Abstract: Objective  To study the relationship between blood glucose variability and prognosis in patients with acute ischemic stroke for the clinical basis in relieving the prognosis in the patients. Methods  Acute ischemic stroke patients in our hospital were enrolled.  All patients were classified by  TOAST subtypes.  According to the standard deviation of the blood glucose level，  the patients were divided into the blood glucose variability group (standard deviation of blood glucose level≥1.4  mmol/L) and control group(standard deviation of blood glucose level ＜1.4 mmol/L).  In addition， neurological function deicits scale was evaluated by the National Institutes of Health Stroke Scale(NIHSS)  and the conditions of prognosis after the 90 days was evaluated by the modified Rankin Scale. The difference in  blood glucose variability group and control group was compared.  Then， the relationship between blood glucose variability and prognosis was analyzed with logistic regressive model. Results  Eightythree patients (47.4%) had blood glucose variability. The average age (74.7±12.7 vs 70.0±13.3 years，P=0.019)， diabetes history(28  cases vs 14 cases，P=0.007)， and stroke history (16 cases vs 6 cases，P<0.01) were significantly different between blood glucose variability group and control group. Logistic regressive model showed that the blood glucose variability (OR=3.235，95%CI=1.1858.130，P=0.022)， advanced age(OR=1.114，95%CI=1.0611.169，P<0.001) and higher admission  NIHSS scores(OR=2.214， 95%CI=1.7132.863，P<0.001) were independent risk factors of  acute ischemic stroke. Conclusion  Blood glucose variability， advanced age and higher admission NIHSS scores were related to 90 days’  prognosis of acute ischemic stroke.

Key words: blood glucose, cerebrovascular disorders, risk factors