新生儿低血糖风险预测模型的系统评价与 Meta 分析

打开文本图片集

中图分类号：R722

文献标识码：A

DOI：10.3969/j.issn.1006-1959.2026.09.011

文章编号：1006-1959（2026）09-0072-07

Systematic Review and Meta-analysis of Neonatal Hypoglycemia Risk Prediction Models

NIE Zhuomiao 1 ， WANG Guorong 2 ， LENG Yingjie 1 ， LI Chengxiang 1 ， XIE Ruonan 1 ， JIANG Xin 1

（1. College of Nursing， Chengdu University of Traditional Chinese Medicine， Chengdu ， Sichuan， China;

2. Department of Nursing， West China School of Public Health， Sichuan University/West China Fourth Hospital，

Sichuan University， Chengdu ， Sichuan， China）

Abstract： Objective To systematically evaluate the risk prediction model of neonatal hypoglycemia. Methods CNKI， Wanfang， VIP， PubMed， Web of Science， Embase， The Cochrane Library databases were searched from the establishment of the database to December 1， 2024. The literature will be included in the systematic evaluation of the neonatal hypoglycemia risk prediction model with the prediction model bias risk assessment tool （PROBAST）. Meta-analysis of neonatal hypoglycemia predictors was performed using RevMan5.3 software. Results A total of 16 risk prediction models were included. The area under the receiver operating characteristic curve （AUC） for these models ranged from 0.657 to 0.944， with 15 models achieving an AUC>0.7 and an average AUC of 0.836. The PROBAST evaluation indicated that all included studies were at high risk of bias. Meta-analysis results identified preterm birth， advanced maternal age， twins， cesarean section， small-for-gestational-age infants， low birth weight neonates， neonatal hypothermia， gestational diabetes mellitus， gestational hypertension as significant predictors of neonatal hypoglycemia （P<0.05）. Conclusion The overall performance of risk prediction models for neonatal hypoglycemia is relatively good; however， high risk of bias and issues with data sources and model performance limit their practical application.

Key words： Neonate; Hypoglycemia; Prediction model; Systematic review; Meta-analysis

新生儿低血糖（neonatal hypoglycemia， NH）是新生儿期最常见的代谢异常之一，指血糖水平低于新生儿正常生理需求的下限[1]。（剩余9638字）