中国麻风皮肤病杂志 ›› 2024, Vol. 40 ›› Issue (3): 157-162.doi: 10.12144/zgmfskin202403157

• 论著 • 上一篇    下一篇

大疱性类天疱疮伴发恶性肿瘤的相关因素分析及预测模型建立

韩君雅,吴维维,何筱君,祝文文,方险峰   

  1. 三峡大学第一临床医学院，宜昌市中心人民医院，湖北宜昌，443003
  • 出版日期:2024-03-15 发布日期:2024-02-26

Related factors analysis and prediction model establishment of malignant tumor in patients with bullous pemphigoid

HAN Junya, WU Weiwei, HE Xiaojun, ZHU Wenwen, FANG Xianfeng   

  1. Yichang Central People's Hospital, The First College of Clinical Medical Science, China Three Gorges University, Yichang 443003,China
  • Online:2024-03-15 Published:2024-02-26

PDF (PC)

167

摘要： 目的：明确大疱性类天疱疮(BP)伴发恶性肿瘤的相关因素并建立风险预测模型。方法：收集2018年1月至2023年6月在我院确诊为BP的住院患者109例,根据是否伴发恶性肿瘤分为单纯BP组和BP合并肿瘤组，对两组患者的临床特征及实验室检查进行统计学分析，经二元Logistic回归分析将筛选出来的危险因素利用R软件建立临床预测模型。结果：109例BP患者中有19例（17.43%）伴发恶性肿瘤，其中肺癌最常见(6例)。单因素和二元Logistic回归分析结果示糖类抗原199和血沉升高是BP伴发肿瘤的危险因素，白蛋白是保护因素。基于以上危险因素建立的列线图预测模型ROC曲线面积为0.917（95% CI 0.8573~0.9766）,校准曲线分析结果显示该预测模型有较好的校准能力。结论：血沉和糖类抗原199是BP伴发肿瘤的影响因素，基于以上危险因素建立的BP伴发恶性肿瘤的风险预测模型预测效能良好，可为BP患者进行系统性肿瘤筛查提供参考。

关键词: 大疱性类天疱疮, 恶性肿瘤, 相关因素, 预测模型, 列线图

Abstract: Objective: To determine the related factors of bullous pemphigoid (BP) complicated with malignant tumors and establish clinical prediction model. Methods: One hundred and nine BP patients hospitalized in our hospital from January 2018 to June 2023 were collected. According to whether there was a concomitant malignant tumor, the patient were divided into BP simple group and BP combined tumor group. Statistical analysis of clinical features and laboratory tests were performed in both groups. Related factors of BP complicated with malignant tumor were analyzed by binary logistic regression, R software was used to construct a clinical prediction model. Results: Of 109 patients, 19 (17.43%) were complicated with malignant tumor, and lung cancer was the most common (6 cases). The univariate analysis and binary logistic regression analysis showed that erythrocyte sedimentation rate and carbohydrate antigen 199 were risk factors of malignancy in BP patients, albumin was a protective factor. The AUC of ROC curve of the nomogram prediction model established based on the above risk factors was 0.917 (95% CI 0.8573-0.9766). The correction curves of the predictive model set indicated that the predictive model had good calibration ability. Conclusion: Erythrocyte sedimentation rate and carbohydrate antigen 199 are risk factors for malignancy in BP patients. The prediction model shows good predictive ability, which can provide a reference for systemic tumor screening in patients with BP.

Key words: bullous pemphigoid, malignant tumor, related factors, prediction model, nomogram model