Signs in Thoracic Imaging

Marshall, Geoffrey B. BSc, MD; Farnquist, Brenda A. BMLSc, MD; MacGregor, John H. MD, FRCP(C); Burrowes, Paul W. MD, FRCP(C)

Journal of Thoracic Imaging:
doi: 10.1097/01.rti.0000189192.70442.7a
Classic Concepts in Thoracic Imaging

Radiologic signs are recognizable, characteristic patterns used to describe abnormalities visualized on imaging modalities that ultimately aid in the diagnosis and subsequent treatment of disease. This pictorial essay discusses 23 classic roentgenographic signs used in thoracic imaging. Its purpose is to be used as an educational review for residents, whether they are beginning their training or preparing for certification exams, and serve as a refresher and a reference to the practicing radiologist.

Author Information

Department of Diagnostic Imaging, Foothills Medical Center, Calgary, Alberta, Canada

Reprints: Dr Paul Burrowes, Department of Diagnostic Imaging, Foothills Medical Centre, 1403-29 Street NW, Calgary, Alberta, Canada T2N 2T9 (e-mail:

Article Outline

Radiologic signs are recognizable, characteristic patterns used to describe abnormalities visualized on imaging modalities that ultimately aid in the diagnosis and subsequent treatment of disease. When the late pulmonary radiologist Benjamin Felson was asked about his propensity for naming signs he responded that “the name saves time, helps you remember the sign, and advertises it… They say what I mean.”1 This pictorial essay discusses 23 classic roentgenographic signs used in thoracic imaging and presents them in alphabetic order. Its purpose is to be used as an educational review for residents, whether they are beginning their training or preparing for certification exams, and serve as a refresher and a reference to the practicing radiologist.

Back to Top | Article Outline

The air crescent sign appears as a variably sized, peripheral crescentic collection of air surrounding a necrotic central focus of infection on thoracic radiographs (Fig. 1A) and CT (Fig. 1B).2–4 It is often seen in neutropenic patients who have undergone bone marrow or organ transplantation and is most characteristic of infection with invasive pulmonary aspergillosis. The fungus invades the pulmonary vasculature, causing hemorrhage, thrombosis, and infarction. With time, the peripheral necrotic tissue is reabsorbed by leukocytes and air fills the space left peripherally between the devitalized central necrotic tissue and normal lung parenchyma.5 Thus, the presence of the air-crescent sign heralds recovery of granulocytic function.4 Other causes of the air crescent include cavitating neoplasms, bacterial lung abscesses, and infections such as tuberculosis or nocardiosis.6

Back to Top | Article Outline

Classically, this sign is associated with consolidation of the right upper lobe (Fig. 2) due to Klebsiella pneumoniae infection.7 Due to the tendency for Klebsiella to produce large volumes of inflammatory exudate, the involved lobe expands and exerts mass effect on the adjacent interlobar fissure.8 The normally straight minor fissure on the lateral view bulges convex posteroinferiorly due to rapid lobar expansion.3,7 Although previously reported in up to 30% of patients with Klebsiella pneumonia,8,9 the finding is identified less commonly today, most likely due to rapid prophylactic implementation of antibiotics.3 Other less common causes of the bulging fissure sign include Hemophilus influenzae, tuberculosis, pneumococcal pneumonia, large lung abscesses, and lung neoplasms.10

Back to Top | Article Outline

This sign is based on the understanding that if a thoracic mass is in direct contiguity with the soft tissues of the neck, the borders delineating their point of contact will be lost or obscured.11 Anatomically, the thoracic inlet parallels the first ribs, and the posterior aspects of the lung apices extend further superiorly than the anterior portions.12 On PA chest radiographs, a mass clearly delineated on all borders above the level of the clavicles lies posterior to the level of the trachea and completely within the lung.13 When the cephalic border of a mass is obscured at or below the level of the clavicles, it is deemed to be a “cervicothoracic lesion”13 involving the anterior mediastinum (Figs. 3A–E). Mediastinal masses posterior to the trachea are well outlined above the level of the clavicles due to the interface with lung in the posterior aspects of the lung apices (Figs. 3F, G).

Back to Top | Article Outline

Identified as a curvilinear opacity arising from a rounded, subpleural opacity on CT (Fig. 4A, arrow), the “comet tail” denotes the distorted curving of pulmonary vessels and bronchi toward a region of round atelectasis.14–17 The atelectatic lung typically abuts an area of pleural effusion or thickening, and there are associated signs of volume loss. Round atelectasis is an unusual form of atelectasis most commonly associated with asbestos-related pleural disease (Fig. 4B, arrow), but also seen with other chronic forms of pleural disease, such as tuberculosis, histoplasmosis, pulmonary infarcts, or congestive heart failure.14–17 It is also known as atelectatic pseudotumor, folded lung, or Blesovsky syndrome, after the physician who initially postulated an association with asbestos exposure.18 Most often found in the posterior aspect of the lower lobes, it may demonstrate significant contrast enhancement and contain air bronchograms.14–17 Histopathology demonstrates fibrous thickening of the visceral pleura with extensive pleural folding and invagination.19 The exact cause is still debated. One theory suggests that a pleural effusion is the inciting event causing passive atelectasis and pleural invagination.20 Fibrous adhesions will eventually develop between the visceral and parietal pleural surfaces, so that when the effusion clears, the entrapped lung folds in on itself. An alternate theory suggests that a focal inflammatory pleural fibrotic response secondary to irritants such as asbestos, is solely responsible for causing invagination of the pleura and regional atelectasis of the subjacent lung.14,19,21

Back to Top | Article Outline

Identifiable on PA chest radiography (Fig. 5, arrows), there is a continuous line of lucency extending across the midline, above the diaphragm.22,23 This is re- presentative of air tracking posterior to the heart in pneumomediastinum and is helpful in differentiating from a pneumothorax.4,22,23

Back to Top | Article Outline

The “crazy paving” pattern12 is described at HRCT of the lungs (Fig. 6). The pattern is caused by thickening of interlobular septa superimposed on areas of ground glass opacification.24–26 It was first used to describe the pattern of pulmonary alveolar proteinosis, but can be seen in a variety of diseases. Histopathologically, the ground glass attenuation in alveolar proteinosis correlates to lipid-rich, intra-alveolar proteinaceous fluid, with thickening of the interlobular septa due to adjacent inflammatory reaction.13 Infectious causes include Pneumocystis carinii pneumonia, organizing pneumonia, usual interstitial pneumonia, non-specific interstitial pneumonia, and exogenous lipoid pneumonia. Other entities, such as respiratory bronchiolitis with interstitial lung disease,27 sarcoidosis, adult respiratory distress syndrome, pulmonary hemorrhage, and mucinous BAC, also less commonly demonstrate the crazy paving pattern.12–15

Back to Top | Article Outline

The CT halo sign appears as a zone of ground-glass attenuation around a nodule or mass (Fig. 7) on computed tomographic (CT) images.2–4,6,28 In febrile neutropenic patients, the sign suggests angioinvasive fungal infection, which is associated with a high mortality rate in the immunocompromised host.2–4 The zone of attenuation represents alveolar hemorrhage,2,4,6,28 whereas the nodules represent areas of infarction and necrosis caused by thrombosis of small to medium sized vessels.2–4,6,28,29 Other infectious causes include candidiasis, cytomegalovirus, herpes simplex virus, and coccidioidomycosis.30 The CT halo sign may also be caused by non-infectious causes, such as Wegener granulomatosis, metastatic angiosarcoma, Kaposi sarcoma, and brochioloalveolar carcinoma (BAC).29,30 Due to the lepidic growth pattern of BAC, where the tumor cells spread along the alveolar walls, the typical ground glass halo visualized with the sign results.29

Back to Top | Article Outline

The presence of radiolucency in a deep costophrenic sulcus on a supine thoracic radiograph (Fig. 8) is characteristic of a pneumothorax in a supine patient.31 Intrapleural air rises to the highest portion of the hemithorax leading to the presence of a lucency in the anteromedial, subpulmonic, and lateral basilar space adjacent to the diaphragm.3,4,31 It is useful in confirming suspected pneumothorax on AP supine radiography in compromised patients, such as those in the intensive care setting.4

Back to Top | Article Outline

On frontal chest radiographs, this sign presents as a curvilinear density projecting over the right retrocardiac region, signifying left atrial enlargement (Fig. 9).32 The curvilinear line represents the inferolateral margin of the left atrium.33 The double density sign may be observed in patients without cardiac disease; however, there is a semiquantitative measurement to estimate the left atrial diameter and better estimate whether it is a real finding.33 Higgins et al found that, on PA radiographs of adult patients, if the left atrial dimension is defined as the distance from the midpoint of the double density to the inferior wall of the left mainstem bronchus, a distance greater than 7 cm was consistent with a diagnosis of left atrial enlargement, confirmed on echocardiography.33 The measurement was found to be an unreliable sign in the evaluation of pediatric patients with a double density sign on PA radiographs.33

Back to Top | Article Outline

Although CT is widely accepted as the primary modality for detecting mediastinal lymphadenopathy, chest radiography is generally the initial radiographic investigation performed. On a normal lateral chest radiograph (Fig. 10A), the aortic arch and right and left pulmonary arteries are visualized in an “inverted horseshoe” configuration.34 In the presence of subcarinal lymphadenopathy (Figs. 10B, C), the inferior portion of the “horseshoe” fills in.34 The lymphadenopathy appears as a mass posterior to the bronchus intermedius and inferior to the tracheal bifurcation, completing the rounded hilar “doughnut” density.34,35

Back to Top | Article Outline

Tubular shadows of soft tissue opacity akin to gloved fingers are seen on thoracic radiographs (Figs. 11A, B) and CT (Fig. 11C) and typically originate in the upper lobes in a bronchial distribution.2–4,36,37 The tubular, fingerlike projections represent dilated, mucoid-impacted bronchi surrounded by aerated lung. When an inciting stenosis or bronchial obstruction occurs, mucous glands will continue to produce fluid, while the secretions are continually taken to the site of narrowing by mucociliary transport.38 As the secretions become inspissated, debris accumulates distal to the point of obstruction, and bronchiectasis ensues. Visualization of the gloved fingers is made possible by collateral air drift through the interalveolar pores of Kohn and canals of Lambert aerating lung distal to the point of mucoid impaction. There are 2 broad etiologic categories: non-obstructive and obstructive.38 Non-obstructive causes, such as allergic bronchopulmonary aspergillosis (ABPA), asthma, or cystic fibrosis, are common considerations. ABPA is seen most commonly in asthmatic patients and occurs after inhaled Aspergillus organisms are trapped in airway mucus, triggering subsequent type I and type III allergic reactions.38 The acute type I response results in bronchoconstriction, heightened vascular permeability, wall edema, and protracted mucus production, whereas the delayed type III response causes immunopathological damage to the involved bronchi.38 Mucoid impaction in the setting of cystic fibrosis is secondary to mucociliary dysfunction and thick mucous secretions. Benign (bronchial hamartomas or lipomas) and malignant (bronchogenic carcinoma or carcinoid tumors) neoplasms are considerations in the obstructive category. Congenital obstructive causes, such as bronchial atresia, intralobar sequestration, or bronchogenic cysts, might also be considered.

Back to Top | Article Outline

Felix Fleischner was a German-born radiologist who immigrated to the United States during the second World War.39 The finding that bears his name refers to enlargement of the proximal pulmonary arteries on plain chest radiography (Fig. 12A) or angiography (Fig. 12B) secondary to pulmonary embolism.40,41 This occurs most commonly in the setting of massive pulmonary embolism (defined angiographically as involving 50% or more of the major pulmonary artery branches)42 and has a relatively low sensitivity in diagnosis.43 The enlargement can be caused by massive embolus enlarging the luminal diameter of the proximal artery in the acute setting but can also be seen in the subacute to chronic setting due to elevated pressures in the pulmonary arterial circulation.44 An important ancillary finding is the abrupt tapering of the occluded pulmonary artery distally, creating the “knuckle sign.”44

Back to Top | Article Outline

Dr Ross Golden was an American-born radiologist who originally described this sign as a form of right upper lobe collapse associated with a central mass.45 When the right upper lobe bronchus is obstructed by an endobronchial lesion, there is elevation and medial displacement of the minor fissure with proximal convexity of the fissure due to the mass (Fig. 13). This creates the “reverse S” characteristic of central obstructing bronchogenic carcinomas.3,4,45 Given the same presentation of a proximal, obstructing endobronchial lesion within the left upper lobe bronchus with associated left upper lobe collapse, the upwardly retracted major fissure will follow an S-shaped contour along its length. Although initially used to describe signs of right upper lobe collapse, the Golden S sign can be applicable to atelectasis involving any lobe.46

Back to Top | Article Outline

On plain radiographs (Fig. 14A) and CT (Figs. 14B, C), pulmonary infarcts are typically multifocal, peripheral in location, contiguous with one or more pleural surfaces, and more commonly confined to the lower lungs.3,47,48 The apex of these rounded or triangularly shaped opacities may point toward the hilum.3 The opacities resolve slowly over a period of several months, akin to “melting ice cubes,” and may leave a residual scar.3 The first documentation of the finding was made by Aubrey Otis Hampton, who was a practicing radiologist in the mid 1920s. He and his co-author Castleman first reported evidence of incomplete pulmonary infarction in the setting of PE in the 1940s.47,48 Autopsy follow-up showed evidence of intra-alveolar hemorrhage without alveolar wall necrosis in the first 2 days of infarction. After 2 days, wall necrosis begins and eventually leads to pulmonary infarction and an organized scar.47,48 Hampton also observed that there were differences in the healing of these incomplete infarcts depending on their premorbid cardiac history.47,48 In patients without heart disease, the incomplete infarcts would generally heal without scarring, whereas patients with congestive failure were more likely to progress to infarction with a persisting pulmonary scar.47,48 When pulmonary embolism results in infarction, airspace opacities typically develop within 12 to 24 hours.3,48

Back to Top | Article Outline

In evaluating densities projecting over the hila on PA chest radiographs, it is important to evaluate whether the opacity blends with the normal hilar shadow of the proximal pulmonary arteries and obscures clear visualization or if the mass and hilar structures are overlapping but distinct from one another. If the hilar vessels are sharply delineated, then it can be assumed that the overlying mass is anterior (Fig. 15) or posterior to the centrally located vascular structures and the intervening aerated lung enables crisp visualization of the pulmonary vessels.12,13 However, when the mass is inseparable from the proximal pulmonary arteries, it is assumed that the structures are immediately adjacent to one another, with the lack of interposing air obscuring the margins of the normal hilar structures. This sign was originally described by Benjamin Felson.12,13

Back to Top | Article Outline

This sign, first documented by Kattan et al in 1980,49 is an ancillary sign of upper lobe collapse, depicted as a triangular opacity projecting superiorly over the medial half of the diaphragm, at or near its highest point (Fig. 16). It is most commonly related to the presence of an inferior accessory fissure.49,50 Although the mechanism is not certain, one theory suggests it is due to the negative pressure of upper lobe atelectasis causing upward retraction of the visceral pleura and extrapleural fat protruding into the recess of the fissure.49 A recent retrospective analysis of patients who have undergone upper lobectomies suggests that the prevalence of the sign increases in the ensuing weeks after intervention and documents its utility in specifically recognizing the type of surgery.51

Back to Top | Article Outline

The word Luftsichel is German for “air crescent.” The finding is seen in the setting of left upper lobe collapse. Due to the absence of a minor fissure on the left, as the left upper lobe collapses, the major fissure assumes a vertical position roughly parallel to the anterior chest wall.4 As volume loss progresses, the fissure continues to migrate more anteriorly and medially until the atelectatic lobe is contiguous with the left heart border, effectively obliterating its contour on the frontal radiograph (Figs. 17A–C). With movement of the apical upper lobe segment anteromedially, the superior segment of the left lower lobe hyperinflates and fills the vacated apical space.4,52 Occasionally this segment will insinuate itself between the aortic arch and the collapsed left upper lobe creating a sharp outline, or periaortic lucency, described as the Luftsichel.3,4,52

Back to Top | Article Outline

On PA chest radiography, a characteristic broad, curved shadow extends inferiorly toward the diaphragm along the right side of the heart (Figs. 18A, B).53 Its appearance resembles a Turkish sword and signifies a partial anomalous pulmonary venous return, most commonly to the infradiaphragmatic inferior vena cava (Figs. 18C–E). It is part of the congenital hypogenetic lung (scimitar) syndrome.4,53

Back to Top | Article Outline

Described on chest CT (Fig. 19, arrows) as a circle of soft-tissue attenuation abutting a lucent ring, this sign depicts the pulmonary artery lying adjacent to a dilated bronchus in cross-section.54–57 The sign is found on CT in patients with bronchiectasis, or irreversible, abnormal bronchial dilatation.56 Accompanying findings, such as peribronchial thickening, lack of bronchial tapering, and visualization of bronchi within 1 cm of the pleura, are all contributing findings confirming the diagnosis. In isolation, an enlarged bronchoarterial ratio may be a spurious finding, as bronchial diameter varies with parameters such as altitude.58 Measurement of the bronchial luminal diameter is made between the inner walls so as not to overestimate in cases with associated peribronchial thickening. Bronchiectasis occurs due to bronchial wall injury, most often due to necrotizing viral or bacterial bronchitis.4,57,58 It can also occur secondary to bronchial obstruction with subsequent mucus production dilating the involved airways, as mentioned previously with the finger-in-glove sign. Finally, in the setting of pulmonary fibrosis or radiation-induced lung injury, traction on the adjacent bronchi will cause dilatation. Other entities, such as cystic fibrosis and Kartagener syndrome, are also associated.4,57

Back to Top | Article Outline

This classic roentgenographic sign first described by Felson in 1950 states that “an intrathoracic lesion touching a border of the heart, aorta, or diaphragm will obliterate that border on the roentgenogram. An intrathoracic lesion not anatomically contiguous with a border of one of these structures will not obliterate that border.”13,59,60 The underlying premise of the silhouette sign is that visualization of a roentgen shadow depends on a difference in radiographic density between adjacent tissues.13 For example, obscuration of the right heart border is often used to differentiate a right middle lobe process from a lower lobe abnormality (Fig. 20).3,4,60 Similarly, if the left heart border is partially or completely obliterated, the lingula is the region of involvement. Conversely, the absence of the silhouette sign can help in further localizing a lesion. With lower lobe disease, the right or left heart border on the side of involvement is preserved, whereas the silhouette of the hemidiaphragm is obliterated in cases of lower lobe collapse or consolidation (Fig. 21). The basic tenets of the silhouette sign are used to explain various other signs discussed previously, such as the cervicothoracic and hilum overlay sign.

Back to Top | Article Outline

Separation and enhancement of the visceral and parietal pleural layers on CT (Fig. 22) is considered strong evidence of empyema.3,61 Normally, individual pleural layers are not discernable as discrete structures.3 Empyemic fluid fills the pleural space, resulting in thickening and enhancement of the pleura with a denotable separation.3,61

Back to Top | Article Outline

Best seen on high resolution computed tomography (HRCT) (Fig. 23), this sign appears as small, peripheral, centrilobular soft tissue nodules connected to multiple contiguous, linear branching opacities.3,4,62,63 This radiologic term represents the mucous plugging, bronchial dilatation, and wall thickening of bronchiolitis.3,4,62,63 The histopathological correlate demonstrates small airway plugging with mucus, pus, or fluid, with dilated bronchioles, peribronchiolar inflammation, and wall thickening. Anatomically, this sign relates to viscous fluid blocking the intralobular bronchiole of the secondary pulmonary lobule.63 Ordinarily the intralobular bronchiole is not visualized at HRCT (<1 cm); however, when filled with fluid and inflamed, the bronchiole becomes visible at the resolution of thin-slice CT.63 Although commonly associated with infectious etiologies such as endobronchial spread of Mycobacterium tuberculosis, recent evidence suggests that the sign is not specific for any one pulmonary disease, but includes other infectious entities such as Pneumocystis jiroveci (Pneumocystis carinii) pneumonia and invasive pulmonary aspergillosis. In addition, various immunodeficiency states, congenital disorders (cystic fibrosis), malignancy (lymphoma), panbronchiolitis,62 and aspiration are other etiologies in which tree-in-bud may be seen and, thus, clinical correlation is important in elucidating a diagnosis.3,4,62,63

Back to Top | Article Outline

Neil Westermark was a 20th-century German radiologist who first discovered that a certain subset of patients diagnosed with pulmonary embolism (PE) were complicated by pulmonary infarction.64 He described the “anemic” or oligemic peripheral regions of lung parenchyma as “wedge-shaped shadows.” Interestingly, he also found that the majority of patients with PE did not have pulmonary infarcts, a finding that has been affirmed in the more recent literature.47 The chest radiograph (Fig. 24A) and CT (Fig. 24B) findings of increased translucency (chest radiograph) or hypoattenuation (CT) corresponding to oligemia in the periphery of the lung distal to an occlusive arterial embolus is typical.3,4,64 Visualization typically signifies either occlusion of a larger lobar or segmental artery or widespread small vessel occlusion.4,64

Back to Top | Article Outline


The authors would like to thank Mr Ronald Hildebrandt for his assistance in preparing the images for the article. Special thanks to Heather Marshall for designing the schematic diagram of the scimitar vein.

Back to Top | Article Outline


1. Felson B. Letter from the editor. Sem Roentgenol. 1969;4:1.
2. Franquet T, Muller NL, Gimenez A, et al. Spectrum of pulmonary aspergillosis: histologic, clinical, and radiologic findings. Radiographics. 2001;21:825–837.
3. Collins J, Stern EJ. Signs in chest radiology. In: Collins J, Stern EJ, eds. Chest Radiology, The Essentials. Philadelphia: Lippincott Williams & Williams; 1999:18–27.
4. Muller NL, Fraser RS, Colman NC, et al. Radiologic Diagnosis of Diseases of the Chest. Philadelphia: W.B. Saunders Company; 2001.
5. Abramson S. Signs in imaging: the air-crescent sign. Radiology. 2001;218:230–232.
6. Curtis AM, Smith GJ, Ravin CE. Air crescent sign of invasive aspergillosis. Radiology. 1979;133:17–21.
7. Felson LB, Rosenberg LS, Hamburger M. Roentgen findings in acute Friedlander's pneumonia. Radiology. 1949;53:559–565.
8. Barnes DJ, Naraqi S, Igo JD. The diagnostic and prognostic significance of bulging fissures in acute lobar pneumonia. Aust N Z J Med. 1988;18:130.
9. Moon WK, Im JG, Yeon KM, et al. Complications of Klebsiella pneumonia: CT evaluation. J Comput Assist Tomogr. 1995;19:176.
10. Francis JB, Francis PB. Bulging (sagging) fissure sign in Hemophilus influenzae lobar pneumonia. South Med J. 1978;71:1452–1453.
11. Felson B. More chest roentgen signs and how to teach them. Radiology. 1968;90:429.
12. Felson B. The mediastinum. Semin Roentgenol. 1969;4:41.
13. Felson B. Localization of intrathoracic lesions. In: Felson B, ed. Chest Roentgenology. Philadelphia: W.B. Saunders Company; 1973.
14. Schneider HJ, Felson B, Gonzalez LL. Rounded atelectasis. AJR Am J Roentgenol. 1980;134:225–232.
15. Glass TA, Armstrong P, Minor GR, et al. Computed tomographic features of round atelectasis. J Comput Tomogr. 1983;7:183–185.
16. Doyle TC, Lawler GA. CT features of rounded atelectasis of the lung. AJR Am J Roentgenol. 1984;143:225–228.
17. Partap VA. The comet tail sign. Radiology. 1999;213:553–554.
18. Blesovsky A. The folded lung. Br J Dis Chest. 1966;60:19–22.
19. Menzies R, Fraser R. Round atelectasis: pathologic and pathogenic features. Am J Surg Pathol. 1987;11:674–681.
20. Hanke R, Kretzschmar R. Round atelectasis. Semin Roentgenol. 1980;15:174–182.
21. Hillerdal G. Rounded atelectasis: clinical experience with 74 patients. Chest. 1989;95:836–841.
22. Levin B. The continuous diaphragm sign. A newly recognized sign of pneumomediastinum. Clin Radiol. 1973;24:337–338.
23. Zylak CM, Standen JR, Barnes GR, et al. Pneumomediastinum revisited. Radiographics. 2000;20:1043–1057.
24. Franquet T, Gimenez A, Bordes R, et al. The crazy-paving pattern in exogenous lipoid pneumonia: CT-pathologic correlation. AJR Am J Roentgenol. 1998;170:315–317.
25. Rossi SE, Erasmus JJ, Volpacchio M, et al. “Crazy-paving” pattern at thin-section CT of the lungs: radiologic-pathologic overview. Radiographics. 2003;23:1509–1519.
26. Johkoh T, Itoh H, Muller NL, et al. Crazy-paving appearance at thin-section CT: spectrum of disease and pathologic findings. Radiology. 1999;211:155–160.
27. O'Donnell WJ, Kradin RL, Evins AE, et al. Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Case 39-2004. A 52-year-old woman with recurrent episodes of atypical pneumonia. N Engl J Med. 2004;351:2741–2749.
28. Kuhlman JE, Fishman EK, Siegelman SS. Invasive pulmonary aspergillosis in acute leukemia: characteristic findings on CT, the CT halo sign, and the role of CT in early diagnosis. Radiology. 1985;157:611–614.
29. Pinto P. Radiology. Signs in imaging: the CT halo sign. Radiology. 2004;230:109–110.
30. Primack SL, Hartman TE, Lee KS, et al. Pulmonary nodules and the CT halo sign. Radiology. 1994;190:513–515.
31. Gordon R. The deep sulcus sign. Radiology. 1980;136:25–27.
32. Kaye J, Van Lingen B, Meyer MJ. The radiological diagnosis of mitral valve disease. Br J Radiol. 1953;26:242–251.
33. Higgins CB, Reinke RT, Jones NE, et al. Left atrial dimension on the frontal thoracic radiograph: a method for assessing left atrial enlargement. AJR Am J Roentgenol. 1978;130:251–255.
34. Andronikou S, Wieselthaler N. Modern imaging of tuberculosis in children: thoracic, central nervous system and abdominal tuberculosis. Pediatr Radiol. 2004;34:861–875.
35. Roche CJ, O'Keeffe DP, Lee WK, et al. Selections from the buffet of food signs in radiology. Radiographics. 2002;22:1369–1384.
36. Greer AE. Mucoid impaction of the bronchi. Ann Intern Med. 1957;46:506–522.
37. Mintzer RA, Neiman HL, Reeder MM. Mucoid impaction of a bronchus. JAMA. 1978;240:1397–1398.
38. Nguyen ET. Signs in imaging: the gloved finger sign. Radiology. 2003;227:453–454.
39. Fraser RG, Mellins RB. The Fleischner Society: a 30th anniversary retrospective. Radiology. 2000;214:631–632.
40. Fleischner FG. Roentgen diagnosis of pulmonary embolism. Heart Bull. 1961;10:104–107.
41. Fleischner FG. Pulmonary embolism. Clin Radiol. 1962;13:169.
42. Kerr IH, Simon G, Sutton GC. The value of the plain radiograph in acute massive pulmonary embolism. Br J Radiol. 1971;44:751.
43. Stein PD, Athanasoulis C, Greenspan RH, et al. Relation of plain chest radiographic findings to pulmonary arterial pressure and arterial blood oxygen levels in patients with acute pulmonary embolism. Am J Cardiol. 1992;69:394.
44. Williams JR, Wilcox WC. Pulmonary embolism: roentgenographic and angiographic considerations. AJR Am J Roentgenol. 1963;89:333.
45. Golden R. The effect of bronchostenosis upon the roentgen-ray shadows in carcinoma of the bronchus. AJR Am J Roentgenol. 1925;13:21–30.
46. Woodring JH, Reed JC. Radiographic manifestations of lobar atelectasis. J Thorac Imag. 1996;11:109–144.
47. Dalen JE. Pulmonary embolism: what have we learned since Virchow? Chest. 2002;122:1440–1456.
48. Hampton AO, Castleman B. Correlation of postmortem chest teleroentgenograms with autopsy findings with special reference to pulmonary embolism and infarction. AJR Am J Roentgenol. 1940;43:305–326.
49. Kattan KR, Eyler WR, Felson B. The juxtaphrenic peak in upper lobe collapse. Radiology. 1980;134:763–765.
50. Davis SD, Yankelevitz DF, Wand A, et al. Juxtaphrenic peak in upper and middle lobe volume loss: assessment with CT. Radiology. 1996;198:143–149.
51. Konen E, Rozenman J, Simansky DA, et al. Prevalence of the juxtaphrenic peak after upper lobectomy. AJR Am J Roentgenol. 2001;177:869–873.
52. Webber M, Davies P. The Luftsichel: an old sign in upper lobe collapse. Clin Radiol. 1981;32:271–275.
53. Halasz NA, Halloran KH, Liebow AA. Bronchial and arterial anomalies with drainage of the right lung into the inferior vena cava. Circulation. 1956;14:826–846.
54. Naidich DP, McCauley DI, Khouri NF, et al. Computed tomography of bronchiectasis. J Comput Assist Tomogr. 1982;6:437–444.
55. McGuinness G, Naidich DT, Leitman BS, et al. Bronchiectasis: CT evaluation. AJR Am J Roentgenol. 1993;160:254–259.
56. Kang EY, Miller RR, Muller NL. Bronchiectasis: comparison of preoperative thin-section CT and pathologic findings in resected specimens. Radiology. 1994;195:649–654.
57. Ouellette H. Signs in imaging: the signet ring sign. Radiology. 1999;212:67–68.
58. Kim JS. Bronchoarterial ratio on thin-section CT: comparison between high altitude and sea level. J Comput Assist Tomogr. 1997;21:306–311.
59. Felson B, Felson H. Localization of intrathoracic lesions by means of the postero-anterior roentgenogram; the silhouette sign. Radiology. 1950;55:363–374.
60. Longuet R, Phelan J, Tanous H, et al. Criteria of the silhouette sign. Radiology. 1977;122:581–585.
61. Stark DD, Federle MP, Goodman PC, et al. Differentiating lung abscess and empyema: radiography and computed tomography. AJR Am J Roentgenol. 1983;141:163–167.
62. Akira M, Kitatani F, Lee YS, et al. Diffuse panbronchiolitis: evaluation with high-resolution CT. Radiology. 1988;168:433–438.
63. Eisenhuber E. Signs in radiology: the tree-in-bud sign. Radiology. 2002;222:771–772.
64. Westermark N. On the roentgen diagnosis of lung embolism. Acta Radiol. 1938;19:357–372.

chest; pulmonary; sign; silhouette; radiograph; computed tomography

© 2006 Lippincott Williams & Wilkins, Inc.