EDITORIAL
Connectivity at a crossroads What white matter integrity can tell us about cognitive impairment
Randolph S. Marshall, MD, MS Yael D. Reijmer, PhD
Correspondence to Dr. Marshall: [email protected] Neurology® 2014;83:296–297
Vascular cognitive impairment is the second most prevalent dementia after Alzheimer disease.1 With the aging of our population, its prevalence will continue to increase, creating an imperative to improve our understanding of this condition. In recent years, disruption of brain connectivity has emerged as a key component to explain cognitive dysfunction: the size and location of ischemic lesions are important, but so are the effects of ischemic injury on connections among functional brain regions.2 The idea that focal brain injury could have consequences beyond the locus of the lesion is not a new one. von Monakow first postulated the diaschisis effect in the early 1900s, and the idea of disconnection producing disruption in higher cognitive function was advanced by Geschwind,3 Mesulam,4 and others. In the modern era of large dataset analysis, the investigation of connectivity within the brain has taken a leap forward with advanced imaging methods. Functional connectivity studies have shown that cognitive functions such as information processing and executive functioning rely on the communication among widely dispersed brain regions that form large-scale networks. Disruption of the structural connections between and among regions, therefore, likely explains cognitive deficits resulting from vascular disease. Diffusion tensor imaging (DTI) can quantify microscopic white matter tissue injury not captured by classic MRI markers of small-vessel disease (SVD). The directionality of the diffusion obtained by DTI makes it possible to reconstruct individual fiber tracts and map the large-scale connectivity pattern of the brain (fiber tractography). Graph analysis, an advanced statistical method to analyze complex networks, can characterize the topologic organization of white matter connections by delineating a network as sets of interacting nodes (brain regions) connected by edges (white matter tracts).5 This approach offers the novel possibility of quantifying the efficiency with which the brain can integrate information among regions, taking into account the integrity as well as the organization of the white matter tracts. With this approach, previous studies have demonstrated that an efficient brain network is related to healthy brain
functioning,6 and that network efficiency is reduced in patients with mild cognitive impairment and dementia.7,8 Lawrence et al.9 in this issue of Neurology® have advanced our understanding of the association between structural connectivity and cognition by using MRI diffusion-based graph analysis to assess 115 patients with SVD compared with 50 strokefree controls. Whole-brain networks were identified by first parcellating the gray matter into 90 cortical and subcortical nodes. The white matter connections among nodes were then reconstructed with fiber tractography. The authors assessed the overall efficiency (defined as the inverse length of the shortest paths between nodes) of the brain network with graph analysis, and subsequently characterized the subnetwork of white matter connections that differed most between the patients with SVD and controls. The authors found that patients with SVD had overall reduced global and local network efficiency (EGlobal and ELocal) compared with controls. The subnetwork of “most impaired” connections was composed of interhemispheric connections of all major subdivisions of the corpus callosum as well as longrange intrahemispheric connections, including frontoparietal, frontotemporal, and temporoparietal connections. Network disruption in patients with SVD correlated highly with cognitive dysfunction, particularly with processing speed and executive function. Using mediation analysis, the authors showed that EGlobal fully mediated the relationship between MRI markers of white matter integrity (white matter hyperintensity, fractional anisotropy, mean diffusivity) and cognition, providing a potential mechanism through which SVD affects processing speed and executive functioning. The authors also pointed out that whereas the conventional MRI finding of white matter hyperintensity burden colocalized with the reduced corpus callosum connections, other regionally reduced connectivity, specifically the fronto-frontal, frontoparietal, and frontotemporal connections, did not overlap with white matter
See page 304 From Columbia University Medical Center (R.S.M.), New York; and Massachusetts General Hospital Stroke Center (Y.D.R.), Boston. Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the editorial. 296
© 2014 American Academy of Neurology
lesions seen with conventional MRI. Thus, the connectivity measures provided important anatomical and functional information that was not apparent from conventional MRI measures alone. In addition, the authors reported that the correlation of normalized brain volume and lacunar load to cognitive function was preserved when the white matter measures were corrected for, suggesting that these factors work by independent mechanisms, although the latter variable had a very low regression coefficient (,0.2) and therefore had a borderline independent effect. Lawrence et al. contribute to our understanding of how SVD might affect certain cognitive functions via white matter injury and its effect on distantly connected cortical regions. Although diffusion tractography has a resolution far lower than the actual axons it is meant to represent, structural connectivity imaging may still be useful as a biomarker in natural history studies of cognitive impairment, or as a means to follow patients longitudinally in treatment trials targeting white matter disease and cognition. The study has limitations. First, there were other basic network properties, such as mean edge density and mean network strength, which were also decreased in patients with SVD, making it difficult to determine which specific network alterations drove the associations with cognition. Second, the tractography method used in this study was not optimal. Newer methods are available to better resolve tracts in crossing fiber regions. Finally, for diffusion network methods to be used in clinical practice or clinical trials, we need to standardize and validate DTI-MRI acquisition and processing protocols, and collect normative data. This is an ongoing process with a number of challenges that need to be resolved.10 Overall, however, advances in structural connectivity analysis as demonstrated in this study are helping to shape a new landscape of investigation that will continue to elucidate the complex relationship between vascular brain injury and cognitive decline.
AUTHOR CONTRIBUTIONS Randolph S. Marshall: drafting/revising the manuscript, study concept or design. Yael Reijmer: drafting/revising the manuscript.
STUDY FUNDING No targeted funding reported.
DISCLOSURE The authors report no disclosures relevant to the manuscript. Go to Neurology.org for full disclosures.
REFERENCES 1. Gorelick PB, Scuteri A, Black SE, et al. Vascular contributions to cognitive impairment and dementia: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2011;42:2672–2713. 2. Reijmer YD, Freeze WM, Leemans A, Biessels GJ; Utrecht Vascular Cognitive Impairment Study Group. The effect of lacunar infarcts on white matter tract integrity. Stroke 2013;44:2019–2021. 3. Geschwind N. Disconnexion syndromes in animals and man. Brain 1965;88:237–294. 4. Mesulam MM. Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Ann Neurol 1990;28:597–613. 5. Rubinov M, Sporns O. Complex network measures of brain connectivity: uses and interpretations. Neuroimage 2010;52:1059–1069. 6. Li Y, Liu Y, Li J, et al. Brain anatomical network and intelligence. PLoS Comput Biol 2009;5:e1000395. 7. Greicius MD, Kimmel DL. Neuroimaging insights into network-based neurodegeneration. Curr Opin Neurol 2012;25:727–734. 8. Reijmer YD, Leemans A, Caeyenberghs K, Heringa SM, Koek HL, Biessels GJ; Utrecht Vascular Cognitive Impairment Study Group. Disruption of cerebral networks and cognitive impairment in Alzheimer disease. Neurology 2013;80:1370–1377. 9. Lawrence AJ, Chung AW, Morris RG, Markus HS, Barrick TR. Structural network efficiency is associated with cognitive impairment in small-vessel disease. Neurology 2014;83:304–311. 10. Fornito A, Zalesky A, Breakspear M. Graph analysis of the human connectome: promise, progress, and pitfalls. Neuroimage 2013;80:426–444.
Neurology 83
July 22, 2014
297
Connectivity at a crossroads: What white matter integrity can tell us about cognitive impairment Randolph S. Marshall and Yael D. Reijmer Neurology 2014;83;296-297 Published Online before print June 20, 2014 DOI 10.1212/WNL.0000000000000627 This information is current as of June 20, 2014 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/83/4/296.full.html
References
This article cites 10 articles, 4 of which you can access for free at: http://www.neurology.org/content/83/4/296.full.html##ref-list-1
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 2014 American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
Connectivity at a crossroads What white matter integrity can tell us about cognitive impairment
Randolph S. Marshall, MD, MS Yael D. Reijmer, PhD
Correspondence to Dr. Marshall: [email protected] Neurology® 2014;83:296–297
Vascular cognitive impairment is the second most prevalent dementia after Alzheimer disease.1 With the aging of our population, its prevalence will continue to increase, creating an imperative to improve our understanding of this condition. In recent years, disruption of brain connectivity has emerged as a key component to explain cognitive dysfunction: the size and location of ischemic lesions are important, but so are the effects of ischemic injury on connections among functional brain regions.2 The idea that focal brain injury could have consequences beyond the locus of the lesion is not a new one. von Monakow first postulated the diaschisis effect in the early 1900s, and the idea of disconnection producing disruption in higher cognitive function was advanced by Geschwind,3 Mesulam,4 and others. In the modern era of large dataset analysis, the investigation of connectivity within the brain has taken a leap forward with advanced imaging methods. Functional connectivity studies have shown that cognitive functions such as information processing and executive functioning rely on the communication among widely dispersed brain regions that form large-scale networks. Disruption of the structural connections between and among regions, therefore, likely explains cognitive deficits resulting from vascular disease. Diffusion tensor imaging (DTI) can quantify microscopic white matter tissue injury not captured by classic MRI markers of small-vessel disease (SVD). The directionality of the diffusion obtained by DTI makes it possible to reconstruct individual fiber tracts and map the large-scale connectivity pattern of the brain (fiber tractography). Graph analysis, an advanced statistical method to analyze complex networks, can characterize the topologic organization of white matter connections by delineating a network as sets of interacting nodes (brain regions) connected by edges (white matter tracts).5 This approach offers the novel possibility of quantifying the efficiency with which the brain can integrate information among regions, taking into account the integrity as well as the organization of the white matter tracts. With this approach, previous studies have demonstrated that an efficient brain network is related to healthy brain
functioning,6 and that network efficiency is reduced in patients with mild cognitive impairment and dementia.7,8 Lawrence et al.9 in this issue of Neurology® have advanced our understanding of the association between structural connectivity and cognition by using MRI diffusion-based graph analysis to assess 115 patients with SVD compared with 50 strokefree controls. Whole-brain networks were identified by first parcellating the gray matter into 90 cortical and subcortical nodes. The white matter connections among nodes were then reconstructed with fiber tractography. The authors assessed the overall efficiency (defined as the inverse length of the shortest paths between nodes) of the brain network with graph analysis, and subsequently characterized the subnetwork of white matter connections that differed most between the patients with SVD and controls. The authors found that patients with SVD had overall reduced global and local network efficiency (EGlobal and ELocal) compared with controls. The subnetwork of “most impaired” connections was composed of interhemispheric connections of all major subdivisions of the corpus callosum as well as longrange intrahemispheric connections, including frontoparietal, frontotemporal, and temporoparietal connections. Network disruption in patients with SVD correlated highly with cognitive dysfunction, particularly with processing speed and executive function. Using mediation analysis, the authors showed that EGlobal fully mediated the relationship between MRI markers of white matter integrity (white matter hyperintensity, fractional anisotropy, mean diffusivity) and cognition, providing a potential mechanism through which SVD affects processing speed and executive functioning. The authors also pointed out that whereas the conventional MRI finding of white matter hyperintensity burden colocalized with the reduced corpus callosum connections, other regionally reduced connectivity, specifically the fronto-frontal, frontoparietal, and frontotemporal connections, did not overlap with white matter
See page 304 From Columbia University Medical Center (R.S.M.), New York; and Massachusetts General Hospital Stroke Center (Y.D.R.), Boston. Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the editorial. 296
© 2014 American Academy of Neurology
lesions seen with conventional MRI. Thus, the connectivity measures provided important anatomical and functional information that was not apparent from conventional MRI measures alone. In addition, the authors reported that the correlation of normalized brain volume and lacunar load to cognitive function was preserved when the white matter measures were corrected for, suggesting that these factors work by independent mechanisms, although the latter variable had a very low regression coefficient (,0.2) and therefore had a borderline independent effect. Lawrence et al. contribute to our understanding of how SVD might affect certain cognitive functions via white matter injury and its effect on distantly connected cortical regions. Although diffusion tractography has a resolution far lower than the actual axons it is meant to represent, structural connectivity imaging may still be useful as a biomarker in natural history studies of cognitive impairment, or as a means to follow patients longitudinally in treatment trials targeting white matter disease and cognition. The study has limitations. First, there were other basic network properties, such as mean edge density and mean network strength, which were also decreased in patients with SVD, making it difficult to determine which specific network alterations drove the associations with cognition. Second, the tractography method used in this study was not optimal. Newer methods are available to better resolve tracts in crossing fiber regions. Finally, for diffusion network methods to be used in clinical practice or clinical trials, we need to standardize and validate DTI-MRI acquisition and processing protocols, and collect normative data. This is an ongoing process with a number of challenges that need to be resolved.10 Overall, however, advances in structural connectivity analysis as demonstrated in this study are helping to shape a new landscape of investigation that will continue to elucidate the complex relationship between vascular brain injury and cognitive decline.
AUTHOR CONTRIBUTIONS Randolph S. Marshall: drafting/revising the manuscript, study concept or design. Yael Reijmer: drafting/revising the manuscript.
STUDY FUNDING No targeted funding reported.
DISCLOSURE The authors report no disclosures relevant to the manuscript. Go to Neurology.org for full disclosures.
REFERENCES 1. Gorelick PB, Scuteri A, Black SE, et al. Vascular contributions to cognitive impairment and dementia: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2011;42:2672–2713. 2. Reijmer YD, Freeze WM, Leemans A, Biessels GJ; Utrecht Vascular Cognitive Impairment Study Group. The effect of lacunar infarcts on white matter tract integrity. Stroke 2013;44:2019–2021. 3. Geschwind N. Disconnexion syndromes in animals and man. Brain 1965;88:237–294. 4. Mesulam MM. Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Ann Neurol 1990;28:597–613. 5. Rubinov M, Sporns O. Complex network measures of brain connectivity: uses and interpretations. Neuroimage 2010;52:1059–1069. 6. Li Y, Liu Y, Li J, et al. Brain anatomical network and intelligence. PLoS Comput Biol 2009;5:e1000395. 7. Greicius MD, Kimmel DL. Neuroimaging insights into network-based neurodegeneration. Curr Opin Neurol 2012;25:727–734. 8. Reijmer YD, Leemans A, Caeyenberghs K, Heringa SM, Koek HL, Biessels GJ; Utrecht Vascular Cognitive Impairment Study Group. Disruption of cerebral networks and cognitive impairment in Alzheimer disease. Neurology 2013;80:1370–1377. 9. Lawrence AJ, Chung AW, Morris RG, Markus HS, Barrick TR. Structural network efficiency is associated with cognitive impairment in small-vessel disease. Neurology 2014;83:304–311. 10. Fornito A, Zalesky A, Breakspear M. Graph analysis of the human connectome: promise, progress, and pitfalls. Neuroimage 2013;80:426–444.
Neurology 83
July 22, 2014
297
Connectivity at a crossroads: What white matter integrity can tell us about cognitive impairment Randolph S. Marshall and Yael D. Reijmer Neurology 2014;83;296-297 Published Online before print June 20, 2014 DOI 10.1212/WNL.0000000000000627 This information is current as of June 20, 2014 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/83/4/296.full.html
References
This article cites 10 articles, 4 of which you can access for free at: http://www.neurology.org/content/83/4/296.full.html##ref-list-1
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 2014 American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
