Logout
 CME Information
 Editorial Board
 Lessons by Volume
   Volume 22
   Volume 21
   Volume 20
   Volume 19
   Volume 18
   Volume 17
   Volume 16
   Volume 15
 
 

Lesson 5, Volume 15—Clinical Indications for Spiral CT of the Lung

By Jud W. Gurney, MD

Effective December 31, 2004, PCCU Volume 15 is available for review purposes only. CME credit for this volume is no longer being offered.

Objectives

  1. To understand the difference between conventional CT and newer spiral CT.
  2. To understand the basic principles of spiral CT.
  3. To judge the significance of subsegmental pulmonary embolus.
  4. To understand the value of spiral CT in diagnosis of pulmonary embolus.
  5. To understand the value of spiral CT in detecting and characterizing pulmonary nodules.

Key words

high-resolution CT; metastatic disease; pulmonary embolus; solitary pulmonary nodule; spiral CT; traumatic aortic injury

Abbreviations

HRCT = high-resolution CT; Hu = Hounsfield unit; MIP = maximum intensity projection; SPN = solitary pulmonary nodule; 3D = three-dimensional

Introduced in the 1970s, CT revolutionized the practice of medicine. Now 30 years old, CT technology continues to improve and innovate. Images that originally were acquired in minutes now can be acquired in less than a second. Concomitant with the increase in temporal resolution, continued progress has been made in spatial resolution and reconstruction algorithms. Spiral CT, introduced into clinical practice 10 years ago,1 is now widely available and offers several advantages over previous conventional CT imaging. In particular, spiral CT is an important tool in the investigation of a wide variety of pulmonary disorders.

Slicing Potatoes

The main technologic difference between spiral CT and conventional CT is best described with a simple analogy. A sliced potato (American fries) is similar to conventional CT acquisition. Each slice is independent of adjacent contiguous slices. Slice collimation (thickness of the slice) is chosen by the radiologist to achieve diagnostic spatial resolution. However, the potatoes may also be sliced with a rotary slicer (curlicue fries). Now the potato is turned into one continuous ribbon. This is what happens in spiral CT. This ribbon forms a continuous volumetric data set and the slice location is chosen electronically. Adjacent images are not independent but derived from a continuous stream of information. The advantages of the spiral data set over conventional CT are many. Multiple overlapping images can be reconstructed, reducing the problems of misregistration and partial volume artifact. The volumetric data set can be used to produce diagnostic multiplanar and three-dimensional (3D) images. In conventional CT, the x-ray tube had to stop and start with each gantry rotation, as did the table. In spiral CT, the tube and table move continuously, scan time is markedly reduced, and the entire thorax may be scanned in one breath-hold. In conjunction with decreased scan time, IV contrast can be followed in a single pass through the arterial or vascular system. The volume of contrast can be reduced for a given level of contrast enhancement.2

In spiral CT, in addition to the choice of collimation, an additional variable is chosen: table speed.3 This is known as pitch, which is equal to table speed divided by the collimator width times the speed of gantry rotation. Using a toy Slinky as an analogy for the spiral acquisition, as the pitch increases, the coils are stretched further apart. With a pitch of 1, slice thickness is equal to the collimator settings. At a pitch of 2, slice thickness is 30% greater than the designated collimation. Advantages of increasing pitch are reduced scan time and reduced radiation dose. The disadvantage is a decrease in spatial resolution. In the thorax, image quality is acceptable up to a pitch of 2, at which point image degradation becomes a problem. In conventional CT, a survey of the thorax was performed at 8- or 10-mm collimation. Modern CT protocols use thinner collimation with a higher pitch for an examination that has a higher resolution in a shorter scan time as compared with conventional CT. For example, using 5-mm collimation at 1.4:1 pitch, the effective slice thickness for a similarly performed conventional CT is 7.0 mm. One of the least noticed benefits of spiral CT is reduced radiation dose. It is estimated that CT accounts for 10% of the population bone marrow dose. As compared with conventional CT, at a pitch of 1.4, there is a 25% reduction in radiation dose; at pitch of 2, there is a 45% reduction in radiation dose.4

Pulmonary Nodules

Spiral CT is the technology of choice to detect and characterize pulmonary nodules.5 Scanning the thorax in a single breath-hold obviates the problem of respiratory misregistration. Respiratory misregistration is especially problematic in the lung bases, the commonest location for pulmonary metastases. Costello et al6 compared standard (8-mm collimation) and spiral CT (8-mm collimation, 1:1 pitch, 4-mm reconstruction) in 20 patients with a suspected solitary pulmonary nodule (SPN) < 1 cm in diameter. Spiral CT detected four additional nodules. Similarly, Remy-Jardin et al7 compared conventional CT with spiral CT in 39 patients. Spiral CT detected more nodules (mean, 18) than conventional CT (mean, 16). Detection of pulmonary nodules is improved by the use of reconstruction intervals smaller than the collimator width. The latter protocol should be optimal in patients with suspected metastatic disease. For example, Buckley et al8 reviewed spiral CT in 63 patients, varying the reconstruction intervals. More nodules were detected with a greater confidence level with overlapped images than with nonoverlapped images.

Spiral CT also plays a role in the evaluation of the SPN. In some patients, it was difficult to evaluate an SPN with thin sections, particularly if the patient had difficulty with breath-holding or difficulty reproducing the same level of inspiration between scans. However, with spiral CT, the entire nodule is scanned at a single breath-hold, ensuring that the center of the nodule is available for characterization. The ability to reliably image the entire nodule at thin sections is particularly advantageous in enhancement studies. Enhancement of SPNs requires the reassessment of the nodule multiple times over a short time interval. Five 3-mm collimation spiral acquisitions are obtained through the entire nodule at 1-min intervals beginning 1 min after injection onset. Such studies would not be possible without spiral CT technology. In a multicenter study of 356 solitary pulmonary nodules, the absence of contrast enhancement (< 15 Hounsfield units [Hu]) was strongly predictive of benignity (likelihood ratio, 0.06), and the presence of contrast enhancement (> 15 Hu) was strongly predictive of malignancy (likelihood ratio, 2.32).9

High-Resolution CT

High-resolution CT (HRCT) has proven to be a valuable technique in the evaluation of chronic infiltrative lung diseases.10 HRCT is more sensitive in detecting disease and diagnostically more accurate than chest radiographs in the evaluation of these disorders. HRCT technique optimizes spatial resolution through a combination of narrow collimation, targeted field of view, and choice of high spatial frequency reconstruction algorithm. One of the limiting factors in spatial resolution is collimation. When HRCT was first introduced, the narrowest collimation was 1.5 mm. Current scanners have collimation of 1 mm and one manufacturer has introduced a scanner with collimation of 0.75 mm. However, the disadvantage of thinner collimation is the lack of normal anatomic landmarks in the image. When the pulmonary vessels run perpendicular to the scan plane (typically both the upper and lower lobes), an infinitely thin slice will contain nothing but dots. The morphology of vessels is defined by their tubular course and branching angles, which are progressively lost as slice thickness decreases. It is much easier to recognize blood vessels in a 10-mm image than a 1-mm image. With thinner collimation, separating normal vessels from pathologic ones will be difficult. A technique known as thin-slab maximum intensity projection (MIP), when applied to the volumetric data set, combines the benefits of the spatial resolution with thin collimation while allowing the easy recognition of airways or blood vessels.11 In this technique, for example, five contiguous 1-mm images are stacked atop one another. Each image is a square grid of 512 pixels. The computer displays the maximum intensity pixel at a given location from the slice with the highest intensity pixel. The slabs can be reconstructed at MIP or minimum intensity projection. MIP is useful for analyzing pathology, especially micronodules in relation to blood vessels (Fig 1). (In one study, the detection of micronodules improved from 73% with HRCT to nearly 100% with MIP.12) Minimum intensity projection is useful in evaluating hypoattenuating airways or areas of emphysema (Fig 2).13

A B
Figure 1. A, HRCT 1-mm slice interpreted as normal. B, thin-slab MIP (five 1-mm slices) right mid lung. Note the easy identification of blood vessels. Not apparent in A are the small centriacinar nodules (arrows) in this patient with sarcoid.
A B
Figure 2. A, HRCT 1-mm slice, diffuse emphysema, at the level of the carina. Multiple centriacinar hypoattenuating areas scattered throughout the lung. B, thin-slab minimum intensity projection (five 1-mm slices). Most of the blood vessels disappear. The extent and severity of emphysema is somewhat better appreciated with this technique.

Pulmonary Embolus

Clinicians have struggled for years with the diagnosis of pulmonary embolus. A magical test that could provide an answer in patients in whom pulmonary embolus is suspected has not been available. Rather, a convoluted cascade of multiple tests has been used to chase this diagnosis. Before spiral CT, anecdotal case reports of pulmonary embolus demonstrated at conventional CT were published. These cases marked the fortuitous scanning at peak contrast through the pulmonary arteries. Spiral CT and electron beam CT both have the advantage of speed and are able to follow a highly concentrated bolus of contrast as it travels through the pulmonary arterial vessels. CT excels at demonstrating clots in the main, lobar, and segmental pulmonary artery (Fig 3).14 Controversy concerning the frequency and significance of segmental emboli cloud the use of CT in patients with suspected pulmonary embolus.

Figure 3. Four contiguous images from CT pulmonary angiogram. Note the clot (arrows) centered in the lumen of the right interlobar pulmonary artery.

Numerous studies have now shown that spiral CT is > 90% sensitive and specific in the diagnosis of pulmonary embolus. Goodman et al15 first noted the problem of subsegmental emboli. In a highly selected group of 20 patients with indeterminate ventilation-perfusion scans, angiography demonstrated emboli in 11 patients, 4 of whom had emboli limited to the subsegmental vessels. In these 4 patients, spiral CT was positive in 1. The authors concluded that CT is of limited value because the sensitivity was only 63% (7/11). These poor results merit further inspection. In this small study, almost all the missed emboli were subsegmental. A single angiographer interpreted the gold standard, pulmonary angiography. The disagreement among angiographers to the presence of subsegmental emboli is large. In the PIOPED study,16 there was only 66% agreement among the angiographers for the presence or absence of emboli at the subsegmental level. In this study, the interpretation of one test assumes large significance in the results. For example, the poor sensitivity of 63% (7/11) rises to 70% (7/10) if one angiogram is interpreted as normal rather than as subsegmental embolus. The binomial expansion is a probability equation that can be used to answer the following question. What is the probability, given that there is a 34% disagreement amongst angiographers for the detection of pulmonary emboli at the subsegmental level, that at least one of the four cases of subsegmental emboli would be changed to normal if interpreted by a different observer? Interestingly, the probability of changing a single answer is 80%.

Pulmonary angiography is not a perfect test. As already noted, the agreement rate for small subsegmental emboli is poor. The rate of false-negative pulmonary angiograms has been estimated to be 1 to 9%.17 This has been highlighted by the comparison studies of CT and pulmonary angiography reporting discordant results: The gold standard of angiography missed a CT-detected clot.14 Clearly, small emboli are missed by pulmonary angiography. The consequences of such misses are, however, known. Patient outcome, the ultimate judge of the usefulness of a test, has demonstrated that a false-negative pulmonary angiogram has few adverse outcomes, suggesting that small subsegmental emboli have little consequence. In the PIOPED study,16 a 1-year surveillance of patients with negative angiograms showed that pulmonary embolus occurred in only four (0.6%). The consequence of missing subsegmental emboli may not be as dire as is usually thought. A single subsegment serves less than 2% of the pulmonary vasculature. This loss in an otherwise healthy adult would have little consequence for pulmonary function. One of the normal functions of the lung is to remove small emboli that would have disastrous consequences if allowed into the arterial circulation. The lung is ideally suited for removing small emboli. Located between the arterial and venous systems, it has a large capillary bed, dual blood supply, and triple oxygen supply that helps sustain an embolized segment. Normal nonsmoking subjects have been shown to have subsegmental defects on ventilation-perfusion scintigraphy, suggesting that such a function occurs.18

The frequency of subsegmental emboli varies, depending on selection bias and study populations. The most accurate estimate is derived from the PIOPED data. Isolated subsegmental emboli were demonstrated in 5.9% (22/375).16 If spiral CT is positive, the diagnosis is made with a high degree of confidence, as false-positive diagnoses are rare.14 However, the usefulness of a negative examination is unknown. Whether similar outcomes occur in patients with negative spiral CT examinations is only now being reported. Garg et al19 followed 78 patients in whom no embolus had been detected with CT and no treatment given over a 6-month period. One patient had microscopic pulmonary embolus at autopsy (negative predictive value, 99%). Confirmation of the good outcomes in patients with a negative spiral CT awaits the publication of similar studies. Newer protocols take advantage of the speed of spiral CT to examine both the central pulmonary arteries and deep venous systems at the same time. Loud et al20 showed perfect concordance between sonography of the femoropopliteal deep venous system and CT. Spiral CT, however, had the advantage of depicting extension into the pelvic veins or IVC, vessels poorly shown on sonography.

Three-Dimensionality

One of the advantages of the volumetric data set is the reconstruction of 3D images.21 Conventional CT data sets suffered from step artifacts due to the discontinuous image acquisition. Although 3D reconstructions contain no more information than the axial scans, 3D images ease the appreciation of complex spatial relationships. Volume-rendering techniques produce images similar to conventional bronchograms (Fig 4). These techniques are paricularly applicable to congenital anomalies and other complex deformations of the airways.22

A B
Figure 4. A, 3D reconstruction of the trachea. Bronchographic view. Multiple reconstruction artifact adjacent to the midportion of the trachea. B, coronal MIP image through the trachea. This better demonstrates the irregular wall thickening in this patient with relapsing polychondritis. Note the absence of stair-step artifact usually associated with reconstructions from conventional CT.

Traumatic Aortic Injury

Traumatic aortic injury is a life-threatening injury if not treated promptly. It occurs most frequently from motor vehicle accidents and high-speed deceleration injuries. The chest radiograph is highly sensitive but nonspecific for the diagnosis of traumatic injury. The traditional test for aortic injury is aortography. Aortography is time-consuming, however, and delays evaluation and treatment of the other injuries that are common in these patients. Conventional CT has been used as an intermediate step in the evaluation of the mediastinum. A normal mediastinum at conventional CT has a high negative predictive value (> 99%) for aortic tear. However, as a method to directly diagnosis the tear, conventional CT is poor. Spiral CT has now been used in three studies to directly image the tear, eliminating the need for aortography.23-25 Although aortic injury is infrequent, the results to date are promising, with a sensitivity of 100% and specificity of 98% (Fig 5). Unanswered is the utility of spiral CT in detection of great vessel injury, which occurs even less frequently than aortic injury.

A B
Figure 5. A, Four contiguous images from CT in patient with blunt chest trauma. Contour abnormality along the undersurface of the aortic arch (arrows) consistent with pseudoaneurysm. B, 3D reconstruction of the aortic arch again demonstrates the pseduoaneurysm in relation to the aortic arch. The patient did not undergo aortography and the defect was surgically repaired.

Summary

For thoracic applications, spiral CT has replaced previous generations of CT scanners. Decreased radiation dose, faster throughput, and better temporal and spatial resolution all have contributed to the clinical acceptance of CT. The indications of CT in clinical practice have expanded to the diagnosis of pulmonary embolus and aortic trauma. New refinements continue to add to the usefulness of CT. Newer multidetector scanners have rows of detectors. Now multiple images are acquired at varying collimation with a single rotation of the x-ray tube. Such scanners are even faster and may aid in the evaluation of subsegmental vessels for the diagnosis of pulmonary embolus.

References

  1. Kalender WA, Seissler W, Klotz E, et al. Spiral volumetric CT with single-breath-hold technique, continuous transport, and continuous scanner rotation. Radiology 1990; 176:181–183
  2. Brink JA, Heiken JP, Forman HP, et al. Hepatic spiral CT: reduction of dose of intravenous contrast material. Radiology 1995; 197:83–88
  3. Leung AN. Spiral CT of the thorax in daily practice: optimization of technique. J Thorac Imag 1997; 12:2–10
  4. Hidajat N, Jürgen M, Schröder R-J, et al. Radiation exposure in spiral computed tomography: dose distribution and dose reduction. Invest Radiol 1999; 34:51–57
  5. Padhani AR. Spiral CT: thoracic applications. Eur J Radiol 1998; 28:2–17
  6. Costello P, Anderson W, Blume D. Pulmonary nodule: evaluation with spiral volumetric CT. Radiology 1991; 179:875–876
  7. Remy-Jardin M, Remy J, Giraud F, et al. Pulmonary nodules: detection with thick-section spiral CT versus conventional CT. Radiology 1993; 187:513–520
  8. Buckley JA, Scott Jr WW, Siegelman SS, et al. Pulmonary nodules: effect of increased data sampling on detection with spiral CT and cinfidence in diagnosis. Radiology 1995; 196:395–400
  9. Swensen SJ, Viggiano RW, Midthun DE, et al. Lung nodule enhancement at CT: multicenter study. Radiology 2000; 21:73–80
  10. Swensen SJ, Aughenbaugh GL, Douglas WW, et al. High-resolution CT of the lungs: findings in various pulmonary diseases. AJR Am J Roentgenol 1992; 158:971–979
  11. Napel S, Geoffrey DR, Jeffrey Jr RB. STS-MIP: a new reconstruction technique for CT of the chest. J Comput Assist Tomogr 1993; 17:832–838
  12. Remy-Jardin M, Remy J, Artand D, et al. Diffuse infiltrative lung disease: clinical value of sliding thin slab maximum intensity projection CT scans in the detection of micronodular patterns. Radiology 1996; 200:333–339
  13. Remy-Jardin M, Remy J, Gosselin B, et al. Sliding thin slab, minimum intensity projection technique in the diagnosis of emphysema: histopathologic-CT correlation. Radiology 1996; 200:665–671
  14. Hansell DM. Spiral computed tomography and pulmonary embolism: current state. Clin Radiol 1997; 52:575–581
  15. Goodman LR, Curtin JR, Mewissen MW, et al. Detection of pulmonary embolism in patients with unresolved clinical and scintigraphic diagnosis: helical CT versus angiography. AJR Am J Roentgenol 1995; 164:1369–1374
  16. Stein PD, Athanasoulis C, Alavi A, et al. Complications and validity of pulmonary angiography in acute pulmonary embolism. Circulation 1992; 85:462–468
  17. Wellman HN. Pulmonary thromboembolism: current status report on the role of nuclear medicine. Semin Nucl Med 1986; 16:236–274
  18. Tetalman MR, Hoffer PB, Heck LL, et al. Perfusion lung scan in normal volunteers. Radiology 1973; 106:593–594
  19. Garg K, Sieler H, Welsh CH, et al. Clinical validity of helical CT being interpreted as negative for pulmonary embolism: implications for patient treatment. AJR Am J Roentgenol 1999; 172:1627–1631
  20. Loud PA, Katz DS, Klippenstein DL, et al. Combined CT venography and pulmonary angiography in suspected thromboembolic disease: diagnostic accuracy for deep venous evaluation. AJR Am J Roentgenol 2000; 174:61–65
  21. Johnson PT, Fishman EK, Duckwall JR, et al. Interactive three-dimensional volume rendering of spiral CT data: current applications in the thorax. Radiographics 1998; 18:165–187
  22. Remy-Jardin M, Remy J, Artaud D, et al. Volume rendering of the tracheobronchial tree: clinical evaluation of bronchographic images. Radiology 1998; 208:761–770
  23. Dyer DS, Moore EE, Mestek MF, et al. Can chest CT be used to exclude aortic injury? Radiology 1999; 213:195–202
  24. Gavant ML, Menke PG, Fabian T, et al. Blunt traumatic aortic rupture: detection with helical CT of the chest. Radiology 1997; 197:125–133
  25. Wicky S, Capasso P, Meuli R, et al. Spiral CT aortography: an efficient technique for the diagnosis of traumatic aortic injury. Eur Radiol 1997; 8:828–833

Copyright ©2001 American College of Chest Physicians

 
 

3300 Dundee Road • Northbrook, Illinois 60062-2348 • (847) 498-1400 • (800) 343-2227
Copyright ©1999-2008 American College of Chest Physicians. All Rights Reserved.
ACCP Privacy Policy | ACCP Web Site Terms of Use