2939
Quantitative Multi-Parametric Mapping reveals specific white-matter tissue changes in Sickle Cell Disease1Developmental Nerosciences, UCL Great Ormond St Institute of Child Health, London, United Kingdom, 2Clinical Nerosciences, UCL Great Ormond St Institute of Child Health, London, United Kingdom, 3Great Ormond Street Hospital, London, United Kingdom
Synopsis
Diffusion studies have revealed loss of white-matter integrity in sickle cell disease (SCD), but the underlying pathophysiology is poorly understood. We combined diffusion tensor imaging with multi-parametric mapping in 23 patients and 23 controls to examine more specific imaging markers of myelin, iron, and water content. Voxel-based region of interest analyses revealed a pattern of decreased MT, R1, FA, and AD, and increased RD and MD in patients, consistent with lower myelination. Indices of lower myelination were associated with oxygen desaturation and processing speed. Our results offer insight into pathology of SCD brain changes and potential biomarkers for future trials.
Introduction
SCD is a genetic disorder which adversely affects red blood cell rheology and oxygen-carrying capacity. Diffusion tensor imaging (DTI) studies have revealed abnormal integrity in normal appearing WM1–3 and associated cognitive impairment,4,5 but the underlying pathophysiology is poorly understood. Increases in corpus callosum (CC) radial diffusivity (RD) have been associated with oxygen desaturation (SpO2),1 which may reflect demyelination or axonal loss secondary to compromised oxygen delivery and extraction.6,7
Although increases in RD are sometimes interpreted as consistent with demyelination, DTI parameters are not specific to tissue properties such as myelin and iron content. Quantitative multi-parameter mapping (MPM)8 can be used to produce maps sensitive to specific tissue elements, including the longitudinal relaxation rate (R1), sensitive to myelin, macromolecular content and water; transverse relaxation rate (R2*), sensitive to ferromagnetic substances; magnetisation transfer (MT), sensitive to myelin and water fraction; and proton density (PD), sensitive to free water. These contrasts have offered insight into WM pathology in several populations,9,10 but have not been examined in SCD.
Aiming to discover biomarkers for tissue property changes in SCD, we analysed MPM and DTI parameters in patients and controls. We performed voxel-based quantification (VBQ) analyses in WM across the whole-brain and, based on previous literature,1,5 across a ROI in the CC. We hypothesised that parameters linked to myelin would be lower in patients than controls, and explored correlations with disease severity and cognitive outcome.
Methods
MRI, cognitive assessment using the Wechsler scales, and pulse oximetry were conducted on the same day. Closest full blood count was extracted from patient medical records.
MRI Acquisition
MRI was conducted on a 3T Siemens Prisma with 80 mT/m gradients, 64 channel receiver coil, and body coil transmit. We used an established MPM method based on three 3D multi-echo FLASH datasets with predominant PD-, T1- and MT-weighting acquired with 1mm3 resolution.11 Maps of the transmit field B1+ were used to correct for transmit inhomogeneities. We also included established fluid-attenuated inversion recovery (FLAIR; voxel size=0.65x1x0.65mm), and diffusion-weighted (b=1000s/mm2 & b=2200s/mm2, voxel size=2x2x2mm) sequences. A neuroradiologist evaluated FLAIR images, and ROIs were drawn around silent cerebral infarctions (SCI).
MRI Processing
As previously described,12,13 MPM maps were estimated using the SPM hMRI14 toolbox. Following correction for susceptibility-induced distortions and eddy-currents, DTI maps were estimated via a weighted least squares method using tractoR15 and FSL.16 DTI maps were affine aligned to R1 maps. Automated tissue classification of MT maps was performed using the CAT12 toolbox, followed by diffeomorphic registration to MNI space using a group specific template. All maps were normalised to MNI using a combined probability weighting and Gaussian smoothing procedure (3mm3 FWHM) to maintain quantitative values.12,13
Statistical Analysis
For VBQ, a full-factorial model was implemented in SPM12, including one factor (group) and two covariates (age, gender). For ROI analyses in the CC body, a mask was derived from the JHU atlas.17 Results were family-wise error (FWE) corrected and thresholded at p<0.05. To explore correlations with disease severity and cognition, MPM metrics that survived FWE correction were extracted and averaged across clusters from the unthresholded image (p<0.001).
Results
Whole-brain VBQ revealed lower FA and higher MD in patients across small cerebellar clusters (pfwe-corr<0.05, Table 2). ROI VBQ revealed lower MT, R1, AD, and FA in patients, along with higher MD and RD ( pfwe-corr<0.05, Table 2, Figure 1) across the CC body. Differences remained when SCI were removed. In patients, mean MT and R1 across CC clusters were positively associated with SpO2, and processing speed (p<0.05, Figure 2). There were no further correlations with anaemia severity or cognitive outcome.Discussion
Our findings confirm previous reports of changes in normal-appearing WM in SCD. The results are in line with studies reporting associations between RD in the CC and SpO2,1 and between FA, MD, and RD, and processing speed.4 Here we demonstrate a pattern of decreased MT and R1 across the CC which is correlated with decreased SpO2 and processing speed. The observed pattern is consistent with lower myelination.
It is possible that hypoxia leads to demyelination or a failure to myelinate in SCD, and that this in turn leads to slower processing. Whilst global cerebral blood flow (CBF) is increased, oxygen demand may exceed delivery in the WM, where regions of CBF nadir overlap with regions of highest SCI density,7 and abnormal oxygen extraction.18 Oxygen metabolism may be further compromised in patients with low SpO2. There is evidence that myelin-producing oligo-dendrocytes, required for fast neural transmission, are particularly vulnerable to hypoxia.19
Whilst validation in a larger sample is required, our findings offer novel insights into SCD WM pathology.
Acknowledgements
No acknowledgement found.References
1. Kawadler JM, Kirkham FJ, Clayden JD, et al. White Matter Damage Relates to Oxygen Saturation in Children With Sickle Cell Anemia Without Silent Cerebral Infarcts. Stroke. 2015;46(7):1793-1799. doi:10.1161/strokeaha.115.008721
2. Balci A, Karazincir S, Beyoglu Y, et al. Quantitative brain diffusion-tensor MRI findings in patients with sickle cell disease. AJR Am J Roentgenol. 2012;198(5):1167-1174. doi:10.2214/AJR.11.7404
3. Sun B, Brown RC, Hayes L, et al. White matter damage in asymptomatic patients with sickle cell anemia: screening with diffusion tensor imaging. AJNR Am J Neuroradiol. 2012;33(11):2043-2049. doi:10.3174/ajnr.A3135
4. Stotesbury H, Kirkham FJ, Kölbel M, et al. White matter integrity and processing speed in sickle cell anemia. Neurology. 2018. doi:10.1212/WNL.0000000000005644
5. Schatz J, Buzan R. Decreased corpus callosum size in sickle cell disease: Relationship with cerebral infarcts and cognitive functioning. J Int Neuropsychol Soc. 2006. doi:10.1017/S1355617706060085
6. Bush AM, Coates TD, Wood JC. Diminished cerebral oxygen extraction and metabolic rate in sickle cell disease using T2 relaxation under spin tagging MRI. Magn Reson Med. December 2017. doi:10.1002/mrm.27015
7. Ford AL, Ragan DK, Fellah S, et al. Silent infarcts in sickle cell anemia occur in the borderzone region and are associated with low cerebral blood flow. Blood. 2018. doi:10.1182/blood-2018-04-841247
8. Weiskopf N, Suckling J, Williams G, et al. Quantitative multi-parameter mapping of R1, PD(*), MT, and R2(*) at 3T: a multi-center validation. Front Neurosci. 2013;7:95. doi:10.3389/fnins.2013.00095
9. Callaghan MF, Freund P, Draganski B, et al. Widespread age-related differences in the human brain microstructure revealed by quantitative magnetic resonance imaging q. 2014. Neurobiol Aging. doi:10.1016/j.neurobiolaging.2014.02.008
10. Kullmann S, Callaghan MF, Heni M, et al. Specific white matter tissue microstructure changes associated with obesity. Neuroimage. 2016;125:36-44. doi:10.1016/J.NEUROIMAGE.2015.10.006
11. Lorio S, Kherif F, Ruef A, et al. Neurobiological origin of spurious brain morphological changes: A quantitative MRI study. Hum Brain Mapp. 2016;37(5):1801-1815. doi:10.1002/hbm.23137
12. Draganski B, Ashburner J, Hutton C, et al. Regional specificity of MRI contrast parameter changes in normal ageing revealed by voxel-based quantification (VBQ). Neuroimage. 2011;55(4):1423-1434. doi:10.1016/J.NEUROIMAGE.2011.01.052
13. Callaghan MF, Freund P, Draganski B, et al. Widespread age-related differences in the human brain microstructure revealed by quantitative magnetic resonance imaging. Neurobiol Aging. 2014;35:1862-1872. doi:10.1016/j.neurobiolaging.2014.02.008
14. Phillips C, Balteau E, Leutritz T, et al. The hMRI toolbox for quantitative imaging and in vivo histology using MRI (hMRI). OHBM Abstr Proc. 2018. http://www.hmri.info. Accessed November 7, 2018. 15. Clayden JD, Muñoz Maniega S, Storkey AJ, King MD, Bastin ME, Clark CA. TractoR: Magnetic Resonance Imaging and Tractography with R. J Stat Softw. 2011;44(8):1-18.
16. Jenkinson M, Beckmann CF, Behrens TEJ, Woolrich MW, Smith SM. FSL. Neuroimage. 2012;62(2):782-790. doi:10.1016/j.neuroimage.2011.09.015
17. Mori S, Wakana S, Van Zijl PCM, Nagae-Poetscher LM. MRI Atlas of Human White Matter. Elsevier; 2005.
18. Fields ME, Guilliams KP, Ragan DK, et al. Regional oxygen extraction predicts border zone vulnerability to stroke in sickle cell disease. Neurology. 2018. doi:10.1212/WNL.0000000000005194
19. Lyons SA, Kettenmann H. Oligodendrocytes and Microglia Are Selectively Vulnerable to Combined Hypoxia and Hypoglycemia Injury in Vitro. J Cereb Blood Flow Metab. 1998;18(5):521-530. doi:10.1097/00004647-199805000-00007
Figures
