UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The relationship between white matter lesion volume with vascular stiffness, mobility and cognitive performance… Al Keridy, Walid Ahmed 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2018_september_alkeridy_walid.pdf [ 4.05MB ]
Metadata
JSON: 24-1.0369732.json
JSON-LD: 24-1.0369732-ld.json
RDF/XML (Pretty): 24-1.0369732-rdf.xml
RDF/JSON: 24-1.0369732-rdf.json
Turtle: 24-1.0369732-turtle.txt
N-Triples: 24-1.0369732-rdf-ntriples.txt
Original Record: 24-1.0369732-source.json
Full Text
24-1.0369732-fulltext.txt
Citation
24-1.0369732.ris

Full Text

    THE RELATIONSHIP BETWEEN WHITE MATTER LESION VOLUME WITH VASCULAR STIFFNESS, MOBILITY AND COGNITIVE PERFORMANCE IN OLDER ADULTS WITH VASCULAR COGNITIVE IMPAIRMENT    by  Walid Ahmed Al Keridy  MBBS King Saud University      A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF      MASTER OF SCIENCE   in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES   (EXPERIMENTAL MEDICINE)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   July 2018     © Walid Ahmed Al Keridy, 2018   ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis/dissertation entitled:  THE RELATIONSHIP BETWEEN WHITE MATTER LESION VOLUME WITH VASCULAR STIFFNESS, MOBILITY AND COGNITIVE PERFORMANCE IN OLDER ADULTS WITH VASCULAR COGNITIVE IMPAIRMENT  submitted by Walid Ahmed Al Keridy  in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE in EXPERIMENTAL MEDICINE  Examining Committee: Teresa Liu-Ambrose Supervisor  Ging-Yuek Robin Hsiung  Supervisory Committee Member  Kenneth Madden Supervisory Committee Member Thalia Field Additional Examiner       iii  Abstract  Background: Sub-cortical ischemic vascular cognitive impairment (SIVCI) is the most prevalent form of Vascular Cognitive Impairment (VCI). Previous studies have shown that Carotid-Femoral Pulse Wave Velocity (CF-PWV), cognitive function, mobility performance and blood pressure are related to White Matter Lesions (WMLs) volume, but whether these associations exist in those with mild SIVCI is unclear.   Thus, in this study of older adults with mild SIVCI, I examined four questions: 1) What is the association between total, deep and periventricular WMLs volume on Magnetic Resonance Imaging (MRI) with CF-PWV? 2) What are the associations between total, deep and periventricular WMLs volume and measures of global cognitive function and executive function? 3) What is the association between total, deep and periventricular WMLs volume with mobility performance?  4) What is the association between total, deep and periventricular WMLs volume with systolic and diastolic blood pressure?  Methods: The data of 34 participants diagnosed with mild SIVCI were used for this cross-sectional analysis. Measures of interest included global cognitive function tests, such as Alzheimer’s disease Assessment Scale Cognitive Subscale (ADAS-cog) and Montreal Cognitive Assessment (MoCA). Executive functions were assessed with paper Stroop test, Trails Making Tests (A&B) and animal fluency. Mobility performance was assessed with the Time Up and Go test (TUG) and usual gait speed. CF-PWV was measured using the Complior System (ALAM   iv  Medical, France). WMLs volume was quantified for the total, deep and periventricular WMLs with a semi-automated technique.  Results: We did not find an association between total or periventricular WMLs volume with CF-PWV, executive function, global cognitive function, mobility performance or blood pressure. Deep WMLs volume was associated with Trails B-A and animal category fluency but not associated with measures of global cognitive function. Deep WMLs volume was significantly associated with diastolic blood pressure.   Conclusion: In this exploratory analysis, deep WMLs volume is associated with executive function and diastolic blood pressure. While systolic hypertension has been strongly linked to large vessel stroke events which are commonly complicated by post stroke cognitive impairment and dementia, our findings provide possible mechanistic insight to previous studies that found association between diastolic blood pressure and cognitive decline.   v  Lay Summary Dementia impacts millions of people worldwide and comes with enormous public health cost. Recent estimates projected about 1 trillion US dollars of economic and societal cost. Cognitive impairment caused by diseased blood vessels in the brain called “vascular cognitive impairment” is second most common cause of dementia. Our work aims to better understand how changes in the brain’s white matter secondary to vascular cognitive impairment relate to cognitive and mobility performance.  vi  Preface The data used in this project were provided by Professor Teresa Liu-Ambrose, PhD, PT from the “Reshaping the Path of Vascular Cognitive Impairment with Resistance Training” (RVCI; UBC CREB #H15-00972), funded by Heart and Stroke Foundation of Canada.   Associate Professors Robin Hsiung and Ken Madden are both co-investigators of the RVCI randomized controlled trial.   My role in the study was to assess and rate the presence of WMLs on Magnetic Resonance Imaging (MRI) using visual rating scale as well as identifying lesions to quantify WMLs volumes using a semi-automated analyses pipeline under the guidance of Dr. Roger Tam, UBC Multiple Sclerosis Research Group and co-investigator of the RVCI trial. I also acquired measures of CF-PWV. With guidance from my research committee members, I developed the four research questions addressed in this thesis.  Research Associate, Dr. John Best, provided the statistical analysis of the data and helped with creating the tables and figures in the results chapter.               vii  List of contents   Abstract .......................................................................................................................................... iii Lay Summary .................................................................................................................................. v Preface............................................................................................................................................ vi List of contents .............................................................................................................................. vii List of tables ................................................................................................................................... xi List of figures ................................................................................................................................ xii List of abbreviation ...................................................................................................................... xiv Acknowledgements ....................................................................................................................... xv Chapter 1: Introduction ............................................................................................................... 1 Chapter 2: Literature review and background ............................................................................. 5 2.1 Vascular Cognitive Impairment (VCI) ......................................................................... 5 2.2 White matter hyperintensity ......................................................................................... 6  Overview ............................................................................................................... 6  Historical background of WMLs........................................................................... 7  Epidemiology of WMLs........................................................................................ 9  Pathophysiology of WMLs ................................................................................. 11  Pathological findings of WMLs in SIVCI........................................................... 13 2.3 Association of white matter hyperintensity and cognition ......................................... 15 2.4 Association of WMLs and mobility performance ...................................................... 17   viii  2.5 Assessment of WMLs ................................................................................................. 19  Qualitative assessment of WMLs ........................................................................ 19  Quantitative assessment of WMLs ...................................................................... 20 2.6 Clinical presentation of VCI ....................................................................................... 22  Post-stroke dementia ........................................................................................... 22  Multi-infarct dementia......................................................................................... 23  Subcortical Ischemic Vascular Dementia (SIVaD) ............................................. 24  Mixed dementias ................................................................................................. 24  Clinical presentation of VCI  based on the recent VICCCS study 4 ................... 25 2.7 Cortical vs sub-cortical presentation .......................................................................... 26  Sub-cortical presentation ..................................................................................... 26  Cortical presentation ........................................................................................... 27 2.8 Arterial stiffness and PWV ......................................................................................... 28  Overview of arterial stiffness and PWV ............................................................. 28  Pathophysiology of arterial stiffness ................................................................... 29  Assessment of PWV ............................................................................................ 30  The PWV in predicting cardiovascular events and clinical outcomes ................ 32 Chapter 3: Methods ................................................................................................................... 35 3.1 Introduction ................................................................................................................ 35 3.2 Randomization ............................................................................................................ 36   ix  3.3 Inclusion and exclusion criteria for the RVCI trial .................................................... 36 3.4 Measurements ............................................................................................................. 38 3.5 Assessment of executive function and global cognitive function .............................. 38 3.6 Mobility measures ...................................................................................................... 40 3.7 Blood pressure measurement ...................................................................................... 41 3.8 PWV measurement ..................................................................................................... 41 3.9 MRI WMLs volume assessment................................................................................. 43  Visual rating for WMLs ...................................................................................... 43  Automated WMLs assessment ............................................................................ 45 3.10 Statistical analyses ...................................................................................................... 47 Chapter 4: Results ..................................................................................................................... 48 4.1 Participants characteristic ........................................................................................... 48 4.2 Correlation between the two independent raters for WMLs ...................................... 50 4.3 Correlations of measures of interest with total, deep and periventricular WMLs volume ................................................................................................................................... 52 4.4 Correlation between total, deep and periventricular WMLs volume and CF-PWV .. 52 4.5 The association between total, deep and periventricular WMLs volume and measures of global cognitive function and executive function ............................................................. 52 4.6 The association between total, deep and periventricular WMLs volume with mobility performance (TUG, gait speed) ............................................................................................. 53   x  4.7 The association between total, deep and periventricular WMLs volume with systolic and diastolic blood pressure .................................................................................................. 53 Chapter 5: Discussion................................................................................................................ 64 Chapter 6: Conclusion ............................................................................................................... 76 References ..................................................................................................................................... 78 Appendices .................................................................................................................................... 88     xi  List of tables Table 2-1 Clinical presentation of VCI ......................................................................................... 25 Table 3-1 Periventricular WMLs scoring using Fazekas scale ..................................................... 43 Table 3-2 Deep WMLs scoring using Fazekas scale .................................................................... 43 Table 4-1 Demographic data and participants characteristics ...................................................... 49 Table 4-2 Correlation between the two independent raters for WMLs. ....................................... 50 Table 4-3 Correlations of measures of interest with total, deep and periventricular WMLs volume........................................................................................................................................... 54 Table 4-4 Correlation of measures of interest with total, deep and periventricular WMLs volume controlling for age, sex and antihypertensive medication. ........................................................... 55 Table 6-1 Fazekas score ................................................................................................................ 88 Table 6-2 Antihypertensive use .................................................................................................... 88 Table 6-3 Baseline WMLs volumes for the RVCI participants .................................................... 89       xii  List of figures  Figure 2-1 Periventricular vs Deep WMLs ..................................................................................... 9 Figure 2-2 White matter blood supply permission granted from author Prof. Joanna M Wardlaw MD 55 ............................................................................................................................................ 12 Figure 3-1 Screenshot of Complior device results including PWV. ............................................. 42 Figure 3-2 Valid CF-PWV ............................................................................................................ 42 Figure 3-3 Visual rating for WMLs with Fazekas scale ............................................................... 44 Figure 4-1 Correlation between the two independent WMLs raters. ............................................ 51 Figure 4-2  Association of CF-PWV with total, periventricular and deep WMLs volume. ......... 56 Figure 4-3 Association of Trails B minus trail A with total, periventricular and deep WMLs volume........................................................................................................................................... 57 Figure 4-4 Association of Animal fluency scores with total, periventricular and deep WMLs volume........................................................................................................................................... 58 Figure 4-5 Association of gait speed  with total, periventricular and deep WMLs volume. ........ 59 Figure 4-6 Association of Time Up and Go test “TUG” with total, periventricular and deep WMLs volume. ............................................................................................................................. 60 Figure 4-7 Association of systolic blood pressure with total, periventricular and deep WMLs volume........................................................................................................................................... 61 Figure 4-8 Association of diastolic blood pressure with total, periventricular and deep WMLs volume........................................................................................................................................... 62 Figure 4-9 Association of age with total, periventricular and deep WMLs volume..................... 63 Figure 6-1 Total WMLs volume distribution for the RVCI cohort. ............................................. 90 Figure 6-2 Log transformation for total WMLs. ........................................................................... 91   xiii  Figure 6-3 Distribution of deep WMLs. ....................................................................................... 92 Figure 6-4 Log transformation of deep WMLs. ............................................................................ 93 Figure 6-5 Distribution of periventricular WMLs ........................................................................ 94 Figure 6-6 Log transformation of periventricular WMLs. ............................................................ 95 Figure 6-7 ADAS –Cog distribution for the RVCI cohort. .......................................................... 96 Figure 6-8 MoCA scores distribution for the RVCI cohort. ......................................................... 97 Figure 6-9 Log transformation for the Trails B-A ........................................................................ 98 Figure 6-10 Log transformation for Trails part A. ........................................................................ 99 Figure 6-11 Average TUG .......................................................................................................... 100 Figure 6-12 Distribution of gait speed. ....................................................................................... 101 Figure 6-13 Log transformation of delta Stroop. ........................................................................ 102 Figure 6-14 Distribution of animal fluency test. ......................................................................... 103 Figure 6-15 Distribution of systolic blood pressure ................................................................... 104 Figure 6-16 Distribution of diastolic blood pressure .................................................................. 105 Figure 6-17 Association of Trails A with total, periventricular and deep WMLs volume. ........ 106 Figure 6-18 Association of delta Stroop with total, periventricular and deep WMLs volume. .. 107 Figure 6-19 Association of ADAS-cog with total, periventricular and deep WMLs volume. ... 108 Figure 6-20 Association of MoCA scores with total, periventricular and deep WMLs volume. 109   xiv  List of abbreviation AD Alzheimer’s Disease ADAS-cog Global cognitive screening tests; Alzheimer’s Disease Assessment Scale Cognitive Subscale AHA American Heart Association (AHA) BAT Balance and Tone Training CAA Cerebral Amyloid Angiopathy CADASIL Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy CERAD Consortium to Establish a Registry for Alzheimer's Disease  CF-PWV Carotid-Femoral Pulse Wave Velocity  CNS Central Nervous System CT Computed Tomography  CV Cardiovascular  DLB Dementia of Lewy Body DWI Diffusion-Weighted Imaging  DWMH Deep White Matter Hyperintensity  FLAIR FLuid Attenuated Inversion Recovery  HTN HyperTeNsion ICH IntraCerebral Hemorrhage  MCI Mild Cognitive impairment  MID Multi Infarct Dementia mm3 Cubic millimeter  MMSE Mini Mental State Examination MoCA Montreal Cognitive Assessment  MRI Magnetic Resonance Imaging  PD Proton Density  PSD Post-Stroke Dementia  PWV Pulse Wave Velocity RT Resistance Training exercise   RVCI Reshaping the Path of Vascular Cognitive Impairment with Resistance Training  SAH SubArachnoid Hemorrhage SIVaD Subcortical Ischemic Vascular Dementia SIVCI Sub-Cortical Ischemic Vascular Cognitive Impairment  SPPB Short Physical Performance Battery  SVM Support Vector Machines  TUG Time Up and go Test VaD Vascular Dementia  VASCOG International Society for Vascular Behavioral and Cognitive Disorders criteria for vascular cognitive disorder VCI Vascular Cognitive Impairment  VICCCS Vascular Impairment of Cognition Classification Consensus Study WMLs White Matter Lesions   xv  Acknowledgements    Professor Teresa Ambrose   Dr. John Best  Dr. Robin Hsiung   My family    1  Chapter 1: Introduction   Dementia is a major public health concern with around 36 million people worldwide currently living with dementia 1. Caring for dementia comes at an enormous cost, with an estimated global economic cost of about 818 billion US (United States) dollars 1. The cost is estimated to surpass 2 trillion US dollars in 2030 1.   Vascular Dementia (VaD) is the second most common type of dementia 2. Vascular Cognitive Impairment (VCI) is a term used to describe the wide variety of vascular pathology causing cognitive impairment 3-5. VCI is a heterogeneous group of diseases caused by different underlying pathophysiological mechanisms 3,6,7. Sub-Cortical Ischemic Vascular Cognitive Impairment (SIVCI) is one of the main categories of VCI 4. White Matter Lesions (WMLs) and lacunar infarcts on MRI are the associated findings attributed to small vessel disease causing SIVCI 3. Ischemia related to small vessel disease is thought to be one of the main underlying pathological causes of WMLs in SIVCI 3,8,9. However, this view was challenged by recent evidence implicating non-vascular pathological changes as a major contributor to WMLs development 10.   As we age the large conduit arteries such as the aorta becomes stiffer and less compliant. These changes are associated with age-related changes and atherosclerosis 11. The term aortic stiffness or arterial stiffness is being used interchangeably for the past years 12. Aortic stiffness is usually defined as “the elastic resistance to deformation” 11. Pulse Wave Velocity (PWV) is considered a surrogate marker for arterial stiffness as higher arterial stiffness is directly proportional to PWV   2  13,14. WMLs have been found to correlate with cardiovascular measures such as hypertension and PWV. PWV predicts cardiovascular events, cardiovascular mortality and overall mortality 15.  Although PWV is thought to reflect arterial stiffness in large conduit arteries, it has been found to also correlate with cerebral small vessel dysfunction 16,17. The Rotterdam scan study found WMLs volume to be associated with pulse wave velocity, but they did not look at the different distribution of WMLs 18.The association of PWV with the different anatomical distribution of WMLs volume is still controversial and not well understood 19-21. Finding an association between cardiovascular measures such as PWV with different locations of WMLs could help us understand the nature of these lesions. It could also be used as a marker of disease progression or to monitor the response to therapeutic interventions targeting vascular risk factors. The first aim of this thesis is to study the association of total, deep and periventricular WMLs volume on MRI with Carotid-Femoral Pulse Wave Velocity (CF-PWV).   VCI has various etiological causes and present with different clinical and cognitive symptoms 4. In particular, SIVCI is commonly associated with executive dysfunction 22. Executive function is a construct that includes a multitude of cognitive processes such as problem solving, planning, working memory and mental flexibility 23. Studies have found an overall association of total WML and executive dysfunction 24.   Some studies found a significant association between periventricular WMLs volume with measures of global cognition and executive function 25. Other studies found that deep WMLs volume is associated with executive function, creating controversy around our understanding of   3  this area 26,27. Therefore, the second aim of this thesis is to study the associations between total, deep and periventricular WMLs volume and measures of global cognitive function and executive function.  Mobility is defined by the World Health Organization’s International Classification of Functioning, Disability and Health (ICF) as: “Moving and changing body position or location or by transferring from one place to another, by carrying, moving or manipulating objects, by walking, running or climbing, and by using various forms of transportation” 28. Total WMLs volume is linked to impaired mobility 29. Mobility measures can predict health outcomes and functional decline 30. Mobility performance is commonly assessed by many validated measures including the Short Physical Performance Battery (SPPB), TUG, Tinetti scale and gait speed 31. WMLs can negatively impact neural networks that are important for proper mobility performance 29,32,33. Therefore, individuals with significant WMLs load usually present with gait abnormalities. The longitudinal Three-City study by Soumare et al involving 1,086 community dwelling individuals completed in France, found that only periventricular WMLs volume but deep WMLs volume to be associated with slower walking speed 29. Identifying mobility measures that are significantly associated with WMLs volume and distribution in a cohort with mild SIVCI may have important clinical implications. They can be used to track disease progression or assess response to therapeutic interventions targeting WMLs. Therefore, the third aim is to study the association of mobility performance measures with total, deep and periventricular WMLs.     4  Systolic hypertension is thought to be closely related to arterial stiffness and atherosclerosis involving large arteries whereas diastolic hypertension is related to small vessel disease and peripheral vascular resistance 34-36. The Northern Manhattan study, a large prospective longitudinal study showed that diastolic blood pressure is closely associated with white matter hyperintensity volume in contrast to systolic blood pressure 36. On the other hand, the study by Van Dijk et al found that systolic hypertension is associated with periventricular WMLs volume and reduced diastolic blood pressure can in fact increase WMLs volume 37. So there is still gap in our understanding of the association between blood pressure and different anatomical distribution of WMLs. Finding a relationship between systolic or diastolic blood pressure and different distribution of WMLs (deep vs periventricular) could help us better understand the mechanism by which WMLs develop, therefore we can develop therapeutic intervention addressing those mechanisms. The fourth aim is to study the association between total, deep and periventricular WMLs volume with systolic and diastolic blood pressure.    5  Chapter 2: Literature review and background  2.1 Vascular Cognitive Impairment (VCI)  Cognitive impairment and dementia are significant and growing public health issues. It is estimated that 115 million people will live with dementia by 2050 worldwide 38. Currently, it is estimated that about 36 million people are living with dementia 38.   VCI is a broad term that comprises multiple etiologies including vascular dementia which is the second most common cause of dementia, accounting for 15% of all cases 39,40. VCI is a heterogeneous group of disorders, therefore, has a wide range of presentations including cognitive decline following a stroke, as well as Mild Cognitive Impairment (MCI) associated with small vessel disease involving the subcortical white matter 41. The cerebral small vessel network dysfunction has been implicated in causing SIVaD, the most common subset of VCI, but the exact mechanism is poorly understood 4,42. VCI also includes CAA (Cerebral Amyloid Angiopathy) and genetic causes such as CADASIL (Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy) 4.   Although many of these causes share similar risk factor profile, they mostly have a distinct pathophysiological mechanism. For example, the large vessel stroke leading to cognitive decline is mainly caused by atherosclerosis or artery to artery embolism and embolism secondary to atrial fibrillation 4. The subcortical ischemic vascular lesions are closely related to small vessel disease and pathologically manifests as lipohyalinosis and arteriosclerosis.   6  Recently, a significant move towards defining VCI has been attempted through the Vascular Impairment of Cognition Classification Consensus Study (VICCCS), and the final consensus was to use the term VCI to describe a broad spectrum of cognitive decline related to vascular pathologies 4. The VICCCS proposed specific parameters to further categories VCI which included the following factors; Location, Etiology, Domains (affected), and Severity, provisionally named “LEDS” criteria 4. These criteria were utilized by participants in the VICCCS study to allow the selection of current VCI subcategories based on previously proposed categories of VCI such as the work of O’Brien et al. 4,7.  2.2 White matter hyperintensity   Overview White matter is the area that lies just beneath the gray matter of the cortex. It is made up of millions of nerve fibers. These nerve fibers connect neurons (cell bodies) located in widely distributed areas in the brain regions, creating distinct functional circuits. The myelin sheath that wraps the nerve fibers is what gives rise to the white color associated with white matter structure. Myelin facilitates fast speed electrical impulse transmission within the functional circuits. Insults affecting myelin lead to impairment in electrical conduction which in turn cause malfunction of cognitive, motor and sensory circuits 43.  WMLs became increasingly recognized by clinicians following the widespread use of MRI 44. These lesions commonly appear on T2weighted images as hyperintense foci 44. They are very common in the general adult population with a prevalence of about 94% for people over the age   7  of 80 years 44.  WMLs are also found to be significantly associated with neurodegenerative diseases such as Alzheimer’s disease (AD) and Lewy body dementia 45.    Historical background of WMLs White matter changes on Computed Tomography (CT) were recognized by Hachinski and colleagues 46 in the late 1980s 46. The term leukoaraiosis was introduced to describe the irregular low attenuation surrounding the lateral ventricles (periventricular or in deep white matter tracts named deep white matter leukoaraiosis) 46.  The irregular white matter changes appear much clearer to the observer using specific sequences examined using MRI 46. The sequences commonly used are fluid-attenuated inversion recovery (FLAIR), T2-weighted (T2) and proton density weighted (PD). The abnormal change in signal intensity seen on MRI reflects the very fine changes in water content of nerve fibers 46.  Deep gray matter can be affected as well, but these signal hyperintensities in the white matter were much more common in the periventricular or deep white matter, so they are often referred to as “white matter hyperintensities”.They appear hyperintense on FLAIR, T2 and  PD sequences, however, they look hypointense on T1-weighted sequences 46.  Previous studies looking into white matter hyperintensities may have been affected by the low sensitivity or fewer sequences produced by the older generation as in 0.2 or 0.5 Tesla field strength MRI compared to 1.5 or 3 T MRI currently widely available 46.   The ability to study the correlation between imaging abnormalities with pathological changes of WMLs has been limited by this poor sensitivity of early CT and less sensitive MRI studies, when   8  mostly advanced cases of white matter pathologies were considered by pathological studies until more sensitive MRI came into practice leading to the widespread interest in white matter changes across many disciplines 46. This also might have affected other types of imaging studies such as imaging-cognitive and imaging-clinical studies potentially overlooking the subtler changes being picked up by newer generation MRI scanners 46. It potentially could have biased our understanding of the pathophysiology of these lesions towards the more advanced lesions promptly identified by pathologists such as lacunar strokes 46.   Lacunar infarcts are described as small  (0.2 to 15 mm in diameter) non-cortical infarctions secondary to occlusion of blood supply mainly a single penetrating blood vessel 47. Classification of WMLs into deep and periventricular were first put forward by Fazekas et al. 48. Periventricular WMLs were defined as WMLs in continuity with the borders of the lateral ventricles of the brain 48,49. On the other hand, deep WMLs are all the other WMLs not contiguous with the ventricles and isolated from the periventricular WMLs 48,49. Confluent lesions are usually used to describe large deep WMLs lesions adjoining together or with nearby periventricular WMLs 49. It usually represents more severe form of WMLs 49.   9    Figure 2-1 Periventricular vs Deep WMLs      Epidemiology of WMLs 2.2.3.1 Prevalence of WMLs in general adult population WMLs on neuroimaging are very common, even in younger adult populations. However, the prevalence increases dramatically with age 50. Some studies have reported a prevalence of 92% in people over 65 years old 50. Other studies have reported a prevalence of 42% and 70% in people of ages between 30-40 and 50-65 years respectively 50.  The nature, the distribution and the severity of the WMLs differ across the age spectrum as discussed above, but also between studies 50. For example, some studies reported high lesion   10  prevalence of around 19.5% at age 65 and 6.5% in those at age 55 50. With regards to distribution, some studies have looked at the association of WMLs distribution with differential risk factors 50. They found that periventricular white matter is associated more with vascular risk factors compared to lesions found in the deep subcortical areas 50.   There were reported limitations in separating confluent periventricular WMLs from deep WMLs that potentially could have affected the interpretation of the results 50. With regards to sex, it is not clear if there is a differential prevalence of WMLs between males and females; however, some studies have shown higher WMLs volumes in females 50. Some studies suggested that estrogen is protective against neuronal death 51.Thus, it has been hypothesized that the reduction of estrogen following menopause in females may aggravate the degenerative damage in their cerebral white matter explaining the increase prevalence of WMLs in females 52.  2.2.3.2 Prevalence of WMLs in neurodegenerative diseases Previous studies looking into the prevalence of WMLs in different neurodegenerative diseases and vascular dementia have found it to be very high 45,53. When the prevalence of periventricular caps was analyzed based on distribution and shape, periventricular frontal caps were common in vascular dementia, AD, dementia with Lewy body and normal controls with a prevalence of 96%, 93%, 92% and 89% respectively 45,53.  The prevalence of periventricular WMLs located in the occipital areas, were more common in neurodegenerative disease than in controls 45. The reported prevalence was 82% in Lewy body dementia, 75% in AD, 80% in vascular dementia and only 31% in controls 45. Periventricular   11  bands were found to be more severe in AD and vascular dementia when compared to normal controls 45.    Pathophysiology of WMLs The pial network is where most of the white matter’s blood supply originate from giving rise to the long penetrating arteries 54. The penetrating arteries that branch from the subarachnoid vessels then run across the cerebral cortex entering to supply the white matter of myelinated fibers 54. The distributing blood vessels are usually short and come off the carrying vessels to supply a solitary white matter metabolic unit which is cylindrically shaped 54.  The periventricular white matter’s blood supply originates from either the choroidal arteries or the terminal branches of the rami striati which supply the periventricular white matter through ventriculofugal vessels coming out of the subependymal 54. The basal ganglia part of the thalamus and the internal capsule get their blood supply from the same origin as the periventricular white matter 54. There are usually rare, if any, anastomoses between the branches supplying the periventricular white matter and penetrating arteries originating from the pial network. For this reason, the periventricular white matter is referred to as watershed area making it vulnerable to changes in the systemic or regional reduction of blood pressure 54,55.    12   Figure 2-2 White matter blood supply permission granted from author Prof. Joanna M Wardlaw MD 55  Age-related changes lead to further elongation and tortuosity which makes those blood vessels even more susceptible to change in systemic or regional blood pressure 54. The U fibers, which are located just beneath the cortex, are unique in that they receive dual blood supply from the penetrating blood vessels and other short blood vessels that supply the cortical areas, as well as the white matter 54. It is worth noting that the U fibers are usually not affected by WMLs that are thought to be secondary to ischemic small vessel disease 54. The sparing of U fibers in ischemic WMLs is thought to be related to the dual blood supply these fibers receive from both the long penetrating arterioles and the short adjacent cortical blood vessels 54.   13    The anatomy of the small blood vessels supplying the white matter, renders the small blood vessels highly vulnerable to further pathological alterations such as arteriolosclerosis, which is thought to play a major role in the formation of WMLs. Lacunar infarcts usually follow the chronic stenosis or occlusion of small blood vessels leading to areas of necrosis 54,56. Arteriolosclerotic changes of small blood vessels can lead to arteriolar stiffness  , thought to play a major role in the WMLs formation, can impair the physiologic autoregulation response rendering the white matter vulnerable to even subtle swings in systemic blood pressure 54,56.  Cerebral autoregulation is a physiological system which aims to maintain the cerebral blood pressure constant at a wide range of systemic blood pressure 54. The autoregulation requires the participation of large and small blood vessels 54. Specifically, small blood vessels can show significant dilatation in response to systemic hypotension 54.      Pathological findings of WMLs in SIVCI 2.2.5.1 Ischemia driven findings WMLs have very broad causes, but the most commonly documented cause is ischemia which is thought to be the primary driver of the pathological changes associated with WMLs 54. Histologically, there is significant heterogeneity underlying the radiological presentation of WMLs. However, common findings include myelin rarefaction which classically does not affect   14  the U fibers 54. Additionally, there is evidence of spongiosis, loss of the axons, astrogliosis and perivascular space enlargement 54.   There is a switch to a fibro-hyaline material of the smooth muscle involving the arterioles and penetrating arteries specifically 54. These changes present as a reduction in the vessel lumen and wall diameter known as arteriolosclerosis 54. The intimate pathological finding of arteriolosclerosis to areas harboring lacunar infarcts manifesting in areas of necrosis, cavity formation, focal ischemic changes and WMLs formed the basis of the ischemic nature of WMLs 54.  2.2.5.2 Non-ischemic related findings Some studies have suggested that axonal loss related to the white matter is an important mechanism for WMLs formation 57. This was based on the notion that axonal loss leads to the atrophy of dependent cortical areas, or is a result of cortical pathology as commonly associated with a neurodegenerative process like AD 57.   It is not clear what triggers the degenerative process that leads to axonal impairment 57. The axonal loss seems to co-occur with cortical (gray matter) shrinkage 57. Another mechanism includes AD related pathology activating calpain induced degradation at the axonal level which eventually leads to axonal transport dysfunction 57. This is also supported by neuroimaging findings of area-specific WMLs especially located at the posterior areas and corpus callosum in patients with AD 57.    15  A recent study looking at the postmortem pathological changes involving the white matter in the parietal area in individuals diagnosed with AD found evidence of wallerian degeneration induced by the AD pathological changes at the cortical areas 10. The finding of Wallerian degeneration involving WLMs differs from the ischemic changes seen in normal individuals, which argues against considering WMLs as a proxy for small vessel disease without considering other common neurodegenerative diseases like AD especially at the parietal white matter 10.  Clear evidence was found in AD mouse models showing axonal transport impairment that preceded the core pathological changes of AD. Additionally, it must be noted that these findings were confirmed in human studies 58. The pathological findings include abnormal swellings that aggregate microtubules associated proteins and molecular motor proteins as well as other vesicles and organelles in excess numbers 58,59. The formation of senile plaques of AD related pathology follows the decrease in microtubule-dependent transport which potentially triggers the proteolytic processing of β-amyloid precursor protein 58,59.  2.3 Association of white matter hyperintensity and cognition Cognition is defined as “The mental action or process of acquiring knowledge and understanding through thought, experience, and the senses” 60. The core cognitive domains commonly assessed with cognitive impairment are attention, memory, executive function and language 61.   Attention is commonly defined as the ability to choose behaviorally relevant elements from our sensory experience to be processed by other cognitive domains while concurrently prohibiting other sensory experience from being conceived at a conscious level 62. Attention comprises a   16  variety of processes such as 1) Selective attention where certain information either internal or external information gets perceived and processed while other information does not; 2) Sustained attention where extra emphasis on a source of information is pursued either in the long or short term; 3) Intensive attention process when variable amount of attention gets dedicated to a particular information source 62. Attention does not localize to solitary cortical or subcortical structure and it is neither a property of a single brain area nor the entire brain 62. It utilizes widespread networks of anatomical areas that execute a specific cognitive function  62.   Memory represents a constellation of mental capabilities that build on many systems inside the brain 63. It can be categorized into working, episodic, procedural and semantic memory 63. What is referred to as the memory system is the means by which the brain is able to process information that will be accessible for future use 63.  The term executive function is a wide construct that includes a multitude of cognitive processes and behavioral abilities including the following abilities 23: a) verbal reasoning; b) problem-solving; c) plan and carry out sequential tasks; d) sustain attention; e) resist interference; f) multi-task; g) cognitive flexibility; and h) deal with novel situations or environments 23.  The severity of progression of white matter hyperintensity has been linked to the faster rate of decline in cognitive function 44,64. The link between white matter hyperintensity and cognitive decline or dementia is thought to be related to the direct defect to subcortical fibers 44,64 . Based on many studies, the main cognitive domain affected by white matter hyperintensity is executive function that primarily depends on intact cortical-subcortical connections which are commonly   17  affected by WMLs 44,64. Other studies found an association between white matter hyperintensity with processing speed and explicit memory proving to be more sensitive to the presence of white matter hyperintensity 44,64.   Some studies also suggest a secondary effect of white matter hyperintensity to cortical involvement by a neurodegenerative process such as AD which can exacerbate the cognitive decline in those circumstances 44. Although the premise that white matter hyperintensity represents small vessel disease and vascular dysfunction, controversies about the link between white matter hyperintensity and vascular risk factors still exists with some studies failing to show a strong association between them after adjusting for stroke and other vascular risk factors 44.  A meta-analysis looking at many studies to explore the association between white matter hyperintensity with cognitive decline and stroke found that white matter hyperintensity is associated with higher risk of strokes, low scores on cognitive screening tests, and long term cognitive decline 44,64. Other studies that looked at the effect of increasing white matter hyperintensities on stroke incidence and diagnosis of dementia suggested a significant link between dementia and stroke with higher development of white matter hyperintensities 44,64.  2.4 Association of WMLs and mobility performance  Mobility performance is commonly assessed by many validated measures including the SPPB. The SPPB battery has been found to be predictive of functional decline and disability 30. The SPPB battery assesses standing balance, repeated standing from a seated position on a chair and   18  walking a short distance with usual pace. Other mobility measures include TUG, Tinetti scale and gait speed 31.   WMLs are related to mobility performance through different mechanisms 32. It was suggested that the increased fall risk and poor mobility is secondary to involvement of long descending motor fibers, corticostriatal and thalamostriatal networks 32. These networks are central in the control of lower extremity motor function, especially when it comes to the speed and duration of movement and mediating the selection and initiation of lower limb movement 32. Moreover, the involvement of frontal-subcortical network that is known for linking different areas of the cerebral cortex together potentially explains the cognitive dysfunction related to WMLs 32.   It is suggested that executive dysfunction can impair mobility following the involvement of frontal lobes, which participates in postural control due to the inability to recruit sufficient attentional resources to process executive step needed to control posture 32,65.  The volume of WMLs was linked to mobility impairment 29. A large longitudinal study assessing 1,702 individuals aged 80 years or younger in France using a fully automated method to detect, localize and measure WMLs volume found that WMLs exceeding the 90th percentile to be associated with slower walking speed 29. The location of WMLs impacted the association this association 29. Specifically, periventricular WMLs were found to have a stronger association with poor mobility 29. On the other hand, recent studies found an association between deep WMLs volume and gait speed under dual-task conditions 66.    19  Gait speed has been found to be predictor of functional decline 67, postoperative complication and mortality which has significant clinical relevance 68. Models incorporating gait speed, cognitive function and WMLs volume among individuals with neuroimaging evidence of small vessel disease predicted mortality 69. Evidence suggest resistance training may reduce WML progression and reduced WML progression was significantly associated with the maintenance of gait speed 70,71.   2.5 Assessment of WMLs   Qualitative assessment of WMLs  White matter hyperintensity can be assessed by either quantitative or qualitative methods. Visual rating is considered to be an example of qualitative assessment. Lesions are assessed based on certain characteristics including the size of the lesion, location, and the number of lesions 72. The Fazekas’ scale is one of the most widely used and validated scales. It can be used to measure the severity of deep white matter (DWMH) and periventricular hyperintensities 72,73.   Using the Fazekas’ scale, periventricular hyperintensities can be scored as the following  (0)  points are given if periventricular hyperintensities are absent, (1) point is given if periventricular hyperintensities are caps or pencil thin lining, (2) points are given if there is smooth halo, and (3) points are given if  irregular or extending into deep white matter. Moreover, DWMH is scored as following: (0) if deep white matter hyperintensities are absent, (1) point if there are punctate foci, (2) points for the beginning of confluence of a white matter focus, and (3) points for large confluent regions 72,73.   20   Other attempts to further standardize the subjective qualitative assessment of WMLs using a standardized protocol were put forward by the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) 74,75. Despite the wide use of qualitative measures they have several limitations including the following points 1) Vulnerability to bias between raters 74; 2) Insensitivity to minor changes in lesion size which can have a significant clinical implication if unaccounted for 73; 3) Having a ceiling effect 76; 4) Higher degree of subjectivity; 5) Not as reliable as the quantitative measure 74.   Quantitative assessment of WMLs  Quantification of WMLs volume has been attempted since the 1990s utilizing basic techniques like the tracing of images in a manual fashion or other methods that were again subjective and labor-intensive 74,77,78. Currently, automated techniques using fuzzy voxel technology are utilized to quantify WMLs 74,79.   Fuzzy voxel approach is utilized for quantification and detection of signals with high intensity on MRI 79. With the fuzzy voxel approach, an element that has high intensity on MRI, is regarded as a fuzzily connected entity. On the other, the combination of two voxels are assumed to have “strength connectivity” 79. The strength connectivity is the universal property determining how an object is formed according to the fuzzy way voxels are joined together 79. This approach has proven to be reliable but requires the presence of an expert rater to select the lesions for analysis 74. Additionally, other volumetric techniques applied a semi-automated   21  approach using diffusion-weighted imaging (DWI) and fast-fluid-attenuated inversion recovery (FLAIR) also prove to have a high degree of reliability in patients with AD 74,80.    Some limitations to quantitative measures include the difficulty in separating periventricular form deep WMLs especially for confluent lesions 74. More recently over the past couple of years, automated machine learning algorithms such as random forest and Support Vector Machines (SVMs) have been widely utilized as an automated method for WMLs quantification 81.  However, there is a remaining challenge with the selection of specific features or patterns for recognition and further processing which has a great influence on the final outcome using this method 81. To overcome this obstacle, a new technology applying “deep learning approach”  such as convolutional neural network was able to resolve this issue and was, therefore, used in more recent studies 81.   Deep learning is a method which permits the understanding of complex data through computational models made up by multiple processing layers 82,83. However, even with the use of new technologies that offer a great deal of convenience, there is a considerable limitation because of the variability and the lack of reproducibility using these novel automated quantitative methods 81,84.   To study the issue of reproducibility and variability between studies and scan-rescan reproducibility in multicenter studies or even a single-center study, a recent systematic review found significant variability between different centers with better reproducibility within a single center. They attributed this variability and lack of reproducibility to many factors including 1)   22  The type of sequence used; 2) The field strength; 3) Patient positioning in the scanner; 4) The manufacturer of the scanner; 5) The effect of the scanner upgrade; 6) The head coil used. These factors need to be considered to improve reproducibility in the future 76,81,84.  2.6 Clinical presentation of VCI  The clinical presentation of individuals with VCI can vary significantly depending on many factors including the different etiological causes, pathophysiological process, age-associated changes, premorbid physical and cognitive function, cognitive reserve and natural course of different underlying disease process associated with VCI 4.   It is also important to consider the historical background mentioned previously in the development of the different criteria used to define VCI which was based on different inclusion and exclusion factors, making it very difficult to find consensus on a holistic definition either for clinical or research criteria 4. However, recently there has been a great attempt for defining VCI through the VICCCS  which included the input from the world experts in this field and considered previous sets of guidelines such as International Society for Vascular Behavioral and Cognitive Disorders (VASCOG) criteria for vascular cognitive disorder, American Heart Association (AHA), and others in their final consensus 4. The clinical presentation below will be based on the different categories included in the VICCCS 4.     Post-stroke dementia This designation is usually used to describe patients manifesting immediate and/or delayed cognitive decline that happens following but within six months of stroke onset that does not   23  improve 4. Patients presenting with findings suggestive of MCI before a stroke can still be described as having post-stroke dementia if developed significant cognitive impairment within six months following stroke as well 4.   Post-stroke dementia encompasses wide etiological causes such as SIVaD, multiple subcortical infarctions, cortical infarctions, strategic infarcts and even neurodegenerative dementia including AD diagnosed within six months of stroke onset. The temporal relation of cognitive decline developing within six months following stroke incident is what separates post-stroke dementia from other VCI categories 4.   Stroke is commonly defined as a neurological deficit following a focal injury of the Central Nervous System (CNS) by a vascular cause happening acutely. It included the following entities A) Cerebral infarction; B) Intracerebral Hemorrhage (ICH); C) SubArachnoid Hemorrhage (SAH). Stroke is well known to be a leading cause of disability and mortality globally 85.    Multi-infarct dementia Multi-infarct dementia is a term used to encompass dementia caused by large multiple cortical infarcts 4. Multi-infarct dementia is commonly related to vascular disease involving the larger blood vessels, and commonly associated with vascular risk factors such as hypertension 4,86. The term Multi-infarct dementia is used to describe cognitive decline associated with multiple large cortical infarcts 4,86.      24   Subcortical Ischemic Vascular Dementia (SIVaD)  This category encompasses cognitive impairment following lacunar infarct and ischemic WMLs. The main underlying cause is the small-vessel disease involving subcortical areas 4. The presentation of SIVaD commonly involve executive dysfunction, decreased processing speed, gait abnormality, limbs rigidity and neurologic sequelae of lacunar strokes 4,87-89.    Mixed dementias  This category is designated to include all above categories when combined with a neurodegenerative process like AD or Lewy body dementia 4. The order of terms should reflect the relative contribution of the underlying pathology that is AD-VCI or VCI-AD 4.               25   Clinical presentation of VCI  based on the recent VICCCS study 4 Table 2-1 Clinical presentation of VCI VCI Mild VCI Major VCI (vascular dementia VaD) Major VCI (vascular dementia VaD)    Post-stroke dementia (PSD)    Subcortical ischemic VaD (SIVaD)    Multi-infarct (cortical) dementia (MID)    Mixed dementias  Subcategory Presentation Mixed pathology allowed Temporal association Comments Post-Stroke Dementia (PSD) Individuals may exhibit immediate and/or delayed cognitive decline that begins after, but within 6 months of stroke that does not recover. PSD-AD PSD-DLB PSD-# (other combinations allowed)  MUST be within 6 months of vascular event. It includes cases with MID, SIVaD, strategic infarcts. May or may not have presented evidence of MCI before stroke. Subcortical ischemic VaD (SIVaD) Small-vessel disease is the main vascular cause of SIVaD. Lacunar infarct and ischemic WMLs are the main types of brain lesions, which are primarily located subcortically. VCI-AD VCI-DLB VCI-#(other combinations allowed) Not required It incorporates the overlapping clinical entities of Binswanger’s disease and the lacunar state. Multi-Infarct (Cortical) Dementia (MID) The involvement, and likely contribution, of multiple large cortical infarcts in the development of dementia VCI-AD VCI-DLB VCI-#(others) Not required  Mixed dementias Considered as standalone umbrella and may include any of the phenotypes specified above. VCI-AD VCI-DLB VCI-#(others) Not required The order of terms should reflect the relative contribution of the underlying pathology that is AD-VCI or VCI-AD.   26  2.7 Cortical vs sub-cortical presentation   Sub-cortical presentation  Pathologies involving the subcortical structure usually present with apathy, depression, poor memory described as forgetfulness which manifests as impairment in the retrieval of learned information and responds to cueing, psychomotor slowing and decreased processing speed 90,91.  The motor symptoms usually present with slow gait speed, gait abnormality related to bradykinesia. Patients traditionally described as presenting with extrapyramidal symptoms such as rigidity and unsteadiness.   Apathy is frequently defined as the loss of motivation and reduction of interest in daily activities 92. Bradykinesia, another common subcortical symptom, is defined as the slowness of movement performance. On the other hand, akinesia is defined as the lack of spontaneous movement 93. The lack of facial expression is an example for akinesia. Another example of akinesia is reduced arm swing while walking. Hypokinesia refers to slower and smaller movement than that desired 93.  Rigidity, on the other hand, is defined as the high resistance of an extremity during passive movement 94. It is usually velocity and direction independent 94. Ataxic gait usually manifests as widened step widths and step lengths that is irregular, with a walking path that shows veering while walking 95.      27   Cortical presentation  Individuals with cortical involvement usually present with aphasia, apraxia and agnosia, which is usually spared in pathology mainly affecting the subcortical structures 90,91. Additionally, cortical dementias present with agnosia, acalculia and agraphia, which is usually not evident in pathologies primarily affecting subcortical structure disorders 91. Aphasia is defined as impairment or even loss of verbal communication, following a brain dysfunction 96. It usually presents with global impairment in most of the verbal abilities including difficulty or impairment for naming, comprehension of spoken or written items difficulty with reading, writing and difficulty in expression 96. Agnosia is defined as the inability to gain access to the semantic knowledge of an object or other stimuli not caused by basic perceptual dysfunction 97.   Apraxia is commonly defined as impaired ability to execute purposeful learned skilled movements in spite of willingness to perform that movement. It can not be related to impairment in comprehension, primary motor or sensory function. It usually signifies a cortical brain dysfunction 98.   Agraphia can be defined as an acquired impairment in writing abilities induced by brain damage 96. Writing mainly encompasses the following components; the ability to spell 96, the ability to process language, the ability to visually perceive and the ability to orient symbols in a visuospatial manner 96. Moreover, it includes the motor control of writing and motor planning 96.   Acalculia is defined as the acquired inability to carry out mathematical calculation whether it is carried out mentally or when attempted through pencil and paper 96.   28    2.8 Arterial stiffness and PWV Cognitive decline has been strongly linked to vascular risk factors including hypertension, hyperglycemia and left ventricular dysfunction 99,100. Other vascular changes related to vascular aging such as “ arterial stiffness” has been found to be linked to cognitive impairment 100. PWV as a measure of aortic stiffness was found to be associated with higher white matter hyperintensity volume, a surrogate marker for cerebral small vessel disease 19.   The transmission of high pressure and flow pulsatility is thought to damage the small blood vessel supplying susceptible white matter tracts, leading to cognitive decline 19. The following section will review relevant background information about arterial stiffness and its relation to clinical outcome and cognitive decline.   Overview of arterial stiffness and PWV As we age, the large conduit arteries such as the aorta becomes stiffer and less compliant. These changes are associated with age-related changes and atherosclerosis 11. The aorta and large conduit vessels play an important role in the transfer of blood supply to end organs and peripheral tissues 11. The ejection of the blood from the heart is associated with frequent, intermittent oscillations that can have a deleterious effect on peripheral tissues if not attenuated by a compliant elastic conduit large vessel such as the aorta 11.   The term aortic stiffness or arterial stiffness has been used interchangeably for the past years 12. Aortic stiffness is usually defined as “the elastic resistance to deformation” 11. It is influenced by   29  many factors interacting with each other in a complex fashion leading to aortic stiffness. These factors include collagen, vascular smooth muscle, fibrillin fibers and the extracellular matrix containing elastin 11.   The PWV can be described as the generation of the speed of the systolic flow wavefront traveling through the large aortic blood vessel, which in fact is a reflection of the elastic properties of its wall 11,13. The PWV is considered a surrogate marker for arterial stiffness; higher arterial stiffness is directly proportional to PWV 13,14.   Pathophysiology of arterial stiffness Large conduit vessels particularly the aorta, can serve as blood reservoir and is able to store almost half of the cardiac output due to its elastic properties. The stored blood is then pushed to the periphery through the elastic recoil property of the aorta during diastole, this is also known as Windkessel function 11.   Under normal conditions, pressure pulse has low PWV ranging between 5-7 meters per second. It is generated by the ejection of blood from the left ventricular. The forward pulse wave then gets reflected back during diastole after reaching distal bifurcation points at the level of smaller arteries and arterioles 11. The interaction between incident wave and the reflected wave creates the dicrotic notch. The summation of these waveforms creates the final shape and waveforms of aortic blood pressure 11.    30  The pulse pressure amplification is explained by the differential increase in peripheral pulse pressure over the central pulse pressure created by the returning of the reflected wave at diastole in a compliant vessel 11. With aging, blood vessels become stiffer leading to the increase in PWV, causing the early return of the reflected wave during systole, which consequently lead to the rise in systolic blood pressure.   Another marker of aortic stiffness is called the “augmentation index” which is calculated by relative increase in systolic blood pressure divided by pulse pressure that correlates with disruption of elastin in arterial wall secondary to the mechanical shear stress 11.  The injury of the brain’s microvasculature  happens due to the exposure of the small vessels to the high pressure fluctuations as well as the high flow through the cerebral circulation 11. Normally, the brain is perfused at high pressure but at a considerably low vascular resistance, which leaves the brain microvasculature susceptible to high pressure fluctuations 11.   Assessment of PWV The accepted gold standard measurement of arterial stiffness is CF-PWV 14. It is regarded the gold standard method by experts in the field because of the large epidemiological evidence supporting its predictive ability for cardiovascular events, because of the its simple use and that it only mandates very little expertise 14. It can be assessed through a non-invasive, easily applied and reproducible methods that incorporate the propagative physiology of the arterial system to estimate arterial stiffness 14.     31  PWV can be calculated by measuring the foot-to-foot velocity of different waveforms on a different site across the vascular system 14. The aortoiliac site is considered the most clinically relevant because the aorta is the initial branch of the vascular system leaving the left ventricle, and it is also considered the predominant contributor to arterial stiffness 14.   One of the most commonly used systems to assess PWV is called the Complior system (ALAM Medical, France). A study validating the Complior device against the reference method involving invasive cardiac catheterization found a strong correlation between the two methods with correlation coefficients more than 0.9 101.   The accuracy of the Complior device, when compared to the gold standard measure at that time involving manual calculation of PWV, was very strong between the two methods with (r = .99, p value < .001) 102. Additionally, the interobserver repeatability coefficient was 0.890 and intraobserver repeatability coefficient was 0.935 for the Complior device 102.  The Complior device utilizes a mechanotransducers applied on a different site commonly the right carotid and right femoral arteries 14. The transit time between the feet of the two waveforms recorded at the same time on different sites can be calculated through an algorithm incorporated in the device 14.   The other commonly used system called Sphygmocor (AtCor medical, Australia) uses an applanation tonometer to record distal and proximal pulse 14. It can be applied on the carotid artery proximally and femoral artery distally 14. This device uses the R wave on the recorded   32  electrocardiogram as a reference point, then the time difference of the recorded pulse and R wave between the two sites are calculated as the transit time 14.   It is important to note that the distance between the two sites of interest, the carotid–femoral distance, for both systems Sphygmocor and Complior has to be measured by tapeline and recorded to obtain an accurate estimation of PWV 14. Obesity leading to change in body habitus, specifically increased abdominal circumference, can be a limitation for acquiring the actual distance between carotid and femoral site resulting in an overestimation of PWV 14.    The PWV in predicting cardiovascular events and clinical outcomes Arterial stiffness is widely accepted as a surrogate marker for CardioVascular (CV) disease 15,103. PWV is the most commonly used surrogate measure of arterial stiffness 14. Mounting evidence has shown that high arterial stiffness can cause downstream target organ damage and pathological changes eventually leading to incident cardiovascular events 104-107. Increased pulsatility secondary to arterial stiffness leads to transmission of this energy to microcirculation which may lead to damage of the microcirculation 104. The higher pulsatility induced by increasing arterial stiffness can induce hypertrophic remodeling at the level of small vessels and microcirculation to protect it from this high transmitted pulsatile energy at the expense of impairing perfusion to vulnerable areas in the brain 104. Another theory to explain the microvascular damage in the brain and kidneys microvasculature proposes that “with ageing the aorta becomes stiffer leading to early pulse wave reflection which in turn cause increase in pulse pressure. This increase in pulse pressure is thus transmitted to the microvascular beds of the   33  brain and kidney who have limited ability to vasoconstrict in face of this high shear stress and circumferential pulsatile pressure causing significant damage to the microvasculature 17.   The strong association between arterial stiffness and cardiovascular risk factors made it an important target to predict cardiovascular events 15. Therefore, many studies have looked into the capacity of arterial stiffness to be used as a predictor of future risk of cardiovascular events such as myocardial infarction and stroke 15.   PWV, as a surrogate measure of arterial stiffness, is being used in many large studies because of the accessibility, reproducibility, affordability and the non-invasive nature of devices used to calculate it 15. A large meta-analysis reported a strong association between PWV and cardiovascular events and mortality 15. The pooled relative risk for cardiovascular events was 2.26 with 95% Confidence Interval (CI): 1.89 to 2.70. On the other hand, the all-cause mortality was 1.90, 95% confidence interval: 1.61 to 2.24. Finally, the relative risk for cardiovascular mortality was 2.02, 95% CI: 1.68 to 2.42 15.  It was found that individuals with higher underlying cardiovascular such as renal disease, hypertension or coronary artery disease had higher relative risk compared to individuals with lower baseline risk factors 15. They reported an increase of 14% in total cardiovascular events for the increase in aortic PWV by one meter per second (1m/second) after adjusting for sex, subject’s age and other risk factors 15. With regards to white matter hyperintensity as a marker of brain small vessel disease, a recent study found a strong association between larger volumes of   34  white matter hyperintensities and higher PWV 20. Higher PWV was also found to be associated with lacunar strokes but not cerebral microbleeds 20.   In a study looking at the association between PWV and cognitive function, it was found that PWV was associated with cognitive decline when assessed by a global cognitive screening test such as the Mini Mental State Examination (MMSE) 108. Individuals with high pulse wave velocity had an annual decline in MMSE by 0.45 points over the follow-up period of nine years 108. Moreover, a meta-analysis looking at the association of PWV and cognitive decline supported the association of worse cognitive outcomes and white matter hyperintensity volume with abnormal PWV 100.   There is still controversy regarding the specific cognitive domains associated with arterial stiffness and high PWV 109. Some studies found that executive dysfunction, as measured by the Stroop test, to be the main cognitive domain associated with increased PWV 110. Other studies found memory to be the only cognitive domain associated with high PWV 19. However, a recent meta-analysis found no statistically significant association with specific cognitive domains and high PWV 109.    35  Chapter 3: Methods 3.1 Introduction The data used in this project were provided by Professor Teresa Liu-Ambrose from the “Reshaping the Path of Vascular Cognitive Impairment with Resistance Training randomized control Trial” (RVCI trial), funded by the Heart and Stroke Foundation of Canada.   The RVCI Trial is an ongoing 12-month, single-blinded randomized controlled trial aimed to examine the impact of resistance training on WML progress in 88 community-dwelling adults with a diagnosis of mild SIVCI 111. Due to the timeline and resource constraints, we include data from 34 RVCI who already completed baseline assessments in this thesis.    The primary hypothesis of the RVCI trial is that participants randomized to the Resistance Training exercise (RT) program will show less WMLs progression at the end of the 12-month intervention period, compared to the control Balance and Tone Training (BAT) Group. The secondary hypothesis is that the subject randomized to the RT group will manifest improvement in measures of executive function, general mobility, mood, and quality of life when compared to participants randomized to the BAT program. Additionally, they hypothesized that participants randomized to the RT intervention group will show evidence of reduced arterial stiffness, cardiometabolic risk factors and physiological falls risk compared to participants randomized to the BAT control group.  Participants with mild SIVCI were recruited from four outpatient clinics in Metro Vancouver, British Columbia, Canada including the flowing clinics: 1) UBCH-CARD; 2) VGH Stroke   36  Clinic; 3) VGH Falls Prevention Clinic; 4) VGH Geriatric Internal Medicine Teaching Clinic. In addition, participants were recruited through newspaper advertisements and posted flyers throughout Metro Vancouver.  3.2 Randomization  Participants were randomly assigned (1:1) to the 12-month twice-weekly RT program or the 12-month twice-weekly BAT program. Research personnel acquiring measurements and the primary outcome measure are blinded. Research personnel acquiring monthly physical activity data are not be blinded.  3.3 Inclusion and exclusion criteria for the RVCI trial Inclusion criteria 1. Meet the diagnostic criteria for SIVCI outlined by Erkinjuntti and colleagues 111. For this study, participants had to have a score < 26/30 on the MoCA to fulfill the cognitive syndrome part of the criteria and have evidence of small vessel disease present on neuroimaging such as CT scan or MRI. 2. MMSE score of > 20/30 at screening. 3. ≥55 years old. 4. Community-dwelling. 5. Live in Metro Vancouver. 6. Able to comply with scheduled visits, treatment plan, and other trial procedures. 7. Able to read, write, and speak English with acceptable visual and auditory acuity.   37  8. On a stable and fixed dose of cognitive enhancer medications (e.g., donepezil, galantamine, rivastigmine memantine, etc.) that is not expected to change during the 12-month study period, or, if they are not on any of these medications, they are not expected to start them during the 12-month study period. 9. Provide a personally signed and dated informed consent document indicating that the individual (or a legally acceptable representative) has been informed of all pertinent aspects of the trial. In addition, an assent form will be provided at baseline, and again at regular intervals. 10.  Able to walk independently. 11.  In sufficient health to participate in the RT program based on medical history, vital signs and written recommendation by a family physician indicating one’s appropriateness to start an exercise program.  Exclusion criteria 1. Absence of relevant small vessel ischemic lesions on an existing brain CT or MRI. 2. Diagnosed with another type of neurodegenerative (e.g., AD) or neurological condition (e.g., multiple sclerosis, Parkinson’s disease, etc.) that affects cognition and mobility.  3. Diagnosed previously with a genetic cause of SIVCI (e.g., CADASIL). 4. Participating in regular RT in the last six months. 5. Clinically important peripheral neuropathy or severe musculoskeletal or joint disease that impairs mobility. 6. Taking medications that may negatively affect cognitive function, such as anticholinergics, including agents with pronounced anticholinergic properties (e.g., amitriptyline), major   38  tranquilizers (typical and atypical antipsychotics) and anticonvulsants (e.g., gabapentin, valproic acid, etc.). 7. Planning to participate, are already enrolled in a clinical drug trial or exercise trial concurrent to this study, or are unable to meet the specific scanning requirements of the UBC 3T MRI Research Centre. Specifically, we will exclude anyone with: pacemaker, brain aneurysm clip, cochlear implant, recent surgery or tattoos within the past 6 weeks, electrical stimulator for nerves or bones, implanted infusion pump, history of any eye injury involving metal fragments, artificial heart valve, orthopedic hardware, other metallic prostheses, coil, catheter or filter in any blood vessel, ear or eye implant, bullets, or other metallic fragments.  3.4 Measurements  The following measures were collected at baseline for the RVCI trial and used in this thesis  3.5 Assessment of executive function and global cognitive function Executive function: Specific domains of executive function were assessed in the RVCI trial. a) Response inhibition tests the deliberate ability to inhabit dominant, automatic or proponent responses, and is tested in the study with the well-validated Stroop Colour-Word Test 112; b) Set shifting requires the participant to go back and forth using multiple tasks or mental sets. Set shifting was tested using the well-validated “Trail Making Tests” 113 (Parts A & B). The time needed to complete each part is scored in seconds. The time needed to complete part B is subtracted from the time needed to complete part A reported as Trails B-A; c) Working memory deals with monitoring incoming information that is related to the task at hand, processing it, prioritizing it and replacing old irrelevant information by newly more relevant information. In   39  this study, working memory was assessed through the verbal digits tests by forward and backward tests 114. Global cognitive function: Global cognitive function was assessed at baseline by multiple well-validated screen test including; a) ADAS-COG 115. It has been used extensively in the research of AD 115. It assesses multiple cognitive and non-cognitive domains with a higher score indicating greater impairment. The cognitive domains include memory function, language, ideational praxis and constructional praxis. On the other hand, the non-cognitive domains consists of mood and behavioral symptoms including psychotic symptoms, agitation, concentration, socialization skills, nocturnal confusion, cooperation, the initiative for activities of daily living and motor activity 115; b) MMSE is a well-validated global cognitive screen that assessed multiple cognitive domains 116. Lower scores reflecting poor cognitive function with a score below 27/30 has a sensitivity of 89% and specificity of 91% in detecting patients with cognitive impairment in individuals with high education 117. The following domains are assessed with the MMSE: orientation, registration, attention, recall, language (reading and writing), copying “visuospatial ability” and three stage commands 116; c) MoCA is another global cognitive screening test that is used to detect MCI. A MoCA score of less than 26/30 has 90% sensitivity and 87% specificity in diagnosing MCI 118. The following are the cognitive domains assessed by the MoCA test: executive function, visuospatial abilities, language, attention, memory and orientation 118.    40  3.6 Mobility measures  The RVCI study collected multiple measures of mobility and balance. For the purpose of this study, we chose mobility measures that can be easily conducted in clinic and do not require extensive training or special equipment. Therefore, we focused on usual gait speed and TUG.  4-meter walk test instructions Participants are instructed to walk at their normal pace and are permitted to use their usual assistive walking aids or device. Participants are then instructed to walk through a 1-meter zone for acceleration followed by the 4-meter testing zone then lastly 1-meter of deceleration. The timer is started with the first footfall following the 0-meter line. The timer is then stopped following the first footfall of the 4-meter line. Gait speed is calculated as following gait speed = 4/ (recorded time to walk 4 meters). The reported unit is meters/seconds.  The TUG test  Participants are instructed to stand up from a standard arm chair 119. They are permitted to have their personal walking devices or aid as needed. Participants are instructed to rest their back against the chair and keep their arm resting on the chair’s arm. Then they are instructed to walk at their own comfortable pace towards the marked three meters line, turn, then finally walk back to their chair. Time for completion was recorded in seconds. Each participant completed two trials and the average of the two was used for data analysis.     41  3.7 Blood pressure measurement The Participants were asked to rest supine in a darkened room for 5 minutes prior taking blood pressure measurements. We used a single blood pressure measuring device for all participants; the Omron HEM-775 automated oscillometric sphygmomanometer. After 5 min of rest in a supine position, the start button on the device was pressed and the cuff inflates automatically. Resting systolic and diastolic blood pressure was recorded from the device screen.  3.8 PWV measurement PWV analysis by the Complior device The participants were instructed to lie down in a quiet room with darkened light for five minutes then the blood pressure was measured in that setting. The blood pressure was measured using an automated blood pressure device. Resting systolic and diastolic blood pressure were taken using the Omron HEM-775 machine. Participants were instructed to lie supine and relaxed to ensure they reach a stable heart rate and blood pressure while the operator is attempting to launch Complior Analyse software and recording basic information such as patient’s age, blood pressure, etc. The operator then palpates the carotid and femoral arteries and marks the point where they will measure the carotid and femoral pulse wave. The operator measures the distance between the carotid and femoral arteries and enters it on the Complior Analyse software. The operator then palpates for the carotid pulse then positions the carotid sensor with the help of its specific holder around the subject’s neck. Next, the operator palpates for the femoral artery and applies the femoral sensor over the femoral artery. When software indicators turn green and the word valid appears on the screen, the operator presses the valid tab button and the acquisition is stopped. The Complior Analyse software displays the PWV and the central (carotid) pressure   42  waveform analysis. Previous results of the same subject can be used as well for comparison. The calculated PWV was then recorded and stored in the database of the Complior’s device software.    Figure 3-1 Screenshot of Complior device results including PWV.    Figure 3-2 Valid CF-PWV   43       3.9 MRI WMLs volume assessment  Visual rating for WMLs For the visual rating assessment the well-validated Fazekas scale was used as following 72:  Periventricular WMLs scoring using Fazekas scale Points 0 1 2 3 Characteristic absent Caps or pencil thin lining. Smooth halo Irregular or extending into the deep white matter Table 3-1 Periventricular WMLs scoring using Fazekas scale  Deep WMLs scoring using Fazekas scale Points 0 1 2 3  Characteristic  Absent  Punctate foci Beginning of confluence of a white matter focus  large confluent regions Table 3-2 Deep WMLs scoring using Fazekas scale                  44   Figure 3-3 Visual rating for WMLs with Fazekas scale  Case courtesy of Dr Bruno Di Muzio, Radiopaedia.org, rID: 36927        45   Automated WMLs assessment  WMLs volume was calculated in cubic millimeter (mm3) using a validated classifier that is based on “Parzen windows” 120. This system processes and automatically delineates the manually marked seed points as WMLs 120.   WMLs selection for automated volume calculation Lesions were selected based on location, size and possible etiology. Lesions adjacent to ventricles were named periventricular and was assigned a different seed point color to be processed later by the software separately as periventricular WMLs. Periventricular caps or pencil line lesions were all considered periventricular WMLs. Moreover, if a periventricular cap is large and confluent with a clear demarcation between the distal border and other adjacent deep WMLs, the border of normal appearing white matter demarcates the distal border of that confluent lesion and any lesion proximal to that line is considered part of the large confluent periventricular lesion. If there is no clear demarcation border between a deep lesion and large confluent periventricular lesion, it is left to the discretion of the radiologist to choose the distal border of the periventricular lesion in that rare case. All other lesions away from the ventricles including areas deep subcortical structures such as the internal capsule, thalamus and basal ganglia were all considered deep lesions and assigned different seed point color to be processed by the software separately. The total WMLs volume was calculated based on all these seed points chosen irrespective of their color. The volume of WMLs was calculated in Cubic millimeter (mm3). White matter hyperintensities surrounding areas of encephalomalacia, focal parenchymal or stroke lesions were not considered as WMLs or chosen for processing by the   46  software. To segment the lesions, the seed points will be fed into a well-validated classifier based on Parzen windows that automatically delineate each marked lesion.  WMLs processing MRI was done at the UBC MRI Research Centre. A 3T MRI Philips Achieva scanner was used. This MRI scanner uses eight-channel sensitivity encoding head coil as well as parallel imaging. For this project, the T2-weighted (T2w) and the proton-density-weighted (PDw) sequences were used for analysis of WMLs 120. The T2w/PDw sequences enable the differentiation of cerebrospinal fluid from the skull which allows proper computation of intradural volume 120.  Preprocessing steps were standardized as the following; a) The MR intensity inhomogeneity was corrected using a multi-scale version of the nonparametric non-uniform intensity normalization method (N3) 121;  b) The SUSAN noise removing structure preserving filter was applied 122; c)  The Brain Extraction Tool (BET) was used to remove all non-brain tissues 123.            47  3.10 Statistical analyses Primary statistical analysis was conducted in SPSS 24.0. Distribution of study variables were visually analyzed and WMLs volume data that were substantially skewed were log-transformed. Pearson product-moment bivariate and partial correlation analyses were used to examine the association between WMLs data and variables of interest. The statistical significance was set as p ≤ 0.05. In the partial correlation analyses, the covariates included age, sex and antihypertensive use.   One major concern controlling for several covariates with a small sample size is running into a potential model overfitting issues which can possibly significantly bias our understanding of the estimated associations between variables of interest 124. Our approach to avoid the possibility of model overfitting was to include one covariate at a time and by observing that the estimated associations do not change much, we accepted this as evidence that overfitting was not a substantial issue in the fully adjusted model. We choose to conduct the basic bivariate correlation first. We then added further covariates in step by step fashion as needed. Given that age is factor known to be associated with WMLs volume we controlled for age in our models. Additionally, we were interested in understanding the association of WMLs volume with cognitive and physical performance independent of age. We understand that other confounders such as sex and antihypertensive medications can affect pulse wave velocity, cognitive and mobility performance so it was important to adjust for them in the fully adjusted model.     48  Chapter 4: Results  4.1 Participants characteristic Of the thirty four participants who completed baseline assessment and randomized, thirty three participants were included in this cross-sectional analysis. One participant was excluded for missing data.   Table 4-1 provides the demographic characteristics for the RVCI participants. 67.6% of the RVCI cohort are females. The mean age is 75.2, and the baseline mean MoCA score was 20.1. Overall, this subset of the RVCI cohort had a mean gait speed of 1.15 meters/second and average TUG of 8.86 seconds. The median PWV for the RVCI cohort at baseline was 10.65 meters/second which lies within reported normal reference range for this age group 125.   The mean total WMLs volume at baseline was 11761.06 mm3. Periventricular WMLs volume accounted for 74.19 % of the total volume with a mean volume of 8725.98 mm3. The mean baseline systolic and diastolic blood pressure were 134.8 and 78 mmHg respectively. Participant number 12 was not included in the analysis for incomplete data at the time of analysis.              49   Table 4-1 Demographic data and participants characteristics   Characteristic  Mean Median SD Biological Sex (number and %)     Females: 23, 67.6%          Males: 11, 32.4%    Age years 75.2 75 6.10 Systolic blood pressure (mmHg) 134.8 135.5 17.38 Diastolic blood pressure (mmHg) 78 77 9.45 Heart rate (bpm) 62.5 61.5 14.58 Height (cm) 162.9 161.8 11.47 Weight (kg) 70.47 69.5 15.60 Average TUG in seconds 8.86 8.35 2.041 Average dual TUG in seconds 41.96 10.64 7.42 Gait speed meter per seconds 1.15 1.15 0.22 CF-PWV meter per seconds 11.31 10.65 5.46 CF tolerance 7.77 3.1 14.39 Fazekas score 1.64 2 0.87 WMLs total (mm3) 11761.06 7235.76 10595.95 WMLs-periventricular (mm3) 8725.98 5419.68 8129.54 WMLs-deep (mm3) 3584.51 1523.16 4926.97 ADAS-COG/ 70 points 13.32 13 5.73 MoCA /30 points 20.17 21.5 4.44  MMSE /30 points  27.14 27 2.04 Delta Stroop seconds 53.48 43.83 36.39 B-A Trails 90.14 53.81 92.50 Trails A seconds 43.33 36.87 18.91 Trails B seconds 133.48 97.11 107.26 Category Fluency Animals N per 60sec 14.97 15 4.96 Category Fluency Vegetables N per 60sec 11.79 11 4.91  (TUG: Time Up and Go test, CF-PWV: Carotid-femoral pulse wave velocity, CF: Carotid-femoral, WMLs: White Matter Lesions, MoCA:  Montreal Cognitive Assessment, MMSE: Mini mental state examination, HTN: hypertension, SD: Standard Deviation, mmHg: millimeter of mercury, bpm: beats per minutes, cm: centimeter, kg: kilograms, mm3: cubic millimeter).        50  4.2 Correlation between the two independent raters for WMLs Two independent raters GZ: 1st rater and WK: 2nd rater, analyzed and selected total WMLs for automated quantification for a sample of 18 participants of the RVCI cohort. Table 4-2 show the mean, median minimum and maximum values for each rater. Figure 4-1 displays the correlation between the two raters showing R2 = 0.99 representing an excellent correlation between the two independent raters.   Table 4-2 Correlation between the two independent raters for WMLs.   This table shows the compression of total WMLs volume by the two independent expert eyes. GZ: 1st rater WK: 2nd rater. mm3: cubic millimeter   Participant Total WMLs mm3 WK Total WMLs mm3 GZ 01 22458.9 22396.0 02 12093.9 12362.5 03 5292.5 4309.5 04 39796.7 39888.2 05 18612.4 20758.5 06 3695.0 880.2 07 1674.6 1803.2 08 3626.5 1597.5 09 17589.3 17283.5 10 11613.8 11422.3 11 16849.2 16714.9 13 2163.3 1311.7 14 21187.2 20727.1 15 1831.8 1300.3 16 19501.1 19063.9 17 27422.8 24979.4 18 1900.4 1334.6 19 6912.8 5529.7 Mean 13012.3 12425.7 SD 10736.1 11071.3 Min 1674.6 880.2 Max 39796.7 39888.2   51       Figure 4-1 Correlation between the two independent WMLs raters. GZ: 1st rater WK: 2nd                   52  4.3 Correlations of measures of interest with total, deep and periventricular WMLs volume The measures of interest include CF-PWV, executive function measures (Trails B, B-A, Stroop test and animal fluency), global cognitive test (MoCA, ADAS-cog), mobility performance (gait speed, TUG) and blood pressure (systolic and diastolic). Table 4-3 shows the correlation values between measures of RVCI cohort, total periventricular and deep WMLs volume. The strongest correlation found was between age and total WMLs volume (r = 0.67 p value < 0.001).   4.4 Correlation between total, deep and periventricular WMLs volume and CF-PWV There were no statistically significant associations between total, deep or periventricular, WMLs volume and CF-PWV. PWV is associated with age 125, but additional analyses controlling for age (i.e., partial correlations) did not change this relationship between PWV and periventricular, deep or total WMLs. Table 4-4 provides a model controlling for sex and antihypertensive medication use known to affect PWV 125. This did not affect the association between CF-PWV, total, deep, and periventricular WMLs volume.  4.5 The association between total, deep and periventricular WMLs volume and measures of global cognitive function and executive function  Table 4-3 shows that there were no associations between total, deep or periventricular WMLs with measures of global cognitive function. There were also no associations between total or periventricular WMLs with measures executive function. The associations between total and periventricular WMLs with measures of global cognitive function or executive function   53  remained not significant after controlling for age, sex and antihypertensive use. On the other hand, deep WMLs volume was significantly associated with animal fluency (r = - 0.39, p = 0.03) but deep WMLs was not significantly associated with Trails B-A or Stroop test. The association between deep WMLs and animal fluency remained significant after controlling for age (r = -0.43, p = 0.01). Table 4-4 provides a model controlling for age, sex and antihypertensive medications use. It shows that the association between animal fluency and deep WMLs was still significant (r = -0.49, p = 0.006).   4.6 The association between total, deep and periventricular WMLs volume with mobility performance (TUG, gait speed) Table 4-3 shows that there were no associations between total, deep or periventricular WMLs with mobility measures (TUG or gait speed). The association of TUG or gait speed remained not significant after controlling for age alone or in a model controlling for age, sex, and antihypertensive use see table 4-4.  4.7 The association between total, deep and periventricular WMLs volume with systolic and diastolic blood pressure Table 4-3 shows that there were no associations between total or periventricular WMLs with systolic or diastolic blood pressure. Deep WMLs volume was found to be significantly associated with diastolic blood pressure (r = 0.44, p = 0.01). This association between deep WMLs and diastolic blood pressure remained significant after controlling for age (r = 0.59, p = 0.001).    54  Table 4-4 provide a model controlling for age, sex, and antihypertensive medication use, and again the correlation between deep WMLs volume and diastolic blood pressure remained significant (r = 0.59, p = 0.001).      Table 4-3 Correlations of measures of interest with total, deep and periventricular WMLs volume.  Log WMLs total Log WMLs periventricular Log WMLs deep Bivariate Correlations r p value r p value r p value Systolic BP (mmHg) .32 .07 .28 .11 .43 .01 Diastolic BP(mmHg) .21 .23 .14 .43 .44 .01 CF-PWV meter per seconds .27 .14 .26 .14 .25 .17 ADAS Cog /70 points -.04 .82 -.05 .77 -.04 .82 Animal fluency per 60 seconds -.22 .21 -.17 .34 -.39 .03 Log Trails B minus A .11 .53 .05 .80 .29 .10 MoCA /30 points .13 .47 .18 .31 -.07 .71 Log Stroop seconds .30 .10 .28 .12 .29 .10 Gait speed meter per seconds -.15 .41 -.16 .36 -.10 .60 Average TUG seconds .27 .13 .27 .14 .24 .18 Age years .67 <.001 .65 <.001 .61 <.001 Partial Correlations Controlling for Age       Systolic BP (mmHg) .13 .47 .08 .66 .29 .10 Diastolic BP(mmHg) .33 .07 .23 .22 .59 <.001 CF-PWV meter per seconds .03 .88 .03 .86 .03 .87 ADAS Cog/70 points -.04 .84 -.05 .78 -.04 .84 Animal fluency per 60 seconds -.23 .20 -.16 .38 -.43 .01 Log Trails B minus A .13 .46 .05 .81 .36 .05 MoCA /30 points -.001 .99 .07 .69 -.24 .19 Log Stroop seconds .13 .48 .11 .56 .14 .43 Gait speed meter per seconds -.18 .33 -.20 .28 -.10 .58 Average TUG seconds .25 .18 .24 .19 .20 .27   (BP: blood pressure, TUG: Time Up and Go test, CF-PWV: Carotid-femoral pulse wave velocity, WMLs: White Matter Lesions, MoCA: Montreal Cognitive Assessment, MMSE: Mini mental state examination, HTN: hypertension.)   55  Table 4-4 Correlation of measures of interest with total, deep and periventricular WMLs volume controlling for age, sex and antihypertensive medication.  Log WMLs total Log WMLs periventricular Log WMLs deep Bivariate Correlations r p value r p value r p value Partial Correlations Controlling for Age, Sex, and HTN Meds       Systolic BP (mmHg) .09 .64 .05 .78 .22 .24 Diastolic BP (mmHg) .36 .05 .27 .16 .59 .001 CF-PWV meter per seconds .02 .91 .04 .82 -.02 .93 ADAS Cog /70 points -.01 .96 -.01 .98 -.07 .73 Animal fluency per 60 seconds -.31 .10 -.25 .19 -.49 .006 Log Trails B minus A .09 .63 .03 .89 .26 .17 MoCA /30 points .02 .93 .06 .74 -.16 .39 Log Stroop seconds  .24 .19 .27 .15 .13 .49 Gait speed meter per second -.18 .34 -.19 .31 -.13 .51 Average TUG seconds .28 .14 .28 .14 .19 .30   (BP: blood pressure, TUG: Time Up and Go test, CF-PWV: Carotid-femoral pulse wave velocity, WMLs: White Matter Lesions, MoCA: Montreal Cognitive Assessment, MMSE: Mini mental state examination, HTN: hypertension.)                        56  CF-PWV CF-PWV CF-PWV  Log Total WMLs Log Periventricular WMLs Log Deep WMLs  Figure 4-2  Association of CF-PWV with total, periventricular and deep WMLs volume.           57   Log Total WMLs Log Periventricular WMLs Log Deep WMLs  Figure 4-3 Association of Trails B minus trail A with total, periventricular and deep WMLs volume.                             58   Log Total WMLs Log Periventricular WMLs Log Deep WMLs  Figure 4-4 Association of Animal fluency scores with total, periventricular and deep WMLs volume.             59   Log Total WMLs Log Periventricular WMLs Log Deep WMLs  Figure 4-5 Association of gait speed  with total, periventricular and deep WMLs volume.           60   Log Total WMLs Log Periventricular WMLs Log Deep WMLs Figure 4-6 Association of Time Up and Go test “TUG” with total, periventricular and deep WMLs volume.          61   Log Total WMLs Log Periventricular WMLs Log Deep WMLs  Figure 4-7 Association of systolic blood pressure with total, periventricular and deep WMLs volume.        62   Log Total WMLs Log Periventricular WMLs Log Deep WMLs  Figure 4-8 Association of diastolic blood pressure with total, periventricular and deep WMLs volume.         63   Log Total WMLs Log Periventricular WMLs Log Deep WMLs  Figure 4-9 Association of age with total, periventricular and deep WMLs volume.            64  Chapter 5: Discussion In this cross-sectional study of older adults with mild SIVCI, I examined the following questions: 1) What is the association between total, deep and periventricular WMLs volume on MRI with CF-PWV? 2) What are the associations between total, deep and periventricular WMLs volume and measures of global cognitive function and executive function?      3) What is the association between total, deep and periventricular WMLs volume with mobility performance? 4) What is the association between total, deep and periventricular WMLs volume with systolic and diastolic blood pressure?  The results of this study show that CF-PWV is not associated with periventricular WMLs volume (bivariate r = 0.26 and p value = 0.14). Additionally, CF-PWV was not associated with either the total or deep WMLs volume (bivariate r = 0.27 and p value = 0.14, r = 0.25 p value = 0.17) respectively.  These results are discordant from other studies that found a significant association between CF-PWV and periventricular, deep or total WMLs on neuroimaging, but they did not include individuals with mild SIVCI and included individuals with normal cognitive function 16,18,19,21,108,126-131. Notably, these previous studies had much larger cohorts. For example, the Rotterdam Scan Study included 1460 participants and used the Complior device to measure the CF-PWV and used a validated automated WMLs segmentation software for volume   65  quantification. It found a significant association between CF-PWV and WMLs volume 18. In comparison, our study only included 34 participants, and thus, we were likely underpowered to detect associations between total, deep or periventricular WMLs and CF-PWV   Our study included individuals who met the criteria for mild SIVCI and able to partake resistance training of moderate intensity for 12 months. Thus, they may be healthier and have better cardiovascular health than the average older adult with mild SIVCI. The mean systolic blood pressure for the RVCI cohort was 134.8 mmHg and diastolic blood pressure was 78 mmHg and almost half of the participant were not on antihypertensive medication, see table 6-2 supplementary appendix. The lower cardiovascular burden in our study participant could potentially explain the lack of the association of CF-PWV and WMLs volume.  The Rotterdam scan study included much younger individuals with mean age of 58.2 years old and average CF-PWV of 9.0 meters/second. In comparison, the RVCI cohort’s mean age was 75.2 and mean CF-PWV was 11.31 meters/second 18. Age is known to be associated with PWV, but despite adjusting for age, the relationship between total, deep or periventricular WMLs and CF-PWV did not change 125. A meta-analysis by Ben-Shlomo et .al looking at 17,635 participants found that PWV has stronger predictive value in younger participants compared to older individuals, which can potentially explain the difference between our results and the results from the Rotterdam scan study which had much younger individuals than ours 132.  Possible methodological difference in quantifying WMLs between studies could explain the difference in the association between WMLs and measures of interest including CF-PWV. We   66  used a semi-automated measure to quantify WMLs. It was originally validated for use in cases of multiple sclerosis and utilizes a different method of volume analysis than the fully automated software used in the Rotterdam study 18,120. Fully automated methods for WMLs segmentation and volume quantification has be criticized for the high false positive rate of WMLs selection 133,134. Supervised semi-automated methods were developed to overcome the problems related to the fully automated WMLs segmentation methods 134.  A previous study found a differential association between PWV and periventricular white matter severity. However, in that study, they only used a visual rating scale to assess the severity of WMLs 21. Moreover, the RVCI cohort included only participants diagnosed with mild SIVCI 111. Other studies that found significant association between PWV and WMLs either included individuals with normal cognitive function or participants diagnosed with AD 110,131. The pathophysiology of SIVCI although still not well understood, it has distinct pathological presentation from individuals with normal cognitive function or individuals with AD, therefore we can not extrapolate associations of PWV with clinical measures in those studies to our cohort 21,135,136.      The association of cognitive measures including executive function with different anatomical location of WMLs is still controversial. Previous studies have found an association between periventricular WMLs volume and executive function including the large population-based Rotterdam scan study 137-141. Another study found an association between high periventricular white matter hyperintensity volume and poor executive function and episodic memory 25. On the other hand, a smaller study found an association between deep WMLs and executive function 66.   67  Our study did not find an association between total and periventricular WMLs volume with measures of executive function or global cognitive function. Executive function was assessed by multiple measures including Trails B-A and delta Stroop. The prevalence of WMLs increases significantly with age 50. We found a significant association between Trails B-A and deep WMLs volume after controlling for age (r = 0.36 p value = 0.05), possibly indicating that the relationship between deep WMLs volume and executive function is independent of age. However, no significant association was found between delta Stroop and deep WMLs. These results may reflect that WMLs in different locations impact different cognitive processes. For example, a recent systematic review by Lam et.al found that different location and spatial distribution of WMLs were associated with different cognitive profiles 142.    Semantic fluency (category fluency animal test) is a common brief neuropsychological test that assesses multiple cognitive aspects including executive function, language and semantic memory 143-145. Previous studies found a significant association between poor performance on animal fluency test and amnestic MCI and AD 145,146. A recent study looking at the white matter confluency sum score as a marker for total white matter shape irregularity also included assessment of white matter hyperintensity volume and found a significant association of both white matter hyperintensity volume and the confluency sum score with poor performance on category fluency animal test 147.   This study found a significant association between deep WMLs volume and poor performance on category fluency animal test (bivariate r = - 0.39 and p value = 0.03). This association   68  remained significant after adjusting for age, sex and antihypertensive use (r = - 0.49 and p value = 0.006). However, there was no association between category fluency animal test and either total or periventricular WMLs volume. This finding is consistent with a previous study finding an association between total white matter hyperintensity volume and poor performance on category fluency animal test. However, that study did not categorize WMLs based on location, deep vs. periventricular and again included participants with a diagnosis of AD, vascular dementia and normal controls 147.   Taken together, these findings show that deep WMLs volume is associated with measures of executive function (Trails B-A) and category fluency animal test. The association with category fluency is an interesting finding, especially that this brief test has been linked more with language and semantic memory impairment than executive function 143.   Based on the previous studies linking poor performance on semantic category fluency for animals to AD and amnestic MCI, It is intriguing to entertain the possibility that this subset of individuals could progress to develop Alzheimer’s dementia mixed with VCI or have a rapid rate of cognitive decline 145. Previous studies have shown that white matter hyperintensity is related to AD pathology and vascular damage can aggravate amyloid pathology 135,148. It is not surprising to see this differential cognitive profile in a disease known to have a heterogeneous background and manifestations such as VCI 3,6,7.      Additionally, this study did not find an association between periventricular, deep or total WMLs volume and measures of global cognition such as ADAS-cog or MoCA, with or without   69  adjustment for age, sex and anti-hypertensive use. These findings are inconsistent with that of other studies that looked at the relationship of WMLs volume and global cognition 138,141,149-152.   One study was looking at regional WMLs volume and found a differential association between regional WMLs at different locations with global cognitive measures, reporting significant association only  between parietal WMLs volume and MoCA scores, whereas the temporal and hippocampal WMLs volume was associated with MMSE  scores 151. Although they were studying a cohort with SIVCI, they included patients with dementia and were a relatively younger cohort (mean age = 66.8) 151.   Another study looking at individuals with MCI found that participants with higher periventricular WMLs volume had rapid decline rate when assessed by MMSE over 3.8 years of follow up 152. That study recruited participants from memory clinics with MCI but not necessarily meeting criteria for SIVCI, possibly including participants with early neurodegenerative dementia such as AD 152. Compared to those studies, this study used strict inclusion criteria allowing only individuals who meet criteria for mild SIVCI, were overall older cohort with a mean age of 75.2, and were assessed by multiple well validated global cognitive screen tests 111,152-154. This study is also consistent with previous studies showing a weak association between WMLs and cognitive performance 155-157.  With regards to the association between total, deep and periventricular WMLs volume with mobility performance measured by gait speed and TUG test. This study did not find an association between periventricular WMLs volume and either gait speed or TUG test.   70  Additionally, there was no association between either the total or deep WMLs volume with either gait speed or TUG test.   This finding is discordant from the Three City longitudinal study that found slower gait speed to be significantly associated with periventricular WMLs volume 29. The Three City study included participants with normal cognitive function and measured gait speed over 6 meters compared to 4 meters in this study 29. Their median gait speed was 1.50 meters/second compared to median gait speed in this study of 1.15 meters/seconds which is significantly slower than that cohort 29. Moreover, the median periventricular WMLs volume in that study was 2.8 cubic centimeter (equivalent to 2800 cubic millimeter), whereas the median periventricular WMLs in this study was 5419.68 cubic millimeter which is significantly higher 29. This difference reflects the different cohort included in this study who had to be diagnosed with mild SIVCI potentially explaining the higher WMLs volume.   Individuals with SIVCI often present with poor mobility and gait abnormalities. However, our RVCI study participants had to be able exercise for 12 months and therefore, likely had better overall mobility than most individuals with SIVCI 89,111. The lack of the association of mobility performance measures and WMLs volume in our cohort could be explained by their overall better mobility performance. Moreover, other studies found that poor mobility performance is only associated with participants who have the most severe degree of WMLs volume 32. Only 6 participants of our cohort had a Fazekas score of 3, therefore we may be looking at a cohort with less WMLs severity burden explaining the lack of the association between their relatively mild WMLs volume and mobility performance 32.    71   Previous studies proposed a plateau effect of the suggested dose-response association of higher WMLs and clinical measures, creating a ceiling effect for WMLs volume in predicting clinical outcomes especially for the severe cases 158,159. Another possibility explaining the discordance of this study from the other studies is the heterogeneity of the underlying pathology involving WMLs or the possibility that similar WMLs volume could represent different stages of damage 158,160. This and the reduced power of this study could explain the lack of association between WMLs volume clinical measures of interest.   Blood pressure is known to be one of the major risk factors for WMLs and VCI 99,161.  Systolic hypertension is thought to be closely related to arterial stiffness and atherosclerosis involving large arteries, whereas diastolic hypertension is related more to small vessel disease and peripheral vascular resistance 34-36. A previous large prospective longitudinal study showed that diastolic blood pressure is closely associated with white matter hyperintensity volume in contrast to systolic blood pressure 36.  This study shows that diastolic blood pressure is significantly associated with deep but no total or periventricular WMLs volume before and after controlling for age (r = 0.59 p value = < 0.001 after controlling for age). This association remained significant after being analyzed in a model controlling for age, sex, and use of anti-hypertensive medications (r = 0.59 p value = 0.001).  The association of diastolic blood pressure with deep WMLs is consistent with a previous study showing similar association pattern 36.     72  With regards to the association between systolic blood pressure and deep WMLs volume, it was statistically significant before controlling for age (bivariate r =0.43 p value =0.01) but become insignificant after controlling just for age or when analyzed in a model controlling for age, sex, antihypertensive use. We had information about the participants’ medications including antihypertensive medications use. Therefore, we were able to control for the use of antihypertensive medication. In fact only 54.5% of our cohort were on one or more antihypertensive medication, the remaining 45.5 % were not using antihypertensive medications. Blood pressure is known to be independently associated with WMLs volume, so potentially successful treatment of high blood pressure with antihypertensive medications could have affected our negative association between systolic blood pressure and WMLs volume 37. However, given the limitations of our sample size we did not control for blood pressure to avoid over fitting of our model.  The differential association of diastolic blood pressure with deep as opposed to periventricular WMLs is inconsistent with a previous large study looking at 1,805 participants. That study found a significant link between diastolic blood pressure and periventricular WMLs severity 37. However, they only used a semi-quantitative measure to assess the severity of periventricular WMLs and the subcortical WMLs volume was only approximated by expert neuroradiologist which differs from this study where we used a semi-automated method to calculate WMLs volume for both periventricular and deep WMLs.   The finding of diastolic blood pressure and deep WMLs volume could be considered a reflection of the underlying pathology in SIVCI preferentially involving cerebral small vessel leading to   73  diastolic hypertension 162. Our results support the previous prospective population study by Guo et. al that found diastolic blood pressure to be significantly associated with WMLs volume, they proposed that increased peripheral resistance in midlife could be the underlying cause of WMLs development 162. With regards to the link between diastolic blood pressure and cognitive function, the study by Tsivgoulis et.al found that diastolic blood pressure was significantly and independently associated with cognitive impairment, they found that 10 mmHg raise in diastolic blood pressure was associated with 7% higher odds of cognitive impairment 163.  Treatment of VCI at advanced stages is proven to be very challenging and not very successful 6. Currently there is a shift in the field of dementia prevention to identify individuals at earlier stages of cognitive decline or even in the pre-symptomatic phase to further characterize their risk profile or find disease specific biomarkers. Our study is one of the few studies that address individuals with mild SIVCI. Therefore, our study could be helpful in enriching the current literature with better understanding of the vascular risk profile associated with WMLs volume at this stage of the disease.   In our cohort, most individuals with high deep WMLs volume also had the highest periventricular and total WMLs volumes (see supplementary table 6-3). It is possible that deep WMLs represent greater severity of small vessel disease. The finding of higher diastolic blood pressure with greater deep but not periventricular WML volume may support this hypothesis. Higher diastolic blood pressure has been shown to be predictive of WML progression and may reflect more advanced small vessel dysfunction 164. Our finding is supported by a previous study by Van Dijk et. al which showed that high diastolic blood pressure is associated with increase in   74  WMLs volume, however, in that same study lowering diastolic blood pressure was associated with increase in WMLs volume 37. They suggested that lowering diastolic blood pressure could further compromise blood perfusion to areas that are vulnerable to hypoperfusion leading to the increase in WMLs volume 37. Although diastolic blood pressure was not associated with higher WMLs volume, it is important to consider the possible negative effect of lowering diastolic blood pressure on WMLs volume when designing future studies that address diastolic blood pressure as potential therapeutic target in SIVCI participants 37.  This study has several limitations. First, the cross-sectional design only allows us to conclude associations between measures of interest such as diastolic blood pressure and deep WMLs volume but can not provide evidence of causality. Second, the small study sample likely limited our ability to detect associations (i.e., Type 2 error). Conversely, in the statistical analysis there were multiple testing without adjustment of p value, therefore, possible increase in Type 1 error rate. Third, this study recruited individuals who had to be eligible for exercise training, which could potentially have introduced a selection bias potentially minimizing the association between mobility performance measures such as gait speed and WMLs volume. Fourth, recruiting individuals from a memory clinic could have affected the selection of the sample included in this study potentially including participants with early mixed dementia or AD. Fifth, the CF-PWV was done by multiple study assistants at baseline which could have introduced a wide inter-rater variability affecting the accuracy and reproducibility of the results. Sixth, although we attempted to control for several confounders such as age, sex and antihypertensive use, several other confounders could have affected our analysis of the results.     75  This study has several strengths such as using stringent criteria to study subcategory of VCI rather than using just imaging surrogate markers for small vessel disease or including other subcategories of VCI. We used the gold standard CF-PWV method to study PWV. Finally, the semi-automated measure used to calculate WMLs has shown a strong inter-rater correlation.     76  Chapter 6: Conclusion  Neither total nor periventricular WMLs volumes were associated with CF-PWV, executive function, global cognitive function, mobility performance or blood pressure. CF-PWV is known to correlate with blood pressure and possibly having and independent association with systolic or diastolic blood pressure, however given the limitations of our sample size we did not control for blood pressure to avoid overfitting of our model. Deep WMLs volume have shown a differential association with executive function and animal fluency. Diastolic blood pressure was significantly associated with only deep WMLs volume. This study could be supportive of the ceiling effect of WMLs volume in predicting clinical outcomes.  Future studies looking into the associations of WMLs volume with clinical measures should take into account the limitations of WMLs volume in predicting clinical outcomes. There should also be a consensus on the gold standard method of quantifying WMLs volume on MRI to minimize the heterogeneity of results between studies. Future studies should use a standardized method in the classifying of WMLs into deep or periventricular or other sub-classifications based on severity or locations.  It would be helpful if future studies develop and validate new biomarkers that are specific and has the ability to reflect the disease severity for SIVCI pathology. This can be extremely helpful for selecting participants for research studies and provide a treatment target beside WMLs volume or other surrogate markers of small vessel disease in SIVCI individuals. Additionally, using validated biomarkers for neurodegenerative disease can help identify individuals with VCI mixed with neurodegenerative diseases such as AD. Doing so could improve our understanding   77  of the pathophysiology of SIVCI and minimize the noise introduced to the data by these age related neurodegenerative diseases commonly associated with VCI.  Currently there has been substantial progress in using new neuroimaging technology to identify microstructural changes that affects white matter integrity 165. Diffusion tensor imaging have shown significant association with cognitive decline and cerebral small vessel disease in multicenter study proving to be a reliable biomarker for disease progression 165. Future studies can utilize diffusion tensor imaging to examine the association of the earlier microstructural changes affecting the white matter in individuals with SIVCI with vascular risk factors.   Future studies should further examine the directionality of diastolic blood pressure in relation to deep WMLs. Although previous studies showed that high blood pressure precedes WMLs development and that treatment of hypertension may reduce the rate of progression of WMLs, future therapeutic studies should be cautious in reducing diastolic blood pressure significantly as it may in fact increase WMLs volume 37,166.    78  References  1. Wimo A, Guerchet M, Ali GC, et al. The worldwide costs of dementia 2015 and comparisons with 2010. Alzheimers & Dementia. 2017;13(1):1-7. 2. Schneider JA, Arvanitakis Z, Bang W, Bennett DA. Mixed brain pathologies account for most dementia cases in community-dwelling older persons. Neurology. 2007;69(24):2197-2204. 3. Sachdev P, Kalaria R, O'Brien J, et al. Diagnostic Criteria for Vascular Cognitive Disorders A VASCOG Statement. Alzheimer Disease & Associated Disorders. 2014;28(3):206-218. 4. Skrobot OA, Black SE, Chen C, et al. Progress toward standardized diagnosis of vascular cognitive impairment: Guidelines from the Vascular Impairment of Cognition Classification Consensus Study. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2017. 5. Smith EE. Clinical presentations and epidemiology of vascular dementia. Clinical Science. 2017;131(11):1059-1068. 6. Dichgans M, Leys D. Vascular Cognitive Impairment. Circulation Research. 2017;120(3):573-591. 7. O'Brien JT, Erkinjuntti T, Reisberg B, et al. Vascular cognitive impairment. Lancet Neurology. 2003;2(2):89-98. 8. Chui HC. Subcortical ischemic vascular dementia. Neurologic Clinics. 2007;25(3):717-+. 9. Jellinger KA. Morphologic diagnosis of "vascular dementia" - A critical update. Journal of the Neurological Sciences. 2008;270(1-2):1-12. 10. McAleese KE, Walker L, Graham S, et al. Parietal white matter lesions in Alzheimer's disease are associated with cortical neurodegenerative pathology, but not with small vessel disease. Acta Neuropathologica. 2017;134(3):459-473. 11. Sethi S, Rivera O, Oliveros R, Chilton R. Aortic stiffness: pathophysiology, clinical implications, and approach to treatment. Integrated blood pressure control. 2014;7:29-34. 12. Safar ME. Arterial stiffness as a risk factor for clinical hypertension. Nature Reviews Cardiology. 2018;15(2):97-105. 13. Tjeerdema N, Van Schinkel LD, Westenberg JJ, et al. Aortic stiffness is associated with white matter integrity in patients with type 1 diabetes. European Radiology. 2014;24(9):2031-2037. 14. Laurent S, Cockcroft J, Van Bortel L, et al. Expert consensus document on arterial stiffness: methodological issues and clinical applications. European Heart Journal. 2006;27(21):2588-2605. 15. Aznaouridis K, Vlachopoulos C, Masoura C, et al. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a meta-analysis of longitudinal studies. European Heart Journal. 2010;31:617-617. 16. Saji N, Toba K, Sakurai T. Cerebral Small Vessel Disease and Arterial Stiffness: Tsunami Effect in the Brain? Pulse. 2015;3(3-4):182-189. 17. O'Rourke MF, Safar ME. Relationship between aortic stiffening and microvascular disease in brain and kidney - Cause and logic of therapy. Hypertension. 2005;46(1):200-204.   79  18. Poels MMF, Zaccai K, Verwoert GC, et al. Arterial Stiffness and Cerebral Small Vessel Disease The Rotterdam Scan Study. Stroke. 2012;43(10):2637-+. 19. Mitchell GF, van Buchem MA, Sigurdsson S, et al. Arterial stiffness, pressure and flow pulsatility and brain structure and function: the Age, Gene/Environment Susceptibility - Reykjavik Study. Brain. 2011;134:3398-3407. 20. Henskens LHG, Kroon AA, van Oostenbrugge RJ, et al. Increased Aortic Pulse Wave Velocity Is Associated With Silent Cerebral Small-Vessel Disease in Hypertensive Patients. Hypertension. 2008;52(6):1120-U1172. 21. Ohmine T, Miwa Y, Yao H, et al. Association between arterial stiffness and cerebral white matter lesions in community-dwelling elderly subjects. Hypertension Research. 2008;31(1):75-81. 22. Cognitive profile of subcortical ischaemic vascular disease. 2006. 23. Chan RCK, Shum D, Toulopoulou T, Chen EYH. Assessment of executive functions: Review of instruments and identification of critical issues. Archives of Clinical Neuropsychology. 2008;23(2):201-216. 24. Parks CM, Iosif AM, Farias S, Reed B, Mungas D, DeCarli C. Executive function mediates effects of white matter hyperintensities on episodic memory. Neuropsychologia Neuropsychologia. 2011;49(10):2817-2824. 25. Smith EE, Salat DH, Jeng J, et al. Correlations between MRI white matter lesion location and executive function and episodic memory. Neurology. 2011;76(17):1492-1499. 26. Soriano-Raya JJ, Miralbell J, Lopez-Cancio E, et al. Deep versus Periventricular White Matter Lesions and Cognitive Function in a Community Sample of Middle-Aged Participants. JOURNAL- INTERNATIONAL NEUROPSYCHOLOGICAL SOCIETY. 2012;18(5):874-885. 27. Baune BT, Roesler A, Knecht S, Berger K. Single and Combined Effects of Cerebral White Matter Lesions and Lacunar Infarctions on Cognitive Function in an Elderly Population. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 2009;64A(1):118-124. 28. Who. International classification of functioning, disability and health. Geneva: World Health Organization; 2001. 29. Soumare A, Elbaz A, Zhu YC, et al. White Matter Lesions Volume and Motor Performances in the Elderly. Annals of Neurology. 2009;65(6):706-715. 30. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. LOWER-EXTREMITY FUNCTION IN PERSONS OVER THE AGE OF 70 YEARS AS A PREDICTOR OF SUBSEQUENT DISABILITY. New England Journal of Medicine. 1995;332(9):556-561. 31. Menezes K, Auger C, Menezes WRD, Guerra RO. Instruments to evaluate mobility capacity of older adults during hospitalization: A systematic review. Archives of Gerontology and Geriatrics. 2017;72:67-79. 32. Zheng JJJ, Delbaere K, Close JCT, Sachdev PS, Lord SR. Impact of White Matter Lesions on Physical Functioning and Fall Risk in Older People A Systematic Review. Stroke. 2011;42(7):2086-2090. 33. Impact of white matter hyperintensity volume progression on rate of cognitive and motor decline. 2008. 34. Blood pressure and ageing. 2007.   80  35. Benetos A, Thomas F, Safar ME, Bean KE, Guize L. Should diastolic and systolic blood pressure be considered for cardiovascular risk evaluation: a study in middle-aged men and women. Journal of the American College of Cardiology Journal of the American College of Cardiology. 2001;37(1):163-168. 36. Marcus J, Gardener H, Rundek T, et al. Baseline and Longitudinal Increases in Diastolic Blood Pressure Are Associated With Greater White Matter Hyperintensity Volume: The Northern Manhattan Study. Stroke Stroke. 2011;42(9):2639-2641. 37. van Dijk EJ, Breteler MM, Schmidt R, et al. The association between blood pressure, hypertension, and cerebral white matter lesions: cardiovascular determinants of dementia study. Hypertension (Dallas, Tex. : 1979). 2004;44(5):625-630. 38. Wortmann M. Dementia: a global health priority - highlights from an ADI and World Health Organization report. Alzheimers Research & Therapy. 2012;4(5). 39. Eskes GA, Lanctot KL, Herrmann N, et al. Canadian Stroke Best Practice Recommendations: Mood, Cognition and Fatigue Following Stroke practice guidelines, update 2015. International Journal of Stroke. 2015;10(7):1130-1140. 40. O'Brien JT, Thomas A. Vascular dementia. The Lancet The Lancet. 2015;386(10004):1698-1706. 41. Thal DR, Grinberg LT, Attems J. Vascular dementia: Different forms of vessel disorders contribute to the development of dementia in the elderly brain. Experimental Gerontology. 2012;47(11):816-824. 42. Prins ND, Scheltens P. White matter hyperintensities, cognitive impairment and dementia: an update. Nature Reviews Neurology. 2015;11(3). 43. Fields RD. Change in the Brain's White Matter. Science. 2010;330(6005):768. 44. Debette S, Markus HS. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. British Medical Journal. 2010;341. 45. Barber R, Scheltens F, Gholkar A, et al. White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer's disease, vascular dementia, and normal aging. Journal of Neurology Neurosurgery and Psychiatry. 1999;67(1):66-72. 46. Wardlaw JM, Hernandez MCV, Munoz-Maniega S. What are White Matter Hyperintensities Made of? Relevance to Vascular Cognitive Impairment. Journal of the American Heart Association. 2015;4(6). 47. Fisher CM. LACUNAR STROKES AND INFARCTS - A REVIEW. Neurology. 1982;32(8):871-876. 48. Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer's dementia and normal aging. AJR. American journal of roentgenology. 1987;149(2):351-356. 49. Kim KW, MacFall JR, Payne ME. Classification of white matter lesions on magnetic resonance imaging in elderly persons. Biological Psychiatry. 2008;64(4):273-280. 50. Launer LJ. Epidemiology of white matter lesions. Topics in magnetic resonance imaging : TMRI. 2004;15(6):365-367. 51. Sawada H, Ibi M, Kihara T, et al. Mechanisms of antiapoptotic effects of estrogens in nigral dopaminergic neurons. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 2000;14(9):1202-1214. 52. Cerebral white matter lesions are not associated with apoE genotype but with age and female sex in Alzheimer's disease. 2000.   81  53. Lindemer ER, Greve DN, Fischl B, Augustinack JC, Salat DH, Alzheimers Dis N. Differential Regional Distribution of Juxtacortical White Matter Signal Abnormalities in Aging and Alzheimer's Disease. Journal of Alzheimers Disease. 2017;57(1):293-303. 54. Pantoni L, Garcia JH. Pathogenesis of leukoaraiosis - A review. Stroke. 1997;28(3):652-659. 55. Wardlaw JM, Smith C, Dichgans M. Mechanisms of sporadic cerebral small vessel disease: insights from neuroimaging. Lancet Neurology. 2013;12(5):483-497. 56. Pantoni L. Pathophysiology of age-related cerebral white matter changes. Cerebrovascular Diseases. 2002;13:7-10. 57. McAleese KE, Alafuzoff I, Charidimou A, et al. Post-mortem assessment in vascular dementia: advances and aspirations. Bmc Medicine. 2016;14. 58. Stokin GB, Lillo C, Falzone TL, et al. Axonopathy and transport deficits early in the pathogenesis of Alzheimer's disease. Science. 2005;307(5713):1282-1288. 59. Joutel A, Chabriat H. Pathogenesis of white matter changes in cerebral small vessel diseases: beyond vessel-intrinsic mechanisms. Clinical Science. 2017;131(8):635-651. 60. Oxford University P. Oxford language dictionaries online. 2007. 61. Sachdev PS, Blacker D, Blazer DG, et al. Classifying neurocognitive disorders: the DSM-5 approach. Nature Reviews Neurology. 2014;10(11):634-642. 62. Thomas S, Rao SL, Devi BI. Standardization of Tests of Attention and Inhibition. Indian journal of psychological medicine. 2016;38(4):320-325. 63. Budson AE, Price BH. Current concepts - Memory dysfunction. New England Journal of Medicine. 2005;352(7):692-699. 64. Gunning-Dixon FM, Raz N. The cognitive correlates of white matter abnormalities in normal aging: A quantitative review. Neuropsychology. 2000;14(2):224-232. 65. Rankin JK, Woollacott MH, Shumway-Cook A, Brown LA. Cognitive influence on postural stability: A neuromuscular analysis in young and older adults. Journals of Gerontology Series a-Biological Sciences and Medical Sciences. 2000;55(3):M112-M119. 66. Ghanavati T, Smitt MS, Lord SR, et al. Deep white matter hyperintensities, microstructural integrity and dual task walking in older people. Brain imaging and behavior. 2018. 67. Gill TM, Allore H, Holford TR, Guo ZC. The development of insidious disability in activities of daily living among community-living older persons. American Journal of Medicine. 2004;117(7):484-491. 68. Afilalo J, Eisenberg MJ, Morin JF, et al. Gait speed as an incremental predictor of mortality and major morbidity in elderly patients undergoing cardiac surgery. J Am Coll Cardiol. 2010;56(20):1668-1676. 69. van der Holst HM, van Uden IWM, Tuladhar AM, et al. Factors Associated With 8-Year Mortality in Older Patients With Cerebral Small Vessel Disease The Radboud University Nijmegen Diffusion Tensor and Magnetic Resonance Cohort (RUN DMC) Study. Jama Neurology. 2016;73(4):402-409. 70. Bolandzadeh N, Tam R, Handy TC, et al. Resistance Training and White Matter Lesion Progression in Older Women: Exploratory Analysis of a 12-Month Randomized Controlled Trial. Journal of the American Geriatrics Society. 2015;63(10):2052-2060.   82  71. Suo C, Singh MF, Gates N, et al. Therapeutically relevant structural and functional mechanisms triggered by physical and cognitive exercise. Molecular psychiatry. 2016;21(11):1633-1642. 72. Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR SIGNAL ABNORMALITIES AT 1.5-T IN ALZHEIMER DEMENTIA AND NORMAL AGING. American Journal of Roentgenology. 1987;149(2):351-356. 73. Boutet C, Rouffiange-Leclair L, Schneider F, Camdessanche JP, Antoine JC, Barral FG. Visual Assessment of Age-Related White Matter Hyperintensities Using FLAIR Images at 3 T: Inter- and Intra-Rater Agreement. Neurodegenerative Diseases. 2016;16(3-4):279-283. 74. Payne ME, Fetzer DL, MacFall JR, Provenzale JM, Byrum CE, Krishnan KRR. Development of a semi-automated method for quantification of MRI gray and white matter lesions in geriatric subjects. Psychiatry Research-Neuroimaging. 2002;115(1-2):63-77. 75. Davis PC, Gray L, Albert M, et al. THE CONSORTIUM TO ESTABLISH A REGISTRY FOR ALZHEIMERS-DISEASE (CERAD) .3. RELIABILITY OF A STANDARDIZED MRI EVALUATION OF ALZHEIMERS-DISEASE. Neurology. 1992;42(9):1676-1680. 76. Hong Y-J, Yang DW, 심 용 수 , 윤 보 라 , 조 아 현 . A Semi-Automated Method for Measuring White Matter Hyperintensity Volume. Dementia and Neurocognitive Disorders. 2013;12(1):21-28. 77. Bondareff W, Raval J, Woo B, Hauser DL, Colletti PM. MAGNETIC-RESONANCE IMAGING AND THE SEVERITY OF DEMENTIA IN OLDER ADULTS. Archives of General Psychiatry. 1990;47(1):47-51. 78. Boone KB, Miller BL, Lesser IM, et al. NEUROPSYCHOLOGICAL CORRELATES OF WHITE-MATTER LESIONS IN HEALTHY ELDERLY SUBJECTS - A THRESHOLD EFFECT. Archives of Neurology. 1992;49(5):549-554. 79. Kumar A, Bilker W, Jin ZS, Udupa J. Atrophy and high intensity lesions: Complementary neurobiological mechanisms in late-life major depression. Neuropsychopharmacology. 2000;22(3):264-274. 80. Hirono N, Kitagaki H, Kazui H, Hashimoto M, Mori E. Impact of white matter changes on clinical manifestation of Alzheimer's disease - A quantitative study. Stroke. 2000;31(9):2182-2188. 81. Blair GW, Hernandez MV, Thrippleton MJ, Doubal FN, Wardlaw JM. Advanced Neuroimaging of Cerebral Small Vessel Disease. Current treatment options in cardiovascular medicine. 2017;19(7):56. 82. Lecun Y, Bengio Y, Hinton G. Deep learning. Nature Nature. 2015;521(7553):436-444. 83. Akkus Z, Galimzianova A, Hoogi A, Rubin DL, Erickson BJ. Deep Learning for Brain MRI Segmentation: State of the Art and Future Directions. Journal of digital imaging : the official journal of the Society for Computer Applications in Radiology. 2017;30(4):449-459. 84. De Guio F, Jouvent E, Biessels GJ, et al. Reproducibility and variability of quantitative magnetic resonance imaging markers in cerebral small vessel disease. Journal of Cerebral Blood Flow and Metabolism. 2016;36(8):1319-1337.   83  85. Sacco RL, Kasner SE, Broderick JP, et al. An Updated Definition of Stroke for the 21st Century A Statement for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke. 2013;44(7):2064-2089. 86. Hachinski VC, Lassen NA, Marshall J. MULTI-INFARCT DEMENTIA:A CAUSE OF MENTAL DETERIORATION IN THE ELDERLY. LANCET</cja:jid> The Lancet. 1974;304(7874):207-209. 87. Olsen CG, Clasen ME. Senile Dementia of the Binswanger's Type. AMERICAN FAMILY PHYSICIAN. 1998;58(9):2068-2076. 88. Brief cognitive assessment for patients with cerebral small vessel disease. 2005. 89. Huisa BN, Rosenberg GA. Binswanger's disease: toward a diagnosis agreement and therapeutic approach. EXPERT REVIEW OF NEUROTHERAPEUTICS. 2014;14(10):1203-1213. 90. Bondi MW, Edmonds EC, Salmon DP. Alzheimer's Disease: Past, Present, and Future. Journal of the International Neuropsychological Society. 2017;23(9-10):818-831. 91. Huber SJ, Shuttleworth EC, Paulson GW, Bellchambers MJG, Clapp LE. CORTICAL VS SUBCORTICAL DEMENTIA - NEUROPSYCHOLOGICAL DIFFERENCES. Archives of Neurology. 1986;43(4):392-394. 92. Lanctot KL, Aguera-Ortiz L, Brodaty H, et al. Apathy associated with neurocognitive disorders: Recent progress and future directions. Alzheimers & Dementia. 2017;13(1):84-100. 93. Berardelli A, Rothwell JC, Thompson PD, Hallet M. Pathophysiology of bradykinesia in Parkinson's disease. Brain. 2001;124:2131-2146. 94. Baradaran N, Tan SN, Liu AP, et al. Parkinson's disease rigidity: relation to brain connectivity and motor performance. Frontiers in Neurology. 2013;4. 95. Zackowski KM. Gait and Balance Assessment. Seminars in Neurology. 2016;36(5):474-478. 96. Sinanovic O, Mrkonjic Z, Zukic S, Vidovic M, Imamovic K. POST-STROKE LANGUAGE DISORDERS. Acta Clinica Croatica. 2011;50(1):79-94. 97. Greene JDW. Apraxia, agnosias, and higher visual function abnormalities. Journal of Neurology Neurosurgery and Psychiatry. 2005;76:V25-V34. 98. Zadikoff C, Lang AE. Apraxia in movement disorders. Brain. 2005;128:1480-1497. 99. 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(9):2672-2713. 100. Singer J, Trollor JN, Baune BT, Sachdev PS, Smith E. Arterial stiffness, the brain and cognition: A systematic review. Ageing Research Reviews. 2014;15:16-27. 101. Pereira T, Maldonado J, Coutinho R, et al. Invasive validation of the Complior Analyse in the assessment of central artery pressure curves: a methodological study. Blood Pressure Monitoring. 2014;19(5):280-287. 102. Asmar R, Benetos A, Topouchian J, et al. Assessment of arterial distensibility by automatic pulse wave velocity measurement. Validation and clinical application studies. Hypertension (Dallas, Tex. : 1979). 1995;26(3):485-490. 103. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of Cardiovascular Events and All-Cause Mortality With Arterial Stiffness A Systematic Review and Meta-Analysis. Journal of the American College of Cardiology. 2010;55(13):1318-1327.   84  104. Cooper LL, Palmisano JN, Benjamin EJ, et al. Microvascular Function Contributes to the Relation Between Aortic Stiffness and Cardiovascular Events: The Framingham Heart Study. Circulation. Cardiovascular imaging. 2016;9(12). 105. Mitchell GF, Hwang SJ, Vasan RS, et al. Arterial Stiffness and Cardiovascular Events: The Framingham Heart Study. Circulation Circulation. 2010;121(4):505-511. 106. Laurent S, Boutouyrie P, Asmar R, et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension (Dallas, Tex. : 1979). 2001;37(5):1236-1241. 107. Meaume S, Benetos A, Henry OF, Rudnichi A, Safar ME. Aortic pulse wave velocity predicts cardiovascular mortality in subjects >70 years of age. Arteriosclerosis, thrombosis, and vascular biology. 2001;21(12):2046-2050. 108. Al Hazzouri AZ, Newman AB, Simonsick E, et al. Pulse Wave Velocity and Cognitive Decline in Elders The Health, Aging, and Body Composition Study. Stroke. 2013;44(2):388-393. 109. van Sloten TT, Protogerou AD, Henry RMA, Schram MT, Launer LJ, Stehouwer CDA. Association between arterial stiffness, cerebral small vessel disease and cognitive impairment: A systematic review and meta-analysis. Neuroscience and Biobehavioral Reviews. 2015;53:121-130. 110. Poels MMF, van Oijen M, Mattace-Raso FUS, et al. Arterial stiffness, cognitive decline, and risk of dementia - The Rotterdam study. Stroke. 2007;38(3):888-892. 111. Erkinjuntti T, Inzitari D, Pantoni L, et al. Research criteria for subcortical vascular dementia in clinical trials. Journal of Neural Transmission-Supplement. 2000(59):23-30. 112. Graf P, Uttl B, Tuokko H. COLOR-WORD AND PICTURE-WORD STROOP TESTS - PERFORMANCE CHANGES IN OLD-AGE. Journal of Clinical and Experimental Neuropsychology. 1995;17(3):390-415. 113. Strauss E, Sherman EMS, Spreen O. A compendium of neuropsychological tests : administration, norms, and commentary. Oxford; New York: Oxford University Press; 2006. 114. Wechsler D, Psychological C. Wechsler Adult Intelligence Scale (WAIS-IIIuk) ; Wechsler Memory Scale (WMS-IIIuk). Sidcup: Psychological Corporation; 1998. 115. Rosen WG, Mohs RC, Davis KL. A NEW RATING-SCALE FOR ALZHEIMERS-DISEASE. American Journal of Psychiatry. 1984;141(11):1356-1364. 116. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. Journal of psychiatric research. 1975;12(3):189-198. 117. O'Bryant SE, Humphreys JD, Smith GE, et al. Detecting dementia with the mini-mental state examination in highly educated individuals. Archives of Neurology. 2008;65(7):963-967. 118. Nasreddine ZS, Phillips NA, Bedirian V, et al. The montreal cognitive assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society. 2005;53(4):695-699. 119. Podsiadlo D, Richardson S. THE TIMED UP AND GO - A TEST OF BASIC FUNCTIONAL MOBILITY FOR FRAIL ELDERLY PERSONS. Journal of the American Geriatrics Society. 1991;39(2):142-148.   85  120. McAusland J, Tam RC, Wong E, Riddehough A, Li DKB. Optimizing the Use of Radiologist Seed Points for Improved Multiple Sclerosis Lesion Segmentation. Ieee Transactions on Biomedical Engineering. 2010;57(11):2689-2698. 121. Sled JG, Zijdenbos AP, Evans AC. A nonparametric method for automatic correction of intensity nonuniformity in MRI data. Ieee Transactions on Medical Imaging. 1998;17(1):87-97. 122. Smith SM, Brady JM. SUSAN - A new approach to low level image processing. International Journal of Computer Vision. 1997;23(1):45-78. 123. Smith SM. Fast robust automated brain extraction. Human Brain Mapping. 2002;17(3):143-155. 124. Babyak MA. What you see may not be what you get: a brief, nontechnical introduction to overfitting in regression-type models. Psychosomatic medicine. 2004;66(3). 125. Mattace-Raso FUS, Hofman A, Verwoert GC, et al. Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: 'establishing normal and reference values'. European Heart Journal. 2010;31(19):2338-2350. 126. Tap L, van Opbroek A, Niessen W, Smits M, Mattace- Raso F. Vascular aging is associated with the severity of cerebral white matter lesion load. ARTRES Artery Research. 2017;20:100-101. 127. Rosano C, Watson N, Chang Y, et al. Aortic Pulse Wave Velocity Predicts Focal White Matter Hyperintensities in a Biracial Cohort of Older Adults. Hypertension. 2013;61(1):160-+. 128. Webb AJS, Simoni M, Mazzucco S, Kuker W, Schulz U, Rothwell PM. Increased Cerebral Arterial Pulsatility in Patients With Leukoaraiosis Arterial Stiffness Enhances Transmission of Aortic Pulsatility. Stroke. 2012;43(10):2631-+. 129. King KS, Chen KX, Hulsey KM, et al. White Matter Hyperintensities: Use of Aortic Arch Pulse Wave Velocity to Predict Volume Independent of Other Cardiovascular Risk Factors. Radiology. 2013;267(3):709-717. 130. Saji N, Shimizu H, Kawarai T, Tadano M, Kita Y, Yokono K. Increased Brachial-Ankle Pulse Wave Velocity Is Independently Associated with White Matter Hyperintensities. Neuroepidemiology. 2011;36(4):252-257. 131. Hughes TM, Wagenknecht LE, Craft S, et al. Arterial stiffness and dementia pathology: Atherosclerosis Risk in Communities (ARIC)-PET Study. Neurology. 2018. 132. Ben-Shlomo Y, Spears M, Boustred C, et al. Aortic Pulse Wave Velocity Improves Cardiovascular Event Prediction An Individual Participant Meta-Analysis of Prospective Observational Data From 17,635 Subjects. Journal of the American College of Cardiology. 2014;63(7):636-646. 133. Gibson E, Gao F, Black SE, Lobaugh NJ. Automatic segmentation of white matter hyperintensities in the elderly using FLAIR images at 3T. Journal of magnetic resonance imaging : JMRI. 2010;31(6):1311-1322. 134. Maria Eugenia C, Paolo P, Antonio A, Federico R, Aldo Q, Andrea C. Automatic Detection of White Matter Hyperintensities in Healthy Aging and Pathology Using Magnetic Resonance Imaging: A Review. Neuroinformatics. 2015;13(3):261-276. 135. Iadecola C. The Pathobiology of Vascular Dementia. Neuron. 2013;80(4):844-866. 136. Hughes TM, Kuller LH, Barinas-Mitchell EJ, et al. Pulse wave velocity is associated with b-amyloid deposition in the brains of very elderly adults. Neurology. 2013;81(19):1711-1718.   86  137. de Groot JC, de Leeuw FE, Oudkerk M, et al. Cerebral white matter lesions and cognitive function: The Rotterdam Scan Study. Annals of Neurology. 2000;47(2):145-151. 138. Prins ND, van Dijk EJ, den Heijer T, et al. Cerebral small-vessel disease and decline in information processing speed, executive function and memory. Brain. 2005;128:2034-2041. 139. Uiterwijk R, van Oostenbrugge RJ, Huijts M, De Leeuw PW, Kroon AA, Staals J. Total Cerebral Small Vessel Disease MRI Score Is Associated with Cognitive Decline in Executive Function in Patients with Hypertension. Frontiers in Aging Neuroscience. 2016;8. 140. de Groot JC, de Leeuw FE, Oudkerk M, et al. Periventricular cerebral white matter lesions predict rate of cognitive decline. Annals of Neurology. 2002;52(3):335-341. 141. Delano-Wood L, Abeles N, Sacco JM, Wierenga CE, Horne NR, Bozoki A. Regional white matter pathology in mild cognitive impairment - Differential influence of lesion type on neuropsychological functioning. Stroke. 2008;39(3):794-799. 142. Lam CLM, Yiend J, Lee TMC. Imaging and neuropsychological correlates of white matter lesions in different subtypes of Mild Cognitive Impairment: A systematic review. Neurorehabilitation. 2017;41(1):189-204. 143. Whiteside DM, Kealey T, Semla M, et al. Verbal Fluency: Language or Executive Function Measure? Applied Neuropsychology: Adult Applied Neuropsychology: Adult. 2016;23(1):29-34. 144. Ruff RM, Light RH, Parker SB, Levin HS. The psychological construct of word fluency. Brain and language. 1997;57(3):394-405. 145. Murphy KI, Rich JB, Troyer AK. Verbal fluency patterns in amnestic mild cognitive impairment are characteristic of Alzheimer's type dementia. Journal of the International Neuropsychological Society. 2006;12(4):570-574. 146. Butters N, Granholm E, Salmon DP, Grant I, Wolfe J. EPISODIC AND SEMANTIC MEMORY - A COMPARISON OF AMNESIC AND DEMENTED PATIENTS. Journal of Clinical and Experimental Neuropsychology. 1987;9(5):479-497. 147. Lange C, Suppa P, Maeurer A, et al. Mental speed is associated with the shape irregularity of white matter MRI hyperintensity load. Brain Imaging and Behavior. 2017;11(6):1720-1730. 148. Kester MI, Goos JDC, Teunissen CE, et al. Associations Between Cerebral Small-Vessel Disease and Alzheimer Disease Pathology as Measured by Cerebrospinal Fluid Biomarkers. Jama Neurology. 2014;71(7):855-862. 149. Carmichael O, Schwarz C, Drucker D, et al. Longitudinal Changes in White Matter Disease and Cognition in the First Year of the Alzheimer Disease Neuroimaging Initiative. Archives of Neurology. 2010;67(11):1370-1378. 150. Garde E, Mortensen EL, Rostrup E, Paulson OB. Decline in intelligence is associated with progression in white matter hyperintensity volume. Journal of Neurology Neurosurgery and Psychiatry. 2005;76(9):1289-1291. 151. Ai Q, Pu YH, Sy C, Liu LP, Gao PY. Impact of regional white matter lesions on cognitive function in subcortical vascular cognitive impairment. Neurological Research. 2014;36(5):434-443. 152. Debette S, Bombois S, Bruandet A, et al. Subcortical hyperintensities are associated with cognitive decline in patients with mild cognitive impairment. Stroke. 2007;38(11):2924-2930.   87  153. Pasi M, Salvadori E, Poggesi A, et al. White matter microstructural damage in small vessel disease is associated with Montreal cognitive assessment but not with mini mental state examination performances: vascular mild cognitive impairment Tuscany study. Stroke. 2015;46(1):262-264. 154. Lawrence AJ, Patel B, Morris RG, et al. Mechanisms of Cognitive Impairment in Cerebral Small Vessel Disease: Multimodal MRI Results from the St George's Cognition and Neuroimaging in Stroke (SCANS) Study. Plos One. 2013;8(4). 155. Sabri O, Ringelstein EB, Hellwig D, et al. Neuropsychological Impairment Correlates With Hypoperfusion and Hypometabolism but Not With Severity of White Matter Lesions on MRI in Patients With Cerebral Microangiopathy. Stroke Stroke. 1999;30(3):556-566. 156. Garde E. Relation between age-related decline in intelligence and cerebral white-matter hyperintensities in healthy octogenarians. The lancet. 2000(9230):628-634. 157. Schmidt R, Schmidt H, Kapeller P, Lechner A, Fazekas F. Evolution of White Matter Lesions. CEREBROVASCULAR DISEASES -BASEL-. 2002;13:16-20. 158. Tomimoto H. White matter integrity and cognitive dysfunction: Radiological and neuropsychological correlations. Geriatrics & Gerontology International. 2015;15:3-9. 159. van Straaten ECW, Fazekas F, Rostrup E, et al. Impact of white matter hyperintensities scoring method on correlations with clinical data - The LADIS study. Stroke. 2006;37(3):836-840. 160. Gouw AA, Seewann A, van der Flier WM, et al. Heterogeneity of small vessel disease: a systematic review of MRI and histopathology correlations. Journal of Neurology Neurosurgery and Psychiatry. 2011;82(2):126-135. 161. Aronow WS. Hypertension and cognitive impairment. Annals of translational medicine. 2017;5(12). 162. Guo X, Pantoni L, Simoni M, et al. Blood Pressure Components and Changes in Relation to White Matter Lesions: A 32-Year Prospective Population Study. Hypertension Hypertension. 2009;54(1):57-62. 163. Association of higher diastolic blood pressure levels with cognitive impairment. 2009. 164. Schmidt R, Fazekas F, Enzinger C, Ropele S, Kapeller P, Schmidt H. Risk factors and progression of small vessel disease-related cerebral abnormalities. 2002. 165. Croall ID, Lohner V, Moynihan B, et al. Using DTI to assess white matter microstructure in cerebral small vessel disease (SVD) in multicentre studies. Clinical Science. 2017;131(12):1361-1373. 166. Verhaaren BFJ, Vernooij MW, de Boer R, et al. High Blood Pressure and Cerebral White Matter Lesion Progression in the General Population. Hypertension. 2013;61(6):1354-+.     88  Appendices : Supplementary tables Table 6-1 Fazekas score   Frequency percent Valid percent Cumulative percent Valid     0 3 9.1 9.1 9.1 1 11 33.3 33.3 42.4 2 13 39.4 39.4 81.8 3 6 18.2 18.2 100 total 33 100 100   Table 6-2 Antihypertensive use   Frequency percent Valid percent Cumulative percent HTN medication 1+ Valid    0 15 45.5 45.5 45.5 1 18 54.5 54.5 100 total 33 100 100  HTN medication 2+      Valid    0 27 81.8 81.8 81.8 1 6 18.2 18.2 100 total 33 100 100  HTN: hypertension                             89  Table 6-3 Baseline WMLs volumes for the RVCI participants                                             Participant number WML total Periventricular WMLs Deep WMLs RVCI 020 39028.0 20807.1 23504.8 RVCI 016 19501.1 7793.0 12576.9 RVCI 017 27422.8 18912.4 11410.9 RVCI 023 19063.9 12322.5 10502.2 RVCI 005 18612.4 11079.4 8664.6 RVCI 001 22458.9 14940.2 8661.8 RVCI 009 17589.3 10867.9 7150.0 RVCI 030 19535.4 13948.6 5938.4 RVCI 004 39796.7 35781.6 4746.7 RVCI 032 28974.5 27108.4 3303.5 RVCI 034 7558.7 5272.5 2634.8 RVCI 014 21187.2 18838.1 2594.8 RVCI 019 6912.8 4955.3 2063.3 RVCI 028 8141.7 6187.0 2063.3 RVCI 003 5292.5 3400.7 2049.0 RVCI 010 11613.8 10033.5 1763.2 RVCI 027 11462.3 9953.5 1557.5 RVCI 011 16849.2 15829.0 1488.9 RVCI 012 4060.8 2949.2 1077.4 RVCI 002 12093.9 11179.4 1074.5 RVCI 022 5915.5 4158.0 960.2 RVCI 026 5089.6 4012.3 803.0 RVCI 018 1900.4 1163.1 774.4 RVCI 029 6398.5 5566.9 763.0 RVCI 006 3695.0 2989.2 754.4 RVCI 008 3626.5 2826.3 660.1 RVCI 007 1674.6 1088.8 628.7 RVCI 033 951.6 560.1 397.2 RVCI 013 2163.3 1831.8 345.8 RVCI 025 1260.3 928.8 325.8 RVCI 024 4435.2 4135.1 285.8 RVCI 031 703.0 557.3 168.6 RVCI 015 1831.8 1654.6 148.6 RVCI 021 3074.9 3052.1 31.4   90    : Supplementary figures       Figure 6-1 Total WMLs volume distribution for the RVCI cohort.  This graph shows the distribution of the total WMLs volume for the RVCI cohort. The skewed distribution of the values can be easily appreciated in this figure. The mean and median values for the total WMLs respectively are 11761.06 & 7235.76 mm3.      91   Figure 6-2 Log transformation for total WMLs. This graph shows the normalized distribution of total WMLs after Log transforming the data for optimal statistical analysis.           92   Figure 6-3 Distribution of deep WMLs. This graph shows the distribution of deep WMLs for the RVCI cohort. The mean and median values respectively are 3584.51&1523.16 mm3. Again the skewed distribution shows here as well.       93   Figure 6-4 Log transformation of deep WMLs. This graph shows the normalized distribution of deep WMLs after Log transforming the data for optimal statistical analysis.   94   Figure 6-5 Distribution of periventricular WMLs This graph shows the distribution of periventricular WMLs for the RVCI cohort. The mean and median values respectively are 8725.98 & 5419.68 mm3. Again the skewed distribution can be appreciated for the periventricular WMLs as well.   95     Figure 6-6 Log transformation of periventricular WMLs. This graph shows the normalized distribution of periventricular WMLs after Log transforming the data for optimal statistical analysis.     96   Figure 6-7 ADAS –Cog distribution for the RVCI cohort. This graph shows the distribution of the ADAS-Cog global cognitive screen measure for the RVCI cohort. The mean and the median scores respectively are 13.3 &13.          97   Figure 6-8 MoCA scores distribution for the RVCI cohort. This graph shows the distribution of the MoCA for the RVCI cohort. The mean and median scores respectively are 20.17&21.5 out of 30.    98   Figure 6-9 Log transformation for the Trails B-A This graph shows the distribution of this measure of executive function Trails B-A. The mean and median score respectively are 90.14 & 53.81 seconds.      99    Figure 6-10 Log transformation for Trails part A. This graph shows the distribution of the Log transformation of the Trails part A. The mean and median scores respectively are 43.33 & 36.87 seconds.       100    Figure 6-11 Average TUG This graph shows the distribution of the “TUG” TUG test for the RVCI cohort. The mean and median scores respectively are 8.86 & 8.35 seconds.     101   Figure 6-12 Distribution of gait speed. This graph shows the distribution of gait speed of the RVCI cohort. The mean and median scores respectively are 1.15 & 1.15 meter/seconds.   102   Figure 6-13 Log transformation of delta Stroop. This graph shows the distribution of the delta strop test a measure of executive function. The mean and median scores respectively are 53.48 & 43.83 seconds.        103   Figure 6-14 Distribution of animal fluency test. This graph shows the distribution for the animal fluency test. The mean and median scores respectively are 14.97 &15 animals named in 1 minute.                    104                                 Figure 6-15 Distribution of systolic blood pressure This graph shows the distribution of systolic blood pressure for the RVCI cohort. The mean and median scores respectively are 134.8 & 135.5-millimeter mercury mmHg.    105   Figure 6-16 Distribution of diastolic blood pressure This graph shows the distribution of diastolic blood pressure for the RVCI cohort. The mean and median scores respectively are78 & 77-millimeter mercury mmHg.                  106                                           Figure 6-17 Association of Trails A with total, periventricular and deep WMLs volume.         log Total WMLs log Periventricular WMLs log Deep WMLs   107    log Total WMLs log Periventricular WMLs log Deep WMLs  Figure 6-18 Association of delta Stroop with total, periventricular and deep WMLs volume.               108   log Total WMLs log Periventricular WMLs log Deep WMLs  Figure 6-19 Association of ADAS-cog with total, periventricular and deep WMLs volume.                    109   log Total WMLs log Periventricular WMLs log Deep WMLs  Figure 6-20 Association of MoCA scores with total, periventricular and deep WMLs volume.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0369732/manifest

Comment

Related Items