Wang Jinyun, Wu Xiaowei, Hu Ge
Hefei Cancer Hospital, Chinese Academy of Sciences, Hefei 230031, People's Republic of China.
Int J Gen Med. 2024 Dec 17;17:6271-6279. doi: 10.2147/IJGM.S490451. eCollection 2024.
To investigate risk factors for esophageal fistula in esophageal squamous cell cancer (ESCC) patients who treated with volumetric modulated arc therapy (VMAT).
A retrospective analysis was performed on 171 ESCC patients treated with VMAT at Hefei Cancer Hospital, Chinese Academy of Sciences, from February 2017 to February 2021. Clinical and dosimetric parameters, including age, gender, feeding channel, tumor location, T stage, ulcerative tumor, were recorded. Univariate and multivariate logistic regression were used to determine risk factors for esophageal fistula. The predictive accuracy of the constructed nomogram was assessed using a receiver operating characteristic (ROC) curve and calibration curves.
Esophageal fistula occurred in 12.87% (22/171) of all the patients. Univariate analysis showed that gender, age, diabetes, T4 stage, ulcerative tumor, total radiation dose, maximum gross tumor volume (GTV) diameter, and GTV length correlated with the incidence of esophageal fistula. Multivariable analysis highlighted gender, age, diabetes, T4 stage, and total radiation dose as significant predictors. A predictive nomogram including these five factors was developed and showed an AUC of 0.876 (95% CI 0.807-0.946), a C-index of 0.847, and a corrected C-index of 0.833.
Gender, age, diabetes, T4 stage, and total radiation dose emerged as significant risk factors for esophageal fistula in ESCC patients undergoing VMAT. The developed nomogram provides a reliable tool to predict the risk of esophageal fistula risk in this cohort.
探讨接受容积调强弧形放疗(VMAT)的食管鳞状细胞癌(ESCC)患者发生食管瘘的危险因素。
对2017年2月至2021年2月在中国科学院合肥肿瘤医院接受VMAT治疗的171例ESCC患者进行回顾性分析。记录临床和剂量学参数,包括年龄、性别、进食途径、肿瘤位置、T分期、溃疡性肿瘤等。采用单因素和多因素logistic回归分析确定食管瘘的危险因素。使用受试者工作特征(ROC)曲线和校准曲线评估所构建列线图的预测准确性。
所有患者中12.87%(22/171)发生食管瘘。单因素分析显示,性别、年龄、糖尿病、T4分期、溃疡性肿瘤、总放疗剂量、最大肿瘤总体积(GTV)直径和GTV长度与食管瘘的发生率相关。多因素分析表明,性别、年龄、糖尿病、T4分期和总放疗剂量是显著的预测因素。开发了一个包含这五个因素的预测列线图,其曲线下面积(AUC)为0.876(95%可信区间0.807 - 0.946),C指数为0.847,校正C指数为0.833。
性别、年龄、糖尿病、T4分期和总放疗剂量是接受VMAT的ESCC患者发生食管瘘的显著危险因素。所开发的列线图为预测该队列中食管瘘风险提供了一个可靠的工具。