Elsevier

NeuroImage

Volume 30, Issue 3, 15 April 2006, Pages 1010-1020
NeuroImage

Declarative memory impairments in Alzheimer's disease and semantic dementia

https://doi.org/10.1016/j.neuroimage.2005.10.008Get rights and content

Abstract

Semantic dementia (SD) and Alzheimer's disease (AD) are both disorders in which early pathology affects the temporal lobe yet they produce distinct syndromes of declarative memory impairment—loss of established semantic knowledge with relatively preserved episodic memory in the former and the converse in the latter. Groups with mild SD and mild AD who showed a double dissociation in these two aspects of declarative memory were studied—the SD group's episodic memory and the AD group's semantic knowledge each being comparable to controls. Positron emission tomography and volumetric magnetic resonance imaging were used to map deficits in regional cerebral metabolic rate and mesial temporal lobe (MTL) atrophy, respectively. Episodic memory impairment in AD was associated with dysfunction of an integrated network (mesial temporal lobe, mamillary bodies, dorso-mesial thalamus and posterior cingulate). Semantic memory impairment in SD was associated with bilateral rostral temporal lobe hypometabolism. The SD group had comparable MTL atrophy and hypometabolism to that found in AD but the remainder of their limbic–diencephalic network was preserved suggesting that the latter explains their ability to acquire new episodic memories. The results challenge the view that amnesia in early AD can be explained by the degree of MTL damage alone while showing that semantic impairment can occur with damage restricted to the rostral temporal lobes.

Introduction

Episodic memory impairment is the first and most severe deficit in Alzheimer's disease (AD). It is often assumed that this deficit is a consequence of mesial temporal lobe (MTL) damage; focal lesions to the MTL are known to cause amnesia (Scoville and Milner, 1957) and this is the site of most severe neurofibrillary tangle deposition in AD (Braak and Braak, 1995, Delacourte et al., 1999). The syndrome of semantic dementia (SD), a variant of fronto-temporal lobar degeneration (Neary et al., 1998), is also associated with neurodegenerative changes in the temporal lobes but gives rise to a different early cognitive profile to that seen in AD. SD cases have progressive loss of previously established semantic knowledge (non-context specific fact, word and object knowledge), yet they are still able to learn new episodic information (Hodges et al., 1992). Thus, in the early stages of disease, a double dissociation in these two aspects of declarative memory can be shown between AD and SD. Preserved episodic memory acquisition in SD was initially interpreted as being due to a relative preservation of MTL structures in this syndrome. This hypothesis was challenged, however, when manual magnetic resonance imaging (MRI) studies found that hippocampal and parahippocampal – including entorhinal cortex – structures were at least as atrophic in SD, when compared to AD (Chan et al., 2001, Davies et al., 2004, Galton et al., 2001). An important caveat is that it was unclear whether the subjects included in these studies exhibited this double dissociation at the time of scanning.

A critical factor in attempting to identify neural explanations for the cognitive deficits seen in these disorders relates to the stage at which cases are studied. Although AD is characterized by episodic memory deficits, semantic memory impairment typically develops with disease progression (Hodges and Patterson, 1995). Likewise in advancing SD, episodic memory impairments are likely to occur although this can be difficult to assess as their semantic deficit may render them untestable on standard neuropsychological measures. However, pathological examinations in either disease are usually carried out in late, or end-stage cases.

Here, we studied cases that displayed the episodic/semantic dissociation at the time of scanning. Functional imaging with SPECT and PET has identified abnormalities in a variety of limbic areas in early AD (Callen et al., 2002, De Santi et al., 2001, Minoshima et al., 1997). In an attempt to understand the neural basis of episodic memory impairment in AD, we had previously found that cases with incipient AD (amnesic mild cognitive impairment, MCI), whose only significant neuropsychological deficit was in episodic memory, had hypometabolism restricted to a limbic–diencephalic network comprising the MTL (including hippocampus, dentate gyrus, subiculum and parahippocampal gyrus), mamillary bodies, dorso-mesial thalamus and posterior cingulate cortex (Nestor et al., 2003a). However, we had been unable to assess whether amnesia was a consequence of damage to any one, or to a combination, of these areas. Similarly, Desgranges et al. (2002) showed that episodic memory performance in mild AD (MMSE 23.8 ± 1.9) correlated with entorhinal, perirhinal and retrosplenial areas. However, these correlations cannot clarify whether the memory problem is caused by a lesion to any single region (with the remaining areas degenerating in concert), combined damage to all correlated areas or albeit less plausibly, damage to a wider and more complex network which happens to generate behavioral data that mimic the variance found in the above structures.

The present study had a two-fold aim. The first was to contrast the profile of brain damage in SD and AD cases that, at the time of scanning, exhibited the dissociation in declarative memory performance described above. The questions being whether (i) the SD group would have preserved MTL function compared to AD and (ii) if not, whether there would be a different profile of MTL-connected limbic network dysfunction in SD to that seen in AD. If there was MTL dysfunction in the present SD group, then the degree to which this was associated with damage to the remaining limbic network could inform understanding of the clinical relevance of lesions at these latter sites. The second aim was to map the extent of brain hypometabolism in an early SD group with FDG-PET. Previous MRI studies have highlighted rostral temporal lobe atrophy in SD (Chan et al., 2001, Davies et al., 2004, Galton et al., 2001); however, functional activation studies examining the neural basis of semantic knowledge in healthy subjects usually activate a more distributed network, notably including prefrontal and caudo-lateral temporal areas (Cabeza and Nyberg, 2000). In contrast, rostral temporal lobe activations in such studies are unusual. As FDG-PET is a highly sensitive marker of dysfunction in degenerative brain diseases, the question addressed here was whether hypometabolism beyond the rostral temporal lobes, would be evident or whether semantic knowledge breakdown in SD was due to rostral temporal lobe pathology in isolation.

Section snippets

Overview

In this study, we selected cases with AD and SD, who at the time of imaging exhibited the double dissociation between episodic and semantic memory. Cases underwent (18F)fluoro-2-deoxy-d-glucose positron emission tomography (FDG-PET) and volumetric MRI. PET data were analyzed with the voxel-based technique Statistical Parametric Mapping (SPM, Wellcome Department of Imaging Neuroscience) and region of interest methods. The latter method was specifically employed to study changes in small brain

Neuropsychology

The memory assessment of each group identified the double dissociation between performance on neuropsychological tests of episodic and semantic memory (Fig. 1). The SD group was significantly impaired on semantic tasks relative to AD with the converse profile being found on the episodic tasks. Compared to healthy, aged controls, the mean z score for the semantic battery was z = −12.17 ± 13.0 for the SD cases and z = −0.18 ± 0.6 for the AD cases (P = 0.02) while for the episodic battery the

Discussion

In this study, cases with SD and AD who, at the time of scanning, exhibited a double dissociation between impairments of new episodic memory learning and established semantic memory were shown to have comparable MTL atrophy and hypometabolism. The groups were, however, differentiated by the distribution of lesions beyond this region. In AD, there were metabolic deficits in limbic–diencephalic connections of the MTL – most significantly in the posterior cingulate region – while in SD, there was

Conclusion

In summary, we found that the episodic memory impairment heralding the onset of AD, is likely to be the result of a summation of damage to the limbic–diencephalic network but cannot be explained by focal MTL dysfunction. Within this network, it is possible that posterior cingulate dysfunction may be of greatest importance in generating the cognitive deficit. In the SD group, relative preservation of episodic memory was associated with relative preservation of this network and occurred in spite

References (67)

  • C. Pennanen et al.

    Hippocampus and entorhinal cortex in mild cognitive impairment and early AD

    Neurobiol. Aging

    (2004)
  • F. Remy et al.

    Verbal episodic memory impairment in Alzheimer's disease: a combined structural and functional MRI study

    NeuroImage

    (2005)
  • D. Tranel et al.

    Neural correlates of naming animals from their characteristic sounds

    Neuropsychologia

    (2003)
  • G.B. Williams et al.

    Neural correlates of semantic and behavioural deficits in frontotemporal dementia

    NeuroImage

    (2005)
  • A.D. Baddeley et al.

    The Doors and People Test: A Test of Visual and Verbal Recall and Recognition

    (1994)
  • H. Braak et al.

    Staging of Alzheimer's disease-related neurofibrillary changes

    Neurobiol. Aging

    (1995)
  • R. Cabeza et al.

    Imaging cognition II: an empirical review of 275 PET and fMRI studies

    J. Cogn. Neurosci.

    (2000)
  • D.J. Callen et al.

    Beyond the hippocampus: MRI volumetry confirms widespread limbic atrophy in AD

    Neurology

    (2001)
  • D.J. Callen et al.

    Limbic system perfusion in Alzheimer's disease measured by MRI-coregistered HMPAO SPET

    Eur. J. Nucl. Med. Mol. Imaging

    (2002)
  • D. Chan et al.

    Patterns of temporal lobe atrophy in semantic dementia and Alzheimer's disease

    Ann. Neurol.

    (2001)
  • G. Chételat et al.

    Dissociating atrophy and hypometabolism impact on episodic memory in mild cognitive impairment

    Brain

    (2003)
  • H. Damasio et al.

    A neural basis for lexical retrieval

    Nature

    (1996)
  • R.R. Davies et al.

    The human perirhinal cortex and semantic memory

    Eur. J. Neurosci.

    (2004)
  • R.R. Davies et al.

    The pathological basis of semantic dementia

    Brain

    (2005)
  • A. Delacourte et al.

    The biochemical pathway of neurofibrillary degeneration in aging and Alzheimer's disease

    Neurology

    (1999)
  • B. Desgranges et al.

    The neural substrates of episodic memory impairment in Alzheimer's disease as revealed by FDG-PET: relationship to degree of deterioration

    Brain

    (2002)
  • A.T. Du et al.

    Higher atrophy rate of entorhinal cortex than hippocampus in AD

    Neurology

    (2004)
  • N.C. Fox et al.

    Correlation between rates of brain atrophy and cognitive decline in AD

    Neurology

    (1999)
  • C.J. Galton et al.

    Differing patterns of temporal atrophy in Alzheimer's disease and semantic dementia

    Neurology

    (2001)
  • G.E. Garrido et al.

    Relation between medial temporal atrophy and functional brain activity during memory processing in Alzheimer's disease: a combined MRI and SPECT study

    J. Neurol. Neurosurg. Psychiatry

    (2002)
  • T.J. Grabowski et al.

    A role for left temporal pole in the retrieval of words for unique entities

    Hum. Brain Mapp.

    (2001)
  • K.S. Graham et al.

    Differentiating the roles of the hippocampal complex and the neocortex in long-term memory storage: evidence from the study of semantic dementia and Alzheimer's disease

    Neuropsychology

    (1997)
  • K.S. Graham et al.

    A reverse temporal gradient for public events in a single case of semantic dementia

    Neurocase

    (1998)
  • Cited by (319)

    View all citing articles on Scopus
    View full text