摘要
目的探讨甲型流行性感冒病毒性肺炎的早期诊断指标及相关危险因素。方法选取2016年11月至2018年11月北部战区总医院呼吸内科收治的101例确诊为成人社区获得性肺炎患者,男63例,女38例,年龄(66.38±18.13)岁,年龄范围为18~80岁。按肺炎类型不同,将患者分为病毒性肺炎组与细菌性肺炎组,病毒性肺炎组患者41例,男27例,女14例;细菌性肺炎组患者60例,男36例,女24例。收集患者白细胞计数、淋巴细胞计数、中性粒细胞计数、血小板计数、动脉氧分压、动脉血乳酸、尿素、肌酐、乳酸脱氢酶、肌酸激酶、C反应蛋白及纤维蛋白原。筛选出具有统计学意义的指标,后采用单因素logistic回归对上述指标进行分析并记录,分析得出具有统计学意义的指标则为与甲型流行性感冒病毒性肺炎相关的危险因素。再通过受试者操作特征(ROC)曲线中各个危险因素的曲线下面积,确定其诊断价值,记录各危险因素的截断值、灵敏度、特异度及诊断符合率。将危险因素组成一个logistic回归模型并进行预测,预测结果与实际诊断结果进行对比,以评估模型的准确度。结果病毒性肺炎组与细菌性肺炎组患者的年龄、白细胞计数、淋巴细胞计数、中性粒细胞计数、氧合指数、尿素氮、肌酐、乳酸脱氢酶、英国胸科协会肺炎(CURB-65)评分、肺炎严重指数(PSI)评分指标比较,差异有统计学意义(P<0.05)。经单因素logistic回归分析,共有6个定量指标具有统计学意义,分别为淋巴细胞计数、中性粒细胞计数、氧合指数、尿素氮、肌酐及乳酸脱氢酶,筛选出3个对提示甲型流行性感冒病毒性肺炎有统计学意义的危险因素,淋巴细胞计数、氧合指数及乳酸脱氢酶,对其绘制ROC曲线,ROC曲线下面积分别为0.879±0.037、0.705±0.058、0.756±0.055,截断值分别为0.95×10^(9)/L、285.50 mmHg(1 mmHg=0.133 kPa)、278.50 U/L,淋巴细胞计数与氧合指�
Objective To explore the early diagnostic indexes and related risk factors of influenza A virus pneumonia.Methods Retrospective analysis of 101 patients with adult community-acquired pneumonia admitted to the respiratory department of the northern theater general hospital from November 2016 to November 2018,including 63 males and 38 females,aged(66.38±18.13)years old,ranging from 18 to 80 years old.According to different types of pneumonia,the patients were divided into virus pneumonia group and bacterial pneumonia group.There were 41 patients in virus pneumonia group,27 males and 14 females;There were 60 patients in bacterial pneumonia group,36 males and 24 females.Collectted patients′leukocyte count,lymphocyte count,neutrophil count,platelet count,arterial oxygen partial pressure,arterial blood lactate,urea,creatinine,lactate dehydrogenase,creatine kinase,C-reactive protein and fibrinogen were measured.Screen out the statistically significant indicators,and then use single factor logistic regression to analyze and record the above indicators.The analysis shows that the statistically significant indicators are the risk factors related to influenza A viral pneumonia.Then,the diagnostic value was determined by the area under the curve of each risk factor in the receiver operating characteristic(ROC)curve,and the cut-off value,sensitivity,specificity and diagnostic coincidence rate of each risk factor were recorded.The risk factors were formed into a logistic regression model and predicted.The predicted results were compared with the actual diagnostic results to evaluate the accuracy of the model.Results There were significant differences in age,leukocyte count,lymphocyte count,neutrophil count,oxygenation index,urea nitrogen,creatinine,lactate dehydrogenase,consciousness,urea nitrogen,respiratory rate,blood pressure and age(CURB-65)score and pneumonia severity index(PSI)score between influenza A virus pneumonia group and bacterial pneumonia group(P<0.05).By univariate logistic regression analysis,there were 6 q
作者
商楠
赵梦姿
刘秋实
翟晶
黄乐为
Shang Nan;Zhao Mengzi;Liu Qiushi;Zhai Jing;Huang Lewei(Graduate training Base of General Hospital of Northern Theater Command of Jinzhou Medical University,Shenyang 110000,China;Department of Respiratory Medicine,General Hospital of Northern Theater Command,Shenyang 110000,China;Department of cardiovascular medicine,Zhongzhi Shengjing geriatric hospital(Shenyang 12th people′s Hospital),Shenyang 110000,China)
出处
《中国临床实用医学》
2021年第6期17-21,共5页
China Clinical Practical Medicine
