实用医学杂志 ›› 2022, Vol. 38 ›› Issue (13): 1652-1656.doi: 10.3969/j.issn.1006⁃5725.2022.13.016

• 临床研究 • 上一篇    下一篇

髋关节翻修术后输血的危险因素分析及列线图预测模型建立

洪尉师1 张亚鑫1 林群1 马艺博1 孙钰2    

  1. 1 大连医科大学(辽宁大连 116000);2 江苏省苏北人民医院关节外科(江苏扬州 225000)

  • 出版日期:2022-07-10 发布日期:2022-07-10
  • 通讯作者: 孙钰 E⁃mail:docsunyu@126.com
  • 基金资助:
    江苏省自然科学基金(编号:BK20201221)

An analysis on the risk factors of blood transfusion after hip revision and establishment of a nomogram prediction model

HONG Weishi*,ZHANG Yaxin,LIN Qun,MA Yibo,SUN Yu.    

  1. Dalian Medical University,Da⁃ lian 116000,China 

  • Online:2022-07-10 Published:2022-07-10
  • Contact: SUN Yu E⁃mail:docsunyu@126.com

摘要:

目的 探讨髋关节翻修患者围手术期输血的危险因素,为临床上早期筛选术后高风险输血 患者提供指导意义。方法 纳入自 2012 8 月至 2021 3 月行髋关节翻修术的 162 例患者,根据术后是 否输血分为输血组 88 例,未输血组 74 例,输血率 54.32%。收集两组患者年龄、性别、BMI、高血压、糖尿 病、骨质疏松、术前红细胞、术前血红蛋白、术前血小板、术前PT、术前APTT、术前INR、术前TT、术前血钙、 术前白蛋白、ASA评分和术后异体输血等资料,利用SPSS软件进行多因素二元logistic回归分析髋关节翻修患者术后输血的危险因素;通过 R 软件(4.1.1)构建列线图预测模型,并绘制出 ROC 曲线及校准曲线来评 价模型的区分度和标准度。结果 多因素二元logistic回归分析显示,高血压(OR = 2.249,P = 0.037)和骨质 疏松(OR = 3.927,P = 0.001)、术前血红蛋白(OR = 0.255,P = 0.026)和术前血钙水平(OR = 0.338,P = 0.038 是髋关节翻修患者术后输血的危险因素。通过 R 软件构建列线图预测模型并绘制出 ROC 曲线及校准曲 线,得到曲线下面积(AUC)为 0.751,校准曲线斜率接近 1,表明该预测模型预测术后输血风险具有较好的 区分度和准确度。结论 基于高血压、骨质疏松、术前血红蛋白以及术前血钙这四项独立危险因素建立 的列线图预测髋关节翻修患者术后输血风险具有较好的区分度和准确度。

关键词:

髋关节翻修术, 输血, 危险因素, 列线图

Abstract:

Objective To investigate the risk factors of perioperative blood transfusion in patients undergoing hip revision,and to provide guidance for early screening the patients at high risks of postoperative blood transfusion. Methods A total of 162 patients who had undergone hip revision surgery during the period from August 2012 to March 2021 were included in this study. They were divided into a transfusion group(88 patients)and a non⁃trans⁃ fusion group(74 patients)according to whether they received blood transfusion,with a transfusion rate of 54.32%. The data on age,gender,BMI,hypertension,diabetes,osteoporosis,preoperative red blood cells,preoperative hemoglobin,preoperative platelets,preoperative PT,preoperative APTT,preoperative INR,preoperative TT preoperative blood calcium,preoperative albumin,ASA score and postoperative allogeneic blood transfusion were collected from both groups. The SPSS software was used to analyze the risk factors of postoperative blood transfusion in patients with hip revision. The R software(4.1.1)was used to build a nomogram prediction model,and the ROC curve and calibration curve were drawn to evaluate the differentiation and standard degree of the model. Results Multivariate binary logistic regression analysis showed that hypertension(OR = 2.249,P = 0.037),osteoporosis (OR = 3.927,P = 0.001),preoperative hemoglobin(OR = 0.255,P = 0.026)and preoperative blood calcium levels (OR = 0.338,P = 0.038)were risk factors for postoperative blood transfusion in hip revision patients. R software was used to construct a nomogram prediction model and draw a ROC curve and a calibration curve. The area under the curve(AUC)was 0.751,and the slope of the calibration curve was close to 1,indicating that the prediction model had good differentiation and accuracy in predicting the risks of postoperative blood transfusion. Conclusions The nomogram has better differentiation and accuracy in predicting the risks of postoperative blood transfusion in hip revision patients,which is established based on five independent risk factors including hypertension,osteoporosis preoperative hemoglobin and preoperative blood calcium.

Key words: hip revision,  , blood transfusion,  , risk factors,  , nomogram