Objectives:To develop a machine learning prediction model for poor outcome of acute minor ischemic stroke (AMIS) at 90 days after onset and to explain the importance of various risk factors.Methods:Patients with AMIS admitted to the Second People's Hospital of Hefei from June 2022 to December 2023 were included retrospectively. AMIS was defined as the National Institutes of Health Stroke Scale (NIHSS) score ≤5 at admission. According to the modified Rankin Scale score at 90 days after onset, the patients were divided into a good outcome group (<2) and a poor outcome group (≥2). Recursive feature elimination (RFE) method was used to screen characteristic variables of poor outcome. Based on logistic regression (LR), supported vector machine (SVM), and extreme Gradient Boosting (XGBoost) machine learning algorithms, prediction models for poor outcome of AMIS were developed, and the predictive performance of the models was compared by the area under the curve (AUC) of receiver operating characteristic (ROC) curve and the calibration curve. Shapley Additive exPlanations (SHAP) algorithm was used to explain the role of characteristic variables in the optimal prediction model. Results:A total of 225 patients with AMIS were included, of which 152 (67.56%) had good outcome and 73 (32.44%) had poor outcome. Multivariate analysis showed that baseline NIHSS score, baseline systolic blood pressure, hypertension, diabetes, low-density lipoprotein cholesterol, homocysteine, body mass index, D-dimer, and age were the characteristic variables associated with poor outcome in patients with AMIS. The ROC curve analysis shows that the LR model had the best predictive performance (AUC=0.888, 95% confidence interval [ CI] 0.807-0.970), the next was the XGBoost model (AUC=0.888, 95% CI 0.796-0.980), while the SVM model had the lowest performance (AUC=0.849, 95% CI 0.754-0.944). The calibration curve showed that the LR model performed the best in terms of calibration accuracy. SHAP showed that baseline systolic blood pressure, baseline NIHSS score, diabetes, hypertension and body mass index were the top five risk factors for poor outcome of patients with AMIS. Conclusions:The LR algorithm has stable and superior performance in predicting poor outcome of patients with AMIS. Baseline systolic blood pressure, baseline NIHSS score, diabetes, hypertension and body mass index are the important risk factors for poor outcome of patients with AMIS.

Objective:To investigate the predictive value of monocyte-to-high-density lipoprotein cholesterol ratio (MHR) for early neurological deterioration (END) and hemorrhagic transformation (HT) after intravenous thrombolysis in patients with acute ischemic stroke (AIS).Methods:Patients with AIS received IVT in Hefei Second People's Hospital from May 2020 to January 2022 were retrospectively enrolled. Blood collection was completed and MHR was calculated before intravenous thrombolysis. END was defined as an increase of ≥2 from the baseline in the National Institutes of Health Stroke Scale (NIHSS) score or ≥1 from the baseline in motor function score at any time within 7 d after admission. HT was defined as intracranial hemorrhage newly found by CT/MRI within 24 h after intravenous thrombolysis. Multivariate logistic regression analysis was used to determine the independent predictors of END and HT, and the receiver operating characteristic (ROC) curve was used to analyze the predictive value of MHR for END and HT. Results:A total of 186 patients with AIS treated with IVT were included, of which 50 (26.9%) had END and 31 (16.7%) had HT. The median MHR was 0.43. The MHR in the END group was significantly higher than that in the non-END group (0.49 vs. 0.40; P=0.008), and the MHR in the HT group was significantly higher than that in the non-HT group (0.52 vs. 0.40; P=0.013). All patients were divided into 4 groups (MHR1, MHR2, MHR3 and MHR4) according to the MHR quartile from low to high. Multivariate logistic regression analysis showed that after adjusting for confounding factors, taking MHR1 as a reference, MHR3 (odds ratio [ OR] 6.317, 95% confidence interval [ CI] 1.465-27.237; P=0.013) and MHR4 ( OR 8.064, 95% CI 1.910-34.051; P=0.005) were the significant independent predictors of END; Taking MHR1 as a reference, MHR4 ( OR 5.147, 95% CI 1.194-22.182; P=0.028) was the significant independent predictor of HT. The ROC curve analysis showed that the area under the curve of MHR for predicting END was 0.628 (95% CI 0.554-0.698; P=0.008). When the optimal MHR cutoff value was 0.41, its sensitivity and specificity for predicting END was 74.0% and 53.7% respectively. The area under the curve of MHR for predicting HT was 0.642 (95% CI 0.569-0.711; P=0.013). When the best cutoff value was 0.44, the sensitivity and specificity of MHR for predicting HT were 77.4% and 58.1% respectively. Conclusion:Higher MHR is a risk factor for END and HT after intravenous thrombolysis in patients with AIS, but the predictive value of MHR for END and HT is limited.

Objective:To develop a short-term prognostic model of spontaneous cerebral hemorrhage based on eXtreme Gradient Boosting (XGBoost) machine learning, and to compare its predictive performance with a Logistic regression model.Methods:Patients with sICH admitted to Department of Neurology, Second People's Hospital of Hefei from January 2018 to March 2022 were chosen; their general demographic characteristics, medical history, laboratory indices and cranial imaging data were retrospectively collected. The prognoses of patients 90 d after discharge were evaluated according to modified Rankin Scale (mRS) scores (good prognosis: mRS scores<3; poor prognosis: mRS scores≥3). XGboost and multiple Logistic regression models were used to screen out the factors for prognoses of patients 90 d after discharge, and area under receiver operating characteristic (ROC) curves, sensitivity, specificity and prediction accuracy of the 2 models were analyzed and compared.Results:A total of 413 patients with sICH were included; 180 patients(43.6 %) had poor prognosis and 233 (56.4%) had good prognosis 90 d after discharge. Multivariate Logistic regression results showed that age≥65 years, hemorrhage into the ventricle, hematoma volume of 20-40 mL, hematoma volume>40 mL and National Institutes of Health Stroke Scale (NIHSS) scores were independent influencing factors for short-term prognoses of sICH ( P<0.05). The variables in the XGBoost model were ranked in order of importance: NIHSS scores, systolic blood pressure at admission, Glasgow coma scale (GCS) scores, age≥65 years, hemorrhage into the ventricle, hematoma volume of 20-40 mL, and hematoma volume>40 mL. AUC of XGBoost model in predicting the prognosis was 0.895 (95% CI: 0.820-0.947), enjoying sensitivity of 68.89%, specificity of 94.83%, and prediction accuracy of 83.5%. AUC of Logistic regression model in predicting the prognosis was 0.894 (95% CI: 0.818-0.946), enjoying sensitivity of 93.33%, specificity of 70.69%, and prediction accuracy of 80.58%. Conclusion:The short-term prognostic model based on XGboost for sICH patients has high predictive efficacy, whose predictive accuracy is better than traditional Logistic regression.

Ischemic stroke is the main cause of death and disability in adults, and its incidence is increasing year by year. Obstructive sleep apnea (OSA) is the most common type of sleep-disordered breathing, which can increase the risk of ischemic stroke and affect the outcomes of patients. There is an increasing amount of research on the relationship between OSA and ischemic stroke. This article reviews the bidirectional relationship between OSA and ischemic stroke, the mechanism of OSA causing ischemic stroke, and the diagnosis and treatment of OSA in patients with ischemic stroke.

Carotid atherosclerosis is closely associated with ischemic stroke. Research shows that the rupture of vulnerable carotid plaque is an important reason for carotid atherosclerosis leading to thromboembolic events. Therefore, early identification of vulnerable carotid plaques is of great significance for the diagnosis, treatment, and prevention of ischemic stroke. This article reviews the pathophysiological features, imaging evaluation of carotid plaque and its relationship with ischemic stroke.

Objective:To investigate the impact of obstructive sleep apnea (OSA) on neurological function outcome in patients with acute ischemic stroke (AIS) at 90 days after onset.Methods:Patients with AIS admitted to Hefei Second People's Hospital from September 2022 to June 2023 were prospectively included. According to the modified Rankin Scale score at 90 days after onset, they were divided into a good outcome group (0-2) and a poor outcome group (>2). The demographic data, vascular risk factors, baseline laboratory tests, National Institutes of Health Stroke Scale (NIHSS) scores at admission, severity of obstructive sleep apnea (OSA), and apnea hypopnea index (AHI) were compared between the two groups. Multivariate logistic regression analysis was used to determine the independent risk factors for poor outcomes. Results:A total of 104 patients with AIS were enrolled, including 62 males (59.6%), with a median age of 65.5 years (interquartile range, 57.0-72.0 years). The median baseline NIHSS score was 3.00 (interquartile range, 2.00-4.00). The median AHI was 18.14/h (interquartile range, 11.34-27.88/h), 43 patients (41.35%) with no/mild OSA and 61 patients (58.65%) with moderate to severe OSA. Seventy-four patients (71.2%) had good outcome, and 30 patients (28.8%) had poor outcome. When introducing AHI as a categorical variable into the logistic regression model, the higher baseline NIHSS score (odds ratio [ OR] 3.041, 95% confidence interval [ CI] 1.797-5.145; P<0.001) and moderate to severe OSA ( OR 4.413, 95% CI 1.032-18.877; P=0.045) were independent risk factors for poor outcome; When introducing AHI as a continuous variable into the logistic regression model, higher baseline NIHSS score ( OR 3.176, 95% CI 1.844-5.472; P<0.001), age ( OR 1.093, 95% CI 1.014-1.177; P=0.020), and AHI ( OR 1.044, 95% CI 1.002-1.089; P=0.042) were independent risk factors for poor outcome. Conclusion:Moderate to severe OSA is an independent risk factor for poor functional outcome in patients with AIS at 90 days after onset, and a higher AHI indicates poor outcome in patients.

Objective:To investigate the efficacy and safety of intravenous thrombolysis based on diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) mismatch in patients with wake-up stroke (WUS).Methods:Patients with acute ischemic stroke received alteplase intravenous thrombolysis in the Stroke Center, the Second People's Hospital of Hefei from July 2019 to June 2021 were enrolled retrospectively. According to the time of finding the symptoms, they were divided into WUS group and non-WUS group. The demographic and baseline clinical data were documented and compared between the two groups. The efficacy endpoint was the clinical outcome assessed by the modified Rankin Scale (MRS) score at 90 d after onset. 0-2 was defined as a good outcome, and >2 were defined as a poor outcome. The primary safety endpoint was symptomatic intracranial hemorrhage (sICH); the secondary safety endpoint was death within 90 d after onset. Multivariate logistic regression analysis was used to determine the independent risk factors for poor outcome. Results:A total of 256 patients with acute ischemic stroke were enrolled, including 155 males (60.5%), aged 63.0±8.53 years. The median time from symptom onset to intravenous thrombolysis was 130.5 min, and the median baseline National Institutes of Health Stroke Scale (NIHSS) score was 7. Forty-eight patients (18.7%) were WUS and 208 (81.3%) were non-WUS; 186 (72.7%) had a good outcome and 70 (27.3%) had a poor outcome. There were no significant differences in 90 d good outcome rate (79.2% vs. 71.2%; χ2=1.260, P=0.262), sICH incidence (4.2% vs. 5.3%; χ2=0.102, P=0.750) and 90 d mortality (2.1% vs. 3.4%; χ2=0.000, P=1.000) between the WUS group and the non-WUS group. The baseline NIHSS score, the time from symptom onset to intravenous thrombolysis and the proportion of patients with cardiogenic embolism in the poor outcome group were significantly higher than those in the good outcome group (all P<0.05). Multivariate logistic regression analysis showed that the baseline NIHSS score (odds ratio 1.670, 95% confidence interval 1.453-1.919; P<0.001) and the time from symptom onset to intravenous thrombolysis (odds ratio 1.007, 95% confidence interval 1.000-1.015; P=0.043) were the independent risk factors for the poor outcome. Conclusion:The efficacy and safety of intravenous thrombolysis in DWI-FLAIR-mismatched wake-up stroke patients are comparable to those of acute ischemic stroke within the time window.

Objective:To investigate the predictive value of platelet-to-neutrophil ratio (PNR) on hemorrhagic transformation (HT) and poor outcomes at 90 d after intravenous thrombolysis (IVT) in patients with acute ischemic stroke (AIS).Methods:Patients with AIS received IVT in Hefei Second People's Hospital from July 2019 to June 2021 were retrospectively enrolled. HT was defined as intracerebral hemorrhage found by CT reexamination within 24 h after IVT, and the poor outcome was defined as the modified Rankin Scale score ≥3 at 90 d after onset. Multivariate logistic regression analysis was used to determine the independent predictors of HT and poor outcomes at 90 d, and the predictive value of PNR on HT and poor outcomes at 90 d was analyzed by receiver operating characteristic (ROC) curve. Results:A total of 202 patients with AIS treated with IVT were included, of which 32 had HT and 50 had poor outcomes at 90 d after onset. Multivariate logistic regression analysis showed that PNR at 24 h after IVT was significantly and independently negatively correlated with the poor outcomes (odds ratio [ OR] 0.959, 95% confidence interval [ CI] 0.928-0.991; P=0.012); PNR at admission ( OR 0.886, 95% CI 0.827-0.948; P<0.001) and PNR at 24 h after IVT ( OR 0.923, 95% CI 0.879-0.969; P=0.001) were significantly independently and negatively correlated with HT. ROC curve analysis showed that the area under the curve of PNR at 24 h after IVT for predicting poor outcomes was 0.733 (95% CI 0.659-0.807; P=0.012), the best cutoff value was 35.03, and the predictive sensitivity and specificity were 70.4% and 74%, respectively. The area under the curve of PNR at admission for predicting HT was 0.830 (95% CI 0.774-0.886; P<0.001), the best cutoff value was 34.81, and the predictive sensitivity and specificity were 70% and 93.7%, respectively. The area under the curve of PNR at 24 h after IVT for predicting HT was 0.770 (95% CI 0.702-0.839; P=0.001), the best cutoff value was 41.73, and the predictive sensitivity and specificity were 53.5% and 96.9%, respectively. Conclusion:For patients with AIS treated with IVT, lower PNR at 24 h after IVT is an independent predictor of the poor outcomes at 90 d, while PNR at admission and 24 h after IVT are the independent predictors of HT.

Objective:To investigate the efficacy and safety of using diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) mismatch to guide intravenous thrombolysis in patients with ischemic stroke beyond a 4.5-h time window.Methods:Patients with acute ischemic stroke received intravenous thrombolysis with alteplase in the Stroke Center of Hefei Second People's Hospital from July 2019 to June 2021 were retrospectively enrolled. According to the time of onset, they were divided into the time window group and the beyond time window group. The demographic and baseline clinical data of both groups were recorded and compared. The primary outcome measure was the clinical outcome assessed by the modified Rankin Scale (mRS) at 90 d after onset. 0-2 points were defined as good outcome, and >2 were defined as poor outcome. The secondary outcome measure was symptomatic intracranial hemorrhage (sICH). Multivariate logistic regression analysis was used to determine the independent risk factors for poor outcomes. Results:A total of 244 patients with acute ischemic stroke were enrollded, including 146 males (58.8%), aged 61.4±8.47 years. The median time from onset to thrombolysis was 142 min, and the median baseline National Institutes of Health Stroke Scale (NIHSS) score was 7. Thirty-six (14.8%) patients exceeded the 4.5 h time window, and 69 (28.3%) patients had poor outcomes. There were no significant differences in the good outcome rate (71.2% vs. 75.0%; χ2=0.224, P=0.636), any intracranial hemorrhage (9.6% vs. 13.9%; χ2=0.233, P=0.629) and the incidence of sICH (5.3% vs. 5.6%; χ2=0.000, P=1.000) between the time window group and the beyond time window group. Univariate analysis showed that the proportion of patients with atrial fibrillation or cardiogenic embolism and the baseline NIHSS score in the poor outcome group were significantly higher than those in the good outcome group (all P<0.05), while there was no statistical difference in the proportion of patients receiving intravenous thrombolysis beyond the time window. Multivariate logistic regression analysis showed that only the baseline NIHSS score was an independent risk factor for poor outcomes (odds ratio 1.681, 95% confidence interval 1.457-1.940; P<0.001). Conclusions:Compared with the patients who received intravenous thrombolysis within 4.5 h after onset, intravenous thrombolysis in patients with acute ischemic stroke beyond the 4.5 h time window guided by DWI-FLAIR mismatch results in similar clinical outcomes, and does not increase the incidence of intracranial hemorrhage.

Ischemic stroke is the second leading cause of disability and death worldwide, which brings heavy burden to society and families. Epidemiological studies have shown that vitamin D level are associated with the prevalence of hypertension, diabetes, atherosclerosis, and cerebrovascular events. This article reviews the relationship between vitamin D level and ischemic stroke risk, infarct volume in acute phase, severity of neurological deficit and functional outcome, and discusses the impact of vitamin D supplementation on ischemic stroke, and expects to provide new ideas for the prevention and treatment of ischemic stroke.