Objective To investigate and analyze the hyperglycemia condition in patients with stroke Neurological Intensive Care Unit (NICU) , and study the relation between blood glucose level and the risk of mortality. Methods A retrospective analysis involving hyperglycemic patients with stroke admitted to NICU from July 1, 2017 to December 3.1, 2018 was conducted. Based on outcomes? these patients were divided into survival group and dead group。The difference of blood glucose condition between the two groups and the relation between blood glucose level and the risk of mortality were analyzed. Results 395 hyperglycemic patients were involved, including 22 in the dead group and 373 in the survival group. Compared with the survival group, the average age (78. 86士 11. 13 vs 67. 81 + 16. 04, P<0. 01) , Acute Physiol ogy and Chronic Health Evaluation K (APACHE H ) (score) (12. 9 士 3. 92 vs .10. 78 士 3, 01, P<0. 05) , mean blood glucose (mmol/L) (12. 93 + 5. 49 vs 10. 7士4. 56, PV0. 01) , incidence of hyperglycemia (blood glucose >7. 8 mmol/L) (83.31% vs 68. 55 %, PV0.01), incidence of severe hyperglycemia (blood glucose >13,9 mmol/L) (36. 0.1 % vs 21. 65 % ? P<0. 01.) , incidence of severe hypoglycemia(blood glucose <3, 0 mmol/L) (0. 28% vs 0.18% , P<0. 05), standard deviation of blood, glucose (mmol/L) (4. 20士 1. 44 vs 2. 84+1. 44, FV0. 01) , largest fluctuation range of blood glucose (mmol/L) (17. 34+8. 48 vs 11. 22士6. 45 , P<0. 01) in the dead group increased significantly? while the control rate (62 47% vs 78. 07%” P<0. 01) decreased. Multivariate analysis showed that, age (OR=1. 083), APACHE H (OR= 1. 282) and mean blood glucose (OR = 1. 424) were independent risk factors for mortality in hyperglycemic patients with stroke. Conclusion The hospitalized patients with stroke in NIGU had a high incidence of hyperglycemia and a large fluctuation of blood glucose, especially in the dead patients. We should pay more attention to blood glucose management.Age, APACHE R and mean blood glucose may be the independent risk factors for mortality.