Keun-Hwa Jung

BACKGROUND: Although preclinical studies and limited clinical evidence suggest potential benefits of sodium-glucose cotransporter 2 inhibitors (SGLT2i) after acute ischemic stroke, their effects in patients with moderate-to-severe acute ischemic stroke requiring endovascular thrombectomy remain unclear. This study aimed to evaluate whether SGLT2i use after endovascular thrombectomy is associated with improved 3-month functional outcomes in diabetic patients with acute ischemic stroke. METHODS: This retrospective cohort study analyzed prospectively collected data from January 2020 to March 2025 across 4 comprehensive stroke centers in Korea, Italy, and Switzerland. Consecutive adult patients with diabetes were grouped according to whether SGLT2i was used after their endovascular thrombectomy. Propensity score–based methods—including overlap weighting, inverse probability of treatment weighting, and 1:1 propensity score matching—were applied. Outcomes included the shift in 3-month modified Rankin Scale score, favorable (modified Rankin Scale score 0–2), and excellent outcomes (modified Rankin Scale score 0–1) at 3 months. Ordinal and binary logistic regression analyses were performed. RESULTS: Of 429 eligible patients (mean [SD] age, 73.5 [11.1] years; 249 [58.0%] men; median [interquartile range] initial National Institutes of Health Stroke Scale score, 12 [6–18]), 51 (11.9%) received SGLT2i poststroke. Successful recanalization was achieved in 85.3%. In the overlap weighting model (weighted n=37 per group), SGLT2i use was associated with improved 3-month modified Rankin Scale score distribution (common odds ratio, 2.31; [95% CI, 1.30–4.10]) and higher odds of a favorable outcome (odds ratio, 2.26; [95% CI, 1.15–4.43]). Results were consistent across inverse probability of treatment weighting and propensity score matching models, with inverse probability of treatment weighting analyses additionally demonstrating significant associations with excellent outcomes. No significant effect modification was observed across key clinical subgroups. CONCLUSIONS: Poststroke SGLT2i therapy after endovascular thrombectomy was associated with improved 3-month functional outcomes in this multicenter retrospective analysis of diabetic patients. These findings suggest a potential therapeutic role for SGLT2i in poststroke diabetes management and support future prospective trials.

BACKGROUND: Obesity increases the risk of stroke and cognitive decline, but its association with intracranial arterial hemodynamics remains unclear. We examined whether obesity score is associated with intracranial signal-intensity gradient, a time-of-flight magnetic resonance angiography–derived surrogate of arterial wall shear conditions, and whether this association differs by vascular territory. METHODS: This cross-sectional study included 5618 adults who underwent comprehensive health screening, including intracranial time-of-flight magnetic resonance angiography, in the Republic of Korea between 2012 and 2020. Body mass index was calculated from measured height and weight, and waist circumference was measured by health-screening staff on the same day as brain magnetic resonance angiography. An obesity score was calculated by assigning 1 point each for general and abdominal obesity, yielding scores of 0 (neither), 1 (either), or 2 (both). Arterial SIG and arterial radius were calculated using semiautomated 3-dimensional vessel segmentation. Lower SIG values were interpreted as reflecting reduced arterial wall shear conditions. Multivariable linear regression models assessed the association between obesity score and SIG across 11 intracranial arteries, adjusting for demographics, arterial radius, and cardiometabolic factors. RESULTS: Among 5618 participants (mean [SD] age, 55.9 [8.6] years; 45.2% women), higher obesity score was associated with lower SIG across all intracranial arteries ( P value for trend <0.001). In adjusted models, each 1-point increase in obesity score was associated with reduced SIG, including in the basilar artery (B=−0.26 [95% CI, −0.31 to −0.21]). The obesity-related reduction in SIG was greater in the posterior than anterior circulation and persisted after adjustment for arterial radius and cardiometabolic factors. CONCLUSIONS: A higher obesity score was associated with reduced intracranial arterial SIG, consistent with adverse time-of-flight magnetic resonance angiography–derived flow-related hemodynamic alterations. Greater reductions in the posterior circulation suggest territory-specific vulnerability, which may be associated with early cerebrovascular compromise in the context of obesity.

Background and Purpose Recurrent ischemic stroke is a sentinel audit point for secondary prevention delivery. We quantified attainment of prevention indicators, sex- and age-related heterogeneity, and projected prevention gains.Methods We analyzed 24,183 recurrent ischemic stroke admissions from 81 Korean Stroke Registry hospitals (2018–2023). Annual attainment of antithrombotic therapy, low-density lipoprotein cholesterol (LDL-C) <70 mg/dL, glycated hemoglobin (HbA1c) <7.0% in diabetes, non-current smoking, and normal weight (body mass index [BMI] 18.5–24.9 kg/m²) was age-standardized descriptively; temporal trends were assessed with hospital-clustered generalized estimating equation models. Projections combined observed implementation gaps with trial-derived relative risk reductions.Results Antithrombotic use increased from 74.6% to 79.4% (odds ratio [OR]/year 1.05, 95% confidence interval 1.01–1.09), but plateaued after 2019, leaving 20.6% untreated in 2023. Among 4,441 patients with known atrial fibrillation (AF), oral anticoagulant use rose from 65.3% to 73.3% (OR/year 1.08; P=0.005), and 26.7% remained untreated. LDL-C <70 mg/dL improved from 32.6% to 41.9% (OR/year 1.08; P<0.001), whereas HbA1c, smoking, and weight indicators showed limited overall progress. Heterogeneity emerged: non-current smoking improved in men but not women (already near ceiling, approximately 95%); LDL-C control did not improve in adults aged 19–49 years; and BMI ≥25 kg/m² rose from 36.2% to 56.2% in this age group. Upper-bound projections estimated approximately 183, 126, and 201 potentially preventable recurrent strokes per year within the registry (not national estimates) from antithrombotic non-use, anticoagulant non-use in AF, and LDL-C gaps in large-artery atherosclerosis, respectively.Conclusions Recurrent-stroke admission revealed persistent, unevenly distributed prevention gaps. Medication reconciliation, AF detection and anticoagulation, lipid intensification, and tailored behavioral pathways should be prioritized.

Background: Current stroke guidelines lack subtype‐specific recommendations for optimal low‐density lipoprotein cholesterol targets. Indeed, most lipid‐lowering trials have not differentiated between stroke subtypes or focused primarily on large artery atherosclerosis, leaving a gap in evidence for other stroke subtypes, such as small vessel occlusion stroke. This study is designed to evaluate whether intensive lipid‐lowering therapy is superior to standard therapy in reducing major adverse cardiovascular events among patients with small vessel occlusion stroke. Methods: SVO70 (Optimal Target Low‐Density Lipoprotein Cholesterol Level for Small Vessel Occlusion Stroke) is a multicenter, prospective, randomized, open, blinded‐end point clinical trial in which adult patients with neuroimaging‐confirmed small vessel occlusion stroke within 180 days of randomization will be enrolled. Key exclusion criteria include predefined low‐density lipoprotein cholesterol targets for other comorbidities, contraindications to statin, and pregnancy or lactation. Eligible participants will be randomized 1:1 to achieve low‐density lipoprotein cholesterol <70 mg/dL (intensive group) or 90 to 110 mg/dL (standard group). The trial aims to enroll 4016 participants over a 2‐year recruitment period, with a minimum follow‐up of 4 years. The primary end point is the occurrence of major adverse cardiovascular events, defined as cardiovascular death, stroke, and acute coronary syndrome, during the follow‐up period. Secondary end points include (1) any stroke, (2) ischemic stroke or transient ischemic attack, (3) hemorrhagic stroke, (4) cardiovascular death, (5) myocardial infarction, (6) all‐cause death, and (7) acute coronary syndrome. Conclusions: This study will provide valuable information for determining optimal low‐density lipoprotein cholesterol targets for patients with small vessel occlusion stroke. Registration: URL: https://ClinicalTrials.gov ; Unique Identifier: NCT06649240, Clinical Research Information Service (KCT0009043).

Older patients aged ≥75 years are underrepresented in major statin trials, leaving the optimal statin intensity after acute ischemic stroke (AIS) undefined. We aimed to compare functional outcomes and short-term safety between high-intensity statin therapy (HIST) and moderate-intensity statin therapy (MIST) in patients aged ≥75 years with AIS or transient ischemic attack. Using a prospective stroke registry at a single tertiary center (2019-2022), we retrospectively analyzed 337 patients aged ≥75 years with AIS or transient ischemic attack who maintained statin therapy for 3 months (HIST n=117; MIST n=220). The primary outcome was a favorable 3-month functional outcome (modified Rankin Scale score 0-2). Secondary outcomes included stroke recurrence, adverse effects, and statin discontinuation. Multivariable logistic regression with pre-specified sensitivity analyses was performed. Favorable outcomes at 3 months were more frequent with HIST (70.9% vs. 55.9%; p=0.010). After multivariable adjustment, HIST was independently associated with a favorable outcome (adjusted odds ratio [aOR] 2.03, 95% confidence interval [CI] 1.17-3.53), consistent across sensitivity analyses: per-protocol (aOR 3.48, 95% CI 1.97-6.17) and atrial fibrillation-adjusted (aOR 2.21, 95% CI 1.26-3.89). No significant differences were observed in statin discontinuation, stroke recurrence, or adverse effects. In older patients with AIS, HIST was independently associated with better functional outcomes without evidence of increased harm, broadly consistent with current guideline recommendations for HIST when tolerated. Prospective studies are needed to confirm a causal relationship.

Previous studies have reported mixed findings regarding sex differences in cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), with some indicating earlier ischemic events in men, whereas others found no significant difference in stroke onset. However, evidence from Asian populations remains limited, where clinical manifestations and NOTCH3 variant distribution differ substantially from those in European cohorts. We investigated sex-specific variations in clinical phenotypes and brain MRI characteristics among Korean patients with CADASIL. In this nationwide retrospective study, 577 patients were screened for CADASIL across 26 centers in South Korea. After excluding those with benign or uncertain variants or insufficient clinical or MRI data, 368 individuals with pathogenic NOTCH3 variants were included in the final analysis. Clinical features, vascular risk factors, and NOTCH3 variant positions classified by epidermal growth factor–like repeat (EGFr) domains were compared between sexes. Brain MRI findings—including white matter hyperintensities (WMHs), lacunes, cerebral microbleeds, and atrophy—were evaluated. Sex differences in the onset of ischemic stroke and dementia were analyzed using Kaplan-Meier and Cox proportional hazards models. The mean age of the patients at data collection was 61.7 ± 11.9 years, and 48.6% were male. Thirty-six distinct pathogenic variants were identified. The most frequent variant was R544C (40.8%) located at the EGFr 13/14 interdomain boundary (corresponding to exon 11). Other recurrent variants were R75P (17.9%), C542R (5.4%), and R587C (4.6%). Men exhibited a higher prevalence of ischemic stroke (70.9% vs 44.4%, p < 0.001) and experienced their first event at an earlier age (median 53 vs 59 years, log-rank p < 0.001). On brain MRI, male sex was independently associated with a greater lacune burden (odds ratio 3.03, 95% CI 1.70–5.38, p < 0.001) while female sex tended to show higher WMH volume. This nationwide study highlights distinct sex differences in CADASIL presentation within a Korean population. Male patients exhibited a higher prevalence and earlier onset of ischemic stroke not fully explained by conventional vascular risk factors. These findings underscore the importance of integrating sex, hormonal, and genetic-epigenetic factors into future research to guide precision phenotyping and sex-informed clinical management in CADASIL.

Purpose: Despite substantial advances in computational cardiovascular modeling, simplifying assumptions often overlook key compensatory mechanisms. To better capture physiological responses under pathological conditions, we incorporate two essential compensatory mechanisms—collateral circulation and arteriolar vasodilation—into a multiscale blood perfusion simulation framework. Methods: Collateral vessels are stochastically generated to supply ischemic regions, with model parameters systematically varied to control vessel density, caliber, and spatial distribution. Arteriolar vasodilation is modeled to represent the combined response of vascular smooth muscle cells to local metabolic and hemodynamic stimuli. To assess their impact on perfusion, artificial stenoses are introduced into both an idealized geometry and a subject-specific cerebrovascular model. Results: Simulation results across varying collateral configurations and vasodilation capacities demonstrate that perfusion enhancement from sub-resolution collateral vessels alone is limited and highly dependent on vessel caliber and density. In contrast, combining collateral flow with vasodilation produces a more pronounced improvement in tissue perfusion. These findings suggest that arteriolar vasodilation serves as the primary mechanism for ischemic compensation, with collateral circulation providing secondary support which is consistent with clinical observations. Conclusion: The proposed methods enable multiscale blood flow simulations while accounting for the main autoregulatory mechanisms. The models have been verified using literature data and are shown to provide accurate results.

Introduction: This multicenter, double-blind, placebo-controlled trial, commissioned by South Korea’s Ministry of Food and Drug Safety, evaluated the effect of oxiracetam for preventing post-stroke cognitive impairment (PSCI) and explored potential interaction with physical activity using neuroimaging. Patients and methods: Patients at high risk of PSCI, reporting subjective cognitive decline ⩾3 months after stroke, were randomized 1:1 to receive oxiracetam or placebo for 36 weeks. Physical activity was tracked via wrist-worn actigraphy. Coprimary endpoints were changes in Mini-Mental State Examination (MMSE) and Clinical Dementia Rating–Sum of Boxes (CDR-SB). Secondary outcomes included neuropsychological assessments and resting-state functional magnetic resonance imaging network metrics. Results: Of 500 enrolled participants (mean age 68.9 years; median 32 months post-stroke), 457 completed the study. There were no statistically significant differences between groups in changes in MMSE (oxiracetam: +0.13 ± 2.27 vs placebo: +0.27 ± 2.09; p = 0.49) or CDR-SB scores (–0.14 ± 0.70 vs −0.08 ± 0.80; p = 0.38). No evidence of interaction was observed between oxiracetam and physical activity. Exploratory analyses suggested favorable trends in functional segregation and CDR-SB scores among highly active oxiracetam participants. Discussion and conclusion: Oxiracetam did not demonstrate benefit in preventing PSCI in high-risk patients. These findings support the recent regulatory decision to suspend its use in South Korea.

Determining blood flow in major cerebral arteries is crucial for understanding cerebrovascular diseases. This study focused on comparing flow volumes derived from the signal intensity gradient (SIG) of Time-of-flight magnetic resonance angiography (TOF-MRA), referred to as SIG volume, and two-dimensional phase-contrast MR (PC-MR) and assessing their correlation in both experimental and human studies. The SIG volume of TOF-MRA could be calculated based on the concept of SIG as velocity gradient correspondingly. This research encompassed an experimental study as well as a prospective cross-sectional study involving human participants. Flow volumes were measured in four varying-sized tubes using PC-MR and TOF-MRA in the experimental study. The internal carotid and vertebral arteries were examined in relatively healthy subjects and the data were prospectively collected. Pearson’s correlation coefficients and coefficients of determination (R²) were calculated, and Bland-Altman plots were used to assess the agreement between the two methods. In the tubal experiments (n = 80), the correlation coefficient of flow volumes measured by two modalities was 0.98, with an R² of 0.96 (p < 0.001). In the human study, 50 internal carotid and 49 vertebral arteries from 25 subjects (mean age ± standard deviation: 63.3 ± 11.5 y) were examined. The coefficient for the cerebral arteries was 0.93, with an R² of 0.87 (p < 0.001). Calculated SIG volumes of TOF-MRA were highly correlated with the blood flow volume from PC-MR in both tubal experiments and human studies on extracranial cerebral arteries. This study was registered at ClinicalTrials.gov (NCT04585971, Date of Registration: 2020-10-14).