非药物干预对中国糖尿病前期人群相关指标的影响:网状Meta分析

Effects of non-pharmacological interventions on relevant indicators of pre-diabetes in China: a network Meta-analysis

目的采用网状Meta分析评价饮食干预、运动干预、健康教育对我国糖尿病前期人群的干预效果。方法计算机检索CNKI、万方、维普、PubMed、Springer Link、Proquest、Highwire press、Uptodate以及Cochrane数据库,纳入2017年1月前中国有关糖尿病前期饮食干预、运动干预、健康教育的随机对照研究,以患病率、空腹血糖、餐后2 h血糖为研究指标,采用ADDIS 1.16.6软件进行数据分析。计数资料采用比值比、计量资料采用标准化均数差及其95%CI为疗效分析统计量。应用排序概率图对各干预措施结果进行直观评估。结果共纳入14项研究。3个结局指标的潜在的标尺缩减参数(PSRF)均接近1,收敛性良好,非一致性检验均无统计学意义(P>0.05),一致性良好。饮食干预较健康教育、运动干预患病率更低,OR值(95%CI)为1.44(0.20,11.69)、1.94(0.42,8.82);饮食干预较健康教育、运动干预的空腹血糖更低,标准化均数差(SMD)(95%CI)分别为0.07(-0.43,0.52),0.32(-0.38,0.94);健康教育较饮食干预、运动干预餐后2 h血糖更低,SMD(95%CI)分别为0.62(-0.91,2.19),0.81(-1.70,3.44)。排序概率图结果显示,患病率方面饮食干预排序优于健康教育,空腹血糖控制效果排序依次为饮食干预、健康教育、运动,餐后2 h血糖控制效果健康教育排序优于饮食干预。结论医务人员在进行针对糖尿病前期患者人群的干预时饮食干预的效果可能更好。

ObjectiveTo evaluate the intervention effects of diet, exercise and health education on pre-diabetes in China by network meta-analysis. MethodsA systematic search of CNKI, Wan-Fang, VIP, PubMed, Springer Link, Proquest, Highwire press, Uptodate, Cochrane database was carried out. Randomized control trials examining the influence of non-drug intervention on pre-diabetes in China were included. The data of prevalence rate, fasting plasma glucose (FPG), 2-hour postprandial plasma glucose (PPG) were analyzed by ADDIS 1.16.6. The count data were used the odds ratio (OR) and its 95% CI, and the measurement data were used the standardized mean difference (SMD) and its 95%CI as the efficacy analysis statistics. The results of each intervention were evaluated visually using rank probability.ResultsA total of 14 studies were included. The potential scale reduced factor of all the three outcomes was close to 1 in the presence of good convergence and consistence (P>0.05). Dietary intervention was better than health education and exercise intervention in prevalence rate[OR(95%CI): 1.44(0.20, 11.69) and 1.94(0.42,8.82), respectively]. Dietary intervention was better than health education and exercise intervention in FPG[SMD(95% CI): 0.07(-0.43, 0.52) and 0.32(-0.38, 0.94), respectively]. Health education was better than dietary intervention and exercise intervention in PPG[SMD(95% CI): 0.62(-0.91, 2.19) and 0.81(-1.70, 3.44), respectively]. The rank probability showed that dietary intervention was better than others in prevalence rate and FPG, health education was better than others in 2PPG.ConclusionMedical intervention may focus on dietary intervention among pre-diabetes in China.