肺恶性肿瘤放射性肺炎危险因素分析及预测模型的构建
Analysis of Risk Factors for Radiation Pneu-monia in Pulmonary Malignancies and Construction of a Predictive Model

目的：探索影响放射性肺炎发生的危险因素，以指导肺恶性肿瘤放疗计划的制定，预防放射性肺炎的发生。方法：选取2021年8月至2021年12月就诊于青岛大学附属医院肿瘤放疗科的101例首次接受胸部放疗的肺恶性肿瘤患者，收集其临床基线资料包括患者的性别、年龄、高血压、糖尿病、吸烟史、饮酒史、既往胸部疾病、肿瘤分叶、病理类型、肿瘤分期，是否化疗或者行免疫、靶向治疗及放射剂量学参数，以放射治疗结束后发生≥2级放射性肺炎作为终点事件进行随访，回顾性分析其临床资料及放疗剂量学参数与放射性肺炎的关系，采用SPSS 26.0进行卡方检验的单因素分析和Logistics回归评估各因素对放射性肺炎的预测价值。结果：101例肺恶性肿瘤的患者中，≥2级的放射性肺炎共发生了37例数，约占36.6%，单因素分析显示糖尿病、既往胸部基础疾病、处方剂量、患肺MLD、患肺V10、V20、V30、双肺V10、V20、V30与放射性肺炎的发生有关，性别、年龄、高血压、吸烟史、饮酒史、肿瘤位置、病理类型、肿瘤分期、是否手术、是否化疗、是否同步放化疗、是否免疫靶向治疗与放射性肺肺炎的发生无统计学意义。通过共线性分析，提示剂量相关因素之间存在共线性，采用逐步分析法进行多因素分析显示既往胸部疾病、处方剂量、患肺V20是放射性肺炎发生的独立危险因素(P < 0.05)。结论：既往胸部疾病、处方剂量、患肺V20可以预测放射性肺炎的发生，根据临床资料和DVH参数，提示我们在制定计划时，充分考虑这些因素，预防放射性肺炎的发生。
Objective: To explore the risk factors affecting the occurrence of radiation pneumonia to guide the development of radiotherapy plan for pulmonary malignancies and prevent the occurrence of radi-ation pneumonia. Methods: One hundred and one patients with pulmonary malignancies who at-tended the Department of Oncology Radiotherapy of the Affiliated Hospital of Qingdao University for the first time from August 2021 to December 2021 were selected, and their clinical baseline data and radiation dosimetric parameters were collected, and the occurrence of grade ≥2 radiation pneumonia at the end of radiation therapy was used as the endpoint event for follow-up, and their clinical data and radiation dosimetric parameters were retrospectively analyzed in relation to radi-ation pneumonia. The relationship between their clinical data and radiotherapy dosimetric param-eters and radiation pneumonia was retrospectively analyzed, and the predictive value of each factor for radiation pneumonia was assessed by one-way analysis with chi-square test and logistic regres-sion using SPSS 26.0. Results: Among 101 patients with pulmonary malignancies, a total of 37 cases of grade ≥2 radiation pneumonia occurred, accounting for about 36.6%. Univariate analysis showed that diabetes mellitus, previous underlying chest disease, prescribed dose, pulmonary MLD, V10, V20, V30 of affected side, and bilateral pulmonary V10, V20, V30 were associated with the occur-rence of radiation pneumonia, and the analysis of covariance suggested that dose-related factors have correlation. The covariance between them and multifactor analysis using stepwise analysis showed that previous chest disease, prescribed dose, and affected lung V20 were independent risk factors for the development of radiation pneumonia (P < 0.05). Conclusion: Previous chest disease, prescribed dose, and having lung V20 can predict the occurrence of radiation pneumonia and, based