血糖变异性与重症急性卒中患者3个月内死亡风险的相关性

Correlation between blood glucose variability and the risk of death within 3 months in patients with severe acute stroke

目的探讨血糖变异性对重症急性卒中患者3个月内死亡风险的影响。方法回顾性分析河北大学附属医院神经重症监护室收治的重症急性卒中患者的临床资料。根据3个月内转归将患者分为存活组和死亡组，比较2组基线美国国立卫生研究院卒中量表（National Institutes of Health Stroke Scale, NIHSS）评分、急性生理与慢性健康评分Ⅱ（Acute Physiology and Chronic Health Evaluation Ⅱ，APACHE Ⅱ）、平均血糖、血糖标准差、平均血糖波动幅度、平均血糖变异系数以及其他临床资料，应用多变量logistic回归分析确定3个月内死亡风险的独立预测因素。结果共纳入176例重症急性卒中患者，其中缺血性卒中124例（70.4%），脑出血52例（29.6%）；27例（15.3%）在3个月内死亡。死亡组基线NIHSS评分[（22.5±2.3）分对（17.1±2.7）分；t＝8.974，P〈0.001]、APACHE Ⅱ评分[（17.5±5.1）分对（6.8±7.1）分；t＝13.136，P〈0.001]、血糖标准差[（3.6±2.2）mmol/L对（1.4±1.1）mmol/L；t＝9.246，P〈0.001]、平均血糖波动幅度[（6.3±3.5）mmol/L对（3.8±2.1）mmol/L；t＝5.989，P〈0.001]和平均血糖变异系数[（43.3±13.1）%对（22.8±11.3）%；t＝8.593，P〈0.001]均显著高于存活组，但平均血糖水平无显著差异[（13.5±1.7）mmol/L对（12.6±1.5）mmol/L；t＝0.990，P＝0.323]。多变量logistic回归分析显示，基线APACHEⅡ评分[优势比（odds ratio, OR）2.911，95%可信区间（confidence interval, CI）1.962～4.560；P〈0.001]、NIHSS评分（OR 12.436，95% CI 5.071～30.495；P〈0.001）、血糖标准差（OR 4.302，95% CI 1.855～9.973；P＝0.001）、平均血糖波动幅度（OR 7.231，95% CI 1.585～32.992；P＝0.011）和平均血糖变异系数（OR 3.408，95% CI 1.363～8.522；P＝0.009）是重症急性卒中患者3个月内死亡的独立预测因素。结论血糖变异性是重症急性卒中患者3个月内死亡风险的独立预测因素，血糖波动范围大提示转归不良，其预测价值与基线APACHE Ⅱ评分相当。相比之下，平均血糖水平与患者转归无显著相关性。

Objective To investigate the correlation between blood glucose variability and the risk of death within 3 months in patients with severe acute stroke.Methods The clinical data of 176 patients with acute severe acute stroke admired to the Neurology Intensive Care Unit, the Affiliated Hospital of Hebei University were analyzed retrospectively. They were divided into a survival group and a death group according to the outcomes within 3 months. The baseline National Institutes of Health Stroke Scale （NIHSS） score, Acute Physiology and Chronic Health Evaluation （APACHE） Ⅱ, mean blood glucose, blood glucose standard deviation, mean amplitude of glycemic excursion, mean coefficient of variation of blood glucose, and other clinical data were compared between both groups. Multivariate logistic regression analysis was used to determine the independent predictors of death within 3 months. Results A total of 176 patients with severe acute stroke were included, 124 （70.4%） with ischemic stroke, 52 （29.6%） with intracerebral hemorrhage, and 27 （15.3%） died within 3 months. The baseline NIHSS scores （22.5 ± 2.3 vs. 17. 1 ± 2. 7; t =8. 974, P 〈 0. 001）, APACHE Ⅱ scores （17.5 ±5. 1 vs. 6.8 ±7.1; t=13.136, P 〈 0, 001）, blood glucose standard deviation （3.6 ± 2.2 mmol/L vs. 1.4 ± 1.1 mmol/L; t = 9. 246, P 〈 0. 001 ）, mean amplitude of glycemic excursion （6. 3 ± 3.5 mmol/L vs. 3.8 ± 2. 1 mmol/L; t = 5. 989, P 〈 0. 001 ）, and mean coefficient of variation of blood glucose （43.3% ± 13.1% vs. 22.8% ± 11.3%; t= 8. 593, P〈 0. 001） in the death group were significantly higher than those in the survival group, however, there was no significant difference in the mean blood glucose （13.5 ± 1.7 mmol/L vs. 12.6 ± 1.5 mmol/L; t = 0. 990, P =0. 323）. Multivariate logistic regression analysis showed that the baseline APACHE Ⅱ scores （odds ratio [ OR ] 2.911,95 % c onfidene e interval [ CI] 1. 962-4. 560; P 〈 0. 001）, NIHS S scores （OR 12. 436, 95 % C1 5. 071-30. 495; P 〈0. 001）, blood glucose standard deviation （OR 4. 302,95% CI 1. 855-9. 973; P =0. 001）, mean amplitude of glycemic excursion （OR 7.231, 95% CI 1. 585-32. 992; P = 0. 011 ）, and mean coefficient of variation of blood glucose （OR 3. 408, 95% CI 1. 363-8. 522; P =0. 009） were the independent predictors of death within 3 months in patients with severe acute stroke. Conclusions The blood glucose variability is an independent predictor of death within 3 months in patients with severe acute stroke. The great range of glycemic excursion indicates poor outcome. Its baseline predictive value is almost the same with the APACHE Ⅱ score. In contrast, the mean blood glucose level do not have significant correlation with the outcomes of patients.