老年患者急性鼻出血内镜下电凝止血列线图预测模型的构建与验证  

Establishment and validation of a nomogram predictive model for endoscopic electrocoagulation hemostasis in elderly patients with acute epistaxis

作　　者：郭明明 刘涵尧 王金辉[1] 欧阳顺林[1] Guo Mingming;Liu Hanyao;Wang Jinhui;Ouyang Shunlin(Department of Otolaryngology,the Third Affiliated Hospital of Guangzhou Medical University,Guangzhou,Guangdong 510000,China)

机构地区：[1]广州医科大学附属第三医院耳鼻喉科,广东广州510000

出　　处：《中国内镜杂志》2024年第2期79-85,共7页China Journal of Endoscopy

摘　　要：目的探讨老年患者急性鼻出血,需内镜下电凝止血术治疗的相关危险因素,建立列线图预测模型,并予以验证,以便治疗时能选择恰当的止血方式。方法收集2018年1月-2022年12月该科228例单侧急性鼻出血老年患者的临床资料。根据是否接受内镜下电凝止血术,分为电凝止血组(n=112)和保守填塞组(n=116),探讨内镜下电凝止血术的独立危险因素,建立列线图预测模型,绘制受试者操作特征曲线(ROC curve),使用校正曲线和临床决策曲线分析(DCA),评价模型的预测性能和一致性。结果单因素分析结果显示,年龄、高血压、心脑血管疾病、服用抗凝药物和出血部位与内镜下电凝止血有关(P<0.05);多因素二元Logistic回归分析显示,年龄增大、存在高血压、长期服用抗凝药物,以及出血部位位于后鼻区或位置不明的患者,接受内镜下电凝止血术的可能性更高(P<0.05);基于上述独立危险因素,建立预测老年患者急性鼻出血内镜下电凝止血风险的列线图模型,曲线下面积(AUC)为0.856(95%CI:0.805~0.907)。校正曲线和DCA显示,该列线图干预收益的阈值范围大。结论基于年龄、高血压、服用抗凝药物和出血部位建立的老年患者急性鼻出血内镜下电凝止血的列线图模型,预测能力和区分度好,对临床有一定的指导意义。Objective To investigate the relevant risk factors for endoscopic electrocoagulation hemostasis in elderly patients with acute epistaxis,and establish and validate a nomogram prediction model to facilitate early selection of appropriate hemostasis methods in clinical practice.Methods Clinical data of 228 elderly patients with unilateral acute epistaxis from January 2018 to December 2022 were collected.There were two groups,the electrocoagulation group(n=112)and the conservative packing group(n=116),based on whether they received endoscopic electrocoagulation hemostasis.Analysis was performed to explore the independent risk factors for requiring endoscopic electrocoagulation hemostasis.A nomogram prediction model was established based on the multivariate results,and receiver operator characteristic curve(ROC curve),calibration curve and clinical decision curve analysis(DCA)were used to evaluate the predictive performance and consistency of the model.Results According to the research results,the univariate analysis showed that age,hypertension,cardiovascular disease,anticoagulant use,and bleeding site were associated with endoscopic electrocoagulation hemostasis(P<0.05).The multivariate binary Logistic regression analysis revealed that older age,the presence of hypertension,long-term use of anticoagulants,and bleeding sites located in the posterior nasal region or unknown location were associated with a higher likelihood of undergoing endoscopic electrocoagulation hemostasis(P<0.05).Based on these independent risk factors,a nomogram model for predicting endoscopic electrocoagulation hemostasis for acute epistaxis in elderly patients was established,the area under the curve(AUC)was 0.856(95%CI:0.805~0.907).The calibration curve and DCA showed that the use of the nomogram model could benefit patients over a wide range of diagnostic threshold probabilities.Conclusion A nomogram model based on age,hypertension,anticoagulant use,and bleeding site to predict the risk of endoscopic electrocoagulation hemostasis in elderly

关 键 词：老年人 急性鼻出血 内镜下电凝止血术 列线图 预测模型 

分 类 号：R765.23[医药卫生—耳鼻咽喉科]