Highlights in Behavioral Neurology
2015 AAN Annual Meeting Research Presentations
At the recent Annual Meeting of the American Academy of Neurology, Kenneth Heilman from the University of Florida and Nicholas Milano from the Medical University of South Carolina presented “Highlights in Behavioral Neurology”. The objective of this session was to highlight and foster discussion of some of the most interesting and important research presented at the AAN meeting. The following is a summary of the abstracts presented.
Cerebral amyloidosis is associated with white matter hyperintensity independent of vascular risk factor history
Julia A. Scott1, Duygu Tosun-Turgut2, Meredith Braskie3, Paul Thompson3, Michael Weiner2, Charles DeCarli1, Owen T. Carmichael4
1University of California Davis; 2Veterans Administration Medical Center, San Francisco; 3University of Southern California; 4Louisiana State University, Pennington Biomedical Research Center
This study aimed to assess the association between cerebral amyloid burden and white matter hyperintensities (WMH) on magnetic resonance images in individuals with varying vascular risk profiles. Using participants from the Alzheimer’s disease Neuroimaging Initiative-2, this study compared WMH volumes with amyloid burden (measured with CSF and PET imaging) and vascular history. The authors found that greater amyloid burden and hypertension history were independently associated with greater WMH volume. Greater amyloid burden was even associated with WMH in individuals with no history of hypertension and normal current blood pressure. WMH are commonly found in patients with Alzheimer’s disease and have been thought of as a marker of cerebral small vessel disease. This study shows that amyloid may be contributing to the generation of WMH and should be considered in their interpretation.
The discussion about this paper focused on the possible causes of these WMH and if it was possible that these changes were caused by either amyloid angiopathy, or perhaps the toxic effect of amyloid on the oligodentrocytes. Furthermore, since alteration of the white matter may induce disconnections, and disconnection may be responsible for cognitive and conative deficits, future research should be directed at gaining a better understanding of the pathophysiology and treatment of these alterations in the white matter.
Neural correlates of action recognition and object knowledge in neurodegenerative disease
Miguel A. Santos1, Ignacio Illan-Gala2, Giulia Vinceti1,3, Maria Luisa Mandelli1, Honey Hubbard1, Zachary A. Miller1, Gil D. Rabinovici1, Bruce L. Miller1, Maria Luisa Gorno-Tempini1
1Memory and Aging Center, University of California San Francisco; 2Hospital Universitario La Paz, Madrid, Spain; 3University of Modena and Reggio Emilia
This study aimed to identify the neuroanatomical correlates of action recognition and its relation to object knowledge. Groups consisted of subjects with 1) Alzheimer’s disease (AD) spectrum neurodegenerative disease (early onset AD, logopenic variant primary progressive aphasia (PPA), posterior cortical atrophy), 2) Non-AD spectrum neurodegenerative disease (nonfluent variant PPA, corticobasal degeneration, progressive supranuclear palsy, behavioral variant frontotemporal dementia), 3) semantic variant PPA, and 4) healthy controls. Subjects completed 3 tasks: an action recognition task where they had to match a video of a pantomime (i.e., a hand using an imaginary key) with the correct object (answer: key), a word-picture matching task, and a hand matching task (match hands in different orientations). All subjects also received a 3T structural MRI of their brain, and a whole brain voxel-wise multiple linear regression analysis was performed to identify voxels highly correlated with performance on the action recognition task. The authors found that all groups were significantly worse than controls on the action recognition task and that it correlated with voxels in the bilateral frontal and left posterior temporal and inferior parietal lobes. When controlling for the word picture and hand matching tasks, only the AD spectrum group remained significantly different from controls on the action recognition task, and the task correlated with voxels in the left mid-posterior temporal and inferior parietal lobes. The authors concluded that left mid-posterior temporal and inferior parietal areas are critical for deriving meaning from dynamic gestures, and that AD spectrum disorders show particular impairment in this aspect of action recognition.
It was noted during the discussion of this paper that studies of patients with strokes have revealed that in righted handed patients the inferior parietal lobe appears to be critical in storing the representations of learned purposeful skilled actions, and patients with strokes that injure this area not only have ideomotor apraxia but also are impaired in gesture recognition and discrimination. Functional imaging studies have also shown that when performing transitive gestures the left inferior parietal lobe shows activation. Ideomotor apraxia, as well as a failure to recognize and discriminate gestures, can be a source of disability and unfortunately many tests for dementia, such as the Mini-Mental Status Exam, do not assess for these disabilities. In addition, there are few studies that have attempted to learn how to best rehabilitate these disorders.
Validation of olfactory deficit as a biomarker of AD
Chaitanya V. Amrutkar1, Matthew R. Woodward1, Harshit C. Shah1, Ralph Benedict1, Sanjanaa Rajakrishnan1, Rachelle S. Doody2, Li Yan3, Kinga Szigeti1
1Alzheimer’s Disease and Memory Disorders Center, Department of Neurology, University at Buffalo, SUNY; 2Alzheimer’s Disease and Memory Disorders Center, Department of Neurology, Baylor College of Medicine; 3Department of Bioinformatics, University at Buffalo, SUNY
An odor identification deficit (OID) has been associated with both AD and amnestic mild cognitive impairment (aMCI). This study aimed to determine a battery of smell identification tests which maximized sensitivity and specificity to detect early AD and aMCI. All subjects received the full University of Pennsylvania Smell Identification Test (UPSIT). The ROC area under the curve (AUC) for identifying an amnestic disorder (AD or aMCI), and AD alone, was similar for the 40 item UPSIT, the top 10 smells in this study, and the 10 item subset previously proposed by Tabert. This is important as the 10-item subtest decreases the time to administer the test by 75% to 5 minutes. Non-smeller status based on a cutoff of 7 (7 or less= nonsmeller) had a sensitivity of 74% and a specificity of 71% for identifying an amnestic disorder (AD or aMCI). In the longitudinal component of this study, 36.4% of non-smeller patients with aMCI converted to AD, while only 17.3% of patients with intact olfaction converted to AD. The ROC AUC for conversion to AD was 0.62, which is comparable to CSF biomarkers (0.63-0.67) and somewhat inferior to structural MRI (0.69-0.73). This study emphasizes that a short 10-item assessment of an OID is a potential cheap, noninvasive, and accurate biomarker for AD.
In the discussion of this paper it was noted how important it may be to have a simple test that could: 1) help to confirm the diagnosis of early AD and 2) could be given to patients with MCI to learn which patients would be most likely to develop AD. Unfortunately, there are other diseases associated with dementia, such as Parkinson’s disease, where patients may have a reduction of smell, and there are other causes of decreased smell which are not related to neurodegenerative disease. Stamps and Heilman reported that by using a simple peanut butter smell detection task, patients with mild AD, as well as some patients with MCI, had a greater reduction of smell in the left than the right nostril. Doty and his colleagues reported that they were not able to replicate these findings; however, these investigators studied patients with more severe AD, and with severe AD this asymmetry may decline or even be absent. Thus further studies are needed to learn if an asymmetrical reduction of smell sensitivity may be a valid test for early AD, as well as a predictor of those patients with MCI who may develop AD.
Psychological Stress and Age-Related Memory decline
Richard J. Caselli1, Amylou C. Dueck2, Dona Locke3, Bryan K. Woodruff1, Charlene Hoffman-Snyder1, Yonas Geda3
Departments of 1Neurology, 2Biostatistics, 3Psychology, and 3Psychiatry, Mayo Clinic Arizona
Chronic stress has been associated with a variety of functional and structural changes in the prefrontal and hippocampal areas of the brain, and has also been associated with dementia. This study aimed to assess the affect of stress vulnerability on age-related memory decline in subjects with and without the APOE e4 allele. Cognitively normal subjects were given the NEO Five Factor Personality Inventory, had APOE genotype testing, and completed a battery of neuropsychological tests ever 1-2 years. One of the components of the NEO Five Factory Personality Inventory is neuroticism. Questions related to neuroticism pertain to depression, anxiety, and anger, which reflect a proneness to stress. Subjects with a high level of neuroticism (clinical cutoff: 56) had a more rapid age-related memory decline, and this effect was more pronounced in APOE e4 carriers.
The discussion of this paper was focused on what may be causing the age related memory decline in patients with stress. One possibility raised was the role of glucocorticoids. Animal studies have revealed that with stress there is increased secretion of glucocorticoids via the hypothalamic-pituitary-adrenocortical axis and high levels of glucocorticoids may injure the neurons in the hippocampus. Based on this study it would be important to reduce stress when possible, and to treat people who are prone to stress using both psychotherapy as well as medications; however, it should be noted that some recent studies have suggested that some anxiolytic medications, such as the benzodiazepines, have an association with the development of dementia.
Differentiating the Logopenic and Nonfluent/Agrammatic Variants of Primary Progressive Aphasia by Selective Word Comprehension Impairment
Collin York, Christopher A. Olm, Sharon Ash, Katya Rascovsky & Murray Grossman
University of Pennsylvania Department of Neurology
This study aimed to characterize and investigate the basis for selective word class impairment in logopenic variant primary progressive aphasia (lvPPA) and nonfluent/agrammatic variant primary progressive aphasia (naPPA). Although both lvPPA and naPPA have relatively spared single word comprehension, naPPA may have action verb deficits. Subjects in 3 groups (lvPPA, naPPA, healthy controls) were given a two-alternative forced-choice lexical comprehension task. The task included verb triads (half motion verbs, half cognition verbs) and noun triads (half concrete nouns, half abstract nouns). For example, subjects were given a stimulus word such as ‘collect’, and then would have to match it with either ‘gather’ or ‘grab’. Subjects had MR imaging of their brains, which was followed by a voxel-based morphometry analysis of gray matter atrophy and a regression analysis relating motion verb accuracy to gray matter thickness. The authors found that subjects with naPPA performed worse than controls on motion verbs, and their motion verb performance correlated with grammatical comprehension and left inferior frontal atrophy. Subjects with lvPPA performed worse than controls on motion verbs and concrete nouns, and their motion verb performance correlated with category fluency and left inferior temporal and fusiform atrophy. The authors concluded that the impairment of motion verbs in naPPA was due to grammatical impairment as the left inferior frontal region is associated with grammatical processing. They hypothesized that the impairment of motion verbs in lvPPA may be related to the visual representation of motion verb knowledge, as it was associated with visual association cortex in the left inferior temporal and fusiform gyrus.
During the discussion of this paper it was noted that the stroke literature has revealed that lesions of the left inferior frontal lobe cause a Broca’s aphasia. Similar to naPPA, Broca’s aphasia is characterized by impaired fluency, and impairment in the expression and comprehension of syntax. Verbs predicate syntax and dysfunction of the left inferior frontal lobe, as in naPPA, may lead to impairment of the comprehension of syntax as well as verb knowledge. In contrast, the stroke literature has revealed that lesions to the posterior temporal lobe cause a Wernicke’s aphasia, which includes impaired comprehension of nouns and verbs due to destruction of the input lexicon. Subjects with lvPPA may have dysfunction of their input lexicon leading toimpaired performance on nouns and verbs. Alternatively visual imagery may also be important for the comprehension of nouns and verbs which is supported by the association with the visual association cortex seen in this study.