Emphysema Imaging

Conventional chest radiography is generally the first imaging procedure performed in patients with respiratory symptoms, and frontal and lateral chest radiographs may reveal changes of emphysema. A chest radiograph is universally available, noninvasive, and inexpensive, and it poses an acceptable radiation exposure. ^{[1, 2, 3, 4]}  Chest radiographic findings are not good indicators of the severity of disease and do not help in identifying patients with chronic obstructive pulmonary disease (COPD) without clinically significant emphysema. ^{[5, 6]}  Although COPD affects close to 30 million individuals in the United States, it has been reported to be underdiagnosed because of underutilization of spirometry and because respiratory symptoms are often attributed to older age or physical deconditioning. ^{[7, 8, 9]}

Pulmonary interstitial emphysema (PIE) is a rare condition that occurs primarily in neonates but can also occur in adults. In PIE, increased air pressure in the alveoli and alveolar airspaces disrupts adjacent lung interstitial tissue and damages lung structures, causing linear and cystic spaces. Leaked air can then collect outside normal air passages and within the interstitium or bronchovascular complexes. ^[10]

High-resolution computed tomography (HRCT) scanning is more sensitive than chest radiography in diagnosing emphysema and in determining its type and extent of disease. ^[11] HRCT also has a high specificity for diagnosing emphysema with virtually no false-positive diagnoses, but in clinical practice, more reliance is placed on patient history, lung function tests, and abnormal chest radiographs to diagnose emphysema. However, some patients with early emphysema, particularly those with early disease, may present with atypical symptoms, and it is in these patients that an HRCT is most rewarding. If significant emphysema is found on HRCT, no further workup is necessary; specifically, lung biopsy is not needed. Imaging information from HRCT does not alter the management of emphysema; therefore, HRCT has no place in the day-to-day care of patients with COPD. In their early stages, the 3 forms of emphysema can be distinguished morphologically by using HRCT, but as the disease becomes more extensive, the distinction becomes difficult or impossible, both radiographically and pathologically. ^{[12, 13]}

A number of studes have assessed the role of computed tomography (CT) in the early detection of lung cancer in patients with COPD and in predicting response to lung-volume–reduction surgery (LVRS). ^[14]  CT scans are used to characterize the emphysema and degree of destruction on a lobar basis, to evaluate the distribution of the emphysema destruction, and to determine the integrity of the lobar fissures. ^[15]  CT characterization of heterogeneous parenchymal abnormalities can provide criteria for selection of the preferable treatment in each patient and improve the  outcome of LVR. In particular, emphysema distribution pattern and fissures integrity can be evaluated to tailor the choice of the most appropriate LVR technique. ^{[16, 17]}  Radionuclide scanning and MRI may have a potential role in patients being assessed for LVRS. 

(Images of emphysema are displayed below.)

Ballile and Laennec described the anatomopathology of emphysema in 1793 and in 1826, respectively. After that, Lynne Reid published one of the landmark works in our understanding of emphysema, The Pathology of Emphysema, ^[18] which provided a detailed description of the anatomy of the lung units and of the anatomopathology and pathophysiology of emphysema, broadening the view of this complex disease. ^{[18, 19]}

Pulmonary emphysema is defined as the permanent enlargement of airspaces distal to the terminal bronchioles and the destruction of the alveolar walls. Pathology reveals a marked increase in the size of the airspaces, resulting in labored breathing and an increased susceptibility to infection. It can be caused by irreversible expansion of the alveoli or by the destruction of alveolar walls. Fibrosis is not associated with this condition.

Pulmonary emphysema and chronic bronchitis are important components of chronic obstructive pulmonary disease. Emphysema often coexists with chronic bronchitis in the COPD population, and from a clinical point of view, they are generally considered as one entity. Although a tissue diagnosis of emphysema is possible, in advanced cases it can usually be confidently diagnosed on the basis of the patient's history, physical findings, pulmonary function, and imaging results.

In moderate to severe emphysema, chest radiographic findings include bilaterally hyperlucent lungs of large volume, flattened hemidiaphragms with widened costophrenic angles, horizontal ribs, and a narrow mediastinum. The peripheral vascular markings are attenuated, but the markings become prominent when the patient has pulmonary hypertension and right-sided heart failure. A lateral view shows increased retrosternal airspace and flattening of the anterior diaphragmatic angle. In addition, bullae and an irregular distribution of the lung vasculature may be present. When pulmonary hypertension develops, the hilar vascular shadows become prominent, with filling of the lower retrosternal airspace due to right ventricular enlargement.

(See the images below.)

In clinical practice, reliance is placed on the patient's history, lung function, and abnormal chest radiographs to diagnose emphysema. Chest radiographic findings generally cannot establish the diagnosis of mild emphysema; however, when emphysema is fully established, classic radiographic findings are typically observed. Findings on routine chest radiographs can suggest emphysema, but this is not a sensitive technique for diagnosis. However, chest radiography is useful to look for complications during acute exacerbations and to exclude other pathologies, such as superadded infection or lung cancer.

The chest radiograph is not a good indicator of the severity of disease and does not help in identifying patients with COPD without significant emphysema. Thurlbeck and Simon found that only 41% of those with moderately severe emphysema and two thirds of those with severe emphysema had evidence of disease on chest radiography. ^[20]

CT scanning of the chest, especially high-resolution CT (HRCT), has a much greater sensitivity and specificity than those of plain chest radiography in diagnosing and assessing the severity of emphysema (see the images below). ^[21]  HRCT is useful in the workup of smokers with new-onset or progressive dyspnea. The severity of emphysematous change may be underestimated on conventional radiography, whereas HRCT depicts combined fibrosis and emphysema. ^[22] Patients with these conditions may have relatively normal lung volumes and spirometric results, but they may have severe dyspnea and a reduced diffusing capacity. In healthy nonsmokers aged 19-40 years, a maximum of 0.35% of the area of emphysema can be detected by means of CT quantification. ^[23]

CT can depict surgically treatable areas of bullous disease that are not evident on plain chest radiography. CT is also useful in predicting the outcome of surgery. HRCT may be useful in diagnosing subclinical or mild emphysema, and HRCT can be used to differentiate the pathologic types of emphysema. However, CT scanning is not yet used to routinely evaluate patients with COPD. Instead, it is reserved for patients in whom the diagnosis is in doubt, to look for coexistent pathologies, and to assess their suitability for surgical intervention. ^{[11, 24, 25, 26, 27, 28, 29, 30, 31, 32]}

Hruban et al ^[33] and Bergin C et al ^[34] have shown an excellent correlation between HRCT and histologic findings. They used low-resolution (10 mm) scans and were still able to show that CT findings were better predictors of emphysema than results of pulmonary function tests.

Semiautomated assessment of emphysema by using HRCT data is possible and can help eliminate interobserver and intraobserver variability and provide a reproducible assessment of the percentage of lung affected. Gould et al ^[35] measured the mean density in vivo of the lowest fifth percentile of the distribution of pixels and compared it with a computed quantification of emphysema on the subsequently excised lungs and identified a strong correlation between lung attenuation and distal airspace size.

Müller et al ^[36] and Kinsella et al ^[37] used a CT attenuation mask to highlight voxels in a given attenuation range to quantitate emphysema and define areas of abnormally low attenuation. They compared different masks, mean lung attenuations, and visual appearances and pathologic grades of emphysema in 28 patients undergoing lung resection for tumor. In each patient, a single representative CT image was compared with corresponding pathologic specimens. They found good correlation between the extent of emphysema as assessed by using the attenuation mask and the pathologic grade. Such methods not only eliminate interobserver and intraobserver variability but also enable reproducible assessment of the percentage of lung that is affected. Although quantitative CT measurements have problems, they may improve our understanding of lung function. ^[38]

Kuwano et al ^[39] visually quantified emphysema on 1- and 5-mm HRCT scans by using resected specimens. They found an excellent correlation; however, as expected, the 5-mm sections tended to cause underestimation of the degree of emphysema. They concluded that HRCT scans could depict mild emphysema in patients without clinical evidence of airflow limitation and that they could be used to exclude emphysema in patients with moderate or severe airflow limitation.

Miller et al ^[40] found that CT can cause underestimation of the extent of emphysema when lesions are less than 0.5 cm. However, in their study, the inflation pressures of the fixed lung specimens were not controlled, and a number of their patients had only thick-section (10 mm) studies. Using 1.5- and 10-mm collimation scans, Miller et al showed that the extent of centriacinar and panacinar emphysema was consistently underestimated with CT, because it missed most lesions less than 0.5 cm in diameter. They concluded that CT is insensitive in detecting the earliest lesions of emphysema.

Increased airway wall thickness and lung parenchymal destruction contribute to airflow imitation in emphysema. Advances in CT postprocessing imaging have helped quantify this feature. Dijkstra et al in a population survey studied the relationship of increased airway thickening, lung function, emphysema, and respiratory symptoms. The authors concluded that postprocessing standardization of airway wall measurements provides a reliable and useful method to assess airway wall thickness. The authors also suggested that increased airway wall thickness contributed more to airflow limitation than emphysema in a smoking male population, even after adjustment for smoking behavior. ^[41]

Mild to moderate degrees of centrilobular emphysema are depicted on HRCT as small, round areas of low attenuation, several millimeters in diameter, grouped near the center of secondary pulmonary lobules, with no discernible walls in many cases. Although the centrilobular location of these lucencies cannot always be appreciated on HRCT, lung parenchymal changes are diagnostic of emphysema. ^[42]

Alpha 1-PI deficiency is classically associated with panlobular emphysema, though panlobular emphysema may also be seen in smokers without alpha 1-PI deficiency, in the elderly, and in people with distal bronchial and bronchiolar obliteration. It is almost always most severe in the lower lobes. In severe panlobular emphysema, the characteristic HRCT appearance is that of decreased lung attenuation, with few visible pulmonary vessels in the abnormal regions; bullae or cysts are characteristically absent. Mild and even moderately severe panlobular emphysema can be subtle and difficult to detect. ^[43]  

The Fleischner Society classified the patterns of emphysema severity. Parenchymal emphysema is classified as trace centrilobular emphysema (CLE), mild CLE, moderate CLE, confluent emphysema, or advanced destructive emphysema. Paraseptal emphysema (PSE) is classified as mild or substantial. Because true panlobular emphysema seems to be uncommon in smoking-related emphysema, this classification applies the terms "confluent emphysema" and "advanced destructive emphysema" to what previously was called panlobular emphysema, and the term panlobular emphysema is used for the emphysema found in patients with alpha-1 antitrypsin deficiency. ^{[44, 45, 46, 47]}

Paraseptal emphysema usually involves the distal part of the secondary lobule and is therefore most obvious in subpleural regions. Paraseptal emphysema may be seen in isolation or in combination with centrilobular emphysema. It is often asymptomatic, but it can be associated with spontaneous pneumothorax in young adults. As demonstrated in the image below, HRCT shows the bullae or air cysts associated with paraseptal emphysema well despite their thin walls.

Bullous emphysema is generally seen in association with centriacinar emphysema and paraseptal emphysema. ^[48] Although a bullous emphysema is not a specific pathologic entity, a syndrome of giant bullous emphysema or vanishing lung syndrome has been described on the basis of clinical and radiologic features. Giant bullous emphysema is often seen in young men in association with large, progressive upper-lobe bullae that occupy a considerable volume of a hemithorax. Most patients with giant bullous emphysema smoke cigarettes, but this entity may also occur in nonsmokers.

Yahaba et al performed a study on 91 patients who underwent inspiratory and expiratory multidetector CT (MDCT) to determine whether emphysematous changes alter the relationships between airflow limitation and airway dimensions. Images were evaluated for mean airway luminal area (Ai); wall area percentage (WA%) from the third to the fifth generation of 3 bronchi (B1, B5, B8) in the right lung; and low attenuation volume percent (LAV%). In patients without emphysema, Ai and WA% from both the inspiratory and expiratory scans were significantly correlated with FEV1, but no correlation was found in patients with emphysema. Also, emphysematous COPD patients with GOLD (Global Initiative for Chronic Obstructive Lung Disease) stage 1 or 2 disease had significantly lower changes in B8 Ai than nonemphysematous patients. ^[49]

In a retrospective cohort study by Kurashima et al, the presence of CT-diagnosed emphysema predicted poor prognosis in patients who had COPD and asthma with chronic airflow obstruction (CAO). Patients with asthma without emphysema had the best prognosis, followed by those with asthma with emphysema, COPD without emphysema, and COPD with emphysema. ^[50]

Using low-dose CT scans of 62,124 current, former, and never smokers, emphysema was identified in 28.5% (6,684) of current smokers, 20.6% (5,422) of former smokers, and 1.6% (194) of never smokers. The prevalence of lung cancer in current smokers was 1.1% for those without emphysema and 2.3% for those with emphysema; in former smokers, 0.9% for those without emphysema and 1.8% for those with emphysema; and in never smokers, 0.4% for those without emphysema and 2.6% for those with emphysema. ^[51]

In a meta-analysis, by Yang et al, of the association between chest CT-defined emphysema and  lung cancer, the probability of lung cancer was found to increase with emphysema severity, with both visual and quantitative CT assessments of emphysema being associated with a higher odds of lung cancer. Only centrilobular emphysema was significantly associated with lung cancer. ^[52]

In the COPDGene retrospective cohort study, an objective CT analysis tool was used to measure interstitial features (reticular changes, honeycombing, centrilobular nodules, linear scar, nodular changes, subpleural lines, and ground-glass opacities) and emphysema in 8266 participants. The combined presence of interstitial features and emphysema was associated with worse clinical disease severity and higher mortality than was emphysema alone. In addition, interstitial features enhanced the deleterious effects of emphysema on clinical disease severity and mortality. ^[53]

Quantitative emphysema measured on low-dose CT (LDCT) has been shown to be associated with lung cancer incidence and mortality. In a study by Labaki et al, every 1% increase in percent low attenuation area (%LAA) was independently associated with higher hazards of lung cancer incidence (hazard ratio [HR], 1.02; 95% CI, 1.01-1.03; P=0.004), lung cancer mortality (HR, 1.02; 95% CI, 1.00-1.05; P=0.045), and all-cause mortality (HR, 1.01; 95% CI, 1.00-1.03; P=0.042). ^{[54, 55]}

Konietzke et al found that quantitative CT can detect progression of emphysema in severe COPD in a 3-month period, quantify slight parenchymal changes that are not detected by spirometry, and reveal inconsistent changes of individual lung lobes and airway generations. ^[56]

Jeyin et al found that dual-energy computed tomographic pulmonary angiography (DECTPA)-derived lobar iodine quantification can provide an accurate estimate of lobar perfusion in patients with severe emphysema. They found a statistically significant  linear relationship between lobar perfusion values using DECTPA and single-photon emission CT perfusion scintigraphy (SPECT-PS). ^[57]

Helical CT

Because of the great variability of the machines, with single helical scanners and several models of multisection CT scanners, no technique has been standardized, and the detection rate of emphysema varies with the technique. Even with thick sections, the detection and quantification of emphysema is better than it is with conventional radiography and pulmonary function tests. The great advantage of helical CT is that the whole chest can be scanned in a single acquisition of less than 20 seconds.

For the visual detection of emphysema, use of a high-definition algorithm (bone or lung settings) is helpful. However, for the automatic detection of emphysema by computer, the standard algorithm is probably best. High-definition filters affect the attenuation measured by the computer, deviating from the values from the real Hounsfield scale and generally increasing the attenuation to variable degrees depending on the air-lung-tissue proportion. This effect is even more important when the attenuation of the lungs is compared for high-definition processing with scanners from different suppliers.

To enhance the margins of adjacent structures with different attenuations, processing artificially changes the original attenuation of the interface planes between the adjacent high- and low-attenuating structures, as in the case of the lung parenchyma and the air content of the lungs. This phenomenon is more obvious in the lung and skin than in solid viscera. This is probably why thresholds for discriminating emphysema differ in the current literature. The authors' personal experience suggests that the threshold -950 HU, as Gevenois suggested, with the standard algorithm without edge enhancement is the most appropriate method. This method may be most consistent and reliable for measuring the lung attenuation by using different machines.

Volumetric quantification of emphysema is based on the Hounsfield scale by using CT pulmonary densitovolumetry (shown in the images below). Some studeis suggest that precocious detection with quantification and 3-dimensional (3D) demonstration of the extension and distribution of emphysema can be helpful in smoking cessation programs or in risk assessments for occupational exposures.

 

Hyperpolarized gases are contrast agents that, when inhaled, provide images of the lung airspaces with high temporal and spatial resolution. The availability of these gases has great potential in the study of diffuse lung disease, particularly emphysema. ^[58]

Preoperative dynamic MRI has been studied as an additional outcome predictor for patient selection in LVRS. ^[59]

Ley et al assessed emphysematous enlargement of distal airspaces and concomitant large- and small-airway disease by using diffusion-weighted helium MRI, high-resolution CT (HRCT), and lung function tests. Helium MRI and HRCT scanning showed better agreement than did HRCT scanning results and functional characterizations of emphysema in terms of hyperinflation and large- and small-airway disease, as assessed on lung function tests. ^[60]

Sergiacomi et al ^[61] used lung-perfusion 2-dimensional (2D) dynamic breath-hold technique in patients with severe emphysema and found a high sensitivity (86.7%) and good specificity (80.0%) in detecting perfusion defects. They observed low peak signal intensities in emphysematous regions and concluded that lung perfusion MRI is a potential alternative to nuclear medicine study in the evaluation of severe pulmonary emphysema.

Functional evaluation of the lungs can be carried out by using xenon-133 (¹³³Xe) lung ventilation scintigraphy before and after lung-volume–reduction surgery (LVRS) in patients with pulmonary emphysema.

Xenon-133 washout curves during lung scintigraphy exhibit a biphasic pattern, as follows ^{[62, 63]} : 

• The first component of the washout curve, m(r), corresponds to an initial rapid phase in washout that reflects emptying of the large airways.
• The second component, m(s), reflects a slower phase of washout that is attributed to gas elimination in the small airways.

Radionuclide ventilation scans enable a useful assessment of lung function before and after LVRS. Travaline et al ^[64] demonstrated that small-airway ventilation in lung regions that were surgically treated, as well as in those areas that were not surgically treated, in the same patient were associated with increased improvement in lung function after LVRS.

In a study by Bonney et al that evaluated the use of gallium-68 ventilation/perfusion PET-CT (⁶⁸Ga-VQ/PET-CT) to assess severe emphysema, measurement of lobar pulmonary function by ⁶⁸Ga-VQ/PET-CT was shown to provide physiologic information that was not evident on CT densitometry, such as ventilation and perfusion specifics and matched defects. ^[65]