远程医疗干预对妊娠期糖尿病孕妇母胎结局的影响：随机对照试验的系统评价与荟萃分析。
完整摘要 · 中文译文
背景：妊娠期糖尿病（GDM）的全球患病率不断上升，给母亲和胎儿带来短期和长期的不良风险。传统的面对面GDM管理受时间和地域限制，常导致依从性不佳和个体化干预延迟。远程医疗越来越多地应用于GDM管理；然而，随机对照试验（RCT）的证据仍不一致。 目的：本研究旨在汇总RCT数据，量化远程医疗对GDM妇女血糖控制和妊娠相关结局的影响，并为优化临床管理提供依据。 方法：我们检索了PubMed、Embase、Cochrane Library、Web of Science和Scopus，检索时间截至2026年8月16日。两名 reviewers（评审者）独立进行文献筛选和数据提取。使用RoB 2评估偏倚风险。Meta分析在Stata 18.0中进行。对于异质性结局，计算95%预测区间（PI）以估计未来研究的效应范围。通过逐一剔除敏感性分析、meta回归和亚组分析探讨异质性。使用GRADE（推荐意见分级评估、制定和评价）框架评定证据确定性。对≥10项研究报告的结局评估发表偏倚。 结果：共纳入23项RCT（3055例GDM妇女）。合并估计表明，远程医疗可能降低空腹血糖（平均差[MD] -0.26，95% CI -0.42至-0.09，95% PI -0.93至0.42；P=.002）、餐后2小时血糖（MD -0.50，95% CI -0.87至-0.14，95% PI -1.84至0.84；P=0.007），以及孕期增重（MD -0.78，95% CI -1.40至-0.17，95% PI -2.83至1.26；P=0.01）。远程医疗还与较低的剖宫产、急诊剖宫产、巨大儿、新生儿低血糖和新生儿呼吸窘迫风险相关。糖化血红蛋白、早产或新生儿重症监护病房入住率未发现显著差异。值得注意的是，空腹血糖、餐后2小时血糖和孕期增重表现出极高的未解释异质性（I²>90%），其95%预测区间跨越零值，尽管合并效应具有统计学显著性。所有评估结局的GRADE证据确定性为极低至低。 结论：远程医疗可能改善妊娠期糖尿病（GDM）女性的部分血糖和围产期结局，并可能作为产前保健的一种有前景的补充选择，尤其是在常规面对面随访难以实施的资源有限环境中。然而，由于极端且未解释的研究间异质性及95%预测区间跨越零值，血糖控制和孕期增重的表面获益仍属初步结论。基于现有证据，远程医疗不应完全替代标准产前保健。未来需要设计良好、大样本的随机对照试验（RCT）；此类试验应采用标准化的远程医疗干预定义，完整报告干预组成部分，并进行长期随访以验证母体和新生儿临床获益的持久性。
完整摘要 · 英文原文
BACKGROUND: Gestational diabetes mellitus (GDM) has a rising global prevalence and brings short-term and long-term adverse risks for mothers and fetuses. Conventional in-person GDM care is limited by time and geographic barriers, often resulting in suboptimal compliance and delayed individualized interventions. Telemedicine is increasingly applied in GDM management; however, randomized controlled trial (RCT) evidence remains inconsistent. OBJECTIVE: The study aims to pool data from RCTs to quantify the effects of telemedicine on glycemic control and pregnancy-related outcomes in women with GDM and to inform optimized clinical management. METHODS: We searched PubMed, Embase, Cochrane Library, Web of Science, and Scopus up to August 16, 2026. Two reviewers independently performed study screening and data extraction. Risk of bias was assessed using RoB 2. Meta-analyses were conducted in Stata 18.0. For heterogeneous outcomes, 95% prediction intervals (PIs) were calculated to estimate effect ranges for future studies. Heterogeneity was explored via leave-one-out sensitivity, meta-regression, and subgroup analyses. The GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework rated evidence certainty. Publication bias was evaluated for outcomes reported in ≥10 studies. RESULTS: Twenty-three RCTs (3055 women with GDM) were included. Pooled estimates suggested that telemedicine may reduce fasting glucose (mean difference [MD] -0.26, 95% CI -0.42 to -0.09, 95% PI -0.93 to 0.42; P=.002), 2-hour postprandial glucose (MD -0.50, 95% CI -0.87 to -0.14, 95% PI -1.84 to 0.84; P=.007), and gestational weight gain (MD -0.78, 95% CI -1.40 to -0.17, 95% PI -2.83 to 1.26; P=.01). Telemedicine was also associated with lower risks of cesarean section, emergency cesarean section, macrosomia, neonatal hypoglycemia, and neonatal respiratory distress. No significant differences were found for hemoglobin A1c, preterm birth, or neonatal intensive care unit admission. Notably, fasting glucose, 2-hour postprandial glucose, and gestational weight gain exhibited very high unexplained heterogeneity (I²>90%), with their 95% PIs crossing the null value despite statistically significant pooled effects. The GRADE certainty of evidence was very low to low across all assessed outcomes. CONCLUSIONS: Telemedicine may improve selected glycemic and perinatal outcomes for women with GDM and may serve as a promising supplementary option for prenatal care, particularly in resource-limited settings where routine in-person follow-up is difficult to implement. Nevertheless, the apparent benefits for glycemic control and gestational weight gain remain tentative due to extreme, unexplained between-study heterogeneity and 95% PIs crossing the null. Based on current evidence, telemedicine should not completely replace standard antenatal care. Future well-designed, large-sample RCTs are warranted; such trials should adopt standardized definitions for telemedicine interventions, fully report intervention components, and conduct long-term follow-up to verify sustained maternal and neonatal clinical benefits.