乳腺癌术后患者淋巴水肿影响因素分析

Analysis of Influencing Factors of Lymphedema in Postoperative Breast Cancer Patients

摘要: 目的：探讨乳腺癌术后淋巴水肿(BCRL)的独立危险因素，构建个体化风险预测模型，为早期识别与精准干预提供循证依据。方法：回顾性纳入2023年1月~2025年4月本院84例行乳腺癌根治/改良根治术并行规范放疗的女性患者。术后随访12个月，以ISL ≥ Ⅰ期且RCE ≥ 10%诊断BCRL。收集临床病理、手术参数及实验室指标，采用单因素与多因素Logistic回归筛选独立影响因素，建立预测模型并评价其区分度与校准度。结果：20例(23.81%)发生BCRL。多因素分析显示，阳性淋巴结数(OR = 1.46)、ALND范围 ≥ Ⅱ水平(OR = 3.49)、术后引流总量每增加100 ml (OR = 1.32)、NLR每升高1 (OR = 2.36)、VEGF-C每升高100 pg·ml⁻¹ (OR = 1.52)及血清白蛋白(OR = 0.79)为独立影响因素；模型AUC = 0.87，Hosmer-Lemeshow P = 0.81，内部验证C指数 = 0.87。结论：基于6项易获取指标构建的预测模型具有良好的区分度与校准度，可用于术前–术后早期识别BCRL高危人群并指导分级康复干预。

Abstract: Objective: To explore the independent risk factors of postoperative lymphedema (BCRL) after breast cancer surgery, construct an individualized risk prediction model, and provide evidence-based basis for early identification and precise intervention. Methods: A retrospective study included 84 female patients who underwent radical mastectomy/modified radical mastectomy for breast cancer and received standardized radiotherapy in our hospital from January 2023 to April 2025. The patients were followed up for 12 months after the operation. BCRL was diagnosed as ISL ≥ stage I and RCE ≥ 10%. Clinicopathological, surgical parameters and laboratory indicators were collected. Univariate and multivariate Logistic regression were used to screen independent influencing factors. A predictive model was established and its discrimination and calibration were evaluated. Result: BCRL occurred in 20 cases (23.81%). Multivariate analysis showed that the number of positive lymph nodes (OR = 1.46), ALND range ≥ level II (OR = 3.49), every 100 ml increase in total postoperative drainage (OR = 1.32), every 1 increase in NLR (OR = 2.36), every 100 pg·ml⁻¹ increase in VEGF-C (OR = 1.52), and serum albumin (OR = 0.79) It is an independent influencing factor; Model AUC = 0.87, Hosmer-Lemeshow P = 0.81, internal validation C-index = 0.87. Conclusion: The prediction model constructed based on six easily accessible indicators has good discrimination and calibration, and can be used to identify high-risk populations of BCRL in the early preoperative and postoperative period and guide hierarchical rehabilitation intervention.