天津医药 ›› 2025, Vol. 53 ›› Issue (3): 287-292.doi: 10.11958/20241375

• 临床研究 • 上一篇    下一篇

全膝关节置换术后神经认知功能障碍的影响因素及列线图模型的构建

贾方(), 李平, 李静, 尹路   

  1. 天津市天津医院麻醉科(邮编300211)
  • 收稿日期:2024-09-18 修回日期:2024-11-27 出版日期:2025-03-15 发布日期:2025-03-31
  • 作者简介:贾方(1977),女,副主任医师,主要从事麻醉学的基础与临床方面研究。E-mail:jiafangjia77@163.com

Influencing factors of neurocognitive dysfunction after total knee arthroplasty and the construction of a nomogram model

JIA Fang(), LI Ping, LI Jing, YIN Lu   

  1. Department of Anesthesiology, Tianjin Hospital, Tianjin 300211, China
  • Received:2024-09-18 Revised:2024-11-27 Published:2025-03-15 Online:2025-03-31

摘要:

目的 探究全膝关节置换术(TKA)患者发生神经认知功能障碍(PND)的影响因素并构建列线图预测模型。方法 收集249例行TKA患者的临床资料并按照7∶3的比例将患者分为建模队列(n=174)和验证队列(n=75)。根据患者是否发生PND,将建模队列分为PND组(n=62)和非PND组(n=112),验证队列分为PND组(n=27)和非PND组(n=48)。分析TKA患者发生PND的影响因素,建立列线图预测模型并进行验证。结果 高衰弱评分、全身麻醉、单一镇痛方案、高疼痛视觉模拟评分(VAS)是PND的危险因素,高白蛋白和血红蛋白水平是PND的保护因素(均P<0.05)。构建列线图模型,预测建模队列和验证队列发生PND的曲线下面积(AUC)及其95%CI分别为0.875(0.824~0.925)、0.905(0.835~0.975);校准曲线分析结果发现,预测患者发生PND的风险和实际发生情况基本吻合,Hosmer-Lemeshow结果显示模型拟合较好;临床决策曲线结果显示该模型临床应用价值较高。结论 高衰弱评分、全身麻醉、单一镇痛方案、高VAS评分、低白蛋白和低血红蛋白水平可增加TKA患者发生PND的风险,本研究构建的列线图预测模型有助于临床识别PND高危人群。

关键词: 关节成形术,置换,膝, 认知功能障碍, Logistic模型, 列线图

Abstract:

Objective To explore the influencing factors of neurocognitive dysfunction (PND) in patients with total knee arthroplasty (TKA) and construct a nomogram prediction model. Methods The clinical data of 249 patients with TKA admitted to our hospital from December 2022 to July 2024 were selected. Patients were divided into the modeling queue (n=174) and the verification queue (n=75) according to the ratio of 7∶3. At the same time, according to whether patients developed PND or not, the modeling queue was divided into the PND group (n=62) and the non-PND group (n=112), and the verification queue was divided into the PND group (n=27) and the non-PND group (n=48). The influencing factors of PND in TKA patients were analyzed, and a nomogram prediction model was established to verify. Results High debilitation score, general anesthesia, single analgesic regimen and high VAS score were risk factors of PND, while high albumin and hemoglobin levels were protective factors of PND (P<0.05). The nomogram model was constructed, and the AUC of PND in the modeling queue and the verification queue was 0.875 (95%CI: 0.824-0.925) and 0.905 (95%CI:0.835-0.975) respectively. The results of calibration curve analysis showed that the predicted risk of PND was basically consistent with the actual situation, and the results of Hosmer-Lemeshow showed that the model fit well. The results of clinical decision curve showed that the model had high clinical application value. Conclusion High debilitating score, general anesthesia, single analgesic regimen, high VAS score and low albumin and hemoglobin levels can increase the risk of PND in TKA patients. The nomogram prediction model constructed in this study is helpful for clinical identification of high-risk groups of PND.

Key words: arthroplasty, replacement, knee, cognitive dysfunction, Logistic models, nomograms

中图分类号: