Prevalence of gestational diabetes mellitus in mainland China: A systematic review and meta-analysis 中国大陆妊娠糖尿病患病率的系统评价和荟萃分析
Chenghan Gao ^(1){ }^{1}, Xin Sun ^(2**){ }^{2 *} (iD, Li Lu ^(2){ }^{2}, Fangwei Liu ^(2){ }^{2}, Jing Yuan² 高成汉 ^(1){ }^{1} ， 孙 ^(2**){ }^{2 *} 欣 （iD， 陆丽 ^(2){ }^{2} ， 刘 ^(2){ }^{2} 方伟 ， 袁静²^(1){ }^{1} The First Clinical College, Liaoning University of Traditional Chinese Medicine, and ^(2){ }^{2} Department of Endocrinology and Metabolism, The People's Hospital of China Medical University, Shenyang, China ^(1){ }^{1} 辽宁中医药大学第一临床学院， ^(2){ }^{2} 中国医科大学人民医院内分泌代谢科，中国沈阳
Aims/Introduction: Pregnant women with gestational diabetes mellitus (GDM) are at a higher risk of adverse pregnancy outcomes. The aim of the present study was to estimate the pooled prevalence of GDM in mainland China according to International Association of Diabetes and Pregnancy Study Groups criteria. Materials and Methods: We carried out a systematic review by searching both English and Chinese literature databases. Random effects models were used to summarize the prevalence of GDM in mainland China. Subgroup and sensitivity analyses were carried out to address heterogeneity. Publication bias was evaluated using Egger’s test. Results: A total of 25 papers were included in the meta-analysis, involving 79,064 Chinese participants. The total incidence of GDM in mainland China was 14.8% (95% confidence interval 12.8-16.7%). Subgroup analysis showed that the age, bodyweight and family history of diabetes mellitus could significantly increase the incidence of GDM. Conclusions: To the best of our knowledge, this systematic review is the first to estimate the pooled prevalence of GDM among women in mainland China according to International Association of Diabetes and Pregnancy Study Groups criteria. The results of our systematic review suggest a high prevalence of GDM in mainland China, indicating that this country might have the largest number of GDM patients worldwide. 目标/简介： 患有妊娠糖尿病 （GDM） 的孕妇发生不良妊娠结局的风险更高。本研究的目的是根据国际糖尿病和妊娠研究组协会标准估计中国大陆 GDM 的汇总患病率。材料和方法： 我们通过检索英文和中文文献数据库进行了系统评价。采用随机效应模型总结中国大陆 GDM 的患病率。进行亚组和敏感性分析以解决异质性。使用 Egger 检验评估发表偏倚。结果： 荟萃分析共纳入 25 篇论文，涉及 79 064 名中国参与者。中国大陆 GDM 的总发生率为 14.8% （95% 置信区间 12.8-16.7%）。亚组分析显示，年龄、体重和糖尿病家族史可显著增加 GDM 的发生率。结论： 据我们所知，本系统综述是首次根据国际糖尿病与妊娠研究组协会标准估计中国大陆女性 GDM 的汇总患病率。我们的系统评价结果表明，中国大陆的 GDM 患病率很高，这表明该国可能拥有全球最多的 GDM 患者。
INTRODUCTION 介绍
Gestational diabetes mellitus (GDM) is defined as carbohydrate intolerance resulting in hyperglycemia with first onset or detection during pregnancy. GDM is seriously harmful to both the woman and the fetus. Pregnant women and puerperae are prone to complications of gestational hypertensive disease, polyhydramnios, premature rupture of fetal membranes, infection and premature birth; in severe cases, ketoacidosis can occur, and puerperae might have long-term postpartum diabetes ^(1,2){ }^{1,2}. In addition, the fetus is prone to spontaneous abortion, malformation and hypoxia; in severe cases, intrauterine death can occur. Hyperglycemia tends to cause fetal macrosomia; the chances of dystocia at parturition are increased, and the newborn is prone to neonatal respiratory distress syndrome, hypoglycemia and other complications after birth, including death in severe instances ^(3){ }^{3}. 妊娠糖尿病 （GDM） 被定义为碳水化合物不耐受导致高血糖症，并在怀孕期间首次发作或检测到。GDM 对女性和胎儿都严重有害。孕妇和产褥容易出现妊娠高血压病、羊水过多、胎膜早破、感染和早产等并发症;在严重的情况下，可能会出现酮症酸中毒，产褥期可能患有 长期产后糖尿病 ^(1,2){ }^{1,2} .此外，胎儿易发生自然流产、畸形和缺氧;在严重的情况下，可能会发生宫内死亡。高血糖往往会导致巨大胎儿;分娩时难产的机会增加，新生儿出生后容易出现新生儿呼吸窘迫综合征、低血糖和其他并发症，包括严重时 ^(3){ }^{3} 死亡。
In 2008, the hyperglycemia and adverse pregnancy outcome (HAPO) study, which involved multiple countries, showed that 2008 年，涉及多个国家的高血糖和不良妊娠结局 （HAPO） 研究表明，
Received 13 November 2017; revised 25 February 2018; accepted 13 April 2018 2017 年 11 月 13 日接收;修订于 2018 年 2 月 25 日;接受日期 2018 年 4 月 13 日
at 24-32 weeks-of-gestation, a higher blood glucose level in the 75-g oral glucose tolerance test (OGTT) indicates a greater risk of adverse gestational outcomes. Indeed, even with a normal blood glucose level, the risk of having an adverse outcome for both mother and baby is greater with an increase in blood glucose level, whereas significant thresholds were not observed for most comorbidities. Based on that study, the International Association of Diabetes and Pregnancy Study Groups (IADPSG) proposed new GDM diagnostic criteria in 2010: boundary blood glucose levels for fasting, 1 and 2 h after oral glucose of 5.1,10.05.1,10.0 and 8.5mmol//L8.5 \mathrm{mmol} / \mathrm{L}, respectively, by 75-g75-\mathrm{g} OGTT. If any one of these three values reaches or exceeds the boundary level, the patient should be diagnosed with GDM ^(4){ }^{4}. The publication of this diagnostic standard had a “milestone” significance. In 2011, the American Diabetes Association (ADA) recommended the IADPSG criteria be adopted as GDM diagnostic criteria, and in August 2013, the World Health Organization (WHO) used the HAPO study results as an important reference to develop new GDM diagnostic 在妊娠 24-32 周，75 克口服葡萄糖耐量试验 （OGTT） 中较高的血糖水平表明不良妊娠结局的风险更大。事实上，即使血糖水平正常，随着血糖水平的升高，母亲和婴儿出现不良后果的风险也更大，而大多数合并症没有观察到显着的阈值。基于该研究，国际糖尿病和妊娠研究组协会 （IADPSG） 于 2010 年提出了新的 GDM 诊断标准：空腹、口服葡萄糖后 1 小时和 2 小时的边界血糖水平分别为 5.1,10.05.1,10.0 和 8.5mmol//L8.5 \mathrm{mmol} / \mathrm{L} ，通过 75-g75-\mathrm{g} OGTT。如果这三个值中的任何一个达到或超过边界水平，则应诊断为患者患有 GDM ^(4){ }^{4} 。该诊断标准的发布具有“里程碑”意义。2011 年，美国糖尿病协会 （ADA） 建议采用 IADPSG 标准作为 GDM 诊断标准，2013 年 8 月，世界卫生组织 （WHO） 将 HAPO 研究结果作为开发新的 GDM 诊断的重要参考
criteria ^(5){ }^{5}. In 2014, the ADA once again noted that although the new diagnostic criteria would increase healthcare costs, they might also reduce the incidence of adverse gestation events, especially for pregnant women with slightly high blood glucose levels. In October 2015, the International Federation of Gynecology and Obstetrics published a practical guide to GDM, which also utilizes the IADPSG criteria to diagnose GDM ^(6){ }^{6}. 标准 ^(5){ }^{5} 。2014 年，ADA 再次指出，虽然新的诊断标准会增加医疗保健成本，但它们也可能降低不良妊娠事件的发生率，尤其是对于血糖水平略高的孕妇。2015 年 10 月，国际妇产科联合会 （International Federation of Gynecology and Obstetrics） 发布了一份 GDM 实用指南，该指南也利用 IADPSG 标准来诊断 GDM ^(6){ }^{6} 。
As a result of economic development and improvements in living standards, together with increased attention to GDM screening, an increase has been observed in the incidence of GDM. China has a high incidence of diabetes, and the increase in GDM incidence in China is also alarming. Furthermore, China encompasses a vast territory, and has a large population with considerable differences in regions, ethnicities, diets and living habits, and these factors lead to differences in the incidence of GDM reported in various regions. For example, studies have found that even if the IADPSG diagnostic criteria are applied, the incidence of GDM in mainland China fluctuates between 5.12%5.12 \% and 33.3%^(7,8)33.3 \%^{7,8}. As there is currently no systematic analysis of the incidence of GDM in China, the present study aimed to explore the incidence of GDM among pregnant women in mainland China, and the impact of relevant factors on GDM incidence through a systematic meta-analysis. 由于经济发展和生活水平的提高，以及对 GDM 筛查的日益关注，观察到 GDM 的发病率有所增加。中国糖尿病发病率高，中国 GDM 发病率的增加也令人担忧。此外，中国幅员辽阔，人口众多，在地区、民族、饮食和生活习惯方面存在很大差异，这些因素导致不同地区报告的 GDM 发病率存在差异。例如，研究发现，即使采用 IADPSG 诊断标准，中国大陆的 GDM 发病率也在 和 33.3%^(7,8)33.3 \%^{7,8} 之间 5.12%5.12 \% 波动。由于目前尚未对我国孕妇的 GDM 发病率进行系统分析，本研究旨在通过系统的荟萃分析探讨中国大陆孕妇 GDM 的发病率，以及相关因素对 GDM 发病率的影响。
METHODS 方法
A completed Preferred Reporting Items for Systematic Review and Meta-Analyses checklist is presented in Data S1. 完整的系统评价和荟萃分析首选报告项目清单在数据 S1 中介绍。
Search strategy 检索策略
We searched for epidemiological studies on GDM in several electronic databases, including Medline, PubMed, China National Knowledge Infrastructure, Wanfang and Chongqing VIP. Each search strategy is listed as follows. Medline: (TS = gestational diabetes mellitus OR TS = GDM) AND ([TS = prevalence] OR TS = epidemi*) AND ([(TS = Chinese) OR TS = China ]] OR TS = mainland); Pubmed: ([gestational diabetes mellitus(Title/ Abstract)] OR GDM(Title/Abstract)] AND [(prevalence(Title/ Abstract)] OR epidemi*[Title/Abstract]) AND ([(Chinese[Title/ Abstract]) OR China[Title/Abstract]) OR mainland[Title/ Abstract]); China National Knowledge Infrastructure: AB=\mathrm{AB}= gestational diabetes mellitus AND(AB=\mathrm{AND}(\mathrm{AB}= prevalence ORAB=\mathrm{OR} A B= epidemiology); Wangfang: Abstract: (gestational diabetes mellitus)* (prevalence + epidemiology). Chongqing VIP: R=\mathrm{R}= gestational diabetes mellitus* (R=(\mathrm{R}= prevalence +R=+\mathrm{R}= epidemiology). All studies published from 1 January 2010 to 30 April 2017, were searched. In addition, the reference lists of the retrieved articles were examined to identify additional eligible studies. Unpublished studies were not retrieved. The search languages were limited to English and Chinese. 我们在几个电子数据库中检索了 GDM 的流行病学研究，包括 Medline、PubMed、中国国家知识基础设施、万方和重庆 VIP。每种搜索策略的列出方式如下。Medline：（TS = 妊娠糖尿病 OR TS = GDM） 和 （[TS = 患病率] OR TS = 流行病*） AND （[（TS = 中国） OR TS = 中国 ]] OR TS = 大陆）;Pubmed： （[妊娠糖尿病（标题/摘要）] OR GDM（标题/摘要）] AND [（患病率（标题/摘要）] OR 流行病*[标题/摘要]） AND （[（中文[标题/摘要]） OR 中国[标题/摘要]） OR 大陆[标题/摘要]）;中国国家知识基础设施： AB=\mathrm{AB}= 妊娠糖尿病 AND(AB=\mathrm{AND}(\mathrm{AB}= 患病流行 ORAB=\mathrm{OR} A B= 病学）;Wangfang： 摘要：（妊娠糖尿病）*（患病率 + 流行病学）。重庆 VIP： R=\mathrm{R}= 妊娠糖尿病* (R=(\mathrm{R}= 患病率 +R=+\mathrm{R}= 流行病学）。检索了 2010 年 1 月 1 日至 2017 年 4 月 30 日发表的所有研究。此外，还检查了检索到的文章的参考文献列表，以确定其他符合条件的研究。未检索到未发表的研究。检索语言仅限于英文和中文。
Inclusion and exclusion criteria 纳入和排除标准
To satisfy the analysis requirements and to reduce selection deviation, studies needed to meet the following criteria for inclusion: (i) a cross-sectional study or retrospective study 为了满足分析要求并减少选择偏差，研究需要满足以下纳入标准：（i） 横断面研究或回顾性研究
collected in mainland China; (ii) sufficient information on the sample size and crude prevalence of GDM; (iii) GDM diagnostic criteria proposed by IADPSG in 2010^(4)2010^{4}; (iv) containing information for at least family history of diabetes mellitus, body mass index (BMI), age, pregnancy history and delivery history. Studies were excluded if they recruited patients with serious and chronic diseases, including thyroid disease, heart disease and overt diabetes mellitus. In the case of multiple articles based on the same population, only the study reporting the most detailed data was included. 在中国大陆收集;（ii） 关于 GDM 样本量和粗略患病率的足够信息;（iii） IADPSG 在 2010^(4)2010^{4} ;（iv） 包含至少糖尿病家族史、体重指数 （BMI）、年龄、怀孕史和分娩史的信息。如果研究招募了患有严重和慢性疾病的患者，包括甲状腺疾病、心脏病和明显的糖尿病，则被排除在外。在基于同一人群的多篇文章的情况下，仅包括报告最详细数据的研究。
Data extraction and quality assessment 数据提取和质量评估
All searched articles from different electronic databases were combined in Endnote, and duplicates were removed. Two researchers independently screened the title and abstract, and reviewed the full text of eligible citations. In the case of disagreement, a third reviewer made the final decision. For each included study, the two researchers independently extracted the following information: general information (e.g., first author and publication year), study characteristics (including study period, study area and sample size) and all possible participant information (e.g., age, family history of diabetes mellitus, BMI, region etc.). The two researchers independently assessed the quality of each included study using the Newcastle-Ottawa Scale recommended by the Cochrane Handbook for Systematic Reviews of Interventions. 将来自不同电子数据库的所有检索到的文章合并到 Endnote 中，并删除重复的文章。两名研究人员独立筛选了标题和摘要，并审查了合格引文的全文。如果出现分歧，则由第三位审稿人做出最终决定。对于每项纳入的研究，两位研究人员独立提取了以下信息：一般信息（例如，第一作者和出版年份）、研究特征（包括研究期、研究区域和样本量）和所有可能的参与者信息（例如，年龄、糖尿病家族史、BMI、地区等）。两位研究人员使用 Cochrane 干预系统综述手册推荐的 Newcastle-Ottawa 量表独立评估每项纳入研究的质量。
Statistical analysis 统计分析
We used a systematic analysis approach to calculate the pooled prevalence of GDM for all eligible studies. A random effects model was selected to summarize the prevalence of GDM; heterogeneity among studies was assessed using Cochran’s QQ test and the I^(2)I^{2} statistic, which shows the percentage of variation across studies. Subgroup analyses by age, family history of diabetes mellitus, BMI, region and so on were carried out to address heterogeneity. Additionally, sensitivity analysis was carried out to examine the influence of any particular study on the pooled estimate. Publication bias was evaluated using Egger’s test, and independent tt-tests were carried out as appropriate. The significance level was set at a PP-value of < 0.05<0.05. All statistical analyses were carried out using Stata version 12.0 (StataCorp, College Station, TX, USA) and SPSS version 20.0 (SPSS Inc., Chicago, IL, USA). 我们使用系统分析方法来计算所有符合条件的研究的 GDM 汇总患病率。选择随机效应模型来总结 GDM 的患病率;使用 Cochran QQ 检验和 I^(2)I^{2} 统计量评估研究之间的异质性，该统计量显示了研究间差异的百分比。按年龄、糖尿病家族史、 BMI 、地区等进行亚组分析，以解决异质性。此外，还进行了敏感性分析以检查任何特定研究对合并估计的影响。使用 Egger 检验评估发表偏倚，并酌情进行独立 tt 检验。显著性水平设置为 PP -value 。 < 0.05<0.05 所有统计分析均使用 Stata 12.0 版（StataCorp，College Station，TX，USA）和 SPSS 20.0 版（SPSS Inc.，Chicago，IL，USA）进行。
RESULTS 结果
The initial search retrieved 2,576 records from Medline, PubMed, China National Knowledge Infrastructure, Wanfang and Chongqing VIP databases, and 508 articles remained after excluding duplicates, reviews and letters. After screening for eligibility based on the title and abstract, 107 articles were selected; of these, 25 articles were included after screening the full text. The main reasons for inclusion in the full-text selection are shown in Figure 1^(7-31)1^{7-31}. The 25 articles that met the requirements and were eventually included in the study covered 初步检索从 Medline、PubMed、中国国家知识基础设施、万方和重庆 VIP 数据库中检索到 2,576 条记录，在排除重复、评论和信件后仍保留 508 篇文章。根据标题和摘要筛选合格后，选择了 107 篇文章;其中，25 篇文章经筛选全文后纳入。包含在全文选择中的主要原因如图 所示 1^(7-31)1^{7-31} 。符合要求并最终被纳入研究的 25 篇文章涵盖
the prevalence of GDM in pregnant women in 21 regions of mainland China between 2010 and 2017, including 79,064 participants. The characteristics of the selected studies are summarized in Table 1. Among the included articles, 24 focused on women of Han nationality, one involved other ethnic groups and two included a multiple pregnancy. The economic levels of the regions in the included papers had per capita annual incomes ranging from less than $US1,000 to $US30,000, and the papers included age, family history of diabetes mellitus, history of pregnancy and delivery, BMI, per capita income, and many other factors that affect GDM. In accordance with the recommended criteria of the Newcastle-Ottawa Scale, the studies included in the present meta-analysis were of acceptable quality; therefore, we did not exclude any article from the meta-analysis for quality reasons. 2010 年至 2017 年中国大陆 21 个地区孕妇的 GDM 患病率，包括 79,064 名参与者。表 1 总结了所选研究的特征。在纳入的文章中，24 篇关注汉族女性，1 篇涉及其他民族，2 篇涉及多胎妊娠。所收论文中各地区的经济水平人均年收入从不到 1,000 $US 到 30,000 $US不等，论文包括年龄、糖尿病家族史、怀孕和分娩史、BMI、人均收入以及许多其他影响 GDM 的因素。根据纽卡斯尔-渥太华量表的推荐标准，本荟萃分析中纳入的研究质量可接受;因此，出于质量原因，我们没有将任何文章排除在meta分析之外。
The total incidence of GDM in mainland China was 14.8% ( 95%95 \% confidence interval [CI] 12.8-16.7%; Figure 2). Table 2 中国大陆 GDM 的总发病率为 14.8% （ 95%95 \% 置信区间 [CI] 12.8-16.7%;图 2）。表 2
shows the results of subgroup analysis in different groups. Subgroup analysis showed an incidence of GDM in older pregnant women of 26.7%26.7 \% ( 95%95 \% CI 23.2-30.3%), whereas that in younger pregnant women was just 13.4% (95% CI 11.0-15.7%), with a significant difference between the two subgroups (P < 0.01)(P<0.01). The incidence of GDM in overweight or obese women was 30.3%30.3 \% ( 95%95 \% CI 25.9-34.7%), which was significantly higher than that of women who had a normal bodyweight ( 14.9%,95%14.9 \%, 95 \% CI 11.7-18.1%; P < 0.01P<0.01 ). The incidence of GDM in women with a family history of diabetes mellitus was 32.9% (95% CI 27.5-38.4%), approximately threefold that in women without a family history ( P < 0.01P<0.01 ). Using the per capita income of $US10,000 as a boundary, the regional economic level did not have a significant impact on the incidence of GDM ( 14.8%14.8 \% and 15.4%,P=0.5315.4 \%, P=0.53 ). We carefully and comprehensively searched the articles in the database. Sensitivity analysis was carried out to examine the influence of any particular 显示了不同组中的子组分析结果。亚组分析显示，老年孕妇的 GDM 发生率为 26.7%26.7 \% （ 95%95 \% CI 23.2-30.3%），而年轻孕妇的 GDM 发生率仅为 13.4% （95% CI 11.0-15.7%），两个亚组 (P < 0.01)(P<0.01) 之间存在显著差异。超重或肥胖女性 GDM 的发生率为 30.3%30.3 \% （ 95%95 \% CI 25.9-34.7%），显著高于体重正常的女性 （ 14.9%,95%14.9 \%, 95 \% CI 11.7-18.1%; P < 0.01P<0.01 ）。有糖尿病家族史的女性的 GDM 发病率为 32.9% （95% CI 27.5-38.4%），大约是无家族史女性的三倍 （ P < 0.01P<0.01 ）。以 10,000 $US的人均收入为界，区域经济水平对 GDM 的发病率没有显著影响 （ 14.8%14.8 \% 和 15.4%,P=0.5315.4 \%, P=0.53 ）。我们仔细全面地检索了数据库中的文章。进行了敏感性分析以检查任何特定
Figure 1\mathbf{1} | Flow chart showing the detailed procedure for the inclusion or exclusion of studies. 图 1\mathbf{1} |显示纳入或排除研究的详细程序的流程图。
