摘要
目的分析老年肺炎患者并发急性肾损伤(Acute kidney injury,AKI)的相关危险因素,构建老年肺炎患者并发AKI的列线图预测模型。方法以2020年1月-2023年12月安徽中医药大学第一附属医院收治的98例老年肺炎患者为研究对象。根据患者是否发生AKI进行分组,将发生AKI的患者纳入合并AKI组共48例,未发生AKI的患者纳入单纯性肺炎组共50例。收集患者入院后24 h内的相关临床数据。采用Logistic回归因素分析老年肺炎患者并发AKI的相关影响因素,采用rms程序包构建老年肺炎患者并发AKI的预测模型,应用受试者工作曲线(Receiver operating characteristic,ROC)和Hosmer-Lemeshow检验验证列线图模型的预测效能。结果与单纯性肺炎组比较,合并AKI组患者患有基础性疾病占比、血乳酸(lac)、白细胞计数(WBC)、血小板(PLT)、中性粒细胞计数(NE)、总胆红素(TBil)、血肌酐(Scr)、尿素氮(BUN)、胱抑素(CysC)、C-反应蛋白(CRP)及降钙素原(PCT)水平较高,差异有统计学意义(P<0.05)。Logistic多因素分析显示,血清中SCr、BUN、CysC、CRP及PCT水平均为影响老年肺炎患者发生AKI的相关因素(P<0.05)。根据Logistic多因素分析结果建立预测老年肺炎患者并发AKI的列线图模型,并对模型进行验证,ROC曲线分析结果显示,预测模型C-index指数为0.905(95%CI:0.832~0.963);经Hosmer-Lemeshow检验列线图模型实测值与预测值基本一致,经决策曲线分析列线图预测价值较高,具有较好的临床有效性。结论血清中SCr、BUN、CysC、CRP及PCT水平均是老年肺炎患者发生AKI的相关影响因素,老年肺炎患者并发AKI的列线图模型对预后具有较好的预测价值。
Objective The risk factors related to acute kidney injury(AKI)in elderly patients with pneumonia were analyzed,and a nomogram prediction model for AKI in elderly patients with pneumonia was established.Methods A total of 98 elderly patients with pneumonia admitted to the hospitalfrom January 2020 to December 2023 were studied.Thepatients were grouped according to whether they had AKI.48 patients with AKI were included in the combined AKI group,and 50 patients without AKI were included in the simple pneumonia group.Clinical data were collected within 24 h after admission.Logistic regression factors were used to analyze the related influencing factors of AKI in elderly patients with pneumonia.rms package was used to construct a prediction model of AKI in elderly patients with pneumonia.Receiver Operating Characteristic was applied.ROC and Hosmer-Lemeshow tests verified the predictive efficiency of the nomogram model.Results Compared with simple pneumonia group,the patients with AKI had the proportion of underlying diseases,blood lactic acid(lac),white blood cell count(WBC),platelets(PLT),neutrophil count(NE),total bilirubin(TBil),blood creatinine(Scr),urea nitrogen(BUN),cystatin(CysC),C-reactive protein(CRP)and procalcitonin(PCT)water The difference was statistically significant(P<0.05).Logistic multivariate analysis showed that serum levels of SCr,BUN,CysC,CRP and PCT were related factors affecting AKI in elderly patients with pneumonia(P<0.05).A nomogram model for predicting AKI in elderly patients with pneumonia was established according to the Logistic multivariate analysis results,and the model was verified.ROC curve analysis results showed that the C-index of the prediction model was 0.905(95%CI:0.832-0.963).According to Hosmer-Lemeshow test,the measured values of the nomogram model were basically consistent with the predicted values,and the nomogram analysis by decision curve had high predictive value and good clinical effectiveness.Conclusions Serum levels of SCr,BUN,CysC,CRP and PCT are related factors to the occurrence of AKI in elderly patients with pneumonia,and the column-graph model of elderly patients with pneumonia complicated with AKI has good predictive value.
作者
殷俊
马娣
洪峰
周大勇
YIN Jun;MA Di;HONG Feng;ZHOU Dayong(Department of Intensive Care Medicine,First Affiliated Hospital of Anhui University of Traditional Chinese Medicine,Hefei 230031,China)
出处
《新疆医科大学学报》
2025年第3期317-321,327,共6页
Journal of Xinjiang Medical University
基金
安徽省临床医学研究转化专项(202304029)。
关键词
老年
肺炎
急性肾损伤
预测模型
列线图
old age
pneumonia
acute kidney injury(AKI)
prediction model
nomograph
