ORIGINAL RESEARCH published: 10 September 2021 doi: 10.3389/fneur.2021.700616 Excessive White Matter Edited by: Hyperintensity Increases Martin Ebinger, Charité – Universitätsmedizin Susceptibility to Poor Functional Berlin, Germany Reviewed by: Outcomes After Acute Ischemic Russell Chander, University of New South Stroke Wales, Australia Joanna M. Wardlaw, Sungmin Hong 1*, Anne-Katrin Giese 2 , Markus D. Schirmer 1,3 , Anna K. Bonkhoff 1 , University of Edinburgh, Martin Bretzner 1,4 , Pamela Rist 5 , Adrian V. Dalca 6,7 , Robert W. Regenhardt 1 , United Kingdom Mark R. Etherton 1 , Kathleen L. Donahue 1 , Marco Nardin 1 , Steven J. T. Mocking 7 , *Correspondence: Elissa C. McIntosh 7 , John Attia 8,9 , Oscar R. Benavente 10 , John W. Cole 11 , Sungmin Hong Amanda Donatti 12 , Christoph J. Griessenauer 13,14 , Laura Heitsch 15,16 , Lukas Holmegaard 17 , shong20@mgh.harvard.edu Katarina Jood 17 , Jordi Jimenez-Conde 18 , Jaume Roquer 18 , Steven J. Kittner 11 , Robin Lemmens 19,20 , Christopher R. Levi 8,21 , Caitrin W. McDonough 22 , James F. Meschia 23 , Specialty section: Chia-Ling Phuah 16 , Arndt Rolfs 24 , Stefan Ropele 25 , Jonathan Rosand 1,7,26 , This article was submitted to Tatjana Rundek 27 , Ralph L. Sacco 27 , Reinhold Schmidt 25 , Christian Enzinger 25 , Stroke, a section of the journal Pankaj Sharma 28,29 , Agnieszka Slowik 30 , Alessandro Sousa 12 , Tara M. Stanne 17 , Frontiers in Neurology Daniel Strbian 31 , Turgut Tatlisumak 32,33 , Vincent Thijs 34,35 , Achala Vagal 36 , Johan Wasselius 37,38 , Daniel Woo 39 , Ramin Zand 40 , Patrick F. McArdle 41 , Received: 26 April 2021 Bradford B. Worrall 42 , Ona Wu 7 , Christina Jern 43 , Arne G. Lindgren 44,45 , Jane Maguire 46 , Accepted: 28 July 2021 Liisa Tomppo 31 , Polina Golland 6 , Natalia S. Rost 1 and the MRI- Published: 10 September 2021 GENIE and GISCOME Investigators and the International Stroke Genetics Consortium Citation: 1 Hong S, Giese A-K, Schirmer MD, J. Philip Kistler Stroke Research Center, Massachusetts General Hospital and Harvard Medical School, Boston, MA, Bonkhoff AK, Bretzner M, Rist P, United States, 2 Department of Neurology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 3 Clinic for Dalca AV, Regenhardt RW, Neuroradiology, University Hospital Bonn, Bonn, Germany, 4 Univ. Lille, Inserm, CHU Lille, U1172 - LilNCog (JPARC) - Lille Etherton MR, Donahue KL, Nardin M, Neurosciences & Cognition, Lille, France, 5 Division of Preventive Medicine, Department of Medicine, Brigham and Women’s Mocking SJT, McIntosh EC, Attia J, Hospital and Harvard Medical School, Boston, MA, United States, 6 Computer Science and Artificial Intelligence Lab, Benavente OR, Cole JW, Donatti A, Massachusetts Institute of Technology, Boston, MA, United States, 7 Athinoula A. Martinos Center for Biomedical Imaging, Griessenauer CJ, Heitsch L, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States, 8 Hunter Medical Research Holmegaard L, Jood K, Institute, Newcastle, NSW, Australia, 9 School of Medicine and Public Health, University of Newcastle, Newcastle, NSW, Jimenez-Conde J, Roquer J, Australia, 10 Division of Neurology, Department of Medicine, University of British Columbia, Vancouver, BC, Canada, 11 Kittner SJ, Lemmens R, Levi CR, Department of Neurology, University of Maryland School of Medicine and Veterans Affairs Maryland Health Care System, McDonough CW, Meschia JF, Baltimore, MD, United States, 12 School of Medical Sciences, University of Campinas (UNICAMP) and the Brazilian Institute of Phuah C-L, Rolfs A, Ropele S, Neuroscience and Neurotechnology (BRAINN), Campinas, Brazil, 13 Department of Neurosurgery, Geisinger, Danville, PA, Rosand J, Rundek T, Sacco RL, United States, 14 Research Institute of Neurointervention, Paracelsus Medical University, Salzburg, Austria, 15 Division of Schmidt R, Enzinger C, Sharma P, Emergency Medicine, Washington University School of Medicine, St. Louis, MO, United States, 16 Department of Neurology, Slowik A, Sousa A, Stanne TM, Washington University School of Medicine & Barnes-Jewish Hospital, St. Louis, MO, United States, 17 Department of Clinical Strbian D, Tatlisumak T, Thijs V, Neuroscience, Institute of Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg and Vagal A, Wasselius J, Woo D, Zand R, Department of Neurology, The Sahlgrenska University Hospital, Gothenburg, Sweden, 18 Department of Neurology, McArdle PF, Worrall BB, Wu O, Neurovascular Research Group (NEUVAS), IMIM-Hospital del Mar (Institut Hospital del Mar d’Investigacions M‘ediques), Jern C, Lindgren AG, Maguire J, Universitat Autonoma de Barcelona, Barcelona, Spain, 19 KU Leuven - University of Leuven, Department of Neurosciences, Tomppo L, Golland P, Rost NS and Experimental Neurology and Leuven Research Institute for Neuroscience and Disease (LIND), Leuven, Belgium, 20 VIB, the MRI-GENIE and GISCOME Vesalius Research Center, Laboratory of Neurobiology, University Hospitals Leuven, Department of Neurology, Leuven, Investigators and the International Belgium, 21 Department of Neurology, John Hunter Hospital, Newcastle, NSW, Australia, 22 Department of Pharmacotherapy Stroke Genetics Consortium (2021) and Translational Research and Center for Pharmacogenomics, University of Florida, Gainesville, FL, United States, 23 Excessive White Matter Hyperintensity Department of Neurology, Mayo Clinic, Jacksonville, FL, United States, 24 Centogene AG, Rostock, Germany, 25 Increases Susceptibility to Poor Department of Neurology, Clinical Division of Neurogeriatrics, Medical University Graz, Graz, Austria, 26 Center for Genomic Functional Outcomes After Acute Medicine, Massachusetts General Hospital, Boston, MA, United States, 27 Department of Neurology and Evelyn F. McKnight Ischemic Stroke. Brain Institute, Miller School of Medicine, University of Miami, Miami, FL, United States, 28 Institute of Cardiovascular Front. Neurol. 12:700616. Research, Royal Holloway University of London (ICR2UL), Egham, United Kingdom, 29 St. Peter’s and Ashford Hospitals, doi: 10.3389/fneur.2021.700616 Egham, United Kingdom, 30 Department of Neurology, Jagiellonian University Medical College, Krakow, Poland Frontiers in Neurology | www.frontiersin.org 1 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes 31 Department of Neurology, Helsinki University Hospital and Clinical Neurosciences, University of Helsinki, Helsinki, Finland, 32 Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden, 33 Department of Neurology, Sahlgrenska University Hospital, Gothenburg, Sweden, 34 Stroke Division, Florey Institute of Neuroscience and Mental Health, Heidelberg, VIC, Australia, 35 Department of Neurology, Austin Health, Heidelberg, VIC, Australia, 36 Department of Radiology, University of Cincinnati College of Medicine, Cincinnati, OH, United States, 37 Department of Clinical Sciences Lund, Radiology, Lund University, Lund, Sweden, 38 Department of Radiology, Neuroradiology, Skåne University Hospital, Malmo, Sweden, 39 Department of Neurology and Rehabilitation Medicine, University of Cincinnati College of Medicine, Cincinnati, OH, United States, 40 Department of Neurology, Geisinger, Danville, PA, United States, 41 Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, University of Maryland School of Medicine, Baltimore, MD, United States, 42 Departments of Neurology and Public Health Sciences, University of Virginia, Charlottesville, VA, United States, 43 Department of Laboratory Medicine, Institute of Biomedicine, The Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden, 44 Department of Neurology and Rehabilitation Medicine, Skåne University Hospital, Lund, Sweden, 45 Department of Clinical Sciences Lund, Neurology, Lund University, Lund, Sweden, 46 School of Nursing and Midwifery, University of Technology Sydney, Sydney, NSW, Australia Objective: To personalize the prognostication of post-stroke outcome using MRI-detected cerebrovascular pathology, we sought to investigate the association between the excessive white matter hyperintensity (WMH) burden unaccounted for by the traditional stroke risk profile of individual patients and their long-term functional outcomes after a stroke. Methods: We included 890 patients who survived after an acute ischemic stroke from the MRI-Genetics Interface Exploration (MRI-GENIE) study, for whom data on vascular risk factors (VRFs), including age, sex, atrial fibrillation, diabetes mellitus, hypertension, coronary artery disease, smoking, prior stroke history, as well as acute stroke severity, 3- to−6-month modified Rankin Scale score (mRS), WMH, and brain volumes, were available. We defined the unaccounted WMH (uWMH) burden via modeling of expected WMH burden based on the VRF profile of each individual patient. The association of uWMH and mRS score was analyzed by linear regression analysis. The odds ratios of patients who achieved full functional independence (mRS < 2) in between trichotomized uWMH burden groups were calculated by pair-wise comparisons. Results: The expected WMH volume was estimated with respect to known VRFs. The uWMH burden was associated with a long-term functional outcome (β = 0.104, p < 0.01). Excessive uWMH burden significantly reduced the odds of achieving full functional independence after a stroke compared to the low and average uWMH burden [OR = 0.4, 95% CI: (0.25, 0.63), p <0.01 and OR = 0.61, 95% CI: (0.42, 0.87), p <0.01, respectively]. Conclusion: The excessive amount of uWMH burden unaccounted for by the traditional VRF profile was associated with worse post-stroke functional outcomes. Further studies are needed to evaluate a lifetime brain injury reflected in WMH unrelated to the VRF profile of a patient as an important factor for stroke recovery and a plausible indicator of brain health. Keywords: white matter hyper intensity, stroke, brain health, brain vulnerability, post-stroke outcomes, functional independence, functional outcome after acute stroke, acute ischemic stroke INTRODUCTION It has not been fully characterized taking into consideration multiple clinical characteristics, such as age, sex, and other The ability of the brain to recover after an acute ischemic vascular risk factors (VRFs), that comprise a “traditional” stroke stroke (AIS) is linked to the pre-stroke burden of white matter risk profile. Because those characteristics might be associated hyperintensity (WMH), a neuroradiological biomarker of brain with stroke outcomes and also with accumulating WMH burden, health (1–6). However, there is a significant variability in quantifying the WMH burden effect on functional outcomes functional outcomes after AIS in association with WMH burden. independent of clinical characteristics is challenging (2, 7–14). Frontiers in Neurology | www.frontiersin.org 2 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes In addition to and compounded by age, a VRF profile is linked framework consists of an automatic whole-brain extraction, to overall brain health as reflected in the WMH burden— intensity normalization, and WMH segmentation that leverages for example, an older patient with a history of hypertension a deep neural network specifically for clinical grade T2-FLAIR is likely to have a larger WMH volume and less likely to images. Total brain volume was quantified automatically by a achieve functional independence after AIS (2, 15). Prior studies similar previously described framework (21). Both WMH and using standard multiple regression models had a limited success total brain segmentation results were evaluated, and the quality in characterizing and quantifying an independent association was controlled by experts and experienced raters. WMHv was between conventionally measured WMH volume and post- adjusted by the total brain volume and log-transformed to correct stroke outcome because the effects of confounding VRFs to its skewed distribution. The brain-volume-adjusted and log- the association of WMH and post-stroke outcomes were not transformed WMHv will be denoted as WMHv for simplicity in explicitly accounted for (1, 2, 8, 9, 11, 12, 14, 15). the rest of this paper. In this study, we hypothesized that the excessive portion of Out of 6,600 patients, the complete clinical characteristics of WMH burden unaccounted for by the clinical characteristics 5,822 patients were reported. Acute stroke severity and long- of an individual patient was an important indicator of brain term functional outcome data were available for 1,097 patients health: i.e., the unaccounted WMH (uWMH) burden would out of those 6,015 patients and measured by the acute NIHSS show the clear association of the lifelong chronic brain damage and the 3- to 6-month modified Rankin Scale (mRS) scores, independent of the VRF profile of a patient and the post- respectively. After the quality control of WMH and total brain stroke functional outcomes, especially in achieving full functional segmentation, 924 AIS patients remained. Thirty-four patients independence after stroke. To test this hypothesis, we first who did not survive after stroke at the time of the mRS score defined the uWMH burden given the VRF profiles of patients evaluation were excluded from the analysis, which resulted in a to extrinsically quantify the portion of WMH burden that was total of 890 patients. The patient selection and the criteria are independent of the VRF profiles. We then analyzed the uWMH summarized in Figure 1. association with post-stroke functional outcomes leveraging a large international multi-site cohort study for AIS, the MRI- Genetics Interface Exploration (MRI-GENIE), for the analyses Unaccounted WMH Burden and validation (16, 17). To measure uWMH independent of the VRF profile of a patient, we first modeled the association of WMH burden and the profile of the patient by multiple log-linear regression. Age, sex, AF, DM, METHODS HTN, CAD, SMK, and PS were considered in the analysis as Standard Protocol Approvals, explanatory variables, and WMHv was the target variable. NIHSS was included in the analysis as well to account for the effect Registrations, and Patient Consents of acute stroke severity on functional stroke outcomes for later The image and clinical data were acquired from the MRI-GENIE analyses. The normality of WMHv data of the AIS population study (16). The MRI-GENIE study is a large international cohort with respect to the multiple log-linear regression model was study of 19 sites, contributing 6,600 AIS patients with phenotypic, tested by the Shapiro–Wilks test. radiographic, and genotypic data. Patients were recruited in a The expected burden of WMH with respect to the VRF profile hospital-based setting through Stroke Genetic Network (SiGN). of a patient was calculated as the trend of the multiple log-linear The informed and written consent forms, including sharing regression model. The uWMH burden measured the amount of of de-identified demographic, clinical phenotypic, and imaging WMH burden that was not accounted for by the clinical profile data, were obtained from all patients or their legally authorized of a patient, i.e., the residual WMH burden of the multiple log- representative (18, 19). Each site received the approval of their linear regression model—for example, a positive uWMH value internal review board. indicated that a patient had excessive WMHv than expected from her or his clinical profile, and negative means lower WMHv than Data, Clinical Characteristics, and WMH expected. The normality of uWMH burden was tested by the Quantification Shapiro–Wilks test. Age, sex, atrial fibrillation (AF), diabetes mellitus (DM), We analyzed the continuous association of uWMH burden hypertension (HTN), coronary artery disease (CAD), smoking and post-stroke functional outcome measured by the 3- to 6- (SMK), and prior stroke history (PS) were included as the clinical month mRS score using the univariate linear regression model characteristics of patients. Each vascular risk factor was recorded for the continuous association between uWMH burden and as a binary variable at hospital admission. The acute stroke post-stroke functional outcomes. severity of a patient was measured according to NIH Stroke Scale We investigated how uWMH affected the possibility of (NIHSS) at hospital admission. functional independence after stroke by trichotomized analysis. T2 FLAIR MR images were obtained within the first 24 h We trichotomized patients into low, expected, and excessive from symptom onset. A more detailed description of the study eMWH burden groups based on uWMH burden. We defined can be found in previous publications (16, 20). White matter the average group as patients within ±1 standard deviation hyperintensity volume (WMHv) was quantified by a previously from the mean uWMH burden, which resulted in 68% of total described automatic WMH segmentation framework (20). The patients. The low and excessive uWMH burden groups were Frontiers in Neurology | www.frontiersin.org 3 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes FIGURE 1 | Flow chart on patient selection and its criteria. defined as lower and higher 16% of patients of the uWMH burden All statistical analyses were performed with Python SciPy 1.5.4 distribution, respectively. (22), Sci-kit Learn 0.23 (23), and StatsModel 0.12 packages (24). The dichotomized 3- to 6-month mRS score (mRS excellent: mRS score 0–1, mRS moderate–poor: mRS score 2–6) was used RESULTS to measure functional independence after AIS. We dichotomized the mRS scores to mRS excellent and mRS moderate–poor to Population Analysis investigate the association of uWMH burden and full functional The population characteristics of the included patients are independence after stroke. We calculated the pair-wise odds summarized in the first column of Table 1. The multiple log- ratios of patients who achieved full functional independence for linear regression model of WMHv with respect to all included patients between all burden groups (e.g., low vs. expected, low clinical characteristics (age, sex, AF, HTN, DM, CAD, PS, and vs. excessive, and expected vs. excessive). The 95% confidence NIHSS) represented an expected WMHv, ŶWMHv , of a patient interval and the statistical significance of the odds ratio were with respect to the VRF profile of the patient. The regression calculated by the standard z-test to investigate if there was a coefficients of the multiple log-linear regression model are significant difference between the groups in achieving functional summarized in Table 2. The multiple log-linear model explained independence after stroke. 43.3% of the total variation of WMHv (R2 = 0.433). The We performed the subgroup analysis regarding acute stroke uWMH burden was not correlated with any included clinical severity by dividing the patients into mild and severe acute characteristics: i.e., the correlation coefficients of the uWMH stroke severity groups. We divided the patients with respect burden to all included clinical characteristics were 0. The uWMH to their acute NIHSS scores, with a threshold defined as the burden passed the normality test by the Shapiro–Wilks test (p < third quartile (Q3) of the NIHSS scores of the population. 0.001). The uWMH burden showed a significant association We performed the same trichotomized analysis of uWMH with continuous functional outcomes (3-month to 6-month mRS burden and investigated the odds ratios of patient who achieved score: 0–5) in the univariate linear regression model β = full functional independence, defined as “excellent mRS” (with 0.104, p < 0.01. residual non-disabling symptoms, mRS 1, or symptom-free, The second to the fourth columns of Table 1 summarize the mRS 0) between the uWMH burden groups of each acute population characteristics of the low, expected, and excessive severity subgroup. uWMH burden groups. The averages of the uWMH burden of Frontiers in Neurology | www.frontiersin.org 4 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes TABLE 1 | Population clinical characteristics of total acute ischemic stroke patients from the MRI-GENIE study and those included the analysis with available admission NIHSS, 3- to 6-month mRS scores and WMHv, and the low, expected, and excessive uWMH burden groups. Total (n = 5,822) Total included in the analysis (n = 890) Low (n = 140) Expected (n = 599) Excessive (n = 151) Age, average (SD) 61.3 (15.2) 63.7 (15.0) 61.6 (14.1) 64.1 (15.5) 64.2 (13.9) Sex, n (%) 2,416 (41.5%) 379 (42.6%) 59 (42.1%) 251 (41.9%) 69 (45.7%) HTN, n (%) 3,849 (66.1%) 541 (60.8%) 88 (62.9%) 355 (59.3%) 98 (64.9%) DM, n (%) 1,480 (25.4%) 180 (20.2%) 24 (17.1%) 131 (21.9%) 25 (16.6%) AF, n (%) 741 (12.7%) 140 (15.7%) 19 (13.6%) 98 (16.4%) 23 (15.2%) CAD, n (%) 1,021 (17.5%) 140 (15.7%) 13 (9.3%) 105 (17.5%) 22 (14.6%) SMK, n (%) 3,342 (57.4%) 475 (53.4%) 70 (50.0%) 331 (55.3%) 74 (49.0%) PS, n (%) 749 (12.9%) 85 (9.6%) 13 (9.3%) 59 (9.8%) 13 (8.6%) NIHSS, median (IQR) N/A 3 (5) 3 (4) 3 (6) 3 (5) uWMH, avg (std) N/A 0 (1.1) −1.8 (0.5) 0.0 (0.6) 1.6 (0.4) TABLE 2 | The regression coefficients of individual univariate and multiple log-linear regression analyses that modeled the associations of age, sex, VRFs, and NIHSS with WMHv. Y = WMHv Age Sex HTN AF CAD DM SMK PS NIHSS β0 Regression coefficients 0.06 −0.19 0.33 −0.01 −0.1 0.27 0.30 0.33 0.01 −9.82 p <0.01 0.01 <0.01 0.96 0.4 <0.01 <0.01 0.01 0.08 <0.01 TABLE 3 | The odds ratios of patients who achieved full functional independence = 6). The results did not change substantially by including non- (mRS excellent: mRS scores 0–1) of the excessive uWMH burden groups to the survivors. The uWMH burden showed a significant association low (excessive/low) and expected (excessive/expected) groups and of the with functional outcomes (3-month to 6-month mRS score: 0– expected group to the low (expected/low) group. 6) in the univariate regression model, β = 0.091, p = 0.036. Population Excessive/low Excessive/expected Expected/low The odds ratios between the low, expected, and excessive uWMH burden groups did not change substantially from the analysis OR 0.39 0.6 0.65 with post-stroke survivors only. The detailed results are reported p, 95% CI <0.01 (0.25, 0.63) <0.01 (0.42, 0.87) 0.03 (0.44, 0.96) in Supplementary Tables 1–3 in the Supplementary Materials. Acute Stroke Severity Subgroup Analysis low, expected, and excessive eWMH burden groups were −1.8, The third quartile of the NIHSS scores of the population was 0.0, and 1.6, respectively. seven. The median NIHSS scores of the mild (NIHSS <7) and The ratios of patients who achieved full functional severe (NIHSS ≥7) acute stroke severity subgroups were 2 and 11, independence (mRS excellent: mRS score 0–1) in the low, respectively. The overall population characteristics of the acute expected, and excessive uWMH burden groups were 65.7, 55.6, stroke severity subgroups are summarized in Table 4. and 43%, respectively. The odds ratios of the mRS excellent In the mild acute stroke severity subgroup, the ratios of mRS patients between the uWMH burden groups are summarized excellent patients in the low, expected, and excessive uWMH in Table 3. Patients in the excessive uWMH burden groups burden groups were 71, 65.2, and 52.3%, respectively. The odds were less likely to achieve full functional independence after ratios of the mRS excellent patients in the excessive uWMH AIS than the expected and excessive uWMH burden groups. burden group with respect to the low and expected uWMH The odds ratios of the excessive uWMH group with respect burden groups were 0.446 (p < 0.01) and 0.585 (p = 0.01), to the expected and low uWMH burden groups were 0.604 respectively. The odds ratio of the expected uWMH burden (p < 0.01) and 0.394 (p < 0.01), respectively. The odds group compared to the low uWMH burden group was 0.763 ratio of the expected uWMH burden group with respect (p = 0.25). to the low group was 0.653 (p = 0.03). Figure 2A shows In the severe acute stroke severity subgroup, the ratios of mRS the distributions of mRS scores of the low, expected, and excellent patients in the low, expected, and excessive uWMH excessive uWMH burden groups. The distribution of mRS burden groups were 48.5, 27.2, and 17.5%, respectively. The odds scores of the excessive uWMH burden groups was consistently ratios of the mRS excellent patients in the excessive and expected shifted toward worse mRS scores compared to the lower uWMH burden groups with respect to the low uWMH burden uWMH groups. group were 0.225 (p < 0.01) and 0.396 (p = 0.02). The odds ratio As an auxiliary analysis, we conducted the same analyses of the excessive uWMH burden group compared to the expected with all patients, including 34 post-stroke non-survivors (mRS uWMH burden group was 0.569 (p = 0.21). The odds ratios of Frontiers in Neurology | www.frontiersin.org 5 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes FIGURE 2 | (A) The distribution of the modified Rankin Scale (mRS) scores of patients in the low, expected, and excessive unaccounted white matter hyperintensity (uWMH) burden groups. The proportions of mRS scores were shifted toward worse mRS scores for the higher uWMH burden groups. (B) The acute stroke severity subgroup analysis. The proportion of patients who achieved the full functional independence (mRS excellent: mRS 0 and 1) in each uWMH burden group was plotted in blue and in red for the mild [NIH Stroke Scale (NIHSS) <7] and severe (NIHSS ≥7) acute stroke severity groups, respectively. TABLE 4 | Population clinical characteristics of acute ischemic stroke patients in TABLE 5 | Acute stroke severity subgroup analysis. the mild acute stroke severity group (mild, NIHSS < 7) and severe stroke severity group (severe, NIHSS ≥ 7). Acute stroke severity group Excessive/ Excessive/ Expected/ low expected low Mild (n = 666) Severe (n = 224) Mild (n = 666) OR 0.45 0.58 0.76 Age, average (SD) 63.4 (15.2) 64.6 (14.3) p, 95% CI <0.01 (0.26, 0.01 (0.38, 0.25 (0.48, 1.21) Sex, n (%) 273 (41.0%) 106 (47.3%) 0.78) 0.89) HTN, n (%) 408 (61.3%) 133 (59.4%) Severe (n = 224) OR 0.23 0.57 0.40 DM, n (%) 134 (20.1%) 46 (20.5%) p, 95% CI <0.01 (0.08, 0.21 (0.23, 0.02 (0.18, 0.86) AF, n (%) 86 (12.9%) 54 (24.1%) 0.65) 1.39) CAD, n (%) 95 (14.3%) 45 (20.1%) SMK, n (%) 367 (55.1%) 108 (48.2%) The odds ratios of patients who achieved full functional independence (mRS excellent: mRS scores 0–1) of the excessive uWMH burden groups to the low (excessive/low) PS, n (%) 63 (9.5%) 22 (9.8%) and expected (excessive/expected) groups and of the expected group to the low NIHSS, median (IQR) 2 (3) 11 (7) (expected/low) group in the mild acute stroke severity group (mild, NIHSS < 7) and the uWMH, average (SD) 0.0 (1.1) 0.0 (1.1) severe acute stroke severity group (severe, NIHSS ≥ 7). DISCUSSION the subgroup analysis are summarized in Table 5 and visualized In this study, we analyzed the association of uWMH, i.e., in Figure 2B. the amount of WMH burden that is not explained by the We also conducted an auxiliary analysis with all traditional VRFs of a patient, and post-stroke functional patients, including the post-stroke non-survivors for outcome. Our results showed that, in stroke survivors, uWMH the subgroup analysis. The odds ratios and p-values was significantly associated with the 3–6 months of functional between the low, expected, and excessive uWMH burden outcomes. Furthermore, our population analysis showed that groups were not changed substantially for both mild patients with higher uWMH burden were less likely to achieve and severe acute stroke severity patients. The detailed full functional independence (mRS < 2) than patients with results are summarized in Supplementary Tables 4, 5 in lower uWMH burden, while their clinical characteristics were the Supplementary Materials. similar. Our study supports the previous suggestions that WMH Frontiers in Neurology | www.frontiersin.org 6 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes severity implicates more than the effects of demographics the other hand, the low uWMH burden was more protective and traditional VRFs and shows the clear association of the on achieving functional independence for those with severe uWMH burden and complete post-stroke functional recovery, stroke severity. with the uWMH burden emerging as a new independent The uWMH burden was not correlated with any of the other variable to utilize in future analyses of post-stroke functional included clinical characteristics. This was expected since uWMH outcomes (25). represent residuals of the multiple log-linear regression model The consistent results of the trichotomized analysis with with respect to the clinical characteristics. The multiple log- respect to the uWMH burden on the entire population showed linear regression modeling of WMHv and the other clinical that excessive uWMH burden was linked to less probability characteristics and the definition of the uWMH burden as the of achieving full functional independence and the opposite for residual of the model can be interpreted as the adjustment of low uWMH. The mean uWMH burden of patients in the low the effects of the demographics and traditional VRF profile uWMH burden group was negative; thus, patients in the low of a patient to the WMHv. The uWMH burden analysis uWMH group had lower WMH severity than expected from enabled the explicit exploration of the association between their demographics and traditional VRF profiles. This might WMH severity and post-stroke functional outcomes, free of the indicate that those patients who were more resilient to a lifelong, traditional VRF patient profile effect (26). We only included the chronic damage to the brain (as measured by WMH severity) univariate linear regression analysis for the uWMH association were more likely to achieve full functional independence after with the continuous post-stroke functional outcome because stroke than those with similar VRF profiles but unexpectedly the uWMH burden was not correlated with the included excessive uWMH. The consistent finding that patients in the clinical factors. excessive uWMH burden group were less likely to achieve full There are several limitations of our study that are worth functional independence compared to those in the other groups considering for further investigations. The clinical characteristics also supports this interpretation. that were considered in this analysis were not exhaustive of The subgroup analysis based on acute stroke severity all known factors associated with WMH and stroke outcomes. showed that the uWMH burden was linked to full functional WMH accumulation depends on many factors: normal aging independence for patients with both mild and severe acute (27), VRF aggregation, blood–brain barrier disruption (28), stroke severity. Because our patient population had milder acute amyloid and tau accumulation (29), hyperlipidemia (8, 14), stroke severity (median NIHSS = 3) than a typical clinical hyperhomocysteinemia (30), and genetics (31). Because the scenario of mixed mild and severe strokes, separate subgroup uWMH burden captured the effects of the demographics analyses, including mild acute (median NIHSS = 2) and and traditional VRFs in a stroke cohort by modeling, it severe (median NIHSS = 11) strokes, provided additional might have captured the effect of unmeasured risk factors, information on the uWMH association with full functional including the aforementioned VRFs, stroke etiologies, or independence after stroke (mRS 0–1). The results were consistent genetics. The binary representations of the traditional VRFs with the overall population analysis; excessive uWMH burden may have undermined their effects on the WMH burden. in these subgroups resulted in a decreased probability of The additional information on a time-based load of VRFs achieving full functional independence and the opposite for may improve the estimation of the expected WMH burden low uWMH. based on the clinical characteristics of the patients and the In the severe stroke subgroup, patients with low uWMH quantification of the unaccounted WMH burden. The education showed significantly greater odds of achieving full functional and personal lifestyle of a patient that may also be related to independence after stroke than any others. In the mild stroke the WMH accumulation will improve the personalized analysis group, the odds of achieving full functional independence by on the association of the WMH burden and functional stroke patients with excessive uWMH was significantly lower compared outcomes. The other possible risk factors will need to be to other groups. These findings might indicate that a patient investigated further by other studies that specifically focus on with the brain resilient to chronic cerebrovascular injury is non-traditional risk factors for WMH and that may explain more robust to achieve full functional recovery even after the remaining unexplained variance of functional outcomes in severe stroke, while a patient with a vulnerable brain (as this study. reflected by excessive uWMH) is more susceptible to post-stroke The post-stroke treatment information of patients was not disability even after a mild stroke (as shown in Figure 2B). included in the analysis because the information was not The additional finding to highlight with regard to these data is available in our study. The post-stroke treatment information that the probability of achieving full functional independence will enable a stratified analysis on patients who received post-stroke (mRS 0–1) in mild stroke patients with excessive post-stroke treatments and those did not, with a careful uWMH and the probability of those with low uWMH burden assessment on the heterogeneity of post-stroke treatments and and severe stroke were similar at 52.3 and 48.5%, respectively their complications. The cognitive decline of a patient will (Table 5 and Figure 2B). We interpret these data as potential also provide an important insight on the interaction between evidence of the strong effect by uWMH on the functional the WMH burden and functional outcomes that can be outcome in patients with different levels of acute stroke severity. investigated further. The excessive uWMH burden was more predictive on lower Patients in the low uWMH burden group were slightly functional outcomes for those with mild stroke severity and, on younger than the expected and excessive uWMH burden groups. Frontiers in Neurology | www.frontiersin.org 7 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes Patients in the low uWMH burden groups were also less likely CONCLUSION to have DM, AF, CAD, and smoking than the expected uWMH groups. Although the differences were not large, it might have The excessive burden of WMH that was not accounted affected the results. However, patients in the excessive uWMH for by the demographics and traditional VRFs of each burden group also had less proportions of having histories of individual patient is associated with post-stroke functional DM, AF, CAD, and smoking, which might suggest that the outcomes, especially with full functional independence after differences were caused by the relatively small sample sizes of stroke. Our analysis suggests that a lifetime injury to the the low and excessive uWMH burden groups. It is worth to white matter that is not explained by traditional VRF profile note that categorizing patients with respect to WMHv would is an important factor for stroke recovery and a plausible have resulted in larger differences in clinical characteristic indicator of brain health. In the future, an individual risk distribution across the groups and further complicated the profile of an AIS patient and its relationship to WMH interpretation on the association of WMH severity and post- should be considered in a personalized approach to post-stroke stroke outcomes. outcome prediction. Being focused on explaining functional outcomes, our analysis does not expand to the predictive capacity of the uWMH DATA AVAILABILITY STATEMENT burden for post-stroke functional outcomes. Our analysis was also not validated with an independent cohort because of The original contributions presented in the study are included the availability. The strong association of the uWMH burden in the article/Supplementary Material, further inquiries can be with post-stroke functional outcomes suggests that the uWMH directed to the corresponding author/s. burden could be of benefit for future clinical prediction models of functional outcomes. We do not suggest the uWMH burden as the strongest predictor of post-stroke functional outcomes. MRI-GENIE, GISCOME AND As shown in the subgroup analysis regarding the acute stroke INTERNATIONAL STROKE GENETICS severity, the acute NIHSS was a stronger predictor of post- CONSORTIUM stroke functional outcomes. However, the subgroup analysis also showed that the uWMH burden could complement the Sungmin Hong, Massachusetts General Hospital, Boston, MA, prediction of functional outcomes with the admission NIHSS. USA; Anne-Katrin Giese, Massachusetts General Hospital, This will be an interesting research direction to investigate Boston, MA, USA; Markus D. Schirmer, Massachusetts further regarding the predictive capability of the uWMH burden, General Hospital, Boston, MA, USA; Anna K. Bonkhoff, combined with other risk factors including the acute stroke Massachusetts General Hospital, Boston, MA, USA; Martin severity, by using predictive modeling (e.g., logistic regression, Bretzner, Massachusetts General Hospital, Boston, MA, USA; decision forests, deep learning models, etc.). The predictive Pamela Rist, Brigham and Women’s Hospital, Boston, MA, power of the uWMH burden and its association to functional USA; Adrian V. Dalca, Massachusetts Institute of Technology, outcomes will be benefited from cross-validation across multiple Cambridge, MA, USA; Robert W. Regenhardt, Massachusetts datasets and also from validation with a dataset from an General Hospital, Boston, MA, USA; Mark R. Etherton, independent cohort. Massachusetts General Hospital, Boston, MA, USA; Kathleen L. To enable the analysis of WMHv with the largest possible Donahue, Massachusetts General Hospital, Boston, MA, USA; number of image data, we used the automatic segmentation Marco Nardin, Massachusetts General Hospital, Boston, MA, framework to obtain WMHv. Although the WMH segmentation USA; Steven J. T. Mocking, Massachusetts General Hospital, masks were validated quantitatively and qualitatively, we did Boston, MA, USA; Elissa C. McIntosh, Massachusetts General not have access to old and new stroke lesion outlines and Hospital, Boston, MA, USA; John Attia, University of New could hence not comprehensively account for either one Castle, New South Wales, Australia; Oscar R. Benavente, (20). It is therefore possible that stroke infarcts with WMH University of British Columbia, Vancouver, British Columbia, similarity might have led to some inaccuracies in WMH Canada; John W. Cole, University of Maryland, Baltimore, MD, delineation. This effect might be especially pronounced in USA; Amanda Donatti, University of Campinas, Campinas, patients with a high acute stroke severity and larger, or Brazil; Christoph J. Griessenauer, Geisinger, Danville, PA, multiple, old and new stroke infarcts. However, in view of USA; Laura Heitsch, Washington University, St. Louis, a median NIHSS of only three, the median size of stroke MO, USA; Lukas Holmegaard, University of Gothenburg, infarcts is expected to be small, as is their impact on Gothenburg, Sweden; Katarina Jood, University of Gothenburg, WMHv estimation. Gothenburg, Sweden; Jordi-Jimenez-Conde, Universitat Lastly, WMHv and whole-brain volume were quantified Autonoma de Barcelona, Barcelona, Spain; Jaume Roquer, by a scalar abstract measure—volume. Future studies should Universitat Autonoma de Barcelona, Barcelona, Spain; Steven investigate how high-dimensional brain features, such as J. Kittner, University of Maryland, Baltimore, USA; Robin WMH, in relation to different white matter subregions Lemmens, University of Leuven, Leuven, Belgium; Christopher and brain anatomical structures (e.g., white matter, striatal R. Levi, Hunter Medical Research Institute, New Castle, complex, and ventricle), are associated with functional South Wales, Australia; Caitrin W. McDonough, University post-stroke outcomes. of Florida, Gainesville, FL, USA; James F. Meschia, Mayo Frontiers in Neurology | www.frontiersin.org 8 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes Clinic, Jacksonville, FL, USA; Chia-Ling Phuah, Washington FUNDING University, St. Louis, MO, USA; Arndt Rolfs, Centogene AG, Rostock, Germany; Stefan Ropele, Medical University Graz, NIH-NINDS (MRI-GENIE: R01NS086905 - PI NR; Graz, Austria; Jonathan Rosand, Massachusetts General K23NS064052, R01NS082285 - NR; SiGN: U01 NS069208 - JR, Hospital, Boston, USA; Tatjana Rundek, University of SK; R01NS059775, R01NS063925, R01NS082285, P50NS051343, Miami, Miami, USA; Ralph L. Sacco, University of Miami, R01NS086905, U01 NS069208 - OW); NIH NIBIB (NAC Miami, USA; Reinhold Schmidt, Medical University Graz, P41EB015902 PG, U01NS030678 Kissela, Kleindorfer; EB015325 Graz, Austria; Christian Enzinger, Medical University Graz, OW); ISGS (R01NS423733 P.I. JM, Swedish Heart and Lung Graz, Austria; Pankaj Sharma, Royal Holloway University of Foundation, Lund University, Region Skåne, the Freemasons London, Egham, UK; Agnieszka Slowik, Jagiellonian University, Lodge of Instruction Eos Lund, Skåne University Hospital, Krakow, Poland; Alessandro Sousa, University of Campinas, the Foundation of Färs &Frosta (One of Sparbanken Skånes Campinas, SP, Brazil; Tara M. Stanne, University of Gothenburg, ownership Foundations), and the Swedish Stroke Association Gothenburg, Sweden; Daniel Strbian, Helsinki University - AL, Swedish Research Council and the Swedish Heart Central Hospital, Helsinki, Finland; Turgut Tatlisumak, and Lung Foundation, the Swedish State under the ALF University of Gothenburg, Gothenburg, Sweden; Vincent agreement CJ, Spanish Ministry of Science and Innovation, Thijs, University of Melbourne, Heidelberg, Australia; Achala Instituto de Salud Carlos III (Funding for Research in Health Vagal, University of Cincinnati, Cincinnati, OH, USA; Johan PI051737, PI10/02064, PI12/01238 and PI15/00451 JJ-C), Wasselius, Skåne University, Lund, Sweden; Daniel Woo, Fondos FEDER/EDRF Red de Investigación Cardiovascular University of Cincinnati, Cincinnati, OH, USA; Rasmin Zand, (RD12/0042/0020 JJ-C), Fundació la Marató TV3 (76/C/2011 - Geisinger, Danville, PA, USA; Patrick F. McArdle, University JJ-C) and Recercaixa13 (JJ086116 JJ-C), Wistron Corporation of Maryland, Baltimore, MD, USA; Bradford B. Worrall, (PG); The European Unions Horizon 2020 research and University of Virginia, Charlottesville, VA, USA; Ona Wu, innovation programme under the Marie Sklodowska-Curie grant Massachusetts General Hospital, Boston, MA, USA; Christina agreement No 753896 - MS; RL is a senior clinical investigator Jern, University of Gothenburg, Gothenburg, Sweden; Arne of FWO Flanders; MB was supported by the ISITE-ULNE Lindgren, Skåne University, Lund, Sweden; Jane Maguire, Fundation, the Société Française de Neuroradiologie, the University of Technology Sydney, Sydney, Autralia; Liisa Société Française de Radiologie, the Thérèse and René Planiol Tomppo, Helsinky University Central Hospital, Helsinki, Fundation; ME was supported by the American Academy of Finland; Polina Golland, Massachusetts Institute of Technology, Neurology and MGH Executive Council on Research; TT was Cambridge, MA, USA; Natalia S. Rost, Massachusetts General supported by the Helsinki University Central Hospital, Sigrid Hospital, Boston, MA, USA. Juselius Foundation, Sahlgrenska University Hospital, and University of Gothenburg. The funders were not involved in the study design, collection, analysis, interpretation of data, AUTHOR CONTRIBUTIONS the writing of this article of the decision to submit it for SH, A-KG, MS, and NR contributed to conception and publication. design of the study. SH performed statistical analysis and drafted the manuscript. SH, AB, MB, PR, and SUPPLEMENTARY MATERIAL NR analyzed and reviewed the analysis and data. All authors did major roles in the acquisition of data, The Supplementary Material for this article can be found contributed to manuscript revision, read, and approved the online at: https://www.frontiersin.org/articles/10.3389/fneur. submitted version. 2021.700616/full#supplementary-material REFERENCES and race. Stroke. (2009) 40:530–6. doi: 10.1161/STROKEAHA.108.5 21906 1. Wardlaw JM, Chappell FM, Valdés Hernández MC, Makin SDJ, Staals J, 5. Debette S, Markus HS. The clinical importance of white matter Shuler K, et al. White matter hyperintensity reduction and outcomes after hyperintensities on brain magnetic resonance imaging: systematic minor stroke. Neurology. (2017) 89:1003–10. doi: 10.1212/WNL.00000000000 review and meta-analysis. BMJ. (2010) 341:c3666. doi: 10.1136/bmj. 04328 c3666 2. Georgakis MK, Duering M, Wardlaw JM, Dichgans M. WMH and 6. Etherton MR, Wu O, Cougo P, Giese AK, Cloonan L, Fitzpatrick long-term outcomes in ischemic stroke: a systematic review and meta- KM, et al. Integrity of normal-appearing white matter and analysis. Neurology. (2019) 92:e1298–308. doi: 10.1212/WNL.00000000000 functional outcomes after acute ischemic stroke. Neurology. (2017) 07142 88:1701–8. doi: 10.1212/WNL.0000000000003890 3. Liou L, Chen C, Guo Y, Cheng H, Lee H, Hsu J, et al. Cerebral white matter 7. Awad IA, Spetzler RF, Hodak JA, Awad CA, Carey R. Incidental subcortical hyperintensities predict functional stroke outcome. Cerebrovasc Dis. (2010) lesions identified on magnetic resonance imaging in the elderly. i. 29:22–7. doi: 10.1159/000255970 Correlation with age and cerebrovascular risk factors. Stroke. (1986) 17:1084– 4. Kissela B, Lindsell CJ, Kleindorfer D, Alwell K, Moomaw CJ, Woo D, 9. doi: 10.1161/01.STR.17.6.1084 et al. Clinical prediction of functional outcome after ischemic stroke: 8. Jimenez-Conde J, Biffi A, Rahman R, Kanaki A, Butler C, Sonni the surprising importance of periventricular white matter disease S, et al. Hyperlipidemia and reduced white matter hyperintensity Frontiers in Neurology | www.frontiersin.org 9 September 2021 | Volume 12 | Article 700616 Hong et al. eWMH and Post-stroke Functional Outcomes volume in patients with ischemic stroke. Stroke. (2010) 41:437– 22. Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau 42. doi: 10.1161/STROKEAHA.109.563502 D, et al. SciPy 1.0: fundamental algorithms for scientific computing 9. Kearney-Schwartz A, Rossignol P, Bracard S, Felblinger J, Fay R, Boivin in Python. Nat Methods. (2020) 17:261–72. doi: 10.1038/s41592-020- J, et al. Vascular structure and function is correlated to cognitive 0772-5 performance and white matter hyperintensities in older hypertensive 23. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel patients with subjective memory complaints. Stroke. (2009) 40:1229– O, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 36. doi: 10.1161/STROKEAHA.108.532853 (2011) 12:2825–30. doi: 10.5555/1953048.2078195 10. Dufouil C, de Kersaint-Gilly A, Besancon V, Levy C, Auffray E, Brunnereau L, 24. Seabold S, Perktold J. Statsmodels: econometric and statistical modeling et al. Longitudinal study of blood pressure and white matter hyperintensities: with Python. In: 9th Python in Science Conference. Austin, TX the EVA MRI cohort. Neurology. (2001) 56:921–6. doi: 10.1212/WNL.5 (2010). doi: 10.25080/Majora-92bf1922-011 6.7.921 25. Etherton MR, Wu O, Rost NS. Recent advances in leukoaraiosis: white matter 11. de Leeuw FE, de Groot JC, Oudkerk M, Kors JA, Hofman A, van structural integrity and functional outcomes after acute ischemic stroke. Curr Gijn J and Breteler MM. Atrial fibrillation and the risk of cerebral Cardiol Rep. (2016) 18:123. doi: 10.1007/s11886-016-0803-0 white matter lesions. Neurology. (2000) 54:1795–801. doi: 10.1212/WNL.54. 26. Belsley DA, Kuh E, Welsch RE. Regression Diagnostics: Identifying Influential 9.1795 Data and Sources of Collinearity. 2nd ed. New York, NY: Wiley (2004). 12. Garde E, Mortensen EL, Krabbe K, Rostrup E, Larsson HBW. Relation 27. Gunning-Dixon FM, Brickman AM, Cheng JC, Alexopoulos GS. Aging of between age-related decline in intelligence and cerebral white-matter cerebral white matter: a review of MRI findings. Int J Geriatr Psychiatry. (2009) hyperintensities in healthy octogenarians: a longitudinal study. Lancet. (2000) 24:109–17. doi: 10.1002/gps.2087 356:628–34. doi: 10.1016/S0140-6736(00)02604-0 28. Wardlaw JM, Smith C, Dichgans M. Mechanisms of sporadic cerebral small 13. Aribisala BS, Morris Z, Eadie E, Thomas A, Gow A, Valdés vessel disease: insights from neuroimaging. Lancet Neurol. (2013) 12:483– Hernández MC, et al. Blood pressure, internal carotid artery 97. doi: 10.1016/S1474-4422(13)70060-7 flow parameters, and age-related white matter hyperintensities. 29. Jack Jr. CR, Wiste HJ, Weigand SD, Therneau TM, Knopman DS, Lowe Hypertension. (2014) 63:1011–8. doi: 10.1161/HYPERTENSIONAHA.113. V, et al. Age-specific and sex-specific prevalence of cerebral β-amyloidosis, 02735 tauopathy, and neurodegeneration in cognitively unimpaired individuals 14. Wardlaw JM, Allerhand M, Doubal FN, Valdés Hernández MC, aged 50–95 years: a cross-sectional study. Lancet Neurol. (2017) 16:435– Morris Z, Gow AJ, et al. Vascular risk factors, large-artery 44. doi: 10.1016/S1474-4422(17)30077-7 atheroma, and brain white matter hyperintensities. Neurology. (2014) 30. Dufouil C, Alperovitch A, Ducros V, Tzouriol C. Homocysteine, white matter 82:1331–8. doi: 10.1212/WNL.0000000000000312 hyperintensities, and cognition in health elderly people. Ann Neurol. (2003) 15. Soriano-Tárraga C, Giralt-Steinhauer E, Mola-Caminal M, Vivanco- 53:214–21. doi: 10.1002/ana.10440 Hidalgo RM, Ois A, Rodríguez-Campello A, et al. Ischemic stroke 31. Sachdev PS, Thalamuthu A, Mather KA, Ames D, Wright MJ, Wen W, et al. patients are biologically older than their chronological age. Aging. (2016) White Matter Hyperintensities are under strong genetic influence. Stroke. 8:2655. doi: 10.18632/aging.101028 (2016) 47:1422–8. doi: 10.1161/STROKEAHA.116.012532 16. Giese AK, Schirmer MD, Donahue KL, Cloonan L, Irie R, Winzeck Conflict of Interest: AR was employed by Centogene AG, Germany. S, et al. Design and rationale for examining neuroimaging genetics in ischemic stroke: the MRI-genie study. Neurol Genet. (2017) The remaining authors declare that the research was conducted in the absence of 3:e180. doi: 10.1212/NXG.0000000000000180 any commercial or financial relationships that could be construed as a potential 17. Wu O, Winzeck S, Giese AK, Hancock BL, Etherton MR, Bouts conflict of interest. MJRJ, et al. Big data approaches to phenotyping acute ischemic stroke using automated lesion segmentation of multi-center magnetic resonance imaging data. Stroke. (2019) 50:1734–41. doi: 10.1161/STROKEAHA.119.0 Publisher’s Note: All claims expressed in this article are solely those of the authors 25373 and do not necessarily represent those of their affiliated organizations, or those of 18. Meschia JF, Arnett DK, Ay H, Brown RD Jr, Benavente OR, Cole JW, et al. the publisher, the editors and the reviewers. Any product that may be evaluated in Stroke genetics network (sign) study: design and rationale for a genome- this article, or claim that may be made by its manufacturer, is not guaranteed or wide association study of ischemic stroke subtypes. Stroke. (2013) 44:2694– endorsed by the publisher. 702. doi: 10.1161/STROKEAHA.113.001857 19. Pulit SL, McArdle PF, Wong Q, Malik R, Gwinn K, Achterberg S, Copyright © 2021 Hong, Giese, Schirmer, Bonkhoff, Bretzner, Rist, Dalca, et al. Ninds stroke genetics network (SiGN); International stroke genetics Regenhardt, Etherton, Donahue, Nardin, Mocking, McIntosh, Attia, Benavente, consortium (ISGC): loci associated with ischaemic stroke and its subtypes Cole, Donatti, Griessenauer, Heitsch, Holmegaard, Jood, Jimenez-Conde, Roquer, (SiGN): a genome-wide association study. Lancet Neurol. (2016) 15:174– Kittner, Lemmens, Levi, McDonough, Meschia, Phuah, Rolfs, Ropele, Rosand, 84. doi: 10.1016/S1474-4422(15)00338-5 Rundek, Sacco, Schmidt, Enzinger, Sharma, Slowik, Sousa, Stanne, Strbian, 20. Schirmer MD, Dalca AV, Sridharan R, Giese AK, Donahue KL, Tatlisumak, Thijs, Vagal, Wasselius, Woo, Zand, McArdle, Worrall, Wu, Jern, Nardin MJ, et al. White matter hyperintensity quantification in Lindgren, Maguire, Tomppo, Golland, Rost and the MRI-GENIE and GISCOME large-scale clinical acute ischemic stroke cohorts–the MRI-genie Investigators and the International Stroke Genetics Consortium. This is an open- study. NeuroimageClin. (2019) 23:101884. doi: 10.1016/j.nicl.2019.1 access article distributed under the terms of the Creative Commons Attribution 01884 License (CC BY). The use, distribution or reproduction in other forums is permitted, 21. Schirmer MD, Donahue KL, Nardin MJ, Dalca AV, Giese AK, Etherton provided the original author(s) and the copyright owner(s) are credited and that the MR, et al. Brain volume: an important determinant of functional original publication in this journal is cited, in accordance with accepted academic outcome after acute ischemic stroke. Mayo Clin Proc. (2020) 95:955– practice. No use, distribution or reproduction is permitted which does not comply 65. doi: 10.1016/j.mayocp.2020.01.027 with these terms. Frontiers in Neurology | www.frontiersin.org 10 September 2021 | Volume 12 | Article 700616