| 摘要: | 背景：
孕期首次出現血糖調節失衡被定義為妊娠糖尿病（Gestational Diabetes Mellitus, GDM），國際糖尿病聯合會（International Diabetes Federation, IDF）估計全球每六為胎兒就有一位受妊娠高血糖影響，其中GDM佔83.6%，相當於14%盛行率。聚焦臺灣，在12年間GDM增加1.8倍（從7.6%至13.4%），相關研究也證實GDM影響著母體自孕期、生產或胎兒出生、成長等過程的健康風險，例如母親子癲前症、剖腹產及日後糖尿病的發生率，也使胎兒呼吸窘迫、動脈粥狀硬化、心臟疾病、巨嬰症及成長受阻的風險增高。肥胖、疾病史、孕期體重過度增加或生活習慣不佳皆會增加風險，臨床研究更從這些面向發展出許多預防方案，調整飲食、運動提升、改變生活型態、營養素或藥物補充等似乎有助於降低發生風險，然過去的回顧性文獻、統合分析結果並不一致亦尚無定論，被認為與納入的族群差異、介入型態有關。一篇網絡統合分析（Network Meta-Analysis, NMA）顯示藥物、運動、益生菌或維生素D等任一介入不足以有效降低GDM發生。爾後另一則考科藍（Cochrane）綜論卻認為運動飲食結合、肌醇、維生素D及二甲雙胍（Metformin）具有降低GDM發生的潛在益處。
有鑒於此議題中仍有許多待解決的困境，且目前尚未有囊括調整飲食等比較多種介入的網絡統合分析，若能統整現有醫學證據中飲食、運動、口服補充品或藥物作為預防妊娠糖尿病之效益，並評估是否須於備孕期即進行實施，對於母親及孕育下一代的健康與公共衛生健康人員擬定策略將能有所貢獻。
目的：
以系統性文獻回顧及網絡統合分析探討飲食、運動、口服補充品或藥物之隨機臨床試驗介入是否有助於降低備孕或孕期婦女發生妊娠糖尿病。
方法：
針對資料庫PubMed, Embase, Medline, Cochrane和CINAHL進行系統性文獻查找，以關鍵字搜尋，納入有孕期妊娠糖尿病發生數據的研究為主。執行研究篩選、資料提取及研究偏差風險評估。本研究分析主要是將納入的隨機試驗文章進行分析以及不同篩檢方式與高風險族群之次分析，進一步探討是否有直接與間接證據所致之混合效應，並試圖釐清介入的有效貢獻程度。以Stata統計軟體16.1版，進行頻率架構的隨機效應模型網絡統合分析模組。
結果：
總收錄98篇研究，納入41,134名女性，年齡平均30.1±5.3歲，BMI平均27.2±5.9，共區分出16種介入方案，可彙整成6組（飲食、運動、藥物、口服營養品及標準婦產程序），統合分析結果如下：相比於基本照護，肌醇、維生素D、飲食及運動皆顯著降低GDM發生率，風險比分別為0.48（95%信賴區間0.37-0.63）、0.51（95%信賴區間0.34-0.76）、0.75（95%信賴區間0.63-0.90）及0.76（95%信賴區間0.64-0.90），於彙整後的6組別中，除藥物外，皆有顯著降低GDM發生之效益。進一步針對高GDM風險婦女進行次族群分析時，介入肌醇、維生素D的顯著性仍存在，而以診斷標準（International Association of Diabetes in Pregnancy Study Group, IADPSG）一致的研究進行次族群分析後，肌醇、飲食調整之顯著性保留較高。在孕期體重控制方面，以飲食調整與基本照護相差約2.17公斤（95%信賴區間0.95-3.39）最為顯著。
結論：
透過本研究的發現，其實當前有許多針對預防的方法被發展，且此次透過我們整合後的數據也再度支持“在孕期甚至孕前應考慮給予預防的策略”，能有助於預防GDM發生。介入需考量個人化，口服補充劑可能是最簡單執行的方式，例如針對高風險族群給予肌醇補充，能顯著降低GDM發生。而飲食調整針對廣納的族群來說不僅降低GDM，更有利於控制孕期體重增加。
Introduction
Gestational diabetes mellitus (GDM) is currently the most common medical complication of pregnancy, which is associated with negative consequences for both the mother and the offspring. Prevention methods might develop from a modifiable risk factor, and striving towards optimal care for women before and during pregnancy is a necessary component of health and well-being.
Objective
To determine the effectiveness of diet, exercise, oral supplementation, and medication on the prevention of GDM among preconception or conception women.
Design
A systematic review and network meta-analysis of randomized controlled trials (RCTs).
Data sources
Electronic databases (PubMed, Medline, Embase, Cochrane and CINAHL) from database inception until Jan 2022 and reference of eligible trials, without language restrictions.
Study selection
To search the literature for RCTs comparing diet, exercise program, oral supplementation, or treatment with medication with usual obstetrics care, in women who were pregnant or planning a pregnancy and the results of the full-text discussed incidence of GDM.
Outcomes and measures
The primary outcome was the incidence of GDM, secondary outcomes was maternal gestational weight gain (GWG).
Review methods
Two reviewers independently extracted data and population description, interventions methods, GDM criteria, trials duration, and assessed risk of bias were involved. All meta-analyses were undertaken using Stata software version 16.1 (StataCorp, College Station, TX) and produced presentation tools with the network graphs package.
Results
The 98 RCTs, which included 41,134 women with a mean age of 30.1 ± 5.3 years, mean body mass index (BMI) of 27.2 ± 5.9. The eligible trials were aggregated to 16 options, and incorporated them into 6 groups. Forty-one (41.8%) studies were judged at low risk of bias in all domains. Most of reasons for high risk are unclear or not apparent for the participants, investigators, and outcome assessors. Otherwise, the outcome measurements make no mention or are inconsistent.
Compared with usual care, inositol significantly lowered the risk ratio (RR) by 0.48 (95% confidence interval (CI) 0.37 to 0.63), followed by vitamin D with RR 0.51 (95% CI 0.34 to 0.76), diet RR 0.75 (95% CI 0.63 to 0.90) and exercise RR 0.76 (95% CI 0.64 to 0.90). While study included only high-GDM risk participants (64.3% trials) or filtering the studies using the diagnostic methods equivalent to International Association of Diabetes in Pregnancy Study Group (IADPSG) criteria (46.9% trials), the surface under the cumulative ranking curve (SUCRA) ranking was the same as the focal result of the main analysis. Among 60.2% of the research RCTs provided the outcome value of weight gain during pregnancy, compared with usual care, diet had the relatively lowest difference in weight gain during pregnancy, among which the diet mean difference (MD) 2.17kg (95% CI 0.95 to 3.39).
Conclusions
Intervention needs to consider individualization, and oral supplementations may be the easiest way to implement the prevention of GDM. For high-risk groups, the administration of inositol significantly has the highest effect. However, for a wide range of the public, dietary adjustment is worthwhile in enhancing the benefits of GWG control, not just for reducing the risk of GDM. |
