Risk factors associated with extracranial atherosclerosis in old patients with acute ischemic stroke

Prediction of recurrence of ischemic stroke within 1 year of discharge based on machine learning MRI radiomics

Purpose

This study aimed to investigate the value of a machine learning-based magnetic resonance imaging (MRI) radiomics model in predicting the risk of recurrence within 1 year following an acute ischemic stroke (AIS).

Methods

The MRI and clinical data of 612 patients diagnosed with AIS at the Second Affiliated Hospital of Nanchang University from March 1, 2019, to March 5, 2021, were obtained. The patients were divided into recurrence and non-recurrence groups according to whether they had a recurrent stroke within 1 year after discharge. Randomized splitting was used to divide the data into training and validation sets using a ratio of 7:3. Two radiologists used the 3D-slicer software to label the lesions on brain diffusion-weighted (DWI) MRI sequences. Radiomics features were extracted from the annotated images using the pyradiomics software package, and the features were filtered using the Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis. Four machine learning algorithms, logistic regression (LR), Support Vector Classification (SVC), LightGBM, and Random forest (RF), were used to construct a recurrence prediction model. For each algorithm, three models were constructed based on the MRI radiomics features, clinical features, and combined MRI radiomics and clinical features. The sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC) were used to compare the predictive efficacy of the models.

Results

Twenty features were selected from 1,037 radiomics features extracted from DWI images. The LightGBM model based on data with three different features achieved the best prediction accuracy from all 4 models in the validation set. The LightGBM model based solely on radiomics features achieved a sensitivity, specificity, and AUC of 0.65, 0.671, and 0.647, respectively, and the model based on clinical data achieved a sensitivity, specificity, and AUC of 0.7, 0.799, 0.735, respectively. The sensitivity, specificity, and AUC of the LightGBM model base on both radiomics and clinical features achieved the best performance with a sensitivity, specificity, and AUC of 0.85, 0.805, 0.789, respectively.

Conclusion

The ischemic stroke recurrence prediction model based on LightGBM achieved the best prediction of recurrence within 1 year following an AIS. The combination of MRI radiomics features and clinical data improved the prediction performance of the model.

Retrospective case-control study on screening risk factors of antibiotic-associated encephalopathy in patients with chronic kidney disease

OBJECTIVE

The renal excretion function of patients with chronic kidney disease (CKD) is reduced, and the nervous system toxic reactions of antibiotics are prone to occur. The purpose of this study is to screen out some risk factors for patients with CKD to suffer from antibiotic-associated encephalopathy (AAE).

DESIGN

A case-control study.

SETTING

A tertiary hospital in China.

PARTICIPANTS

The medical records of patients who were hospitalised for CKD and infectious diseases in our hospital from January 2010 to December 2019. All patients used antibiotics to treat infectious diseases during hospitalisation. All patients were divided into two groups according to whether they developed AAE during hospitalisation. The patients with CKD without AAE were selected as the control group (n=120), and the patients with CKD with AAE were regarded as the AAE group (n=102).

INTERVENTIONS

This study systematically analysed its clinical manifestations, laboratory examinations, prognosis, etc, and summarised the risk factors related to AAE in patients with CKD.

PRIMARY OUTCOME

Screening risk factors of AAE in patients with CKD.

RESULTS

Logistic regression analysis showed that coronary heart disease, as well as abnormal indicators of haemoglobin, albumin, uric acid and blood phosphorus were independent risk factors for patients with CKD with AAE (OR values were 4.137, 0.963, 0.849, 0.996 0.161, respectively, all p<0.05). The case fatality rate (Pearson χ²=7.524, p=0.006), rehospitalisation rate (Pearson χ²=6.187, p=0.013) and treatment costs (t=-8.44, p<0.001) in encephalopathy group are significantly higher than the control group.

CONCLUSIONS

Patients with CKD with AAE will increase the case fatality rate and cause poor prognosis. Coronary heart disease, as well as decreased levels of haemoglobin, albumin, uric acid, and blood phosphorus are independent risk factors for patients with CKD with AAE. Timely intervention of these risk factors may reduce the incidence of AAE and improve the prognosis.

Effect of electroacupuncture on glial fibrillary acidic protein and nerve growth factor in the hippocampus of rats with hyperlipidemia and middle cerebral artery thrombus

Electroacupuncture (EA) has been shown to reduce blood lipid level and improve cerebral ischemia in rats with hyperlipemia complicated by cerebral ischemia. However, there are few studies on the results and mechanism of the effect of EA in reducing blood lipid level or promoting neural repair after stroke in hyperlipidemic subjects. In this study, EA was applied to a rat model of hyperlipidemia and middle cerebral artery thrombosis and the condition of neurons and astrocytes after hippocampal injury was assessed. Except for the normal group, rats in other groups were fed a high-fat diet throughout the whole experiment. Hyperlipidemia models were established in rats fed a high-fat diet for 6 weeks. Middle cerebral artery thrombus models were induced by pasting 50% FeCl₃ filter paper on the left middle cerebral artery for 20 minutes on day 50 as the model group. EA1 group rats received EA at bilateral ST40 (Fenglong) for 7 days before the thrombosis. Rats in the EA1 and EA2 groups received EA at GV20 (Baihui) and bilateral ST40 for 14 days after model establishment. Neuronal health was assessed by hematoxylin-eosin staining in the brain. Hyperlipidemia was assessed by biochemical methods that measured total cholesterol, triglyceride, low-density lipoprotein and high-density lipoprotein in blood sera. Behavioral analysis was used to confirm the establishment of the model. Immunohistochemical methods were used to detect the expression of glial fibrillary acidic protein and nerve growth factor in the hippocampal CA1 region. The results demonstrated that, compared with the model group, blood lipid levels significantly decreased, glial fibrillary acidic protein immunoreactivity was significantly weakened and nerve growth factor immunoreactivity was significantly enhanced in the EA1 and EA2 groups. The repair effect was superior in the EA1 group than in the EA2 group. These findings confirm that EA can reduce blood lipid, inhibit glial fibrillary acidic protein expression and promote nerve growth factor expression in the hippocampal CA1 region after hyperlipidemia and middle cerebral artery thrombosis. All experimental procedures and protocols were approved by the Animal Use and Management Committee of Beijing University of Chinese Medicine, China (approval No. BUCM-3-2018022802-1002) on April 12, 2018.