DOI: 10.1148/radiol.2392050261
(Radiology 2006;239:586-590.)
© RSNA, 2006
CT Screening for Lung Cancer:Prevalence and Incidence of Mediastinal Masses1
Claudia I. Henschke, PhD, MD,
In-Jae Lee, MD,
Ning Wu, MD,
Ali Farooqi, MD,
Arfa Khan, MD,
David Yankelevitz, MD,
Nasser K. Altorki, MD, For the ELCAP and NYELCAP Investigators
1 From the Weill Medical College of Cornell Univ, 525 E 68th St, New York, NY 10021 (C.I.H., A.F., D.Y., N.K.A.); Medical College of Hallym Univ, Seoul, Korea (I.J.L.); Cancer Hosp, Chinese Academy of Medical Sciences, Beijing, China (N.W.); and North Shore–Long Island Jewish Health System, New Hyde Park, NY (A.K.). Received Feb 15, 2005; revision requested Apr 12; revision received May 2; accepted June 3; final version accepted July 5. ELCAP supported in part by NIH grant R01-CA-63393, Eastman-Kodak, General Electric, and the National Cancer Institute. NYELCAP supported by the City of New York Dept of Health and Mental Hygiene; Starr Foundation; Empire Blue Cross and Blue Shield; New York Community Trust; New York State Office of Science, Technology, and Academic Research; the Rogers Family Fund; Weill Medical College of Cornell Univ; Cornell Univ; and AMDeC Foundation. Members of the ELCAP and NYELCAP study groups are listed in the Acknowledgments.
Address correspondence to C.I.H. (e-mail: chensch{at}med.cornell.edu).
 |
ABSTRACT
|
|---|
Purpose: To determine the frequency and natural course of mediastinal masses in asymptomatic people at high risk for lung cancer who were undergoing computed tomographic (CT) screening.
Materials and Methods: Informed consent and institutional review board approval for this HIPAA-compliant study were obtained at each participating institution. All documented mediastinal masses among the 9263 baseline and 11 126 annual repeat screenings performed in the Early Lung Cancer Action Project (ELCAP) and its successor project, the New York ELCAP, were identified. Two radiologists confirmed all cases, identified the location and measured the diameter (average of length and width) of each mass, and reviewed all subsequent CT and clinical and surgical results. The prevalence and incidence of mediastinal masses were then determined.
Results: Of the 9263 individuals, 71 had a mediastinal mass seen at baseline screening (prevalence of 0.77%). Of the 71 masses, 41 were thymic, 16 were thyroidal, two were esophageal cancers, six were tracheal-esophageal diverticula, and six were other masses. Among the 11 126 annual repeat screenings, only one new mediastinal mass was identified (incidence of 0.01%). This suggests a long average duration for mediastinal masses in asymptomatic people. Among the 41 thymic masses, five were larger than 3.0 cm in diameter, and all five were resected; of these five, one was a thymic carcinoma and four were noninvasive thymomas. Of the remaining 36 thymic masses, 25 were evaluated at follow-up CT 1 year later: Five had increased in diameter, two had decreased, and 18 remained unchanged. All 16 thyroid masses were due to goiter; none of these were changed at follow-up CT 1 year later.
Conclusion: Mediastinal masses found in the context of CT screening for lung cancer in asymptomatic people should be approached in a "conservative" manner; this includes thymic masses smaller than 3 cm in diameter, as most of these remain unchanged or even decrease in size.
© RSNA, 2006
|
INTRODUCTION
|
|---|
In computed tomographic (CT) screening for lung cancer, despite a primary diagnostic focus on nodules in the lung, mediastinal masses are found incidentally.
Mediastinal masses (abnormal by definition) occur anywhere in the mediastinum, including in the thymus, heart, and esophagus, and a mass in the neck, such as in the thyroid, may extend into the mediastinum. Descriptions of the normal and abnormal CT appearances of the thymus and thyroid were first published in the early 1980s (1–5)—soon after the introduction of CT scanners. Subsequently, additional information on normal and abnormal findings has been provided (6–9), including information on the frequency of abnormal findings, but this information typically comes from CT performed for diagnostic purposes rather than in the context of screening. Thus, the purpose of our study was to determine the frequency and natural course of mediastinal masses in asymptomatic people at high risk for lung cancer who were undergoing CT screening.
|
MATERIALS AND METHODS
|
|---|
Screening Cohort
This report is based on our experience with CT screening for lung cancer. This experience encompasses 9263 baseline screenings, 2968 of which were performed in the Early Lung Cancer Action Project (ELCAP) in 1993–2002 (10–12) and 6295 of which were performed in the subsequent New York ELCAP (NYELCAP) in 2001–2003. This experience also encompasses 11 126 repeat screenings performed within 6–18 months of the preceding screening, 5547 of which were performed in the ELCAP in 1994–2003 and 5579 of which were performed in the NYELCAP in 2001–2003. At baseline, the ages of the 9263 people ranged from 40 to 92 years (median, 65 years), pack-years of cigarette smoking ranged from 1 to 267 pack-years (median, 40 pack-years); 1019 participants (11%) reported asbestos exposure, and 4724 participants (51%) were women. Patient informed consent and institutional review board approval were obtained for this study at each of the participating institutions. The study was performed according to Health Insurance Portability and Accountability Act requirements.
The ELCAP was partially funded by a National Institutes of Health grant and by Eastman-Kodak, General Electric, and the National Cancer Institute. The NYELCAP was funded by the City of New York Department of Health and Mental Hygiene; the Starr Foundation; Empire Blue Cross and Blue Shield; the New York Community Trust; New York State Office of Science, Technology, and Academic Research; the Rogers Family Fund; Weill Medical College of Cornell University; Cornell University; and AMDeC Foundation. The authors had control of data and information submitted for publication.
CT Imaging
The images for both baseline and repeat screenings in the ELCAP (1993–1999) were initially obtained by using a CT scanner (High-Speed Advantage; GE Medical Systems, Milwaukee, Wis) at 140 kVp and 40 mA, with a pitch of 2 (as defined by the International Electrotechnical Commission) and a collimation of 10 mm. Later (1999–2001), images were obtained by using a scanner (Light-Speed; GE Medical Systems) at 120 kVp and 40 mA, with a pitch of 1.5 and a collimation of 2.5 mm. Most recently (2001–2003), images were obtained by using an eight-section scanner (Ultra 8; GE Medical Systems) at 120 kVp and 80 mA, with a pitch of 1.35 and a collimation of 1.25 mm. In the NYELCAP, the images were obtained by using single-section scanners at five institutions and multi–detector row CT scanners at seven, with the settings as used in the ELCAP. All CT scans were obtained, with patients in the supine position, from above the thoracic inlet (the level of the first thoracic vertebra, first ribs, and of the manubrium) down to the upper kidney in a single breath hold (about 15–20 seconds) at end inspiration following hyperventilation. Contrast material was not used.
Imaging Evaluation
In the ELCAP, images from the first 1000 (10) CT scans were read by at least two board-certified chest radiologists, whereas those from the last 1968 were read by a radiology fellow and a board-certified chest radiologist (11). In the NYELCAP, all CT scans were read locally by one board-certified chest radiologist and then, at the Coordinating Center at Weill Medical College of Cornell University, by another board-certified radiologist experienced in CT screening. All instances in which a mediastinal mass had been documented when the CT scan was initially read were again reviewed for the present study by two chest radiologists (I.J.L. and C.I.H., with 10 and 25 years of experience, respectively); images were first read independently and then reviewed for consensus. All cases of lymph node enlargement and vascular abnormalities, although tabulated, were excluded from this report. The review by both radiologists confirmed each of the prior findings of mediastinal mass.
The location of each mass was classified according to the mediastinal compartment—superior or inferior, with the latter subclassified as anterior, middle, or posterior—and also according to the likely organ involved on the basis of the location of the mass. The length and width of the mass were measured, and the "diameter" (an average of the length and width) was calculated. CT numbers were obtained for each mass to determine whether it was cystic (homogeneous CT numbers consistent with fluid) or fatty (CT numbers consistent with fat). Calcifications in and necrosis of the mass, as well as infiltration of the mediastinal fat surrounding it, were documented. All subsequent CT scans were reviewed to assess change in size (increase, no change, decrease). If the mass was resected, pathologic diagnosis was retrieved from the pathology report.
The thyroid area was always included on the CT scan, but for the purposes of this report we focused on thyroid masses below the thoracic inlet, where thyroid tissue may also be found (8). A mediastinal thyroid mass was classified as ectopic thyroid tissue if the CT number exceeded 100 HU on an unenhanced scan and the mass was not in continuity with the thyroid gland (in the neck), as a goiter if the CT number exceeded 100 HU and the mass was in continuity with the thyroid gland (9), and as a thyroid cyst if it had homogeneous CT numbers consistent with fluid and was in continuity with the thyroid gland. Anything else required further work-up.
Each thymic mass was first classified according to its shape: ovoid, bi-lobed, or arrowhead; the latter two represented benign appearance. If the shape was ovoid, the length, width, and height of the mass were defined and assessed in the usual way (1–3). If the mass was bi-lobed or arrowhead shaped, then the length was measured along the longest axes of the lobes, the width laterally to these, and the height perpendicular to both length and width. Among bi-lobed and arrowhead masses, only those with a short-axis measurement greater than 13 mm were considered masses (1–3), and among ovoid masses, only those with the short-axis measurement greater than 7 mm were considered masses (1–3). If the CT attenuation was consistent with that of fluid, the mass was classified as a thymic cyst. If the attenuation was consistent with that of fat with intermingling soft tissue, the mass was classified as a thymolipoma (7); otherwise, the mass was classified as a thymic tumor. If the mass was not a cyst and the size was larger than 3 cm or necrosis or infiltration of mediastinal fat was seen, further work-up was required.
Masses other than those in the thyroid and thymic gland were also documented as to length, width, height, and the average of these, as well as their location in the mediastinum.
Data Analysis
The prevalence of any given type of mediastinal mass was addressed in terms of findings on the initial CT scan at baseline screening and its incidence in terms of findings on the initial CT scan at annual repeat screenings following no such finding in the preceding screening. The average duration was calculated by using the well-known relationship of prevalence = incidence density x average duration (13). Note that sufficient significant digits, not the rounded incidence and prevalence values, need to be used in this calculation.
|
RESULTS
|
|---|
Masses: Prevalence and Incidence
In the 9263 baseline screenings, a mediastinal mass was identified in 71 cases, or 0.77%. The most common finding was an enlarged thymus (41 cases; prevalence, 0.45%). Thyroid mass was found in 16 cases (0.17%), and some other type of mass was found in 14 cases (0.15%; Table 1).
Among the 11 126 annual repeat screenings in those persons in whom no mediastinal mass had previously been identified, a mediastinal mass of any type was identified in one case (0.01%). This was a case of an ovoid thymic mass.
Taken together, the overall prevalence (0.77%) and the 1-year overall incidence (0.01%) suggest that for identifiable asymptomatic mediastinal masses, the average duration is 85 years; for thymic masses, it is shorter—about 49 years.
Thymic Masses
Of the 41 thymic masses identified at baseline screening, most were located at the anterior mediastinum (n = 39; Table 1) and were ovoid in shape (n = 39; Table 2). None of the masses with an arrowhead or bi-lobed appearance were larger than 3.0 cm in width, and none of these were resected. Five of 41 masses were greater than 3.0 cm in width and ovoid in appearance: One was a cyst. Three were immediately resected, of which two were noninvasive thymoma and the other was a thymic carcinoma (Fig 1). The fifth mass continued to increase in width over 3 years, despite the recommendation for resection; when it was finally resected, it was found to be a noninvasive thymoma. Among the remaining 36 of 41 masses, which were 0.7–3.0 cm in width, only two had an arrowhead or bi-lobed appearance (one decreased in size at 1-year follow-up and the other was not evaluated at follow-up CT; documented follow-up of both confirmed that neither were diagnosed as malignant), while the other 34 of 36 masses had an ovoid appearance. Of these 34 ovoid masses that were smaller than 3.0 cm in width and were evaluated at 1-year follow-up, 18 were unchanged in width, five had increased in size (Fig 2), two had decreased (Fig 3) (Table 3), and nine were not evaluated at follow-up CT (one was resected and diagnosed as thymoma, and the others have not been resected nor has there been a diagnosis of malignancy). Focal calcification was found in only two ovoid masses: One was an unresected mass, 0.8 cm in diameter, and the other was found in the single case of thymic carcinoma (Fig 1).
View larger version (118K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1: Baseline low-dose transverse CT scan in a 61-year-old man shows a mass (M) with calcifications in the anterior mediastinum. Diagnosis based on biopsy of the mass showed thymic carcinoma.
|
|
View larger version (76K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a: (a) Baseline low-dose transverse CT scan in a 64-year-old man shows an ovoid thymic mass (M) with a width of 8 mm. (b) On repeat screening CT scan 1 year later, the width of the mass (M) is increased to 10 mm.
|
|
View larger version (77K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b: (a) Baseline low-dose transverse CT scan in a 64-year-old man shows an ovoid thymic mass (M) with a width of 8 mm. (b) On repeat screening CT scan 1 year later, the width of the mass (M) is increased to 10 mm.
|
|
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a: (a) Baseline low-dose transverse CT scan in a 60-year-old man shows an ovoid thymic mass (M) with a width of 20 mm. (b) On annual repeat screening low-dose CT scan 1 year later, the width of the mass (M) is decreased to 10 mm.
|
|
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b: (a) Baseline low-dose transverse CT scan in a 60-year-old man shows an ovoid thymic mass (M) with a width of 20 mm. (b) On annual repeat screening low-dose CT scan 1 year later, the width of the mass (M) is decreased to 10 mm.
|
|
Thyroid Masses
All of the 16 thyroid masses in the mediastinum had a goiter. Of these 16, four also had a mass in addition to the goiter. None of these four masses was evaluated at biopsy or was resected; all of them had well-defined margins and CT findings suggestive of fluid, and all were unchanged at follow-up CT 1 year later.
Other Masses
Of the remaining 14 masses, eight were tracheal-esophageal masses. Two of these eight were esophageal cancers and six were diverticula. Of the other six of 14 mediastinal masses (none of which has been resected), two were solid and four were cystic. Of the two solid masses, one was a lipoma in the superior mediastinum, and the other was a cardiac lipoma. Of the four cystic masses, two in the superior and anterior mediastinum were considered a lymphangioma and pericardial cyst, respectively, based on the CT findings; one in the middle mediastinum was a bronchogenic cyst; and one in the posterior mediastinum in the paraspinal region was either a lymphangioma, bronchogenic cyst, or neurenteric cyst.
|
DISCUSSION
|
|---|
Our study addresses the occurrence of mediastinal masses in asymptomatic individuals at high risk enrolled in a program of CT screening for lung cancer. On the basis of 71 cases, the prevalence was 0.77%, and on the basis of only one case, the annual incidence was 0.01%. While these data are sparse from a statistical point of view, they do represent the largest prospectively collected series available. The data indicate a long average duration for an asymptomatic mediastinal mass and even for a thymic mass.
Overall, finding mediastinal masses in asymptomatic people, including masses in the thymus gland, should be approached in a "conservative" manner, particularly because the evidence presented here points to very slow growth in some people and regression in others.
Mediastinal thyroid masses are usually caused by goiters. None in our series grew or showed malignancy. However, malignancies have been found within these goiters (14), and thus careful follow-up is needed to assess change.
Thymic masses were the most frequent type of mediastinal mass, with a prevalence of 0.45% at baseline CT screening, but most remained unchanged in size 1 year later. Of the 25 cases of ovoid thymic mass that were evaluated at 1-year follow-up CT (all with an initial width of <3 cm at baseline), five masses (20%) increased in size, two (8%) decreased in size, and 18 (72%) remained unchanged in width 1 year later. The single case of an unresected mass with an initial width of larger than 3 cm continued to grow. Incidental findings of thymic masses typically call for further work-up to determine their cause. Although biopsy may typically be recommended (15), positron emission tomography or follow-up for growth are alternatives. For thymic masses that are larger than 3 cm in diameter, resection is recommended even when the mass is found to be a benign thymoma, because the surgery becomes more difficult (15) and, once they reach this size, they tend to grow. With respect to thymic masses of smaller than 3 cm in diameter, the long-term management remains unclear. We found growth as well as regression, with no adverse effect from waiting for at least 1 year of follow-up, and, while their histologic characteristics are unknown, it is likely that many are benign thymomas. However, sometimes tumors that appear both radiologically and pathologically benign, in fact, behave in a malignant fashion and thus further follow-up data will be necessary before reaching a conclusion as to the need for surgical intervention.
|
ADVANCE IN KNOWLEDGE
|
|---|
- Mediastinal masses found in the context of CT screening for lung cancer in asymptomatic people should be approached in a "conservative" manner; this includes thymic masses smaller than 3 cm, as most of these remained unchanged or even decreased in size.
|
ACKNOWLEDGMENTS
|
|---|
Maria C. Shiau, MD, participated in the radiologic readings at Columbia University Medical Center, New York, NY. ELCAP Investigators: Claudia I. Henschke, PhD, MD, principal investigator, Nasser Altorki, MD, Daniel Libby, MD, Dorothy I. McCauley, MD, Mark W. Pasmantier, MD, James P. Smith, MD, Madeline Vazquez, MD, and David Yankelevitz, MD (Weill Medical College of Cornell University); Olli S. Miettinen, MD, PhD (Weill Medical College of Cornell University and McGill University); and Georgeann McGuinness, MD, and David Naidich, MD (New York University Medical Center). NYELCAP Investigators: Claudia I. Henschke, principal investigator, David Yankelevitz, MD, and Dorothy McCauley, MD (Weill Medical College of Cornell University); Matthew Rifkin and Edward S. Fiore (State University of New York at Stony Brook, NY); Samuel Kopel (Maimonides Medical Center, Brooklyn, NY); Donald Klippenstein, Alan Litwin, and Peter A. Loud (Roswell Park Cancer Institute, Buffalo, NY); Leslie Kohman and Ernest M. Scalzetti (State University of New York, Upstate Medical Center, Syracuse, NY); Arfa Khan and Rakesh Shah (North Shore-Long Island Jewish Health System, New Hyde Park, NY); John H. M. Austin and Gregory D. N. Pearson (Columbia University Medical Center, New York, NY); David S. Mendelson (Mount Sinai School of Medicine, New York, NY); Robert T. Heelan and Michelle S. Ginsberg (Memorial Sloan-Kettering Cancer Center, New York, NY); Terence A. S. Matalon (New York Medical College, Valhalla, NY); Peter H. Wiernik (Our Lady of Mercy Cancer Center, New York Medical College, Bronx, NY).
|
FOOTNOTES
|
|---|
Abbreviations: ELCAP = Early Lung Cancer Action Project • NYELCAP = New York ELCAP
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, C.I.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, C.I.H.; clinical studies, all authors; statistical analysis, C.I.H.; and manuscript editing, all authors
|
References
|
|---|
- Baron RL, Lee JK, Sagel SS, Peterson RR. Computed tomography of the normal thymus. Radiology 1982;142:121–125.[Abstract/Free Full Text]
- Baron RL, Lee JK, Sagel SS, Levitt RG. Computed tomography of the abnormal thymus. Radiology 1982;142:127–134.[Abstract/Free Full Text]
- Francis IR, Glazer GM, Bookstein FL, Gross BH. The thymus: reexamination of age-related changes in size and shape. AJR Am J Roentgenol 1985;145:249–254.[Abstract/Free Full Text]
- Glazer GM, Axel L, Moss AA. CT diagnosis of mediastinal thyroid. AJR Am J Roentgenol 1982;138:495–498.[Abstract/Free Full Text]
- Binder RE, Pugatch RD, Faling LJ, Kanter RA, Sawin CT. Diagnosis of posterior mediastinal goiter by computed tomography. J Comput Assist Tomogr 1980;4:550–552.[Medline]
- Choi YW, McAdams HP, Jeon SC, et al. Idiopathic multilocular thymic cyst: CT features with clinical and histopathologic correlation. AJR Am J Roentgenol 2001;177:881–885.[Abstract/Free Full Text]
- Rosado-de-Christenson ML, Pugatch RD, Moran CA, Galobardes J. Thymolipoma: analysis of 27 cases. Radiology 1994;193:121–126.[Abstract/Free Full Text]
- Naik KS, Bury RF. Imaging the thyroid. Clin Radiol 1998;53:630–639.[CrossRef][Medline]
- Weber AL, Randolph G, Aksoy FG. The thyroid and parathyroid glands. Radiol Clin North Am 2000;38:1105–1129.[CrossRef][Medline]
- Henschke CI, McCauley DI, Yankelevitz DF, et al. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999;354:99–105.[CrossRef][Medline]
- Henschke CI, Naidich DP, Yankelevitz DF, et al. Early Lung Cancer Action Project: initial results of annual repeat screening. Cancer 2001;92:153–159.[CrossRef][Medline]
- Henschke CI, Yankelevitz DF, Smith JP, et al. CT screening for lung cancer: assessing a regimen's diagnostic performance. Clin Imaging 2004;28:317–321.[Medline]
- MacMahon B, Trichopoulos D. Epidemiology: principles and methods. Boston, Mass: Little, Brown, 1996; 57.
- Crandall R, Henschke CI, Godwin TA, Yankelevitz DF, Skinner DB. MR imaging of retrocaval insular thyroid carcinoma. Comput Surg 1995;14:175–178.
- Shields TW. Thymic tumors. In: Shields TW, Locicero J, Ponn RB, Rusch VW, eds. General thoracic surgery. 6th ed. Vol 2. Philadelphia, Pa: Lippincott, Williams & Wilkins, 2005; 2581–2616.
This article has been cited by other articles:
|
|
|
|
 
R. K. Gill, M. F. Vazquez, A. Kramer, M. Hames, L. Zhang, K. Heselmeyer-Haddad, T. Ried, K. Shilo, C. Henschke, D. Yankelevitz, et al.
The Use of Genetic Markers to Identify Lung Cancer in Fine Needle Aspiration Samples
Clin. Cancer Res.,
November 15, 2008;
14(22):
7481 - 7487.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. S. Gierada, T. K. Pilgram, M. Ford, R. M. Fagerstrom, T. R. Church, H. Nath, K. Garg, and D. C. Strollo
Lung Cancer: Interobserver Agreement on Interpretation of Pulmonary Findings at Low-Dose CT Screening
Radiology,
December 1, 2007;
246(1):
265 - 272.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
