Abstract:

Objective The purpose of this systematic review was to evaluate the effect of different lengths of use of continuous glucose monitoring（CGM） system on glycaemic control in patients with diabetes,compared with routine nursing or self-monitoring of blood glucose（SMBG）. Methods 8 Chinese and English databases,such as PubMed,Web of Science,Cochrane Library and Embase,were searched to include randomized controlled trials on the effects of CGM on glycated hemoglobin（HbA1c）,time above range（TAR）,time in range（TIR）,time below range（TBR） and mean glucose level（MG） in diabetic patients. The randomised controlled trials were collected from inception until August 2021,and the meta-analysis was performed using RevMan 5.4 and Stata 16 software. Results A total of 11 publications were included,including 1 745 diabetic patients. The >16 weeks[MD=-0.23,95%CI（-0.35,-0.11）,P<0.001] and ≤16 weeks[MD=-0.40,95%CI（-0.42,-0.38）,P<0.001] glycated hemoglobin levels in diabetic patients using CGM,>16 weeks[MD=91.53,95%CI（65.77,117.28）,P<0.001] and ≤16 weeks[MD=128.68,95%CI（112.95,144.42）,P<0.001] glucose time in range,>16 weeks[MD=-45.10,95%CI（-76.34,-13.87）,P=0.005] and ≤16 weeks[MD=-80.37,95%CI（-96.05,-64.69）,P<0.001] glucose time above range,>16 weeks[MD=-31.83,95%CI （-44.11,-19.54）,P<0.001] and ≤16 weeks[MD=-20.90,95%CI（-37.45,-4.35）,P=0.010] glucose time below range,and ≤16 weeks mean glucose levels[MD=-0.40,95%CI（-0.54,-0.26）,P<0.001] have statistically significant differences but no statistically significant difference for >16 weeks mean glucose levels[MD=-0.22,95%CI（-0.48,0.04）,P=0.090]. Conclusion CGM was able to reduce HbA1c,increase TIR,and reduce TBR and TAR to improve glycaemic control in diabetic patients at different durations of use. As the duration of use increased,the improvement in TBR was more significant,while the improvement in HbA1c,TIR,TAR and MG gradually decreased,but the effect on long-term MG improvement was not significant,and more high-quality studies are still needed to further prove this.

Key words: Blood Glucose Monitoring, Diabetes, Implantable CGM System, Continuous Blood Glucose Monitoring, Meta-Analysis

摘要：

目的 评价连续血糖监测（continuous glucose monitoring,CGM）与常规护理或自我血糖监测（self-monitoring of blood glucose,SMBG）相比,不同使用时长对糖尿病患者血糖控制的效果。 方法 检索PubMed、Web of Science、Cochrane Library、Embase、中国知网、万方数据库、维普数据库、中国生物医学文献数据库中有关CGM对糖尿病患者糖化血红蛋白（HbA1c）、血糖高于范围时间（time above range,TAR）、目标血糖范围内时间（time in range,TIR）、血糖低于范围时间（time below range,TBR）和平均葡萄糖水平（mean glucose,MG）干预效果的随机对照试验,检索时限为建库至2021年8月。由2名研究者独立进行文献筛选、资料提取、文献质量评价后,采用RevMan 5.4及Stata 16软件进行Meta分析。 结果 共纳入11篇文献,包括1 745例糖尿病患者。糖尿病患者使用CGM>16周[MD=-0.23,95%CI（-0.35,-0.11）,P<0.001]及≤16周[MD=-0.40,95%CI（-0.42,-0.38）,P<0.001]对HbAlc水平,使用CGM>16周[MD=91.53,95%CI（65.77,117.28）,P<0.001]及≤16周[MD=128.68,95%CI（112.95,144.42）,P<0.001]对TIR,使用CGM>16周[MD=-45.10,95%CI（-76.34,-13.87）,P=0.005]及≤16周[MD=-80.37,95%CI（-96.05,-64.69）,P<0.001]对TAR,使用CGM>16周[MD=-31.83,95%CI（-44.11,-19.54）,P<0.001]及≤16周[MD=-20.90,95%CI（-37.45,-4.35）,P=0.010]对TBR,使用CGM≤16周对MG[MD=-0.40,95%CI（-0.54,-0.26）,P<0.001]的影响,差异有统计学意义;使用CGM>16周对MG的影响,差异无统计学意义[MD=-0.22,95%CI（-0.48,0.04）,P=0.090]。 结论 CGM不同使用时长均能够降低糖尿病患者的HbA1c,升高TIR并降低TBR、TAR,改善血糖控制。随着时长的增加,TBR的改善更为明显,HbA1c、TIR、TAR、MG的改善效果逐渐降低,MG改善长期效果不显著,仍需要更多高质量的研究进一步论证。

关键词: 血糖监测, 糖尿病, 植入式CGM系统, 连续血糖监测系统, Meta分析