中国组织工程研究 ›› 2025, Vol. 29 ›› Issue (33): 7137-7142.doi: 10.12307/2025.801

• 人工假体 artificial prosthesis • 上一篇    下一篇

骨质疏松症患者髋关节置换预后相关因素分析及Nomogram预测模型的构建

王荣强,杨  柳,吴向坤,尚立林   

  1. 南阳市第二人民医院,河南省南阳市   473000
  • 收稿日期:2024-03-07 接受日期:2024-06-01 出版日期:2025-11-28 发布日期:2025-04-12
  • 作者简介:王荣强,男,1983年生,河南省南阳市人,汉族,2007年新乡医学院毕业,副主任医师,主要从事关节置换方向的研究。

Analysis of factors associated with prognosis of osteoporosis patients after hip arthroplasty and construction of Nomogram prediction model

Wang Rongqiang, Yang Liu, Wu Xiangkun, Shang Lilin   

  1. Nanyang Second People’s Hospital, Nanyang 473000, Henan Province, China
  • Received:2024-03-07 Accepted:2024-06-01 Online:2025-11-28 Published:2025-04-12
  • About author:Wang Rongqiang, Associate chief physician, Nanyang Second People’s Hospital, Nanyang 473000, Henan Province, China

摘要:


文题释义:

Nomogram预测模型:也称为列线图(Alignment Diagram)或诺莫图(Nomogram图),是一种用于可视化预测模型结果的图形工具。它是建立在多因素回归分析的基础上,将多个预测指标进行整合,并通过带有刻度的线段,按照一定的比例绘制在同一平面上,从而用以表达预测模型中各个变量之间的相互关系。
二分类logistic回归分析:是一种分类方法,主要用于预测离散的二元输出结果,例如“是/否”或“正/负”。这种方法基于Logistic回归模型,并使用逻辑函数来模拟结果的概率,广泛应用于数据挖掘和机器学习领域,特别是在预测某个疾病的病因、分析客户是否会购买某种产品等方面。


背景:骨质疏松症患者髋关节置换预后不良严重影响患者的生活质量。准确预测骨质疏松症患者髋关节置换预后不良的危险因素仍然是骨科医生面临的重大挑战。

目的:探讨骨质疏松症患者髋关节置换预后不良的危险因素并构建Nomogram预测模型。 
方法:选择2020年7月至2022年6月于南阳市第二人民医院行髋关节置换的192例骨质疏松症患者为研究对象,术后6个月行髋关节Harris评分,将Harris评分≥80分的患者纳入预后良好组(n=142),Harris评分< 80分的患者纳入预后不良组(n=50)。收集两组患者临床资料并进行单因素分析;受试者工作特征曲线分析计量指标对骨质疏松症患者髋关节置换预后不良的预测价值;二分类logistic回归分析影响骨质疏松症患者髋关节置换预后不良的危险因素;构建骨质疏松症患者髋关节置换预后不良的Nomogram预测模型,采用校正曲线行内部验证并计算一致性指数,决策曲线行临床预测效能评估。 

结果与结论:①两组患者在年龄、体质量指数、手术时间、术中出血量、血清白蛋白、外周血淋巴细胞计数、预后营养指数、并发症方面的差异有显著意义(P < 0.05);②年龄、体质量指数、手术时间、术中出血量、血清白蛋白、外周血淋巴细胞计数、预后营养指数的曲线下面积为0.813,0.780,0.787,0.764,0.777,0.785,0.818;③年龄、体质量指数、术中出血量、并发症是影响骨质疏松症患者髋关节置换预后不良的危险因素;④Nomogram预测模型的校正、原始曲线与理想曲线接近,一致性指数为0.851(0.815-0.886),模型拟合度较高,Nomogram预测模型的阈值> 0.12,可提供临床净收益,且临床净收益均高于独立预测因子;⑤提示年龄、体质量指数、术中出血量、并发症是影响骨质疏松症患者髋关节置换预后不良的危险因素,基于此构建的Nomogram预测模型可帮助临床医生对骨质疏松症患者髋关节置换的预后情况进行评估,制定个性化干预措施,改善预后,提高生活质量。

https://orcid.org/0009-0003-1311-2343 (王荣强) 

中国组织工程研究杂志出版内容重点:人工关节;骨植入物;脊柱;骨折;内固定;数字化骨科;组织工程

关键词: 骨质疏松症, 髋关节置换, 预后, Nomogram预测模型, 并发症

Abstract: BACKGROUND: Poor prognosis of hip arthroplasty in patients with osteoporosis seriously affects the patients’ quality of life. Accurately predicting the risk factors for poor prognosis of hip arthroplasty in patients with osteoporosis remains a major challenge for orthopedic surgeons.
OBJECTIVE: To explore risk factors for poor prognosis after hip arthroplasty in patients with osteoporosis and construct a Nomogram prediction model. 
METHODS: A total of 192 patients with osteoporosis who underwent hip arthroplasty in Nanyang Second People’s Hospital from July 2020 to June 2022 were selected as study subjects. Harris hip function scale was performed 6 months after operation. Patients with Harris score ≥ 80 were included in the good prognosis group (n=142), while patients with Harris score < 80 were included in the poor prognosis group (n=50). Clinical data of the two groups were collected and subjected to univariate analysis. Receiver operating characteristic curves were used to analyze the predictive value of the measures for poor prognosis after hip arthroplasty in patients with osteoporosis. Binary logistic regression was used to analyze the risk factors affecting poor prognosis after hip arthroplasty in patients with osteoporosis. The Nomogram prediction model for poor prognosis after hip arthroplasty in patients with osteoporosis was constructed. The calibration curve was internally validated and the concordance index was calculated, and the decision curve was evaluated for clinical predictive efficacy. 
RESULTS AND CONCLUSION: (1) The differences between the two groups were statistically significant in terms of age, body mass index, operative time, intraoperative bleeding, serum albumin, peripheral blood lymphocyte count, prognostic nutritional index, and complications (P < 0.05). (2) Area under the curve for age, body mass index, operative time, intraoperative bleeding, serum albumin, peripheral blood lymphocyte count, and prognostic nutritional index were 0.813, 0.780, 0.787, 0.764, 0.777, 0.785, and 0.818. (3) Age, body mass index, intraoperative bleeding, and complications were risk factors for poor prognosis after hip arthroplasty in patients with osteoporosis. (4) The corrected, raw curve of the nomogram prediction model was close to the ideal curve with a concordance index of 0.851 (0.815-0.886) and a good model fit, with a threshold of > 0.12 for the Nomogram prediction model to provide a net clinical benefit, and all net clinical benefits were higher than the independent predictors. (5) It is concluded that age, body mass index, intraoperative bleeding, and complications are risk factors affecting the poor prognosis of osteoporotic patients after hip arthroplasty. The Nomogram prediction model constructed based on this can help clinicians assess the prognosis of osteoporotic patients after hip arthroplasty, develop personalized interventions, improve prognosis, and enhance the quality of life.

Key words: osteoporosis, hip arthroplasty, prognosis, Nomogram prediction model, complication

中图分类号: