重型颅脑损伤患者胃肠功能衰竭风险预测模型的建立  

Establishment of a risk prediction model for gastrointestinal failure in patients with severe head injury

作　　者：林莎莎[1] 周灵静 唐秀琼 潘梅英[1] Lin Sasha;Zhou Lingjing;Tang Xiuqiong;Pan Meiying(Department of Neurosurgery,Dazhou Integrated Traditional Chinese and Western Medicine Hospital,Sichuan 635000,China)

机构地区：[1]达州市中西医结合医院神经外科,四川635000

出　　处：《脑与神经疾病杂志》2024年第7期428-433,共6页Journal of Brain and Nervous Diseases

基　　金：四川省中医药管理局科学技术研究专项课题(2021MS006)。

摘　　要：目的建立重型颅脑损伤(sTBI)患者胃肠功能衰竭风险预测模型,并进行验证。方法选择2019年1月至2023年3月达州市中西医结合医院神经外科接诊的147例sTBI患者资料进行分析。收集可能影响sTBI患者胃肠功能衰竭的相关因素,根据急性胃肠道损伤(AGI)分级将患者分为胃肠功能衰竭组与非胃肠功能衰竭组,以LASSO筛选出潜在影响因素后行多因素Logistic回归分析,根据多因素分析结果建立列线图模型并进行验证。结果本研究模型组102例sTBI患者中共有33例(32.35%)出现胃肠功能衰竭,多因素Logistic回归分析结果显示:机械通气情况、休克、通便措施、肺部感染情况、血钠、人肠脂肪酸结合蛋白(iFABP)、D-乳酸及二胺氧化酶(DAO)为sTBI患者胃肠功能衰竭的独立性影响因素(P<0.05)。ROC分析显示模型组曲线下面积(AUC)为0.942,95%CI为0.896~0.988;验证组AUC为0.908,95%CI为0.849~0.966。校准曲线结果显示,预测曲线与标准曲线基本拟合。决策曲线分析结果显示,当列线图模型预测sTBI患者胃肠功能衰竭风险的概率阈值为0.10~0.90时,患者的净收益率大于0。结论sTBI患者胃肠功能衰竭主要受机械通气情况、休克、通便措施等相关因素的影响,本研究基于上述因素建立的列线图预测模型用于预测sTBI患者胃肠功能衰竭具有较高的准确度与区分度。Objective To establish a risk prediction model for gastrointestinal failure in patients with severe head injury and validate it.Methods A retrospective analysis was conducted on the data of 147 patients with severe traumatic brain injury(sTBI)admitted to the neurosurgery department of Dazhou Integrated Traditional Chinese and Western Medicine Hospital from January 2019 to March 2023.Collect relevant factors that may affect gastrointestinal failure in sTBI patients,and divide them into gastrointestinal failure group and non gastrointestinal failure group based on acute gastrointestinal injury(AGI)grading.Use LASSO regression to screen out potential influencing factors and conduct multivariate Logistic regression analysis.Based on the results of multivariate analysis,establish a column chart model and validate it.Results A total of 33(32.35%)of the 102 patients with severe head injury in this study model group experienced gastrointestinal failure.The results of multivariate Logistic regression analysis showed that mechanical ventilation,shock,defecation measures,pulmonary infection,blood sodium,human intestinal fatty acid binding protein(IFABP),D-lactate,and diamine oxidase(DAO)were independent influencing factors for gastrointestinal failure in sTBI patients(P<0.05).ROC analysis showed that the area under the curve(AUC)of the model group was 0.942,with a 95%CI of 0.896-0.988.The AUC of the validation group was 0.908,with a 95%CI of 0.849-0.966.The calibration curve results show that the predicted curve basically fits the standard curve.The decision curve analysis results show that when the probability threshold of the column graph model predicting the risk of gastrointestinal failure in sTBI patients is 0.10-0.90,the net return rate of patients is greater than 0.Conclusion Gastrointestinal failure in sTBI patients is mainly influenced by relevant factors such as mechanical ventilation,shock,and bowel movements.The column chart prediction model established based on these factors in this study has high accuracy and

关 键 词：重型颅脑损伤 胃肠功能衰竭 多因素分析 列线图模型 

分 类 号：R651[医药卫生—外科学]