DOI: 10.1148/rg.256055037
RadioGraphics 2005;25:1639-1652
© RSNA, 2005
Multi–Detector Row CT and Postprocessing Techniques in the Assessment of Diffuse Lung Disease1
Catherine Beigelman-Aubry, MD,
Catherine Hill, FRCR,
Aymeric Guibal, MD,
Julien Savatovsky, MD and
Philippe A. Grenier, MD
1 From the Department of Radiology, Pitié-Salpêtrière Hospital, 47/83 Boulevard de l’Hôpital, 75013 Paris, France. Recipient of a Cum Laude award for an education exhibit at the 2004 RSNA Annual Meeting. Received March 4, 2005; revision requested March 29 and received May 16; accepted May 16. All authors have no financial relationships to disclose.
Address correspondence to C.B.A. (e-mail: catherine.beigelman{at}psl.ap-hop-paris.fr).
 |
Abstract
|
|---|
Many acute and chronic lung diseases are characterized by diffuse infiltration of the lung parenchyma. High-resolution computed tomography (CT) has been widely accepted as the imaging standard of reference for the assessment of these diseases. However, only approximately 10% of the lung parenchyma is scanned with high-resolution CT, and characteristic foci of disease may be missed. With use of the established characteristic high-resolution CT patterns, multi–detector row chest CT has revolutionized the evaluation of diffuse lung disease. Multi–detector row CT generates isotropic volumetric high-resolution data, allowing contiguous three-dimensional (3D) visualization of the lung parenchyma, with the capacity to create high-quality two-dimensional (2D) and 3D reformatted images. Minimum intensity projection is the postprocessing technique of choice for the detection and characterization of most patterns of diffuse lung disease. Maximum intensity projection (MIP) allows the detection and characterization of micronodules; the recognition of enlarged pulmonary veins, which is extremely useful in the diagnosis of pulmonary edema and the assessment of mosaic perfusion; and differentiation between perilymphatic, miliary, and centrilobular distribution. MIP can also help differentiate between constrictive bronchiolitis and mixed emphysema. Two-dimensional reformatted images are now of equal importance with the 2D axial images in diagnosing specific diffuse lung diseases. In the future, 3D reformatted images may be used to help quantify these disorders.
© RSNA, 2005
|
Introduction
|
|---|
A wide spectrum of acute and chronic lung diseases are characterized by diffuse infiltration of the lung parenchyma. High-resolution computed tomography (CT) with 1-mm-thick sections obtained at 10-mm intervals has been widely accepted as the imaging standard of reference for assessing diffuse lung disease. In certain clinical settings, the presence of a typical pattern at high-resolution CT may be sufficient to make a presumptive diagnosis.
High-resolution CT of diffuse lung disease is based on the principle that representative areas of disease will be present on images acquired at selected levels; however, because only approximately 10% of the lung parenchyma is scanned, characteristic foci of disease may be missed. Multi–detector row CT generates isotropic volumetric high-resolution data and allows contiguous visualization of the lung parenchyma.
The various patterns of diffuse lung disease seen at high-resolution CT have all been described. This information is now being used for volumetric multi–detector row CT (1,2). Although the various high-resolution CT patterns are still valid, multi–detector row CT allows (a) visualization in three dimensions by providing an isotropic volumetric data set and (b) the ability to create two-dimensional (2D) and three-dimensional (3D) reformatted images of excellent quality (3,4) and significance.
In this article, we (a) review image acquisition and reconstruction techniques; (b) describe image interpretation and postprocessing tools, including 2D reformation (multiplanar volume-rendered [VR] averaging, minimum intensity projection [mIP], maximum intensity projection [MIP]) and 3D reformation (volume intensity projection [VIP], 3D VR); and (c) discuss and illustrate multi–detector row CT findings in diffuse lung disease (linear, nodular, and ground-glass patterns; decreased lung attenuation).
|
Image Acquisition and Reconstruction
|
|---|
Because the lung parenchyma has a unique natural contrast, a low radiation dose (100–120 kV, 80–160 mAs) is used for multi–detector row CT.
Images are acquired with a 40– or 64–detector row CT scanner during a single breath hold lasting about 4–10 seconds, making respiratory motion artifacts very unusual. A high-frequency algorithm, a 512 x 512 or 768 x 768 matrix, and a 325-mm field of view are used, with a rotation time of approximately 500 msec that allows a marked decrease in cardiac pulsation artifacts. With a detector size of 0.625 mm, images with a section thickness of approximately 1 mm are reconstructed at intervals of approximately 0.5 mm, thereby producing a voxel of almost cubic dimensions and allowing the creation of excellent 2D and 3D reformatted images.
|
Image Interpretation and Postprocessing Tools
|
|---|
The interpretation of multi–detector row CT scans of the chest goes far beyond the standard assessment of axial sections, since multiplanar reformatted (MPR) images are easily obtained in real time in all orientations (5). At the workstation, the radiologist moves through the volumetric data set using a mouse, much as an ultrasonographer uses a probe. The objective is to find the planes in which key features are displayed that can help confidently identify the main pattern of diffuse lung disease, any associated findings, and the distribution of lesions relative to the secondary pulmonary lobule anatomy.
Two-dimensional Reformation
A significant decrease in the number of sections to be analyzed is achieved by comparing longitudinal reformatted images with the axial images.
This technique may be applied to a wide variety of diffuse lung diseases (Fig 1), allowing rapid assessment of the regional distribution of nodules in both the craniocaudal and axial dimensions (Fig 2). Multiplanar VR corresponds to a thickness of several pixels, immediately generating a better signal-to-noise ratio, with the possibility of using various rendering tools, including multiplanar VR averaging, mIP, and MIP.
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Langerhans cell histiocytosis in a 51-year-old man with chronic dyspnea. (a) Axial CT scan demonstrates multiple irregular cysts in both upper lung lobes. (b) Coronal MPR image clearly shows that the cysts are located predominantly in the upper and middle lung zones. Note the normal size of the bronchi. MPR helped confirm the diagnosis in less time than it took to review the more numerous axial images.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Langerhans cell histiocytosis in a 51-year-old man with chronic dyspnea. (a) Axial CT scan demonstrates multiple irregular cysts in both upper lung lobes. (b) Coronal MPR image clearly shows that the cysts are located predominantly in the upper and middle lung zones. Note the normal size of the bronchi. MPR helped confirm the diagnosis in less time than it took to review the more numerous axial images.
|
|
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Nonspecific interstitial pneumonitis in a 64-year-old woman with systemic sclerosis. (a) Axial CT scan of the lung bases shows ground-glass attenuation and reticular lines (arrows). (b, c) Longitudinal coronal (b) and sagittal (c) reformatted images clearly depict lesions in a subpleural location (arrows), with the sagittal image demonstrating their posterobasal location. These findings allowed immediate recognition of the craniocaudal and axial distribution of diffuse lung disease.
|
|
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Nonspecific interstitial pneumonitis in a 64-year-old woman with systemic sclerosis. (a) Axial CT scan of the lung bases shows ground-glass attenuation and reticular lines (arrows). (b, c) Longitudinal coronal (b) and sagittal (c) reformatted images clearly depict lesions in a subpleural location (arrows), with the sagittal image demonstrating their posterobasal location. These findings allowed immediate recognition of the craniocaudal and axial distribution of diffuse lung disease.
|
|
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c. Nonspecific interstitial pneumonitis in a 64-year-old woman with systemic sclerosis. (a) Axial CT scan of the lung bases shows ground-glass attenuation and reticular lines (arrows). (b, c) Longitudinal coronal (b) and sagittal (c) reformatted images clearly depict lesions in a subpleural location (arrows), with the sagittal image demonstrating their posterobasal location. These findings allowed immediate recognition of the craniocaudal and axial distribution of diffuse lung disease.
|
|
Multiplanar VR Averaging.—
In multiplanar VR averaging, the mean attenuation value of the voxels on every view throughout the volume is projected onto a 2D image (Fig 3). This technique allows concomitant evaluation of the tracheobronchial tree in the setting of diffuse lung disease.
View larger version (50K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Diagram illustrates multiplanar VR averaging technique. The mean attenuation value of the voxels on each view throughout the volume is projected onto the 2D image. White cubes represent voxels with the highest attenuation value; black cubes represent voxels with the lowest attenuation value.
|
|
Minimum Intensity Projection.—
mIP consists of projecting the voxel with the lowest attenuation value on every view throughout the volume onto a 2D image (Fig 4) (6,7). This technique displays only 10% of the data set. The subtle difference in density between the endobronchial (pure) air and the lung parenchyma, corresponding to a difference in attenuation of 50–150 HU, permits visualization of the bronchi below the subsubsegmental level. mIP is the optimal tool for the detection, localization, and quantification of ground-glass (Fig 5) and linear attenuation patterns.
View larger version (52K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Diagram illustrates mIP technique. The lowest attenuation value of the voxels on each view throughout the volume is projected onto the 2D image. White cubes represent voxels with the highest attenuation value; black cubes represent voxels with the lowest attenuation value.
|
|
View larger version (162K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Added clinical value of mIP. (a) Multivisceral Kaposi sarcoma in a 33-year-old man with human immunodeficiency virus (HIV) infection. Coronal mIP reformatted image (4.9-mm-thick slab) demonstrates ground-glass attenuation due to pulmonary hemorrhage in the left upper lobe. Note the normal difference in attenuation between air and lung parenchyma in the right upper lobe. Excellent visualization of air bronchograms helps the endoscopist select the optimal site for bronchoalveolar lavage. (b) Nonspecific interstitial pneumonitis due to systemic sclerosis in a 59-year-old woman. Oblique mIP–multiplanar VR image (10-mm-thick slab) clearly depicts small reticular opacities with concomitant, subtle bronchiolectasis.
|
|
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Added clinical value of mIP. (a) Multivisceral Kaposi sarcoma in a 33-year-old man with human immunodeficiency virus (HIV) infection. Coronal mIP reformatted image (4.9-mm-thick slab) demonstrates ground-glass attenuation due to pulmonary hemorrhage in the left upper lobe. Note the normal difference in attenuation between air and lung parenchyma in the right upper lobe. Excellent visualization of air bronchograms helps the endoscopist select the optimal site for bronchoalveolar lavage. (b) Nonspecific interstitial pneumonitis due to systemic sclerosis in a 59-year-old woman. Oblique mIP–multiplanar VR image (10-mm-thick slab) clearly depicts small reticular opacities with concomitant, subtle bronchiolectasis.
|
|
Maximum Intensity Projection.—
MIP consists of projecting the voxel with the highest attenuation value on every view throughout the volume onto a 2D image (Fig 6) (6,8). Like mIP, this technique displays only 10% of the data set. The primary clinical application of MIP is to improve the detection of pulmonary nodules and assess their profusion. MIP also helps characterize the distribution of small nodules (Fig 7) (9). In addition, MIP sections of variable thickness are excellent for assessing the size and location of vessels, including the pulmonary arteries and veins. Pulmonary edema associated with enlarged pulmonary veins can easily be differentiated from other causes of diffuse ground-glass attenuation. MIP also facilitates the differentiation between ground-glass attenuation and mosaic perfusion in cases of a mosaic attenuation pattern.
View larger version (50K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Diagram illustrates MIP technique. The highest attenuation value of the voxels on each view throughout the volume is projected onto the 2D image. White cubes represent voxels with the highest attenuation value; black cubes represent voxels with the lowest attenuation value.
|
|
View larger version (184K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7. Sarcoidosis in a 51-year-old man. Coronal MIP image (4.4-mm-thick slab) demonstrates the typical distribution of small nodules in the upper and middle lung zones seen in sarcoidosis.
|
|
Three-dimensional Reformation
Volume Intensity Projection.—
VIP is a new method of 3D reformation. VIP assigns the highest attenuation values to the voxels that are closest to the viewer and lower values to those that are farther away, thereby providing perspective information (Fig 8).
View larger version (205K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Emphysema in a 52-year-old male smoker. Coronal VIP image (15.2-mm-thick slab) allows excellent evaluation of the rearrangement of the pulmonary vasculature, providing perspective information on the volumetric distribution of the remaining vessels.
|
|
Three-dimensional VR.—
Three-dimensional VR assigns a color and opacity to each attenuation threshold chosen. All data in the volume are integrated, which may sometimes produce an image that closely resembles the macropathologic view (Fig 9).
View larger version (177K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Waldenström macroglobulinemia in an 82-year-old woman. Coronal MPR (a) and VR (b) images demonstrate a "crazy-paving" pattern secondary to biopsy-proved hypersensitivity pneumonitis. Note that the VR image resembles the macropathologic view.
|
|
View larger version (196K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Waldenström macroglobulinemia in an 82-year-old woman. Coronal MPR (a) and VR (b) images demonstrate a "crazy-paving" pattern secondary to biopsy-proved hypersensitivity pneumonitis. Note that the VR image resembles the macropathologic view.
|
|
|
Multi–Detector Row CT Findings in Diffuse Lung Disease
|
|---|
Diffuse lung disease may have an acute or a chronic cause. Acute causes include pulmonary edema, diffuse alveolar damage (adult respiratory distress syndrome), diffuse pulmonary hemorrhage, hypersensitivity pneumonitis, acute interstitial pneumonia, and acute infectious bronchiolitis. Chronic causes include chronic infiltrative lung disease, emphysema, chronic obstructive pulmonary disease, and obliterative bronchiolitis.
Multi–detector row CT facilitates the detection of the various patterns of diffuse lung disease. With this modality, it is easier to recognize the predominant pattern of distribution, which is important for developing the differential diagnosis. Abnormalities are also more easily identified in relation to the underlying vascular, bronchial, and lobular anatomy (Fig 10). Actually, the assessment of the perilobular, centrilobular, or pan-lobular distribution of findings is crucial to making a correct diagnosis.
View larger version (65K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10. Diagram illustrates the three main compartments of the pulmonary interstitium at the level of the secondary pulmonary lobule. The axial compartment surrounds the lobular artery and bronchiole, whereas the peripheral compartment is located at the level of the interlobular septum. These compartments are connected to each other by the intralobular interstitium.
|
|
Linear Pattern
A linear pattern may correspond to (a) thickening of the interstitial fiber network of the lung owing to fluid, (b) fibrous tissue, or (c) interstitial infiltration by cells or other material. Longitudinal reformation provides additional information of great value in this setting (Fig 11). These patterns are best imaged with mIP, whatever their cause (Fig 12). Abnormal septal lines, which are easily assessed with this technique, may be related to a variety of disorders such as rejection, pulmonary edema, fibrosis, or lymphangitis carcinomatosis. In cases of perilymphatic distribution, consisting of peribronchovascular thickening and abnormal septal lines, MIP is the reformatting technique of choice for evaluating the size of pulmonary vessels. When enlarged pulmonary veins are responsible for the nodular appearance of the thickened septa, a definite diagnosis of pulmonary edema may be made (Fig 13).
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. Pulmonary fibrosis caused by dermatomyositis in a 91-year-old woman. (a) Axial CT scan shows irregular thickened septal lines in the left lower lobe and focal subpleural honeycombing in the right lower lobe. (b) Coronal MPR image depicts diffuse distorted septal lines (arrows), which are characteristic of pulmonary fibrosis.
|
|
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. Pulmonary fibrosis caused by dermatomyositis in a 91-year-old woman. (a) Axial CT scan shows irregular thickened septal lines in the left lower lobe and focal subpleural honeycombing in the right lower lobe. (b) Coronal MPR image depicts diffuse distorted septal lines (arrows), which are characteristic of pulmonary fibrosis.
|
|
View larger version (173K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12. Lymphangitis carcinomatosis in a 74-year-old man. Oblique mIP–multiplanar VR image (3.2-mm-thick slab) clearly delineates lesions with a perilymphatic distribution. Peribronchovascular thickening (arrows) and abnormal thickened nodular septal lines (arrowheads) are also well demonstrated.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. Pulmonary edema in a 64-year-old man who presented with breathlessness. (a) Axial CT scan (3-mm-thick slab) demonstrates thickened and nodular septal lines (arrow). (b) Sagittal MIP image (10-mm-thick slab) reveals enlarged pulmonary veins (arrows), thereby helping make the diagnosis of pulmonary edema.
|
|
View larger version (208K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. Pulmonary edema in a 64-year-old man who presented with breathlessness. (a) Axial CT scan (3-mm-thick slab) demonstrates thickened and nodular septal lines (arrow). (b) Sagittal MIP image (10-mm-thick slab) reveals enlarged pulmonary veins (arrows), thereby helping make the diagnosis of pulmonary edema.
|
|
Reticular patterns, which usually reflect intralobular interstitial thickening, are most commonly seen in patients with pulmonary fibrosis. The great advantage of mIP in this setting is its capacity to help correctly evaluate the distribution of reticular opacities, while at the same time helping assess for possible traction bronchiolectasis or discrete "honeycombing," both of which entities are difficult to appreciate on axial images alone (Fig 14).
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14a. Lung fibrosis of unknown origin in a 71-year-old man. (a) Sagittal reformatted image shows small reticular opacities in the subpleural area of the left lower lobe (arrow) and focal honeycombing in the anterior subpleural region of the upper lobe (arrowheads). (b) Oblique mIP–multiplanar VR image (7.2-mm-thick slab) demonstrates apparent traction bronchiolectasis in the area of reticular opacities, a finding that was not evident on the sagittal image.
|
|
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14b. Lung fibrosis of unknown origin in a 71-year-old man. (a) Sagittal reformatted image shows small reticular opacities in the subpleural area of the left lower lobe (arrow) and focal honeycombing in the anterior subpleural region of the upper lobe (arrowheads). (b) Oblique mIP–multiplanar VR image (7.2-mm-thick slab) demonstrates apparent traction bronchiolectasis in the area of reticular opacities, a finding that was not evident on the sagittal image.
|
|
Nodular Pattern
A nodular pattern is often difficult to evaluate solely on the basis of its attenuation and definition. More useful is the assessment of its regional distribution, both in the craniocaudal and axial dimensions and within the secondary pulmonary lobule. MIP with progressively increasing slab thickness is very helpful because it improves the detection of small nodules and the estimation of their profusion, thereby helping in the recognition of their characteristic distribution relative to the landmarks of the secondary pulmonary lobule (Figs 15–17).
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15a. Sarcoidosis in a 41-year-old man. (a) Sagittal reformatted image demonstrates micronodules distributed predominantly in the upper and middle portions of the lungs. (b) MIP image (4.8-mm-thick slab) more clearly depicts the characteristic perilymphatic distribution of micronodules along the septa (arrows) and the major fissure (arrowhead). Note also the retraction of the major fissure, a finding that represents fibrosis.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15b. Sarcoidosis in a 41-year-old man. (a) Sagittal reformatted image demonstrates micronodules distributed predominantly in the upper and middle portions of the lungs. (b) MIP image (4.8-mm-thick slab) more clearly depicts the characteristic perilymphatic distribution of micronodules along the septa (arrows) and the major fissure (arrowhead). Note also the retraction of the major fissure, a finding that represents fibrosis.
|
|
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16a. Miliary tuberculosis in a 45-year-old HIV-positive woman. (a) Axial CT scan demonstrates a few micronodules. (b) MIP image (6-mm-thick slab) shows a random distribution of nodules, which are much more clearly depicted than on the axial image.
|
|
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16b. Miliary tuberculosis in a 45-year-old HIV-positive woman. (a) Axial CT scan demonstrates a few micronodules. (b) MIP image (6-mm-thick slab) shows a random distribution of nodules, which are much more clearly depicted than on the axial image.
|
|
View larger version (81K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17. Tuberculosis in a 56-year-old woman who was being evaluated for chronic fever. MIP–multiplanar VR images with increasing slab thickness (left to right) clearly depict the characteristic centrilobular location of micronodules seen in endobronchial spread of tuberculosis. The use of progressively increasing slab thickness improves the detection and assessment of the profusion of micronodules and allows progressive assessment of the margins of the secondary pulmonary lobule, which are recognized from pulmonary veins lying along the septa (arrows). Note that the fissure is free of nodules. Bronchoscopy helped confirm the diagnosis of tuberculosis.
|
|
Ground-Glass Pattern
The ground-glass pattern is defined as an abnormal increased attenuation of the lung parenchyma without obscuration of vessels or bronchi. It may correspond to partial or, rarely, total alveolar filling, thickening of alveolar septa with partial collapse of alveoli (with or without lung fibrosis), or an increase in blood capillary size. Ground-glass attenuation can be focal or diffuse and homogeneous or heterogeneous and can develop in an acute or chronic context. It is sometimes difficult to assess on axial images alone but may be more easily demonstrated with mIP as an abnormal contrast between endobronchial air and hyperattenuating lung parenchyma (Fig 18). Moreover, it can be used to guide the endoscopist in selecting the best area in which to perform bronchoalveolar lavage (Fig 5a). MIP must also be performed in this setting to differentiate ground-glass attenuation with a mosaic pattern from mosaic perfusion. In cases of ground-glass attenuation with a mosaic pattern, the vessels will be of equal size throughout the lungs (Fig 19). MIP should also be systematically performed in cases of perihilar or dependent ground-glass attenuation. In these cases, hydrostatic pulmonary edema may readily be distinguished from other causes of ground-glass attenuation on the basis of enlarged pulmonary veins (Fig 20).
View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 18a. Drug-induced lung disease in a 59-year-old woman who was undergoing chemotherapy for a retroperitoneal malignancy. (a) Axial CT scan shows inhomogeneous lung attenuation. (b) mIP image (5.2-mm-thick slab) shows diffuse abnormal increased attenuation of the lung parenchyma compared with endobronchial air. This finding is characteristic of ground-glass attenuation, which in this case is secondary to drug-induced lung disease.
|
|
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 18b. Drug-induced lung disease in a 59-year-old woman who was undergoing chemotherapy for a retroperitoneal malignancy. (a) Axial CT scan shows inhomogeneous lung attenuation. (b) mIP image (5.2-mm-thick slab) shows diffuse abnormal increased attenuation of the lung parenchyma compared with endobronchial air. This finding is characteristic of ground-glass attenuation, which in this case is secondary to drug-induced lung disease.
|
|
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 19a. Pneumocystis carinii pneumonia in a 31-year-old HIV-positive man. (a) Axial CT scan shows ground-glass attenuation in the left upper lobe. (b) Sagittal MIP image (4.9-mm-thick slab) more clearly demonstrates the size of vessels in both the normal and abnormal areas. In this case, the similarity of vessel size in the two areas allowed a definite diagnosis of mosaic attenuation and the exclusion of mosaic perfusion. P carinii pneumonia was diagnosed by means of bronchoalveolar lavage directed to the left upper lobe.
|
|
View larger version (177K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 19b. Pneumocystis carinii pneumonia in a 31-year-old HIV-positive man. (a) Axial CT scan shows ground-glass attenuation in the left upper lobe. (b) Sagittal MIP image (4.9-mm-thick slab) more clearly demonstrates the size of vessels in both the normal and abnormal areas. In this case, the similarity of vessel size in the two areas allowed a definite diagnosis of mosaic attenuation and the exclusion of mosaic perfusion. P carinii pneumonia was diagnosed by means of bronchoalveolar lavage directed to the left upper lobe.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 20a. Hydrostatic pulmonary edema in a 73-year-old man who presented with acute breathlessness. (a) Axial CT scan demonstrates bilateral patchy areas of ground-glass attenuation associated with enlarged pulmonary vessels and right-sided pleural effusion. (b) Coronal mIP image (4.9-mm-thick slab) provides a simplified global display of the medullary distribution of ground-glass attenuation. (c) Coronal MIP image (12-mm-thick slab) clearly depicts enlarged pulmonary veins. All of these findings, in association with a pleural effusion, are characteristic of hydrostatic pulmonary edema.
|
|
View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 20b. Hydrostatic pulmonary edema in a 73-year-old man who presented with acute breathlessness. (a) Axial CT scan demonstrates bilateral patchy areas of ground-glass attenuation associated with enlarged pulmonary vessels and right-sided pleural effusion. (b) Coronal mIP image (4.9-mm-thick slab) provides a simplified global display of the medullary distribution of ground-glass attenuation. (c) Coronal MIP image (12-mm-thick slab) clearly depicts enlarged pulmonary veins. All of these findings, in association with a pleural effusion, are characteristic of hydrostatic pulmonary edema.
|
|
View larger version (102K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 20c. Hydrostatic pulmonary edema in a 73-year-old man who presented with acute breathlessness. (a) Axial CT scan demonstrates bilateral patchy areas of ground-glass attenuation associated with enlarged pulmonary vessels and right-sided pleural effusion. (b) Coronal mIP image (4.9-mm-thick slab) provides a simplified global display of the medullary distribution of ground-glass attenuation. (c) Coronal MIP image (12-mm-thick slab) clearly depicts enlarged pulmonary veins. All of these findings, in association with a pleural effusion, are characteristic of hydrostatic pulmonary edema.
|
|
The crazy-paving pattern consists of a combination of ground-glass attenuation and reticular lines in the same area (Fig 9) and often reflects chronic filling of the alveoli from various causes. This pattern was originally described in pulmonary alveolar proteinosis. However, it may occur in many other diseases, including hypersensitivity pneumonitis, bronchoalveolar cell carcinoma, lipoid pneumonia, P carinii pneumonia, chronic eosinophilic pneumonia, pulmonary edema, and pulmonary hemorrhage. The crazy-paving pattern is a clinical indication for bronchoalveolar lavage, which provides the key to identifying its cause.
Decreased Lung Attenuation
Decreased lung attenuation is a characteristic finding in lung cysts (Fig 1), honeycombing, emphysema, and mosaic perfusion. A pattern of multiple cysts distributed throughout the lungs is suggestive of Langerhans cell histiocytosis or lymphangioleiomyomatosis. In patients with Langerhans cell histiocytosis, nodules are often associated with cysts, and both involve the upper two-thirds of the lungs, sparing the costophrenic angles. In patients with lymphangioleiomyomatosis, the lung cysts are thin walled, numerous, of varying size, and often large, involving the lung diffusely without any topographic predominance. The cysts are surrounded by relatively normal lung parenchyma.
mIP in the optimal oblique plane—in most cases along the long axis of the bronchus—permits differentiation of cystic bronchiectasis from lung cysts (Fig 21). mIP is also the best technique for depicting centrilobular emphysema. However, multiplanar VR averaging is the best postprocessing technique for depicting a central dot, which represents the remaining centrilobular artery, within a round hypoattenuating area, findings that are characteristic of centrilobular emphysema (Fig 22). Such a central dot is not seen in lung cysts.
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 21a. Emphysema in a 61-year-old man with exertional dyspnea. (a) Axial CT scan shows hypoattenuating areas and cystic lesions (arrows), findings that are best evaluated further with mIP in the optimal oblique plane. (b) On an oblique mIP image (5.2-mm-thick slab), the bronchiectatic nature of the cystic lesions in the left upper lobe becomes obvious (arrow). Note the presence of additional cystic lesions, which are related to emphysema.
|
|
View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 21b. Emphysema in a 61-year-old man with exertional dyspnea. (a) Axial CT scan shows hypoattenuating areas and cystic lesions (arrows), findings that are best evaluated further with mIP in the optimal oblique plane. (b) On an oblique mIP image (5.2-mm-thick slab), the bronchiectatic nature of the cystic lesions in the left upper lobe becomes obvious (arrow). Note the presence of additional cystic lesions, which are related to emphysema.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 22. Centrilobular emphysema in a 46-year-old male smoker. Multiplanar VR image (3.2-mm-thick slab) shows a centrilobular artery appearing as a central dot (arrow), a finding that is characteristic of centrilobular emphysema and helps differentiate it from cysts. Centrilobular emphysema is most clearly depicted with multiplanar VR averaging.
|
|
The term honeycombing refers to the characteristic appearance of extensive end-stage pulmonary fibrosis, resulting from lung destruction and obliteration of the acinar architecture. At multi–detector row CT, the cystic spaces in honeycombing commonly share walls, are predominantly sub-pleural, and occur in several layers. Honeycombing is most commonly caused by idiopathic pulmonary fibrosis, collagen vascular disease, end-stage hypersensitivity pneumonitis, or asbestosis. It may have an atypical distribution, particularly in sarcoidosis, drug-related fibrosis, and hypersensitivity pneumonitis. In such cases, a focal area of honeycombing may mimic paraseptal or cicatrizing emphysema. Multiplanar VR–mIP is helpful in showing the airway with distal bronchiolectasis converging into a honeycombing pattern (Fig 23). Such findings are not present in cases of emphysema.
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 23. Idiopathic pulmonary fibrosis in a 45-year-old woman. Oblique multiplanar VR–mIP image (5.2-mm-thick slab) clearly demonstrates bronchiolectasis within areas of cysts (arrow). The juxtaposed areas of cysts with intervening walls (arrowheads) are characteristic of honeycombing.
|
|
MIP helps evaluate the vascular architecture in areas of decreased lung attenuation. This capacity is particularly valuable in differentiating emphysema from constrictive bronchiolitis. The latter is characterized by a poor but preserved vascular architecture. Conversely, a disorganized architecture is demonstrated in most cases of emphysema (Fig 24). Sometimes it is impossible to differentiate between the two disease processes; in such cases, texture-based image analysis may be attempted (10).
View larger version (171K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 24a. (a) Emphysema in a 52-year-old man. Coronal MIP image (6.8-mm-thick slab) shows a heterogeneous distribution of the remaining vessels within hypoattenuating areas. (b) Constrictive (obliterative) bronchiolitis related to vasoconstriction or remodeling secondary to hypoxia. Axial MIP image (5.6-mm-thick slab) shows harmonious reduction of vessels, a finding that is quite unlike that seen in a. Note the superimposed centrilobular nodules in the apical segment of the right lower lobe (arrow), which are related to acute infectious bronchiolitis.
|
|
View larger version (105K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 24b. (a) Emphysema in a 52-year-old man. Coronal MIP image (6.8-mm-thick slab) shows a heterogeneous distribution of the remaining vessels within hypoattenuating areas. (b) Constrictive (obliterative) bronchiolitis related to vasoconstriction or remodeling secondary to hypoxia. Axial MIP image (5.6-mm-thick slab) shows harmonious reduction of vessels, a finding that is quite unlike that seen in a. Note the superimposed centrilobular nodules in the apical segment of the right lower lobe (arrow), which are related to acute infectious bronchiolitis.
|
|
Finally, MIP is the tool of choice in facilitating a definite diagnosis of mosaic perfusion. This pattern is characterized by hypoattenuating areas with small vessels that correspond to pathologic change and by hyperattenuating areas with enlarged vessels that correspond to normal lung with redistribution of perfusion. These findings may be related to small airway disease or a primary vascular abnormality, mainly due to chronic thromboembolic pulmonary disease.
|
Conclusions
|
|---|
Multi–detector row CT of the chest has revolutionized the evaluation of diffuse lung disease. Two-dimensional reformatted images are now of equal importance with the 2D axial images in making the final diagnosis. mIP is the tool of choice for the detection, assessment of distribution, evaluation of extent, and characterization of linear or ground-glass attenuation and mosaic perfusion. MIP allows the recognition of pulmonary edema in cases of linear attenuation associated with enlarged pulmonary veins. Moreover, mosaic perfusion may be differentiated from mosaic attenuation on the basis of vessel size. MIP permits the detection and characterization of micronodules and differentiation between perilymphatic, miliary, and centrilobular distribution. MIP can also help differentiate between constrictive bronchiolitis and mixed emphysema. In the future, 3D reformatted images could be used to quantify these diffuse disorders.
|
Footnotes
|
|---|
Abbreviations: HIV = human immunodeficiency virus, MIP = maximum intensity projection, mIP = minimum intensity projection, MPR = multi-planar reformation, 3D = three-dimensional, 2D = two-dimensional, VIP = volume intensity projection, VR = volume rendering
|
References
|
|---|
- Hunsaker AR. Multidetectorrow CT and interstitial lung disease. Semin Roentgenol
2003;38:176–185.[CrossRef][Medline]
- Honda O, Johkoh T, Tomiyama N, et al. High-resolution CT using multidetector CT equipment: evaluation of image quality in 11 cadaveric lungs and a phantom. AJR Am J Roentgenol
2001;177: 875–879.[Abstract/Free Full Text]
- Ravenel JG, McAdams HP, Remy-Jardin M, Remy J. Multidimensional imaging of the thorax. J Thorac Imaging
2001;16:269–281.[CrossRef][Medline]
- Kauczor HU. MDCT in diffuse lung disease. In: Schoepf UJ, ed. Multidetector-row CT of the thorax. Berlin, Germany: Springer, 2004; 81–105.
- Arakawa H, Sasaka K, Lu WM, Hirayanagi N, Nakajima Y. Comparison of axial high-resolution CT and thin-section multiplanar reformation (MPR) for diagnosis of diseases of the pulmonary parenchyma. J Thorac Imaging
2004;19:24–31.[CrossRef][Medline]
- Bhalla M, Naidich DP, McGuinness G, Gruden JF, Leitman BS, McCauley DI. Diffuse lung disease: assessment with helical CT—preliminary observations of the role of maximum and minimum intensity projection images. Radiology
1996; 200:341–347.[Abstract/Free Full Text]
- Remy-Jardin M, Remy J, Gosselin B, Copin MC, Wurtz A, Duhamel A. Sliding thin slab, minimum intensity projection technique in the diagnosis of emphysema: histopathologic-CT correlation. Radiology
1996;200:665–671.[Abstract/Free Full Text]
- Remy-Jardin M, Remy J, Artaud D, Deschildre F, Duhamel A. Diffuse infiltrative lung disease: clinical value of sliding–thin-slab maximum intensity projection CT scans in the detection of mild micronodular patterns. Radiology
1996;200:333–339.[Abstract/Free Full Text]
- Gruden JF, Ouanounou S, Tigges S, Norris SD, Klausner TS. Incremental benefit of maximum-intensity-projection images on observer detection of small pulmonary nodules revealed by multidetector CT. AJR Am J Roentgenol
2002;179:149–157.[Abstract/Free Full Text]
- Chabat F, Yang GZ, Hansell DM. Obstructive lung diseases: texture classification for differentiation at CT. Radiology
2003;228:871–877.[Abstract/Free Full Text]
Top
Abstract
Introduction
