Yang Shangmin, Sun Yanmeng, Wang Mengyuan, Xu Huan, Wang Shifu
Department of Microbiology Laboratory, Children's Hospital Affiliated to Shandong University (Jinan Children's Hospital), Jinan, China.
Department of Clinical Microbiology, Shandong Provincial Clinical Research Center for Children's Health and Disease, Jinan, China.
Front Cell Infect Microbiol. 2025 Jun 27;15:1616773. doi: 10.3389/fcimb.2025.1616773. eCollection 2025.
This study aimed to identify the independent risk factors and develop a predictive model for pulmonary aspergillosis (PA) in pediatric populations.
This retrospective study compromised 97 pediatric patients with pulmonary infections (38 PA cases and 59 non-PA cases) at Children's Hospital Affiliated to Shandong University between January 2020 and October 2024. Multivariate binary logistic regression was used to identify PA-associated risk factors. Receiver operating characteristic (ROC) curves, calibration plots, and Brier scoring were used to evaluate the diagnostic model.
8 clinical variables significantly differed between the PA and non-PA groups. Multivariate binary logistic regression analysis identified six significant independent risk factors: a history of surgery (OR: 9.52; 95% CI: 1.96-46.23; = 0.005), hematologic diseases (OR: 11.68; 95% CI: 0.89-153.62; = 0.062), absence of fever (OR: 8.244; 95% CI: 1.84-36.932; = 0.006), viral coinfection (OR: 15.99; 95% CI: 3.55-72.00; < 0.001), elevated (1, 3) -β -D-glucan levels (BDG, > 61.28 pg/mL; OR: 7.38; 95% CI: 1.26-43.31; = 0.027), and shorter symptom-to-admission interval (< 4.5 days; OR: 38.68; 95% CI: 5.38-277.94; < 0.001) were risk factors for PA. The predictive model demonstrated excellent discrimination (AUC 0.93, 95% CI 0.88-0.98) and calibration (Hosmer-Lemeshow p=0.606, R²=0.96, Brier score 0.097). metagenomic next - generation sequencing (mNGS) revealed significantly higher rates of polymicrobial infections in PA cases (86.84% vs 18.64%, p<0.001).
This study established and validated a high-performance predictive model incorporating six clinically accessible parameters for the diagnosis of pediatric PA.
本研究旨在确定儿童人群中肺曲霉病（PA）的独立危险因素，并建立预测模型。
本回顾性研究纳入了2020年1月至2024年10月期间山东大学附属儿童医院的97例肺部感染儿童患者（38例PA病例和59例非PA病例）。采用多变量二元逻辑回归确定与PA相关的危险因素。采用受试者操作特征（ROC）曲线、校准图和Brier评分来评估诊断模型。
PA组和非PA组之间有8个临床变量存在显著差异。多变量二元逻辑回归分析确定了6个显著的独立危险因素：手术史（OR：9.52；95%CI：1.96 - 46.23；P = 0.005）、血液系统疾病（OR：11.68；95%CI：0.89 - 153.62；P = 0.062）、无发热（OR：8.244；95%CI：1.84 - 36.932；P = 0.006）、病毒合并感染（OR：15.99；95%CI：3.55 - 72.00；P < 0.001）、（1,3）-β -D-葡聚糖水平升高（BDG，> 61.28 pg/mL；OR：7.38；95%CI：1.26 - 43.31；P = 0.027）以及症状出现至入院间隔较短（< 4.5天；OR：38.68；95%CI：5.38 - 277.94；P < 0.001）是PA的危险因素。该预测模型显示出良好的区分度（AUC 0.93，95%CI 0.88 - 0.98）和校准度（Hosmer-Lemeshow p = 0.606，R² = 0.96，Brier评分0.097）。宏基因组下一代测序（mNGS）显示PA病例中多重微生物感染的发生率显著更高（86.84%对18.64%，P < 0.001）。
本研究建立并验证了一个包含6个临床可及参数的高性能预测模型，用于儿童PA的诊断。
