Abstract:
Ischemic strokes disrupt brain networks, leading to remote effects in key regions like the thalamus, a critical hub for brain functions. However, non-invasive methods to quantify these remote consequences still need to be explored. This study aimed to demonstrate that MRI-derived R2* changes can capture iron accumulation linked with inflammation secondary to stroke-induced disconnection. In order to link remote R2* changes to stroke-induced disconnection, we first conducted a secondary analysis of 156 prospectively included stroke patients who underwent MRI at baseline and 1-year follow-up. We mapped fibers disconnected by baseline infarcts to compare the R2* changes over 1 year according to the disconnectivity status in specific thalamic nuclei groups. We also identified the variables associated with elevated R2* at 1 year in a multivariate context through linear regressions. In parallel, to understand the biological underpinning of the remote R2* changes, we set up a translational mouse model through photothrombotic induction of focal cortical infarcts or sham procedures in 110 C57BL/6J mice. We explored the mice through combinations of in vivo MRI at 72h, 2-, 4-, and 8-weeks, histology, qPCR for gene expression, mass spectrometry for iron concentration quantification, and additional ex vivo high-resolution diffusion tensor imaging. In stroke patients, we found a significant increase of R2* within severely disconnected medial and lateral thalamic nuclei groups from baseline to 1 year. At the same time, no change occurred if these structures were not disconnected. We also showed that the disconnectivity status at baseline was significantly associated with R2* at follow-up, independently from confounders, establishing a direct and independent relationship between baseline disconnection and the subsequent R2* increase within the associated locations. In mice, we recapitulated the patients' conditions by observing increased R2* in the stroke groups, specifically within the disconnected thalamic nuclei. Such remote and focal R2* changes peaked at 2 weeks, preceding and correlating with longer-term atrophy at 8 weeks. We established that the remote R2* increase was spatially and temporally correlated with a significant increase of chemically determined iron load bound to ferritin within activated microglial cells. This study provides critical evidence that R2* is a sensitive marker of inflammation secondary to network disconnection, potentially informing future neuroprotective strategies targeting remote brain regions after stroke.

Abstract:
CONTEXT: Obesity is accompanied by damages to several tissues, including the brain. Pathological data and animal models have demonstrated an increased inflammatory reaction in hypothalamus and hippocampus. OBJECTIVE: We tested whether we could observe such pathological modifications in vivo through quantitative MRI metrics. DESIGN: This prospective study was conducted between May 2019 and November 2022. SETTING: The study was conducted in the Specialised Center for the Care of Obesity in a French Universitary Hospital. PATIENTS: Twenty seven patients with obesity and 23 age and gender-paired normal-weight controls were prospectively recruited. INTERVENTIONS: All participants were explored on brain MRI. Anthropometric and biological data, eating behavior, anxiety, depression and memory performance were assessed on both groups. MAIN OUTCOME MEASURE: The main outcome measure was brain MRI with the following parametric maps: quantitative susceptibility mapping (QSM), mean diffusivity (MD), fractional anisotropy (FA), magnetization transfer ratio map (MTR) and T2 relaxivity map (R2). RESULTS: In the hypothalamus, patients with obesity had higher FA, lower QSM compared to normal-weight controls. In the hippocampus, patients with obesity had higher FA and lower MD. There was no correlation between imaging biomarkers and eating behavior or anxiety. CONCLUSION: Our findings are consistent with the presence of neuro-inflammation in brain regions involved in food intake. In vivo brain biomarkers from quantitative MRI appear to provide an incremental information for the assessment of brain damages in patients with obesity.

Abstract:
Microstructural changes after an ischemic stroke (IS) have mainly been described in white matter. Data evaluating microstructural changes in gray matter (GM) remain scarce. The aim of the present study was to evaluate the integrity of GM on longitudinal data using mean diffusivity (MD), and its influence on post-IS cognitive performances. A prospective study was conducted, including supra-tentorial IS patients without pre-stroke disability. A cognitive assessment was performed at baseline and 1 year, including a Montreal Cognitive Assessment, an Isaacs set test, and a Zazzo cancelation task (ZCT): completion time and number of errors. A 3-T brain MRI was performed at the same two time-points, including diffusion tensor imaging for the assessment of GM MD. GM volume was also computed, and changes in GM volume and GM MD were evaluated, followed by the assessment of the relationship between these structural changes and changes in cognitive performances. One hundred and four patients were included (age 68.5 +/- 21.5, 38.5% female). While no GM volume loss was observed, GM MD increased between baseline and 1 year. The increase of GM MD in left fronto-temporal regions (dorsolateral prefrontal cortex, superior and medial temporal gyrus, p < 0.05, Threshold-Free Cluster Enhancement, 5000 permutations) was associated with an increase time to complete ZCT, regardless of demographic confounders, IS volume and location, GM, and white matter hyperintensity volume. GM integrity deterioration was thus associated with processing speed slowdown, and appears to be a biomarker of cognitive frailty. This broadens the knowledge of post-IS cognitive impairment mechanisms.

Abstract:
BACKGROUND AND PURPOSE: An early understanding of stroke mechanism may improve treatment and outcome in patients presenting with large vessel occlusion stroke (LVOS) treated with mechanical thrombectomy (MT). We aimed to investigate whether spontaneous external carotid artery (ECA) embolism detection during MT is associated with stroke etiology and clinical outcome. METHODS: We retrospectively reviewed our prospectively maintained institutional database including consecutive patients with anterior circulation LVOS treated with MT between January 2015 and August 2020. RESULTS: An ECA embolus was detected in 68 of 1298 patients (5.2%). The kappa coefficient for interobserver agreement was 0.89 (95% confidence interval [CI] 0.82-0.95). ECA embolism was significantly associated with intracranial internal carotid artery (ICA) occlusion (p < 0.001), cardioembolic etiology (p < 0.001) and a lower clot burden score (p < 0.001). Day-1 variation of National Institutes of Health Stroke Scale score (adjusted odds ratio [OR] -2.7, 95% CI -4.9 to 0.3; p = 0.021) and delta Alberta Stroke Program Early Computed Tomography Score (adjusted OR 0.9, 95% CI 0.2 to 1.5; p = 0.004) were worse among patients with ECA emboli. There was no significant difference in 90-day functional outcome between groups (adjusted OR 0.8, 95% CI 0.42 to 1.52; p = 0.50). CONCLUSION: In patients with anterior circulation LVOS treated with MT, ECA embolism was significantly associated with cardioembolic etiology, high thrombus burden and proximal intracranial ICA occlusions. This underexplored angiographic pattern might provide a valuable etiologic clue to the underlying cause of anterior circulation LVOS and may also help determine the appropriate revascularization strategy.

Abstract:
Background A target mismatch profile can identify good clinical response to recanalization after acute ischemic stroke, but does not consider region specificities. Purpose To test whether location-weighted infarction core and mismatch, determined from diffusion and perfusion MRI performed in patients with acute stroke, could improve prediction of good clinical response to mechanical thrombectomy compared with a target mismatch profile. Materials and Methods In this secondary analysis, two prospectively collected independent stroke data sets (2012-2015 and 2017-2019) were analyzed. From the brain before stroke (BBS) study data (data set 1), an eloquent map was computed through voxel-wise associations between the infarction core (based on diffusion MRI on days 1-3 following stroke) and National Institutes of Health Stroke Scale (NIHSS) score. The French acute multimodal imaging to select patients for mechanical thrombectomy (FRAME) data (data set 2) consisted of large vessel occlusion-related acute ischemic stroke successfully recanalized. From acute MRI studies (performed on arrival, prior to thrombectomy) in data set 2, target mismatch and eloquent (vs noneloquent) infarction core and mismatch were computed from the intersection of diffusion- and perfusion-detected lesions with the coregistered eloquent map. Associations of these imaging metrics with early neurologic improvement were tested in multivariable regression models, and areas under the receiver operating characteristic curve (AUCs) were compared. Results Data sets 1 and 2 included 321 (median age, 69 years [IQR, 58-80 years]; 207 men) and 173 (median age, 74 years [IQR, 65-82 years]; 90 women) patients, respectively. Eloquent mismatch was positively and independently associated with good clinical response (odds ratio [OR], 1.14; 95% CI: 1.02, 1.27; P = .02) and eloquent infarction core was negatively associated with good response (OR, 0.85; 95% CI: 0.77, 0.95; P = .004), while noneloquent mismatch was not associated with good response (OR, 1.03; 95% CI: 0.98, 1.07; P = .20). Moreover, adding eloquent metrics improved the prediction accuracy (AUC, 0.73; 95% CI: 0.65, 0.81) compared with clinical variables alone (AUC, 0.65; 95% CI: 0.56, 0.73; P = .01) or a target mismatch profile (AUC, 0.67; 95% CI: 0.59, 0.76; P = .03). Conclusion Location-weighted infarction core and mismatch on diffusion and perfusion MRI scans improved the identification of patients with acute stroke who would benefit from mechanical thrombectomy compared with the volume-based target mismatch profile. Clinical trial registration no. NCT03045146 (c) RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Nael in this issue.

Abstract:
OBJECTIVES: Investigating differential vulnerability of thalamic nuclei in multiple sclerosis (MS). METHODS: In a secondary analysis of prospectively collected datasets, we pooled 136 patients with MS or clinically isolated syndrome and 71 healthy controls all scanned with conventional 3D-T1 and white-matter-nulled magnetization-prepared rapid gradient echo (WMn-MPRAGE) and tested for cognitive performance. T1-based thalamic segmentation was compared with the reference WMn-MPRAGE method. Volumes of thalamic nuclei were compared according to clinical phenotypes and cognitive profile. RESULTS: T1- and WMn-MPRAGE provided comparable segmentations (0.84 +/- 0.13 < volume-similarity-index < 0.95 +/- 0.03). Medial and posterior thalamic groups were significantly more affected than anterior and lateral groups. Cognitive impairment related to volume loss of the anterior group. CONCLUSION: Thalamic nuclei closest to the third ventricle are more affected, with cognitive consequences.

Abstract:
Normal-appearing white matter (NAWM) is a hub of plasticity, but data relating to its influence on post-ischemic stroke (IS) outcome remain scarce. The aim of this study was to evaluate the relationship between NAWM integrity and cognitive outcome after an IS. A longitudinal study was conducted including supra-tentorial IS patients. A 3-Tesla brain MRI was performed at baseline and 1 year, allowing the analyses of mean fractional anisotropy (FA) and mean diffusivity (MD) in NAWM masks, along with the volume of white matter hyperintensities (WMH) and IS. A Montreal Cognitive Assessment (MoCA), an Isaacs set test, and a Zazzo's cancellation task were performed at baseline, 3 months and 1 year. Mixed models were built, followed by Tract-based Spatial Statistics (TBSS) analyses. Ninety-five patients were included in the analyses (38% women, median age 69 +/- 20). FA significantly decreased, and MD significantly increased between baseline and 1 year, while cognitive scores improved. Patients who decreased their NAWM FA more over the year had a slower cognitive improvement on MoCA (beta = - 0.11, p = 0.05). The TBSS analyses showed that patients who presented the highest decrease of FA in various tracts of white matter less improved their MoCA performances, regardless of WMH and IS volumes, demographic confounders, and clinical severity. NAWM integrity deteriorates over the year after an IS, and is associated with a cognitive recovery slowdown. The diffusion changes recorded here in patients starting with an early preserved white matter structure could have long term impact on cognition.

Abstract:
BACKGROUND AND PURPOSE: Accurate quantification of WM lesion load is essential for the care of patients with multiple sclerosis. We tested whether the combination of accelerated 3D-FLAIR and denoising using deep learning-based reconstruction could provide a relevant strategy while shortening the imaging examination. MATERIALS AND METHODS: Twenty-eight patients with multiple sclerosis were prospectively examined using 4 implementations of 3D-FLAIR with decreasing scan times (4 minutes 54 seconds, 2 minutes 35 seconds, 1 minute 40 seconds, and 1 minute 15 seconds). Each FLAIR sequence was reconstructed without and with denoising using deep learning-based reconstruction, resulting in 8 FLAIR sequences per patient. Image quality was assessed with the Likert scale, apparent SNR, and contrast-to-noise ratio. Manual and automatic lesion segmentations, performed randomly and blindly, were quantitatively evaluated against ground truth using the absolute volume difference, true-positive rate, positive predictive value, Dice similarity coefficient, Hausdorff distance, and F1 score based on the lesion count. The Wilcoxon signed-rank test and 2-way ANOVA were performed. RESULTS: Both image-quality evaluation and the various metrics showed deterioration when the FLAIR scan time was accelerated. However, denoising using deep learning-based reconstruction significantly improved subjective image quality and quantitative performance metrics, particularly for manual segmentation. Overall, denoising using deep learning-based reconstruction helped to recover contours closer to those from the criterion standard and to capture individual lesions otherwise overlooked. The Dice similarity coefficient was equivalent between the 2-minutes-35-seconds-long FLAIR with denoising using deep learning-based reconstruction and the 4-minutes-54-seconds-long reference FLAIR sequence. CONCLUSIONS: Denoising using deep learning-based reconstruction helps to recognize multiple sclerosis lesions buried in the noise of accelerated FLAIR acquisitions, a possibly useful strategy to efficiently shorten the scan time in clinical practice.

Abstract:
OBJECTIVES: 3D-fluid attenuation inversion recovery (FLAIR) collected 4 h after intravenous gadolinium injection can delineate the perilymphatic space (PLS) from the endolymphatic space (ELS) to capture endolymphatic hydrops, the pathological counterpart of Meniere's disease. We aimed to optimize visualization of such inner ear internal anatomy using 3D-FLAIR without injection. METHODS: 3D-FLAIR signal from different fluid compartments such as PLS and ELS was first simulated. Then, twenty-two healthy subjects were scanned at 3.0-T MRI with non-injected 3D-FLAIR using variable T2 preparations (T2Preps) (OFF, 200, 400, and 600 ms) and variable inversion times (TIs) (from 224 to 5000 ms) and different resolutions (1.0 x 1.0 x 1.5, 0.6 x 0.6 x 0.8, and 0.6 x 0.6 x 0.6 mm(3)). The relative contrast between PLS and ELS and the visibility of the saccule and utricle were assessed. Additionally, non-injected 3D-FLAIR with the optimal setting was tested in a Meniere patient and compared with gadolinium-injected 3D-FLAIR. RESULTS: The PLS and ELS were differentiated when T2Prep was used but not without. The relative contrast was larger with T2Prep at 400 ms than at 200 or 600 ms (0.72 +/- 0.22 vs. 0.44 +/- 0.11, p = 0.019; and 0.72 +/- 0.22 vs. 0.46 +/- 0.28, p = 0.034, respectively). The saccule and utricle were best delineated in 87. % cases with T2Prep = 400 and TI = 2100 ms at the highest resolution. Visualization of the saccule and utricle in the optimized non-injected 3D-FLAIR was similar to conventional injected 3D-FLAIR in a patient. CONCLUSIONS: Combining a specific T2Prep and TI in non-injected 3D-FLAIR could separate PLS and ELS and even the saccule and utricle, paving the way toward future application to diagnose Meniere's disease. KEY POINTS: * MRI can capture the internal anatomy of inner ear without injection of contrast media. * Specific parameters consisting of a T2 preparation of 400 ms and an inversion time of 2100 ms must be used to visualize the saccule and utricle on non-injected 3D-FLAIR.

Abstract:
BACKGROUND: Gadolinium leakage in ocular structures (GLOS) was recently observed in fluid-attenuated inversion recovery (FLAIR) images obtained the day after an initial gadolinium injection in stroke patients. The specificity of GLOS to stroke and its mechanisms remain unclear. OBJECTIVE: We investigated the factors associated with GLOS in a cohort of patients presenting with acute neurological deficits. MATERIALS AND METHODS: This retrospective study included consecutive patients admitted to our stroke unit for acute neurological deficit between July 2017 and August 2018 who underwent baseline brain magnetic resonance imaging with the injection of a macrocyclic gadolinium agent and another scan without injection within 72 hours. The patients were separated into a stroke group and a stroke mimic group based on diffusion-weighted images. Gadolinium leakage in ocular structures was defined as a bright signal in the vitreous in follow-up FLAIR compared with baseline FLAIR (pregadolinium). Clinical data were collected together with imaging features from the baseline scans, including the volume of the infarct and of hypoperfusion if applicable, white matter hyperintensities, the number of lacunes, and the number of microbleeds, which were combined to yield a small vessel disease (SVD) score. We compared the prevalence of GLOS in both groups using the chi2 test. In the entire cohort, univariate and multivariate regression models were used to test the associations between GLOS and the collected data. RESULTS: Among the 467 patients included in the study, GLOS was observed in similar proportions in the stroke group (32.2%, 136/422) and the stroke mimic group (28.9%, 13/45; mean difference, 3.3%; 95% confidence interval, -10.9 to 17.6; P = 0.65). In univariate analysis, GLOS was associated with older age, increased prevalence of vascular risk factors, brain imaging features of SVD (white matter hyperintensities, lacunes, microbleeds), as well as with impairment of renal function and increased dose of gadolinium. No associations were found with factors related to stroke, such as its volume, acute treatment, or rate of recanalization. Multivariate analyses showed that aging (P < 0.001), diabetes (P = 0.010), severe renal failure (P = 0.004), and increased dose of gadolinium (P < 0.001) were independent contributors to GLOS. CONCLUSIONS: Gadolinium leakage in ocular structures, which occurs more commonly at higher concentrations of gadolinium, is not specific to stroke and may represent increased permeability of the blood-retinal barrier associated with age- and vascular risk factor-related SVD.