摘要
目的构建低温等离子扁桃体切除术后出血的预测模型, 并进行模型的验证。方法回顾性分析2018年1月至2022年12月在陕西省第二人民医院接受低温等离子扁桃体切除术治疗的患者243例的临床资料, 对术后出血患者的相关因素进行单因素、多因素分析, 采用列线图的方法构建低温等离子扁桃体切除术后出血的预测模型, 并进行预测效能分析。结果 243例患者中, 男147例, 女96例;年龄(25.94±11.47)岁。纳入研究的患者中, 共有21例发生术后出血, 术后出血发生率为8.64%。单因素分析结果显示, 患者的年龄、扁桃体包埋程度、既往出血史、术后继发呼吸道感染、术后进食不当可能与低温等离子扁桃体切除术后出血有关(t=-4.73、χ^(2)=6.58、16.55、14.49、13.19, 均P<0.05)。logistic多因素分析结果显示, 年龄、既往出血史、术后继发呼吸道感染是低温等离子扁桃体切除术后出血的独立影响因素(OR=0.911、4.154、6.891, 均P<0.05)。依据多因素分析所筛选出来的变量构建列线图风险模型, C-index为0.713。利用回归模型的独立影响因素以及P值预测概率, 采用受试者操作特征曲线(ROC曲线)对低温等离子扁桃体切除术后出血的情况的列线图模型预测效能实施分析, 约登指数分别为0.53、0.40、0.29、0.64。结论年龄、既往出血史和术后继发呼吸道感染是低温等离子扁桃体切除术后出血的独立影响因素, 据此建立的列线图模型具有较为良好的预测效能, 有望为临床工作提供有效的指导。
Objective To construct a predictive model for postoperative bleeding after low-temperature plasma tonsillectomy and validate the model.Methods A retrospective analysis was conducted on the clinical data of 243 patients who underwent low-temperature plasma tonsillectomy at the Second People's Hospital of Shaanxi Province from January 2018 to December 2022.Univariate and multivariate analyses were conducted on the relevant factors in patients experiencing postoperative bleeding.A predictive model for postoperative bleeding after low-temperature plasma tonsillectomy was constructed using a nomogram method,and predictive performance analysis was conducted.Results Among the 243 patients,there were 147 males and 96 females.Their average age was(25.94±11.47)years.Among the included patients,21 experienced postoperative bleeding,with an incidence rate of 8.64%.The results of univariate analysis showed that the patient's age,degree of tonsil embedding,history of bleeding,postoperative secondary respiratory infections,and improper eating may be related to postoperative bleeding after low-temperature plasma tonsillectomy(t=-4.73),χ^(2)=6.58,16.55,14.49,13.19,all P<0.05).The results of logistic multivariate analysis showed that age,previous bleeding history,and postoperative secondary respiratory infections were independent influential factors for postoperative bleeding after low-temperature plasma tonsillectomy(OR=0.911,4.154,6.891,all P<0.05).Based on the variables selected from multivariate analyses,a nomogram model was constructed,achieving a C-index of 0.713.Utilizing the independent influential factors and the prediction probability derived from the P-values of the regression model,a receiver operating characteristic curve was created to assess the predictive performance of the nomogram model in relation to postoperative bleeding following a low-temperature plasma tonsillectomy.The Jordan indices were 0.53,0.40,0.29,and 0.64,respectively.Conclusion Age,previous bleeding history,and postoperative secondary respiratory infections are independent influential factors for postoperative bleeding after low-temperature plasma tonsillectomy.The nomogram model established based on this finding has good predictive efficacy and is expected to provide effective guidance for clinical work.
作者
何珍
张滟
He Zhen;Zhang Yan(Department of Otolaryngology,Second People's Hospital of Shaanxi Province,Xi'an 710005,Shaanxi Province,China)
出处
《中国基层医药》
CAS
2024年第7期994-999,共6页
Chinese Journal of Primary Medicine and Pharmacy
关键词
扁桃体切除术
手术后出血
预测
比例危险度模型
列线图
Tonsillectomy
Postoperative hemorrhage
Forecasting
Proportional hazards models
Nomograms
