PDF(Pubmed)
Abstract：
UNASSIGNED: Pneumonia is a common infection in the intensive care unit (ICU), and gram-negative bacilli are the most common bacterial cause. The purpose of the study was to investigate the risk factors for 30-day mortality in patients with gram-negative bacillary pneumonia in the ICU, construct a predictive model, and stratify patients based on risk to assess their short-term survival.
UNASSIGNED: Patients admitted to the ICU with gram-negative bacillary pneumonia at Fujian Medical University Affiliated First Hospital between January 2018 and September 2020 were selected. Patients were divided into deceased and survivor groups based on whether death occurred within 30 days. Multifactorial logistic regression analysis was used to identify independent risk factors for 30-day mortality in these patients, and a predictive nomogram model was constructed based on these factors. Patients were categorized into low-, medium-, and high-risk groups according to the model\'s predicted probability, and Kaplan-Meier survival curves were plotted to assess short-term survival.
UNASSIGNED: The study included 305 patients. Lactic acid (odds ratio [OR], 1.524, 95% CI: 1.057-2.197), tracheal intubation (OR: 4.202, 95% CI: 1.092-16.169), and acute kidney injury (OR:4.776, 95% CI: 1.632-13.978) were identified as independent risk factors for 30-day mortality. A nomogram prediction model was established based on these three factors. Internal validation of the model showed a Hosmer-Lemeshow test result of X2=5.770, P=0.834, and an area under the ROC curve of 0.791 (95% CI: 0.688-0.893). Bootstrap resampling of the original data 1000 times yielded a C-index of 0.791, and a decision curve analysis indicated a high net benefit when the threshold probability was between 15%-90%. The survival time for low-, medium-, and high-risk patients was 30 (30, 30), 30 (16.5, 30), and 17 (11, 27) days, respectively, which were significantly different.
UNASSIGNED: Lactic acid, tracheal intubation, and acute kidney injury were independent risk factors for 30-day mortality in patients in the ICU with gram-negative bacillary pneumonia. The predictive model constructed based on these factors showed good predictive performance and helped assess short-term survival, facilitating early intervention and treatment.
UNASSIGNED: Patients admitted to the ICU with gram-negative bacillary pneumonia at Fujian Medical University Affiliated First Hospital between January 2018 and September 2020 were selected. Patients were divided into deceased and survivor groups based on whether death occurred within 30 days. Multifactorial logistic regression analysis was used to identify independent risk factors for 30-day mortality in these patients, and a predictive nomogram model was constructed based on these factors. Patients were categorized into low-, medium-, and high-risk groups according to the model\'s predicted probability, and Kaplan-Meier survival curves were plotted to assess short-term survival.
UNASSIGNED: The study included 305 patients. Lactic acid (odds ratio [OR], 1.524, 95% CI: 1.057-2.197), tracheal intubation (OR: 4.202, 95% CI: 1.092-16.169), and acute kidney injury (OR:4.776, 95% CI: 1.632-13.978) were identified as independent risk factors for 30-day mortality. A nomogram prediction model was established based on these three factors. Internal validation of the model showed a Hosmer-Lemeshow test result of X2=5.770, P=0.834, and an area under the ROC curve of 0.791 (95% CI: 0.688-0.893). Bootstrap resampling of the original data 1000 times yielded a C-index of 0.791, and a decision curve analysis indicated a high net benefit when the threshold probability was between 15%-90%. The survival time for low-, medium-, and high-risk patients was 30 (30, 30), 30 (16.5, 30), and 17 (11, 27) days, respectively, which were significantly different.
UNASSIGNED: Lactic acid, tracheal intubation, and acute kidney injury were independent risk factors for 30-day mortality in patients in the ICU with gram-negative bacillary pneumonia. The predictive model constructed based on these factors showed good predictive performance and helped assess short-term survival, facilitating early intervention and treatment.
摘要：
■肺炎是重症监护病房（ICU）的常见感染，革兰阴性杆菌是最常见的细菌病因。目的探讨ICU革兰阴性杆菌肺炎患者30d死亡的危险因素,构建一个预测模型，并根据风险对患者进行分层以评估其短期生存率。
■选择2018年1月至2020年9月在福建医科大学附属第一医院入住ICU的革兰阴性杆菌肺炎患者。根据是否在30天内死亡，将患者分为死亡组和幸存者组。多因素logistic回归分析用于确定这些患者30天死亡率的独立危险因素。并基于这些因素构建了预测列线图模型。患者被分类为低，medium-,根据模型的预测概率，高危人群，绘制Kaplan-Meier生存曲线以评估短期生存。
■该研究包括305名患者。乳酸(比值比[OR]，1.524，95%CI：1.057-2.197），气管插管（OR：4.202，95%CI：1.092-16.169），和急性肾损伤(OR:4.776,95%CI:1.632-13.978)被确定为30天死亡率的独立危险因素.基于这三个因素建立了列线图预测模型。模型的内部验证显示，Hosmer-Lemeshow检验结果X2=5.770，P=0.834，ROC曲线下面积为0.791（95％CI：0.688-0.893）。对原始数据进行1000次的Bootstrap重采样得出的C指数为0.791，当阈值概率在15％-90％之间时，决策曲线分析表明具有很高的净收益。生存时间低，medium-,高危患者为30（30，30），30(16.5,30),和17（11，27）天，分别,有很大的不同。
■乳酸，气管插管,急性肾损伤是ICU革兰阴性杆菌肺炎患者30天死亡的独立危险因素.基于这些因素构建的预测模型显示出良好的预测性能，并有助于评估短期生存率，促进早期干预和治疗。
■选择2018年1月至2020年9月在福建医科大学附属第一医院入住ICU的革兰阴性杆菌肺炎患者。根据是否在30天内死亡，将患者分为死亡组和幸存者组。多因素logistic回归分析用于确定这些患者30天死亡率的独立危险因素。并基于这些因素构建了预测列线图模型。患者被分类为低，medium-,根据模型的预测概率，高危人群，绘制Kaplan-Meier生存曲线以评估短期生存。
■该研究包括305名患者。乳酸(比值比[OR]，1.524，95%CI：1.057-2.197），气管插管（OR：4.202，95%CI：1.092-16.169），和急性肾损伤(OR:4.776,95%CI:1.632-13.978)被确定为30天死亡率的独立危险因素.基于这三个因素建立了列线图预测模型。模型的内部验证显示，Hosmer-Lemeshow检验结果X2=5.770，P=0.834，ROC曲线下面积为0.791（95％CI：0.688-0.893）。对原始数据进行1000次的Bootstrap重采样得出的C指数为0.791，当阈值概率在15％-90％之间时，决策曲线分析表明具有很高的净收益。生存时间低，medium-,高危患者为30（30，30），30(16.5,30),和17（11，27）天，分别,有很大的不同。
■乳酸，气管插管,急性肾损伤是ICU革兰阴性杆菌肺炎患者30天死亡的独立危险因素.基于这些因素构建的预测模型显示出良好的预测性能，并有助于评估短期生存率，促进早期干预和治疗。
