重度哮喘急性加重的预测通路：一项贝叶斯网络分析研究
2025/06/16
摘要
背景：精准预测急性发作风险是重度哮喘管理的关键。尽管已知多种风险因素，但其预测路径尚未明确。
目的：临床相关预测因素如何相互作用导致重度哮喘患者的重度急性加重。
研究设计与方法：从国际重度哮喘登记系统（2017-2021年）中纳入未接受过生物制剂治疗的重度哮喘患者（6814例，≥18岁），并进行为期12月跟踪随访。预测因子涵盖人口学特征、肺功能指标、炎症生物标志物、医疗资源使用情况、用药记录、急性发作史及合并症。通过结合专家共识与机器学习算法构建贝叶斯网络（BN），解析重度急性加重预测路径并进行内部验证。该影响图模型整合了决策节点与效用节点。
结果：BN分析显示，血嗜酸性粒细胞计数（BEC）、呼出气一氧化氮分数（FeNO）水平及一秒用力呼气容积占预计值百分比（FEV1%）直接调控既往到未来重度急性发作的风险转化。慢性鼻窦炎（CRS）通过直接影响BEC、FeNO及FEV1%，间接参与此过程。大环内酯类药物则通过独立作用于既往急性加重史来影响未来重度发作风险。模型在十折交叉验证与留一国家交叉验证中区分度中等，且在训练-测试数据集校准度良好。
结论：本研究揭示了重度哮喘急性加重的关键预测通路：CRS通过调控BEC、FeNO和FEV1%等直接预测因子介导风险向未来风险转化。大环内酯类药物使用构成另一独立预测通路。该发现为重度哮喘的临床共同决策提供了循证证据。
关键词：哮喘；贝叶斯网络；影响图；模型验证；重度急性发作。
文献来源：（Yadav CP, Chakraborty A,et, al. Prediction pathway for severe asthma exacerbations: a Bayesian Network analysis. Chest. 2025 May 19:S0012-3692(25)00647-6. doi: 10.1016/j.chest.2025.04.046. Epub ahead of print. PMID: 40398558.）
Background: Accurate risk prediction of exacerbations is pivotal in severe asthma management. Multiple risk factors are at play, but the pathway of risk prediction remains unclear.
Research question: How do the interplays of clinically relevant predictors lead to severe exacerbations in patients with severe asthma?
Study design and methods: Severe asthma patients (N=6814, ≥18 years), biologic naïve, were identified from the Severe Asthma Registry (2017-2021). Relevant predictors covered demographics, lung function, inflammation biomarkers, healthcare utilization, medications, exacerbation history, and comorbidities. Bayesian network (BN), representing the prediction process of severe exacerbations, was obtained by combining expert knowledge and machine learning algorithms. Internal validation was performed. The proposed influence diagram integrated decision and utility nodes into the prediction pathway.
Results: The BN analysis revealed that blood eosinophil count (BEC), fractional exhaled nitric oxide (FeNO) level, and % predicted forced expiratory volume in 1 second (FEV1) directly influenced the transition between prior and future severe exacerbations. The presence of chronic rhinosinusitis (CRS) indirectly affected such transition by directly influencing BEC, FeNO, and % predicted FEV1. Macrolide use independently affected history of exacerbations to influence future severe asthma exacerbations. Model discrimination was moderate in 10-fold cross-validation and leave-one-country-out cross-validation, and model calibration was high in train-test data.
Interpretation: An essential prediction pathway of severe exacerbation was identified, which involves the influence of CRS on the immediate predictors of risk transition from current to future severe asthma exacerbations. Macrolide use was another essential prediction pathway. The findings support shared clinical decision-making in severe asthma treatment.
Keywords: Asthma; Bayesian network; Influence diagram; Model validation; Severe exacerbation.
背景：精准预测急性发作风险是重度哮喘管理的关键。尽管已知多种风险因素，但其预测路径尚未明确。
目的：临床相关预测因素如何相互作用导致重度哮喘患者的重度急性加重。
研究设计与方法：从国际重度哮喘登记系统（2017-2021年）中纳入未接受过生物制剂治疗的重度哮喘患者（6814例，≥18岁），并进行为期12月跟踪随访。预测因子涵盖人口学特征、肺功能指标、炎症生物标志物、医疗资源使用情况、用药记录、急性发作史及合并症。通过结合专家共识与机器学习算法构建贝叶斯网络（BN），解析重度急性加重预测路径并进行内部验证。该影响图模型整合了决策节点与效用节点。
结果：BN分析显示，血嗜酸性粒细胞计数（BEC）、呼出气一氧化氮分数（FeNO）水平及一秒用力呼气容积占预计值百分比（FEV1%）直接调控既往到未来重度急性发作的风险转化。慢性鼻窦炎（CRS）通过直接影响BEC、FeNO及FEV1%，间接参与此过程。大环内酯类药物则通过独立作用于既往急性加重史来影响未来重度发作风险。模型在十折交叉验证与留一国家交叉验证中区分度中等，且在训练-测试数据集校准度良好。
结论：本研究揭示了重度哮喘急性加重的关键预测通路：CRS通过调控BEC、FeNO和FEV1%等直接预测因子介导风险向未来风险转化。大环内酯类药物使用构成另一独立预测通路。该发现为重度哮喘的临床共同决策提供了循证证据。
关键词：哮喘；贝叶斯网络；影响图；模型验证；重度急性发作。
文献来源：（Yadav CP, Chakraborty A,et, al. Prediction pathway for severe asthma exacerbations: a Bayesian Network analysis. Chest. 2025 May 19:S0012-3692(25)00647-6. doi: 10.1016/j.chest.2025.04.046. Epub ahead of print. PMID: 40398558.）
（南方医科大学南方医院 樊可可 朱邦 赵海金）
AbstractBackground: Accurate risk prediction of exacerbations is pivotal in severe asthma management. Multiple risk factors are at play, but the pathway of risk prediction remains unclear.
Research question: How do the interplays of clinically relevant predictors lead to severe exacerbations in patients with severe asthma?
Study design and methods: Severe asthma patients (N=6814, ≥18 years), biologic naïve, were identified from the Severe Asthma Registry (2017-2021). Relevant predictors covered demographics, lung function, inflammation biomarkers, healthcare utilization, medications, exacerbation history, and comorbidities. Bayesian network (BN), representing the prediction process of severe exacerbations, was obtained by combining expert knowledge and machine learning algorithms. Internal validation was performed. The proposed influence diagram integrated decision and utility nodes into the prediction pathway.
Results: The BN analysis revealed that blood eosinophil count (BEC), fractional exhaled nitric oxide (FeNO) level, and % predicted forced expiratory volume in 1 second (FEV1) directly influenced the transition between prior and future severe exacerbations. The presence of chronic rhinosinusitis (CRS) indirectly affected such transition by directly influencing BEC, FeNO, and % predicted FEV1. Macrolide use independently affected history of exacerbations to influence future severe asthma exacerbations. Model discrimination was moderate in 10-fold cross-validation and leave-one-country-out cross-validation, and model calibration was high in train-test data.
Interpretation: An essential prediction pathway of severe exacerbation was identified, which involves the influence of CRS on the immediate predictors of risk transition from current to future severe asthma exacerbations. Macrolide use was another essential prediction pathway. The findings support shared clinical decision-making in severe asthma treatment.
Keywords: Asthma; Bayesian network; Influence diagram; Model validation; Severe exacerbation.
上一篇：
特泽利尤单抗（Tezepelumab）治疗成人重度哮喘实现临床缓解及生物缓解：一项真实世界研究
下一篇：
度普利尤单抗治疗慢阻肺病的临床疗效和安全性：一项为期7年的人群队列研究
