多发性骨髓瘤患者合并心包积液危险因素分析及列线图预测模型构建  

Risk factors of multiple myeloma patients with pericardial effusionand construction of nomogram prediction model

作　　者：赵思捷 余航 党珮珠 王海洋 董杨 贺鹏程[1] ZHAO Sijie;YU Hang;DANG Peizhu;WANG Haiyang;DONG Yang;HE Pengcheng(Department of Hematology,the First Affiliated Hospital of Xi’an Jiaotong University,Xi’an 710061,China)

机构地区：[1]西安交通大学第一附属医院血液内科,陕西西安710061 [2]西安交通大学第一附属医院心血管内科,陕西西安710061 [3]西安交通大学第一附属医院心血管外科,陕西西安710061

出　　处：《陕西医学杂志》2024年第12期1635-1639,共5页Shaanxi Medical Journal

基　　金：国家自然科学基金资助项目(81270236)。

摘　　要：目的:探究多发性骨髓瘤(MM)患者合并心包积液的危险因素并构建列线图预测模型。方法:纳入MM患者238例,根据有无心包积液分为心包积液组(39例)和非心包积液组(199例)。比较两组患者临床资料和实验室指标,分析合并心包积液的影响因素,构建列线图模型并使用受试者工作特征(ROC)曲线和校准曲线评估模型预测效能。结果:与非心包积液组比较,心包积液组女性、ISSⅢ期及合并高血压占比以及D-二聚体水平更高,入院收缩压更低(均P<0.05)。Logistic回归分析显示,女性、合并高血压、D-二聚体升高和MM国际分级体系(ISS)Ⅲ期为MM患者合并心包积液的独立危险因素,入院收缩压升高为MM患者发生心包积液的保护因素(均P<0.05)。将性别、入院时收缩压、高血压、D-二聚体及ISS分期作为预测因子构建列线图预测模型,ROC曲线和校准曲线显示该模型具有良好的预测能效和判别能力。结论:女性、合并高血压、D-二聚体升高和ISSⅢ期为MM患者合并心包积液的独立危险因素,依据这些因素构建的列线图模型能够有效预测患者合并心包积液的风险。Objective:To investigate the risk factors of multiple myeloma(MM)patients with pericardial effusion and to construct a nomogram prediction model.Methods:A total of 238 MM patients were enrolled and divided into pericardial effusion group(39 cases)and non-pericardial effusion group(199 cases)according to the presence or absence of pericardial effusion.The clinical data and laboratory indexes of the two groups were compared,and the influencing factors of pericardial effusion were analyzed.A nomogram model was constructed,and ROC curve and calibration curve were used to evaluate the predictive efficacy of the model.Results:Compared with the non-pericardial effusion group,the pericardial effusion group had a higher proportion of women,ISS stageⅢ,hypertension,and D-dimer level,and a lower systolic blood pressure on admission(all P<0.05).Logistic regression analysis showed that female,hypertension,elevated D-dimer and ISS stageⅢwere independent risk factors for MM patients with pericardial effusion,and elevated systolic blood pressure on admission was a protective factor(all P<0.05).Gender,systolic blood pressure on admission,hypertension,D-dimer and ISS stage were used as predictors to construct a nomogram prediction model.ROC curve and calibration curve showed that the model had good predictive energy efficiency and discrimination ability.Conclusion:Female,hypertension,elevated D-dimer and ISS stageⅢare independent risk factors of MM patients with pericardial effusion.The nomogram model based on these risk factors can effectively predict the risk of pericardial effusion in MM patients.

关 键 词：多发性骨髓瘤 心包积液 危险因素 列线图模型 校准曲线 

分 类 号：R733.3[医药卫生—肿瘤]