目的 探讨肺脏超声应用于新生儿肺疾病随访的可行性。 方法 回顾性选取2025年3—7月于郑州大学第一附属医院新生儿科随访的102例新生儿期患肺疾病患儿为研究对象，按随访时肺脏超声结果分为正常组与异常组，比较两组临床特征；采用多因素logistic回归分析筛选随访时肺脏超声异常的影响因素，并构建预测模型，通过受试者操作特征曲线验证其预测患肺疾病新生儿随访时肺脏超声异常的效能。 结果 中位随访时长32 d，肺超声异常58例（56.9%）。正常组与异常组胎龄、出生体重、随访时长等比较差异有统计学意义（P<0.05）。多因素logistic回归分析显示，胎龄（OR=0.617，P<0.001）、随访时长（OR=0.976，P<0.001）是患肺疾病新生儿随访时肺超声异常的独立保护因素。受试者操作特征曲线分析显示，以胎龄和随访时长构建联合模型logit（P）=17.001-0.025×随访时长-0.483×胎龄（其中P为患肺疾病新生儿随访时肺超声异常的发生概率），对预测随访时肺脏超声异常效能的曲线下面积为0.833（95%CI：0.758~0.909，P<0.001），灵敏度为62.1%，特异度为93.2%，拟合优度良好（P>0.05）。 结论 肺脏超声可应用于新生儿肺疾病随访。

Objective To explore the feasibility of lung ultrasound in the follow-up of neonatal lung disease. Methods A retrospective study was conducted of 102 infants with neonatal lung disease who were followed up in the Department of Neonatology, the First Affiliated Hospital of Zhengzhou University from March to July 2025. According to follow-up lung ultrasound results, infants were classified into a normal group and an abnormal group. Baseline and clinical characteristics were compared between groups. Multivariable logistic regression was performed to identify independent influencing factors of abnormal lung ultrasound at follow-up, and a prediction model was constructed. Its performance for predicting abnormal lung ultrasound at follow-up was evaluated using receiver operating characteristic (ROC) curve analysis. Results The median follow-up duration was 32 days, and 58 infants (56.9%) had abnormal lung ultrasound. Significant differences were found between the normal and abnormal groups in gestational age, birth weight, and follow-up duration (all P<0.05). Multivariable logistic regression showed that gestational age (OR=0.617, P<0.001) and follow-up duration (OR=0.976, P<0.001) were independent protective factors against abnormal lung ultrasound at follow-up. ROC analysis indicated that the combined model constructed with gestational age and follow-up duration, logit(P)=17.001-0.025×follow-up duration-0.483×gestational age (where P is the probability of abnormal lung ultrasound at follow-up), achieved an area under the curve of 0.833 (95%CI: 0.758-0.909, P<0.001), with a sensitivity of 62.1% and a specificity of 93.2%. The goodness of fit was good (P>0.05). Conclusions Lung ultrasound is feasible for the follow-up of neonatal lung diseases.