摘要
目的构建老年膝骨关节炎患者膝关节置换术后谵妄风险预测模型。方法采用回顾性队列研究,选取2021年2月至2023年2月西安医学院第二附属医院和西安市红会医院收治的98例膝骨关节炎患者,根据术后72 h内意识模糊评估法(CAM)结果分为谵妄组(n=35)和对照组(n=63);另选取2023年3月至2025年3月本院收治的98例膝骨关节炎患者临床资料作为验证集。比较两组患者的基本临床资料、实验室指标及手术指标;采用多因素Logistic回归分析对影响术后谵妄发生的因素进行分析;采用受试者操作特征(ROC)曲线分析独立危险因素诊断术后谵妄的效果。结果谵妄组患者平均年龄,糖尿病、痴呆、慢性阻塞性肺疾病病史人数占比及麻醉时长分别为(73.45±5.54)岁、57.14%、22.86%、11.43%、(190.54±20.43)min,均高于对照组[(70.77±5.37)岁、23.81%、4.76%、1.59%、(175.24±23.17)min],白蛋白水平为(33.86±4.25)g/L,低于对照组[(36.75±4.33)g/L],差异均有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,年龄增高、白蛋白水平降低、麻醉时长、糖尿病、痴呆均是影响术后谵妄发生的影响因素(P<0.05)。联合预测模型诊断谵妄的敏感度和特异度达到80.00%和70.33%,曲线下面积(AUC)值为0.790,显著高于年龄单独诊断(Z=3.65,P=0.001)、白蛋白单独诊断(Z=3.12,P=0.002)、麻醉时长单独诊断(Z=2.89,P=0.004)。在验证集中,联合预测模型敏感度为80.00%,特异度为68.97%,AUC为0.801,与模型集结果类似(Z=0.158,P=0.567)。结论构建年龄、低白蛋白水平、麻醉时长、糖尿病史和痴呆史5个因素的联合预测模型,对老年膝骨关节炎患者膝关节置换术后谵妄风险具有良好的预测效能,有助于实现围手术期风险的个体化评估与精准干预,从而改善患者预后。
Objective To construct a risk prediction model of delirium after knee replacement in elderly patients with knee osteoarthritis.Methods Retrospective cohort study was conducted to select 98 patients with knee osteoarthritis admitted to the Second Affiliated Hospital of Xi'an Medical University and Xi'an Honghui Hospital from February 2021 to February 2023.The results were divided into the delirium group(n=35)and the control group(n=63)according to the consciousness blurring assessment method(CAM)within 72 hours after surgery.Additionally,clinical data from 98 knee osteoarthritis patients admitted to our hospital from March 2023 to March 2025 were selected as a validation set.The basic clinical information,laboratory indicators,and surgical outcomes were compared between the two groups.Multivariate Logistic regression analysis was used to analyze the factors affecting the occurrence of postoperative delirium.The receiver operator characteristic(ROC)curve was used to analyze the effect of independent risk factors on the diagnosis of postoperative delirium.Results The average age,proportion of diabetes,proportion of dementia,proportion of chronic obstructive pulmonary disease history,and duration of anesthesia in the delirium group were(73.45±5.54)years,57.14%,22.86%,and 11.43%,(190.54±20.43)min,respectively,which were higher than those in the control group[(70.77±5.37)years,23.81%,4.76%,1.59%,(175.24±23.17)min],the albumin level was(33.86±4.25)g/L,which was lower than that in the control group[(36.75±4.33)g/L],the differences were statistically significant(P<0.05).Multivariate Logistic regression analysis demonstrated that factors including advanced age,decreased albumin levels,prolonged anesthesia duration,diabetes mellitus,and dementia significantly influenced postoperative delirium development(P<0.05).The sensitivity and specificity of the combined prediction model for diagnosing delirium were 80%and 70.33%,respectively,with an area under the curve(AUC)value of 0.790,which was significantly higher than that of the sole diagnosis of age(Z=3.65,P=0.001),albumin(Z=3.12,P=0.002),and anesthesia duration(Z=2.89,P=0.004).In the validation cohort,the combined diagnosis sensitivity was 80.00%and specificity was 68.97%,with AUC 0.801,similar to the model cohort results(Z=0.158,P=0.567).Conclusion A combined prediction model of age,low albumin level,duration of anesthesia,history of diabetes and history of dementia was constructed,which had a good predictive effect on the risk of delirium after knee arthroplasty in elderly patients with knee osteoarthritis,which was helpful to realize the individualized assessment and precise intervention of perioperative risk,so as to improve the prognosis of patients.
作者
王璐
赵永斌
李静
马世军
WANG Lu;ZHAO Yongbin;LI Jing(Department of Anesthesiology and Perioperative Medicine,The Second Affiliated Hospital of Xi'an Medical University,Xi'an Shaanxi 710038,China;Department of Anesthesiology,Xi'an Honghui Hospital,Xi'an Shaanxi 710054,China)
出处
《临床和实验医学杂志》
2025年第21期2283-2287,共5页
Journal of Clinical and Experimental Medicine
基金
陕西省重点研发计划项目(编号:2022SF-517)。
