采用肠内营养支持的重症肺炎患者住院死亡风险因素分析

Risk factors of in-hospital death in severe pneumonia patients receiving enteral nutrition support

目的重症肺炎患者营养状态下降可能增加住院死亡率,肠内营养支持能够改善患者营养状态,同时具有相对容易管理、成本较低、严重并发症较少等优势。另一方面,胃潴留、胃内微生物异位等不良反应可能增加院内获得性肺炎的发生率,增加了患者预后的不确定性。目前尚无针对肠内营养支持的重症肺炎患者住院死亡的预测模型。研究旨在探究经肠内营养支持的重症肺炎患者住院死亡的风险因素,并建立此类患者预后的预测模型。方法该项单中心回顾性研究于2015年1月1日至2020年12月31日对北京协和医院住院治疗且采用肠内营养支持的重症肺炎患者进行分析,主要终点为医院死亡及非医嘱离院率。通过单因素和多因素逻辑回归分析确定独立风险因素,构建列线图评分模型,并绘制决策曲线分析曲线及临床决策曲线。结果共纳入632例采用肠内营养支持的重症肺炎患者。根据患者是否发生住院死亡,将患者分为死亡组和存活组,发现24个参数在分割后显著。以下9个参数:呼吸机使用时间、合并恶性增殖性疾病、住院期间血小板、凝血酶原时间最大值及谷丙转氨酶、血清白蛋白、血钠、血钾、血糖的最低水平是死亡率的独立预测因素。基于这些变量建立了风险预测评分模型:ROC曲结下面积为0.782;95%CI 0.744~0.819,模型一致性指数为0.772。并绘制校准曲线、决策曲线分析及临床影响曲线评估预测列线图在训练和测试集中的功能优度、准确性和适用性。结论研究描述了采用肠内营养支持的重症肺炎患者的临床特征,且开发了一种评分模型来明确危险因素并建立预后预测模型。

Objective The decline in nutritional status in patients with severe pneumonia may contribute to an increase in in-hospital mortality.Enteral nutrition support can improve the nutritional status of patients,and is relatively easy to manage,with low cost and fewer serious complications.On the other hand,adverse reactions such as gastric retention and gastric microbiota translocation may increase the incidence of nosocomial pneumonia and increase the uncertainty of patient prognosis.There is no predictive model for in-hospital death in severe pneumonia patients receiving enteral nutrition support.The objective of this study was to investigate the risk factors of in-hospital death in patients with severe pneumonia receiving enteral nutrition support and to establish a prognostic model for such patients.Methods This was a single-center retrospective study.Patients with severe pneumonia who were hospitalized in Peking Union Medical College Hospital and received enteral nutrition support were included from January 1,2015 to December 31,2020.The primary endpoints were in-hospital mortality rate and unordered discharge rate.The independent risk factors were determined using univariate and multifactorial logistic regression analysis,the nomogram scoring model was constructed,and the decision curve analysis(DCA)was performed.Results A total of 632 severe pneumonia patients who received enteral nutrition support were included.Patients were divided into death and survival groups according to the presence or absence of in-hospital death,and 24 parameters were found with significant differences between groups.Nine parameters were independent predictors of mortality,namely the duration of ventilator use,the presence of malignant hyperplasia diseases,the maximal levels of platelet and prothrombin during hospitalization,and the nadir levels of alanine aminotransferase,serum albumin,sodium,potassium,and blood glucose.Based on these variables,a risk prediction scoring model was established(ROC=0.782;95%CI:0.744 to 0.819,concordance index:0.772).Calibration curves,DCA,and clinical impact curve were plotted to evaluate the goodness of function,accuracy,and applicability of the predictive nomogram,using the training and test sets.Conclusion This study summarized the clinical characteristics of patients with severe pneumonia receiving enteral nutrition support and developed a scoring model to identify risk factors and establish prognostic models.