Abstract：
OBJECTIVE: Limited research exists on translation of in-vitro glucose measurement interfering compounds to the in-vivo situation. We investigated whether Point-of-Care glucose measurements by Accu Chek Inform II (ACI II) were accurate to monitor glucose concentrations during surgery with general anesthesia by comparing with the reference laboratory hexokinase plasma glucose test.
METHODS: Patients undergoing surgery with general anesthesia were included. Anesthesia was maintained with either Sevoflurane or Total intravenous anesthesia (TIVA). Prior to and after induction, blood glucose was measured with ACI II and the hexokinase test. Bland-Altman analysis was performed to assess method agreement. Subgroup analyses on glucose measurement differences per type of maintenance anesthesia were performed.
RESULTS: Thirty-nine patients were included, and 78 measurements were performed. All paired measurements had clinically acceptable agreement with a percentage error of 10.0% (95% CI 8.0 to 11.9). The mean difference (95% limits of agreement) between ACI II and hexokinase for all measurements was 0.0 mmol/L (-0.7 to 0.7 mmol/L). Before induction (n = 39), mean difference was -0.1 mmol/L (-0.6 to 0.4 mmol/L), and after induction (n = 39), mean difference was 0.1 mmol/L (-0.8 to 0.9 mmol/L). Further investigation showed the difference varied per test for patients receiving Sevoflurane compared to patients receiving TIVA (-0.2 ± 0.4 mmol/L vs. 0.4 ± 0.3 mmol/L, p < 0.001). Before and after induction, the difference between ACI II and hexokinase measurements increased for patients receiving Sevoflurane compared to patients receiving TIVA (0.4 ± 0.4 mmol/L vs. -0.4 ± 0.3 mmol/L, p < 0.001).
CONCLUSIONS: The agreement between glucose measurements using ACI II and the reference laboratory hexokinase test was clinically acceptable with a percentage error of 10.0% (95% CI 8.0 to 11.9). The use of TIVA may negatively affect the measurement performance of the ACI II.
METHODS: Patients undergoing surgery with general anesthesia were included. Anesthesia was maintained with either Sevoflurane or Total intravenous anesthesia (TIVA). Prior to and after induction, blood glucose was measured with ACI II and the hexokinase test. Bland-Altman analysis was performed to assess method agreement. Subgroup analyses on glucose measurement differences per type of maintenance anesthesia were performed.
RESULTS: Thirty-nine patients were included, and 78 measurements were performed. All paired measurements had clinically acceptable agreement with a percentage error of 10.0% (95% CI 8.0 to 11.9). The mean difference (95% limits of agreement) between ACI II and hexokinase for all measurements was 0.0 mmol/L (-0.7 to 0.7 mmol/L). Before induction (n = 39), mean difference was -0.1 mmol/L (-0.6 to 0.4 mmol/L), and after induction (n = 39), mean difference was 0.1 mmol/L (-0.8 to 0.9 mmol/L). Further investigation showed the difference varied per test for patients receiving Sevoflurane compared to patients receiving TIVA (-0.2 ± 0.4 mmol/L vs. 0.4 ± 0.3 mmol/L, p < 0.001). Before and after induction, the difference between ACI II and hexokinase measurements increased for patients receiving Sevoflurane compared to patients receiving TIVA (0.4 ± 0.4 mmol/L vs. -0.4 ± 0.3 mmol/L, p < 0.001).
CONCLUSIONS: The agreement between glucose measurements using ACI II and the reference laboratory hexokinase test was clinically acceptable with a percentage error of 10.0% (95% CI 8.0 to 11.9). The use of TIVA may negatively affect the measurement performance of the ACI II.
摘要：
目的：关于将体外葡萄糖测量干扰化合物转化为体内情况的研究有限。我们通过与参考实验室己糖激酶血浆葡萄糖测试进行比较，研究了AccuChekInformII（ACIII）在全身麻醉手术期间监测葡萄糖浓度是否准确。
方法：纳入全麻手术患者。使用七氟醚或全静脉麻醉（TIVA）维持麻醉。在诱导之前和之后，使用ACIII和己糖激酶试验测量血糖.进行Bland-Altman分析以评估方法一致性。对每种维持麻醉类型的葡萄糖测量差异进行亚组分析。
结果：纳入了39例患者，进行了78次测量。所有配对测量均具有临床上可接受的一致性，误差百分比为10.0％（95％CI8.0至11.9）。所有测量结果的ACIII和己糖激酶之间的平均差(95%一致限度)为0.0mmol/L(-0.7至0.7mmol/L)。诱导前（n=39），平均差为-0.1mmol/L（-0.6至0.4mmol/L），诱导后(n=39)，平均差为0.1mmol/L（-0.8至0.9mmol/L）。进一步的调查显示，与接受TIVA的患者相比，接受七氟醚的患者的每种测试差异（-0.2±0.4mmol/Lvs.0.4±0.3mmol/L，p<0.001)。诱导前后,与接受TIVA的患者相比，接受七氟醚的患者的ACIII和己糖激酶测量值之间的差异增加（0.4±0.4mmol/Lvs.-0.4±0.3mmol/L,p<0.001)。
结论：使用ACIII和参考实验室己糖激酶测试的葡萄糖测量结果之间的一致性在临床上是可接受的，误差百分比为10.0％（95％CI8.0至11.9）。TIVA的使用可能会对ACIII的测量性能产生负面影响。
方法：纳入全麻手术患者。使用七氟醚或全静脉麻醉（TIVA）维持麻醉。在诱导之前和之后，使用ACIII和己糖激酶试验测量血糖.进行Bland-Altman分析以评估方法一致性。对每种维持麻醉类型的葡萄糖测量差异进行亚组分析。
结果：纳入了39例患者，进行了78次测量。所有配对测量均具有临床上可接受的一致性，误差百分比为10.0％（95％CI8.0至11.9）。所有测量结果的ACIII和己糖激酶之间的平均差(95%一致限度)为0.0mmol/L(-0.7至0.7mmol/L)。诱导前（n=39），平均差为-0.1mmol/L（-0.6至0.4mmol/L），诱导后(n=39)，平均差为0.1mmol/L（-0.8至0.9mmol/L）。进一步的调查显示，与接受TIVA的患者相比，接受七氟醚的患者的每种测试差异（-0.2±0.4mmol/Lvs.0.4±0.3mmol/L，p<0.001)。诱导前后,与接受TIVA的患者相比，接受七氟醚的患者的ACIII和己糖激酶测量值之间的差异增加（0.4±0.4mmol/Lvs.-0.4±0.3mmol/L,p<0.001)。
结论：使用ACIII和参考实验室己糖激酶测试的葡萄糖测量结果之间的一致性在临床上是可接受的，误差百分比为10.0％（95％CI8.0至11.9）。TIVA的使用可能会对ACIII的测量性能产生负面影响。
