体外循环心脏瓣膜置换术后低蛋白血症相关危险因素与受试者工作特征曲线预测模型建立OA

Objective To investigate the risk factors associated with hypoproteinemia after cardiac valve replacement with cardiopulmonary bypass(CPB),and to establish a predictive model using receiver operating characteristic(ROC)curve analysis.Methods A total of 250 patients undergoing cardiac valve replacement under CPB in our hospital were enrolled.According to serum albumin(ALB)levels within 48 hours postoperatively,patients were divided into a hypoproteinemia group(n=65)and a non-hypoproteinemia group(n=185).Multivariate logistic regression was used to analyze risk factors,and ROC curve was applied to evaluate predictive performance and construct a predictive model.Results Multivariate logistic regression analysis revealed that advanced age(OR=1.037),female(OR=2.147),fatty liver disease(OR=4.402),prolonged lengthy CPB time(OR=1.016),extended postoperative cannulation(OR=1.182),pulmonary infection(OR=6.552),and gastrointestinal im-pairment(OR=3.957)were independent risk factors for hypoproteinemia following CPB cardiac valve replacement(P<0.05).ROC curve analysis demonstrated that the combined predictive model achieved an area under the curve(AUC)of 0.850(95%CI of 0.791-0.909),with a sensitivity of 0.846,and specificity of 0.762,indicating high predictive value.Conclusion Advanced age,female,fatty liver disease,prolonged CPB time,extended postoperative catheterization time,pulmonary infection and gas-trointestinal dysfunction are significant risk factors for hypoproteinemia after CPB cardiac valve replacement.