Circulating inflammatory markers are associated with MR visible Perivascular Spaces
but not directly with White Matter Hyperintensities
Benjamin S. Aribisala, PhD1,2,3,4, Stewart Wiseman, BSc1, Zoe Morris, FRCR1, Maria C.
Valdés-Hernández, PhD1,2,3, Natalie A. Royle, BSc1,2,3, Susana M. Maniega, PhD1,2,3,
Alan J. Gow, PhD2,5, Janie Corley, BSc 6, Mark E. Bastin, PhD1,2,3, John Starr, MD2,7
Ian J. Deary, PhD2,6, Joanna M. Wardlaw, MD1,2,3
1
Brain Research Imaging Centre, University of Edinburgh, UK
2
Centre for Cognitive Ageing and Cognitive Epidemiology, University of Edinburgh, UK
3
Scottish Imaging Network, A Platform for Scientific Excellence (SINAPSE)
4
Department of Computer Science, Lagos State University, Nigeria
5
Department of Psychology, Heriot-Watt University, UK
6
Department of Psychology, University of Edinburgh, UK
7
Geriatric Medicine Unit, University of Edinburg, UK
Author for Correspondence:
Prof. JM Wardlaw
Brain Research Imaging Centre
University of Edinburgh
Western General Hospital
Crewe Road, Edinburgh
EH4 2XU, U.K.
Tel:+441315373093,Email:joanna.wardlaw@ed.ac.uk
1
Abstract
Background and Purpose: White matter hyperintensities (WMH) and perivascular spaces
(PVS) are features of small vessel disease (SVD), found jointly on magnetic resonance
imaging (MRI) of older people. Inflammation is a prominent pathological feature of SVD.
We examined the association between inflammation, PVS and WMH in the Lothian Birth
Cohort 1936 (N=634).
Methods: We measured plasma fibrinogen, C-reactive protein (CRP) and interleukin-6 (IL-6)
and rated PVS in three brain regions. We measured WMH volumetrically and visually using
the Fazekas scale. We derived latent variables for ‘PVS’, ‘WMH’ and ‘Inflammation’ from
measured PVS, WMH and inflammation markers, and modelled associations using structural
equation modelling.
Results: After accounting for age, sex, stroke and vascular risk factors, ‘PVS’ were
significantly associated with ‘WMH’ (β=0.47, p<0.0001); ‘Inflammation’ was weakly but
significantly associated with ‘PVS’ (β=0.12, p=0.048); but not with ‘WMH’ (β=0.02, p=NS).
Conclusions: Circulating inflammatory markers are weakly associated with MR-visible PVS
but not directly with WMH. Longitudinal studies should examine whether visible PVS
predate WMH progression and whether inflammation-modulators can prevent SVD.
2
Introduction
The pathogenesis of small vessel disease (SVD) is poorly understood. It is thought to result
from arteriolosclerosis in the penetrating arterioles leading to ischemia with diffuse
rarefaction, necrosis and cavitation in the subcortical tissues seen as white matter
hyperintensities (WMH) and lacunes on MR imaging[1]. Although perivascular inflammation
is a prominent well-established pathological feature in WMH[2] and lacunar stroke[3], the
nature of the inflammation and its role in the pathogenesis of SVD is uncertain.
Perivascular spaces (PVS), another marker of SVD, are thought to be associated with
elevated plasma inflammatory markers in older subjects[4] and in patients with small
subcortical stroke[5]. The association between plasma markers of inflammation and WMH is
less clear[6]. We tested if inflammatory markers had a direct and potentially causal
relationship with WMH, or if any relationship was via an association with PVS.
Methods
Subjects
Participants are members of the Lothian Birth Cohort 1936 (LBC1936;[7]), all born in 1936.
At mean age 73 years, inflammatory markers (N=866) were measured and brain MRI was
performed (N=700). Participants provided demographic information, medical history and
informed consent.
3
Measurement of Inflammatory Markers
CRP and IL-6 were analysed using high sensitivity ELISA (R&D Systems, Oxford, UK).
Fibrinogen was measured using an automated Clauss assay (TOPS coagulometer, Instumentation Laboratory, Warrington, UK). See Supplementary Materials for details.
Brain Magnetic Resonance Imaging
Brain MRI data were acquired on a 1.5T GE Signa Horizon HDx scanner (GE, Milwaukee,
WI, USA) and included[8]: T1-W, T2-W, T2*-W and Fluid Attenuated Inversion Recovery
(FLAIR) brain imaging.
WMH were segmented using MCMxxxVI[9] and visually rated by a neuroradiologist on
FLAIR images using the Fazekas scale[10]. Another neuroradiologist cross-checked 20%.
The intra-class correlation coefficient was 0.96. Intracranial volume (ICV) was extracted
using Analyze 9.0TM.
PVS were rated in the hippocampus, basal ganglia (BG) and centrium semiovale (CS). 20%
were cross checked by another neuroradiologist. Intra- and inter-rater kappa statistics ranged
from 0.68-0.90. See supplementary materials for details.
4
Statistical analyses
Bivariate associations were assessed between markers of inflammation, WMH, PVS and
covariates using Pearson correlation implemented in SPSS 19.0. Covariates were sex, age at
scanning, and history of cardiovascular disease, diabetes, hypertension, smoking,
hypercholesterolemia and stroke.
Multivariate associations were investigated using structural equation modelling (SEM,[11])
implemented in Amos 18.0.0 (Amos Development Corporation, Florida, USA; see
Supplementary materials for full details). This allowed derivation of latent variables for
WMH, PVS and Inflammation (‘WMH’, ‘PVS’ and ‘Inflammation’ respectively), and also the
assessment of relationships between the latent variables while including covariates.
Results
Of the 700 subjects who underwent brain MRI, 66 had incomplete data reducing the final
sample to 634 (Supplementary Table I). Almost half of the participants had a history of
hypertension, smoking or hypercholesterolemia while 11% had diabetes and 6.9% had a
history of stroke.
In the bivariate correlation analysis (Supplementary Table II), all measures of WMH (i.e.
%WMH volume in ICV, Fazekas periventricular and Fazekas deep WMH scores) were
significantly associated with PVS (in the hippocampus, BG and CS, r range 0.11 to 0.32,
P<0.001). No significant association was found between WMH measures and inflammation
markers. CS-PVS were significantly associated with CRP (r=0.10, P=0.010) but not
5
fibrinogen or IL-6. All p-values were Bonferroni corrected to account for multiple
comparisons.
Multivariate analysis using SEMs showed that ‘PVS’ was significantly associated with
‘WMH’ (β=0.47, P<0.0001), accounting for about 22% variation in ‘WMH’ (Figure 1).
‘Inflammation’ was weakly but significantly associated with ‘PVS’ (β=0.13, P=0.048),
explaining 1.6% of the variation in ‘PVS’, Figure 2. There was no significant association
between ‘inflammation’ and ‘WMH’ (Figure 3).
Discussion
We demonstrate a strong association between increased numbers of visible PVS and
increased amounts of WMH in older adults aged 71-74 years. We demonstrate that variation
in PVS accounted for about 22% of the variation in WMH. However, despite previous reports
of raised plasma inflammatory markers in subjects with SVD[5,12], we found only a weak
association between inflammatory markers and PVS and no association between
inflammatory markers and WMH. The narrow-age cohort may have allowed us to unmask
several relationships that are actually co-associations rather than direct associations as
suggested in wider age-range cohorts.
The weak association between ‘Inflammation’ and ‘PVS’ is consistent with the hypothesis
that inflammation influences SVD through effects on the small perforating arterioles, which
in turn precipitate WMH, but substantially more work is required to determine the direction
and strength of the association.
6
The lack of association between inflammatory markers and WMH contrasts with some
previous studies which found associations between CRP (n=6518[4,12]), or IL-6
(n=3644[12]), and WMH. However it agrees with other studies (n=1699[5,13-15]), a total of
2333 subjects not showing an association between CRP and WMH including the present
study. Fewer studies have addressed associations between IL-6 or fibrinogen and WMH or
PVS. The studies of inflammation and WMH are dominated by the Cardiovascular Health
Study[12] which contributes two thirds of the data on CRP and all of the data on IL-6 and
WMH. The study[12] used MR images acquired from 1992-1994 which may have been less
sensitive to WMH than current scanners; the wider age cohort may have suggested an
inflammation-WMH association that was in part a residual age co-association despite
correction for age.
The strengths of the present work include the use of both volumetric and visual WMH scores,
the use of SEMs for a robust multivariate analysis, the cohort with a narrow age range and
little ethnic diversity to reduce confounding. Weaknesses include the relative health of the
participants (about half were hypertensive), and the limited number of inflammatory variables
examined (but the three used are well understood and key to pathological processes).
Future studies should consider examining inflammatory markers in younger subjects with
SVD, as they may show a stronger differential inflammatory plasma profile. Studies should
also examine for direct evidence of inflammation in the brain.
7
Acknowledgements:
The image acquisition and analysis was performed at the Brain Research Imaging Centre,
University of Edinburgh (http://www.bric.ed.ac.uk). The work was undertaken as part of the
Cross Council and University of Edinburgh Centre for Cognitive Ageing and Cognitive
Epidemiology (http://www.ccace.ed.ac.uk).
Funding:
Research Into Ageing, Age UK, MRC, the Scottish Funding Council SINAPSE Collaboration
(http://www.sinapse.ac.uk) and the Row Fogo Charitable Trust.
Disclosures
None
8
References
1.
Wardlaw J, Smith C, Dichgans M. Mechanisms of sporadic cerebral small vessel
disease: insights from neuroimaging. Lancet Neurol. 2013;12:483-497.
2.
Gouw AA, Seewann A, van der Flier WM, Barkhof F, Rozemuller AM, Scheltens P, et
al. Heterogeneity of small vessel disease: a systematic review of MRI and
histopathology correlations. J Neurol Neurosurg Psychiatry. 2011;82:126-135.
3.
Bailey EL, Smith C, Sudlow CLM, Wardlaw JM. Pathology of lacunar ischemic stroke
in humans- a systematic review. Brain Pathol. 2012;22:583-591.
4.
Satizabal CL, Zhu YC, Mazoyer B, Dufouil C, Tzourio C. Circulating IL-6 and CRP are
associated with MRI findings in the elderly: The 3C-Dijon Study. Neurology.
2012;78:720-727.
5.
Rouhl RP, Damoiseaux JG, Lodder J, Theunissen RO, Knottnerus IL, Staals J, et al.
Vascular inflammation in cerebral small vessel disease. Neurobiol Aging.
2012;33:1800-1806.
6.
Knottnerus ILH, Ten CH, Lodder J, Kessels F, van Oostenbrugge RJ. Endothelial
dysfunction in lacunar stroke: a systematic review. Cerebrovasc Dis. 2009;27:519-526.
7.
Deary IJ, Gow AJ, Taylor MD, Corley J, Brett C, Wilson V, et al: The Lothian Birth
Cohort 1936: a study to examine influences on cognitive ageing from age 11 to age 70
and beyond. BMC Geriatr. 2007;7:28.
8.
Wardlaw JM, Bastin ME, Valdes-Hernandez MC, Maniega SM, Royle NA, Morris Z, et
al. Brain aging, cognition in youth and old age and vascular disease in the Lothian Birth
Cohort 1936: rationale, design and methodology of the imaging protocol. Int J Stroke.
2011:6:547-559.
9
9.
Hernandez MdelC, Ferguson KJ, Chappell FM, Wardlaw JM. New multispectral MRI
data fusion technique for white matter lesion segmentation: method and comparison
with thresholding in FLAIR images. Eur Radiol. 2010;20:1684-1691.
10.
Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities
at 1.5 T in Alzheimer’s dementia and normal aging MR. AJR Am J Roentgenol.
1987;149:351-356.
11.
Hooper D, Coughlan J, Mullen M. Structural equation modelling: guidelines for
determining model fit. Electronic Journal of Business Research Methods. 2008;6:53-60.
12.
Fornage M, Chiang YA, O'Meara ES, Psaty BM, Reiner AP, Siscovick DS, et al.
Biomarkers of inflammation and MRI-defined small vessel disease of the brain - The
Cardiovascular Health Study. Stroke. 2008;39:1952-1959.
13.
Schmidt R, Schmidt H, Pichler M, Enzinger C, Petrovic K, Niederkorn K, et al: Creactive protein, carotid atherosclerosis, and cerebral small-vessel disease: results of the
Austrian Stroke Prevention Study. Stroke. 2006;37:2910-2916.
14.
Wright CB, Moon Y, Paik MC, Brown TR, Rabbani L, Yoshita M, et al: Inflammatory
biomarkers of vascular risk as correlates of leukoariosis. Stroke. 2009;40:3466-3471.
15.
Wersching H, Duning T, Lohmann H, Mohammadi S, Stehling C, Fobker M. Serum Creactive protein is linked to cerebral microstructural integrity and cognitive function.
Neurology. 2010;74:1022-1029.
10
Figure 1: SEM of the association between measures of PVS and WMH. Model fit: Root
Mean Square Error Approximation (RMSEA)=0.044, Tucker-Lewis Index (TLI)=0.952,
Comparative Fit Index (CFI)=0.968 and Incremental Fit index (IFI)=0.968. A model is said
to give a good fit if RMSEA < 0.06, TLI > 0.90, CFI >0.90 and IFI >0.90.
Note. ‘PVS’=latent variable derived from measures of PVS, ‘WMH’=latent variable derived
from measures of WMH. Models included covariates that were significantly associated with
measures of WMH or PVS. BG=basal ganglia PVS, HIP=hippocampal PVS, CS=centrum
semiovale PVS, high BP=hypertension, WMH v=%WMH in ICV, Faz=Fazekas,
Peri=Perivascular.
Shape representation: Rectangular boxes=measured observations, ellipses=latent variables,
single-headed arrows=hypothesised causal pathways, and double headed arrows=correlations,
dashed arrows=non-significant associations and the solid arrows=significant associations at
p<0.05, and numbers adjacent to arrows are beta-weights. The numbers on the arrows
pointing from the latent to the measured variables represent the strength of the association
between the measured and latent variables, +/- 1 indicate the strongest association. Thick
arrows pointing to the measured variables mean that that error term was accounted for in the
modelling and the numbers adjacent represent the r2.
11
Figure
2:
SEM
of
the
association
between
markers
of
Inflammation,
and
PVS.RMSEA=0.003, TLI=1, CFI=1 and IFI=1.
Abbreviations: See Figure 1.
Figure 3: SEM of the association between inflammation and WMH load. RMSEA=0.024,
TLI=0.987, CFI=0.991 and IFI=0.991.
Abbreviations: See Figure 1.
12