预防老年髋部骨折术中压力性损伤风险模型构建及验证

    Construction and validation of a risk model for preventing intraoperative stress injury in elderly hip fractures

    • 摘要:
      目的: 探讨老年髋部骨折病人术中压力性损伤影响因素,建立预防性风险预测列线图模型。
      方法: 选取2022年6月至2024年12月接受髋部骨折手术的老年病人252例,根据术中是否发生压力性损伤,分为发生组(n = 42)与未发生组(n = 210)。采用差异性分析、多因素logistic回归分析老年髋部骨折病人术中压力性损伤发生的影响因素。建立术中压力性损伤发生风险预测列线图模型,采用ROC曲线分析列线图模型的效能,并对其预测稳定性与决策效能予以验证。
      结果: 2组病人糖尿病、骨折至手术时间、手术用时、术中低体温、局部潮湿、体位移动比较,差异有统计学意义(P < 0.05)。多因素logistic回归分析显示,糖尿病、手术用时越长、术中低体温、局部潮湿、体位移动为老年髋部骨折病人术中压力性损伤的独立危险因素(P < 0.05)。建立风险预测列线图模型并经ROC分析发现,糖尿病、手术用时、术中低体温、局部潮湿、体位移动及列线图模型的曲线AUC分别为0.629、0.754、0.638、0.629、0.698、0.905。DeLong法检验显示,列线图模型的曲线AUC高于各单项因素的AUC(P < 0.05)。当取Cut–off值时,列线图模型的灵敏度、特异度分别为0.905、0.810。Bootstrap(B = 1000)内部验证显示,列线图模型有良好稳定性,且决策分析发现,列线图模型有正向净收益率。
      结论: 糖尿病、手术用时越长、术中低体温、局部潮湿、体位移动为老年髋部骨折病人术中压力性损伤的独立危险因素,所构建的列线图模型具有较高的预测准确性和临床决策价值。

       

      Abstract:
      Objective To investigate the influencing factors of intraoperative stress injury in elderly patients with hip fractures, and establish a preventive risk prediction nomogram model.
      Methods A total of 252 elderly patients who underwent hip fracture surgery from June 2022 to December 2024 were selected, and divided into the occurrence group (n = 42) and non-occurrence group (n = 210) according to whether stress injury occurred during the operation. The influencing factors of intraoperative stress injury in elderly patients with hip fractures were analyzed by multivariate logistic regression and differential analysis. The nomogram model for predicting the risk of intraoperative pressure injury was established, the ROC curve was used to analyze the efficiency of nomogram model, and its prediction stability and decision-making efficiency were verified.
      Results The differences of the diabetes, fracture to operation time, operation time, intraoperative hypothermia, local moisture and body position movement were statistically significant between two groups (P < 0.05). The results of multivariate logistic regression analysis showed that the diabetes mellitus, longer operation time, intraoperative hypothermia, local moisture and body position movement were the independent risk factors of intraoperative stress injury in elderly patients with hip fractures (P < 0.05). The risk prediction nomogram model and ROC analysis revealed that the curve AUC of diabetes, operation time, intraoperative hypothermia, local dampness, body position movement and nomogram model were 0.629, 0.754, 0.638, 0.629, 0.698 and 0.905, respectively. The results of DeLong test showed that the curve AUC of the nomogram model was higher than that of each single factor (P < 0.05). When the cut-off value is taken, the sensitivity and specificity of nomogram model were 0.905 and 0.810, respectively. Bootstrap (B = 1000) internal verification showed that the nomogram model had good stability, and the decision analysis found that the nomogram model had a positive net return rate.
      Conclusions Diabetes mellitus, longer operation time, intraoperative hypothermia, local moisture and body position movement are the independent risk factors of intraoperative stress injury in elderly patients with hip fracture. The constructed nomographic model has high prediction accuracy and clinical decision value.

       

    /

    返回文章
    返回