CHAPTER 1

Introduction

12 | CHAPTER 1

Purpose and motivationTo improve patient care, and to target the socio-economical burden of demen-

tia that is to be expected due to an increase in mean age of the population,

several important goals in present clinical neuroimaging studies can be identi-

fied. These goals become even more relevant with the prospect of possible

future development of disease-modifying therapies. The first of these goals is

to develop and evaluate the applicability of clinical tools that enable an early

diagnosis. Secondly, prediction of disease progression and improvement of

prognosis are necessary to optimize patient management. Finally, especially

with the prospect of future therapies that target disease-specific mechanisms,

it becomes important to predict the presence of underlying neuropathological

mechanisms that lead to dementia, to be able to select patients that might

benefit from these treatments. This thesis explores the clinical value of hip-

pocampal atrophy, measured on MRI, in the cognitive continuum of normal

aging, mild cognitive impairment (MCI) and Alzheimer’s disease (AD). Ex-

panding insights in the clinical value of this MRI marker might lead to earlier

recognition and increased specificity in the diagnosis of AD, and thus increase

the effectiveness of a patient’s management.

General introductionClassically, the role of neuroimaging in the clinical work-up of dementia has

been to exclude other causes of dementia, such as possible ‘treatable’ causes

(tumours, hemorrhage) and vascular lesions. This role has shifted towards

the recognition of findings on MRI that increase the specificity of a diagnosis

of AD.5-8 Neuroimaging allows in vivo measurement of the brain tissue loss

that occurs in AD. There are different approaches to measure this volume

loss. Although MRI has been the modality of choice because of its high resolu-

tion and possibility to obtain three-dimensional volume scans, advances in

computed tomography (CT) scanners have increased the resolution of this

modality, potentially making CT a suitable alternative for MRI. Furthermore,

volume measurements using MRI can be divided into cross-sectional (using a

13INTRODUCTION |

scan from one time point per patient) and longitudinal measures (comparing

scans from different time points of one patient), and can focus on different

brain structures that are important in AD.

Figure 1. Coronal T1-weighted images. (a) control, no atrophy (b) patient with AD, showing widening of the gyri and ventricles, with pronounced atrophy of the hippocampus on both sides (arrows)

There are different cross-sectional methods to estimate the volume of brain

structures on MRI, ranging from simple (but clinically very efficient) visual rat-

ing scales, to manual segmentation and semi-automated segmentation tech-

niques. An advantage of MRI over many other biomarkers in AD, is its ability

to monitor changes over time within a patient. Longitudinal studies use one or

more scans from different time points to calculate change in volume over time

within a patient. Methods to calculate longitudinal volume changes include the

subtraction of volumes, derived by manual segmentation of brain structures

at different time points, as well as more sophisticated, automated registration

methods.

14 | CHAPTER 1

Figure 2. Images showing the hippocampus at (a) baseline and (b) at follow-up, as well as (c) the color-overlay, derived from non-linear fluid registration, of (1) a patient with MCI that did not progress, and (2) a patient with MCI who had progressed to AD at follow-up. Colours indicate the relative local volume change between the two scans, as indicated by the color bar. The patient with MCI that progressed to AD at follow-up shows more atrophy of the hippocampus at follow-up, which is illustrated by the brigth blue areas in the color overlay.

In AD, two important neuroimaging markers are whole brain atrophy and re-

gional atrophy of the hippocampus. Hippocampal atrophy on MRI correlates with

pathologically assessed volumes and number of neurons,12,13 and with neuro-

fibrillary changes in the medial temporal lobes.14 Furthermore, hippocampal

volume loss, measured on MRI, is associated with cognitive decline15,16 and

with a clinical diagnosis of AD, and predicts progression from mild cognitive

impairment (MCI) to AD.17,18 Although studied less extensively, more or less

the same results have been found for whole brain volume and especially whole

brain volume change. Longitudinal studies showed that whole brain atrophy

rates are higher in AD patients than in controls,18-20 and predict progression

from MCI to AD.21,22 Whole brain atrophy rate also seems to correlate with

pathological neuro-fibrillary changes.23 It is not clear whether one of the two

markers, whole brain atrophy or regional hippocampal atrophy, is superior to

the other, and how these two markers relate to each other.

15INTRODUCTION |

MRI findings that often occur in older age in general and in dementia in par-

ticular, are findings related to cerebrovascular disease. These findings can

be divided into large vessel disease (infarcts) and small vessel disease. MRI

findings of the latter are white matter hyperintensities, lacunar infarcts and

microbleeds. The presence of extensive vascular neuroimaging findings is

a requirement in the clinical diagnostic criteria of VaD,24 and MRI therefore

plays an important role in distinguishing AD from VaD.7,8 Nevertheless, neu-

ropathological studies show that in the majority of subjects with dementia,

both AD-related pathology and vascular pathological changes are found, and

these studies even suggest that these two types of pathology might have a

synergistic effect on clinical disease severity.25,26 These findings are supported

by neuroimaging studies that show that vascular findings occur more often

in patients with AD than in controls,27 and that presence of medial temporal

lobe atrophy is often found in VaD.28 The clinical relevance of the presence of

vascular neuroimaging findings in AD is not clear.

The hippocampusThe hippocampus is a brain structure, located at the infero-medial temporal lobe, and is part of the limbic system. It plays a crucial role in learning and memory, and is also in-volved in systems controlling emotional behaviour and motor control.1The hippocampus has been identified to be a primary location of AD pathology, eventually leading to loss of neurons.2 This volume loss can be detected in vivo by MRI. Different methods exist to assess hippocampal atrophy. Visual rating scales have been developed that allow semi-quantitative assessment of atrophy according to predefined criteria. The scale we use in this thesis (range 0-4), is based on characteristics of the hippocampus and features of surrounding structures of the medial temporal lobe.3 Interrater agreement of this method is moderate to good.4 A second method consists of segmentation of the hippocampus on MRI. Although (semi) automated segmentation methods have been described, the gold standard still consists of manual delineation of a region of interest (ROI) around the hippocampus on coronal slices, and calculation of the volume by multiplying the total surface on these slices by slice thickness. Calculation of longitudinal change in volume within the hippocampus can be performed by comparing segmented hippocampal vol-umes. However, more sophisticated, semi-automated registration methods have been developed. After a registration algorithm matches a serial scan onto the reference scan, the registration matrix can be used to describe the difference in volume between these two scans. In this thesis, we use a regionally optimized non-linear registration method of serial scans onto corresponding baseline scans, which we quantify within manually constructed ROIs of the hippocampus on the baseline scan to measure hippocampal atrophy rate.9-11 Because the deformations applied by the registration algorithm are constrained by general laws of fluid dynamics, this registration method is called ‘fluid’ registration.

16 | CHAPTER 1

Aims and outlineThe general objective of this thesis was to extend the insights in the clinical

applicability of MRI markers in AD and its preceding clinical stages, with an

emphasis on the measurement of hippocampal atrophy. We evaluate the use of

different methods, comparing CT with MRI and comparing the measurement of

regional hippocampal atrophy with the measurement of whole brain atrophy.

Furthermore, we focus on the use of hippocampal atrophy as a marker of

disease progression, and try to identify the prognostic value of MRI markers.

Thirdly, we investigate the relevance of vascular MRI findings in dementia in

relation to markers of atrophy.

Outline per chapterThe presence of dementia is associated with increased mortality,29,30 and within

a dementia population, several clinical and demographic risk factors have been

identified that are associated with an increased risk of mortality.29,31 However,

the role of neuroimaging in predicting mortality is not clear. In chapter 2, we

investigate the predictive value of MRI variables on mortality in patients that

had visited our memory clinic. Baseline MRI findings of atrophy and vascular

damage were assessed by visual rating scales and information about whether

patients were still alive was attained.

In chapter 3, we use the screening data from a clinical trial to investigate the

significance of vascular findings in AD by comparing demographic, clinical and

neuropsychological characteristics of AD patients with and without these find-

ings. In addition, we investigate associations between vascular MRI findings

and MRI markers of atrophy with cognitive performance in AD.

Because of its higher resolution, MRI has been the neuroimaging modality of

choice in dementia, over CT. However, advances in technology have improved

the resolution of CT scanning. In chapter 4, we investigate whether high-reso-

17INTRODUCTION |

lution CT can be used in the diagnostic process of dementia, by calculating the

agreement of MRI and CT for visual rating scales of global and regional atrophy

and small vessel disease.

In chapter 5, we compare the applicability of measurements of regional hip-

pocampal volume and whole brain volume by comparing their ability to distin-

guish between subjective complaints, MCI and AD, and their ability to predict

progression to AD within patients with subjective complaints and MCI. We

compare cross-sectional and longitudinal measurement of the hippocampus

and whole brain.

In chapter 6, we use the rate of hippocampal atrophy over time, calculated

using a regional, non-linear registration algorithm,9,10 as a proxy for disease

progression in a population of patients with AD, MCI and subjective complaints.

We assessed which baseline data from neuropsycological examination, APOE

genotyping, cerebrospinal fluid biomarkers and baseline MRI parameters were

associated with subsequent hippocampal atrophy rates.

Chapter 7 comprises a pilot study in the feasibility of setting up a collaborative

database with clinical, biochemical and neuroimaging data of AD patients in

Europe. These data were gathered from patients with AD, MCI and subjective

complaints, from seven European centres. Results from basic visual rating of

the scans were compared with results from a large database in the United

States as validation.

In chapter 8, the main findings are discussed in the light of recent literature

and recommendations for future research are given.

18 | CHAPTER 1

Reference ListDuvernoy HM. The human hippocampus: functional anatomy, vascularization and serial 1.

sections with MRI. 2 ed. Berlin Heidelberg New york: Springer-Verlag, 1998.

Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta Neuro- 2.

pathol 1991;82:239-259.

Scheltens P, Leys D, Barkhof F et al. Atrophy of medial temporal lobes on MRI in “probable” 3.

Alzheimer’s disease and normal ageing: diagnostic value and neuropsychological correlates. - J

Neurol Neurosurg Psychiatry 1992 Oct;55(10):967-72 1992.

Scheltens P, Launer LJ, Barkhof F, Weinstein HC, van Gool WA. Visual assessment of 4.

medial temporal lobe atrophy on magnetic resonance imaging: interobserver reliability. J Neurol

1995;242:557-560.

Dubois B, Feldman HH, Jacova C et al. Research criteria for the diagnosis of Alzheimer’s 5.

disease: revising the NINCDS-ADRDA criteria. Lancet Neurol 2007;6:734-746.

Knopman DS, DeKosky ST, Cummings JL et al. Practice parameter: diagnosis of dementia 6.

(an evidence-based review). Report of the Quality Standards Subcommittee of the American Acad-

emy of Neurology. Neurology 2001;56:1143-1153.

Scheltens P, Fox N, Barkhof F, De Carli C. Structural magnetic resonance imaging in the 7.

practical assessment of dementia: beyond exclusion. Lancet Neurol 2002;1:13-21.

Waldemar G, Dubois B, Emre M et al. Recommendations for the diagnosis and manage- 8.

ment of Alzheimer’s disease and other disorders associated with dementia: EFNS guideline. Eur J

Neurol 2007;14:e1-26.

Crum WR, Scahill RI, Fox NC. Automated hippocampal segmentation by regional 9.

fluid registration of serial MRI: validation and application in Alzheimer’s disease. Neuroimage

2001;13:847-855.

Freeborough PA, Fox NC. Modeling brain deformations in Alzheimer disease by fluid 10.

registration of serial 3D MR images. J Comput Assist Tomogr 1998;22:838-843.

van de Pol LA, Scahill RI, Frost C et al. Improved reliability of hippocampal atrophy rate mea- 11.

surement in mild cognitive impairment using fluid registration. Neuroimage 2007;34:1036-1041.

Barkhof F, Polvikoski TM, van Straaten ECW et al. The significance of medial temporal lobe 12.

atrophy. Neurology 2007;69:1521-1527.

Bobinski M, de Leon MJ, Wegiel J et al. The histological validation of post mortem mag- 13.

netic resonance imaging-determined hippocampal volume in Alzheimer’s disease. Neuroscience

2000;95:721-725.

Whitwell JL, Josephs KA, Murray ME et al. MRI correlates of neurofibrillary tangle pathology 14.

at autopsy - A voxel-based morphometry study. Neurology 2008;71:743-749.

Petersen RC, Jack CR, Jr., Xu yC et al. Memory and MRI-based hippocampal volumes in 15.

aging and AD. Neurology 2000;54:581-587.

19INTRODUCTION |

Rusinek H, De SS, Frid D et al. Regional brain atrophy rate predicts future cognitive decline: 16.

6-year longitudinal MR imaging study of normal aging. Radiology 2003;229:691-696.

Devanand DP, Pradhaban G, Liu X et al. Hippocampal and entorhinal atrophy in mild cogni- 17.

tive impairment: prediction of Alzheimer disease. Neurology 2007;68:828-836.

Jack CR, Jr., Petersen RC, Xu y et al. Rates of hippocampal atrophy correlate with change in 18.

clinical status in aging and AD. Neurology 2000;55:484-489.

Fox NC, Cousens S, Scahill R, Harvey RJ, Rossor MN. Using serial registered brain magnetic 19.

resonance imaging to measure disease progression in Alzheimer disease: power calculations and

estimates of sample size to detect treatment effects. Arch Neurol 2000;57:339-344.

Whitwell JL, Przybelski SA, Weigand SD et al. 3D maps from multiple MRI illustrate chang- 20.

ing atrophy patterns as subjects progress from mild cognitive impairment to Alzheimer’s disease.

Brain 2007;130:1777-1786.

Jack CR, Shiung MM, Weigand SD et al. Brain atrophy rates predict subsequent clinical 21.

conversion in normal elderly and amnestic MCI. Neurology 2005;65:1227-1231.

Sluimer JD, van der Flier WM, Karas GB et al. Whole-Brain Atrophy Rate and Cognitive 22.

Decline: Longitudinal MR Study of Memory Clinic Patients. Radiology 2008.

Josephs KA, Whitwell JL, Ahmed z et al. beta-amyloid burden is not associated with rates 23.

of brain atrophy. Annals of Neurology 2008;63:204-212.

Roman GC, Tatemichi TK, Erkinjuntti T et al. Vascular dementia: diagnostic criteria for re- 24.

search studies. Report of the NINDS-AIREN International Workshop. Neurology 1993;43:250-260.

Pathological correlates of late-onset dementia in a multicentre, community-based popu- 25.

lation in England and Wales. Neuropathology Group of the Medical Research Council Cognitive

Function and Ageing Study (MRC CFAS). Lancet 2001;357:169-175.

Snowdon DA, Greiner LH, Mortimer JA et al. Brain infarction and the clinical expression of 26.

Alzheimer disease. The Nun Study. JAMA 1997;277:813-817.

Scheltens P, Barkhof F, Valk J et al. White matter lesions on magnetic resonance imag- 27.

ing in clinically diagnosed Alzheimer’s disease. Evidence for heterogeneity. Brain 1992;115 ( Pt

3):735-748.

Bastos-Leite AJ, van der Flier WM, Straaten E et al. The contribution of medial tempo- 28.

ral lobe atrophy and vascular pathology to cognitive impairment in vascular dementia. Stroke

2007;38:3182-3185.

Ganguli M, Dodge HH, Shen C, Pandav RS, DeKosky ST. Alzheimer disease and mortality: a 29.

15-year epidemiological study. Arch Neurol 2005;62:779-784.

Tschanz JT, Corcoran C, Skoog I et al. Dementia: the leading predictor of death in a defined 30.

elderly population: the Cache County Study. Neurology 2004;62:1156-1162.

Claus JJ, van Gool WA, Teunisse S et al. Predicting survival in patients with early Alzheimer’s 31.

disease. Dement Geriatr Cogn Disord 1998;9:284-293.