Page 69

Karger_ESC London_2013

London, United Kingdom 2013 14 Brain imaging B 17:10 - 17:20 Relationships between plasma inflammatory markers, perivascular spaces and white mat-ter Cerebrovasc Dis 2013; 35 (suppl 3)1-854 69 hyperintensities J.M. Wardlaw1, B.S. Aribisala2, Z Morris3, M.C. Valdes Hernadez4, N.A. Royle5, S.M. Maniega6, A. Gow7, J.M. Starr8, M.E. Bastin9, I.J. Deary10 University of Edinburgh, Edinburgh, UNITED KINGDOM1,University of Edinburgh, Ed-inburgh, UNITED KINGDOM2, University of Edinburgh, Edinburgh, UNITED KINGDOM3, University of Edinburgh, Edinburgh, UNITED KINGDOM4, University of Edinburgh, Edin-burgh, UNITED KINGDOM5, University of Edinburgh, Edinburgh, UNITED KINGDOM6, University of Edinburgh, Edinburgh, UNITED KINGDOM7, University of Edinburgh, Edin-burgh, UNITED KINGDOM8, University of Edinburgh, Edinburgh, UNITED KINGDOM9, University of Edinburgh, Edinburgh, UNITED KINGDOM10, University of Edinburgh, Edin-burgh, UNITED Background: Perivascular spaces (PVS) seen on MRI, are associated with white matter hyper-intensities (WMH). PVS may also be a marker of inflammation and microvascular inflamma-tion is prominent pathologically in small vessel disease (SVD). We examined the association between plasma markers of inflammation, PVS and WMH in a large older cohort. Methods: We studied subjects aged 71-73 from the Lothian Birth Cohort 1936. All had structural MRI; plasma samples for fibrinogen, C-reactive protein and interleukin-6 (IL-6); history of vascular disease and risk factors. We rated PVS in the basal ganglia, hippocampus and centrum semiova-le with a validated scale and measured WMH by volume and Fazekas scale blind to other fac-tors. We derived latent variables for PVS, inflammatory markers and WMH from the respective measured values, and modelled multivariate associations using structural equation modelling correcting for age, gender, smoking, diabetes, hypertension, hypercholesterolaemia and stroke. Results: We included 634 subjects (mean age 72.5±0.71 years). The latent variables all repre-sented the measured PVS, WMH and inflammatory values well (factor loadings: g-inflamma-tion range 0.61 to 0.67, mean 0.63; g-WMH range 0.70 to 0.94, mean 0.80;g-EPVS range 0.41 to 0.67, mean 0.56). After accounting for age, sex, stroke and vascular risk factors, g-inflamma-tion was significantly associated with g-PVS (β = 0.12, p=0.047) and g-PVS was significantly associated with g-WMH (β = 0.47, p<0.0001), but g-inflammation was not associated with g-WMH, with or without accounting for PVS. Conclusions: MR-visible PVS are associated with raised levels of circulating inflammatory markers and WMH. Inflammation may affect ce-rebral perforating arterioles, manifest as MR-visible PVS, which in turn influence WMH, but inflammation is not directly associated with WMH. Long term studies are required to see if vis-ible PVS predate WMH progression and whether inflammation-modulators can reduce SVD. 13 Brain imaging B 17:00 - 17:10 Enlarge Perivascular Spaces in in Lacunar Stroke Patients; Results of The SPS3 trial H. Obha1, T.S. Field2, L.A. Pearce3, C. Bazan4, G. Potter5, R.G. Hart6, O.R. Benavente7 The SPS3 Investigators Iawate University, Morioka, JAPAN1,University of British Columbia, Vancouver, CANA-DA2, Minot, Minot, USA3, University of Texas, San Antonio, USA4, University of Manchester, Manchester, UNITED KINGDOM5, University of McMaster, Hamilton, CANADA6, University of British Columbia, Vancouver, CANADA7 Background Enlarged perivascular spaces (EPVS) are common on MRIs patients with cerebral small vessel disease (CVSD).Their pathophysiology and clinical significance is uncertain. We analysed as-sociations between EPVS in the basal ganglia (BG) and centrum semiovale (CS) with baseline characteristics and outcomes in the SPS3 trial (NCT00059306). Methods All participants were included in the neuroimaging cohort. All SPS3 patients had an MRI-con-firmed lacunar infarct within 6 months of enrolment. T2 sequence images were rated for the number of EPVS in the basal ganglia (BG) and centrum semiovale (CS) on a 5 point scale. Results Of 3020 participants,average age was 63 y, 63% of patients were male and 2781 had T2 se-quences available. Numbers of EPVS in each participant were symmetric in theBG and CS. In the BG, 50% had >10 EPVS on each side; 74%in the CS had >10. After multivariable analysis, presence of >10 EPVS in the BG was associated with increasing age, diabetes, hypertension and diastolic blood pressure (BP) (p<0.05).In the CS, >10 EPVS was associated with age, hy-pertension, systolic BP, hyperlipidemia, and race-ethnicity (p<0.05). There was a significant association between>10 EPVS and white matter hypertensities rated by the ARWMC score (OR 1.6; 1.5-1.7 for BG, OR 1.2, 1.1-1.3 for CS, p<0.001) and presence of multiple subcortical in-farcts on MRI (OR 1.5, 1.3-1.8 for BG, OR 1.2, 1.2-1.7; p<0.001). Association of EPVS with stroke recurrence, mortality and cognitive outcome will be presented. Conclusions: In this large, well-defined cohort of lacunar stroke patients, increased numbers of EPVS seem to be an independent neuroradiological marker for CVSD and associated vascular risk factors. Ongoing analysis of SPS3 will likely elucidate if EPVS are independently predictors of stroke recurrence and cognitive function.


Karger_ESC London_2013
To see the actual publication please follow the link above