Sunnybrook Health Sciences, University of Toronto
Toronto, Ontario

Subjective cognitive decline (SCD), defined as perception of memory or other cognitive decline with normal test performance, is conceptualized as a possible preclinical stage of Alzheimer's disease (AD) (1). Hippocampal volume reduction is well-established as a biomarker of AD risk (2). The hippocampal complex includes functionally and structurally distinct subfields (cornu ammonis (CA), dentate gyrus, and subiculum) which show selective atrophy in aging and in AD (3), as well as psychiatric conditions such as major depressive disorder (4). Previous work demonstrating patterns of subfield atrophy in Mild Cognitive Impairment (MCI), a prodrome of AD, support the utility of hippocampal subfield volumetry, over measurement of total hippocampus, to detect risk for AD (5).

Evidence of volumetric alterations in hippocampal subfields in SCD is emerging but inconsistent (6; 7). One potential reason for the discrepancy in findings is unaccounted previous diagnosis of depression in SCD participants. Depression frequently co-occurs with SCD (1) and itself is associated with hippocampal atrophy, yet the majority of studies of SCD failed to ascertain the presence of previous depression. In the current study, we measured hippocampal subfield volumes using structural magnetic resonance imaging (MRI) at 3 Tesla with the goals of identifying group differences in subfield volumes between older adults with SCD, who were free of current or past depression, and cognitively unimpaired (CU) older adults, and examining associations between hippocampal subfield volumes and subjective memory ability.

The sample included 26 SCD (13M, age 70.6 SD=5.1 years) and 25 CU (10M, age 71.4 SD=7.4 years) older adults. All participants performed within normal on a neuropsychological battery and screened negative for past or current psychiatric illnesses during psychiatric assessment. SCD was established based on affirmative responses to "Do you feel your memory is becoming worse?" "If so, are you concerned?"(1). Subjective memory ability was further quantified using the Memory Functioning Questionnaire (MFQ)(8) MR scanning was performed on a 3T Siemens Trio scanner using a 12-channel head coil. A high-resolution (voxel size 0.43 x 0.43 x 3 mm) T2-weighted scan of the medial temporal lobe was acquired in an oblique-coronal plane, perpendicular to the hippocampal long axis. The Automatic Segmentation of Hippocampal Subfields (ASHS) processing pipeline was used to extract bilateral CA1, CA2, CA3, dentate gyrus, and subiculum volumes (9). Independent t-tests were used to compare hippocampal subfields between groups. Partial correlations were performed within the total sample to examine the association between subfield volumes and MFQ Frequency of Forgetting ("How often do you...?") subscale scores, which has been linked to AD neuroimaging markers,(10) as well as long delayed recall scores on the California Verbal Learning Test (CVLT-LDR). Covariates were age, sex, and total intracranial volume. Bonferroni correction for multiple comparisons was applied.

Compared to CU, left CA2 volume was reduced in SCD [t (49) = 2.00, P =.025], while right CA3 volume was greater in SCD [t (49) = -1.81, P =.038] (Figure 1), although these findings were not statistically significant following correction for multiple comparisons. Frequency of Forgetting was significantly correlated with left CA2 [r = .42, P =.003] and right subicular volume [r = .45, P =.001. In contrast, CVLT-LDR was not correlated with any hippocampal subfield volume (Figure 2).

These findings suggest a role for CA2, a region linked to social memory in animal models, in SCD. The association between volume of subiculum, a region implicated in AD, and subjective, but not objective memory, suggests that self-awareness of change in memory in everyday life may be sensitive to subtle disease-related volumetric changes within the hippocampus in the earliest stages of the AD course.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) ¹

Higher Cognitive Functions Other

Learning and Memory Other ²

Anatomical MRI

HIGH FIELD MR

Memory

Neurological

STRUCTURAL MRI