摘要
目的:分析急性上消化道出血(Acute Upper Gastrointestinal Bleeding,AUGIB)住院患者不良预后风险因素,构建AUGIB患者短期不良预后风险预测模型,为AUGIB管理及风险分层提供参考.方法:回顾性分析由2016年9月至2022年6月来自广州医科大学附属第二医院由急诊入院的AUGIB患者,按7∶3分为训练集及验证集.通过Lasso回归、多因素Logistic回归构建预测模型并进行预测能效评估.结果:消化系统恶性肿瘤、休克指数>1、GCS评分<14分、血浆白蛋白<24 g/L、血钙<1.95 mmol/L和血肌酐>103 μmol/L与AUGIB患者预后不良相关.使用上述变量组成的评分量表,与GBS评分在预测不良预后上有统计学差异(0.826 vs 0.700,P= 0.002),与AIMS 65评分无统计学差异(0.826 vs 0.768,P=0.260).结论:由消化系统恶性肿瘤、休克指数>1、GCS评分<14分、血浆白蛋白<24 g/L、血钙<1.95 mmol/L以及血肌酐>103 μmol/L五个变量构建的预测模型对于AUGIB患者不良预后具有良好的预测能效,其预测价值与AIMS 65评分相当,可能优于GBS评分.
Abstract
Objective:To analyze the risk factors for adverse prognosis in hospitalized patients with Acute Upper Gastrointestinal Bleeding(AUGIB)and to develop a short-term adverse outcome risk prediction model for AUGIB pa-tients,providing reference for the management and risk stratification of AUGIB.Methods:A retrospective analysis was conducted on AUGIB patients admitted through the emergency department at the Second Affiliated Hospital of Guang-zhou Medical University from September 2016 to June 2022.They were divided into a training set and a validation set in a 7∶3 ratio.The predictive model was constructed using Lasso regression and multiple factor logistic regression,and its predictive performance was evaluated.Results:Malignant digestive system tumors,shock index>1,GCS score<14,plasma albumin<24 g/L,blood calcium<1.95 mmol/L,and blood creatinine>103 μmol/L were associated with adverse prognosis in AUGIB patients.A scoring system composed of the above variables was compared with GBS and AIMS 65 scores.The results showed statistical differences between the scoring system and GBS score in predicting adverse out-comes(0.826 vs 0.700,P=0.002),but no statistical difference was observed between the scoring system and AIMS 65 score(0.826 vs 0.768,P=0.260).Conclusion:The predictive model based on five variables,including malignant diges-tive system tumors,shock index>1,GCS score<14,plasma albumin<24 g/L,blood calcium<1.95 mmol/L,and blood creatinine>103 μmol/L,demonstrated good predictive performance for adverse prognosis in AUGIB patients.Its predictive value is comparable to that of the AIMS 65 score and may be superior to the GBS score.
基金项目
广东省广州市急诊医学重点学科项目(2021-2023)()
NETL
NSTL
维普
万方数据