Predictive value of bronchoscopy combined with CT score for refractory mycoplasma pneumoniae pneumonia in children

Abstract

Introduction Mycoplasma pneumoniae pneumonia (MPP) is prevalent in paediatric patients and can progress to refractory mycoplasma pneumoniae pneumonia (RMPP). Objective To assess the predictive value of bronchoscopy combined with computed tomography (CT) score in identifying RMPP in children. Methods A retrospective analysis was conducted on 244 paediatric patients with MP, categorising them into RMPP and general mycoplasma pneumoniae pneumonia (GMPP) groups. A paired t-test compared the bronchitis score (BS) and CT score before and after treatment, supplemented by receiver operating characteristic (ROC) analysis. Results The RMPP group showed higher incidences of extrapulmonary complications and pleural effusion (58.10% and 40%, respectively) compared with the GMPP group (44.60%, p = 0.037 and 18.71%, p < 0.001, respectively). The CT scores for each lung lobe were statistically significant between the groups, except for the right upper lobe (p < 0.05). Correlation analysis between the total CT score and total BS yielded r = 0.346 and p < 0.001. The ROC for BS combined with CT score, including area under the curve, sensitivity, specificity, and cut-off values, were 0.82, 0.89, 0.64, and 0.53, respectively. Conclusion The combined BS and CT score method is highly valuable in identifying RMPP in children.

Model based on the automated AI-driven CT quantification is effective for the diagnosis of refractory Mycoplasma pneumoniae pneumonia

“…In contrast to the reporting with the visual quantitative estimation of lesions provided by a radiologist, AI analysis of radiologic images possesses good reproducibility and reduces the radiologist’s reading time. To the best of our knowledge, Weihong Lu et al apply quantitative CT scoring to the chest imaging assessment of MPP, which was the latest study of radiological scoring tools for predicting RMPP 14 . However, they require radiologists or clinicians to determine CT scores on the image, which is time consuming.…”

“…CT imaging plays a vital role in diagnosing and monitoring disease progression in patients with MPP by offering detailed visualization of lung tissue and adjacent structures involved in the disease process 12 , 13 . Weihong Lu et al elevated CT scores manually in identifying RMPP and obtained an AUC of 0.781 14 . By combining lung CT characteristics with clinical indicators, a hybrid model could be developed to enhance the accuracy of RMPP prediction beyond that of existing models.…”