中南医学科学杂志2025，Vol.53Issue(3)：462-465,520,5.DOI:10.15972/j.cnki.43-1509/r.2025.03.020
肱骨近端骨折术后肩关节功能恢复不良的列线图预测模型构建
Development and validation of a Nomogram prediction model for poor shoulder joint func-tional recovery after proximal humeral fracture surgery
摘要
Abstract
Aim To analyze the influencing factors of poor shoulder joint functional recovery after proximal humeral fracture(PHF)surgery and establish a nomogram prediction model.Methods Totally 163 PHF patients undergoing locking plate internal fixation surgery were divided into good recovery and poor recovery groups based on the 6-month Constant-Murley score,then divided in-to a training set(n=114)and a validation set(n=49)at a ratio of 7∶3.Clinical data were collected.Multivariable Logistic re-gression was used to identify risk factors and construct the nomogram model.ROC curve and Hosmer-Lemeshow test were used to e-valuate model performance.Results The poor recovery group had higher proportions of age≥60 years,poor reduction quality,BMD T-score≤-2.5,nerve injury,Neer type 3/4 fracture,vascular injury,and no early rehabilitation(P＜0.05).These factors were independent risk factors(P＜0.05).The Lasso-nomogram model showed AUC of 0.850 and 0.826 in training and validation sets respectively,and H-L test P=0.603.Conclusion Age,bone mineral density,vascular nerve injury,reduction quality and early functional exercise are related to the recovery of shoulder joint function after PHF.The established prediction model can early identify high-risk patients of poor functional recovery after PHF surgery.关键词
肱骨近端骨折/肩关节功能/列线图/预测模型Key words
proximal humeral fracture/joint function/Nomogram/prediction model分类
医药卫生引用本文复制引用
胡军,虎师元,张晓俊,鲜光波,王杰..肱骨近端骨折术后肩关节功能恢复不良的列线图预测模型构建[J].中南医学科学杂志,2025,53(3):462-465,520,5.基金项目
四川省医创会2024年度第二期"医创之巅"专项科研项目(YCH-KY-YCZD2024-077) （YCH-KY-YCZD2024-077）
