The Preliminary Exploration On Prediction Models Of Lung Function After VATS Lobectomy Based On QCT Analysis

Objective: This study is aimed to combine the related parameters of quantitative computerized tomography(QCT)analysis with the related factors affecting postoperative lung function to construct the prediction model of postoperative lung function in patients with lung cancer after lobectomy by video-assisted thoracoscopic surgery(VATS)in different periods.To explore an objective,effective and reliable method for the prediction of postoperative lung function in such patients.Methods: The general data of patients who underwent thoracoscopic lobectomy in the Thoracic Surgery Department of The Affiliated Hospital of Guizhou Medical University from October 2020 to December 2021 and perioperative data and postoperative pulmonary ventilation function were prospectively collected,and there were 36 modeling samples and 12 validation samples.The data collected for pulmonary ventilation function included pre-operation forced expiratory volume in the first second(FEV1)and pre-operation percentage of forced expiratory volume in the first second(FEV1%)(T0),1 month after operation(T1),3 months after operation(T2),and 6 months after operation(T3).Multiple linear regression analysis was used to construct relevant results based on QCT analysis,such as the low-attenuation areas less than-950 Hounsfield Units(LAA%-950),the 15 th percentile density(Perc 15),lung volume and other parameters were used to predict postoperative lung function in patients with lung cancer after lobectomy by VATS in different periods.Results:1.The general trend of lung function changes after lobectomy was significantly decreased in T1,significantly increased in T2,and gradually recovered in T3.2.Among the related results of QCT analysis,the resected lung volume was correlated with lung function indexes at T1-T3,and negatively correlated with FEV1(T1)、FEV1%(T1)、FEV1(T2)、FEV1%(T2)、FEV1(T3)and FEV1%(T3)(β =-0.645,P < 0.001;β =-0.112,P = 0.004;β =-0.537,P < 0.001;β =-0.194,P = 0.001;β =-0.651,P < 0.001;β =-0.491,P < 0.001);LAA%-950 of the resected lung was positively correlated with FEV1%(T1)、FEV1(T2)and FEV1(T3)(β = 0.018,P < 0.001;β = 0.049,P < 0.001;β = 0.029,P = 0.013).3.The prediction model of lung function indicators at each time point from T1-T3 was verified.Paired T-test results showed that there was no significant difference between the measured and predicted values of the verified samples,and P values were all greater than 0.05.Pearson correlation analysis showed that the correlation between the measured values and the predicted values was good,and all P values were less than 0.05.Conclusions: 1.The postoperative recovery of FEV1 and FEV1% was nonlinear in T1-T3,so traditional prediction methods could not be used to predict postoperative lung function.2.The factors affecting FEV1 and FEV1% recovery at different time points of T1-T3 were different.3.The prediction of FEV1(T1)= 3.491-0.645 × the resected lung volume(R2 = 0.892,P < 0.001);The prediction of FEV1%(T1)= 1.267-0.112 × the resected lung volume + 0.018 × LAA%-950 of the resected lung-0.017 × the days with thoracic drainage tube(R2 = 0.943,P < 0.001);The prediction of FEV1(T2)= 2.646-0.537 × the resected lung volume + 0.049 × LAA%-950 of the resected lung(R2 = 0.934,P < 0.001);The prediction of FEV1%(T2)= 1.176-0.194 × the resected lung volume(R2 = 0.859,P < 0.001);The prediction of FEV1(T3)= 2.757-0.651 × the resected lung volume + 0.029× LAA%-950 of the resected lung(R2 = 0.872,P < 0.001);The prediction of FEV1%(T3)= 3.512-0.491 × the resected lung volume-0.077 × BMI(R2 = 0.933,P < 0.001).However,all of the prediction models still need further research to confirm and verify them.