结核与肺部疾病杂志  2026, Vol. 7 Issue (5): 615-623    DOI: 10.19983/j.issn.2096-8493.20262040

• 论著 • 上一篇    下一篇

肺癌化疗患者重度骨髓抑制风险预测模型的构建与验证

刘素玲1, 陈华2, 丁文萍1, 王番1()   

  1. 1 呼吸疾病全国重点实验室/广州医科大学结核病防治研究院/广东药科大学附属广州胸科医院肺结节诊疗中心, 广州 510095
    2 呼吸疾病全国重点实验室/广州医科大学结核病防治研究院/广东药科大学附属广州胸科医院非结核分枝杆菌病科, 广州 510095
  • 收稿日期:2026-03-06 出版日期:2026-10-20 发布日期:2026-10-19
  • 通信作者: 王番 E-mail:414679388@qq.com
  • 基金资助:
    广州市卫生健康科技一般引导项目(20241A010036)

Construction and validation of a risk prediction model for severe myelosuppression in lung cancer patients undergoing chemotherapy

Liu Suling1, Chen Hua2, Ding Wenping1, Wang Fan1()   

  1. 1 State Key Laboratory of Respiratory Disease, Guangzhou Medical University Institute of Tuberculosis/Guangzhou Chest Hospital Affiliated to Guangdong Pharmaceutical University/Pulmonary Nodule Diagnosis and Treatment Center, Guangzhou 510095, China
    2 State Key Laboratory of Respiratory Disease, Guangzhou Medical University Institute of Tuberculosis/Department of Nontuberculous Mycobacterial Diseases, Guangzhou Chest Hospital Affiliated to Guangdong Pharmaceutical University, Guangzhou 510095, China
  • Received:2026-03-06 Online:2026-10-20 Published:2026-10-19
  • Contact: Wang Fan E-mail:414679388@qq.com
  • Supported by:
    General Guiding Project of Guangzhou Health Science and Technology(20241A010036)

摘要:

目的: 探讨影响肺癌化疗患者发生重度骨髓抑制的危险因素,构建肺癌化疗患者发生重度骨髓抑制的风险预测模型。方法: 回顾性分析2023年9月至2024年12月在广东药科大学附属广州胸科医院住院的145例肺癌化疗后发生骨髓抑制患者的资料,统计并分析患者发生重度骨髓抑制的危险因素。结果: 145例肺癌化疗后骨髓抑制患者中,重度骨髓抑制患者48例(33.1%)。Spearman相关分析显示,肺癌化疗后重度骨髓抑制与肺部手术史、含铂类化疗方案呈正相关,与体质量指数(BMI)、化疗周期、化疗前红细胞计数、血红蛋白、血小板计数、化疗前血清白蛋白水平、谷氨酰基转移酶呈负相关。多因素回归分析结果显示,非首次化疗是肺癌化疗患者发生重度骨髓抑制的保护因素(OR=0.069,95%CI:0.019~0.246),化疗前血清白蛋白水平降低(OR=10.789,95%CI:4.012~29.015)和肺部手术史(OR=3.193,95%CI:1.173~8.695)是肺癌化疗患者发生重度骨髓抑制的独立危险因素。应用logistic回归构建预测模型,采用受试者工作特征(ROC)曲线下面积检验模型预测效果,最终构建的模型ROC曲线下面积为0.839(95%CI:0.775~0.903),最大约登指数为 0.566,敏感度为 81.3%,特异度为75.3%。结论: 模型预测效果良好,可为临床治疗和制订预防重度骨髓抑制的护理措施提供参考依据。

关键词: 肺肿瘤, 药物疗法, 骨髓, 血清白蛋白, 预测

Abstract:

Objective: To explore the risk factors associated with severe myelosuppression in lung cancer patients receiving chemotherapy and to develop a risk prediction model for severe myelosuppression in these patients. Methods: A retrospective analysis was performed on the relevant data of 145 lung cancer patients who experienced myelosuppression following chemotherapy during their hospitalization at Guangzhou Chest Hospital from September 2023 to December 2024.The risk factors contributing to severe myelosuppression were systematically analyzed. Results: Among 145 lung cancer patients with myelosuppression following chemotherapy, 48 cases (33.1%) developed severe myelosuppression. Spearman correlation analysis revealed that severe myelosuppression was significantly positively correlated with the history of pulmonary surgery and platinum-containing chemotherapy regimens, and was negatively correlated with body mass index (BMI), chemotherapy cycles, pre-chemotherapy red blood cell count, hemoglobin level, platelet count, serum albumin level, and gamma-glutamyl transferase. Multivariate regression analysis showed that noninitial chemotherapy was a protective factor against severe myelosuppression in patients with lung cancer undergoing chemotherapy (OR=0.069,95%CI:0.019-0.246), while decreased serum albumin level (OR=10.789,95%CI:4.012-29.015) and history of pulmonary surgery (OR=3.193,95%CI:1.173-8.695) were independent risk factors for severe myelosuppression in these patients. A logistic regression model was established and the predictive efficacy of the model was verified via the area under the ROC curve, which was 0.839 (95% confidence interval (CI): (0.775-0.903) of the final model, with the maximum Youden index 0.566, the sensitivity 81.3%, and the specificity 75.3%. Conclusion: This model exhibits excellent predictive performance and can serve as a valuable reference for clinical treatment and nursing interventions to prevent severe myelosuppression.

Key words: Lung neoplasms, Drug therapy, Bone marrow, Serum albumin, Forecasting

中图分类号: