We aimed to develop a novel prediction model for early neurological deterioration (END) based on an interpretable machine learning (ML) algorithm for atrial fibrillation (AF)-related stroke and to evaluate the prediction accuracy and feature importance of ML models. Data from multicenter prospective stroke registries in South Korea were collected. After stepwise data preprocessing, we utilized logistic regression, support vector machine, extreme gradient boosting, light gradient boosting machine (LightGBM), and multilayer perceptron models. We used the Shapley additive explanation (SHAP) method to evaluate feature importance. Of the 3,213 stroke patients, the 2,363 who had arrived at the hospital within 24 h of symptom onset and had available information regarding END were included. Of these, 318 (13.5%) had END. The LightGBM model showed the highest area under the receiver operating characteristic curve (0.772; 95% confidence interval, 0.715–0.829). The feature importance analysis revealed that fasting glucose level and the National Institute of Health Stroke Scale score were the most influential factors. Among ML algorithms, the LightGBM model was particularly useful for predicting END, as it revealed new and diverse predictors. Additionally, the effects of the features on the predictive power of the model were individualized using the SHAP method.

Background: Statins lower the risk of recurrent stroke and mortality and improve outcomes in patients with ischemic stroke. However, the effects of statins on atrial fibrillation (AF)-related stroke are not well-established. Our study aims to investigate the effects of statin therapy on the clinical outcomes in patients with AF-related stroke. Methods: Electronic databases (MEDLINE, Embase, and Scopus) were searched for previous studies on the effects of pre- and post-stroke statins on the clinical outcomes in AF-related stroke patients. The primary outcome was all-cause mortality. Secondary outcomes included recurrent ischemic stroke, acute coronary events, major adverse cardiovascular events (MACE), and short-term functional outcomes. We extracted hazard ratios (HRs) or odds ratios (ORs) with 95% confidence interval (CI) from each study and pooled them through a meta-analysis. Results: A total of eight studies (five on post-stroke statins and three on pre-stroke statins) with 12,216 patients were included in the analysis. Post-stroke statin therapy reduced the risk of all-cause mortality (HR, 0.63; 95% CI, 0.55–0.74). This beneficial effect was sustained regardless of statin intensity. However, no significant associations were observed between statin therapy and a reduction in the risk of recurrent ischemic stroke, acute coronary events, or MACE. Pre-stroke statin use was associated with a lower risk of poor short-term functional outcomes (OR, 0.63; 95% CI, 0.47–0.85). Conclusions: Statin therapy for AF-related stroke may reduce all-cause mortality and improve functional outcomes. Randomized controlled studies are warranted to confirm the effects of statins on the outcomes of AF-related stroke.

Purpose: Several studies have examined the clinical impact of poststroke and stroke-related sarcopenia on stroke recovery. However, few studies have investigated the effect of sarcopenia detected shortly after stroke on functional prognosis. We predicted functional outcomes using early screening for sarcopenia in patients with acute ischemic stroke. We also examined the effect of sarcopenia detected shortly after stroke on functional prognosis. Patients and Methods: Patients diagnosed with acute ischemic stroke within 2 days of symptom onset were consecutively enrolled at a tertiary university hospital. Appendicular skeletal muscle mass (ASM) was measured using dual-energy X-ray absorptiometry during early hospitalization. Sarcopenia was diagnosed based on low ASM and strength criteria of the Asian Working Group for Sarcopenia (AWGS) and European Working Group on Sarcopenia in Older People (EWGSOP2). The primary outcome was poor functional outcome, defined as a modified Rankin score of 4– 6 and all-cause mortality at 3 months. Results: Of the 653 patients, 214 (32.8%) and 174 (26.6%) had sarcopenia according to the AWGS and EWGSOP2 criteria, respectively. Irrespective of the definition, the sarcopenia group had a significantly higher proportion of patients with poor functional outcomes and all-cause mortality. Multivariate logistic regression analysis revealed that height-adjusted ASM was independently associated with poor functional outcomes (odds ratio: 0.61; 95% confidence interval: 0.40– 0.91; P < 0.005), and they were negatively correlated. However, the association between 3-month mortality, skeletal muscle mass, and sarcopenia was not sustained in multivariate analyses. Conclusion: Height-adjusted ASM associated with sarcopenia is a potential predictor of poor functional outcomes at 3 months in patients with acute stroke. However, owing to the limitations of this study, further research is required to confirm these findings.

Importance: Cardiovascular benefits of mild to moderate alcohol consumption need to be validated in the context of behavioral changes. The benefits of reduced alcohol consumption among people who drink heavily across different subtypes of cardiovascular disease (CVD) are unclear.Objective: To investigate the association between reduced alcohol consumption and risk of major adverse cardiovascular events (MACEs) in individuals who drink heavily across different CVD subtypes.Design, Setting, and Participants: This cohort study analyzed data from the Korean National Health Insurance Service–Health Screening database and self-reported questionnaires. The nationally representative cohort comprised Korean citizens aged 40 to 79 years who had national health insurance coverage on December 31, 2002, and were included in the 2002 to 2003 National Health Screening Program. People who drank heavily who underwent serial health examinations over 2 consecutive periods (first period: 2005-2008; second period: 2009-2012) were included and analyzed between February and May 2023. Heavy drinking was defined as more than 4 drinks (56 g) per day or more than 14 drinks (196 g) per week for males and more than 3 drinks (42 g) per day or more than 7 drinks (98 g) per week for females.Exposures: Habitual change in heavy alcohol consumption during the second health examination period. People who drank heavily at baseline were categorized into 2 groups according to changes in alcohol consumption during the second health examination period as sustained heavy drinking or reduced drinking.Main Outcomes and Measures: The primary outcome was the occurrence of MACEs, a composite of nonfatal myocardial infarction or angina undergoing revascularization, any stroke accompanied by hospitalization, and all-cause death.Results: Of the 21 011 participants with heavy alcohol consumption at baseline (18 963 males [90.3%]; mean [SD] age, 56.08 [6.16] years) included in the study, 14 220 (67.7%) sustained heavy drinking, whereas 6791 (32.2%) shifted to mild to moderate drinking. During the follow-up of 162 378 person-years, the sustained heavy drinking group experienced a significantly higher incidence of MACEs than the reduced drinking group (817 vs 675 per 100 000 person-years; log-rank P = .003). Reduced alcohol consumption was associated with a 23% lower risk of MACEs compared with sustained heavy drinking (propensity score matching hazard ratio [PSM HR], 0.77; 95% CI, 0.67-0.88). These benefits were mostly accounted for by a significant reduction in the incidence of angina (PSM HR, 0.70; 95% CI, 0.51-0.97) and ischemic stroke (PSM HR, 0.66; 95% CI, 0.51-0.86). The preventive attributes of reduced alcohol intake were consistently observed across various subgroups of participants.Conclusions and Relevance: Results of this cohort study suggest that reducing alcohol consumption is associated with a decreased risk of future CVD, with the most pronounced benefits expected for angina and ischemic stroke.