Analysis of risk factors for enteral nutrition feeding intolerance in ICU patients with mechanical ventilation and construction of risk prediction model
ICU机械通气患者肠内营养喂养不耐受危险因素分析及风险预测模型构建

摘要

目的 对ICU机械通气患者在进行肠内营养喂养过程中出现喂养不耐受（ENFI）危险因素进行分析，并针对风险预测模型构建。方法 本研究采用便利抽样的方法，选取2023年8月至2024年12月在新疆维吾尔自治区人民医院外科重症监护室住院的机械通气患者作为研究对象，共计572例。按照纳入病例是否出现喂养不耐受划分为ENFI组、非ENFI组，最危险因素进行分析。并按照7:3的比例随机分配为训练样本集和检验样本集。建模数据400例，建立决策树模型；检验数据172例，测试决策树模型，针对风险预测模型构建并验证。结果 导致该部分患者喂养不耐受的因素包括平均动脉压、白蛋白、血糖、使用抗生素、使用镇静镇痛药、使用血管收缩药、机械通气模式、PEEP、肠内营养方式、营养液种类、营养液泵入速度、床头抬高角度，P<0.05。风险预测模型构建，可以较为准确对该部分患者出现ENFI风险进行预测。结论 导致ICU机械通气患者肠内营养喂养不耐受危险因素较多，会直接影响到患者恢复，在针对该部分患者进行治疗的过程中，需结合该部分患者病症特点，做好各方面管理工作，针对ENFI进行预防。

Abstract

Objective To analyze the risk factors of feeding intolerance (ENFI) in ICU patients with mechanical ventilation during enteral nutrition feeding and to construct a risk prediction model.
Methods This study adopted the method of convenience sampling to select 572 patients with mechanical ventilation who were hospitalized in the surgical intensive care unit of Xinjiang Uygur Autonomous Region People's Hospital from August 2023 to December 2024 as the research subjects. The included cases were divided into the ENFI group and the non-ENFI group according to whether feeding intolerance occurred, and the most risk factors were analyzed. And they were randomly assigned as the training sample set and the test sample set in a ratio of 7:3. Model 400 cases of data and establish a decision tree model; A total of 172 test data were examined, the decision tree model was tested, and the risk prediction model was constructed and verified.
Results The factors leading to feeding intolerance in this group of patients included mean arterial pressure, albumin, blood glucose, the use of antibiotics, the use of sedatives and analgesics, the use of vasoconstrictors, mechanical ventilation mode, PEEP, enteral nutrition method, type of nutrient solution, pumping speed of nutrient solution, and the Angle of headstock elevation, P<0.05. The construction of a risk prediction model can accurately predict the risk of ENFI in this group of patients.
Conclusion   There are many risk factors leading to enteral nutrition feeding intolerance in ICU patients with mechanical ventilation, which will directly affect the recovery of patients. During the treatment of these patients, it is necessary to combine the characteristics of their diseases and do a good job in all aspects of management, and prevent ENFI.