危重症患者血糖变异度与预后的关系

目的:通过统计危重症患者的血糖变异度,分析其对预后的影响。方法:选择2008年3月-2012年2月入住我科的812例危重患者,统计每例患者的血糖值,计算其平均血糖(average glucose,GLUav)水平及标准差(glucosestandard deviation,GLUsd),以标准差来反映血糖变异度。统计不同平均血糖及血糖标准差患者的死亡率,并比较平均血糖、血糖标准差、平均急性生理与慢性健康(acute physiology and chornic health evalution,APACHE)Ⅱ评分、平均序贯器官衰竭估计(sequential organ failure assessment,SOFA)评分对死亡率的预测价值。结果:812例患者APACHEⅡ评分为(21.0±7.7),SOFA评分(8.7±3.8),总死亡率为26.5%,死亡率从平均血糖水平70～99 mg/dL的18.2%到血糖超过180 mg/dL的36.4%。血糖变异度与死亡率之间存在显著相关性。整个研究队列中,血糖变异度最小的1/8人群死亡率为6.2%,其余依次增加,分别为13.5%、17.8%、26.2%、32.1%、38.4%、45.6%和50.1%。血糖变异度最小组ICU住院时间也显著短于其他各组(P<0.001)。死亡组APACHEⅡ评分、SOFA评分、平均血糖及血糖标准差均明显高于存活组。ROC曲线显示,用血糖标准差反映的血糖变异度预测死亡率敏感性与APPACHEⅡ评分相当,高于SOFA评分及平均血糖。结论:血糖变异度是预测危重患者死亡的独立危险因素。对实施血糖控制的ICU患者,应研究以降低血糖变异度为目标的策略以达到改善其预后的目的。

The relationship of glucose variability and prognosis in critical patients

Objective: To analyze the influence of the prognosis through the statistics of the critically ill patients with glucose variability.Methods: Eight hundred and twelve critically ill patients admitted to our ICU from March 2008 to February 2012,were chosen into our study.All glucose results of each patient were recorded and calculated the average glucose and standard deviation,standard deviation was used as an indicator of glycemic variability.Statistical Analysis the hospital mortality of different average glucose and standard deviation,compare the valve of predicting mortarity of the different factors(average glucose,standard deviation,APPACHEⅡ,SOFA).Results: The mean APACHE Ⅱ score of the 812 patients was(21.0 ± 7.7),the mean SOFA score was(8.7 ± 3.8),total mortality was 26.5%,ranging from 18.2% among patients with mean glucose level 70 mg/dL to 99 mg/dL to 36.4% among patients with mean glucose level 180 mg/dL.The relationship between glucose variability and mortality was strongest in the euglycemic range.Mortality among patients in the entire cohort with the lowest quartile of glucose variability was 6.2%,increasing to quartiles.Intensive care unit length of stay was shorter among patients in the first quartile of glucose variability compared with those in the other three(P<0.001).The APACHE Ⅱ scores,SOFA score,average glucose,standard deviation of dead group were higher than those of survival group.ROC curve indicated that the sensitivity of standard deviation predict mortality was as hight as APPACHE Ⅱ.Conclusions: This study demonstrates that glucose variation is a independent risk factors of death in critical patients.We should take measures to ensure a low degree of glucose variability so as to improve outcomes in intensive care unit’s implementing glycemic control.