心脏瓣膜置换术后营养不良风险预测模型的构建

Construction of the risk prediction model of malnutrition after heart valve replacement

目的:构建心脏瓣膜置换术后营养不良风险预测模型,为护理干预提供策略支持。方法:采用便利抽样法,选取2019年11月至2021年10月入院并行人工心脏瓣膜置换术病人83例,根据病人术后第1天血清白蛋白检测值,分为观察组和对照组,收集2组一般资料和生化营养指标。应用NRS 2002评估2组病人手术前和手术后营养风险。采用logistic回归分析构建心脏瓣膜置换术后营养不良风险预测模型。结果:心脏瓣膜术后共32例(38.6%)病人发生营养不良。单因素分析显示观察组与对照组在病程、置换部位、体质量指数(BMI)、血清白蛋白(ALB)、血红蛋白、前白蛋白、总胆固醇、术前NRS 2002评分和术后NRS 2002评分比较,差异均有统计学意(P<0.05~P<0.01);构建模型方程为:LogitP=-4.132+3.512BMI+4.124ALB+4.911术前NRS 2002+5.109术后NRS 2002。BMI、ALB值和术前、术后NRS 2002评分是心脏瓣膜置换术后营养不良风险的独立预测因素(P<0.01)。结论:成功构建心脏瓣膜置换术后营养不良风险预测模型,可以利用预测模型对瓣膜置换术的病人进行营养风险预测。

Objective:To construct a risk prediction model of malnutrition after heart valve replacement,and provide the strategic support for nursing intervention.Methods:A total of 83 patients treated with prosthetic heart valve replacement from November 2019 to October 2021 were selected by convenience sampling method.According to the serum albumin(ALB) detection value on the first day after operation,the patients were divided into the case group and control group,and the general information and biochemical nutrition indexes of two groups were collected.The NRS 2002 was used to assess the nutritional risk in two groups before and after surgery.The risk prediction model of malnutrition after heart valve replacement was established by logistic regression analysis.Results:A total of 32 patients(38.6%) suffered from malnutrition after heart valve replacement.The results of univariate analysis showed that the differences of the course of disease,replacement site,body mass index(BMI),serum ALB,hemoglobin,prealbumin,total cholesterol and preoperative and postoperative NRS 2002 scores were statistically significant between two groups(P<0.05 toP< 0.01).The model equation was LogitP =-4.132 + 3.512BMI + 4.124ALB + 4.911 preoperative NRS 2002 + 5.109postoperative NRS 2002.BMI,ALB value,and preoperative and postoperative NRS 2002 scores were the independent predictors of the risk of malnutrition after heart valve replacement(P< 0.01).Conclusions:A predicition model for the risk of malnutriton afte heart value replacemenat is successfully constructed which can be used to predict the nutritional risk of patients undergoing heart value replacement.