The Clinical Study Of The Impact Of Glucose Variability On Outcome In Septic Patients

[Objective]Glucose variability is an independent risk factor associated with mortality in critically ill patients, but in China, it has not been applied to the sepsis trials in intensive care unit, and related research also has not defined sepsis severity. So this study tried to analyze the glucose variability in septic patients to understand the relationship between glucose variability and outcomes, and expect to know if the severity of sepsis is related to the higher glucose variability.[Methods]All adult patients admitted from September1,2010to September30,2011. and who fulfilled the criteria of sepsis either on admission or at any time during their ICU stay were included in the study, extracting patients'blood glucose values and other relevant clinical data from medical records. We calculated6indices of glucose variability in each patient (the standard deviation of blood glucose, coefficient of variation of blood glucose, mean absolute blood glucose changes per hour, maximum increase between two successive blood glucose and maximum decrease between two successive blood glucose); Statistically assess the relation between these indices of glucose variability and28days mortality.The area under the receiver operator characteristics (ROC) curves was calculated for the Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) score within first24hours diagnosing sepsis and for the Sepsis-related Organ Failure Assessment (SOFA) score. Through the comparison of the area under the ROC curves to determine a more superior index to access sepsis severity. And then analyzes it's relationship between glucose variability.[Results]Of the methods used to measure glucose variability, the standard deviation of glucose had the best discrimination for mortality (area under the curve=0.634, P=0.042). Multivariate logistic regression analysis showed that there was a significant interaction between standard deviation of glucose and the ICU mortality (odds ratio=4.216, p=0.023). SOFA score (area under the curve=0.787, p <0.001) had a greater area under the ROC curve than APACHE Ⅱ score (area under the curve=0.753, p<0.001). Further analysis showed that the median standard deviation in the highest quartile of the SOFA (2.37±0.90mmol/L) was greater than the median standard deviation in the lowest SOFA quartile (1.90±0.70mmol/L)(p=0.043).IConclusions]Glucose variability is independently associated with28days mortality of patients with sepsis. Sepsis severity has positive correlation with glucose variability.