PURPOSE To provide a systematic review and meta-analysis synthesizing the findings of randomized controlled trials (RCTs) of continuous glucose monitors (CGMs) in the management of adults with type 2 diabetes mellitus (T2DM) on glucose control and clinical outcomes. METHODS MEDLINE, Embase, and Cochrane were searched for RCTs that assessed the effectiveness of real-time CGM (rt-CGM) or flash CGM (FGM) in adults (≥18 years) with T2DM that reported on at least 1 of the following outcomes: hemoglobin A1c (HbA1c), time in range, time in hyperglycemia, or time in hypoglycemia. The GRADE approach was used to assess certainty of evidence for primary outcomes. RESULTS Fourteen RCTs assessing CGM were included, with 825 patients in 9 RCTs using rt-CGM and 822 in 5 RCTs using FGM. Moderate certainty of evidence indicated that use of CGM had a modest but statistically significant reduction in HbA1c levels of about 0.32%. Our analyses of each device type separately showed similar reductions in HbA1c (0.34% and 0.33%, respectively, for rt-CGM and FGM), with trends for improvement in other glucose metrics favoring rt-CGM over self-monitored blood glucose. CONCLUSION Both rt-CGM and flash CGM led to modest but statistically significant declines in HbA1c among individuals with T2DM, with little heterogeneity in the results. However, the duration of the included RCTs was relatively short and few studies reported on important clinical outcomes, such as adverse events, emergency department use, or hospitalization. Longer term studies are needed to determine if the short-term improvements in glucose control leads to improvements in clinically important outcomes.

中文翻译：

---

连续血糖监测对 2 型糖尿病血糖控制指标的有效性：随机对照试验的系统回顾和荟萃分析。

目的 提供系统评价和荟萃分析，综合连续血糖监测仪 (CGM) 在成人 2 型糖尿病 (T2DM) 管理中的血糖控制和临床结果的随机对照试验 (RCT) 的结果。方法 检索了 MEDLINE、Embase 和 Cochrane 的随机对照试验，评估实时 CGM (rt-CGM) 或快速 CGM (FGM) 对患有 T2DM 的成人（≥18 岁）的有效性，并报告至少 1 个以下结果：血红蛋白 A1c (HbA1c)、范围内时间、高血糖时间或低血糖时间。GRADE 方法用于评估主要结局证据的确定性。结果 纳入了 14 项评估 CGM 的随机对照试验，其中 9 项使用 rt-CGM 的随机对照试验纳入了 825 名患者，5 项使用 FGM 的随机对照试验纳入了 822 名患者。中等质量的证据表明，使用 CGM 可使 HbA1c 水平略有降低，但具有统计显着性，降低幅度约为 0.32%。我们对每种设备类型的分析分别显示 HbA1c 的降低程度相似（rt-CGM 和 FGM 分别为 0.34% 和 0.33%），与自我监测血糖相比，其他血糖指标的改善趋势有利于 rt-CGM。结论 rt-CGM 和快速 CGM 均导致 T2DM 患者的 HbA1c 略有下降，但具有统计显着性，结果异质性很小。然而，纳入的随机对照试验的持续时间相对较短，并且很少有研究报告重要的临床结局，例如不良事件、急诊室使用或住院治疗。需要进行更长期的研究来确定血糖控制的短期改善是否会导致临床重要结果的改善。