目的 探讨妊娠期糖尿病（GDM）患者血清微小核糖核酸-15a（miR-15a）、微小核糖核酸-195-5p（miR-195-5p）、微小核糖核酸-33（miR-33）表达及其与母婴结局的关系。 方法 研究对象选自在我院并完成分娩的GDM患者，共纳入208例，病例纳入时间为2021年9月至2024年10月，根据母体妊娠结局将其分为母体结局不良组和母体结局良好组，同时根据围生儿妊娠结局将其分为围生儿结局不良组和围生儿结局良好组。统计母体及围生儿妊娠结局，比较不同母体/围生儿妊娠结局患者临床资料及血清FPG、miR-15a、miR-195-5p、miR-33表达，GDM患者发生母体/围生儿不良妊娠结局的危险因素采用多因素logistic回归分析法进行分析，血清FPG、miR-15a、miR-195-5p、miR-33对GDM患者母体/围生儿妊娠结局的预测价值采用受试者工作特征（ROC）曲线进行分析。 结果 208例GDM患者中有81例发生母体不良妊娠结局，发生率为38.94%。192例围生儿中51例发生围生儿不良妊娠结局，发生率为26.56%。多因素logistic回归分析结果显示，胰岛素抵抗指数高（OR = 2.197）、血糖控制不良（OR = 2.482）、有不良孕产史（OR = 2.838）、血清FPG水平升高（OR = 1.910）、血清miR-15a表达水平升高（OR = 1.988）、血清miR-195-5p表达水平升高（OR = 2.273）、血清miR-33表达水平升高（OR = 2.479）是GDM患者发生母体不良妊娠结局的独立危险因素（P < 0.05）。miR-15a、miR-195-5p、miR-33单一检测及三者联合检测预测GDM患者母体妊娠结局的AUC均高于FPG单一检测，且miR-15a、miR-195-5p、miR-33联合检测的AUC最高（P < 0.05）。多因素logistic回归分析结果显示，胰岛素抵抗指数高（OR = 2.155）、血糖控制不良（OR = 2.408）、有不良孕产史（OR = 2.838）、血清FPG水平升高（OR = 2.018）、血清miR-15a表达水平升高（OR = 2.197）、血清miR-195-5p表达水平升高（OR = 2.246）、血清miR-33表达水平升高（OR = 3.043）是GDM患者发生围生儿不良妊娠结局的独立危险因素（P < 0.05）。miR-15a、miR-195-5p、miR-33单一检测及三者联合检测预测GDM患者围生儿妊娠结局的AUC均高于FPG单一检测，且miR-15a、miR-195-5p、miR-33联合检测的AUC最高（P < 0.05）。 结论 GDM患者母体及围生儿不良妊娠结局发生风险较高，而胰岛素抵抗指数高、血糖控制不良、有不良孕产史及血清miR-15a、miR-195-5p、miR-33表达水平升高均为GDM患者发生母体及围生儿不良妊娠结局的独立危险因素，且miR-15a、miR-195-5p、miR-33单一检测及三者联合检测对GDM患者妊娠结局的预测比FPG单一检测更具优势，三者联合检测的预测价值最高。miR-15a、miR-195-5p、miR-33联合检测对母体及围生儿妊娠结局具有较高预测价值。

Objective To investigate the expression levels of serum microRNA-15a （miR-15a）， microRNA-195-5p （miR-195-5p）， and microRNA-33 （miR-33） in pregnant women with gestational diabetes mellitus （GDM） and to explore their associations with maternal and neonatal outcomes. Methods The research subjects were selected from 208 patients with GDM who completed delivery at our hospital between September 2021 and October 2024. Based on maternal pregnancy outcomes， the subjects were categorized into either the poor maternal outcome group or the good maternal outcome group. Similarly， they were classified into the poor perinatal outcome group and the good perinatal outcome group according to perinatal outcomes. Clinical data and serum levels of fasting plasma glucose （FPG）， miR-15a， miR-195-5p， and miR-33 were collected and compared between groups with different maternal and perinatal outcomes. Multivariate logistic regression analysis was conducted to identify risk factors associated with adverse maternal and perinatal outcomes among GDM patients. Additionally， the predictive value of these biomarkers for adverse maternal and perinatal outcomes was evaluated using receiver operating characteristic （ROC） curve analysis. Results Among 208 patients with GDM， 81 experienced adverse maternal pregnancy outcomes （38.94%）. Of the 192 newborns， 51 cases were associated with adverse perinatal outcomes， resulting in an incidence rate of 26.56%. Multivariate logistic regression analysis revealed that a high insulin resistance index （OR = 2.197）， poor glycemic control （OR = 2.482）， a history of adverse pregnancy or childbirth （OR = 2.838）， elevated FPG levels （OR = 1.910）， increased serum expression of miR-15a （OR = 1.988）， miR-195-5p （OR = 2.273）， and miR-33 （OR = 2.479） were independent risk factors for adverse maternal pregnancy outcomes in GDM patients （P < 0.05）. The area under the receiver operating characteristic curve （AUC） for the combined detection of miR-15a， miR-195-5p， and miR-33 was higher than that of FPG alone in predicting maternal pregnancy outcomes， with the combined detection showing the highest AUC （P < 0.05）. Similarly， multivariate logistic regression results also indicated that elevated levels of the aforementioned biomarkers—insulin resistance index （OR = 2.155）， poor glycemic control （OR = 2.408）， history of adverse pregnancy or childbirth （OR = 2.838）， FPG （OR = 2.018）， miR-15a （OR = 2.197）， miR-195-5p （OR = 2.246）， and miR-33 （OR = 3.043）—were independent risk factors for adverse perinatal outcomes in GDM patients （P < 0.05）. The AUC for combined detection of miR-15a， miR-195-5p， and miR-33 was significantly higher than that of FPG alone in predicting perinatal outcomes， with the combination yielding the highest predictive accuracy （P < 0.05）. Conclusions Patients with GDM are at increased risk of adverse maternal and perinatal outcomes. Elevated insulin resistance index， poor glycemic control， a history of adverse pregnancy or childbirth， and increased serum levels of miR-15a， miR-195-5p， and miR-33 were identified as independent risk factors for adverse pregnancy outcomes in individuals with gestational diabetes. The individual measurement of miR-15a， miR-195-5p， and miR-33， as well as their combined detection， demonstrated greater predictive value for pregnancy outcomes in GDM patients compared to FPG alone， with the combined detection showing the highest predictive accuracy.