通讯作者: 乐静, nblejing@aliyun.com

DOI：10.12201/bmr.202410.00057

Construction and effect evaluation of risk prediction model for grades 3-4 myeloma bone disease in newly diagnosed multiple myeloma patients

• 摘要：目的 探讨初诊多发性骨髓瘤（New Diagnosis of Multiple Myeloma, NDMM）患者发生3-4级骨髓瘤骨病（Myeloma Bone Disease, MBD）的影响因素，并建立Nomogram风险预测模型。 方法 选取我院血液内科2015年1月至2021年12月NDMM患者，根据初诊时是否存在3-4级MBD分为A组（MBD 0-2级，110例）、B组（MBD 3-4级，151例），使用logistic回归分析，来确定初诊时发生3-4级MBD的危险因素并构建风险预测模型。为确保模型效果有效，使用受试者工作特征（ROC）曲线进行综合评价。 结果 共纳入261例NDMM患者，150例为3-4级MBD，发病率为56.39%；根据多因素回归分析结果，我们发现年龄、血清磷、血清C-反应蛋白（C-Reactive Protein, CRP）、血清球蛋白（Serum Globulin , GLB）以及骨髓浆细胞比例（Bone Marrow Plasma Cell Percentage, BMPCp）是NDMM患者发生3-4级MBD的独立危险因素（均P<0.05）。基于这些发现，我们构建了3-4级MBD的风险预测模型：logit（P）=-15.092+0.107（年龄）+1.150（血清磷）+0.057（CRP）+0.040（GLB）+0.212（BMPCp）。通过Lemeshow拟合优度检验，我们发现发生3-4级MBD的预测概率和实际发生率之间没有显著差异（P=0.770）。此外，我们还对模型的准确率进行了验证，结果表明模型的准确率达到了90.40%。最后，我们用ROC曲线下的面积（Area Under Curve, AUC）衡量模型的预测能力，结果显示AUC为0.957（95%CI为0.932～0.981），说明该模型的预测能力较为可靠。结论 年龄、血清磷、CRP、GLB、BMPCp是NDMM患者发生3-4级MBD的独立危险因素，运用logistic回归分析构建的风险预测模型对NDMM患者发生3-4级MBD的预测效果较好。

  关键词： 多发性骨髓瘤; 骨髓瘤骨病 logistic回归分析; 风险预测模型; 

   

  Abstract: Objective To investigate the influencing factors of developing Grade 3-4 myeloma bone disease (MBD) in newly diagnosed multiple myeloma (NDMM) patients and establish a Nomogram risk prediction model. Methods We selected NDMM patients from January 2015 to December 2021 at Ningbo Li Huili Hospital. Patients were divided into group A (MBD grades 0-2, 110 cases) and group B (MBD grades 3-4, 151 cases) based on the presence of Grade 3-4 MBD at diagnosis. Logistic regression analysis was used to determine the risk factors for the occurrence of Grade 3-4 MBD at diagnosis and construct a risk prediction model. To ensure the effectiveness of the model, we used the and receiver operating characteristic (ROC) curve for comprehensive evaluation. Results A total of 261 NDMM patients were included, with 150 cases with Grade 3-4 MBD, and the incidence rate was 56.39%. According to the results of the logistic multivariate regression analysis, we found that age, serum phosphorus, C-reactive protein (CRP), serum globulin (GLB), and bone marrow plasma cell proportion (BMPCp) were independent risk factors for the occurrence of Grade 3-4 MBD in NDMM patients (all P<0.05). Based on these findings, we established a risk prediction model for Grade 3-4 MBD: logit(P)=-15.092+0.107(age)+1.150(serum phosphorus)+0.057(CRP)+0.040(GLB)+0.212(BMPCp). The Hosmer-Lemeshow goodness-of-fit test showed no significant difference (P=0.770) between the predicted probability of developing Grade 3-4 MBD and the actual incidence rate. In addition, the accuracy of the model was verified, and the result showed that the accuracy rate of the model reached 90.40%. Finally, we measured the predictive ability of the model using the area under the ROC curve (AUC), and the result showed that the AUC was 0.957 (95% CI 0.932-0.981), indicating that the model had good predictive reliability. Conclusion Age, serum phosphorus, CRP, GLB, and BMPCp are independent risk factors for the occurrence of Grade 3-4 MBD in NDMM patients. The risk prediction model constructed using logistic regression analysis is effective for predicting the occurrence of Grade 3-4 MBD in NDMM patients.

  Key words: Multiple myeloma bone; myeloma bone disease; logistic regression analysis; risk prediction model

  提交时间：2024-10-20

  版权声明：作者本人独立拥有该论文的版权，预印本系统仅拥有论文的永久保存权利。任何人未经允许不得重复使用。
• html

• 图表

• 韦云师, 莫伟, 向娅, 廖秋姣, 何柳, 凌朝灵, 陆启祥, 刘芳印. 老年髋部骨折患者术后谵妄风险预测模型的研究进展. 2024. doi: 10.12201/bmr.202409.00029

  施明杰, 叶洪舟, 袁琛. 大环内酯类耐药肺炎支原体肺炎患儿发展为重症肺炎的危险因素分析及预测模型构建. 2024. doi: 10.12201/bmr.202409.00021

  王一凡, 石超君, 马安宁. Ⅱ型糖尿病并发动脉粥样硬化风险预测模型比较. 2024. doi: 10.12201/bmr.202404.00007

  杜雪杰, 葛辉. 基于BP神经网络方法构建手足口病预测预警模型设计研究. 2021. doi: 10.12201/bmr.202102.00002

  阮旭凌, 刘琦, 郭志恒, 晏峻峰. 基于LDA和XGBoost算法的乳腺癌预测模型构建研究. 2022. doi: 10.12201/bmr.202106.00007

  张红吉, 刁婷婷, 李晓非. 布鲁氏菌致骨损伤机制的研究进展. 2024. doi: 10.12201/bmr.202411.00041

  许峻川, 费琦. BFH-OST和OSTA对于绝经后妇女骨质疏松风险预测的对比性研究. 2024. doi: 10.12201/bmr.202407.00019

  邹倩, 陈美华, 张志迪. 中青年淋巴瘤患者二元应对的影响因素研究. 2024. doi: 10.12201/bmr.202410.00013

  陈思, 王立杰. 原发性直肠黑色素瘤病例报道. 2023. doi: 10.12201/bmr.202310.00004

  谢芬、余娇通讯作者、宋思蓓、黄琳. 高甘油三酯血症性急性胰腺炎患者短期内复发再入院的风险预警模型构建与验证. 2024. doi: 10.12201/bmr.202408.00010

• 序号	提交日期	编号	操作
  1	2024-09-18	bmr.202410.00057V1	下载

• 公开评论  匿名评论  仅发给作者 提交评论