Patients admitted to Ruian People's Hospital within symptom onset for the treatment of first-ever ischemic strokes were recruited to this study between March 2017 and October 2018. Patients were eligible for the study if they were admitted within 7 days of stroke onset and exhibited evidence of acute ischemic stroke on MRI and MRA. Patients with 1) extracranial or intracranial atherosclerosis causing ≥50% luminal stenosis or occlusion of arteries supplying the ischemia area, 2) no major-risk cardiogenic embolism by history, electrocardiography, echocardiography and 24 h dynamic electrocardiogram, and 3) no other specific causes of ischemic stroke were placed into the LAA group. Patients with AF were confirmed by electrocardiography and 24 hour dynamic electrocardiogram, the criterion and classification of HF were used according to the European Society of Cardiology (ESC) HF guidelines. And patients with 1) absence of extracranial or intracranial atherosclerosis causing ≥50% luminal stenosis or occlusion of arteries supplying the ischemia area, 2) no other specific causes of ischemic stroke were placed into the AF group, and 3) absence of valvular heart disease.
The exclusion criteria were 1) malignancies, 2) intracerebral hemorrhage, 3) recent acute coronary syndrome, 4) renal or hepatic diseases, 5) autoimmune diseases, and 6) any other concomitant terminal disease. Patients with other causes of stroke such as valvular AF, cardioembolic stroke without AF, undetermined stroke etiology had been excluded from the study. This study conformed to the Declaration of Helsinki and was approved by ethics committee of Wenzhou People's Hospital and Ruian People's Hospital. Written informed consent was obtained from all subjects or their immediate family members.
Acute ischemic lesion analysis
The MRI and MRA parameters in this study were measured as previous described (18). The ischemic lesions on diffusion-weighted imaging (DWI) were evaluated including size, composition, and distribution. The infarct volume was calculated by ABC/2 (where A is longest dimension in axis x, B is longest perpendicular dimension to axis x (y), and C is total length in z dimension) (19). Two experienced neurologists independently reviewed neuroimages and reported the results. Patients were divided to three infarct composition pattern groups according to pattern of the ischemic lesion(s): 1) small (< 10 mm in diameter) lesions only, 2) mixed small and large lesions, and 3) large lesions only（≥10 mm in diameter on DWI）(20). Patients were also categorized based on topography pattern of ischemic lesion(s): 1) subcortical-only pattern, 2) small cortical-only pattern, and 3) large cortical/corticaldeep pattern (21).
Baseline Data Collection
Demographic data (age and sex) and history of risk factors (hypertension, diabetes mellitus, congestive heart failure, history of vascular disease, systolic blood pressure; diastolic blood pressure, smoking, and alcohol abuse) were collected at admission via in-person interviews with the patients or their family members. Blood samples of patients were obtained in the next morning of the day of admission. After centrifugation, aliquots of the samples were immediately stored at − 80 °C before assay. Routine blood biomarkers, including triglyceride, cholesterol, high-density lipoproteinm (HDL), low-density lipoprotein (LDL), Lipoprotein (a), high-sensitivity C-reactive protein (hs-CRP), fasting blood glucose (FBG), albumin and Creatinine (Cr), were examined using standard detection methods. MPV as well as leukocyte, erythrocyte, and platelet counts were tested by a Sysmex XE-2100 hematology analyzer (Sysmex, Kobe, Japan). We collected echocardiographic parameters from all patients according to the current guidelines (22), including left ventricular ejection fraction (LVEF), Left atrial diameter (LAD), left ventricular end-diastolic diameter (LVEDd), left ventricular end-systolic diameter (LVEDd) and inter-ventricular septal thickness at end diastole (IVSd).
All the data are expressed as mean ± SD. χ2 was used to examine differences in discrete variables among the groups. Differences in continuous variables were examined by using Mann–Whitney test and Kruskal–Wallis test. In addition, independent factors for ischemic lesion volume were evaluated using linear regression. Adjustment variables in the multivariable regression models were chosen from potential outcome determinants with significant clinical relevance in univariate analysis. Statistical significance was defined as p<0.05. Statistical analysis was performed by using SPSS version 20.0.