| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Depts of Radiology (J.C.) and Cardiothoracic Surgery (R.B.L.), Univ of Wisconsin Hospital and Clinics, E3/311 Clinical Science Center, 600 Highland Ave, Madison, WI 53792-3252; Dept of Radiology, Univ of Michigan Medical School, Ann Arbor (E.A.K.); Dept of Radiology, Univ of Pittsburgh Medical School, Pa (J.L.); Dept of Radiology, Duke Univ Medical School, Durham, NC (H.P.M.); Dept of Radiology, Stanford Univ Medical School, Stanford, Calif (A.N.L.); Dept of Radiology, College of Physicians and Surgeons of Columbia Univ, New York, NY (M.S.); and Mallinckrodt Inst of Radiology, Washington Univ School of Medicine, St Louis, Mo (J.S.). Received July 13, 2001; revision requested August 13; revision received October 11; accepted October 25. Address correspondence to J.C. (e-mail: jcollin4@facstaff.wisc.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The study population was composed of 2,168 consecutive patients at seven lung transplantation centers who survived longer than 1 month after lung transplantation. Medical records, chest radiographs, and computed tomographic (CT) scans obtained at the time of diagnosis and prior images when available were reviewed for various items of information and imaging features.
RESULTS: Twenty-four (1%) of the 2,168 patients, all with single-lung transplants, developed cancer in the native lung. Eighteen patients had emphysema, and six had pulmonary fibrosis. The frequencies of cancer in patients with emphysema and fibrosis were 2% (18 of 859 patients) and 4% (six of 147 patients), respectively. Twelve (50%) of their 24 cancers were detected at chest radiography. Fourteen (58%) patients had clinical symptoms. Twenty-one (88%) of the 24 patients had one (n = 11) or more (n = 10) nodules, and nine (38%) had one (n = 8) or more (n = 1) masses visible on CT scans. Nodules and masses were visible on 12 (50%) and seven (29%) of 24 chest radiographs, respectively. Eleven (48%) of 23 cancers for which prior chest radiographs were available were seen retrospectively on prior chest radiographs.
CONCLUSION: Bronchogenic carcinoma develops in the native lung of transplant recipients with emphysema and pulmonary fibrosis with frequencies of 2% and 4%, respectively. The carcinomas most commonly manifest as a pulmonary nodule or mass on chest radiographs, with more nodules seen on CT scans.
© RSNA, 2002
Index terms: Lung, abnormalities, 60.3211, 60.3212, 60.3213, 60.3215, 60.3217, 60.3235, 3237 • Lung, CT, 60.12112, 60.12115, 60.12118, 60.12119 • Lung neoplasms, 60.3211, 60.3212, 60.3213, 60.3215, 60.3217, 60.3235, 3237 • Lung transplantation, 60.458 • Transplantation, 60.458
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
The Registry of the International Society for Heart and Lung Transplantation has reported the incidence of malignancy at 1-year follow-up after lung transplantation to be 4.3%, with 50.7% due to lymphoproliferative disorders, and at 5-year follow-up to be 7.7%, with 17.8% due to lymphoproliferative disorders and 50.7% due to skin malignancies (3). A number of studies have shown the incidence of lymphoproliferative disease in lung transplant recipients to be 1.8%–20.0% (4–8). Recently, the authors of several published reports (2,9–12) have described a total of eight cases of bronchogenic carcinoma developing after lung transplantation. The purpose of this study was to determine the frequency, clinical characteristics, and imaging findings of bronchogenic carcinoma in patients surviving longer than 1 month after lung transplantation at seven lung transplantation centers.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
Immunosuppression protocols differed among patients and institutions. All patients received prednisone. Other immunosuppressive agents prescribed included cyclosporine (16 patients), azathioprine (13 patients), mycophenolate mofetil (seven patients), and tacrolimus (six patients). The most common regimen was cyclosporine, azathioprine, and prednisone (seven patients).
The seven institutions (the University of Wisconsin; the University of Michigan; the University of Pittsburgh, Penn; Duke University in North Carolina; Stanford University in California; Columbia University in New York; and the Mallinckrodt Institute of Radiology in Missouri) had different protocols for obtaining chest radiographs after transplantation. In the immediate postoperative period at all institutions, chest radiographs were obtained daily as needed. Additional protocols included chest radiography every 3 weeks for 12 weeks, then at 6, 8, and 12 months and annually thereafter; every month for 6 months, then every 3 months; every week for 10 weeks, then every month for 1 year, followed by every 2–3 months thereafter; every 3 months for 1 year; every week for 2 months, then every 2 weeks for 2 months, followed by every 9 months thereafter; and at 6 weeks, then every 3 months for 21 months. Routine surveillance CT scans were obtained at only one institution as part of an ongoing study at 3, 6, 12, 18, and 24 months after transplantation. At all institutions, CT scans were obtained to further evaluate clinical or chest radiographic findings.
CT scans were obtained with helical single–detector row (19 patients), helical multi–detector row (three patients), and conventional CT scanners (two patients). High-resolution (thin-section) CT images were obtained in only one patient. Collimation varied as follows: 5.00 (10 patients), 7.00 (eight patients), 10.00 (four patients), 8.00 (one patient), and 1.00 mm (one patient). Pitch varied as follows: 1.00 (11 patients), 1.60 (four patients), 1.50 (two patients), 1.70 (one patient), and 2.00 (one patient). Pitch was not reported for patients undergoing conventional or multi–detector row CT. Thin sections through nodules were obtained in four patients by using 2.00- (two patients), 1.25- (one patient), or 1.50-mm (one patient) collimation. Contrast material was intravenously administered to eight patients and was not administered to 16 patients prior to scanning.
Patient charts were reviewed by the radiologist collaborator at each institution (J.C., E.A.K., J.L., H.P.M., A.N.L., M.S., J.S.) to determine each patient’s age and sex, lung disease, method of detection and diagnosis of cancer, cancer staging work-up, clinical symptoms at the time of cancer detection, time from transplantation to cancer diagnosis, history of cigarette smoking and time from quitting smoking to transplantation, cancerous histologic findings, cancer treatment, cancer stage at thoracotomy, disease-free interval after diagnosis, time from diagnosis to death, and outcome.
Posteroanterior and lateral chest radiographs and CT scans obtained at the time of diagnosis were available for 22 of the 24 patients. For two patients, only a frontal chest radiograph (not labeled as posteroanterior or anteroposterior) and CT scan were available. Hard copies were available for all patients except one, for whom only electronic images were available. Prior chest radiographs and CT scans were available for 23 (96%) of the 24 patients. The examinations were analyzed by the contributing chest radiologist (J.C., E.A.K., J.L., H.P.M., A.N.L., M.S., J.S.) and a second chest radiologist (E.A.K.) separately, and any instances of disagreement were resolved by a third chest radiologist (J.C.). Chest radiographs were analyzed for the number, size, border characteristics (irregular, circumscribed), calcification, and lobar location of nodules (
3 cm in diameter) and masses (>3 cm in diameter); pleural effusion; hilar or mediastinal lymph node enlargement; and other findings related to the cancer. CT scans were analyzed for nodules and masses (number, size, border characteristics [irregular, circumscribed], calcification, halo of ground-glass attenuation, air bronchograms, and lobar location), pleural effusion (thickness measured manually with calipers), hilar or mediastinal lymph node enlargement (short-axis diameter > 1 cm in diameter), endobronchial lesions, tumor invasion (chest wall, diaphragm, heart, great vessels, trachea, esophagus, vertebral body, carina, mediastinal fat), postobstructive atelectasis or pneumonia, and other findings related to the cancer. For both radiographs and CT scans, it was noted whether a lesion was in the native or transplant lung and whether prior images were available, what time interval between prior and detection examinations was, whether the cancer could be seen retrospectively on prior images, and why cancers had been missed on prior images.
Institutional review board approval was required and obtained at three institutions (the University of Michigan, Columbia University, and the University of Pittsburgh). Approval is currently required but was not required at the time of the study at two institutions (Duke University and the University of Wisconsin) and was not required at two institutions (the Mallinckrodt Institute of Radiology and Stanford University). Patient consent was not required at any of the seven institutions.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
Clinical Data
The time from transplantation to diagnosis was 4–86 months (mean, 34 months). The method of detection was chest radiography in 12 (50%) of the 24 patients; CT in eight (33%); symptoms of pneumonia in one (4%); chest radiography, CT, and symptoms of pneumonia in one (4%); chest radiography and symptoms of pneumonia in one (4%); and right shoulder pain in one (4%). Fourteen (58%) of the 24 patients had clinical symptoms, including increased dyspnea (n = 3); cough, fatigue, and weakness (n = 2); fatigue and weight loss (n = 1); wheezing (n = 1); increased dyspnea and cough (n = 1); right shoulder pain (n = 1); cough and left-sided chest pain (n = 1); cough and fever (n = 1); fever, night sweats, anorexia, and weight loss (n = 1); hemoptysis, cough, congestion, and fatigue (n = 1); and sternal pain (n = 1).
Information regarding cigarette smoking history was available for 21 patients. Nineteen (90%) of these 21 patients had a history of cigarette smoking, and two (10%) had no history of smoking. Information on pack-years (number of packs per day x number of years smoking) was available for 16 patients, and pack-years ranged from 20 to 200 (mean, 53). Information regarding the time interval between when patients quit smoking and the date of transplantation was available for 16 patients. The interval range was 1–25 years (mean, 8 years). Only one patient was reported to be smoking cigarettes after transplantation.
The methods of diagnosis were fine-needle aspiration (10 patients), transbronchial biopsy (seven patients), mediastinotomy (one patient), left lower lobe resection (one patient), right lower lobe resection (one patient), video-assisted thoracotomy (one patient), pleural fluid cytologic examination (one patient), right upper lobe resection (one patient), and sternal biopsy (one patient).
The pathologic diagnoses were squamous cell carcinoma (eight patients), adenocarcinoma (seven patients), non–small cell poorly differentiated carcinoma (four patients), anaplastic carcinoma (one patient), adenocarcinoma with small cell component (one patient), adenosquamous carcinoma (one patient), bronchioloalveolar cell carcinoma (one patient), and carcinosarcoma with chondroid and liposarcomatous differentiation (one patient).
Nine (38%) of the 24 patients were treated with surgery, two (8%) were treated with surgery and chemotherapy, eight (33%) received no treatment, two (8%) were lost to follow-up after diagnosis, two (8%) were treated with radiation alone, and one (4%) was treated with chemotherapy alone. The patients not treated had advanced disease and died 3 days to 2 months after diagnosis (two patients) or had recently received a diagnosis, with treatment yet to be determined (two patients). Surgical procedures included pneumonectomy (two patients), wedge resection (two patients), right lower lobectomy (two patients), right upper lobectomy (two patients), left upper lobectomy (two patients), and surgery not otherwise specified (one patient).
The Revised International System for Staging Lung Cancer was utilized in staging the carcinomas (13). Information on surgical staging was available for 16 patients. Patients whose cancer was not staged had advanced disease and died 3 days to 3 months after diagnosis (n = 6); had advanced disease, refused staging, and died 21 months after diagnosis (n = 1); or had recently received a diagnosis, with staging yet to be performed (n = 1). Of patients whose cancer was staged, five had cancer staged as T1N0M0 (IA); two, T1N0MX (IA); two, T2N0M0 (IB); one, IIIA not otherwise defined; one, T4N0MX (IIIB); one, IIIB not otherwise defined; two, T4 not otherwise defined; and two, M1 (IV) not otherwise defined. This resulted in nine (56%) of 16 patients with stage I cancer and seven (44%) with stage III or IV cancer. Seven (39%) of the 18 patients with emphysema had stage I tumors, and two (33%) of the six patients with pulmonary fibrosis had stage I tumors. When the cancer was detected at surveillance chest radiography, patients had stage I (seven patients) (Fig 1), III (one patient), or IV (one patient) tumors. Seven (78%) of the nine tumors were stage I. When the cancer was detected with other methods, patients had stage I (n = 2), III (n = 2), or IV (n = 3) tumors. Only two (29%) of the seven tumors were stage I. Twenty-three patients underwent work-up for metastases. Thirteen (56%) of the 23 patients had metastases, and 10 (43%) did not. Metastases involved the contralateral lung (three patients); liver (two patients); brain (one patient); brain and liver (one patient); scapula (one patient); bone not otherwise specified (one patient); chest wall, contralateral lung, heart, and adrenal gland (one patient); contralateral lung, diaphragm, chest wall, and pericardium (one patient); ipsilateral lung (one patient); or both lungs (one patient).
|
|
At the time this article was written, 15 (65%) of 23 patients whose status was known were dead, and eight (35%) were alive. One patient was lost to follow-up, and this information could not be obtained. The time from diagnosis to death was 3 days to 21 months (mean, 7 months).
Chest Radiographic Findings
The two reviewers disagreed on eight findings on four chest radiographs, and consensus was achieved with the input of a third reviewer (J.C.). Twelve (50%) of the 24 patients had one (10 patients) or more (two patients) nodules visible at chest radiography. Nodules were 21–30 mm in diameter in five patients, 11–20 mm in diameter in five patients, and 5–10 mm in diameter in two patients. Nodule borders were irregular in nine patients and circumscribed in three. All nodules were noncalcified. The location of the nodules was in the right upper lobe in four patients, left lower lobe in three patients, right lower lobe in two patients, left upper lobe in one patient, right and left upper and left lower lobes in one patient, and right upper and middle lobes in one patient.
Seven (29%) of the 24 patients had a mass 3.1–5.0 cm (four patients) or 5.1–7.0 cm (three patients) in diameter that was visible at chest radiography. The borders were irregular in four masses and circumscribed in three. All masses were noncalcified. The masses were located in the right upper lobe in two patients, left upper lobe in two patients, right lower lobe in one patient, left lower lobe in one patient, or right upper and middle lobes in one patient. One patient with a mass also had several nodules.
Three patients had pleural effusion. Four patients had right hilar (n = 1), left hilar (n = 1), right hilar and mediastinal (n = 1), or bilateral hilar and mediastinal (n = 1) lymph node enlargement.
Prior chest radiographs were available for 23 patients. Eleven (48%) of their 23 tumors were seen retrospectively on prior chest radiographs. Cancers were missed due to superimposition of changes of pulmonary fibrosis (two tumors); a combination of small size, overlapping rib, and low opacity (two tumors); perceptual distraction by surgical staple lines (two tumors) (Fig 2); perceptual error due to subtlety of changes, as compared with prior chest radiographs (two tumors); misinterpretation of radiographic findings as progression of pulmonary fibrosis (one tumor) (Fig 3); misinterpretation of radiographic findings as pneumonia (one tumor); and visualization of the radiographic abnormality on only the lateral view (one tumor) (Fig 4). The time interval between prior and detection chest radiographs was 1
–28 months (mean, 9 months).
|
|
|
|
|
|
|
|
|
Nine (38%) of the 24 patients had one (eight patients) or more than one (one patient) mass visible at CT. Masses were 5.1–7.0 cm in diameter in four patients, 3.1–5.0 cm in three patients, both 3.1–5.0 and 5.1–7.0 cm in one patient, and larger than 7 cm in one patient. The borders were irregular in seven patients and circumscribed in two patients. Eight masses were noncalcified, and one had punctate and amorphous calcification. The locations of the masses were the right upper lobe in three patients, left upper lobe in two patients, left lower lobe in two patients, right lower lobe in one patient, and right upper and middle lobes in one patient. One mass contained air bronchograms. Five patients had both a mass and nodules.
Pleural effusion was visible in 11 patients (on the right side in seven, on the left side in two, and bilaterally in two) at CT and was 0.1 to several centimeters in thickness. In six patients, right hilar and mediastinal (three patients), bilateral hilar and mediastinal (one patient), mediastinal (one patient), or left hilar and mediastinal lymph node enlargement (one patient) was seen. One endobronchial lesion was identified. Three patients had tumors invading the carina (one patient), mediastinal fat (one patient), or both the chest wall and mediastinal fat (one patient). One patient each had postobstructive atelectasis, postobstructive pneumonia, and both postobstructive atelectasis and pneumonia. Beaded septal thickening suggestive of lymphangitic tumor spread was associated with four tumors in four patients.
Prior CT scans were available for 23 (96%) of the 24 patients. Three (13%) of their 23 tumors were seen retrospectively on prior CT scans. The interval between acquisition of prior and current CT scans was 9–24 months (mean, 15 months). The missed tumors were subtle small lesions on prior CT scans (one tumor) (Fig 5) or were obscured by pulmonary fibrosis on prior scans (two tumors). One tumor was detected on CT scans as a subtle change from the CT scan obtained 4 months earlier (Fig 6). Biopsy was not performed until a chest radiograph obtained 9 months later showed marked enlargement of the tumor.
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
Bronchogenic cancer was detected at chest radiography in 12 (50%) of the 24 patients in the present study. This compares with chest radiographic detection in five (71%) of seven heart or lung recipients with bronchogenic cancer reported by Choi et al (9). In their study, one patient with pulmonary fibrosis and a history of cigarette smoking developed bronchogenic cancer in the native lung. This cancer was not detected radiographically, as it was obscured by pulmonary fibrosis. La Fianza et al (12) reported a case of bronchogenic carcinoma detected at chest radiography that developed in the native lung of a single-lung recipient with pulmonary fibrosis. The smoking history of this patient was not reported. Fourteen (58%) of the 24 patients in the present study had clinical symptoms. Most asymptomatic cancers would have been undetected or detected later if routine surveillance chest radiography had not been performed after transplantation.
Most cases (85%) of lung cancer are associated with cigarette smoking (19). In the cohort of patients in the current study, 19 (90%) of 21 patients had a history of cigarette smoking prior to transplantation. There is also a correlation between advancing age and the risk of developing lung cancer (20). The mean age of the patients in the current study was 58 years. These risk factors, in concert with impaired surveillance mechanisms resulting from long-term immunosuppression in transplant recipients (which promotes development of cancer) and the increased risk of lung cancer in patients with pulmonary fibrosis, put many lung transplant recipients into a high-risk category for developing bronchogenic carcinoma.
The most common chest radiographic finding of bronchogenic carcinoma was a solitary pulmonary nodule larger than 10 mm in diameter, seen in 10 (42%) of the 24 patients. Seven (29%) of these 24 patients had a mass larger than 3 cm in diameter that was visible at chest radiography. Many more nodules were seen at CT, as compared with chest radiography. Twenty-one (88%) of the 24 patients had one or more nodules visible at CT, as compared with 12 (50%) visible at chest radiography. This is in keeping with other studies (17) showing that CT markedly enhances detection of pulmonary nodules. At CT, the most common findings were one or more pulmonary nodules larger than 10 mm in diameter, seen in 17 (71%) of the 24 patients, or a pulmonary mass, seen in nine (38%) patients. Postobstructive atelectasis or pneumonia was not a common radiologic finding, seen in only three (13%) patients.
At retrospective review, 11 (48%) of the 23 cancers were seen on prior chest radiographs. Other studies (21–23) have shown a 44%–82% rate of missed bronchogenic cancer. Perceptual error and lesion conspicuity have been cited as reasons for missing cancers on chest radiographs (24–26). A solitary uncalcified lesion smaller than 6 mm in diameter is rarely appreciated and then usually only in retrospect when the lesion has grown and has been detected on subsequent radiographs. A solitary nodule 8–10 mm in diameter has an approximately 50% chance of being detected (24,27). Larger nodules and even masses 3–4 cm in diameter can also be overlooked if situated over the convexity of the lung or in the paramediastinal areas. Pulmonary fibrosis or other diffuse lung abnormalities also can contribute to missed cancers (28). In the cohort of patients in the current study, two cancers were partially obscured by pulmonary fibrosis on prior chest radiographs, and another cancer was missed because of misinterpretation of the radiographic findings as progression of pulmonary fibrosis in the native lung. Distracting radiographic findings, such as staple lines from prior surgery, perceptual errors due to the subtlety of changes from the prior radiograph to the detection radiograph, overlapping bone markings, visualization of radiographic abnormalities on only the lateral view, and misinterpretation of the radiographic findings as pneumonia, accounted for the remainder of the misses.
The prognosis for patients with non–small cell lung cancer relates to many factors, one of the most important being the stage of disease at diagnosis. Individuals who have small T1 (<3 cm) peripheral tumors that can be surgically removed and who do not have clinical or imaging evidence that the tumor has spread (surgical stage I) have survival rates of 60%–80% (13,29–32). Unfortunately, more than 50% of patients will have distant metastases at the time of diagnosis, and only 20%–25% will be potentially resectable for cure (14). In the present series, nine (56%) of the 16 patients with surgical staging information had stage I disease, and seven (44%) had stage III or IV disease. Of the cases for which surgical staging information was available, cancers detected with surveillance chest radiography (n = 9) had a higher incidence of stage I tumors (n = 7), as compared with cancers detected with other methods (n = 2). The cancers detected with surveillance chest radiography also had a lower incidence of stage III or IV tumors (n = 2), as compared with cancers detected with other methods (n = 5). This supports the practice of surveillance chest radiography in the lung transplant recipient population. Since surveillance CT was not performed at most of the centers participating in the current study, data regarding the utility of this test in the lung transplant recipient population are not available from the current study.
In summary, bronchogenic carcinoma develops in the native lung in lung transplant recipients with emphysema and pulmonary fibrosis with frequencies of 2% and 4%, respectively. This rate is similar to that in other high-risk populations (eg, elderly smokers with emphysema or other chronic lung disease). A majority of cancers are associated with a poor prognosis. The most common imaging manifestations are a solitary pulmonary nodule or mass. Further studies are needed to determine whether either surveillance chest radiography or CT result in a decrease in patient mortality.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T J Williams and J W Wilson Challenges in pulmonary fibrosis: 7 {middle dot} Novel therapies and lung transplantation Thorax, March 1, 2008; 63(3): 277 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P N Chhajed, L Bubendorf, H Hirsch, A Boehler, W Weder, and M Tamm Mesothelioma after lung transplantation. Thorax, October 1, 2006; 61(10): 916 - 917. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Ahmed, M. B. Marshall, J. C. Kucharczuk, L. R. Kaiser, and J. B. Shrager Lung Cancer in Transplant Recipients: A Single-Institution Experience Arch Surg, August 1, 2004; 139(8): 902 - 906. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Kotloff, V. N. Ahya, and S. W. Crawford Pulmonary Complications of Solid Organ and Hematopoietic Stem Cell Transplantation Am. J. Respir. Crit. Care Med., July 1, 2004; 170(1): 22 - 48. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.M. Kotloff and V.N. Ahya Medical complications of lung transplantation Eur. Respir. J., February 1, 2004; 23(2): 334 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Konen, G. L. Weisbrod, S. Pakhale, T. Chung, N. S. Paul, and M. A. Hutcheon Fibrosis of the Upper Lobes: A Newly Identified Late-Onset Complication After Lung Transplantation? Am. J. Roentgenol., December 1, 2003; 181(6): 1539 - 1543. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
