2071
Cingulum tractography in old subjects presenting low or high white matter lesion burden1UMR5287, Aquitaine Institute for Cognitive and Integrative Neuroscience, Bordeaux, France, 2INSERM U897, Bordeaux Population Health (BPH Center), Bordeaux, France
Synopsis
Tractography frequently fails in aging brain because diffusion parameters dramatically decrease in regions of WM hyperintensities (WMH) which one are very common in this population. We developed here a pipeline taking into account this pitfall to truly investigate the microstructural properties of the cingulum bundles in presence and absence of WMH.
Introduction
Aging is associated with widespread brain structural modifications both in gray matter (GM) and in white matter (WM) compartments. In WM, the presence of the so called WM Hyperintensities (WMH), as observed on a T2-weighted MRI has been described for a long time in older individuals. More recently, Diffusion Imaging (DI) reveals age-related microstructural changes of WM. Tractographic algorithm based on DI parameters allows to construct the main tract of the brain. However, the construction of the tracts in the aging brain failed or is at least erroneous due to the presence of the WMH. Indeed, in these WM lesions, diffusion parameters dramatically decrease conducting to a wrong start or stop of the algorithm. We developed here a specific pipeline to constraint the algorithm to cross these WM lesions. Using this pipeline adapted to the aging brain, we reinvestigated the microstructural state of well-constructed cingulum bundle in a population of aging subjects presenting a high level of WMH and one presenting a low level of WMH.Methods
We included 45 subjects from the Bordeaux site of the 3-City study, an elderly cohort with 12 years cognitive follow-up (Table1). MRI were obtained using a Philips ACHIEVA 3T scanner. For each subject, we have a diffusion weighted images with 22 directions (b = 1000s/mm² and voxel size: 2x2x2 mm³) and a 1mm isotropic T1-weighted image and a 0.72×1.20×5 mm3 FLAIR sequence. Severity of WMH was evaluated by two trained operators according to the Fazekas1 rating scale. Using this visual scale, we selected subjects with an extensive halo of WMH (grade 3). We computed WM, GM and CSF maps with SPM2 and a WMH probability mask with the LST3. We computed the fODF4 with maximal Spherical Harmonic order of 65. We then generate include, exclude and WM tracking masks. This mask was corrected by removing the voxel of WMH from the seeding mask and we add them accordingly in the include and exclude masks. Then, we performed anatomically constrained probabilistic tractography using the PFT6 with 10 seed per voxel from the WM seeding. Aberrant streamlines were removed by using a clustering algorithm that classifies streamlines from their shapes. We used ROI-based approach to extract cingulum bundles from the whole tractogram using the White Matter Query Language7. In this analysis, we included 7 ROIs from parcellation of anatomical images processing with Freesurfer’s Fan atlas. Finally, we extracted fractional anisotropy (FA) and mean diffusivity (MD) in the whole tract and in 15 different segments all along the tract so as to construct an antero-posterior tract profile8. We compared MD metrics of the whole cingulum and its profile between low WMH group and high WMH group using a t-test. Secondly, in the WMH group, we investigated the relationship between the amount of WMH burden and diffusion parameters using partial Pearson correlations. Results were considered significant when p<0.05.Results
Older subjects with low WMH present lower MD values compared to older subjects with high WMH (p<0.005), no difference for FA values was found. Considering tract profile, we found that the tract FA and MD profiles had a similar shape for the two groups. Moreover, older subjects with low WMH had higher FA than the high WMH at specific locations on the Tract (Figure 2A, B). Older subjects with low WMH had lower MD than the high WMH at specific locations on the Tract Profiles (figure 2C, D, p< 0.05). For the high WMH group, a significant correlation was found between the WMH burden and mean MD values (left: r = 0.479, p = 0.012, right: r = 0.381, p = 0.04). No significant correlations were found for the whole mean FA values. Moreover, in segments significantly different between the two group, a significant correlation was found between the WMH burden and the MD (p < 0.05) and FA values (p < 0.05).Discussion
Here, we used a lesion mask segmentation to bypass the critical step to seed and apply tractography with anatomical constraints in aging brain presenting WMH. In this study, we confirmed that the WMH burden is associated to microstructural properties of the cingulum tract. Moreover, WMH seems to have a greater impact on MD all along the cingulum than on FA metric, this one being different only in the frontal and posterior part of the bundle.Conclusion
Our results highlight the need to consider the presence of WMH to conduct tractography in elderly. In future analysis, we will investigate the relationship between the microstructure alteration of the cingulum presented here and the cognitive performances of the subjects.Acknowledgements
No acknowledgement found.References
1. F. Fazekas, J. B. Chawluk, A. Alavi, H. I. Hurtig, and R. A. Zimmerman, “MR Signal Abnormalities at 1.5 T in Alzheimer’s Dementia and Normal Aging,” AJNR Am J Neuroradiol, vol. 8, no. 3, pp. 421–426, May 1987.
2. W. D. Penny, K. J. Friston, J. T. Ashburner, S. J. Kiebel, and T. E. Nichols, Statistical Parametric Mapping: The Analysis of Functional Brain Images. Academic Press, 2011.
3. P. Schmidt et al., “An automated tool for detection of FLAIR-hyperintense whitematter lesions in Multiple Sclerosis,” NeuroImage, vol. 59, no. 4, pp. 3774–3783, Feb. 2012.
4. J.-D. Tournier, F. Calamante, and A. Connelly, “Robust determination of the fibre orientation distribution in diffusion MRI: non-negativity constrained super-resolved spherical deconvolution,” Neuroimage, vol. 35, no. 4, pp. 1459–1472, May 2007.
5. E. Garyfallidis et al., “Dipy, a library for the analysis of diffusion MRI data,” Front. Neuroinform, vol. 8, p. 8, 2014.
6. G. Girard, K. Whittingstall, R. Deriche, and M. Descoteaux, “Towards quantitative connectivity analysis: reducing tractography biases,” Neuroimage, vol. 98, pp. 266–278, Sep. 2014.
7. Wassermann, D., Makris, N., Rathi, Y., Shenton, M., Kikinis, R., Kubicki, M., Westin, C.F., 2013. On describing human white matter anatomy : the white matter query language, in : International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer. pp. 647–654.
8. Yeatman, J.D., Dougherty, R.F., Myall, N.J., Wandell, B.A., Feldman, H.M., 2012. Tract profiles of white matter properties : automating fiber-tract quantification. PloS one 7, e49790.
Figures
