【摘要】 目的：乳腺癌相关淋巴水肿(breast cancer-related lymphedema,BCRL)是乳腺癌治疗的常见并发症，既往研究认为其与辅助化疗有关。但由于混杂因素干扰，辅助化疗中常用的紫杉醇类药物是否独立影响BCRL尚存争议。倾向性评分匹配(propensity score matching,PSM)可有效控制混杂偏倚，本研究旨在采用该法探讨紫杉醇类辅助化疗与BCRL之间的关联。方法：本研究为回顾性设计，数据来自中南大学湘雅三医院的电子病历系统。BCRL的诊断依据为患肢周长的客观测量结果。紫杉醇类辅助化疗定义为含多西他赛、紫杉醇或白蛋白结合型紫杉醇的化疗方案。通过PSM控制混杂因素，并对匹配后样本进行Logistic回归分析，评估紫杉醇类化疗与BCRL的关系。结果：本研究共纳入1 494例患者。PSM前，紫杉醇类辅助化疗与BCRL未表现出显著相关性(OR=0.979,95%CI 0.587～1.580;P=0.932)。PSM后共匹配363例，结果仍无统计学意义(OR=0.959,95%CI 0.540～1.672;P=0.885)。结论：无论是否控制其他已知危险因素，紫杉醇类辅助化疗均非BCRL的独立危险因素。未来需开展多中心、前瞻性研究进一步验证。更多还原

【Abstract】 Objective: Breast cancer-related lymphedema(BCRL) is a common complication of breast cancer treatment. Previous studies have suggested an association between BCRL and adjuvant chemotherapy; however, whether taxane-based regimens independently influence the risk of BCRL remains controversial due to potential confounding factors. Propensity score matching(PSM) can effectively reduce confounding bias. This study aims to investigate the association between taxane-based adjuvant chemotherapy and BCRL using PSM.Methods: This retrospective study was conducted using data from the electronic medical record system of Xiangya Third Hospital, Central South University. BCRL was diagnosed based on objective measurements of limb circumference. Taxane-based adjuvant chemotherapy was defined as chemotherapy regimens including docetaxel, paclitaxel, or albumin-bound paclitaxel. PSM was applied to control for potential confounders, followed by Logistic regression analysis in the matched cohort to evaluate the association between taxane-based chemotherapy and BCRL.Results: A total of 1 494 patients were included in this study. Before PSM, taxane-based adjuvant chemotherapy was not significantly associated with BCRL(OR=0.979, 95% CI 0.587 to 1.580; P=0.932). After PSM, 363 matched patients were obtained, and the association remained non-significant(OR=0.959, 95% CI 0.540 to 1.672; P=0.885).Conclusion: Taxane-based adjuvant chemotherapy is not an independent risk factor for BCRL, regardless of adjustment for known confounding factors. Further multicenter, prospective studies are warranted to further validate these findings.更多还原