Cerebral white matter hypoperfusion increases with small vessel disease burden. Data from the IST-3 trial.

Francesco Arba1,2,3 MD, PhD; Grant Mair2,3 MBChB; Trevor Carpenter2 PhD; Eleni Sakka2,3 BSc; Peter AG Sandercock3 MD; Richard I Lindley4 MD; Domenico Inzitari1 MD; Joanna M Wardlaw2 ,3 MD; on behalf of the IST-3 Trial Collaborators.

1- NEUROFARBA Department, University of Florence, Florence, Italy

2- Division of Neuroimaging Sciences, Brain Research Imaging Centre, University of Edinburgh, UK ; Brain Research Imaging Centre, SINAPSE Collaboration, UK

3- Centre for Clinical Brain Sciences, University of Edinburgh, Western General Hospital, Edinburgh, United Kingdom

4- Westmead Hospital Clinical School and the George Institute for Global Health, University of Sydney, Sydney, New South Wales, Australia

Running title: white matter perfusion in acute ischaemic stroke

Key words: cerebral small vessel disease, CT perfusion, stroke, white matter, cerebral blood

flow

Tables: 3

Figures: 2

Word count: 4465

Corresponding author: Wardlaw JM, Division of Neuroimaging Sciences, University of

Edinburgh, Western General Hospital, Crewe Rd, Edinburgh EH4 2XU, United Kingdom

Fax: +441313325150

Phone: +441315372943

Email:joanna.wardlaw@ed.ac.uk

1

ABSTRACT

Introduction: Leukoaraiosis is associated with impaired cerebral perfusion but the effect of

individual and combined small vessel disease (SVD) features on white matter perfusion is

unclear.

Methods: We studied patients recruited with perfusion imaging in the IST-3 trial. We rated

individual SVD features (leukoaraiosis, lacunes) and brain atrophy on baseline plain CT or

MR imaging. Separately, we assessed white matter at the level of the lateral ventricles in the

cerebral hemisphere contralateral to the stroke for visible areas of hypoperfusion

(present/absent) on four time-based perfusion imaging parameters. We examined associations

between SVD features (individually and summed) and presence of hypoperfusion using

logistic regression adjusted for age, sex, baseline NIHSS, hypertension, and diabetes.

Results: 115 patients, median (IQR) age 81 (72-86) years, 78 (52%) males had complete

perfusion data. Hypoperfusion was most frequent on mean transit time (MTT; 63 patients,

55%), and least frequent on Tmax (19 patients, 17%). The SVD score showed stronger

independent associations with hypoperfusion (e.g. MTT, OR=2.80; 95% CI=1.56-5.03), than

individual SVD markers (e.g. white matter hypoattenuation score, MTT, OR 1.49,

95%CI=1.09, 2.04). Baseline BP did not differ by presence/absence of hypoperfusion or

across strata of SVD score. Presence of white matter hypoperfusion increased with SVD

summed score.

Conclusions: The SVD summed score was associated with hypoperfusion more consistently

than individual SVD features, providing validity to the SVD score concept. Increasing SVD

burden indicates worse perfusion in the white matter.

2

Introduction

Small vessel disease (SVD) is a clinical and radiological term that refers to pathological

processes that affect microvessels in the brain1. SVD is the most common vascular cause of

dementia and causes about one fifth of all strokes worldwide1. Magnetic resonance (MR) is

the recommended imaging technique for research purposes, but computed tomography (CT)

is widely used in clinical practice, especially in acute stroke, and may also detect important

markers of SVD such as leukoaraiosis, lacunes and brain atrophy2.

SVD is associated with chronic cerebral hypoperfusion3,45. Severe white matter changes have

been associated with reduction of cerebral blood flow in longitudinal studies6, and an

impairment in cerebral blood flow has been found in patients with brain atrophy in the

presence of leukoaraiosis7. However, it is not known whether other SVD features or

increasing severity of any or all SVD features worsen white matter perfusion or not.

Both MR perfusion8 and CT perfusion imaging9 have been used for assessment of cerebral

perfusion in acute ischaemic stroke. However, clinical studies so far evaluated eligible

patients for reperfusion through identification of acute ischaemic but salvageable tissue, and

SVD has not been investigated much as a concurrent factor that may influence cerebral

perfusion in the acute stroke phase. Pre-existing SVD is a common finding in stroke, and the

estimate of imaging features consistent with SVD may convey important information on the

brain’s response to acute stroke10. Moreover, the hypothesis that global rather than single

features of SVD may better represent the burden of pathology in the brain has been

proposed11.

We investigated the relationship between single and summed radiological features of SVD

detected with plain CT or MR, and cerebral perfusion detected with CT or MR perfusion

imaging, using data from a large randomised trial of patients with acute ischaemic stroke.

3

Methods

We analysed data from patients recruited to the Third International Stroke Trial (IST-3) who

underwent CT or MR perfusion imaging in the acute phase. The full trial details have been

published12. Briefly, IST-3 was a randomised, open-label trial of intravenous recombinant-

tissue Plasminogen Activator (rt-PA) versus control given within 6 hours of onset in patients

with symptoms and signs of clinically definite acute stroke in whom brain CT or MR had

excluded intracranial haemorrhage or non-stroke lesions13.The patient or a valid proxy gave

informed consent. Severity of stroke at baseline was graded with the National Institutes of

Health Stroke Scale (NIHSS), the presenting stroke subtype according to the Oxfordshire

Community Stroke Project Classification (OCSP)14.

Functional outcome was assessed at six-months from stroke onset by the Oxford Handicap

Scale (OHS; comparable to the modified Rankin Score15). We defined disability/death as

OHS 2-6, functional dependency/death as OHS 3-6, and death as OHS=6 at six months.

Symptomatic intracranial haemorrhage was defined as CT or MR evidence of significant

intracranial bleeding and one of the following: a) neurological deterioration; b) new

headache; c) new acute hypertension; d) new nausea or vomiting; e) sudden decrease in

conscious level13.

In the present study, we analysed data from patients recruited in IST-3 with perfusion

imaging. A trained stroke neurologist (FA), blinded to all clinical and previously assessed

imaging data, rated the available CT or MR scans for SVD features according to STRIVE

(STandards for ReportIng Vascular changes on nEuroimaging) recommendations2 and

separately analysed the perfusion images. The perfusion readings were blinded to other

radiological features and clinical data except for the side of symptoms. Two experienced

neuroradiologists (GM, JW) cross-checked the ratings. Angiographic information about the

4

affected (i.e. ischaemic) hemisphere was also available, but not with regards to the

contralateral ischaemic side.

Similar to the expert panel review13, we rated white matter changes with the Van Swieten

Scale (VSS), then combined the posterior (range 0-2) and anterior (range 0-2) scores into a

five-point ordinal scale (0-4)16. We recorded the presence and number of lacunes, defined as

round or ovoid shaped small CSF-attenuation areas <1.5cm in diameter in subcortical white

and deep grey matter2. We defined brain atrophy as deep or cortical, and rated with a three-

point ordinal scale as none, moderate or severe against a reference CT brain template12.

We created an aggregate SVD score by summing the white matter changes, lacunes, and

brain atrophy scores. Although previous versions of the SVD score included also

microbleeds11, we only considered radiological features of SVD detectable with CT. We

assigned 1 point for each of the following if present: severe lucencies (VSS=2) in anterior or

posterior periventricular white matter, lacunes>=2, and severe (=2) central or cortical

atrophy. The combined four-point ordinal score therefore assessed the global burden of SVD

from 0 (no imaging features of severe SVD) to 3 (imaging features of SVD scored as severe

for each imaging variable).

The perfusion data were processed centrally using in-house software12,17,18 corrected for bolus

arrival time to produce a range of perfusion parameter maps, as described in detail

previously19. In the present work, we assessed four time-based perfusion parameters without

thresholds because, after blinded evaluation of the cerebral blood flow (CBF), cerebral blood

volume (CBV) and time-based parameters, white matter hypoperfusion was evident in the

contralateral hemisphere primarily with time-based parameters20. We qualitatively examined

four different parameters on standard colour perfusion maps as described in detail

previously12,17,18: mean transit time (MTT; first moment of the concentration time curve), time

to maximum flow (Tmax; time to maximum up slope of contrast bolus curve), time to peak

5

(TTP; time of maximum contrast concentration minus arrival time fitted), arrival time fitted

(ATF; delay from bolus passage through internal carotid artery to arrival of contrast in each

voxel). Firstly, on each perfusion map separately, we examined the hemisphere contralateral

to the acute ischaemic lesion to identify if there were areas in the frontal and occipital white

matter where the perfusion appeared reduced compared with the rest of the white matter in

that hemisphere (Figure 1). Then we compared this white matter with perfusion in the

hemisphere ipsilateral to the acute ischaemic stroke (i.e. acute ischaemic) hemisphere. If the

frontal/occipital periventricular white matter in the contralateral hemisphere was of similar

colour to the acute perfusion deficit in the ipsilateral hemisphere, we scored that subject’s

contralateral white matter as ‘hypoperfused’. If the contralateral periventricular white matter

appeared uniform throughout that hemisphere and not reduced as in the perfusion deficit on

the ipsilateral hemisphere, we scored the white matter as ‘normally perfused’. We used

reference examples for each perfusion parameter for patients who did not have a clearly

visible perfusion deficit in the perfusion map of the cerebral hemispheres (e.g. in patients

with lacunar or posterior circulation territory infarcts in the brainstem or cerebellum).

We tested whether the general characteristics of the study population with perfusion data

differed from the whole IST-3 trial. In patients with perfusion data, we compared the

perfusion characteristics against the baseline characteristics (including blood pressure) by

different grades of SVD using ANOVA. We assessed univariate associations between each

SVD feature, SVD sum score and hypoperfusion in the contralateral hemisphere for each

perfusion parameter. We retained univariate associations with a p-value <0.1 and ran separate

logistic regression models for each radiological feature of SVD as explanatory variables

adjusting for age, sex, baseline NIHSS, hypertension and diabetes. We further examined

associations between presence of hypoperfusion on each perfusion parameter and clinical

outcomes using logistic regression models. A p-value <0.05 was considered statistically

6

significant. Statistical analysis was carried out using SPSS for Windows (version 22.0; SPSS,

Armonk NY, IBM Corp.).

Results

In IST-3, 150 patients had perfusion imaging performed with CT (n=123/150, 82%) or MR

(n=27/150, 18%), but 35 patients (32 with CT, 3 with MR) had non-assessable perfusion maps

due to either incomplete acquisition, severe movement artefact, or problems during data

transfer. Therefore complete perfusion maps and concurrent baseline plain CT or MR

imaging for the assessment of SVD were available in 115 (77%) patients (91 CT, 24 MR).

Demographic and clinical variables were similar to those from the IST-3 trial population as a

whole. Median (IQR) age was 81 (72-85) years, 60 patients (52%) were male (Table 1).

Baseline median NIHSS was 11 (IQR=6-19), 45 (39%) patients had total anterior circulation

syndrome (TACS) according to the OCSP. The most frequent pre-stroke risk factor was

hypertension.

For the radiological SVD features, 35 (30%) patients had severe central or cortical brain

atrophy, 30 (26%) had severe white matter changes, and 22 patients (19%) had two or more

pre-existing lacunes. There were no differences in detection of SVD features between CT and

MR (severe white matter changes=28% vs 23%, p=0.815; two or more lacunes=22% vs 10%,

p=0.272; severe atrophy=28% vs 29%, p=0.959, respectively). A total of 69 (46%) patients

had no severe SVD signs (SVD score=0) on baseline CT of MR scan, and 81 (55%) had at

least one radiological feature rated as severe.

Evidence of acute hypoperfusion in the affected hemisphere was present in 87 (76%) patients,

with 51 (45%) patients having ischaemia in the left hemisphere, 31 (27%) patients in the right

hemisphere, 2 bilateral signs of acute ischaemia and 3 on the midline. There was considerable

variability in the presence of hypoperfusion in the contralateral hemisphere between

7

perfusion parameters. More than half the patients (55%) had hypoperfusion in the

contralateral hemisphere according to MTT, whereas only 19 patients (17%) had impaired

perfusion according to Tmax (Table 2).

The presence of hypoperfusion in the contralateral hemisphere was weakly associated with

age (significant only for TTP, p=0.016, Supplemental material, figure 1), but not with stroke

severity or delay from stroke onset to CT or MR perfusion imaging (Supplemental material,

figures 2 and 3).

Baseline blood pressure did not differ with increasing SVD score (e.g. systolic median blood

pressure: 156,157, 155 and 150 mmHg for SVD score 0, 1, 2, 3 respectively, p=0.875,

Supplemental material, figure 4), nor with the presence versus absence of hypoperfusion in

the subcortical white matter on any of the four time-based parameters (Supplemental

material, figure 5).

On univariate analysis, lacunes were the only individual radiological feature of SVD not

associated with hypoperfusion category on any of the perfusion parameters, whereas

leukoaraiosis, brain atrophy and the SVD summed score were each associated with

hypoperfusion (Supplemental material, table 1). Across all perfusion parameters, the

frequency of hypoperfusion in the contralateral hemisphere increased as SVD sum score

increased (Figure 2).

In the multivariable model adjusted for age, NIHSS, sex, hypertension and diabetes, the SVD

score was associated with hypoperfusion on most perfusion parameters, the strongest

association being with ATF (OR=3.59; 95% CI=1.92-6.75; p<0.001). Individual SVD

components were associated with hypoperfusion on some but not all perfusion parameters,

slightly more with ATF than the other parameters (Table 3).

8

The presence, versus absence, of hypoperfusion in the contralateral hemisphere white matter

on any of the four parameters was not associated with any early or late clinical outcomes

(Supplemental material, table 2).

Discussion

We demonstrated that, in patients with acute ischaemic stroke, the presence of radiological

markers of SVD on baseline brain imaging, mostly with CT scanning, were associated with

impaired – i.e. prolonged – time-based perfusion in the white matter of the contralateral

hemisphere. Furthermore, we showed that increasing severity of total SVD score was the

strongest predictor of hypoperfusion in the contralateral hemisphere. There was no relation

between hypoperfusion and either NIHSS or time after stroke, hypertension or admission BP.

This suggests that the summed score encompassing all SVD features detectable with plain CT

scans provides a better indication of the impact of background changes of SVD on the brain

than individual features alone, and indicates that CT assessment of SVD burden may be

useful where MR is not available. Our findings support hypoperfusion as a key finding in

association with SVD markers.

We confirm and expand to acute stroke the previous studies that showed reduced cerebral

perfusion in patients with features of SVD. In patients without evidence of stroke, Markus et

al., using quantitative MR, showed impaired cerebral blood flow in white matter in 8 patients

with leukoaraiosis3 compared to 9 controls; similar findings were reported in a case-control

study with 37 patients4, supporting the association between impaired cerebral circulation and

SVD. A recent systematic review of CBF in patients with WMH without stroke showed that

CBF fell with increasing WMH in cross sectional studies with older age and dementia both

worsening perfusion compared with similar-aged controls21. Two studies using transcranial

Doppler (TCD) showed a) reduced MCA velocities in the acute phase bilaterally in patients

9

with lacunar infarcts22 but unilateral impairment in patients with middle cerebral artery

occlusion and b) that MCA flow velocity decreased by 3.75 cms-1 per point increase in white

matter hyperintensity score (p=0.004) when adjusted for age23. In a CT perfusion study with

35 patients with transient ischaemic attack or rapidly resolving stroke symptoms, cerebral

blood flow decreased with increasing severity of white matter changes24.

We found that baseline pre-randomisation blood pressure did not differ with SVD severities,

or with the presence or absence of hypoperfusion, suggesting that the hypoperfusion in white

matter was not directly related to concurrent high or low systemic blood pressure. Instead,

increasing SVD score was clearly related to worsening perfusion.

The association between increasing severity of the SVD score and time-based perfusion

parameters suggests that perfusion slowing, or increase in transit time, is a key feature of the

affected microcirculation. There may be several explanations for this. SVD affects both

arterioles and venules, with variation in diameter and course of the micro-vessels (i.e.

narrowing of arterioles and dilation of micro-venules), which might slow blood transit25. SVD

is also associated with rheological modifications such as increased platelet activation26 and a

pro-coagulative status27 that could further slow transit. Finally, SVD microvascular

alterations are associated with blood brain barrier dysfunction28 and raised plasma markers of

endothelial dysfunction29,30,31, consistent with neurovascular unit dysfunction and

dysregulation of microcirculation in response to metabolic demands32.

The combined SVD features were more strongly associated with the perfusion parameters

than individual SVD features, providing support for the SVD burden score concept. Previous

scoring for the global burden of SVD was made using MR imaging and either non-acute

stroke patients or older community-dwelling subjects with SVD11, 33. Quantification of SVD

burden appears to be feasible using plain CT scans in patients with acute ischaemic stroke,

making it readily applicable in clinical practice in the acute stroke setting. However, we

10

acknowledge that the score requires validation in different and larger datasets against

different outcomes and additional statistical modelling to test the weighting of the different

SVD features before being implemented in research or clinical practice. The clinical

relevance is high, given that individual SVD features (leukoaraiosis, brain atrophy, old

infarcts including lacunes) detected on baseline CT scan were independent predictors of poor

outcome after stroke34; the combined SVD score might increase power to predict early and

late outcomes and plan acute therapy.

The study had limitations. Even with 115 patients, the sample size was small and the study

lacked power to determine if increasing hypoperfusion, and worsening SVD score, influenced

clinical outcomes. However, the direction of effect was consistent with the presence of white

matter hypoperfusion in the contralateral hemisphere worsening outcomes.

We used a practical qualitative (i.e. visual) method to assess perfusion scans. We

acknowledge that quantitative assessment might better estimate the entire spectrum of

perfusion values. However, the qualitative assessment was quick and immediately applicable

in clinical practice since it requires less complex image post-processing. We confirmed

hypoperfusion in the contralateral hemisphere by reference to the acute ischaemic defect,

which has reduced perfusion by definition. We did not have information about angiographic

findings on the contralateral hemisphere, however future studies could address associations

between the hemisphere contralateral to the ischaemic side and angiographic appearance.

In conclusion, we showed that single SVD features such as leukoaraiosis and central brain

atrophy are associated with hypoperfusion (i.e. prolonged transit time). The association is

stronger when SVD markers are considered as a combined score, and with worsening of the

SVD score. A summed SVD score of radiological markers on baseline CT scan could act as

an imaging surrogate of chronic white matter hypoperfusion with further tests. Further studies

11

to validate the SVD score in other datasets and to estimate the impact of global burden of

SVD on stroke outcomes are warranted.

12

Acknowledgements: The complete list of funding sources and centres in IST-3 that recruited

to the perfusion imaging study are listed in Appendix 1.

Conflict of interest: F. Arba, G. Mair, T. Carpenter, E. Sakka, D. Inzitari report no conflict

of interest. P. Sandercock has received funding from Boehringer Ingelheim. R. Lindley has

received funding from Boehringer Ingelheim and Covidien. J. Wardlaw has received funding

from Medical Research Council, Efficacy and Mechanisms Evaluation, Chest Heart Stroke

Scotland. The views and opinions expressed therein are those of the authors and do not

necessarily reflect those of the funding agencies or UK Department of Health.

Source of funding: See Appendix 1.

13

References:

1- Wardlaw JM, Smith C, Dichgans M. Mechanisms of sporadic cerebral small vessel

disease: insights from neuroimaging. Lancet Neurol. 2013;12:483-97.

2- Wardlaw JM, Smith EE, Biessels GJ, et al.; STandards for ReportIng Vascular

changes on nEuroimaging (STRIVE v1). Neuroimaging standards for research into

small vessel disease and its contribution to ageing and neurodegeneration. Lancet

Neurol. 2013;12:822-38.

3- Markus HS, Lythgoe DJ, Ostegaard L, O'Sullivan M, Williams SC. Reduced cerebral

blood flow in white matter in ischaemic leukoaraiosis demonstrated using quantitative

exogenous contrast based perfusion MRI. J Neurol Neurosurg Psychiatry.

2000;69:48-53.

4- O'Sullivan M, Lythgoe DJ, Pereira AC, Summers PE, Jarosz JM, Williams SC, et al.

Patterns of cerebral blood flow reduction in patients with ischaemic leukoaraiosis.

Neurology. 2002;59:321-6.

5- Vernooij MW, van der Lugt A, Ikram MA, et al. Total cerebral blood flow and total

brain perfusion in the general population: the Rotterdam Scan Study. J Cereb Blood

Flow Metab. 2008;28:412-9.

6- Van der Veen PH, Muller M, Vincken KL, et al.; SMART Study Group. Longitudinal

relationship between cerebral small-vessel disease and cerebral blood flow: the

second manifestations of arterial disease-magnetic resonance study. Stroke.

2015;46:1233-8.

7- Appelman AP, van der Graaf Y, Vincken KL, et al. Total cerebral blood flow, white

matter lesions and brain atrophy: the SMART-MR study. J Cereb Blood Flow Metab.

2008;28:633-9.

14

8- Davis SM, Donnan GA, Parsons MW, et al.; EPITHET investigators. Effects of

alteplase beyond 3 h after stroke in the Echoplanar Imaging Thrombolytic Evaluation

Trial (EPITHET): a placebo-controlled randomised trial. Lancet Neurol. 2008;7:299-

309.

9- Bivard A, Levi C, Krishnamurthy V, et al. Perfusion computed tomography to assist

decision making for stroke thrombolysis. Brain. 2015;138:1919-31.

10- IST-3 collaborative group. Association between brain imaging signs, early and late

outcomes, and response to intravenous alteplase after acute ischaemic stroke in the

third International Stroke Trial (IST-3): secondary analysis of a randomised controlled

trial. Lancet Neurol. 2015;14:485-96.

11- Staals J, Makin SD, Doubal FN, Dennis MS, Wardlaw JM. Stroke subtype, vascular

risk factors, and total MRI brain small-vessel disease burden. Neurology.

2014;83:1228-34.

12- Sandercock P, Lindley R, Wardlaw J, et al.; IST-3 Collaborative Group. Third

International stroke trial (IST-3) of thrombolysis for acute ischaemic stroke. Trials.

2008;17;9-37. 

13- IST-3 collaborative group. The benefits and harms of intravenous thrombolysis with

recombinant tissue plasminogen activator within 6 h of acute ischaemic stroke (the

third international stroke trial [IST-3]): a randomised controlled trial. Lancet. 2012;

379:2352-63.

14- Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural

history of clinically identifiable subtypes of cerebral infarction. Lancet.

1991;337:1521-6.

15- Bamford JM, Sandercock P, Warlow C, Slattery J. Interobserver Agreement for the

Assessment of Handicap in Stroke Patients.Stroke. 1989;20:828-828.

15

16- Van Swieten JC, Hijdra A, Koudstaal PJ, van Gijn J. Grading white matter lesions on

CT and MRI: a simple scale. J Neurol Neurosurg Psychiatry. 1990;53:1080-3.

17- Wardlaw JM, von Kummer R, Carpenter T, et al. Protocol for the perfusion and

angiography imaging sub-study of the Third International Stroke Trial (IST-3) of

alteplase treatment within six-hours of acute ischaemic stroke. Int J Stroke.

2015;10:956-68.

18- Wardlaw JM, Carpenter T, Sakka E, et al. Imaging perfusion deficits, arterial patency

and thrombolysis safety and efficacy in acute ischaemic stroke. An observational

study of the effect of advanced imaging methods in The Third International Stroke

Trial (IST-3), a randomised controlled trial. Efficacy and Mechanism

Evaluation. 2014;1.

19- Kane I, Carpenter T, Chappell F, et al. Comparison of 10 different magnetic

resonance perfusion imaging processing methods in acute ischaemic stroke. Effect on

lesion size, proportion of patients with diffusion/perfusion mismatch, clinical scores

and radiological outcomes. Stroke. 2007;38;3158-3164.

20- Carpenter T, Armitage PA, Bastin ME, Wardlaw JM. DSC perfusion MRI –

quantification and reduction of systematic errors arising in areas of reduced cerebral

blood flow. Magn Reson Med. 2006;56:1342-1349.

21- Shi Y, Thrippleton MJ, Makin SD, et al. Cerebral blood flow in small vessel disease:

a systematic review and meta-analysis. J Cereb Blood Flow Metab, in press.

22- Immink RV, van Montfrans GA, Stam J, Karemaker JM, Diamant M, van Lieshout JJ.

Dynamic cerebral autoregulation in acute lacunar and middle cerebral artery territory

ischaemic stroke. Stroke. 2005;36:2595-600.

16

23- Wardlaw JM, Doubal FN, Eadie E, Chappell F, Shuler K, Cvoro V. Little association

between intracranial arterial stenosis and lacunar stroke. Cerebrovasc Dis.

2011;31:12-18.

24- Huynh TJ, Murphy B, Pettersen JA, et al. CT perfusion quantification of small-vessel

ischaemic severity. Am J Neuroradiol. 2008;29:1831-6.

25- Black S, Gao F, Bilbao J. Understanding white matter disease: imaging-pathological

correlations in vascular cognitive impairment. Stroke. 2009;40(3 Suppl):S48-52.

26- Iwamoto T, Kubo H, Takasaki M. Platelet activation in the cerebral circulation in

different subtypes of ischaemic stroke and Binswanger's disease. Stroke. 1995;26:52-

6.

27- Tomimoto H, Akiguchi I, Wakita H, Osaki A, Hayashi M, Yamamoto Y. Coagulation

activation in patients with Binswanger disease. Arch Neurol. 1999;56:1104-8.

28- Wardlaw JM, Doubal F, Armitage P, et al. Lacunar stroke is associated with diffuse

blood-brain barrier dysfunction. Ann Neurol. 2009;65:194-202.

29- Poggesi A, Pasi M, Pescini F, Pantoni L, Inzitari D. Circulating biologic markers of

endothelial dysfunction in cerebral small vessel disease: a review. J Cereb Blood

Flow Metab. 2015.doi:10.1038/jcbfm.2015.116.

30- Stevenson SF, Doubal FN; Shuler K, Wardlaw JM. A systematic review of dynamic

cerebral and peripheral endothelial function in lacunar stroke versus controls. Stroke.

2010;41:e434-e442.

31- Wiseman S, Marlborough F, Doubal F, Webb DJ, Wardlaw J. Blood markers of

coagulation, fibrinolysis, endothelial dysfunction and inflammation in lacunar stroke

versus non-lacunarstroke and non-stroke: systematic review and meta-analysis.

Cerebrovasc Dis. 2014;37:64-75.

17

32- Ostergaard L, Sondergaard T, Moreton F, et al. Cerebral small vessel disease:

capillary pathways to stroke and cognitive decline. J Cereb Blood Flow Metab.

2015;36:302-25.

33- Klarenbeek P, van Oostenbrugge RJ, Rouhl RP, Knottnerus IL, Staals J. Ambulatory

blood pressure in patients with lacunar stroke: association with total MRI burden of

cerebral small vessel disease. Stroke. 2013;44:2995-9.

34- IST-3 collaborative group. Association between brain imaging signs, early and late

outcomes, and response to intravenous alteplase after acute ischaemic stroke in the

third International Stroke Trial (IST-3): secondary analysis of a randomised controlled

trial. Lancet Neurol. 2015; 14:485-96.

18

Table 1. Baseline characteristics of study population. *Patients with complete perfusion data.

Variables All patients with perfusion

N=150

Patients in analyses*

N=115

Age, median (IQR) 81 (72-86) 81 (72-85)

Sex, male 78 (52) 60 (52)

Weight, mean (±SD) 73 (±15.7) 73 (14.9)

Systolic BP, mean (±SD) 155 (±23.4) 155 (±23.1)

Diastolic BP, mean (±SD) 81 (±18.3) 82 (±18.5)

Baseline Glucose, mean (±SD) 7.3 (±2.3) 7.2 (±2.3)

TACS by OCSP 60 (40) 45 (39)

Atrial fibrillation 52 (21) 38 (33)

Hypertension 105 (70) 81 (70)

Diabetes 20 (13) 15 (13)

NIHSS, median (IQR) 11 (7-18) 11 (6-19)

Severe leukoaraiosis 60 (39) 30 (26)

Two or more lacunes 26 (10) 22 (19)

Severe brain atrophy 55 (37) 35 (30)

Data are n (%) unless otherwise stated. IQR=Interquartile Range; SD=Standard deviation; BP=Blood Pressure; TACS=Total Anterior Circulation Syndrome; OCSP=Oxfordshire Community Stroke Project; NIHSS=National Institutes of Health Stoke Scale; VSS=Van Swieten Scale; SVD=Small Vessel Disease.

N (%) for SVD score: grade 0=69 pts (46%), grade 1=55 pts (36%), grade 2=21 pts (15%), grade 3=5 pts (3%).

19

Table 2. Frequency of impaired perfusion in contralateral periventricular cerebral white matter by perfusion parameter and SVD features.

Variable Total

N=115 (%)

WMC VSS>2

N=35 (%)

Lacunes≥2

N=21 (%)

Brain atrophy>2

N=69 (%)

MTT 66 (58) 21 (78) 15 (71) 47 (68)

Tmax 19 (17) 8 (30) 6 (29) 16 (23)

TTP 39 (34) 15 (56) 10 (48) 30 (44)

ATF 56 (50) 21 (81) 15 (71) 40 (59)

WMC=White Matter Changes; VSS=Van Swieten Scale for WMC; MTT=Mean-Transit-Time; TTP=Time-to-Peak; ATF=Arrival-Time-Fitted.

20

OR (95% CI)

MTT Tmax TTP ATF

Lacunes - - - -

WMC, anterior VSS 1.54 (0.86-2.76) 1.62 (0.76-3.44) 1.89 (1.01-3.51) 2.61(1.39-4.88)

WMC, posterior VSS 2.39 (1.35-4.23) 1.59 (0.87-2.92) 2.07 (1.24-3.48) 2.28 (1.32-3.93)

WMC, total VSS 1.49 (1.09-2.04) 1.32 (0.92-1.91) 1.51 (1.11-2.07) 1.70 (1.23-2.35)

Central atrophy 2.29 (1.22-4.30) 3.72 (1.44-9.59) 1.49 (0.77-2.85) 2.19 (1.17-4.11)

Cortical atrophy 1.79 (0.86-3.73) - 1.73 (0.78-3.80) 1.36 (0.66-2.82)

SVD score 2.80 (1.56-5.03) 2.36 (1.37-4.09) 1.52 (0.93-1.05) 3.59 (1.92-6.75)*

Table 3. Multivariate associations between SVD markers and hypoperfusion in the contralateral hemisphere white matter by four time-based perfusion parameters.

OR=Odds ratio; CI=Confidence Interval; WMC=White Matter Changes; VSS=Van Swieten Scale; MTT=Mean Transit Time; TTP=Time to Peak; ATF=Arrival Time Fitted; SVD=Small Vessel Disease.*p<0.001Each row is the result of a separate logistic regression model adjusted for: age, sex, NIHSS, hypertension, diabetes.

21