A Preliminary Study on the Relationship Between High-resolution Computed Tomography and Lung Function in People at Risk of Developing Chronic Obstructive Pulmonary Disease

Background: There is high morbidity and mortality for patients with chronic obstructive pulmonary disease (COPD) in China. The aim of our study was to explore the differences in high-resolution computed tomography (HRCT) emphysema parameters, air trapping parameters, and lung density parameters between patients at high risk and low risk of developing COPD and evaluate their correlation with lung function indicators. Methods: In this retrospective, single-center cohort study, we enrolled outpatients from the Physical Examination Center and Respiratory Medicine of The First Aliated Hospital of Wenzhou Medical University. The patients at risk of developing COPD were ≥ 40 years-old, had chronic cough or sputum production, and/or had exposure to risk factors for the disease and had not reached the diagnostic criteria. Patients were divided into a low-risk group and high-risk group according to FEV1/FVC ≥ 80% and 80%>FEV1/FVC ≥ 70%. The data on clinical characteristics, clinical symptom score, lung function, and HRCT were recorded. Results: 72 COPD high-risk patients and 86 COPD low-risk patients were enrolled in the study, and the air trapping index of the left, right, and bilateral lungs of the high-risk group were signicantly higher than those of the low-risk group . Additionally, the mean lung density during expiration was signicantly lower than that of the low-risk group. The emphysema index of left, right, and bilateral lungs was negatively correlated with FEV1/FVC (correlation coecients were -0.33-0.22-0.26). There was a negative correlation between the air trapping index of left and right lungs and bilateral lungs and FEV1/FVC(correlation coecients were -0.33-0.23-0.28, respectively), and the mean expiratory lung density of left and right lungs and bilateral lungs was positively correlated with FEV1/FVC (correlation coecients were 0.31, 0.25, 0.29, respectively). Conclusions: The air trapping index and the mean expiratory lung density obtained by HRCT combined with post-processing technology can be used as a basis for distinguishing between people at high risk


Background
Chronic obstructive pulmonary disease (COPD) is a common preventable and treatable disease that is characterized by persistent air ow limitation that is usually progressive and not completely reversible [1]. The late stage of the disease is often accompanied by systemic multi-system chronic diseases, including cardiovascular diseases, metabolic syndrome, osteoporosis, depression, anxiety, and lung cancer, which contribute to the overall severity in patients [2,3].
The clinical diagnosis of COPD is primarily based on pulmonary function tests (PFTs) [4], but some constraints remain in detecting early changes in lung structure or function. As a general function test, the accurate assessment of local pulmonary function damage is di cult. Also, it has been reported in the literature that clinical symptoms will be present or changes in pulmonary function tests (PFTs) will appear when more than 30% of the total lung parenchyma has been destroyed [5]. Although the forced expiratory volume in one second percentage (FEV1%) may be unchanged, there can be signi cant changes in imaging ndings [6]. The Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease 2020 report suggests that further in-depth research is needed to study those patients without evidence of air ow limitation but who have evidence of structural lung disease upon chest imaging, such as emphysema, that is consistent with what is found in patients with COPD [7].
Computed tomography (CT) is the modality of choice for the imaging characterization of COPD patients, and it can be used to describe the changes in lung parenchyma in patients with COPD [8]. In the past years, with the rapid development of high-resolution computed tomography (HRCT) technology and continuous development of image post-processing and reconstruction techniques, CT imaging has been recognized as an important method for the assessment of COPD [9]. The emphysema index (EI) [10,11], air trapping index (ATI) [12][13][14], and mean lung density (MLD) [15,16] are the most common functional imaging parameters. However, these have rarely been evaluated in a population of patients at risk for developing COPD.
The primary aim of our research was therefore to evaluate the differences in HRCT emphysema parameters, air trapping parameters, and lung density parameters between patients with high and low risk of developing COPD, and to investigate the relationships between CT metrics and lung function indicators.

Methods
This retrospective and single-center cohort study was carried out at The First A liated Hospital of Wenzhou Medical University in China. Ethical approval was obtained, and all participants consented to be included in the trial.

Recruitment and enrolment
There were 158 participants aged 40-79 years with no prior diagnosis of COPD who were recruited from the Physical Examination Center, Department and Respiratory Medicine and other departments between March 2018 and May 2019 ( Figure 1). They underwent respiratory biphasic HRCT, pulmonary function tests, and lled out three detailed questionnaires. The risk of developing COPD was determined by the forced expiratory volume in the rst second of expiration to forced vital capacity (FEV1/FVC): low risk (FEV1/FVC≥80%), or high risk (80%>FEV1/FVC≥70%).
Exclusion criteria were as follows: asthma, severe pneumonia, tuberculosis, pulmonary brosis, or other concomitant chronic diseases, including cardiovascular disease, skeletal muscle dysfunction, metabolic syndrome, osteoporosis, depression, anxiety, and lung cancer.
Medical information and patient characteristics including age, sex, height (m), weight (kg), smoking index, and BMI were obtained from screening questionnaires. All data were stored in an encrypted database.

Clinical Scores and Questionnaires
All subjects were asked to independently complete three questionnaires, the modi ed Medical Research Council (mMRC), the chronic obstructive pulmonary disease Assessment Test (CAT), and the clinical chronic obstructive pulmonary disease questionnaire (CCQ). After exhaustive examination, the researchers then recorded the data from the completed questionnaires.

Spirometry
Spirometry was performed using a Vostro15 portable spirometer according to the ATS/ERS standard. The percentage predicted values were calculated using the equations for Asian adults supplied in the user's manual. After a 10-min rest, the subject assumed a seated position with feet in full contact with the oor, breathed for three cycles, then forcefully, rapidly, and deeply inhaled through the interface device and suddenly, continuously, and steadily exhaled to reach a maximum amount of breathing. It was necessary for the exhalation process to not be spontaneously interrupted by unsolicited coughing. Measurements were performed in triplicate for each subject, and each curve was coincident, as far as possible, to qualify the PFT, which included FVC, FEV1, FEV1%prep, and FEV1/FVC.

Chest HRCT examination and image analysis
Prior to HRCT scanning, patients underwent breathing training to optimize the measurement of maximum inspiration. HRCT was performed at suspended full inspiration and expiration using a Gem Energy Spectroscopy CT (Discovery HD750, GE, USA). Scanning parameters were as follows: tube voltage 120 kV, automatic tube current (mA automatic modulation technique), pitch 0.984, slice thickness 5 mm with 1.25 mm reconstruction interval, detector coverage 40 mm, X-ray tube rotation speed 0.6 slice/rotation, DFOV 30 cm, SFOV 50 cm, and a 512 ×512 matrix.
Quantitative assessments of emphysema were performed using Volume Viewer 11.3 software. We established limits, and the computer program calculated the attenuation as the mean lung density (MLD) of the whole lung. Then, we calculated the total lung area, and the lung area occupied by attenuation values lower than previously xed thresholds (-950 HUs at inspiration and -856 HUs at expiration) by Lung VCAR.

Statistical analysis
Statistical analyses were performed using IBM SPSS (Statistical Package for the Social Sciences) Statistics Software (version 22.0, IBM). The measurement data were tested for normality by the Kolmogorov-Smirnov method. Continuous variables of normal distribution are presented as the mean± standard deviation and were compared by two-sample t-test. Continuous variables of skewed distribution are expressed as median values (interquartile ranges) and were compared by the Mann-Whitney U-test. Categorical variables are expressed as a number (%) and were compared by the χ² test or Fisher's exact test. The correlation between HRCT parameters and pulmonary function parameters was assessed by the Spearman rank correlation test. p<0.05 was considered statistically signi cant.

Baseline Characteristics
The mean age of 158 subjects was 52.18±8.56 years, ranging from 40 to 76 years (Table 1). There were 123 males and 35 females. Overall, 86 (54.4%) participants were under a low risk of developing COPD and 72 were (45.5%) in the high-risk group. There was no statistical difference in BMI, smoking index, or exposure to dust between the two groups. Compared with subjects in the high-risk group, those in the lowrisk group had similar scores for mMRC, CAT, and CCQ (p=0.53, p=0.47, and p=0.84, respectively). The FEV1/FVC and FEV1%prep in thte low-risk group were signi cantly lower than those in the high-risk group (0.82±0.07 vs. 0.79±0.08, 0.92±0.14 vs. 0.86±0.13, p<0.05), while no difference was found in FEV1 and FVC (p=0.47 and p=0.93, respectively).

Correlations Between Quantitative Measurements And Pulmonary Function Parameters
The correlations of various quantitative HRCT parameters at suspended full inspiration with the pulmonary function test variables are shown in Table 3  The LAA-856exp% showed a signi cantly strong correlation with FEV1, FEV1/FVC, and FEV1%prep (p<0.05) in Table 4. Among them, the most optimal correlation between LAA-856exp% was found with

Discussion
Our observational study on 158 patients who were at risk of developing COPD demonstrates that HRCT measurements of emphysema index and air trapping index were negatively correlated with FEV1/FVC from PFT, whereas the mean expiratory lung density showed a positive correlation. Patients in the high-risk group exhibited a signi cantly higher air trapping index but lower mean expiratory lung density than those in the low-risk group.
COPD is a chronic in ammatory airway disease that is characterized by air ow limitations that are not fully reversible [17]. At the early stage of the disease, patients are often asymptomatic or exhibit only mild chronic cough or dyspnea [18,19]. However, as the disease progresses, patients may experience chest pain, expectoration, fatigue, weight loss, and can also develop acute lower respiratory infections, cardiovascular disease, or lung cancer that continue to place an enormous burden on society [20,21]. According to the latest research, the prevalence of COPD in China is continually increasing, and the subgroup of patients over 40 years of age has nearly 99 million people, accounting for 13.7% [3].
Therefore, early detection and timely management are imperative.
There has been interest in the diagnostic value of HRCT for COPD, and its combined use with advanced postprocessing software will provide important clinical applications for COPD. Several studies [22][23][24] have established the relationship between lung function indicators such as FEV1/FVC, RV/TLC, and CT ndings. Some CT scans have been previously used to classify the severity in COPD patients. However, articles discussing CT ndings in patients at risk of developing COPD are scarce. The present study shows that the HRCT emphysema index, air trapping index, and the mean expiratory lung density each contain independent diagnostic information for patients at risk of developing COPD, and represent independent imaging biomarkers [25,26].
We nd from Table 1 that compared with the low-risk group, there was no statistically signi cant difference in the clinical features (age, sex, BMI categories, dust exposure, and smoking index) and the clinical symptom scores (mMRC, CAT, and CCQ) for the high-risk group. However, the differences in the forced expiratory volume in the rst second/forced vital capacity and forced expiratory volume in the rst second as a percentage of the expected value showed signi cant difference (P < 0.01) between the two groups. This suggests that based on behavioral observation and recordings, the severity of COPD in the different states was indistinguishable. These results were in agreement with the ndings of Sunmin Kim et al [27]. Their conclusion shows that the choice of symptom scale can alter the group assignment of COPD. The clinical features and the clinical symptom scores can be used to assist in gauging the condition, but there were limitations to the evaluation at the individual level. It is clear that additional indicators need to be comprehensively assessed in combination.
Comparing the CT parameters of the low-risk and high-risk COPD groups, the CT images of patients from the high-risk group exhibit certain clinical characteristics and typical imaging features. Higher lung volume and emphysema index, and lower mean lung density in the inspiratory phase and expiratory phase were present. This result demonstrates that in COPD high-risk patients, pulmonary hyperin ation and lung volume increased. Patients in the high-risk group have a signi cantly higher air trapping index but lower mean expiratory lung density as compared to the low-risk group. As for the emphysema index and the mean inspiratory lung density, this trend in the high and low-risk group was not statistically signi cant. Therefore, it is suggested that the air trapping index was more accurate at predicting the risk state of COPD and lung function decline than the emphysema index. In other words, using HRCT imaging for the study of lung pathologies in the early stage of the disease, the CT scans obtained at expiration were more informative than those at inspiration. In previous studies, the choice of an end inspiration or end expiration image remained controversial [28]. Some demonstrated the greater diagnostic sensitivity of inspiratory lung volumes for COPD [29]. CT measurements of airway dimensions and emphysema are useful and complementary in the evaluation of the lung with COPD.
Our study on the correlation between CT parameters and lung function shows that the left, right, and bilateral lung emphysema index and air trapping index were signi cantly inversely correlated with FEV1/FVC. This indicates that the emphysema index theoretically has the potential to use HRCT to assess evidence of emphysema and small airway disease. Our results indicate that the presence of increasing lung volume, persistent air ow limitation, and air trapping can also be present in those with undiagnosed COPD, which can in turn lead to lung function changes. In our study, we also found that the left, right, and bilateral lung mean expiratory lung density showed signi cantly positive correlation with FEV1/FVC. This is in accordance with the conclusion arrived at by previously published studies [30][31][32].
There was high predictive value of CT density parameters for detecting pulmonary ventilation.
Our study has several limitations. First, the study was conducted in a single center in China and was carried out at an outpatient clinic where patients had morbid conditions, and thus, inpatients were not included in the study. Second, the study population was relatively small and predominantly male. The results might be different in a larger population. Third, we did not enroll patients with con rmed COPD as controls, and therefore, the extrapolation of these ndings to COPD cohorts must be performed with care.
Last, no follow-up of the patients was performed.

Conclusions
In summary, our original study demonstrates that the combination of the air trapping index and the mean expiratory lung density obtained by HRCT can be used as an indicator for distinguishing between high and low risk of developing COPD. Emphysema index, air trapping index, and the mean expiratory lung density exhibited signi cant correlation with FEV1/FVC, which can be used to assess the pulmonary function status of people at risk of developing COPD, and provide a useful supplement for the early and comprehensive assessment of the disease.

Declarations
Ethics approval and consent to participate This retrospective and single-center cohort study was carried out at The First A liated Hospital of Wenzhou Medical University in China. Ethical approval was obtained, and all participants consented to be included in the trial.

Consent for publication
All list authors consent to the submission and all data are used with the consent of the person generating the data.

Competing interests
The authors declare that they have no competing interests.
Authors' contributions LY, XH and LW conceived and designed the study, responsible for the integrity and accuracy of the data, and had full access to the study. RL and HJ contributed to drafting and writing this paper. MX took responsibility for obtaining written consent from patients, obtaining ethical approval, collecting samples, and con rming data accuracy. PS, YW and DY made substantial contributions to data acquisition, analysis, and interpretation. All the authors had strictly revised the manuscript and agreed to be responsible for all aspects of the work, and nally approved the version to be published. Study ow diagram