Abstract: Objective To analyze the clinical risk factors of hemorrhagic transformation (HT) of cardiogenic cerebral embolism and the influence of HT on outcome. Methods The clinical data of 115 inpatients were reviewed from May, 2012 to December, 2015. They were divided into HT group (n=58) and non-HT group (n=57). The age, anticoagulant therapy, thrombolytic therapy, infarction diameter, diabetes, coronary heart disease, hyperlipidemia, the National Institutes of Health Stroke Scale (NIHSS) score and HAS-BLED score were compared.The risk factors for HT was screened with the multivariate Logistic regression. NIHSS score and Modified Rankin Scale (mRS) score as hospitalization, and one month and three months after stroke were compared. Results There were significant difference in NIHSS score (t=-2.991, P=0.003) and HAS-BLED score (t=-2.499, P=0.014), as well as infarction diameter (χ2=8.355, P=0.004) between HT group and non-HT group. NIHSS score (OR=1.127, P=0.027), HAS-BLED score (OR=1.783, P=0.03) and infarction diameter (OR=4.390, P=0.035) were the risk factors for HT. The incidence of HT was less in low-risk group (HAS-BLED score=0-2) than in high-risk group (HAS-BLED score≥3) (χ2=4.643, P=0.031). The NIHSS score as hospitalization, and one month and three months after stroke were all more in HT group than in non-HT group (t>2.387, P<0.05). The mRS score was more in HT group as hospitalization (t=-2.262, P=0.026), but not significant one and three months later (t<1.468, P>0.05). Conclusion HT tends to happen in the patients of cerebral embolism patients after atrial fibrillation with severe neural function defect, large infarction diameter and high HAS-BLED score. The neural function is poor in those with HT.

Key words: cerebral embolism, atrial fibrillation, hemorrhagic transformation, risk factor, outcome