Chronic Obstructive Pulmonary Disease (COPD) represents a leading cause of chronic morbidity and mortality worldwide, accounting for over 3 million deaths annually [1]. Characterized by persistent, often progressive airflow limitation, the clinical course of COPD is driven by an abnormal inflammatory response of the lungs to noxious particles or gases, most commonly tobacco smoke [2]. Traditionally, the diagnosis and severity staging of COPD have relied on Spirometry, specifically the post-bronchodilator Forced Expiratory Volume in 1 second (FEV_1) and its ratio to Forced Vital Capacity (FEV_1/FVC) [3].

However, standard pulmonary function tests (PFTs) provide an aggregate assessment of total lung function. They are inherently insensitive to early structural alterations in the pulmonary parenchyma and peripheral airways [4]. Two patients presenting with an identical FEV_1 of 50% predicted may exhibit vastly different underlying pathological profiles: one dominated by panlobular or centrilobular alveolar destruction (emphysema-predominant phenotype) and the other by chronic bronchitis, mucus hypersecretion, and luminal narrowing of small airways (airway-predominant phenotype) [5]. Because therapeutic efficacy varies significantly between these physiological phenotypes, accurate phenotypic characterization is vital for modern precision medicine [6].

High-Resolution Computed Tomography (HRCT) of the chest has revolutionized the objective quantification of structural lung disease in COPD [7]. Densitometric HRCT analysis allows for automated calculation of low-attenuation areas (LAA%), typically defined as voxels with attenuation values lower than -950 Hounsfield Units (HU) on full-inspiration scans (EI_{-950}), which corresponds directly to microscopic emphysematous tissue destruction [8]. Simultaneously, advanced airway analysis software enables high-precision measurement of bronchial luminal area, wall thickness (WT), and wall area percentage (WA%) across proximal and intermediate subsegmental bronchi [9].

Despite technological advancements in quantitative CT, the degree of correlation between specific densitometric parameters, bronchial morphometry, and comprehensive pulmonary physiological metrics (FEV_1, lung volumes, and diffusing capacity) across diverse severity stages remains an active area of investigation [10]. This study was undertaken to examine the quantitative relationship between HRCT-derived emphysema/airway metrics and PFT parameters in stable COPD patients, and to evaluate the diagnostic utility of HRCT in phenotypic categorization.

Study Design and Patient Selection This prospective, cross-sectional observational study was conducted at KBN Teaching and General Hospital attached to KBNU-Faculty of Medical Sciences - a tertiary care university teaching hospital over a 12-month period (April 2025 to March 2026). The study protocol was approved by the Institutional Ethics Committee, and written informed consent was obtained from all participants prior to enrollment in accordance with the Declaration of Helsinki. Inclusion Criteria: Adults aged ≥ 40 years with a established diagnosis of stable COPD according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria [3]. Post-bronchodilator FEV_1/FVC < 0.70. Absence of acute exacerbation of COPD for at least 8 weeks prior to study entry. Ability to perform reliable, reproducible spirometry, body plethysmography, and HRCT breath-holding maneuvers. Exclusion Criteria: Primary diagnosis of bronchial asthma, bronchiectasis, active pulmonary tuberculosis, interstitial lung disease, or lung malignancy. History of lung volume reduction surgery, lobectomy, or metallic chest implants. Hemodynamic instability or severe comorbid conditions (e.g., congestive heart failure, end-stage renal disease). Pregnancy or lactation. Pulmonary Function Testing Pulmonary function testing was performed using a calibrated master-screen PFT system (Body Plethysmograph; Jaeger, CareFusion, Germany) in accordance with the American Thoracic Society/European Respiratory Society (ATS/ERS) task force guidelines [11]. Spirometry: Baseline and post-bronchodilator (400µg inhaled Salbutamol via spacer) forced expiratory maneuvers were performed. Parameters recorded included FVC, FEV_1, FEV_1/FVC, and forced expiratory flow between 25% and 75% of vital capacity (FEF_25-75%). Plethysmography: Functional Residual Capacity (FRC), Total Lung Capacity (TLC), and Residual Volume (RV) were measured using constant-volume body plethysmography. Airway resistance (Raw) and RV/TLC% ratio were determined. Diffusing Capacity: Diffusing capacity of the lung for carbon monoxide (DLCO) was measured using the single-breath hold technique (10 seconds holding time) and adjusted for hemoglobin concentration. Values were expressed as percentage of predicted values (DL_CO% predicted). High-Resolution Computed Tomography Acquisition All CT examinations were performed on a 64-slice multidetector CT scanner (Somatom Sensation, Siemens Healthineers, Erlangen, Germany) without intravenous contrast administration. Scans were acquired from the lung apices to the diaphragm during a single deep-inspiration breath-hold at total lung capacity (TLC). Patient coaching was conducted prior to scanning to ensure maximal inflation. Scan Parameters: Tube voltage: 120kVp Tube current: 100mAs (using automatic exposure control, CareDose4D) Collimation: 64 X 0.6 mm Pitch: 1.1 Rotation time: 0.5 seconds Reconstruction slice thickness: 1.0mm with 0.7mm overlap Reconstruction kernel: High-spatial frequency sharp kernel (B70f) for airway detail and smooth kernel (B30f) for densitometric assessment. Quantitative HRCT Image Analysis Image data were processed using dedicated automated thoracic quantitative analysis software (Pulmo-CMS / Syngo.via, Siemens Healthineers). Emphysema Quantification: Automated lung segmentation was performed to isolate the left and right lung parenchyma from the chest wall, mediastinum, and large central airways. Total lung volume on CT (V_CT) was calculated. Low-Attenuation Areas (LAA) were identified using a fixed densitometric threshold of -950 Hounsfield Units (HU) [8]. The Emphysema Index (EI_{-950}) was calculated as the volume of lung parenchyma attenuation below -950HU divided by total lung volume, expressed as a percentage: 〖EI〗_{-950} =(.⁡〖(Volume o Lung<-950 HU)/(Total Lung Volume)〗 )X 100 Mean Lung Density (MLD) was recorded in HU. Airway Morphometry: Airway analysis was conducted on segmental and subsegmental bronchi (3rd, 4th, and 5th generation branches of the apical segment of the right upper lobe and lateral segment of the right lower lobe). Cross-sectional images orthogonal to the bronchial long axis were constructed. Parameters measured included Outer Lumen Diameter (Do), Inner Lumen Diameter (Di), Lumen Area (LA), and Total Lumen Area (WA + LA). Bronchial Wall Area Percentage (WA%) was calculated as: WA%=(.⁡〖(Wall area)/(Total Cross sectional area)〗 )X 100 Mean Wall Thickness (WT) was recorded in millimeters. Statistical Analysis Statistical evaluation was executed using SPSS version 28.0 (IBM Corp., Armonk, NY). Sample size calculation indicated that 210 patients were required to achieve 90% power (β = 0.10) at a significance level of α = 0.05 to detect a correlation coefficient of r ≥ 0.25 between EI_{-950} and FEV_1. To account for potential dropouts or technically inadequate CT/PFT data, 240 participants were recruited. Data normality was tested using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Continuous variables were expressed as mean ± standard deviation (SD) for normally distributed data, or median (interquartile range, IQR) for skewed data. Categorical variables were expressed as frequencies and percentages. Comparisons across GOLD severity stages (I–IV) were performed using One-Way Analysis of Variance (ANOVA) with Tukey’s post-hoc test or the Kruskal-Wallis test where appropriate. Correlations between HRCT quantitative metrics (EI_{-950}, WA%, WT) and PFT parameters (FEV_1, FVC, FEV_1/FVC, RV/TLC, DLCO) were assessed using Pearson’s correlation coefficient (r) or Spearman’s rank correlation coefficient (rho). Multivariate linear regression models were constructed using stepwise variable selection to identify independent CT predictors of post-bronchodilator FEV_1% predicted and DL_CO% predicted. Two-tailed p-values < 0.05 were considered statistically significant. Study Design and Patient Selection This prospective, cross-sectional observational study was conducted at KBN Teaching and General Hospital attached to KBNU-Faculty of Medical Sciences - a tertiary care university teaching hospital over a 12-month period (April 2025 to March 2026). The study protocol was approved by the Institutional Ethics Committee, and written informed consent was obtained from all participants prior to enrollment in accordance with the Declaration of Helsinki. Inclusion Criteria: Adults aged ≥ 40 years with a established diagnosis of stable COPD according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria [3]. Post-bronchodilator FEV_1/FVC < 0.70. Absence of acute exacerbation of COPD for at least 8 weeks prior to study entry. Ability to perform reliable, reproducible spirometry, body plethysmography, and HRCT breath-holding maneuvers. Exclusion Criteria: Primary diagnosis of bronchial asthma, bronchiectasis, active pulmonary tuberculosis, interstitial lung disease, or lung malignancy. History of lung volume reduction surgery, lobectomy, or metallic chest implants. Hemodynamic instability or severe comorbid conditions (e.g., congestive heart failure, end-stage renal disease). Pregnancy or lactation. Pulmonary Function Testing Pulmonary function testing was performed using a calibrated master-screen PFT system (Body Plethysmograph; Jaeger, CareFusion, Germany) in accordance with the American Thoracic Society/European Respiratory Society (ATS/ERS) task force guidelines [11]. Spirometry: Baseline and post-bronchodilator (400µg inhaled Salbutamol via spacer) forced expiratory maneuvers were performed. Parameters recorded included FVC, FEV_1, FEV_1/FVC, and forced expiratory flow between 25% and 75% of vital capacity (FEF_25-75%). Plethysmography: Functional Residual Capacity (FRC), Total Lung Capacity (TLC), and Residual Volume (RV) were measured using constant-volume body plethysmography. Airway resistance (Raw) and RV/TLC% ratio were determined. Diffusing Capacity: Diffusing capacity of the lung for carbon monoxide (DLCO) was measured using the single-breath hold technique (10 seconds holding time) and adjusted for hemoglobin concentration. Values were expressed as percentage of predicted values (DL_CO% predicted). High-Resolution Computed Tomography Acquisition All CT examinations were performed on a 64-slice multidetector CT scanner (Somatom Sensation, Siemens Healthineers, Erlangen, Germany) without intravenous contrast administration. Scans were acquired from the lung apices to the diaphragm during a single deep-inspiration breath-hold at total lung capacity (TLC). Patient coaching was conducted prior to scanning to ensure maximal inflation. Scan Parameters: Tube voltage: 120kVp Tube current: 100mAs (using automatic exposure control, CareDose4D) Collimation: 64 X 0.6 mm Pitch: 1.1 Rotation time: 0.5 seconds Reconstruction slice thickness: 1.0mm with 0.7mm overlap Reconstruction kernel: High-spatial frequency sharp kernel (B70f) for airway detail and smooth kernel (B30f) for densitometric assessment. Quantitative HRCT Image Analysis Image data were processed using dedicated automated thoracic quantitative analysis software (Pulmo-CMS / Syngo.via, Siemens Healthineers). Emphysema Quantification: Automated lung segmentation was performed to isolate the left and right lung parenchyma from the chest wall, mediastinum, and large central airways. Total lung volume on CT (V_CT) was calculated. Low-Attenuation Areas (LAA) were identified using a fixed densitometric threshold of -950 Hounsfield Units (HU) [8]. The Emphysema Index (EI_{-950}) was calculated as the volume of lung parenchyma attenuation below -950HU divided by total lung volume, expressed as a percentage: 〖EI〗_{-950} =(.⁡〖(Volume o Lung<-950 HU)/(Total Lung Volume)〗 )X 100 Mean Lung Density (MLD) was recorded in HU. Airway Morphometry: Airway analysis was conducted on segmental and subsegmental bronchi (3rd, 4th, and 5th generation branches of the apical segment of the right upper lobe and lateral segment of the right lower lobe). Cross-sectional images orthogonal to the bronchial long axis were constructed. Parameters measured included Outer Lumen Diameter (Do), Inner Lumen Diameter (Di), Lumen Area (LA), and Total Lumen Area (WA + LA). Bronchial Wall Area Percentage (WA%) was calculated as: WA%=(.⁡〖(Wall area)/(Total Cross sectional area)〗 )X 100 Mean Wall Thickness (WT) was recorded in millimeters. Statistical Analysis Statistical evaluation was executed using SPSS version 28.0 (IBM Corp., Armonk, NY). Sample size calculation indicated that 210 patients were required to achieve 90% power (β = 0.10) at a significance level of α = 0.05 to detect a correlation coefficient of r ≥ 0.25 between EI_{-950} and FEV_1. To account for potential dropouts or technically inadequate CT/PFT data, 240 participants were recruited. Data normality was tested using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Continuous variables were expressed as mean ± standard deviation (SD) for normally distributed data, or median (interquartile range, IQR) for skewed data. Categorical variables were expressed as frequencies and percentages. Comparisons across GOLD severity stages (I–IV) were performed using One-Way Analysis of Variance (ANOVA) with Tukey’s post-hoc test or the Kruskal-Wallis test where appropriate. Correlations between HRCT quantitative metrics (EI_{-950}, WA%, WT) and PFT parameters (FEV_1, FVC, FEV_1/FVC, RV/TLC, DLCO) were assessed using Pearson’s correlation coefficient (r) or Spearman’s rank correlation coefficient (rho). Multivariate linear regression models were constructed using stepwise variable selection to identify independent CT predictors of post-bronchodilator FEV_1% predicted and DL_CO% predicted. Two-tailed p-values < 0.05 were considered statistically significant.

Patient Baseline Characteristics

A total of 240 stable COPD patients completed all PFT maneuvers and HRCT imaging protocols without protocol violations. Table 1 summarizes the baseline clinical, functional, and radiological characteristics of the study cohort stratified by GOLD airflow limitation stages.

The study population comprised 174 males (72.5%) and 66 females (27.5%), with a mean age of 62.4 ± 8.2 years and a mean smoking history of 38.4 ± 14.2 pack-years. According to GOLD spirometric staging, 36 patients (15.0%) were classified as GOLD Stage I (Mild), 84 (35.0%) as GOLD Stage II (Moderate), 72 (30.0%) as GOLD Stage III (Severe), and 48 (20.0%) as GOLD Stage IV (Very Severe).

Table 1. Baseline Clinical, PFT, and HRCT Characteristics Stratified by GOLD Severity Stages (N = 240)

Correlation Between Emphysema Index (EI_{-950}) and Pulmonary Function Tests

A strong, statistically significant inverse correlation was demonstrated between the Emphysema Index (EI_{-950}) and spirometric markers of airflow limitation (Figure 1). EI_{-950} correlated negatively with post-bronchodilator FEV_1% predicted (r = -0.682, p < 0.001) and FEV_1/FVC ratio (r = -0.614, p < 0.001).

The most robust linear correlation was observed between EI_{-950} and DL_CO% predicted (r = -0.746, p < 0.001), reflecting progressive alveolar capillary membrane destruction as low-attenuation emphysematous parenchymal destruction increased.

Regarding plethysmographic static lung volumes, EI_{-950} exhibited a strong positive correlation with lung hyperinflation metrics, specifically RV/TLC% (r = 0.674, p < 0.001) and total lung capacity (TLC predicted) (r = 0.512, p < 0.001).

[ Bivariate Scatter Plots - Core Correlations ]

Correlation Between Airway Morphometry (WA%, WT) and Pulmonary Function Tests

Bronchial wall measurements demonstrated distinct physiological relationships compared to parenchymal densitometry metrics. As shown in Table 2, Wall Area Percentage (WA%) of subsegmental bronchi correlated moderately and inversely with post-bronchodilator FEV_1% predicted (r = -0.528, p<0.001) and mid-expiratory flow FEF_25-7% predicted (r = -0.586, p < 0.001).

Unlike EI_{-950}, WA% showed only a weak-to-moderate inverse correlation with DL_CO% predicted (r = -0.312, p < 0.001). However, WA% correlated significantly with airway resistance (Raw) (r = 0.598, p < 0.001) and air trapping as reflected by RV/TLC% (r=0.631, p < 0.001). Absolute Wall Thickness (WT) exhibited similar, albeit slightly weaker, correlation coefficients with functional metrics compared to the normalized WA% parameter.

Table 2. Bivariate Correlation Matrix Between HRCT Parameters & Physiological Functional Indices (N=240)

*Correlation is significant at the 0.05 level (2-tailed)  **Correlation is significant at the 0.01 level (2-tailed).

Multivariate Regression Analysis

To establish independent predictors of spirometric airflow obstruction and gas exchange capacity, stepwise multiple linear regression models were constructed (Table 3).

In Model 1, with post-bronchodilator FEV_1% predicted as the dependent continuous variable, both EI_{-950} (β = -0.482, p < 0.001) and WA% (β = -0.314, p < 0.001) emerged as significant independent negative predictors, accounting for 58.4% of the total variance (R² = 0.584, Adjusted R² = 0.580, p < 0.001).

In Model 2, with DL_CO% predicted as the dependent variable, EI_{-950} was the predominant independent predictor (β = -0.668, p < 0.001), while WA% exerted a minor contribution (β = -0.124, p = 0.018), yielding an overall model R² of 0.578 (p < 0.001).

Table 3. Stepwise Multiple Linear Regression Models for Predicting FEV_1% predicted and DL_CO% predicted

Phenotypic Distribution Based on HRCT

Utilizing established threshold cutoffs for EI_{-950} (≥ 10%) and WA% (≥ 65%), the cohort of 240 COPD patients was successfully categorized into three distinct imaging phenotypes:

1. Emphysema-Predominant Phenotype (EI_{-950} ≥ 10%, WA% < 65%): 68 patients (3%). Characterized by severe reduction in DL_CO% predicted (44.2 ± 12.8%), low BMI (21.2 ± 3.1kg/m²), and profound hyperinflation (RV/TLC = 54.8 ± 8.2%).
2. Airway-Predominant Phenotype (EI_{-950} < 10%, WA% ≥ 65%): 52 patients (7%). Characterized by preserved DL_CO% predicted (74.8 ± 11.4%), higher BMI (26.8±3.6kg/m²), elevated Raw (0.48±0.12 kPa s/L), and marked small airway flow impairment (FEF_{25-75%} = 26.4 ± 8.2%).
3. Mixed Phenotype (EI_{-950} ≥ 10%, WA% ≥ 65%): 88 patients (7%). Displayed the most severe physiological impairment across all PFT metrics.
4. Mild / Undifferentiated Phenotype (EI_{-950} < 10%, WA% < 65%): 32 patients (3%). Predominantly composed of GOLD Stage I patients.

The primary objective of this prospective study was to investigate the quantitative relationships between objective HRCT parameters reflecting parenchymal destruction and airway remodeling and functional physiological metrics measured by comprehensive PFTs in patients with COPD. Our findings confirm that densitometric emphysema metrics (EI_{-950}) and bronchial wall morphometry (WA%) correlate strongly with airflow obstruction, air trapping, and impaired gas diffusion, providing essential structural insights that spirometry alone cannot capture [12]. Parenchymal Emphysema vs. Physiological Impairment We observed a strong inverse correlation between EI_{-950} and FEV_1% predicted (r = -0.682), which aligns with large-scale cohort observations from the COPDGene and ECLIPSE studies [13, 14]. Structural emphysema degrades radial traction on peripheral airways, causing premature dynamic airway collapse during forced expiration [15]. This mechanical degradation directly links alveolar loss to physiological airflow limitation. Notably, the strongest correlation in our study occurred between EI_{-950} and DL_CO% predicted (r = -0.746). Carbon monoxide diffusing capacity serves as a direct functional surrogate for the total surface area of the capillary bed. As low-attenuation areas expand, loss of interalveolar septa depletes the pulmonary capillary network, reducing DLCO [16]. Thus, a disproportionate reduction in DLCO relative to FEV_1 should raise immediate suspicion of an emphysema-predominant phenotype, warranting formal HRCT quantification [17]. Small Airway Disease and Bronchial Wall Remodeling Although emphysema has historically received primary focus in CT quantification, peripheral small airway remodeling (bronchi < 2mm internal diameter) represents the principal site of increased airway resistance in COPD [18]. Because modern CT scanners cannot directly visualize bronchi smaller than 2mm, intermediate subsegmental bronchial morphometry (WA% and WT at 3rd–5th generations) serves as a reliable structural surrogate [19]. In our cohort, subsegmental WA% correlated significantly with mid-expiratory flow rate FEF_25-75% predicted (r = -0.586) and plethysmographic airway resistance R_{aw} (r = 0.598). Bronchial wall thickening, secondary to chronic mucosal inflammation, smooth muscle hypertrophy, and peribronchial fibrosis narrows the central lumen, creating fixed anatomical resistance [20]. Furthermore, multivariate linear regression demonstrated that WA% independently predicts FEV_1% predicted alongside EI_{-950}, proving that parenchymal destruction and airway wall thickening exert additive, complementary effects on global airflow obstruction [21]. Clinical and Therapeutic Implications of HRCT Phenotyping Spirometric evaluation alone fails to distinguish between the pathophysiological drivers of airflow limitation [22]. By integrating EI_{-950} and WA%, we successfully classified over 86% of symptomatic COPD patients into discrete structural phenotypes (emphysema-predominant, airway-predominant, or mixed). Phenotypic stratification carries profound therapeutic implications: • Emphysema-Predominant Phenotype: Patients with high EI_{-950} and severe static hyperinflation are prime candidates for lung volume reduction therapies, including endobronchial valves (EBV), lung volume reduction surgery (LVRS), or targeted alpha-1 antitrypsin augmentation where indicated [23]. • Airway-Predominant Phenotype: Patients with elevated WA% and preserved DLCO are more likely to benefit from maximal dual long-acting bronchodilator therapy (LAMA/LABA), targeted anti-inflammatory agents (e.g., inhaled corticosteroids or PDE4 inhibitors), and aggressive bronchial hygiene [24]. • Mixed Phenotype: Represents an advanced disease state requiring comprehensive multi-modal pharmacotherapy and pulmonary rehabilitation [25]. LIMITATIONS Several limitations of this study warrant consideration. First, HRCT scans were acquired exclusively at full inspiration (TLC). Additional expiratory CT scanning was not routinely performed to minimize patient radiation exposure. Expiratory densitometry (EI_{-856} or E/Itext{ ratio}) allows direct quantification of dynamic air trapping, which may further improve correlations with RV/TLC% [26]. Second, manual and semi-automated segmentation of 5th-generation and smaller subsegmental airways remains technically challenging due to motion and partial volume artifacts. Third, our study design was cross-sectional; longitudinal studies tracking changes in EI_{-950} and WA% over time are needed to establish the predictive value of HRCT metrics for disease progression and mortality.

Densitometric and morphometric HRCT parameters (EI_{-950}, WA%, and WT) correlate significantly with pulmonary function test metrics (FEV_1, DLCO, RV/TLC, and R_{aw}) in patients with stable COPD across all GOLD severity stages. Emphysema index drives reductions in diffusing capacity and lung hyperinflation, whereas bronchial wall area percentage primarily correlates with increased airway resistance and small airway flow impairment.

Quantitative HRCT provides clinical value beyond standard spirometry by non-invasively dissecting COPD into distinct structural phenotypes. Integrating automated quantitative HRCT analysis into routine clinical evaluation enables individualized phenotyping, supporting targeted pharmacological and interventional management for COPD patients.

Patient Baseline Characteristics

A total of 240 stable COPD patients completed all PFT maneuvers and HRCT imaging protocols without protocol violations. Table 1 summarizes the baseline clinical, functional, and radiological characteristics of the study cohort stratified by GOLD airflow limitation stages.

The study population comprised 174 males (72.5%) and 66 females (27.5%), with a mean age of 62.4 ± 8.2 years and a mean smoking history of 38.4 ± 14.2 pack-years. According to GOLD spirometric staging, 36 patients (15.0%) were classified as GOLD Stage I (Mild), 84 (35.0%) as GOLD Stage II (Moderate), 72 (30.0%) as GOLD Stage III (Severe), and 48 (20.0%) as GOLD Stage IV (Very Severe).

Table 1. Baseline Clinical, PFT, and HRCT Characteristics Stratified by GOLD Severity Stages (N = 240)

Correlation Between Emphysema Index (EI_{-950}) and Pulmonary Function Tests

A strong, statistically significant inverse correlation was demonstrated between the Emphysema Index (EI_{-950}) and spirometric markers of airflow limitation (Figure 1). EI_{-950} correlated negatively with post-bronchodilator FEV_1% predicted (r = -0.682, p < 0.001) and FEV_1/FVC ratio (r = -0.614, p < 0.001).

The most robust linear correlation was observed between EI_{-950} and DL_CO% predicted (r = -0.746, p < 0.001), reflecting progressive alveolar capillary membrane destruction as low-attenuation emphysematous parenchymal destruction increased.

Regarding plethysmographic static lung volumes, EI_{-950} exhibited a strong positive correlation with lung hyperinflation metrics, specifically RV/TLC% (r = 0.674, p < 0.001) and total lung capacity (TLC predicted) (r = 0.512, p < 0.001).

[ Bivariate Scatter Plots - Core Correlations ]

Correlation Between Airway Morphometry (WA%, WT) and Pulmonary Function Tests

Bronchial wall measurements demonstrated distinct physiological relationships compared to parenchymal densitometry metrics. As shown in Table 2, Wall Area Percentage (WA%) of subsegmental bronchi correlated moderately and inversely with post-bronchodilator FEV_1% predicted (r = -0.528, p<0.001) and mid-expiratory flow FEF_25-7% predicted (r = -0.586, p < 0.001).

Unlike EI_{-950}, WA% showed only a weak-to-moderate inverse correlation with DL_CO% predicted (r = -0.312, p < 0.001). However, WA% correlated significantly with airway resistance (Raw) (r = 0.598, p < 0.001) and air trapping as reflected by RV/TLC% (r=0.631, p < 0.001). Absolute Wall Thickness (WT) exhibited similar, albeit slightly weaker, correlation coefficients with functional metrics compared to the normalized WA% parameter.

Table 2. Bivariate Correlation Matrix Between HRCT Parameters & Physiological Functional Indices (N=240)

*Correlation is significant at the 0.05 level (2-tailed)  **Correlation is significant at the 0.01 level (2-tailed).

Multivariate Regression Analysis

To establish independent predictors of spirometric airflow obstruction and gas exchange capacity, stepwise multiple linear regression models were constructed (Table 3).

In Model 1, with post-bronchodilator FEV_1% predicted as the dependent continuous variable, both EI_{-950} (β = -0.482, p < 0.001) and WA% (β = -0.314, p < 0.001) emerged as significant independent negative predictors, accounting for 58.4% of the total variance (R² = 0.584, Adjusted R² = 0.580, p < 0.001).

In Model 2, with DL_CO% predicted as the dependent variable, EI_{-950} was the predominant independent predictor (β = -0.668, p < 0.001), while WA% exerted a minor contribution (β = -0.124, p = 0.018), yielding an overall model R² of 0.578 (p < 0.001).

Table 3. Stepwise Multiple Linear Regression Models for Predicting FEV_1% predicted and DL_CO% predicted

Phenotypic Distribution Based on HRCT

Utilizing established threshold cutoffs for EI_{-950} (≥ 10%) and WA% (≥ 65%), the cohort of 240 COPD patients was successfully categorized into three distinct imaging phenotypes:

1. Emphysema-Predominant Phenotype (EI_{-950} ≥ 10%, WA% < 65%): 68 patients (3%). Characterized by severe reduction in DL_CO% predicted (44.2 ± 12.8%), low BMI (21.2 ± 3.1kg/m²), and profound hyperinflation (RV/TLC = 54.8 ± 8.2%).
2. Airway-Predominant Phenotype (EI_{-950} < 10%, WA% ≥ 65%): 52 patients (7%). Characterized by preserved DL_CO% predicted (74.8 ± 11.4%), higher BMI (26.8±3.6kg/m²), elevated Raw (0.48±0.12 kPa s/L), and marked small airway flow impairment (FEF_{25-75%} = 26.4 ± 8.2%).
3. Mixed Phenotype (EI_{-950} ≥ 10%, WA% ≥ 65%): 88 patients (7%). Displayed the most severe physiological impairment across all PFT metrics.
4. Mild / Undifferentiated Phenotype (EI_{-950} < 10%, WA% < 65%): 32 patients (3%). Predominantly composed of GOLD Stage I patients.

The primary objective of this prospective study was to investigate the quantitative relationships between objective HRCT parameters reflecting parenchymal destruction and airway remodeling and functional physiological metrics measured by comprehensive PFTs in patients with COPD. Our findings confirm that densitometric emphysema metrics (EI_{-950}) and bronchial wall morphometry (WA%) correlate strongly with airflow obstruction, air trapping, and impaired gas diffusion, providing essential structural insights that spirometry alone cannot capture [12]. Parenchymal Emphysema vs. Physiological Impairment We observed a strong inverse correlation between EI_{-950} and FEV_1% predicted (r = -0.682), which aligns with large-scale cohort observations from the COPDGene and ECLIPSE studies [13, 14]. Structural emphysema degrades radial traction on peripheral airways, causing premature dynamic airway collapse during forced expiration [15]. This mechanical degradation directly links alveolar loss to physiological airflow limitation. Notably, the strongest correlation in our study occurred between EI_{-950} and DL_CO% predicted (r = -0.746). Carbon monoxide diffusing capacity serves as a direct functional surrogate for the total surface area of the capillary bed. As low-attenuation areas expand, loss of interalveolar septa depletes the pulmonary capillary network, reducing DLCO [16]. Thus, a disproportionate reduction in DLCO relative to FEV_1 should raise immediate suspicion of an emphysema-predominant phenotype, warranting formal HRCT quantification [17]. Small Airway Disease and Bronchial Wall Remodeling Although emphysema has historically received primary focus in CT quantification, peripheral small airway remodeling (bronchi < 2mm internal diameter) represents the principal site of increased airway resistance in COPD [18]. Because modern CT scanners cannot directly visualize bronchi smaller than 2mm, intermediate subsegmental bronchial morphometry (WA% and WT at 3rd–5th generations) serves as a reliable structural surrogate [19]. In our cohort, subsegmental WA% correlated significantly with mid-expiratory flow rate FEF_25-75% predicted (r = -0.586) and plethysmographic airway resistance R_{aw} (r = 0.598). Bronchial wall thickening, secondary to chronic mucosal inflammation, smooth muscle hypertrophy, and peribronchial fibrosis narrows the central lumen, creating fixed anatomical resistance [20]. Furthermore, multivariate linear regression demonstrated that WA% independently predicts FEV_1% predicted alongside EI_{-950}, proving that parenchymal destruction and airway wall thickening exert additive, complementary effects on global airflow obstruction [21]. Clinical and Therapeutic Implications of HRCT Phenotyping Spirometric evaluation alone fails to distinguish between the pathophysiological drivers of airflow limitation [22]. By integrating EI_{-950} and WA%, we successfully classified over 86% of symptomatic COPD patients into discrete structural phenotypes (emphysema-predominant, airway-predominant, or mixed). Phenotypic stratification carries profound therapeutic implications: • Emphysema-Predominant Phenotype: Patients with high EI_{-950} and severe static hyperinflation are prime candidates for lung volume reduction therapies, including endobronchial valves (EBV), lung volume reduction surgery (LVRS), or targeted alpha-1 antitrypsin augmentation where indicated [23]. • Airway-Predominant Phenotype: Patients with elevated WA% and preserved DLCO are more likely to benefit from maximal dual long-acting bronchodilator therapy (LAMA/LABA), targeted anti-inflammatory agents (e.g., inhaled corticosteroids or PDE4 inhibitors), and aggressive bronchial hygiene [24]. • Mixed Phenotype: Represents an advanced disease state requiring comprehensive multi-modal pharmacotherapy and pulmonary rehabilitation [25]. LIMITATIONS Several limitations of this study warrant consideration. First, HRCT scans were acquired exclusively at full inspiration (TLC). Additional expiratory CT scanning was not routinely performed to minimize patient radiation exposure. Expiratory densitometry (EI_{-856} or E/Itext{ ratio}) allows direct quantification of dynamic air trapping, which may further improve correlations with RV/TLC% [26]. Second, manual and semi-automated segmentation of 5th-generation and smaller subsegmental airways remains technically challenging due to motion and partial volume artifacts. Third, our study design was cross-sectional; longitudinal studies tracking changes in EI_{-950} and WA% over time are needed to establish the predictive value of HRCT metrics for disease progression and mortality.