Background: Admission hyperglycemia is defined as any blood glucose level greater than 140 mg/dl (> 7.8 mmol/l) in critical care unit according to American Diabetes Association 2015. Stress hyperglycemia is related to multiple causes as inflammatory and neuro-endocrine derangements in critically ill patients,which lead to insulin resistance and high hepatic glucose output.Aim: To determine the frequency,in-hospital mortality and length of stay in a cohort of patients with admission hyperglycemia in unselected acute medically ill patients admitted to Medical intensive care units (medical ICU and CCU) attending Assuit university hospital, Internal medicine department. And to evaluate whether admission hyperglycemia or other comorbid conditions responsible for outcome of critical ill patients.Patients and methods: Aprospective,observational, non-interventional study involved 170 patients were admitted at ICU unit of Internal Medicine department at Assiut University Hospitals during the period between july1st 2016 and 30th December 2016 were enrolled. Measurement of random blood glucose on admission to ICU if less than 140 mg/dl :catrgorized as normoglycemic. If more than 140mg/dl: catrgorized as hyperglycemic:Follow up blood glucose on days of admission to ICU every 8 hours till either discharge or death or maximum four day in addition to Fasting ,postprandial glucose ,HbA1C,liver function test (LFT), complete blood count (CBC)&kidney function tests ,serum sodium , serum potassium ,Arterial blood gases and Calculation of APACHE II score.Results: The present study included 170 patient,35.3% were normoglycemic and 64.7% were hyperglycemic which further subdivided into 43% known diabetics ,14.1%stress hyperglycemia and 7.6% newly discovered diabetics.Percentage of survivors was 72.9% versus non survivors was 27.1%. Median hospital stay for all patients was 6 (4- 28) days with in hospital mortality was 46 (27.1%) patients. It was noticed that frequency of non- survivors was higher in patients with Hyperglycemic versus survivors.Conclusion: stress hyperglycemia and diabetes were independent predictors for in hospital mortality in patients with admission hyperglycemia attending ICU.