Intact human brain hemispheres from 19 patients with definite CAA20 (link) and 5 non-CAA control cases were included from an ongoing post-mortem MRI study at MGH. Details on the inclusion process, scanning procedures and tissue block sampling have been described previously.8 (link),30 (link) In short, the hemispheres were fixed in 10% formalin for several weeks after autopsy. Prior to ex vivo 3T MRI, the hemispheres were placed in a plastic bag filled with periodate-lysine-paraformaldehyde and vacuum sealed. The samples were scanned overnight on a 3T MR system (Siemens, Magnetom trioTim syngo) using a 32-channel head coil. The protocol included a T2-weighted turbo-spin echo sequence (TR, 1800 ms; TE, 61 ms; voxel size, 500 μm3 isotropic) and a gradient-echo fast low-angle shot (FLASH) sequence (TR, 20 ms; voxel size, 500 μm3 isotropic). Presumed ischaemic and haemorrhagic CC lesions were identified by the same raters (W.M.F. and S.J.v.V.) who were blinded to CAA severity or other histopathological findings. Inter-rater reliability for the presence of CC lesions was substantial [κ (95% CI) = 0.75 (0.48–1)].
Free full text: Click here