AIMS/HYPOTHESIS: The purpose of this study was to examine the relationship among the blood glucose levels and mortality rates between the patients admitted to Intensive care unit. METHODS: In a Retrospective Observational study, Blood glucose levels were measured on patients admitted to the Intensive care unit. These measurements were matched against demographic data and hospital mortality rates. RESULTS: Overall in-hospital mortality was 33%. Blood glucose levels during admission were an

more »

... ndent predictor of mortality. There was a significant interaction among baseline blood glucose status and mortality (p<0.001). Higher Blood glucose levels were associated with greater mortality evens between non-diabetes patients also. Between non-diabetes the lowest mortality rate (3.0%) was observed in a group of people with BGL between 100-140mg/dl. Compared with this group patients with a BGL of >140mg/dl had increased mortality rate (p<0.001, after adjustment for age and sex). The risk increased farther at higher glucose levels.