Risk factors analysis according to regional distribution of white matter hyperintensities in a stroke cohort

Deep learning-based grading of white matter hyperintensities enables identification of potential markers in multi-sequence MRI data

BACKGROUND

White matter hyperintensities (WMHs) are widely-seen in the aging population, which are associated with cerebrovascular risk factors and age-related cognitive decline. At present, structural atrophy and functional alterations coexisted with WMHs lacks comprehensive investigation. This study developed a WMHs risk prediction model to evaluate WHMs according to Fazekas scales, and to locate potential regions with high risks across the entire brain.

METHODS

We developed a WMHs risk prediction model, which consisted of the following steps: T2 fluid attenuated inversion recovery (T2-FLAIR) image of each participant was firstly segmented into 1000 tiles with the size of 32 × 32 × 1, features from the tiles were extracted using the ResNet18-based feature extractor, and then a 1D convolutional neural network (CNN) was used to score all tiles based on the extracted features. Finally, a multi-layer perceptron (MLP) was constructed to predict the Fazekas scales based on the tile scores. The proposed model was trained using T2-FLAIR images, we selected tiles with abnormal scores in the test set after prediction, and evaluated their corresponding gray matter (GM) volume, white matter (WM) volume, fractional anisotropy (FA), mean diffusivity (MD), and cerebral blood flow (CBF) via longitudinal and multi-sequence Magnetic Resonance Imaging (MRI) data analysis.

RESULTS

The proposed WMHs risk prediction model could accurately predict the Fazekas ratings based on the tile scores from T2-FLAIR MRI images with accuracy of 0.656, 0.621 in training data set and test set, respectively. The longitudinal MRI validation revealed that most of the high-risk tiles predicted by the WMHs risk prediction model in the baseline images had WMHs in the corresponding positions in the longitudinal images. The validation on multi-sequence MRI demonstrated that WMHs were associated with GM and WM atrophies, WM micro-structural and perfusion alterations in high-risk tiles, and multi-modal MRI measures of most high-risk tiles showed significant associations with Mini Mental State Examination (MMSE) score.

CONCLUSION

Our proposed WMHs risk prediction model can not only accurately evaluate WMH severities according to Fazekas scales, but can also uncover potential markers of WMHs across modalities. The WMHs risk prediction model has the potential to be used for the early detection of WMH-related alterations in the entire brain and WMH-induced cognitive decline.

Cortical Thickness and Brain Connectivity Mediate the Relation Between White Matter Hyperintensity and Information Processing Speed in Cerebral Small Vessel Disease

White matter hyperintensities of presumed vascular origin (WMH) are the most common imaging feature of cerebral small vessel disease (cSVD) and are associated with cognitive impairment, especially information processing speed (IPS) deficits. However, it is unclear how WMH can directly impact IPS or whether the cortical thickness and brain connectivity mediate such association. In this study, it was evaluated the possible mediating roles of cortical thickness and brain (structural and functional) connectivity on the relationship between WMH (also considering its topography distribution) and IPS in 389 patients with cSVD from the RUN-DMC (Radboud University Nijmegen Diffusion tensor and Magnetic resonance imaging Cohort) database. Significant (p < 0.05 after multiple comparisons correction) associations of WMH volume and topography with cortical thickness, brain connectivity, and IPS performance in cSVD individuals were found. Additionally, cortical thickness and brain structural and functional connectivity were shown to mediate the association of WMH volume and location with IPS scores. More specifically, frontal cortical thickness, functional sensorimotor network, and posterior thalamic radiation tract were the essential mediators of WMH and IPS in this clinical group. This study provided insight into the mechanisms underlying the clinical relevance of white matter hyperintensities in information processing speed deficits in cSVD through cortical thinning and network disruptions.

Epigenetic Clock Explains White Matter Hyperintensity Burden Irrespective of Chronological Age

In this manuscript we studied the relationship between WMH and biological age (B-age) in patients with acute stroke. We included in this study 247 patients with acute stroke recruited at Hospital del Mar having both epigenetic (DNA methylation) and magnetic resonance imaging data. WMH were measured using a semi-automated method. B-age was calculated using two widely used methods: the Hannum and Horvath formulas. We used multiple linear regression models to interrogate the role of B-age on WMH volume after adjusting for chronological age (C-age) and other covariables. Average C-age of the sample was 68.4 (±11.8) and we observed a relatively high median WMH volume (median = 8.8 cm³, Q1-Q3 = 4.05-18.8). After adjusting for potential confounders, we observed a significant effect of B-age_Hannum on WMH volume (β_Hannum = 0.023, p-value = 0.029) independently of C-age, which remained significant (β_C-age = 0.021, p-value = 0.036). Finally, we performed a mediation analysis, which allowed us to discover that 42.7% of the effect of C-age on WMH is mediated by B-age_Hannum. On the other hand, B-age_Hoarvath showed no significant associations with WMH after being adjusted for C-age. In conclusion, we show for the first time that biological age, measured through DNA methylation, contributes substantially to explain WMH volumetric burden irrespective of chronological age.

Periventricular rather than deep white matter hyperintensities mediate effects of hypertension on cognitive performance in the population-based 1000BRAINS study

OBJECTIVES

White matter hyperintensities (WMH) of presumed vascular origin are frequent in cerebral MRI of older people. They represent a sign of small vessel disease, are promoted by arterial hypertension, and relate to cognitive deficits. The interdependence of blood pressure and its treatment, WMH, and cognitive performance has not systematically been studied in population-based studies.

METHODS

Consequently, we analysed the interdependence of SBP, DBP, and antihypertensive medications, as well as BP/treatment category, with WMH and cognitive performance in 560 participants of the population-based 1000BRAINS study.

RESULTS

BP, its treatment, and BP/treatment category were moderately associated with cognitive performance (e.g. unadjusted β  = -0.10, 95%CI = -0.19 to -0.02 for the association of SBP (per standard deviation of 17.2 mmHg) with global cognition (per standard deviation of 0.5 z score)]. The harmful effect of BP on cognition was strongly mediated by periventricular hyperintensities (PVH), which were significantly associated with both SBP [ β  = 0.24, 95% CI = 0.14-0.34 (per 1-point-increase in Fazekas score)] and global cognition ( β  = -0.22, 95%CI =  -0.32 to -0.13). Thus, PVH mediated as much as 52% of the effects of SBP on cognitive performance. Mediation was less strong for deep white matter hyperintensities (DWMH, 16%), which showed less association with SBP ( β  = 0.14, 95% CI = 0.05-0.24) and global cognition ( β  = -0.12, 95%CI = -0.21 to -0.03). Regarding different cognitive domains, PVH most strongly mediated effects of SBP on nonverbal memory (94%) and executive function (81%).

CONCLUSION

Our results indicate that PVH may predispose to cognitive impairment associated with hypertension, especially in the domains of nonverbal memory and executive function.