Chiu Chien-Hung, Chang Wei-Yang, Yang Lan-Yan, Chao Yin-Kai, Chen Wei-Hsun, Liu Yun-Hen, Tsao Ya-Tzu, Chang Yu-Ling, Huang Shu-Chun
Division of Thoracic Surgery, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan.
Clinical Trial Center, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan.
BMC Sports Sci Med Rehabil. 2025 Apr 30;17(1):106. doi: 10.1186/s13102-025-01158-7.
Esophagectomy following neoadjuvant chemoradiotherapy or definitive chemoradiotherapy has been shown to yield favorable oncological outcomes in patients with locally advanced esophageal squamous cell carcinoma. However, postoperative complications are frequent and can adversely affect patient survival. This study aimed to identify physical fitness factors across multiple domains associated with major postoperative complications.
This study enrolled patients with esophageal squamous cell carcinoma who were treated with neoadjuvant chemoradiotherapy or definitive chemoradiotherapy and underwent esophagectomy between 2020 and 2022. Multivariate logistic regression analysis was conducted to identify the factors associated with major postoperative complications. Additionally, a tree-based learning process was employed to assess all the variables influencing major complications.
A total of 142 esophageal cancer patients who underwent esophagectomy were screened. Of these, 72 eligible patients were included in the study, and 29 (40.2%) experienced major postoperative complications. In the full model, factors such as low body weight, low body mass index, low peak oxygen consumption (V̇O), low V̇O/skeletal muscle mass (SMM), low appendicular skeletal muscle index (ASMI), and low ASMI-V̇O/SMM product (AV̇P) were found to be significantly associated with major postoperative complications. However, in the parsimonious model, only low AV̇P (P < 0.01) was associated with major complications. Additionally, AV̇P, EqOnadir, and hand grip strength emerged as the key predictors of major post-esophagectomy complications in the tree-based learning analysis, which showed a sensitivity of 0.448, specificity of 0.977, and accuracy of 76.4%, with false negative and false positive rates of 55.2% and 2.3%. AV̇P alone yielded similar results.
Patients with low AV̇P, high EqOnadir, and low HGS are at a very high risk of experiencing major postoperative complications, with low AV̇P showing the strongest correlation. Preoperative physical fitness screening can help identify high-risk patients and guide appropriate perioperative management.
新辅助放化疗或根治性放化疗后行食管切除术已被证明在局部晚期食管鳞状细胞癌患者中可产生良好的肿瘤学结局。然而，术后并发症很常见，且会对患者生存产生不利影响。本研究旨在确定与术后主要并发症相关的多个领域的体能因素。
本研究纳入了2020年至2022年间接受新辅助放化疗或根治性放化疗并接受食管切除术的食管鳞状细胞癌患者。进行多因素逻辑回归分析以确定与术后主要并发症相关的因素。此外，采用基于树的学习过程来评估影响主要并发症的所有变量。
共筛选出142例行食管切除术的食管癌患者。其中，72例符合条件的患者纳入研究，29例（40.2%）发生术后主要并发症。在全模型中，发现低体重、低体重指数、低峰值耗氧量（V̇O）、低V̇O/骨骼肌质量（SMM）、低四肢骨骼肌指数（ASMI）以及低ASMI-V̇O/SMM乘积（AV̇P）等因素与术后主要并发症显著相关。然而，在简约模型中，只有低AV̇P（P < 0.01）与主要并发症相关。此外，在基于树的学习分析中，AV̇P、EqOnadir和握力成为食管切除术后主要并发症的关键预测因素，其敏感性为0.448，特异性为0.977，准确性为76.4%，假阴性率和假阳性率分别为55.2%和2.3%。单独的AV̇P产生了类似的结果。
AV̇P低、EqOnadir高和握力低的患者发生术后主要并发症的风险非常高，其中低AV̇P的相关性最强。术前体能筛查有助于识别高危患者并指导适当的围手术期管理。
