A preliminary ex vivo diffusion tensor imaging study of distinct aortic morphologies

Abstract

Changes in the microstructure of the aortic wall precede the progression of various aortic pathologies, including aneurysms and dissection. Current clinical decisions with regards to surgical planning and/or radiological intervention are guided by geometric features, such as aortic diameter, since clinical imaging lacks tissue microstructural information. The aim of this proof-of-concept work is to investigate a non-invasive imaging method, diffusion tensor imaging (DTI), in ex vivo aortic tissue to gain insights into the microstructure. This study examines healthy, aneurysm and a type B chronic dissection aortae, via DTI. DTI-derived metrics, such as the fractional anisotropy, mean diffusivity, helical angle and tractography, were examined in each morphology. The results from this work highlighted distinct differences in fractional anisotropy (healthy, 0.24 ± 0.008; aneurysmal, 0.19 ± 0.002; dissected, 0.13 ± 0.006) and a larger variation in the helical angle in the dissected aorta compared to healthy (39.28 ± 11.93° vs. 26.12 ± 4.60°, respectively). These differences were validated by histological characterisation. This study demonstrates the sensitivity of DTI to pathological changes in aortic tissue, highlighting the potential of this methodology to provide improved clinical insight.

Keywords: aneurysm; aortic disease; diffusion tensor imaging; dissection; magnetic resonance imaging.

FIGURE 1

Example Masson's Trichrome stained cross‐sections of the (a) healthy and (b, c) thoracic aortic aneurysmal segments. Scale bars are 200 μm.

FIGURE 2

Example Masson's Trichrome stained cross‐section of the complex diseased descending thoracic aortic segments. T1‐weighted MRI image highlights where the histological images are representing. The true lumen and false lumen locations are highlighted in grey text; however the false lumen is not visualised here. Scale bars are 500 μm.

FIGURE 3

H&E, Verhoeff's elastin and picrosirius red (brightfield and polarised light microscopy) images of TAA sample 1.

FIGURE 4

H&E, Verhoeff's elastin and picrosirius red (brightfield and polarised light microscopy) images of TAA sample 2.

FIGURE 5

DTI metrics across three separate aortic morphologies. Parametric maps of (a) fractional anisotropy (FA) and (b) mean diffusivity (MD). (c–e) Violin plots of (c) FA, (b) MD, and (c) helical angle (HA). Dashed lines within the violin are the median and each data point is the mean metric per MR slice; n = 5 slices for healthy, n = 18 slices for TAA1 and TAA2, and n = 31 slices for TBAD. (f) Distinct regions within a chronic dissected aorta were looked at – the media (M, blue), dissected media (DM, yellow), septum (S, green) and luminal thrombus (LT, in red). (f, i–iii) Violin plots of the DTI‐derived metrics for each slice within each region. The dashed line in each plot is the average metric for healthy aorta.

FIGURE 6

Tractography of aortic samples. Three views for the (a) healthy, (b, c) thoracic aortic aneurysmal and (d) type B dissected aortic samples. Red‐green tractography corresponds to circumferential orientation while blue represents more axial. White arrows in point to areas of less densely packed tracts in the aneurysmal samples (b, c) and to the septum (middle row) and false lumen (bottom row) in the TBAD sample (d).