Imaging Markers of Chronic Obstructive Pulmonary Disease on Computed Tomography of the Lungs: A Narrative Review

Abstract

Chronic obstructive pulmonary disease (COPD) is a heterogeneous disorder in which spirometry reflects airflow limitation but incompletely captures the underlying structural injury. Patients with similar spirometric impairment may have markedly different patterns of emphysema, small-airway disease, and pulmonary vascular involvement. Computed tomography (CT) can visualize these abnormalities in vivo and may detect structural changes before overt airflow limitation develops. This narrative review examined whether CT-derived imaging markers provide clinically meaningful information beyond spirometry, particularly for phenotyping, early structural detection, prognostic enrichment, and selective clinical decision support.

A focused narrative review of diverse publications in PubMed, published between January 2000 and April 2025, was conducted using COPD-related MeSH terms and free-text keywords. Across the reviewed literature, emphysema, quantified by low-attenuation area and emphysema index, and air-trapping measures showed consistent associations with expiratory volume measures and identified structural abnormalities in symptomatic smokers with preserved spirometry. Airway wall metrics such as wall area percentage distinguished airway-predominant phenotypes and were associated with airflow limitation and chronic bronchitis. CT vascular markers, including small-vessel cross-sectional area and the pulmonary artery-to-aorta ratio, reflected microvascular disease; a pulmonary artery enlargement to aorta ratio greater than 1 independently predicted severe exacerbations. Xenon CT and dual-energy CT added regional ventilation and perfusion information but remain investigational.

CT imaging markers add complementary structural, phenotypic, and prognostic information beyond spirometry for established COPD and pre-COPD. Their most supported current role is phenotypic characterization and prognostic enrichment in selected clinical settings, whereas advanced functional CT techniques remain research tools. Routine CT for all patients is not supported by current Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidance.

Introduction

Definition and Epidemiology of Chronic Obstructive Pulmonary Disease

Chronic obstructive pulmonary disease (COPD) is defined by the Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2025 report as a heterogeneous lung condition characterized by chronic respiratory symptoms (dyspnea, cough, sputum production, exacerbations) due to abnormalities of the airways (bronchitis, bronchiolitis) and/or alveoli (emphysema) that cause persistent, often progressive, airflow obstruction [1]. The diagnosis requires post-bronchodilator forced expiratory volume at 1 second/forced vital capacity (FEV1/FVC) below 0.70 on spirometry [1].

COPD is a major cause of global morbidity and mortality. The global economic burden was estimated at approximately US$2.1 trillion in 2019, with projections indicating continued growth due to aging populations and sustained risk-factor exposure [2]. Typical symptoms include chronic dyspnea, cough, and sputum production. Acute exacerbations represent key events that accelerate lung function decline and increase mortality risk [1].

Limitations of Spirometry and the Case for Computed Tomography

Spirometry—particularly post-bronchodilator FEV1/FVC—remains mandatory for COPD diagnosis and GOLD grading [1]. However, spirometric indices reflect only a fraction of the underlying structural pathology. There is substantial heterogeneity in symptom burden, exacerbation risk, and prognosis among patients with similar FEV1 values [1,3]. Regan et al demonstrated in the COPDGene cohort that a significant proportion of current and former smokers with preserved spirometry have clinically meaningful respiratory symptoms and structural abnormalities, including emphysema and air trapping, that are detectable by computed tomography (CT) [3]. CT provides in-vivo, 3-dimensional assessment of lung parenchyma, airway morphology, and pulmonary vasculature that complements spirometric evaluation [4,5,6].

Computed Tomography Terminology: From High Resolution Computed Tomography to Multidetector Computed Tomography

The term ‘high-resolution computed tomography’ (HRCT) historically referred to a specific acquisition technique using thin slices (1–1.5 mm), high-frequency kernels, and intermittent scanning—a compromise imposed by the detector configuration of early single-detector scanners [7]. With modern multidetector CT (MDCT), submillimeter isotropic voxels (≤ 0.625 mm) can be reconstructed from any standard volumetric chest acquisition using sharp kernels and iterative reconstruction algorithms. The Fleischner Society acknowledges that ‘HRCT’ is now effectively synonymous with thin-section volumetric CT on contemporary MDCT scanners, and that the functional definition depends on reconstruction parameters rather than a distinct scan protocol [7,8]. Accordingly, throughout this review, the term ‘CT’ or ‘quantitative CT (QCT)’ is used in place of ‘HRCT’, in keeping with current terminology [7–9].

Towards Multidimensional Chronic Obstructive Pulmonary Disease Assessment

Bhatt et al proposed a multidimensional diagnostic schema that integrated respiratory symptoms, quality-of-life scores, spirometry, and CT evidence of emphysema and airway wall thickening, and investigated this diagnostic schema in 9416 participants in the COPDGene and CanCOLD trials [10]. Participants newly classified as having COPD under this schema showed significantly higher all-cause and respiratory-specific mortality, more frequent exacerbations, and faster FEV1 decline [10].

Aims and Scope of the Present Review

This narrative review addresses a central question: do CT-derived imaging markers provide clinically meaningful information beyond spirometry in COPD patients, and if so, at which stages of the disease and for which clinical purposes? Four subordinate questions are examined: (1) whether CT markers improve phenotyping of emphysema-predominant vs airway-predominant disease; (2) whether structural CT abnormalities precede overt spirometric obstruction in at-risk individuals; (3) whether CT biomarkers enrich prognostic assessment beyond GOLD staging; and (4) whether evidence supports selective clinical use or whether markers remain investigational tools. Prior imaging reviews have addressed individual marker domains—emphysema quantification, airway morphometry and vascular imaging—separately. These predated the 2025 GOLD update and recent data on vascular and functional CT techniques. The contribution of this review is integrative: it synthesizes emphysema, airway, vascular, xenon CT, and dual-energy CT markers within a unified clinical framework, positions findings in the context of GOLD 2025 recommendations, and explicitly delineates markers with established adjunctive value from those that remain research tools. This incremental synthesis is intended to support clinically calibrated interpretation rather than to claim novel discovery. The review covers the following topics: (1) quantitative emphysema indices and mean lung density (MLD); (2) expiratory air trapping, functional small airway disease, and parametric response mapping; (3) airway wall remodeling metrics; (4) pulmonary vascular imaging biomarkers; (5) xenon ventilation CT; and (6) dual-energy CT perfusion imaging.

Material and Methods

The following sections details the methods used for literature search and study selection.

Search Strategy

A focused narrative literature search was conducted in PubMed for the period January 2000 to April 2025, combining COPD-specific medical search heading (MeSH) with free-text terms. The MeSH terms were: “Pulmonary Disease, Chronic Obstructive”[MeSH] AND (“Tomography, X-Ray Computed”[MeSH] OR “Tomography, Emission-Computed”[MeSH]). The free-text expansion terms were: (“COPD” OR “chronic obstructive pulmonary disease”) AND (“computed tomography” OR “CT” OR “quantitative CT” OR “QCT” OR “MDCT”) AND (“emphysema” OR “mean lung density” OR “air trapping” OR “small airway” OR “airway wall” OR “pulmonary vessels” OR “pulmonary vascular” OR “microvascular blood flow” OR “dual-energy CT” OR “xenon ventilation” OR “parametric response map”.

The search was restricted to human subjects and English-language publications. Reference lists of included articles, recent imaging reviews, and key imaging cohort studies, including COPDGene, Multi-Ethnic Study of Atherosclerosis (MESA) COPD, Subpopulations and Intermediate Outcome Measures in COPD Study (SPIROMICS), and Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE), were screened for additional studies [4,5,8,9,11,12]. Study selection was performed by the lead author using a 2-stage process: (1) titles and abstracts were screened against the inclusion and exclusion criteria presented in Table 1; (2) the screening was followed by full-text review of potentially eligible records. No formal Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram was applied, as this review is explicitly narrative and selective rather than systematic. The weighting criteria applied to included studies are described in the Limitations section.

Table 1.

Inclusion and exclusion criteria for the literature search.

Inclusion criteria	Exclusion criteria
Studies including adults (≥ 18 years) with established COPD, at-risk individuals, or symptomatic smokers CT or QCT markers of emphysema, air trapping, airway caliber, or pulmonary vasculature Studies relating imaging markers to spirometry, symptoms, GOLD stage, or clinical outcomes Original articles, systematic reviews, and major imaging cohort reports	Pediatric or animal studies Plain radiography only; no CT Non-English publications Case reports (n < 5) Studies focused exclusively on asthma without COPD comparison Studies without lung function data

CT, computed tomography; COPD, chronic obstructive pulmonary disease; GOLD, Global Initiative for Chronic Obstructive Lung Disease; QCT, quantitative CT.

Inclusion and Exclusion Criteria

Table 1 presents the inclusion and exclusion criteria applied.

Limitations

This review is narrative and selective. Priority was given to pivotal original studies [13–17], landmark cohort analyses [8,10,18,19], and comprehensive imaging reviews [4,5,7,9,11,12]. Selection bias is inherent; the review was not pre-registered and no formal PRISMA flow was applied. When drawing interpretive conclusions, studies were weighted according to design quality and clinical relevance. Greater emphasis was placed on longitudinal cohort data, large multi-center studies, and externally validated findings. Cross-sectional studies, single-center reports, and results from specialist techniques with limited availability were treated with commensurate caution and conclusions from these sources are presented as preliminary or hypothesis-generating. Where evidence was primarily cross-sectional or derived from specialist-center techniques, statements are qualified accordingly to avoid overstating clinical readiness.

Computed Tomography Markers of Parenchymal Disease: Emphysema and Mean Lung Density

Definitions and Quantitative Methodology

Emphysema on CT is quantified by densitometry. The most widely used metric is the emphysema index, defined as the percentage of lung voxels with CT attenuation below −950 HU on an inspiratory acquisition (%LAA-950) [4,5,7]. Mean lung density (MLD) is the average voxel attenuation within the segmented lung; increasingly negative values reflect greater air content and parenchymal destruction [4,12].

An alternative densitometry metric is the 15th percentile (Perc15) of the lung attenuation histogram, which is less sensitive to breath-hold depth variability and correlates strongly with histological emphysema extent [8,9]. Measurements can be performed at whole-lung, lobar, or regional level. The Fleischner Society [7] recommends lobe-level analysis as the minimum reporting standard, given the prognostic relevance of lobar distribution (Figure 1).

Figure 1.

Representative inspiratory CT of the chest (axial, coronal, and sagittal planes) demonstrating quantitative emphysema mapping (%LAA-950 color overlay). Areas meeting the −950 HU threshold are highlighted, illustrating upper-lobe-predominant centrilobular emphysema in a GOLD II COPD patient. Reproduced with permission from [4]. CT, computed tomography;%LAA-950,% lung voxels ≤-950 Hounsfield units; GOLD, Global Initiative for Chronic Obstructive Lung Disease; COPD, chronic obstructive pulmonary disease.

Pre-Chronic Obstructive Pulmonary Disease and High-Risk Populations

Lv et al conducted a retrospective single-center cohort study in 158 individuals at risk of COPD (age ≥ 40 years, with chronic symptoms or risk-factor exposure, and with an FEV1/FVC ratio between 0.70 and 0.80) [13]. The high-risk group (FEV1/FVC 0.70–0.80, n=72) had significantly higher emphysema index, higher air-trapping index (ATI) and lower expiratory MLD than the low-risk group (FEV1/FVC ≥ 0.80, n=86; all P<0.05). In multivariable analysis, emphysema index, ATI, and expiratory MLD were independent predictors of reduced FEV1/FVC [13].

Zhang et al studied 140 participants with FEV1/FVC >0.70 but who had respiratory symptoms [14]. Using QCT and machine learning, participants classified as high-risk (FEV1 0.8–0.95 predicted) had significantly greater%LAA-950 in the left lower lobe, which correlated with FEV1 (r=−0.41, P<0.001) and dyspnea scores [14].

Emphysema in Established Chronic Obstructive Pulmonary Disease

Hartley et al enrolled 81 COPD patients, 171 asthmatic patients, and 49 healthy controls undergoing QCT [15]. Emphysema extent and reduced MLD correlated moderately with FEV1 (r=−0.54, P<0.001) and FEV1/FVC (r=−0.48, P<0.001) independently of airway wall metrics and air trapping in multivariable regression [15].

Ley-Zaporozhan and van Beek reviewed CT imaging phenotypes and reported that emphysema-predominant COPD is associated with worse carbon monoxide diffusing capacity (DLCO), more severe dyspnea, and higher long-term mortality, especially with lower-lobe-predominant centrilobular disease [5].

Longitudinal Emphysema Progression

Pompe et al analyzed 5-year progression of emphysema (%LAA-950) in 4117 COPDGene participants [17]. Participants with pre-existing emphysema who continued smoking had the greatest annual lung density decline (mean −0.94 g/L/year vs −0.32 g/L/year in non-smokers). Emphysema progression independently predicted FEV1 decline and all-cause mortality at 10 years [17].

Synthesis

To summarize: (1) quantitative emphysema measures show consistent moderate inverse correlations with FEV1 and FEV1/FVC [4,5,13–15]; (2) quantitative emphysema measures can detect structural abnormalities in symptomatic smokers with preserved spirometry [3,13,14]; (3) lobar distribution patterns carry prognostic information [5,7,8]; (4) longitudinal progression correlates with smoking status and predicts mortality [17]; (5) measurements are sensitive to acquisition parameters and require standardization [4,7,9,11,12].

Air Trapping and Functional Small Airway Disease

Concept and Quantitative Measurement

Air trapping refers to the retention of gas in distal lung units during expiration due to incomplete emptying of poorly ventilated lobules. On paired inspiratory–expiratory CT, it manifests as regions of relatively low attenuation on expiratory images compared with adjacent lung [4,11]. Quantitative indices of air trapping include the percentage of lung voxels below −856 HU on expiration (%LAA-856exp); the expiratory-to-inspiratory mean lung density ratio (E/I ratio); and the ATI, reflecting lobular gas redistribution attributable to small airway obstruction [4,11,13] (Figure 2).

Figure 2.

Paired inspiratory-expiratory CT images demonstrating air trapping and PRM. Voxels classified as normal lung (green), fSAD (yellow), and emphysema (red) are shown in a GOLD II patient with airway-predominant disease. Reproduced with permission from [14]. CT, computed tomography; PRM, parametric response mapping; fSAD, functional small airway disease; GOLD, Global Initiative for Chronic Obstructive Lung Disease.

Parametric Response Mapping for Small Airway Disease

An important methodological advance was parametric response mapping (PRM) by Galbán et al [20], which co-registers inspiratory and expiratory CT volumes at the voxel level to classify each voxel as normal lung, emphysema, or functional small airway disease (fSAD). PRM demonstrated that fSAD and emphysema are spatially distinct processes that co-exist in variable proportions. fSAD burden was independently associated with FEV1 decline and risk of exacerbation in COPDGene participants even after accounting for emphysema extent [20].

Air Trapping in Pre-Chronic Obstructive Pulmonary Disease

In the 158 at-risk individuals studied by Lv et al, expiratory ATI differentiated high-risk from low-risk groups more clearly than emphysema index (correlation with FEV1/FVC: r=−0.44 for ATI vs r=−0.26 for emphysema index, P<0.01 for both) [13]. Zhang et al similarly reported significantly higher%LAA-856exp and greater ATI in their high-risk group, with ATI correlating with dyspnea scores (r=0.37, P<0.01) [14].

Air Trapping in Established Chronic Obstructive Pulmonary Disease

Hartley et al demonstrated that QCT air-trapping indices were independently associated with FEV1 (r=−0.61, P<0.001), residual volume, and GOLD stage, after adjustment for emphysema extent and wall area percentage (WA%) [15]. Benlala et al concluded that expiratory CT is the most robust CT surrogate of functional small airway disease, while cautioning that lobular air trapping occurs in up to 10 to 15% of healthy non-smokers and is highly sensitive to expiratory lung volume and reconstruction kernel [11].

Synthesis

To summarize: air-trapping metrics detect functional small airway disease before overt spirometric obstruction [13,14]; PRM separates fSAD from emphysema at the voxel level, providing phenotypic resolution unavailable from spirometry [20]; ATI and E/I correlate with physiological hyperinflation and obstruction [11,13–15]; accurate quantification requires standardized lung volumes and reconstruction kernels [7,9,11].

Computed Tomography Markers of Airway Disease

Quantitative Airway Morphometry

Direct visualization of small conducting airways (<2 mm) lies below clinical CT resolution [4,5,11]. Airway disease is therefore assessed indirectly using central cartilaginous bronchi (third to sixth generation) as surrogates. Principal metrics include WA% (WA% = 100 × wall area / total bronchial cross-sectional area), wall-thickness to outer-diameter ratio, and luminal area [4,5,15]. WA% measured at fifth-generation airways correlates more strongly with FEV1 than measurements at third-generation airways, reflecting the greater contribution of distal remodeling to flow limitation [4,5].

Airway Wall Metrics in Established Chronic Obstructive Pulmonary Disease

In the QCT study by Hartley et al (n=81 COPD, n=171 asthma, n=49 controls), proximal WA% was significantly increased in both COPD (62.7% ± 2.3%) and asthma (62.5% ± 2.2%) patients, compared with controls (60.3% ± 2.2%, P<0.001) [15]. After adjusting for emphysema and air trapping, higher WA% and smaller luminal area remained independently associated with reduced FEV1 (β=−0.31, P<0.01) [15]. Airway-predominant COPD phenotypes showed relatively greater wall thickening and less emphysema, with stronger associations between WA% and FEV1 in this subgroup [15].

Heterogeneity of Airway Remodeling

Benlala et al highlighted evidence that, after careful size-matching, some COPD phenotypes—particularly those with advanced emphysema—exhibit thinner rather than thicker central airway walls compared with healthy controls [11]. This counterintuitive finding may reflect emphysema-related over-distension of small airways or alveolar wall destruction spreading to bronchial adventitia. It underscores the need for phenotype-specific analysis rather than global WA% thresholds [11].

Bronchiectasis in Chronic Obstructive Pulmonary Disease

Bronchiectasis is increasingly recognized as a CT-definable comorbidity in COPD, with a prevalence of 30–50% in some series. Its presence is associated with more frequent exacerbations, higher sputum bacterial load, and worse health-status scores, and may warrant specific therapeutic approaches [4,11].

Synthesis

To summarize: WA% and luminal area quantify distal airway remodeling and distinguish airway-predominant from emphysema-predominant COPD [4,5,15]; both thickened and thinned airway walls occur depending on disease subtype [11]; bronchiectasis co-existence identifies a higher-risk subgroup [4,11]; standardized measurement protocols specifying airway generation and reconstruction kernel are needed [7,9].

Pulmonary Vascular Imaging Markers

Computed Tomography-Derived Vascular Pruning and Small-Vessel Cross-Sectional Area

Estépar et al measured cross-sectional area (CSA) of vessels <5 mm² in 9156 COPDGene participants, and found that lower small-vessel CSA was independently associated with greater emphysema extent, worse airflow limitation (β=0.33, P<0.001), and increased 5-year mortality [18]. Matsuoka et al reported significant negative correlations between small-vessel CSA and FEV1 (r=−0.65, P<0.001) and DLCO (r=−0.72, P<0.001), and positive correlations with mean pulmonary artery pressure on right heart catheterization [21].

Pulmonary Artery-to-Aorta Diameter Ratio and Exacerbation Risk

In the COPDGene and ECLIPSE analysis by Wells et al (n=3266), pulmonary artery-to-aorta diameter ratio (PA: A) >1 was associated with a 4-fold increase in the odds of previous severe exacerbations (OR 4.78, 95% CI 3.43–6.65) and a 3.4-fold increase in the risk of future severe exacerbations (OR 3.44, 95% CI 2.78–4.25), independently of established risk factors [19]. Iyer et al validated PA: A >1 against echocardiographic pulmonary artery systolic pressure in severe COPD, finding a sensitivity of 74% and specificity of 81% for diagnosing pulmonary hypertension [22].

Pulmonary Microvascular Blood Flow

Hueper et al used dynamic gadolinium-enhanced MRI to measure pulmonary microvascular blood flow (PMBF) in 144 smokers with and without COPD in the MESA COPD Study [16]. PMBF was reduced by 30% in mild COPD (GOLD I), by 29% in moderate COPD (GOLD II), and by 52% in severe COPD (GOLD III–IV), compared with controls (P<0.01 for each). PMBF was reduced even in lung regions not meeting CT thresholds for emphysema, suggesting that microvascular damage may precede overt parenchymal destruction (Figure 3) [16].

Figure 3.

Dynamic gadolinium-enhanced MRI perfusion maps (MESA COPD protocol) in a healthy non-smoker (left) vs a GOLD II COPD patient (right), illustrating globally reduced PMBF extending beyond regions of CT-defined emphysema. Reproduced with permission from [16]. MRI, magnetic resonance imaging; MESA, Multi-Ethnic Study of Atherosclerosis; COPD, chronic obstructive pulmonary disease; GOLD, Global Initiative for Chronic Obstructive Lung Disease; PMBF, pulmonary microvascular blood flow; CT, computed tomography.

Vascular Markers and Pulmonary Hypertension

Benlala et al summarized evidence that CT small-vessel CSA and PA: A ratio correlate with invasively measured mean pulmonary artery pressure (mPAP) in COPD. In mild-to-moderate pulmonary hypertension secondary to COPD, small-vessel CSA negatively correlates with mPAP (r=−0.58 to −0.71 in pooled analyses) [11,18,21].

Synthesis

To summarize: Small-vessel CSA and PA: A ratio are objective vascular markers available from standard non-contrast CT [18,19,21,22]; PA: A >1 robustly predicts severe COPD exacerbations independently of spirometric severity [19]; MRI-derived PMBF demonstrates microvascular impairment extending beyond structural emphysema [16]; vascular markers may guide pulmonary hypertension evaluation, though no CT-specific therapeutic recommendations currently exist [1,11].

Xenon Ventilation Computed Tomography

Principle and Technique

Regional ventilation can be assessed using stable, non-radioactive xenon as an inhaled contrast agent. Inhaled xenon attenuates the X-ray beam in proportion to its local concentration, so dynamic CT acquisition during xenon wash-in and washout generates voxel-level ventilation maps. Chae et al provided the first clinical demonstration of xenon CT feasibility in humans using a dual-source scanner with a dual-energy technique, showing that regional density changes during xenon inhalation could be reliably quantified [23]. This technique is physiologically distinct from dual-energy CT perfusion imaging (Figure 4).

Figure 4.

Xenon ventilation dual-energy CT in a GOLD III COPD patient. The functional map (right) demonstrates multifocal ventilation defects, several localized to areas appearing structurally normal on the corresponding conventional CT image (left). Note: xenon CT maps regional ventilation using inhaled xenon as contrast; this is physiologically distinct from dual-energy CT perfusion imaging, which maps regional pulmonary blood volume using intravenous iodinated contrast. Reproduced with permission from [24]. CT, computed tomography; GOLD, Global Initiative for Chronic Obstructive Lung Disease; COPD, chronic obstructive pulmonary disease.

Clinical Chronic Obstructive Pulmonary Disease Studies

Park et al applied xenon ventilation dual-energy CT (DECT) across GOLD stages I–IV and demonstrated that ventilation defect volume correlated significantly with FEV1 (r=−0.68, P<0.001) and FEV1/FVC (r=−0.72, P<0.001). Ventilation defects were more extensive and spatially heterogeneous in higher GOLD stages. Xenon-based ventilation maps were able to identify functional impairment within lobes that appeared structurally normal on conventional CT [24].

Limitations

The limitations of this approach include: (1) additional radiation exposure beyond standard chest CT; (2) requirement for specialized dual-energy or dual-source CT equipment; (3) inhaled xenon contrast with controlled breathing protocols; (4) limited availability outside specialist centers; and (5) absence of large outcomes-based studies demonstrating clinical benefit [12,24].

Synthesis

To summarize: xenon ventilation CT provides voxel-level ventilation maps revealing functional impairment beyond structural imaging [23,24]; ventilation defect volume correlates with spirometric obstruction [24]; however, xenon ventilation CT is currently only a research tool due to radiation, cost, availability, and limited clinical evidence [12,24].

Dual-Energy Computed Tomography Perfusion Imaging

Principle and Technique

DECT uses simultaneous acquisition at 2 different tube voltages (typically 80/140 kVp) to exploit differential X-ray attenuation of iodine vs soft tissue. Combined with contrast-enhanced CT, DECT generates iodine distribution maps as surrogates for regional pulmonary blood volume and perfusion [12]. This technique is physiologically distinct from xenon ventilation CT. Xenon ventilation CT maps regional ventilation, whereas DECT with iodinated contrast maps regional perfusion. Combining both provides ventilation–perfusion (V/Q) information from a single CT session [12,24].

Chronic Obstructive Pulmonary Disease Applications

Perfusion maps derived from DECT in COPD show regional blood flow defects overlapping with, but extending beyond, emphysematous regions on structural CT. Reductions in small-vessel CSA on structural CT correspond to DECT iodine distribution deficits [12,18]. Early work reviewed by Hoffman et al demonstrated that DECT perfusion deficits are associated with worse FEV1 and residual volume/total lung capacity, and that combined V/Q assessment may identify patients with disproportionate vascular involvement, though this remains investigational [12].

Limitations and Synthesis

To summarize: Limitations of DECT are that it requires dual-energy or spectral CT hardware, and intravenous iodinated contrast, which is contraindicated in patients with renal impairment and allergy. Furthermore, there are no established clinical thresholds or outcome-based evidence for DECT-guided COPD management. Advantages are that DECT perfusion provides regional blood volume maps that are complementary to structural emphysema assessment and that it enables V/Q mapping when combined with xenon ventilation CT. However, it remains solely a research tool [12,24].

Clinical Integration, Limitations, and Guideline Context

Key Overarching Observations

Emphysema, MLD, and air trapping show consistent moderate correlations with airflow limitation, including FEV1 and FEV1/FVC, but substantial overlap exists between GOLD stages [5,13–15,17]. Smokers with preserved spirometry frequently have CT evidence of emphysema or air trapping and report significant dyspnea [3,13,14]. Airway wall metrics identify airway-predominant phenotypes associated with chronic bronchitis symptoms, higher exacerbation risk, and systemic inflammation [4,5,11,15]. Vascular pruning and PMBF reduction indicate microvascular disease—an integral, early pathological component of COPD, with implications for pulmonary hypertension risk and prognosis [16,18,19,21,22]. Multidimensional schemas integrating CT evidence with symptoms and spirometry identify structural and phenotypic heterogeneity not captured by spirometry alone, supporting prognostic enrichment. However, evidence of any direct management benefit in routine practice remains limited [10].

Summary Table of Computed Tomography Imaging Markers

Table 2 summarizes the main CT imaging markers, their metrics, key associations, clinical utility, and principal limitations.

Table 2.

Summary of CT imaging markers of COPD.

CT Marker	Metric	Key associations	Clinical use	Main limitation
Emphysema / MLD	%LAA-950, Perc15, MLD	FEV1, FEV1/FVC, DLCO, dyspnea; progression with smoking [4,5,6,23]	LVRS candidate selection; pre-COPD detection [1,19]	Volume sensitivity; scanner variability [19,24]
Air trapping / fSAD	%LAA-856exp, E/I, ATI, PRM-fSAD	FEV1, RV/TLC, exacerbations; detects disease before obstruction [4,5,6,22]	Research; early COPD phenotyping [22]	Expiratory lung volume control [8]
Airway morphometry	WA%, luminal area, TD ratio	FEV1, FEV1/FVC, chronic bronchitis, exacerbations [2,3,6]	Airway-predominant phenotype identification [3]	No standard threshold; generation-dependent [19]
Vascular pruning	Small-vessel CSA (<5 mm2)	Emphysema, airflow limitation, mPAP, mortality [13,14]	PH evaluation; prognostication [13]	Software required; observer-dependent [24]
PA: A ratio	PA diameter / aorta diameter	Severe exacerbations (OR 4.78), PH [15,16]	Exacerbation risk stratification [15]	Not specific for PH [16]
MRI PMBF	Gadolinium MRI perfusion	Reduced in mild COPD (−30%); beyond emphysema regions [7]	Research only [7]	MRI access; contrast required [7]
Xe ventilation CT	Ventilation defect volume	FEV1, FEV1/FVC; regional heterogeneity [17]	Research only [17,18]	Radiation, cost, Xe availability [10,17]
DECT perfusion	Iodine distribution map	Perfusion defects in emphysema; V/Q mismatch [10]	Research only [10]	Contrast required; DECT hardware [10]

ATI, air-trapping index; COPD, chronic obstructive pulmonary disease; CSA, cross-sectional area; CT, computed tomography; DECT, dual-energy CT; DLCO, diffusing capacity for carbon monoxide; E/I, expiratory-to-inspiratory ratio; fSAD, functional small airway disease; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; LVRS, lung volume reduction surgery; MLD, mean lung density; mPAP, mean pulmonary artery pressure; MRI, magnetic resonance imaging; PA, pulmonary artery; PA: A, pulmonary artery-to-aorta ratio; Perc15, 15th percentile of the lung attenuation histogram; PH, pulmonary hypertension; PMBF, pulmonary microvascular blood flow; PRM, parametric response mapping; RV, residual volume; TD ratio, wall-thickness to outer-diameter ratio; TLC, total lung capacity; V/Q, ventilation–perfusion; WA%, percentage airway wall area; Xe, xenon; %LAA-950, % lung voxels ≤-950 Hounsfield units; %LAA-856exp, % lung voxels ≤-856 Hounsfield units on expiration.

Evidence Interpretation Framework

Interpreting the CT imaging literature requires distinguishing between 4 levels of evidence: (1) structural association, in which a CT marker correlates with spirometric or physiological measures; (2) prognostic enrichment, in which a marker independently predicts exacerbations, lung function decline, or mortality; (3) mechanistic insight, when imaging reveals pathophysiological processes not accessible through spirometry; and (4) demonstrated clinical utility, when CT-guided decisions improve patient-centered outcomes. Most CT imaging markers have well-established evidence at levels 1 and 2. Evidence at level 3 is emerging, particularly for microvascular markers and functional CT. Evidence at level 4—demonstrated benefit on quality of life, hospitalization, or survival through CT-guided therapeutic strategies—is currently absent for all marker classes, including emphysema quantification, airway morphometry, and vascular imaging. Where this review uses phrases such as “clinically meaningful” or “supports multidimensional assessment,” these refer to phenotypic and prognostic enrichment rather than established management-changing utility, unless directly supported by interventional evidence.

Methodological Limitations

One limitation is that most of the included studies were cross-sectional. In addition, quantitative CT markers are sensitive to scanner type, kVp, kernel, slice thickness, and lung volume [4,5,7,9,11,12].

No universally accepted acquisition or analysis standard exists, although Fleischner Society and COPDGene protocols are important steps towards harmonization [7,8]. Finally, this review is narrative; selection bias is inherent.

Clinical Limitations and Guideline Context

CT involves ionizing radiation, so serial use requires justification. GOLD 2025 guidelines do not require CT for COPD diagnosis, which remains spirometry-based [1]. CT is recommended selectively for alternative diagnoses, including lung volume reduction surgery candidacy assessment and investigation of complications or comorbidities [1,6]. No randomized clinical trial evidence currently demonstrates that routine CT-based phenotyping improves quality of life, exacerbation rate, hospitalizations, or survival.

Conclusions

CT-derived imaging markers provide complementary structural and functional information extending beyond spirometry for COPD patients and individuals at risk of COPD. On the basis of the evidence reviewed, quantitative emphysema (%LAA-950) and air-trapping indices (%LAA-856exp, ATI, PRM-fSAD) correlate with FEV1 and FEV1/FVC, and can identify structural lung disease in symptomatic smokers with preserved spirometry [3,13,14,17,20]. Airway wall morphometry (WA%, luminal area) defines airway-predominant COPD phenotypes, associates these phenotypes with chronic bronchitis symptoms and exacerbation risk, and reveals both thickened and thinned remodeling patterns [5,11,15].

Vascular imaging markers—CT small-vessel CSA, PA: A ratio, and MRI-derived PMBF—highlight pulmonary microvascular remodeling as an early, integral component of COPD pathophysiology; PA: A >1 independently predicts severe exacerbations [16,18,19,21,22]. Multidimensional assessment schemas incorporating CT into workups based on symptoms and spirometry could identify structural and phenotypic heterogeneity not captured by spirometry alone, supporting prognostic enrichment in selected populations. However, direct management benefit remains to be demonstrated in prospective trials [10]. Xenon ventilation CT and DECT perfusion are complementary but physiologically distinct functional techniques: the former provides regional ventilation maps, and the latter provides regional perfusion maps. Both currently remain research tools because of radiation, cost, and the absence of evidence on outcomes [12,23,24].

CT should be regarded as a targeted adjunct to spirometry, with its strongest current role lying in phenotypic and prognostic enrichment in selected clinical settings rather than routine universal implementation. Whereas standard CT markers (emphysema quantification, airway morphometry, vascular indices) have established adjunctive value for phenotyping and prognostication, advanced functional techniques (xenon CT, DECT perfusion) remain investigational. Future research priorities include standardization of QCT protocols [7,8], prospective validation of CT biomarkers against patient-centered outcomes, and investigation of imaging-guided therapeutic strategies in vascular and small airway disease subphenotypes [6,9].

Abbreviations

COPD

chronic obstructive pulmonary disease

CT

computed tomography

MDCT

multidetector computed tomography

QCT

quantitative computed tomography

FEV1

forced expiratory volume in 1 second

FVC

forced vital capacity

DLCO

diffusing capacity for carbon monoxide

MLD

mean lung density

%LAA-950

percentage of lung voxels with attenuation ≤-950 HU

ATI

air-trapping index

fSAD

functional small airway disease

PRM

parametric response mapping

WA%

percentage airway wall area

CSA

cross-sectional area

PA: A

pulmonary artery-to-aorta diameter ratio

PMBF

pulmonary microvascular blood flow

DECT

dual-energy computed tomography

GOLD

Global Initiative for Chronic Obstructive Lung Disease

PH

pulmonary hypertension

mPAP

mean pulmonary artery pressure

V/Q

ventilation–perfusion