极早产儿中/重度支气管肺发育不良发生风险的预测模型构建OACSTPCD
Development of a predictive nomogram for the risk of moderate-to-severe bronchopulmonary dysplasia in ex-tremely preterm infants
目的 分析胎龄＜32周极早产儿发生中/重度支气管肺发育不良(bronchopulmonary dysplasia,BPD)的危险因素,并构建中/重度BPD的早期列线图预测模型.方法 回顾性选取2019年1月至2020年12月期间在广州医科大学附属第三医院出生的胎龄＜32周且发生BPD的极早产儿作为建模组,收集围生期资料以及患儿住院期间疾病信息,通过logistic回归分析探讨极早产儿发生中/重度BPD的危险因素,并据此建立列线图预测模型,采用Bootstrap法进行模型内部验证.根据时间序列选取2021年1月至2021年6月在广州医科大学附属第三医院出生且发生BPD的极早产儿作为外部验证组,其纳入和排除标准同建模组.同时对训练组以及验证组模型进行区分度和准确性的效能评估.结果 建模集最终共纳入266例BPD极早产儿,其中轻度BPD组189例(71.1％),中/重度BPD组77例(28.9％).多因素logistic回归分析示母子痫前期、肺炎、有创通气时间及血流动力学的动脉导管未闭(hemodynamically significant patent ductus arteriosus,hsP-DA)是极早产儿发生中/重度BPD的危险因素(P＜0.05).基于上述危险因素构建中/重度BPD的列线图预测模型,建模组和验证组的曲线下面积(AUC)分别为0.810(95％CI:0.750～0.870)和0.900(95％CI:0.826～0.975),列线图校准曲线接近理想对角线,模型具有良好的校准度.结论 母子痫前期、肺炎、有创通气时间及hsPDA为极早产儿发生中/重度BPD的独立危险因素,将4个指标联合构建模型能够对中/重度BPD的发生有较好的预测价值,有助于临床早期识别中/重度BPD患儿以改善防治效果.
Objective To identify risk factors for moderate-to-severe bronchopulmonary dysplasia(BPD)in extremely preterm infants born at less than 32 weeks'gestational age and to develop an early predictive nomogram model.Methods A retrospective analysis was conducted on extremely preterm infants born between January 2019 and December 2020 at the Third Affiliated Hospital of Guangzhou Medical University who developed BPD.Perinatal data and clinical in-formation during hospitalization were collected,and risk factors for moderate-to-severe BPD were identified using logis-tic regression analysis.Based on these findings,a nomogram prediction model was constructed and internally validated u-sing the Bootstrap method.An external validation cohort was selected from infants born between January 2021 and June 2021 meeting the same inclusion and exclusion criteria.Both the training and validation groups were used to assess the model's discrimination and accuracy.Results A total of 266 extremely preterm infants with BPD were included in the model development cohort,comprising 189 cases of mild BPD(71.1％)and 77 cases of moderate-to-severe BPD(28.9％).Multivariate logistic regression analysis indicated that preeclampsia,pneumonia,duration of invasive ventila-tion,and hemodynamically significant patent ductus arteriosus(hsPDA)were significant risk factors for moderate-to-severe BPD(P＜0.05).The nomogram model constructed using these four factors demonstrated an area under the curve(AUC)of 0.810(95％CI:0.750-0.870)in the model development cohort and 0.900(95％CI:0.826-0.975)in the validation cohort,with calibration curves close to the ideal diagonal line,indicating good calibration.Conclusion Preeclampsia,pneumonia,duration of invasive ventilation,and hsPDA are independent risk factors for moderate-to-se-vere BPD in extremely preterm infants.The nomogram model combining these four indicators demonstrates good predictive value,aiding in early identification of infants at risk of moderate-to-severe BPD to improve prevention and management outcomes.
李彩环;古健;樊雨薇;范斯潆;李苑;钟鑫琪;黄为民
南方医科大学第一临床医学院(南方医院)新生儿科(广东 广州 510515)广东省产科重大疾病重点实验室、广东省妇产疾病临床医学研究中心、粤港澳母胎医学高校联合实验室、广州医科大学附属第三医院新生儿科(广东 广州 510150)
临床医学
极早产儿,支气管肺发育不良影响因素预测效能
extremely preterm infantsbronchopulmonary dysplasiarisk factorspredictive model
《广东医学》 2024 (012)
1594-1600 / 7
广东省自然科学基金项目(2022A1515010289);广州市科技计划局基础与应用基础研究项目(2024A03J1163)
评论