Arjen Lindenholz

144 CHAPTER 6 Table 3. Associations between the number of parenchymal changes and the intracranial vessel wall lesion burden. Outcome as count Unadjusted relative risk (±95%CI) P-value Adjusted relative risk (±95%CI) P-value Infarcts Any anterior circulation infarct 1.15 (1.03-1.29) 0.016 1.08 (0.96-1.21) 0.189 Cortical infarcts 1.12 (0.98-1.28) 0.102 1.08 (0.94-1.24) 0.290 Infarcts often caused by SVD 1.12 (0.97-1.31) 0.134 1.05 (0.90-1.23) 0.531 Small subcortical infarcts NA NA NA NA Lacunes of presumed vascular origin 1.09 (0.90-1.31) 0.370 1.00 (0.79-1.26) 0.981 Deep grey matter infarcts 1.21 (0.91-1.61) 0.190 1.15 (0.84-1.58) 0.396 Cortical microinfarcts 1.23 (0.95-1.60) 0.111 1.16 (0.86-1.56) 0.333 White matter hyperintensities (Fazekas grade) Periventricular (0,1,2 or 3) NA NA NA NA Deep (0,1,2 or 3) NA NA NA NA Table 3 . Included for analysis, n = 82. The unadjusted and adjusted (for age and sex) relative risks including their 95% confidence interval (CI) for the number of cerebral parenchymal changes with the total number of anterior vessel wall lesions as included variable. A p-value < 0.05 was considered to indicate a statistically significant difference. For continuous count data as outcome variable a log-binomial regression model was used with a robust variance estimator. For ordinal data as outcome variable an ordinal regression model was used. Ordinal Fazekas grade is described as: 0 = absence or single punctate white matter hyperintensity, 1 = ‘caps’ or pencil-thin lining or multiple punctate lesions, 2 = smooth ‘halo’ or beginning confluency of lesions, 3 = large confluent lesions or irregular hyperintensities extending into the deep white matter. 28 CMI; cortical microinfarct, NA; not applicable (groups too small for reliable statistical analyses), SVD; small vessel disease.