CAT SCAN FILE
| CAT SCANS 1 TO 20
|
|
Pitch is defined as the ratio between table speed and collimator width (Pitch= Table speed [mm/sec]/Collimator width [mm]). Increasing pitch allows a greater volume of tissue to be scanned per unit time. Additionally, a pitch greater than one enables narrower collimation which ultimately results in improved resolution. Using 10 mm collimation and a 10 second breath-hold, a table incrementation of 10 mm/sec results in the scan covering a 10 cm volume (pitch = 1), whereas a table incrementation of 20 mm/sec (pitch = 2) would scan twice the distance. Because it can survey the entire lung volume continuously during a single breath hold, helical scanning enhances the detection of lung nodules (through the use of retrospective reconstruction of overlapping sections) and reduces motion artifacts. The slice, or reconstruction interval, is the distance between the midpoints of adjacent slices. The interval can be post-processed to be either more than, equal to, or less than the collimation. If the slice interval is more than the collimation, the slices will have gaps in between. If the interval is less than the collimation, the slices will be overlapping- this latter capability of obtaining overlapping slices without increasing the radiation dose to the patient is one of the major benefits of helical CT and overlapping section reconstruction decreases partial volume artifacts. A limitation of helical CT is that of "stair-step artifact" which affects structures running obliquely through the plane of section. This artifact also affects the lung fissures which appear blurred.
Another benefit of helical CT is the ability to obtain more uniformly consistent opacification of vessels with smaller volumes of contrast, primarily because of the shorter imaging times. Using 60 ml of 60% contrast material injected via a power injector at 2 to 3 ml/sec typically produces high quality vascular opacification typically between 20 to 30 seconds following initiation of the injection. There is patient variability, however, and peak enhancement may not occur until 50 to 60 seconds in some patients.
A recommendation for routine helical chest CT is to perform 7 mm slices with a pitch of 1 from the level of C5 to the carina, then 5 mm sections with a pitch of 1 (or 3 mm sections with a pitch of 2) from the carina to the level of the inferior pulmonary veins, followed by 7 mm sections with a pitch of 1 through the remainder of the lungs.
In the evaluation of hemoptysis an initial sequence of 7 mm scans with a pitch of 1 obtained from the apex to the carina, followed by 3-5 mm sections with a pitch of 1 to 1.6 from the carina through the inferior pulmonary veins, and then 1-2 mm sections every 10 mm to the bases is an adequate approach.. This sequence is based on the fact that most cases of hemoptysis (in patients with non-localizing CXR's) are due to peripheral bronchiectasis or a central endobronchial lesion. Only about 5% of patients that present with hemoptysis and a negative CXR are found to have lung cancer, and less than 1% of patients are found to develop cancer on 2 year follow-up.
No definitive study has been performed to determine the optimal volumetric technique for scanning the airways. Routine 7-8 mm sections acquired with a pitch of 1.6 to 2.0 through the entire lungs with the use of I.V. contrast will suffice for most cases. For more detailed analysis, 5 mm sections should be obtained.
Overlapping reconstruction with at least 50-60% overlap (4 or 5 mm sections reconstructed at a minimum of 2 to 3 mm intervals) will minimize aliasing and stairstep artifacts on 3D and MPR views.
Maximum Intensity Projection Reconstructions:
The MIP technique (maximum intensity projection) depicts only the picture units (voxels) between a pre-selected threshold along the observers line of sight. If the threshold of brightness is selected for IV contrast material, all other superimposed structures that are above (such as bone) or below (such as soft tissue and lung) the threshold values are eliminated and only a two dimensional angiographic image will remain. With this technique, there is improved blood vessel conspicuity along greater portions of the vessel length. It is the only view that allows differentiation of calcification from contrast material. This technique can also aid the detection and differentiation of micronodules from small vessels.
Minimum Intensity Projection Reconstructions:
The minimum intensity algorithm enables pixels to encode the minimum encountered voxel value during reconstruction along each ray. Such images can aid in the detection of areas of emphysema, however, the images are more susceptible to motion and pulsation artifacts. Motion and pulsation will create areas of hypoattenuation that can mimic emphysematous change. Cardiac pulsation artifact is usually recognized by its paramediastinal / paracardiac location. Breathing artifacts cannot be accurately recognized unless standard CT images are also available for review.
Ray-Sum Projection Reconstructions:
The ray-sum projection displays the sum of pixel values along vectors or rays projected through the model and it is the most analogous in appearance to conventional arteriograms. It has a relative translucent appearance that allows superimposed vessels to be seen through one another. It is important to remember that thresholding and disarticulation used for reconstruction may introduce artifacts such as pseudostenoses. The lower threshold should not be set above approximately 50-60% of the peak vascular enhancement- raising the threshold further may erode the vessels and create false areas of stenosis.
HRCT is most often performed in the evaluation of patients with suspected diffuse lung disease. Up to 10% of patients with interstitial lung disease may have a normal CXR. Between 20 to 30% of these individuals will be shown to have interstitial disease by HRCT. Additionally, HRCT has been shown to be of great value in directing biopsy to regions with the appearance of most active disease. It is important to remember that HRCT provides macroscopic, not microscopic information. Findings on HRCT do however often reflect histologic appearances and distribution of disease.
High resolution CT of the lungs utilizes thin collimation (1-1.5 mm), image reconstruction with a high spatial frequency (bone algorithm), and increased kVp or mA technique (kVp 120-140, mA 140-240, and mAs 240-400). The high spatial frequency reconstruction reduces image smoothing and increases spatial resolution which allows structures to appear sharper. Although this technique does make image noise more apparent, much of the noise is quantum related and will decrease with increased technique (kVp, mA). Low dose HRCT using 120 kVp and 20-60 mA has also been performed and provides diagnostic information comparable to the conventional high resolution technique. Still, for large patients, patients with suspected posterior lung disease, and for initial patient evaluation, the use of conventional technique is recommended. The largest matrix available should be used for image reconstruction (typically 512 x 512). Retrospectively targeting reconstruction to a single lung instead of the entire thorax by using a smaller field of view, will significantly reduce the image pixel size and increase spatial resolution. Current scanners are capable of providing a spatial resolution of 0.5 mm (pixel size of 0.25 mm).
Lung attenuation is greater in dependent lung than in non-dependent lung (typically between 50-100 HU) . Because patients that are positioned supine within the scanner may demonstrate dependent densities (sub-pleural lines) that can mimic interstitial lung disease, a number of protocols have been described to permit adequate evaluation of the lungs. One option is to scan the lungs in full inspiration in the supine position, and then obtain 3 selected slices (arch, carina, and 2 cm above the right hemidiaphragm) with the patient prone. Alternatively, supine and prone scanning each at 2 cm spacing could be performed. Expiratory scans at 3 selected levels are usually sufficient to demonstrate significant air trapping.
Images should be photographed using "lung windows" which are typically a level of -600 to -700 and a window of 1000 to 1500. For evaluating the hila, mediastinum, or pleura, a level of 50 and window of 350 is best. A low level setting (-800 to -900) and a narrow width (500) can aid in the detection of emphysema. These recommendations should serve as guidelines- individuals should select settings which they find to be the most useful for their equipment. A 6 on 1 format using 14 x 17 film is used to make the images large enough to view easily.
HRCT has been shown to be of value in confirming the presence of air trapping. In the majority of normal subjects, lung attenuation increases in a homogeneous fashion during exhalation with an average increase of about 200 HU. In the presence of air-trapping lung parenchyma remains lucent on expiration and shows little change in volume. Trapping is considered to be present when the lung fails to increase normally in attenuation during exhalation. An increase in attenuation of less than 100 HU is indicative of air-trapping. Air trapping is associated with bronchiolitis obliterans, asthma, chronic bronchitis, sarcoidosis, and hypersensitivity pneumonitis . Normal subjects (between 52-84% ) can show focal areas of relative lucency (typically fewer than 3 adjacent secondary pulmonary lobules ) on expiratory HRCT images that are indicative of air trapping- most commonly localized to the lingula or superior segments of the lower lobes. The frequency of air trapping increases with age and is also higher in patients with a greater than 10 pack year smoking history . The extent of air trapping is usually small (under 5% of the total lung area) in asymptomatic patients . Larger areas of air trapping (segmental and lobar) are more commonly found in smokers or ex-smokers. An increased incidence of focal air trapping on expiratory CT has also been observed in HIV-positive individuals. The severity of focal air trapping is related to duration of HIV infection, and inversely related to CD4 count, suggesting that airway damage is progressive in these individuals.
Thickening of the intralobular interstitium produces a fine reticular or mesh-like pattern in the sub-pleural lung periphery. It is also referred to as the small reticular pattern. Intralobular bronchioles are often visible in patients with this type of fibrosis because of traction bronchiolectasis.
Ground glass opacity is defined as a region of increased lung attenuation not obscuring the underlying vessels. Ground glass opacification is a very non-specific finding. It is produced by any infiltrative process within the interstitium of the lung or partial filling of the air spaces by fluid, cells, or fibrosis so that the CT attenuation of the affected lung increases compared with that of the normal lung parenchyma. The caliber and number of vessels are not appreciably different between the normal and abnormal regions of lung. Although it is non-specific, ground-glass opacity usually (80%) indicates an acute, active, and potentially treatable process.
Ground glass mimicks: Ground-glass opacities (GGO) may occur artifactually. Certain technical parameters may also produce GGO. Narrow window width and level can also result in GGO- a recommended range for viewing the lung parenchyma is a window width of 1500-2000 and a level of -500 to -700. A tube current that is too low will produce excessive noise which can mimic GGO- generally a current of 240-400 mA is acceptable. Collimation should be 1.0-1.5 mm because GGO may occur when areas of parenchymal opacification are volume averaged with the thicker slices, or it may not be visualized at all, also due to volume averaging [10]. Low lung volumes may produce areas of ground glass attenuation and all HRCT exams should be performed at deep inspiration to avoid this problem. The tracheal configuration changes from round in inspiration to flat or cresant-shaped in expiration and it can be used to determine whether the exam was poerformed in inspiration or expiration. Bronchoalveolar lavage has also been shown to produce areas of ground glass attenuation for up to 24 hours following the procedure. Cardiac and respiratory motion can create GGO, but blurring and double images of vessels and fissures will help to confirm the presence of motion.
CT performed at suspended full expiration and sometimes on inspiratory scans can show the physiologic consequence of small airway (bronchiolar) disease- air trapping. Lung regions that retain air during exhalation remain more lucent and show less decrease in volume than lung supplied by normal airways. Adjacent normal areas of lung may appear to have increased attenuation (GGO) in this situation. Frequently, these patients will have associated findings of airways disease such as bronchiectasis in the hyperlucent regions. Hypoxic vasocontriction will result in decreased vascular size and markings within the hyperlucent regions.
In pulmonary thromboembolic disease regions of hyperemic lung (higher attenuation) mimic ground glass infiltrates when seen adjacent to oligemic (lower attenuation) regions of lung. The oligemic lung, however, will show a decrease in the caliber and number of pulmonary vessels compared to normal lung. (See also "mosaic pattern of lung attenuation" below)
Findings which aid in the identification of true areas of ground glass attenuation include prominent septal lines within the region of abnormality, a poorly marginate/non-anatomic distribution, and normal vessel caliber within the involved region. Occasionally, a mixed pattern of both ground glass opacification and air-trapping can be seen. Etiologies which can produce this combination of findings include infection (mycoplasma), hypersensitivity penumonitis, and sarcoid.
Honeycombing is defined by the presence of small airway cystic spaces, generally lined by bronchiolar epithelium, and having thick walls composed of fibrous tissue. Honeycombing indicates the presence of end-stage pulmonary fibrosis. The cystic spaces average about 1 cm in size (range from 3 mm to several centimeters) and have clearly defined walls of 1 to 3 mm in thickness. Honeycombing typically predominates in the peripheral and sub-pleural lung. Honeycombing is often associated with other findings of lung fibrosis such as intralobular interstitial thickening and traction bronchiectasis. Superimposition of emphysema and dependent density may occasionally mimic honeycombing when the patient is supine. This finding should clear when the patient is scanned in the prone position.
A mosaic pattern is produced by the patchy distribution of areas of increased and decreased lung attenuation. A moasic attenuation pattern may reflect the presence of underlying small airway disease, pulmonary vascular disease, or ground-glass interstitial or air space infiltrates. The mosaic pattern observed in small airway disease is related to both air trapping and hypoxic vasoconstriction (which can result in small or fewer visible pulmonary vessels). Expiratory images can be used to confirm the presence of air trapping. Up to 20% of patients with evidence of air trapping on expiratory images will have normal inspiratory exams. Airway abnormalities (bronchial wall thickening or dilatation) are a common finding in these cases.
The mosaic pattern produced from underlying pulmonary vascular disease (chronic thromboembolic disease or pulmonary arterial hypertension) results from regions of hyperemic lung (higher attenuation) seen adjacent to oligemic regions (low attenuation). The oligemic lung will demonstrate a decrease in the caliber and number of vessels in comparison to normal lung, but will demonstrate increased attenuation on expiratory images (ie: show no evidence of air trapping). Also, signs of pulmonary arterial hypertension (enlarged arteries) can aid in the diagnosis.
Finally, areas of ground glass attenuation due to either air-space or interstitial infiltrates when viewed next to normal lung may also produce a mosaic pattern. In this situation, however, the caliber and number of vessels is similar to that in normal lung.
In some patients inspiratory scans can show a mixture of several types of abnormality- ground-glass opacity, consolidation, normal lung, and mosaic attenuation. This combination has been termed the "head cheese" sign because of its resemblence to a particular sausage . The mixed pattern is usually indicative of a combination of infiltrative and obstructive lung disease (best appreciated on expiratory images). The most common causes of this pattern are hypersensitivity pneumonitis, sarcoidosis, and infections associated with bronchiolitis.
This term is used to describe non-tapering linear densities within the lungs from 2 to 5 cm in length that are often located in the lung periphery and frequently contact the pleural surface. The bands may represent continuous thickened interlobular septa, but those that are several millimeters thick, may represent coarse scars or atelectasis.
The peribronchovascular interstitium is the connective tissue that surrounds the bronchi and pulmonary vessels and it is present from the hilar regions to the level of the terminal or respiratory bronchiole (although it thins as it travels peripherally in the lung). Within the peribronchovascular interstitium is an abundant network of bronchial arteries, veins, and lymphatics. The interstitium usually appears as a thin band of low density attenuation adjacent to the major bronchi and vessels. It typically has a concave or straight margin and it is typically not visualized beyond the segmental bronchi even with the use of HRCT. Diseases that spread along lymphatic channels such as lymphangitic carcinomatosis and sarcoid commonly produce abnormalities of the peribronchovascular interstitium .
A subpleural line is a thin (1-2 mm) curvilinear opacity which is less than 1 cm from the pleural surface and parallels the pleura. These lines may be the result of the confluence of peribronchiolar interstitial abnormalities- representing early fibrosis with alveolar collapse. However, these lines may be seen in normal patients as a result of atelectasis within dependent lung (the abnormality disappears in these patients when placed prone).Tree-in-bud Appearance:
On HRCT abnormal bronchioles filled with fluid, mucus, or pus can appear as centrilobular tubular, branching, or nodular structures. These can be associated with centrilobular opacities if inflammation of the adjacent lung is present. This combination of fluid or pus filled bronchioles and surrounding inflammation has been described as the "tree-in-bud" appearance . Disorders associated with a tree-in-bud appearance include infection such as endobronchial spread of mycobacterium tuberculosus (classically), atypical mycobacterial infection, as well as bacterial, viral, or fungal infection . Congenital disorders associated with small airways disease can also produce a tree-in-bud appearance and include cystic fibrosis, dyskinetic cilia syndrome, yellow nail syndrome, and immunodeficiency states