消化道穿孔术后重症监护室患者的死亡相关危险因素分析

Analysis of death related risk factors in intensive care unit after gastrointestinal perforation

目的探讨影响消化道穿孔(DTP)术后转入重症监护室患者的死亡相关因素,建立预测模型并对模型的预测效能进行评估。方法回顾性分析2021年1月—2024年1月在山西白求恩医院(山西医学科学院)行DTP手术且术后转入重症监护室的306例患者病历资料,其中男性176例,女性130例,年龄28~92岁,平均(66.07±16.03)岁。根据预后分为生存组(n=264)和死亡组(n=42),比较两组患者的临床特征,使用单因素及多因素Logistic回归分析围手术期死亡的危险因素,并选取相关危险因素建立列线图预测模型,绘制受试者工作曲线,计算曲线下面积(AUC),评估其预测效能;进一步运用校准图及临床决策曲线分析评估模型的预测准确性及临床应用价值。结果死亡组在年龄、白细胞计数、降钙素原、乳酸水平、术前休克、术前基础疾病(脑梗死、激素史)、术中出血量、术后肺部感染等方面均高于生存组(P<0.05),而在血红蛋白方面低于生存组(P<0.05);多因素Logistic回归分析显示年龄(OR=1.422,95%CI:1.205~1.680,P<0.001)、血红蛋白(OR=0.945,95%CI:0.904~0.987,P=0.012)、白细胞计数(OR=1.832,95%CI:1.341~2.501,P<0.001)、降钙素原(OR=1.099,95%CI:1.012~1.192,P=0.024)、乳酸水平(OR=16.435,95%CI:3.729~72.425,P<0.001)、术前休克(OR=172.358,95%CI:13.059~2274.773,P<0.001)、术中出血量(OR=1.041,95%CI:1.017~1.065,P=0.001)及术后肺部感染(OR=38.670,95%CI:3.449~433.553,P=0.003)是DTP术后重症监护患者围术期死亡的独立危险因素,基于筛选的相关独立危险因素(P<0.05),建立列线图模型并绘制受试者工作特征(ROC)曲线,该曲线下模型面积0.985,较准图显示该模型预测结果和实际临床结果具有较高一致性,临床决策曲线分析提示该模型具有较高的临床预测价值。结论年龄>71.5岁、血红蛋白<10^(9)g/L、白细胞计数>17.9×10^(9)/L、降钙素原>6.225 ng/mL、乳酸水平>2.25 mmol/L、术前休克、术中出血量>45 mL及术后肺部感染是DTP术后重症监护患者围手术期死亡的独立危险因素。

Objective To investigate the mortality-related factors affecting patients with gastrointestinal perforation who are transferred to the intensive care unit(ICU)and to establish a prediction model,and to evaluate the predictive performance of the model.Methods A retrospective analysis was performed on the medical records of 306 patients who underwent gastrointestinal perforation surgery in Shanxi Bethune Hospital(Shanxi Academy of Medical Sciences)from January 2021 to January 2024 and were transferred to intensive care unit after surgery,including 176 males and 130 females,aged from 28 to 92 years with the average of(66.07±16.03)years.According to the prognosis,patients were divided into survival group(n=264)and death group(n=42).Clinical characteristics of the two groups were compared,univariate and multivariate Logistic regression was used to analyze the risk factors of perioperative death,and the related risk factors were selected to establish a nomogram prediction model,the subject work curve was drawn,and the area under the curve(AUC)was calculated.Evaluate its predictive effectiveness;The calibration chart and clinical decision curve were further used to evaluate the prediction accuracy and clinical application value of the model.Results Clinical data analysis showed that age,white blood cell count,procalcitonin,lactic acid level,preoperative shock,preoperative underlying diseases(cerebral infarction,hormone history),intraoperative blood loss,postoperative lung infection in the death group were higher than those in the survival group(P<0.05),and hemoglobin was lower than those in the survival group(P<0.05).Multivariate Logistic regression analysis showed age(OR=1.422,95%CI:1.205-1.680,P<0.001),hemoglobin(OR=0.945,95%CI:0.904-0.987,P=0.012),white blood cell count(OR=1.832,95%CI:1.341-2.501,P<0.001),procalcitonin(OR=1.099,95%CI:1.012-1.192,P=0.024),lactic acid level(OR=16.435,95%CI:3.729-72.425,P<0.001),reoperative shock(OR=172.358,95%CI:13.059-2274.773,P<0.001),intraoperative blood loss(OR=1.041,95%CI:1.017-1.065,P=0.001)and postoperative pulmonary infection(OR=38.670,95%CI:3.449-433.553,P=0.003)was an independent risk factor for perioperative death in intensive care patients after DTP.Based on the screened independent risk factors(P<0.05),a nomogram model was established and receiver operating characteristic(ROC)curve was drawn.The model area under the curve was 0.985.The accurate graph shows that the predicted results of the model are in good agreement with the actual clinical results,and the analysis of clinical decision curve indicates that the model has high clinical prediction value.Conclusion Age>71.5 years,hemoglobin<10^(9)g/L,white blood cell count>17.9×10^(9)/L,procalcitonin>6.225 ng/mL,lactate level>2.25 mmol/L,preoperative shock,intraoperative blood loss>45 mL and postoperative pulmonary infection are independent risk factors for perioperative death in intensive care patients after DTP.