MR Imaging in Carotid Artery Atherosclerosis Plaque Characterization

Abstract

Aim: To evaluate the potential role of carotid artery atherosclerosis plaque magentic resonance (MR) microimaging as magnetic resonance imaging (MRI) marker, ex vivo MR images were acquired at optimized parameters on 9.4T Bruker animal imager for occluded tissue resected by carotid endarterectomy (CEA) and corresponding histopathological analysis was made.Methods and Materials: For imaging, CEA tissues of size 2-6 cm long and 0.5-1.5 cm wide, were transferred to 15 ml co-polymer laboratory culture tubes containing either 10z formalin in phosphate buŠered saline (PBS) or in 50z glycerol in PBS. Imaging protocol was set at TE＝30 ms, TR＝1.5 s, matrix size＝265×512, NEX＝128, slice thickness＝1 mm and in-plane resolution＝0.1 mm for total sample size 2.5 cm. Soon after imaging done, carotid artery tissues were cut into 5-mm segments and processed for histological section for successive 5-micrometer slices. To compare morphology of 5 mm thin CEA section with that of 1 mm MR slices, registration was obtained between histologic sections and MR slices. Contrast and magnetic resonance relaxation characteristics were analyzed.Results: Total carotid artery area computed by MR imaging was correlated with areas determined from histologic sections (r 2 ＝0.989, p＝0.0001). For the lumen area, the correlation between MR images and histologic area was (r 2 ＝0.942, p＝0.0001). Relaxation times and T 2 parametric images of diŠerent plaque components were determinant for contrast resolution. Scan parameters were optimized forˆbrous cap and atheroma. Scan parameters were characteristic for comparison at 1.5T and 9.4T MR imagers.Conclusion: The observed correlation validated MR microimaging to assess morphological features of carotid artery plaques and contrast resolution highlighted the potential of in vivo MR imaging as non-invasive MRI marker to monitor carotid artery plaque morphometry and plaque composition.

Ex vivo study of carotid endarterectomy specimens: quantitative relaxation times within atherosclerotic plaque tissues

“…Our study also finds no difference in T 1 between plaque components, this observation concurs with previous studies [19,24,25].…”

Microimaging of hairless rat skin by magnetic resonance at 900 MHz

“…The histology-MRI correlation was done by using a nonlinear regression curve fit method using sensitivityspecificity criteria, reproducibility test to avoid subjective bias and Student t test at two-tailed Pb.0001 as level of significant match [27]. The sensitivity, specificity and prediction values were calculated by true-or-false, negative-or-positive (TP, TN, FP, FN) statistics of comparing skin hair and epithelium structures as described earlier [8].…”

“…The denoising and surface-fitting segmentation were done by the 'level set' method to create a new volume from the input data to obtain a user-defined feature extraction as described elsewhere [26,27].…”

“…The MRI signal intensity and % MT activity profiles were assigned to anatomically identified skin layers (stratum corneum remnants, epidermis, papillary dermis and reticular dermis). For measurement accuracy, delineated areas on T2-weighted MR images were compared with delineated areas on coregistered histology digital images by the ImagePro software [27].…”

“…In brief, 4-μm-thick skin samples were dehydrated in a sequence of 70%, 90%, 100% ethanol and 20 s of hematoxylin staining followed by brief treatment of eosin and extra stain removal by xyline as described elsewhere [27]. Distinct skin layers of the epidermis, dermis, hair follicle, hair and sebaceous gland were assessed for skin viability and grade evaluation [27].…”