国际肿瘤学杂志 ›› 2026, Vol. 53 ›› Issue (8): 464-469.doi: 10.3760/cma.j.cn371439-20251015-00075

• 论著 • 上一篇    下一篇

高压氧舱治疗乳腺癌急性放射性皮肤炎疗效的列线图预测模型构建

牛萍, 田龙, 崔玲()   

  1. 河北北方学院附属第一医院放疗科张家口 075000
  • 收稿日期:2025-10-15 出版日期:2026-08-08 发布日期:2026-07-21
  • 通讯作者: 崔玲,Email: zjkxhtl@126.com
  • 作者简介:

    牛萍:数据收集、整理,统计学分析及论文撰写;田龙:研究指导、论文修改;崔玲:论文修改

  • 基金资助:
    张家口市科技计划项目(2522119D)

Construction of a nomogram prediction model for evaluating the therapeutic efficacy of hyperbaric oxygen chamber therapy in managing acute radiation dermatitis among breast cancer patients

Niu Ping, Tian Long, Cui Ling()   

  1. Department of Radiation OncologyFirst Affiliated Hospital of Hebei Northern UniversityZhangjiakou 075000, China
  • Received:2025-10-15 Online:2026-08-08 Published:2026-07-21
  • Contact: Cui Ling, Email: zjkxhtl@126.com
  • Supported by:
    S&T Program of Zhangjiakou(2522119D)

摘要:

目的 分析影响高压氧舱治疗乳腺癌放疗致急性放射性皮肤炎(ARD)效果的因素并构建列线图预测模型。方法 选取河北北方学院附属第一医院放疗科2021年12月至2024年12月接受改良根治术和放疗的乳腺癌患者720例为研究对象,采用随机抽签法,按照7∶3的比例将患者分为模型组(n=504)和验证组(n=216)。在常规ARD预防措施基础上,患者接受15次高压氧舱治疗。采用多因素logistic回归分析模型组患者第7周时ARD为1级的影响因素并构建疗效列线图预测模型。采用C-index和校准曲线评价模型一致性,绘制模型组和验证组患者ARD为1级的受试者操作特征(ROC)曲线并采用DeLong检验比较曲线下面积(AUC)。结果 放疗第7周时发生1级与2级及以上ARD患者的年龄(t=-9.05,P<0.001)、化疗史(χ2=4.47,P=0.035)、高血压(χ2=5.40,P=0.020)、糖尿病(χ2=5.34,P=0.021)、微循环障碍(χ2=4.14,P=0.042)、动脉粥样硬化(χ2=4.12,P=0.042)差异均有统计学意义。多因素分析显示,年龄(OR=1.56,95%CI为1.09~1.93,P=0.031)、化疗史(OR=2.30,95%CI为1.66~4.78,P<0.001)、高血压(OR=1.79,95%CI为1.24~2.78,P=0.026)、糖尿病(OR=1.88,95%CI为1.39~3.25,P=0.019)、微循环障碍(OR=2.15,95%CI为1.58~4.46,P=0.004)、动脉粥样硬化(OR=2.01,95%CI为1.49~4.27,P=0.010)均为乳腺癌患者根治术后放疗第7周发生1级ARD的独立影响因素。基于上述因素构建的列线图预测模型C-index为0.862(95%CI为0.824~0.895),校准曲线一致性良好(χ2=5.89,P=0.177)。列线图模型预测模型组和验证组乳腺癌患者根治术后放疗第7周ARD为1级的AUC分别为0.852、0.839,差异无统计学意义(Z=0.48,P=0.328)。结论 对于年龄≤51岁、无化疗史、无高血压、无糖尿病、无微循环障碍、无动脉粥样硬化的乳腺癌根治术后放疗致ARD患者,高压氧舱治疗效果较为理想。基于上述因素构建的列线图预测模型有助于及时、准确地筛选适合接受该治疗的患者,具有一定的临床应用和参考价值。

关键词: 乳腺肿瘤, 化放疗, 放射性皮炎, 高压氧

Abstract:

Objective To analyze the factors affecting the efficacy of hyperbaric oxygen chamber in treating acute radiation dermatitis (ARD) caused by breast cancer radiotherapy, and to construct a nomogram prediction model. Methods A total of 720 breast cancer patients who underwent modified radical mastectomy and radiotherapy in the Department of Radiation Oncology of the First Affiliated Hospital of Hebei North University from December 2021 to December 2024 were selected as research subjects. Using the random drawing method, patients were divided into the model group (n=504) and the validation group (n=216) in a ratio of 7∶3. On the basis of conventional ARD prevention measures, the patients received 15 sessions of hyperbaric oxygen chamber therapy. At week 7, multivariate logistic regression was used to analyze the influencing factors for patients with ARD grade 1 in the model group. A efficacy nomogram prediction model was constructed. The model's consistency was assessed using the C-index and calibration curves. Receiver operator characteristic (ROC) curve was plotted for patients with grade 1 ARD in both the model group and the validation group, and the area under the curve (AUC) was compared between groups using the DeLong test. Results There were statistically significant differences in age (t=-9.05, P<0.001), history of chemotherapy (χ2=4.47, P=0.035), hypertension (χ2=5.40, P=0.020), diabetes (χ2=5.34, P=0.021), microcirculation disorder (χ2=4.14, P=0.042), and arteriosclerosis (χ2=4.12, P=0.042) between patients with grade 1 and grade 2 or above ARD at the 7th week of radiotherapy. Multivariate analysis revealed that age (OR=1.56, 95%CI: 1.09-1.93, P=0.031), history of chemotherapy (OR=2.30, 95%CI: 1.66-4.78, P<0.001), hypertension (OR=1.79, 95%CI: 1.24-2.78, P=0.026), diabetes (OR=1.88, 95%CI: 1.39-3.25, P=0.019), microcirculation disorder (OR=2.15, 95%CI: 1.58-4.46, P=0.004), and atherosclerosis (OR=2.01, 95%CI: 1.49-4.27, P=0.010) were all independent influencing factors for grade 1 ARD in breast cancer patients at the 7th week of radiotherapy after radical mastectomy. The C-index of the nomogram prediction model constructed based on the above factors was 0.862 (95%CI: 0.824-0.895), and the calibration curve showed good consistency (χ2=5.89, P=0.177). The AUCs of the nomogram model for predicting grade 1 ARD in breast cancer patients at the 7th week of radiotherapy after radical mastectomy in the model group and validation group were 0.852 and 0.839, respectively, with no statistically significant difference (Z=0.48, P=0.328). Conclusions For patients aged 51 or younger, without a history of chemotherapy, without hypertension, without diabetes, without microcirculation disorders, without arteriosclerosis, and who have developed ARD after radical mastectomy for breast cancer, hyperbaric oxygen chamber therapy demonstrates a relatively favorable therapeutic outcome. A nomogram prediction model developed based on the above factors is helpful for timely and accurately screening patients suitable for this intervention, offering certain clinical applicability and reference value.

Key words: Breast neoplasms, Chemoradiotherapy, Radiodermatitis, Hyperbaric oxygen