Quantitative computed tomography analysis of emphysema severeness and distribution in chronic obstructive lung disease and its correlation with clinical findings

Purpose: We aimed to evaluate correlation between different densitometric parameters and pulmonary function tests (PFT) in chest computed tomography (CT) of patients with emphysema predominant chronic obstructive disease (COPD). Material and Method: We retrospectively evaluated patients with COPD. Patients who have no more than 2 months interval between chest CT and pulmonary function were included. Densitometric values, anthropometric measurements, peripheral venous blood cell values, PFT values, smoking habits of all patients are recorded. Image data of the patients are obtained in DICOM format and transferred to the 3D Slicer program for quantitative measurements. Lungs are segmented into 9 segmenst as whole lung, right lung, left lung and upper, middle, lower zones of both lungs. Normal and emphysematous parenchyma in segmented zones are measured using low attenuation area percent (%LAA), percentile density (PD), mean lung density (MLD) parametres. Effects of emphysema distribution on PFT are evaluated using %LAA, PD, MLD, kurtosis, skewness methods. Statistical analyses were performed using SPSS 21.0 program. Pearson Test was used for correlation of parametric properties of the continuous variables, and Spearman Test was used for the correlation of those that do not have the parametric properties of the continuous variables. All values of p <0.005 were considered significant. Results: 56 patients are included in our study. All patients were diagnosed with COPD. %LAA - 910 method showed higher correlation with forced expiratory volume in one second (FEV1) compared to other densitometric parameters (p= <0,001, r= - 0,556). Other densitometric parametres had differences in sensitivity with (FEV1) between segments. We found involment of 82 right lung mid - zone segment affects FEV1 value more than other segments when using %LAA - 910 method (p= <0,001, r= - 0,569). Patients are divided into 2 groups as GOLD 1 - 2 and GOLD 3 - 4. GOLD 3 - 4 had higher emphysema values in all densitometric parametres compared to GOLD 1 - 2 (p= <0,001). No correlation was observed in any segment between skewness, kurtosis values and FEV1. Weight and body mass index variables showed strong inverse correlation with densitometric values (p= <0,001). Strong positive correlation was found with smoking (p= <0,001). There were no correlation between age and densitometric values. All densitometric values showed strong correlation with FEV1 / FVC and mid - expiratory flow (MEF 25 - 75) values (p= <0,001). No correlation was observed between peak expiratory flow (PEF) and densitometric measurements. When correlation between peripheral venous blood cell values and densitometry evaluated, lymphocyte percentage showed poor correlation with %LAA - 950 and %LAA - 925 methods (p= 0,044, r= - 0,270). Conclusion: Quantitative analysis of chest CT is an objective method for determining distribution and severity of emphysema. There was a significant correlation between densitometry and PFT values. We think that quantitative analysis can be used to evaluate the severity of emphysema and to predict the clinical findings of the patients. Keywords: Emphysema, densitometry, pulmonary function test