Clinical Study On Blood Glucose In Children Sepsis

Objective:To investigate the association of glucose variability with pathogenetic condition and prognosis of children with sepsis.Methods:Retrospective study of 140 cases children with sepsis in Children's Hospital of Chongqing Medical University from July 2013 to December 2014. Divided the patients into 3 groups according to the severity of the disease:sepsis group, severe sepsis group and septic shock group. And divided the patients into 2 groups according to prognosis of the disease:death group and survival group. Collected the relevant clinical data of patients during staying in PICU from medical records. Extracting all the blood glucose values, which included the initial blood glucose (GLUadm), blood glucose maximum (GLUmax), blood glucose minimum (GLUmin). To calculate the average blood glucose (GLUave), standard deviation of blood glucose (GLUsd), coefficient of variation of blood glucose (GLUcv). Analyzed the occurrence rate of stress hyperglycemia and hypoglycemia. Compared the differences between death and survival groups in each index related to blood glucose, PCIS score, length of hospital stay, etc. To find the association of each index with pathogenetic condition and prognosis of patients. Applied the receiver-operating characteristic (ROC) curve to evaluate the value in predicting adverse outcomes of GLUsd and GLUcv. Used the logistic regression analysis the effect on the prognosis of statistically significant indexes.Results:The incidence rate of Stress hyperglycemia and hypoglycemia in death group were higher than survival group. The GLUadm, GLUave, duration of mechanical ventilation and PICU hospitalization days were not different between groups of different prognosis and groups of different pathogenetic condition(P>0.05). The GLUmin, GLUdif, GLUsd, GLUcv, PCIS score and length of hospital stay were significantly different between groups of different prognosis and groups of different pathogenetic condition(P<0.05). The GLUmax was significantly different between groups of different prognosis, but no different between groups of different pathogenetic condition. From the correlation analyses about indexes which were statistically significant in the groups of different prognosis, we found that GLUmax, GLUdif, GLUsd, GLUcv were positively related with prognosis, GLUmin, PCIS score, the length of hospital stay were negatively correlated with prognosis. To take the thirteen indexes about this study into single-factor and multi-factors Logistic regression analysis, found that GLUcv was the independent risk factor of adverse outcomes.Conclusion:The average blood glucose(GLUave) had no correlation with the disease prognosis. The standard deviation of blood glucose (GLUsd) and coefficient of variation of blood glucose (GLUcv) had significant correlation with the disease prognosis, and the GLUcv was the independent risk factor of adverse outcomes, means With the increase of GLUcv, the risk of death increased. Therefore,maybe it is an important way that reducing the glucose variability to improve the prognosis of children with sepsis.