| ObjectiveDual-energy CT is an examination equipment that can produce X-ray beams of two different energy levels(high energy and low energy),scanning the lesions with different attenuation of X-rays at two different energy levels can make the lesions develop more clearly.at the same time,it can provide quantitative parameters that reflect the blood perfusion of the lesions,it is possible to make a quantitative evaluation of the qualitative diagnosis of the lesions,therefore,it can be widely used to assist in the differential evaluation of benign and malignant pulmonary nodules.The purpose of this study was to analyze the independent risk factors for the differential diagnosis of benign and malignant pulmonary nodules by combining the relevant clinical features,imaging features and dual-energy CT quantitative parameters,and to establish a suitable clinical prediction model,and to further analyze the application value of quantitative parameters of iodine map in the differential diagnosis of benign and malignant pulmonary nodules.To evaluate the significance of tuberculosis infection T cell examination and tumor markers in the qualitative diagnosis of pulmonary nodules;multi-disciplinary team(MDT)plays an important role in the diagnosis and treatment of pulmonary nodules,we will analyze and evaluate the efficacy of MDT in the diagnosis and treatment of pulmonary nodules and compare it with other departments at the same time.MethodsFrom our hospital’s outpatient and inpatient electronic medical record system,scientific research big data platform and the lung cancer joint outpatient system,the data of pulmonary nodules examined by small nodule analysis(also known as dual-energy CT)were collected retrospectively from January 2015 to June 2021,and the pathological diagnosis of patients was clear,including clinical records,imaging data and pathological results.Then the patients were divided into benign group and malignant group according to the pathological results as the gold standard for diagnosis.The statistical data were analyzed by SPSS 23.0,and the samples were randomly divided into modeling group and verification group according to the proportion of 2:1,univariate analysis was used to evaluate the independent risk predictors for the differentiation of benign and malignant pulmonary nodules,t-test or nonparametric test was used to analyze the measurement data according to whether the normal distribution was satisfied,chi-square test was used for counting data.The independent risk factors in univariate analysis were substituted into multivariate Logistic regression for multivariate analysis and a clinical predictive model was constructed;the diagnostic efficacy of the model was evaluated by receiver operating characteristic curve(ROC);the stability of the model will be evaluated using data from the verification group.The correlation analysis was used to evaluate the relationship among the indicators.Results1.A total of 872 eligible pulmonary nodules were collected from 844 cases,aged from20 to 84 years(mean age 54.41±10.698),including 357 males(42.3%)and 487 females(57.7%);among the 872 pulmonary nodules,182(20.9%)were benign nodules and 690(79.1%)were malignant nodules.The main benign nodules were inflammations,accounted for 44.0%(80/182),followed by granulomas(17.0%,31/182),hamartomas(11.5%,21/182),tuberculosis(7.7%,14/182),fungi(3.9%,7/182)and others(15.9%,29/182);adenocarcinomas(95.2%,657/690)is the main pathological type of malignant nodules,followed by squamous cell carcinomas(2.9%,20/690)and metastatic carcinomas(1.2%,8/690),neuroendocrine carcinomas(0.4%,3/690)and other types of primary lung cancer(0.3%,2/690)are rare.2.According to the proportion of 2:1,872 pulmonary nodules were randomly divided into model group and verification group,there were 581 pulmonary nodules in the model group(111 benign and 470 malignant nodules,respectively),and 291 pulmonary nodules in the verification group(71 benign and 220 malignant,respectively).The results of univariate analysis in the model group showed that age,sex,number of nodules,nodule density,CT value,vacuole sign,air bronchus sign,iodine concentration and iodine ratio were independent risk factors for differentiating benign and malignant pulmonary nodules(p < 0.05);multivariate analysis showed that iodine concentration,nodule density,vacuole sign and air bronchus sign were independent predictors of lung cancer(p<0.05),based on this,a clinical predictive model was established and its ROC curve was drawn,the results showed that the area under the curve(AUC)of the model group was 0.790(cut-off value was 0.471,diagnostic sensitivity was 81.3%,specificity was 65.8%,and the positive predictive value(PPV)at threshold was 91.0%,the negative predictive value(NPV)was 45.3%,and the diagnostic accuracy was 78.3%(p= 0.000).The AUC=0.794 of the verification group(cut-off value is 0.478,sensitivity = 88.6%,specificity = 59.2%)(p= 0.000).3.The ROC curves of iodine concentration and iodine ratio of 872 pulmonary nodules showed that when iodine concentration ≥ 1.05mg/ml(AUC=0.632,sensitivity = 77.4%,specificity = 45.1%,the PPV at the threshold was 84.2%,the NPV was 34.5%,and the diagnostic accuracy was 70.6%),and(or)iodine ratio ≥ 13.9%(AUC=0.604,sensitivity =89.9%,specificity = 29.1%,the PPV at the threshold was 82.8%,the NPV was 43.1%,and the diagnostic accuracy is 77.2%),the pulmonary nodules were more likely to be malignant.Both iodine concentration and iodine ratio were valuable in the qualitative judgment of solid nodules(p<0.05),and iodine concentration could also be used to judge the malignant tendency of mixed ground glass nodules(m GGN)(p<0.05).4.A total of 272 tuberculosis infected T cells were tested(31.2%),with a positive rate of39.7%,chi-square test showed that tuberculosis infected T cells may be associated with lung cancer(p= 0.000).5.A total of 696 nodules(79.8%)completed the examination of tumor markers before the pathological samples were obtained,with an increase rate of 16.8%,of the 131 nodules with diameter ≤10mm,only 8 nodules(6.1%)had elevated tumor markers,while 109(19.3%)nodules with diameter of 10-30mm(n=565)were abnormally elevated,the rising rate of tumor markers in large nodules(10-30mm)was significantly higher than that in small nodules(≤10mm)(p<0.05);although tumor markers did not show statistical difference in the differential diagnosis of benign and malignant pulmonary nodules(p>0.05).However,subgroup analysis showed that carcinoembryonic antigen(CEA)had auxiliary diagnostic value in the qualitative diagnosis of lung adenocarcinoma with solid nodules(p= 0.000),especially in solid nodules with the longest diameter of 10-30 mm.6.A total of 136 patients participated in the MDT consultation,of which 92(67.6%)nodules were diagnosed as malignant,and the positive diagnostic coincidence rate was89.32%.ConclusionAge,sex,number of nodules,nodule density,CT value,vacuole sign,air bronchial sign,iodine concentration and iodine ratio are independent risk factors for differential diagnosis of benign and malignant pulmonary nodules,and when iodine concentration and iodine ratio are1.05mg/ml and 13.9%,respectively,pulmonary nodules are more likely to be malignant;the established clinical prediction model has good and stable diagnostic performance,although the contribution of iodine concentration to the whole model is relatively small.It is not recommended to detect tumor markers in pulmonary nodules whose diameter is less than10mm;CEA is of clinical value in the differential diagnosis of benign and malignant pulmonary nodules with solid density,especially for solid nodules with the longest diameter exceeding 10 mm.MDT is helpful for the early diagnosis of malignant nodules and is suitable for popularization and application. |
PDF Full Text Request