ARDS患者血糖异常波动程度与和短期临床预后之间的相关性分析

Correlation analysis of glucose fluctuation,critical illness scores and 28-day mortality in patients with ARDS

目的探讨急性呼吸窘迫综合征(ARDS)患者血糖变异程度和短期临床预后的相关性。方法本研究共收录92例ARDS患者(2015年1月至2017年6月),分为存活组(n=56)和死亡组(n=36)。所有患者于入组时检测初始血糖(blood glucose on admission,BGad)、血糖变异系数(coefficient of blood glucose variability,BGcv)、48小时平均血糖(mean blood glucose,BGm)及其标准差(standard deviation blood glucose,BGsd),并计算血糖不稳定指数(blood glucose instability index,BGI)用于评价血糖波动情况。所有ARDS患者分别于入组时和死亡或转出ICU当日,两个时间点进行急性生理与慢性健康评分II评分(APACHE II)等评估病情危重程度。结果死亡组患者BGsd(2. 41±0. 38 mmol/Lvs. 1. 26±0. 25 mmol/L,P <0. 001),BGcv(40. 12±7. 32%vs. 20. 11±3. 63%,P <0. 001)和BGI(11. 21±2. 02(mmol/L) 2/h/dvs. 4. 67±1. 45(mmol/L) 2/h/d,P <0. 001)水平均明显高于存活组患者。死亡组中糖尿病亚组患者BGsd(2. 89±0. 45 mmol/L vs.2. 14±0. 28 mmol/L,P <0. 05)和BGI(13. 15±2. 52(mmol/L) 2/h/d) vs. 10. 51±1. 81(mmol/L) 2/h/d),P <0. 05)明显高于非糖尿病亚组。糖尿病亚组间比较可见,死亡组糖尿病亚组患者BGsd(2. 89±0. 45 mmol/Lvs. 1. 36±0. 61mmol/L,P <0. 001),BGcv (42. 45±11. 61%vs. 22. 53±7. 90%,P <0. 001),BGI (13. 15±2. 52(mmol/L) 2/h/d) vs. 4. 98±1. 42(mmol/L) 2/h/d),P <0. 001)均高于存活组糖尿病亚组。非糖尿病亚组间比较可见,死亡组非糖尿病亚组患者BGsd(2. 14±0. 28 mmol/Lvs. 1. 13±0. 28 mmol/L,P <0. 001),BGcv(39. 54±9. 33%vs. 20. 71±6. 16%,P <0. 001),BGI(10. 51±1. 81(mmol/L) 2/h/d) vs. 4. 43±2. 21(mmol/L) 2/h/d),P <0. 001)均高于存活组非糖尿病亚组。相关性分析提示BGcv(r=0. 687,P <0. 001)和BGI(r=0. 699,P <0. 001)均与28天死亡事件呈正相关。BGcv和BGI之间存在正相关(r=0. 654,P <0. 001)。结论对于ARDS患者来说,血糖剧烈波动可能预示不良短期预后,BGcv和BGI有助于早期预后评估。

Objective To evaluate the correlation of glucose fluctuation, critical illness scores and 28-day mortality in patients with acute respiratory distress syndrome （ARDS）. Methods 92 patients with ARDS were enrolled in this study and divided into two group： the death group （ n =36） and the survival group （ n =56）. All patients received glucose fluctuation and critical illness score evaluation. The blood glucose on admission （BGad）, mean blood glucose （BGm）, standard deviation blood glucos e（BGsd） and coefficient of variation （GLUcv） were calculated in each patient, and coefficient of blood glucose variability （BGcv） and blood glucose instability index （BGI） were used as glucose swing. Results The levels of BGsd （2.41±0.38 mmol/L0x09vs. 1.26±0.25 mmol/L, P〈 0.001）, BGcv （40.12±7.32% vs. 20.11±3.63%, P〈 0.001） and BGI （11.21±2.02 （mmol/L） 2/h/d0x09vs. 4.67±1.45 （mmol/L） 2/h/d, P〈 0.001） in the death group were significantly higher than those in the survival group. The levels of BDsd （2.89±0.45 mmol/L vs. 2.14±0.28 mmol/L, P〈 0.05） and BGI （13.15±2.52 （mmol/L） 2/h/d） vs.10.51±1.81 （mmol/L）2/h/d）, P〈 0.05） in the DM subgroup of death were higher than those in the non-DM subgroup of death. The levels of BGsd （2.89±0.45 mmol/L vs. 1.36±0.61mmol/L, P〈 0.001）, BGcv （42.45±11.61%0x09vs. 22.53±7.90%, P〈 0.001） and BGI （13.15±2.52 （mmol/L） 2/h/d vs. 4.98±1.42 （mmol/L） 2/h/d, P〈 0.001） in the DM subgroup of death were higher than those in the DM subgroup of survival. Correlation analysis showed that BGcv （ r =0.687, P〈 0.001） and BGI （ r =0.699, P〈 0.001） were positively related with 28-day mortality. Conclusion Glucose variability （BGcv and BGI） may be a marker for short-time prognosis in ARDS patients.