Zhao Tong, Yang Qize, Feuerbacher Joshua F, Yu Bizhu, Brinkmann Christian, Cheng Sulin, Bloch Wilhelm, Schumann Moritz
Department of Molecular and Cellular Sports Medicine, German Sport University Cologne, Cologne, Germany.
Exercise Translational Medicine Centre, Shanghai Centre for Systems Biomedicine Shanghai, Shanghai Jiao Tong University, Shanghai, Shanghai, China.
Br J Sports Med. 2024 Dec 2;58(23):1452-1460. doi: 10.1136/bjsports-2024-108127.
To compare the efficacy of exercise, metformin and their combination on glucose metabolism in individuals with abnormal glycaemic control.
Systematic review and network meta-analysis.
Embase, Web of Science, PubMed/MEDLINE and SPORTDiscus.
Randomised controlled trials involving exercise, metformin or their combined treatments in individuals with prediabetes or type 2 diabetes mellitus (T2DM) were included. Outcomes included haemoglobin A1c (HbA1c), 2-hour glucose during oral glucose tolerance test, fasting glucose, fasting insulin and homeostasis model assessment of insulin resistance (HOMA-IR).
407 articles with 410 randomised controlled trials (n=33 802) were included. In prediabetes, the exercise showed greater efficacy than metformin on HbA1c levels (mean difference -0.16%, 95% CI (-0.23 to -0.09) vs -0.10%, 95% CI (-0.21 to 0.02)), 2-hour glucose (-0.68 mmol/L, 95% CI (-0.97 to -0.39) vs 0.01 mmol/L, 95% CI (-0.38 to 0.41)) and HOMA-IR (-0.54, 95% CI (-0.71 to -0.36) vs -0.23, 95% CI (-0.55 to 0.10)), while the efficacy on fasting glucose was comparable (-0.26 mmol/L, 95% CI (-0.32 to -0.19) vs -0.33 mmol/L, 95% CI (-0.45 to -0.21)). In T2DM, metformin was more efficacious than exercise on HbA1c (-0.88%, 95% CI (-1.07 to -0.69) vs -0.48%, 95% CI (-0.58 to -0.38)), 2-hour glucose (-2.55 mmol/L, 95% CI (-3.24 to -1.86) vs -0.97 mmol/L, 95% CI (-1.52 to -0.42)) and fasting glucose (-1.52 mmol/L, 95% CI (-1.73 to -1.31) vs -0.85 mmol/L, 95% CI (-0.96 to -0.74)); exercise+metformin also showed greater efficacy in improving HbA1c (-1.23%, 95% CI (-2.41 to -0.05)) and fasting glucose (-2.02 mmol/L, 95% CI (-3.31 to -0.74)) than each treatment alone. However, the efficacies were modified by exercise modality and metformin dosage.
Exercise, metformin and their combination are efficacious in improving glucose metabolism in both prediabetes and T2DM. The efficacy of exercise appears to be superior to metformin in prediabetes, but metformin appears to be superior to exercise in patients with T2DM.
CRD42023400622.
比较运动、二甲双胍及其联合应用对血糖控制异常个体糖代谢的疗效。
系统评价和网状Meta分析。
Embase、Web of Science、PubMed/MEDLINE和SPORTDiscus。
纳入涉及糖尿病前期或2型糖尿病（T2DM）个体的运动、二甲双胍或其联合治疗的随机对照试验。结局指标包括糖化血红蛋白（HbA1c）、口服葡萄糖耐量试验中的2小时血糖、空腹血糖、空腹胰岛素和胰岛素抵抗稳态模型评估（HOMA-IR）。
纳入407篇文章，共410项随机对照试验（n=33802）。在糖尿病前期，运动在降低HbA1c水平（平均差值-0.16%，95%可信区间（-0.23至-0.09）vs -0.10%，95%可信区间（-0.21至0.02））、2小时血糖（-0.68 mmol/L，95%可信区间（-0.97至-0.39）vs 0.01 mmol/L，95%可信区间（-0.38至0.41））和HOMA-IR（-0.54，95%可信区间（-0.71至-0.36）vs -0.23，95%可信区间（-0.55至0.10））方面比二甲双胍更有效，而在空腹血糖方面疗效相当（-0.26 mmol/L，95%可信区间（-0.32至-0.19）vs -0.33 mmol/L，95%可信区间（-0.45至-0.21））。在T2DM中，二甲双胍在降低HbA1c（-0.88%，95%可信区间（-1.07至-0.69）vs -0.48%，95%可信区间（-0.58至-0.38））、2小时血糖（-2.55 mmol/L，95%可信区间（-3.24至-1.86）vs -0.97 mmol/L，95%可信区间（-1.52至-0.42））和空腹血糖（-1.52 mmol/L，95%可信区间（-1.73至-1.31）vs -0.85 mmol/L，95%可信区间（-0.96至-0.74））方面比运动更有效；运动+二甲双胍在改善HbA1c（-1.23%，95%可信区间（-2.41至-0.05））和空腹血糖（-2.02 mmol/L，95%可信区间（-3.31至-0.74））方面也比单独使用每种治疗方法更有效。然而，疗效会因运动方式和二甲双胍剂量而有所改变。
运动、二甲双胍及其联合应用在改善糖尿病前期和T2DM患者的糖代谢方面均有效。运动在糖尿病前期的疗效似乎优于二甲双胍，但在T2DM患者中二甲双胍似乎优于运动。
PROSPERO注册号：CRD42023400622。
