脓毒性休克患者动脉血及末梢血床旁即时血糖检测准确性的研究

目的:探讨脓毒性休克患者动脉血及末梢血床旁即时血糖检测准确性及影响准确性的相关因素。方法:前瞻性纳入2016-03-2019-04期间我院急诊科收治的且预期住院时间>48 h的严重脓毒症患者122例。根据是否出现脓毒性休克分为研究组(伴有脓毒性休克)与对照组(不伴有脓毒性休克)。每组均采用末梢血-血糖仪法及动脉血-血糖仪法测量血糖,同时以实验室测量的血糖作为标准进行准确性评估。对可能影响末梢血-血糖仪法检测准确性的各因素进行多元线性回归分析。结果:所有患者末梢血-血糖仪法及动脉血-血糖仪法检测的血糖值均高于实验室检测的血糖值,但差异无统计学意义(P>0.05)。Bland-Altman分析结果显示:研究组末梢血-血糖仪法检测的血糖值95%一致性界限范围较宽(7.7^-6.0 mmol/L),超上、下限范围发生率为5.7%(可接受限度5%)。研究组末梢血-血糖仪法及动脉血-血糖仪法准确率分别为84.9%及95.7%,两者差异有统计学意义(F=4.220,P=0.040)。对照组末梢血-血糖仪法及动脉血-血糖仪法准确率分别为95.4%及97.8%,两者差异无统计学意义(P>0.05)。研究组实验室检测的血糖值与末梢血-血糖仪法及动脉血-血糖仪法检测的血糖值的相关系数分别为0.637及0.966,两者差异有统计学意义(Z=6.378,P<0.01)。对照组实验室检测的血糖值与末梢血-血糖仪法及动脉血-血糖仪法检测的血糖值的相关系数分别为0.950及0.974,两者差异无统计学意义(Z=1.913,P>0.05)。多元线性回归分析提示高血糖水平、低动脉血氧饱和度及高去甲肾上腺素用量对研究组患者末梢血-血糖仪法血糖检测的准确性有明显影响(均P<0.05)。结论:脓毒性休克患者末梢血-血糖仪法检测的血糖值准确率性较差,建议应用动脉血进行床边即时血糖检测,这有助于避免不当调整胰岛素治疗和低血糖发作。

The accuracy of point-of-care testing for blood glucose monitoring with arterial and peripheral blood samples in patients with septic shock

Objective: To investigate the accuracy and realted factors of point-of-care testing for blood glucose monitoring with arterial and peripheral blood samples in patients with septic shock. Method: This prospective study carried out in consecutively enrolled 122 patients stayed more than 48 hours in the emergency department of the Second Affiliated Hospital of Hainan Medical College from March 2016 to April 2019. They were divided into study group(with septic shock) and control group(without septic shock). The peripheral blood/glueometer and arterial blood/glueometer were used to determine blood glucose in each group, compared with arterial blood/blood biochemical meter. Multiple linear regression analysis was used to evaluate the possible influencing factors of the peripheral blood/glueometer. Result: The levels of blood glucose of the peripheral blood/glueometer and arterial blood/glueometer in all patients were higher than those of the arterial blood/blood biochemical meter, but there was no significant difference(P>0.05). The results of Bland-Altman analysis showed that the 95% limits of agreement of blood glucose determined by peripheral blood/glueometer in the study group was more wider(7.7--6.0 mmol/L), and the incidence of exceeding the upper and lower limits ranged was 5.7%(acceptable limit was 5%). The accuracy rates of peripheral blood/glueometer and arterial blood/glueometer in the study group were 84.9% and 95.7%, respectively. There was significant difference between the two methods(F=4.220, P=0.040). The accuracy rates of peripheral blood/glueometer and arterial blood/glueometer in the control group were 95.4% and 97.8%, respectively. There was no significant difference between the two methods(P>0.05). The correlation coefficients of blood glucose determined by laboratory with blood glucose determined by peripheral blood/glueometer and arterial blood/glueometer in the study group were 0.637 and 0.966, respectively. There was significant difference between them(Z=6.378, P<0.01). The correlation coefficients of blood glucose determined by laboratory with blood glucose measured by peripheral blood/glueometer and arterial blood/glueometer in the control group were 0.950 and 0.974, respectively. There was no significant difference between them(Z=1.913, P>0.05). Multivariate linear regression analysis showed that hyperglycemia, low arterial oxygen saturation and high norepinephrine dosage had significant effects on the accuracy of blood glucose determined by peripheral blood/glueometer in the study group(all P<0.05). Conclusion: The accuracy of point-of-care testing for blood glucose monitoring with peripheral blood samples in patients with septic shock is poor. And it is recommended to use arterial blood for point-of-care testing for blood glucose monitoring, which helps to avoid improper adjustment of insulin therapy and hypoglycemic attacks.