The Analysis Of Independent Risk Factors For Isolated Pulmonary Nodules And Establishment Of A Prediction Model Of Benign And Malignant

Objective:To analyze the general clinical data,laboratory examination data and imaging characteristics of patients with solitary pulmonary nodules,explore the risk factors and independent risk factors affecting their nature,establish a prediction model of malignant pulmonary nodules,and calculate the probability of malignant pulmonary nodules,So as to guide clinicians and related workers to deal with the disease scientifically and effectively.Methods:The general clinical data,laboratory examination data and imaging characteristics of 821 patients diagnosed by percutaneous lung biopsy or surgical pathology in jiangsu subei people’s hospital from 2015 to 2018 were sorted out.SPSS 17.0 software is used to make statistics and Analysis on all the collected data.The relevant data conforming to the normal distribution shall be represented by the mean standard deviation and t for comparison test,while the relevant data of non normal distribution shall be represented by Z(P25,P75)and nonparametric comparison test.X2 test was used to analyze the count data,logistic regression was used to analyze the relevant independent risk factors,and a regression equation model was constructed to predict the possibility of malignant pulmonary nodules.At the same time,the ROC curve of the probability of malignant pulmonary nodules was drawn,the accurate critical value of the probability of benign and malignant pulmonary nodules was found,and the sensitivity,specificity,positive and negative predictive values of patients were calculated correctly,P<0.05.Results:There were significant differences in smoking index,age,spiculati on,nodule diameter,P-LCR,PDW,carcinoembryonic antigen and neuron specif ic enolase between benign and malignant pulmonary nodules(P<0.05).At the same time,there were also significant differences between benign and malignan t pulmonary nodules in patients’ age,nodule diameter,spiculation,NSE,P-LCR and PDW(P<0.05).The predictive regression equation of SPN is:P=ex/(1+ex),e is the natural logarithm,X=-2.85+(0.03×Age)+(0.07×Nodule diameter)+(0.30×spiculation)+(0.12×NSE)+(0.15×P-LCR)+(0.28×PDW).Calculate the malignant prediction rate of each patient,and draw the ROC curve with full reference to the pathological results.Calculate the area under the ROC curve as 0.72(0.66-0.78),and select the appropriate critical value of 0.80 as the standard value fo r judging the benign and malignant of pulmonary nodules.The specificity,sens itivity,positive and negative predictive values of the model are 64%,73%,70%and 68%respectively.Conclusion:Patient age,nodule diameter,spiculation,NSE,P-LCR,PDW are important basis for the determination of malignant SPN,and the establishment of prediction model is helpful to predict the benign and malignant SPN,so as to greatly reduce the error of human factors,and finally continuously improve the clinical diagnosis efficiency of SPN.